Application of proteomics to identify biomarkers of … · Application of proteomics to identify...
261
Application of proteomics to identify biomarkers of myopathy Thesis submitted for the degree of DOCTOR OF PHILOSOPHY at by Theophilus Olusegun Dare 2009 Department of Experimental Medicine and Toxicology Division of Investigative Science Faculty of Medicine Imperial College London Hammersmith Hospital Campus Du Cane Road London, W12 ONN United Kingdom
Transcript of Application of proteomics to identify biomarkers of … · Application of proteomics to identify...
Application of proteomics to identify biomarkers of myopathy
Thesis submitted for the degree of
DOCTOR OF PHILOSOPHY
at
by
Theophilus Olusegun Dare
2009
Department of Experimental Medicine and Toxicology
Division of Investigative Science
Faculty of Medicine
Imperial College London
Hammersmith Hospital Campus
Du Cane Road
London, W12 ONN
United Kingdom
2
For Hayley, Lauren and Cameron
3
Acknowledgements
I would like to express my sincere gratitude to Dr Robert Edwards whose expertise,
guidance and patience have been invaluable during this PhD and whose tutorship
has changed my approach to science almost beyond recognition. His advice and
kindness got me through some of my most difficult periods and I am sure that I
would not have completed this thesis without his help. I would also like to thank my
industrial supervisor, Dr Brian Sweatman, and the head of the Division of
Investigative Science at Imperial College London, Professor Martin Wilkins, for their
guidance and support during this period.
I would like to thank GlaxoSmithKline R&D for giving me the opportunity to study for
this PhD and also for the financial support. My thanks also go to my friends and
colleagues within the department of Safety Assessment at GlaxoSmithKline R&D
who were also extremely supportive during this period of study. In particular, I
would like to thank Dr Andy Nicholls, Su Moore and Dr John Haselden for their
patience and understanding during the writing-up period of this thesis. I would also
like to express my gratitude to Malcolm York, Tom Williams and Dr Ian Pyrah who
encouraged me to register for this PhD, Dr Cathy Waterfield and Dr Kevin Buchan
for their guidance in the early stages and Dr David Hassall for introducing me to the
Experimental Medicine and Toxicology group at Imperial College London. Thanks
also go to Aubrey Swain, Ian Roman, Paul McGill and Nikol Simecek for their
valuable technical help and also to my fellow part-time students at GlaxoSmithKline
R&D for their general encouragement. I would also like to thank Jim Armitage and
the rest of the Investigative Preclinical Toxicology department for making the
workplace a more enjoyable experience for 5 of the 7 days spent there every week,
and the soon-to-be Dr Mark Hodson for making the other 2 days almost bearable.
Thanks also go to my friends and colleagues within the Faculty of Medicine and the
Division of Investigative Science at Imperial College London, whose help and
fellowship was greatly appreciated. In particular, I would like to thank Dr Vahitha
Abdul Salam and Dr Zheying Zhu for their advice and practical help, and also Dr
John Cupitt for his programming expertise. Thanks also to the rest of the
department for making me feel like part of the team.
4
In amongst all of these messages of gratitude, I would also like to take the
opportunity to apologise to all my family and friends for all the get-togethers I
missed, the birthdays I forgot and the phone calls I ignored. Thank you for
understanding.
A big ‘thank-you’ goes to Stephanie and Jarlath (Granma and Grampus) for their
day-to-day help. It has meant so much to me and I appreciate it immensely. Lastly,
my largest thanks are reserved for my wife, Hayley, and also for my two lovely
children, Lauren and Cameron, who supported me all the way and made huge
sacrifices in order for me to complete this PhD. I am indebted to you for your love,
support and patience, and I dedicate this thesis to the three of you in recognition of
this.
5
Declaration This thesis is a summary of research conducted within the Division of
Investigative Science, Imperial College London (Hammersmith, UK), and at
GlaxoSmithKline R&D (Ware, UK). The work detailed in this thesis
demonstrates the novel uses of surface-enhanced laser desorption/ionization
and also label-free quantitative proteomics for the discovery of biomarkers of
compound-induced skeletal muscle toxicity in cultured myoblasts or myotubes.
This thesis is the result of my own work and includes nothing which is the
outcome of work done in collaboration, except where specifically indicated in
the text. None of the research described, in its entirety or in part, has been
submitted for any other qualification at any other university.
6
Abstract
The assessment of skeletal muscle toxicity is important in the safety evaluation of
new chemical entities, particularly those used as lipid-lowering agents. As current
strategies for assessing myopathy are based on in vivo measurements, which
sometimes lack sensitivity and specificity, there is a need for biomarkers that better
predict myopathy and also for an approach that is more amenable to high-
throughput screening. Here, the use of cultured rat L6 myoblasts and myotubes
was explored in combination with a variety of techniques to identify potential
biomarkers, following exposure to a known experimental myotoxicant, 2,3,5,6
tetramethyl-p-phenylenediamine (TMPD), or a highly selective peroxisome
proliferator-activated receptor δ agonist, 4-[2-(3-fluoro-4-trifluoromethyl-phenyl)-4-
methyl-thiazol-5-ylmethylsulfanyl]-2-methyl-phenoxy-acetic acid (GWδ). The use of candidate biomarkers, based on the utilisation of a variety of established
and putative markers of myopathy, proved to be of little additional value in
myoblasts treated with low concentrations of TMPD. In myotubes, however,
intracellular levels of aldolase and creatine kinase appeared to be responsive
markers. Proteomic analysis of TMPD and GWδ-treated skeletal muscle cells using
surface-enhanced laser desorption/ionization-time of flight-mass spectrometry
(SELDI-TOF-MS) showed changes in the levels of 16 protein ions in myoblasts and
12 changes in myotubes, although SELDI-TOF-MS did not lend itself to the
conclusive identification of these ions. Analysis of the muscle cell proteins using
label-free quantitative proteomics was able to simultaneously quantify TMPD-
induced changes and identify the proteins involved. In this way, 8 proteins in
myoblasts and 10 proteins in myotubes were discovered to be the most responsive
proteins to treatment. Changes in the levels of Rho GDP dissociation inhibitor (GDI)
alpha in myoblasts and Myosin light chain 2 in myotubes were confirmed by
immunoblotting, and these proteins were also responsive to treatment with GWδ
and pravastatin.
In conclusion, cultured skeletal muscle cells were responsive to the effects of
potentially myotoxic agents, and proteomics was successful in identifying and
quantifying proteins that may act as biomarkers of effect.
7
Contents
Acknowledgements .......................................................................................3
Declaration .....................................................................................................5
Abstract ..........................................................................................................6
Contents .........................................................................................................7
List of figures ...............................................................................................11
List of tables.................................................................................................15
Abbreviations ...............................................................................................17
1. Introduction ..............................................................................................22
1.1 Pharmaceutical safety assessment..........................................................22
1.2 Adverse drug reactions ............................................................................24
1.2.1 Target organ toxicity ......................................................................26
1.2.2 Skeletal muscle toxicity..................................................................26
1.2.3 Assessment of myopathy...............................................................30
1.2.4 Models used in toxicology..............................................................32
1.3 Tools used in biomarker discovery...........................................................33
1.3.1 Metabolic profiling..........................................................................35
1.3.2 Transcriptomics .............................................................................36
1.4 Proteomics ...............................................................................................37
1.4.1 Enzymatic digestion of proteins .....................................................40
1.4.2 Protein or peptide separation.........................................................41
1.4.3 Mass spectrometry ........................................................................43
1.4.3.1 Matrix-assisted laser desorption/ionization..............................44
1.4.3.2 Electrospray ionisation ............................................................45
1.4.3.3 Time of flight mass analysis ....................................................47
1.4.3.4 Ion trap mass analysis.............................................................50
1.4.4 Surface-enhanced laser desorption/ionization ...............................51
1.4.5 Quantitative proteomics .................................................................53
1.4.5.1 Absolute or relative quantitation ..............................................53
8
1.4.5.2 Multidimensional protein identification technology...................54
1.4.5.3 GeLC-MS ................................................................................55
1.5 Summary..................................................................................................56
1.6 Hypothesis ...............................................................................................56
1.7 Project aims .............................................................................................56
2. Materials and methods ............................................................................57
2.1 Cell culture and treatment ........................................................................59
2.2 Cell morphology .......................................................................................60
2.3 Total intracellular protein..........................................................................61
2.4 MTT reduction..........................................................................................62
2.5 Biochemical analyses ..............................................................................62
2.5.1 Alanine aminotransferase ..............................................................63
2.5.2 Aspartate aminotransferase...........................................................64
2.5.3 Creatine kinase..............................................................................64
2.5.4 Lactate dehydrogenase .................................................................65
2.5.5 Aldolase .........................................................................................65
2.5.6 Meso Scale discovery markers ......................................................66
2.6 SELDI-TOF-MS analysis..........................................................................66
2.6.1 ProteinChip® preparation ...............................................................66
2.6.2 ProteinChip® mass spectrometry ...................................................68
2.6.3 SELDI-TOF-MS data pre-processing and analysis ........................69
2.6.4 SELDI-TOF-MS statistical analysis ................................................72
2.7 Label-free quantitative proteomics ...........................................................73
2.7.1 One-dimensional sodium dodecyl sulphate polyacrylamide gel-
electrophoresis .......................................................................................73
2.7.2 In-gel protein digestion ..................................................................73
2.7.3 LC-MS/MS analysis .......................................................................74
2.7.4 LS-MS/MS data analysis................................................................76
2.7.5 Immunoblotting ..............................................................................80
3. Results ......................................................................................................82
3.1 Establishment of the cell culture system ..................................................82
3.2 Effect of the test compounds on skeletal muscle cells .............................87
9
3.2.1 Effect of test compounds on myoblasts .........................................87
3.2.2 Effect on test compounds myotubes..............................................91
3.3 Measurement of candidate biomarkers to discover protein biomarkers of
myopathy ...............................................................................................96
3.3.1 Effect of drug treatment on the leakage of candidate markers.......96
3.3.2 Drug-induced changes in the intracellular levels of candidate
markers...................................................................................................99
3.4 Differential protein expression profiling using SELDI-TOF-MS ..............102
3.4.1 Optimisation of SELDI-TOF-MS protein ion detection settings ....104
3.4.2 Discovery of protein profiles that discriminate between control and
treated muscle cells ..............................................................................106
3.4.3 Selection of TMPD or GWδ responsive protein ions in skeletal
muscle cells ..........................................................................................112
3.4.4 Reproducibility of SELDI-TOF-MS protein ions in muscle cells ...116
3.4.5 Validation of TMPD and GWδ responsive protein ions in skeletal
muscle cells ..........................................................................................121
3.4.6 Selection of responsive protein ion combinations in skeletal
muscle cells ..........................................................................................134
3.4.7 Identification of SELDI-TOF-MS putative biomarkers of myopathy143
3.5 Discovery of protein biomarkers of myopathy using label-free
quantitative proteomics ........................................................................152
3.5.1 1D gel-electrophoresis of rat L6 muscle cells ..............................152
3.5.2 Liquid chromatography-tandem mass spectrometry analysis of rat
muscle cell homogenates .....................................................................156
3.5.2.1 Effect of TMPD-treatment on biological processes in myoblasts
and myotubes....................................................................................176
3.5.3 Biomarker discovery in TMPD-treated skeletal muscle cells using
LC-MS/MS ............................................................................................184
3.5.3.1 Optimisation of Decyder™ MS peptide detection settings......184
3.5.3.2 Comparison of enzyme activity and protein levels.................185
3.5.3.3 Selection of putative biomarkers of TMPD-induced myopathy in
myoblasts ..........................................................................................189
10
3.5.3.4 Selection of putative biomarkers of TMPD-induced myopathy in
myotubes...........................................................................................197
4. Discussion..............................................................................................206
4.1 The use of rat L6 skeletal muscle cells ..................................................207
4.2 The assessment of drug-induced cytotoxicity ........................................209
4.3 The candidate biomarker approach to biomarker discovery ..................211
4.4 SELDI-TOF-MS as a tool for biomarker discovery .................................212
4.5 The label-free approach to biomarker discovery ....................................217
4.6 Summary................................................................................................223
5. Conclusions ...........................................................................................227
References..................................................................................................229
11
List of figures
Figure 1: A haematoxylin and eosin stained section of skeletal muscle........... 30
Figure 2: The differentiation of myoblasts into myotubes................................. 33
Figure 3: Strategies for MS-based protein biomarker discovery ...................... 39
Figure 4: Schematic of the basic components of a mass spectrometer ........... 43
Figure 5: Matrix-assisted laser desorption/ionization ....................................... 44
Figure 6: Electrospray ionisation...................................................................... 46
Figure 7: Time of flight mass spectrometry ...................................................... 47
Figure 8: The path of ions through a quadrupole ............................................. 48
Figure 9: Hybrid quadrupole-time of flight mass spectrometer......................... 49
Figure 10: Schematic of an ion trap ................................................................. 50
Figure 11: Schematic of SELDI-TOF-MS......................................................... 52
Figure 12: A SELDI-TOF-MS ProteinChip® array............................................. 66
Figure 13: Cluster generation using the Biomarker Wizard in Ciphergen Peaks
software .................................................................................................... 70
Figure 14: Biomarker Wizard detection settings............................................... 71
Figure 15: Noise calculation in Ciphergen Peaks software .............................. 72
Figure 16: Sectioning of one-dimensional electrophoresis gels ....................... 75
Figure 17: Schematic of quantitative Decyder™ MS differential peptide analysis
.................................................................................................................. 79
Figure 18: Set up of PVDF membrane/gel sandwich for protein transfer ......... 80
Figure 19: Establishment of the rat L6 myblast cell culture system.................. 83
Figure 20: Morphology of rat L6 myotubes treated with vehicle control alone . 85
Figure 21: Baseline biochemical levels in skeletal muscle cells....................... 86
Figure 22: Effect of a range of concentrations of TMPD on rat L6 myoblasts
following 24 hours of TMPD treatment...................................................... 88
12
Figure 23: Degenerative ultrastructural changes in myoblasts caused by TMPD
treatment................................................................................................... 89
Figure 24: Effect of a range of concentrations of TMPD and GWδ on myoblasts
.................................................................................................................. 90
Figure 25: Degenerative ultrastructural changes in myotubes ......................... 92
Figure 26: Effect of a range of concentrations of TMPD on rat L6 myoblasts
following 24 hours of TMPD treatment...................................................... 93
Figure 27: Effect of a range of concentrations of TMPD and GWδ on myotubes
.................................................................................................................. 94
Figure 28: TMPD-induced changes in leakage into myoblast conditioned
medium..................................................................................................... 97
Figure 29: TMPD-induced changes in leakage into myotube conditioned
medium..................................................................................................... 98
Figure 30: TMPD-induced changes in intracellular enzymes in myoblasts .... 100
Figure 31: TMPD-induced changes in intracellular enzymes in myotubes..... 101
Figure 32: Representative SELDI-TOF-MS spectra of skeletal muscle cells from
different ProteinChip® arrays .................................................................. 103
Figure 33: Influence of Biomarker Wizard detection settings on protein ion
detection ................................................................................................. 105
Figure 34: Peaks detected in SELDI-TOF-MS spectra using selected Biomarker
Wizard detection settings........................................................................ 105
Figure 35: Principal components analysis comparing control and treated
myoblasts following SELDI-TOF-MS analysis......................................... 107
Figure 36: Principal components analysis comparing control and treated
myotubes following SELDI-TOF-MS analysis ......................................... 108
Figure 37: Partial least squares discriminant analysis (PLS-DA) comparing
control and treated myoblasts................................................................. 110
13
Figure 38: Partial least squares discriminant analysis (PLS-DA) comparing
control and treated myotubes ................................................................. 111
Figure 39: Intra and inter-experiment reproducibility of SELDI-TOF-MS protein
ions in myoblasts .................................................................................... 117
Figure 40: Intra and inter-experimental reproducibility of SELDI-TOF-MS protein
ions in myotubes..................................................................................... 119
Figure 41: Responsive SELDI-TOF-MS protein ions in myoblasts................. 123
Figure 42: Responsive SELDI-TOF-MS protein ions in myotubes ................. 126
Figure 43: Levels of TMPD-responsive protein ions in myoblasts.................. 128
Figure 44: Levels of GWδ-responsive protein ions in myoblasts.................... 129
Figure 45: Representative SELDI-TOF-MS spectra showing putative biomarkers
in TMPD or GWδ-treated myoblasts ....................................................... 130
Figure 46: Levels of TMPD-responsive protein ions in myotubes .................. 131
Figure 47: Levels of GWδ-responsive protein ions in myotubes .................... 132
Figure 48: Representative SELDI-TOF-MS spectra showing putative biomarkers
of myopathy in myotubes........................................................................ 133
Figure 49: Two-dimensional scatterplots of TMPD and GWδ responsive protein
ions selected by regression analysis in myoblasts.................................. 139
Figure 50: Two-dimensional scatterplots of TMPD and GWδ responsive protein
ions selected by regression analysis in myotubes .................................. 140
Figure 51: Effect of TMPD or GWδ on rat L6 myoblasts ................................ 142
Figure 52: Effect of TMPD or GWδ on rat L6 myotubes................................. 143
Figure 53: Fractionation of 1DE gels.............................................................. 153
Figure 54: 1D gel of rat L6 myoblasts treated with vehicle control or 25µM
TMPD ..................................................................................................... 154
Figure 55: 1D gel of rat L6 myotubes treated with vehicle control or 25µM TMPD
................................................................................................................ 155
14
Figure 56: Number of proteins identified in myoblasts and myotubes............ 157
Figure 57: Normal biological processes in rat L6 myoblasts .......................... 177
Figure 58: Normal biological processes in rat L6 myotubes........................... 178
Figure 59: Comparison of the normal biological process in myoblasts and
myotubes ................................................................................................ 180
Figure 60: Optimisation of detection settings in the PepDetect module of
Decyder™ MS software ........................................................................... 185
Figure 61: Relative levels of peptides used to quantify Rho GDP dissociation
factor Inhibitor (GDI) alpha...................................................................... 193
Figure 62: Example spectrum obtained following the MS/MS analysis of a
peptide from Rho GDI............................................................................. 194
Figure 63: TMPD-induced changes in the levels of Rho GDI in myoblasts.... 195
Figure 64: Rho GDI levels in myoblasts treated with GWδ or pravastatin...... 196
Figure 65: Relative levels of peptides used to quantify myosin light chain
polypeptide 2 in myotubes ...................................................................... 201
Figure 66: Example spectrum obtained following the MS/MS analysis of a
peptide from Myl2 ................................................................................... 202
Figure 67: TMPD-induced changes in the levels of myosin light chain 2 in
myotubes ................................................................................................ 203
Figure 68: Myosin light chain 2 levels in myotubes following treatment with GWδ
or Pravastatin.......................................................................................... 204
Figure 69: Summary of protein changes in myoblasts and myotubes............ 226
15
List of tables
Table 1: Binding and washing buffers used for ProteinChip® preparation........ 68
Table 2: Ion trap mass spectrometer settings used for the analysis of skeletal
muscle cells .............................................................................................. 76
Table 3: Detection settings used in Decyder™ MS software ............................ 77
Table 4: Number of responsive proteins ion discovered in myoblasts ........... 113
Table 5: Number of responsive proteins ion discovered in myotubes ............ 114
Table 6: Summary of protein expression changes in myoblasts .................... 122
Table 7: Summary of protein expression changes in myotubes..................... 125
Table 8: Summary of forward stepwise regression of responsive protein ions in
control and treated myoblasts................................................................. 136
Table 9: Summary of forward stepwise regression of responsive protein ions in
control and treated myotubes ................................................................. 137
Table 10: Identification of SELDI-TOF-MS biomarkers of myopathy in myoblasts
(TMPD responders) ................................................................................ 144
Table 11: Identification of SELDI-TOF-MS putative biomarkers of myopathy in
myoblasts (GWδ responders).................................................................. 145
Table 12: Identification of SELDI-TOF-MS putative biomarkers of myopathy in
myoblasts (common responders)............................................................ 146
Table 13: Identification of SELDI-TOF-MS putative biomarkers of myopathy in
myotubes (TMPD responders)................................................................ 148
Table 14: Identification of SELDI-TOF-MS putative biomarkers of myopathy in
myotubes (GWδ responders) .................................................................. 149
Table 15: Identification of SELDI-TOF-MS putative biomarkers of myopathy in
myotubes (common responders) ............................................................ 150
Table 16: Proteins identified in rat L6 myoblasts............................................ 158
Table 17: Proteins identified in rat L6 myotubes ............................................ 166
16
Table 18: Biological processes affected in TMPD-treated myoblasts ............ 182
Table 19: Biological processes affected in TMPD-treated myotubes ............. 183
Table 20: Comparison of biochemical levels with protein levels in myoblasts 187
Table 21: Comparison of biochemical levels with protein levels in myotubes 188
Table 22: Identified myoblasts cell proteins that showed differential expression
following treatment with TMPD ............................................................... 191
Table 23: Identified myotubes cell proteins that showed differential expression
following treatment with TMPD ............................................................... 198
Chapter 1: Introduction
Abbreviations
1DE One dimensional gel electrophoresis
2DE Two dimensional gel electrophoresis
3D Three dimensional
21CFR Title 21 of the code of federal regulations
AA Amino acid
ADME Absorption distribution metabolism excretion
ADP Adenosine diphosphate
ADR Adverse drug reaction
ALD Aldolase
ALT Alanine aminotransferase
AMU Atomic mass unit
AST Aspartate aminotransferase
ATP Adenosine triphosphate
BCA Bicinchoninic acid
BH Benjamini-Hochberg
BMW Biomarker Wizard
Ca2+ Calcium ion
CAD Cationic amphiphilic drug
CHAPS 3-[(3-Cholamidopropyl)dimethylammonio]-1-propane sulfonate
CHCA Alpha-cyano-4-hydroxycinnamic acid
CK Creatine kinase
CM10 Weak cation exchange (ProteinChip® array)
CO2 Carbon dioxide
CoA Coenzyme A
CRM Charged residue model
CV Coefficient of variation
Da Dalton
DAP Dihydroxyacetone phosphate
DMD Duchenne muscular dystrophy
DMEM Dulbecco's minimum essential medium
DMSO Dimethyl sulfoxide
18
DNA Deoxyribonucleic acid
D-PBS Dulbecco's phosphate buffered saline
DTT Dithiothreitol
EAM Energy-absorbing molecule
EDTA Ethylenediaminetetraacetic acid
ELISA Enzyme-linked immunosorbent assay
EM Electron microscopy
ESI Electrospray ionisation
ExPASy Expert protein analysis system
FABP3 Fatty acid binding protein 3
FBS Foetal bovine serum
FDA Food and drug administration
FDR False discovery rate
FTP File transfer protocol
GAP Glyceraldehyde-3-phosphate
GCP Good clinical practice
GDH glycerolphosphate dehydrogenase
GDP Guanosine diphosphate
GOT Glutamic oxaloacetic transaminase
GTP Guanosine triphosphate
GWδ 4-[2-(3-Fluoro-4-trifluoromethyl-phenyl)-4-methyl-thiazol-5-
ylmethylsulfanyl]-2-methyl-phenoxy}-acetic
H2O Water
H50 Hydrophobic (ProteinChip® array)
HMG 3-Hydroxy-3-methylglutarate
HPLC High performance liquid chromatography
HRP Horseradish peroxidase
ICAT Isotope-coded affinity tags
IEM Ion evaporation model
IgG Immunoglobulin G
IMAC 30 Immobilised metal affinity capture (ProteinChip® array)
IND Investigative new drug
IPG Immobilized pH gradient
19
IRB Institutional review board
ITRAQ Isobaric tag for relative and absolute quantitation
LC Liquid chromatography
LDH Lactate dehydrogenase
LFQP Label-free quantitative proteomics
LM Light microscopy
LTQ Linear trap quadrupole
m/z mass:charge
MALDI Matrix-assisted laser desorption/ionization
MDH Malate dehydrogenase
MEF Myocyte enhancing factor
MES 2-(N-morpholino) ethanesulphonic acid
Mg2+ Magnesium ion
Min Minute
MRF Myogenic regulatory factor
MRM Multiple reaction monitoring
mRNA Messenger ribonucleic acid
MS Mass spectrometry
MS/MS Tandem mass spectrometry
MTT 3-(4,5-Dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide
Mw Molecular weight
Myl2 Myosin light chain 2
Myl3 Myosin light chain 3
NAD+ Nicotinamide adenine dinucleotide
NADP+ Nicotinamide adenine dinucleotide phosphate
NCBI National center for biotechnology information
NCE Novel chemical entity
NMR Nuclear magnetic resonance
NP20 Normal phase (ProteinChip® array)
NSAID Non-steroidal anti-inflammatory drug
PAGE Polyacrylamide gel electrophoresis
PBS Phosphate buffered saline
PBS IIc Protein biology system IIc
20
PCA Principal component analysis
PLS-DA Partial least squares discriminant analysis
Pos Position
PPAR Peroxisome proliferator-activated receptor
PVDF Polyvinylidene fluoride
Q-PCR Quantitative polymerase chain reaction
QQQ Triple quadrupole
R&D Research and development
RefSeq Reference sequence
RER Rough Endoplasmic Reticulum
Rev pos Reverse position
Rho GDI Rho GDP dissociation inhibitor (GDI) alpha
S/N Signal:Noise
SDS Sodium dodecyl sulphate
SELDI Surface-enhanced laser desorption/ionization
SEM Standard error of the mean
SILAC Stable isotope labelling with amino acids in cell culture
SIM Selected ion monitoring
SKDM Skeletal muscle differentiation medium
SKGM Skeletal muscle growth medium
SRF Search results file
sTnI Skeletal troponin I
TFA Trifluoroacetic acid
TIC Total ion current
TIM Triosphosphate isomerase
TMPD 2,3,5,6 Tetramethyl-p-phenylenediamine
TOF Time of flight
VC Vehicle control
VIP Variable importance of projection
Xcorr Cross-correlation
21
Amino acid abbreviations
Amino acid Single letter code
Alanine AArginine RAspargine NAspartic acid DCysteine CGlutamic acid EGlutamine QGlycine GHistidine HIsoleucine ILeucine LLysine KMethionine MPhenylalanine FProline PSerine SThreonine TTryptophan WTyrosine YValine V
22
Chapter 1
1. Introduction
1.1 Pharmaceutical safety assessment
The process of taking a compound from synthesis in the laboratory to the
treatment of patients in the clinic, otherwise known as drug development, is a
lengthy, complex and often ethically-questioned process, which takes place at
huge costs to research and development (R&D) companies1-6. Despite the fact
that advances in high-throughput technologies (e.g. combinatorial chemistry)
have facilitated the identification of a large number of chemical targets in the
early stages of drug development, the number of new compounds making it
through to successful product launch has reduced significantly over the past few
years, whilst the costs of research have increased7-13. These costs are
exacerbated by the fact that only a small percentage of drugs initially evaluated
make it through the approval process, due to unexpected findings in the later
stages of drug development.
Late stage drug attrition is combined with the fact that only a small proportion of
drugs that complete the regulatory process make sufficient profits to recover the
huge investments made by research companies. This could lead to sponsors
investing in drugs that are not necessarily in the therapeutic areas of greatest
need but that have the highest likelihood of being profitable14-16. As such, the
rigorous safety assessment of new drugs is paramount, as adverse drug
reactions (ADRs) are most frequently to blame for the failure of new compounds
to be registered for clinical use. In addition to the duty of drug manufacturers to
the protection of potential patients, research companies must also protect
themselves (and the industry) against negative public perception and possible
litigation arising from unexpected ADRs when new compounds are licensed for
general use17-21.
23
Drug development is highly regulated by authorities such as the Food and Drug
Administration (FDA) and guided by title 21 of the code of federal regulations
(also known as 21CFR), which ensures that qualified investigators produce
appropriate regulatory data and documents in a manner compliant with good
clinical practice (GCP). In combination, scientific investigators, the FDA and
institutional review boards (IRBs) certify the credibility of research work
contributing to the registration of a novel chemical entity (NCE). Together, they
provide the public with a measure of confidence in the science behind drugs
available for clinical therapeutics22-24. Before an NCE is approved by the FDA,
pharmacology and toxicology data is generated through a series of preclinical
and clinical experiments in order to identify the most appropriate chemical
targets, and also to test the effects of potential therapeutic drugs. Information on
the efficacy of a molecule must be considered whilst, at the same time, ensuring
that there is a sufficient therapeutic index (the ratio between the toxic dose and
the therapeutic dose) so that the drug is most likely to be safe in humans during
clinical trials and also post-marketing.
The safety assessment of new compounds is typically conducted using a
combination of disciplines looking at the chemistry of the molecule during
candidate selection and the pharmacological interactions between the molecule
and its intended or unintended targets. Complications that may arise from the
bioavailability and biodistribution of the drug are investigated in absorption,
distribution, metabolism and excretion (ADME) studies. Toxicological drug
effects are usually explored preclinically through carcinogenicity, teratogenicity
and toxicity studies and the aim of these investigations is to select molecules
with the lowest probability of being associated with adverse events. Following
the submission of data from preclinical studies, a successful investigational new
drug (IND) application to the FDA allows drugs to enter the clinical phases of
investigation.
24
In Phase I clinical trials, the emphasis is on testing the safety of novel chemicals
in humans before Phase II trials explore the efficacy of compounds for a specific
condition. Once these aspects of the drug have been examined, a larger
population of patients is included in Phase III studies so that the effects of the
drug on a wider, more varied set of patients can be investigated and more
information can be gained which may contribute to advertising or product
labelling. In these studies, the assessment of any other risk factors such as age,
gender or environment may also be undertaken. In some cases, a further Phase
(IV) may be included voluntarily following FDA approval, or at the request of the
FDA, in which post-marketing trials are conducted in order to provide additional
data on the drug on an even larger cohort of patients. This process ensures that
reasonable steps are taken by drug developers to ensure the safety of
patients25-32.
1.2 Adverse drug reactions
ADRs are those that may require intervention or an interruption in treatment in
order to ensure the safety of the test species under investigation, and are the
most common reason for drug attrition during the drug development process.
Despite most clinical trials being completed without major findings, due to the
comprehensive portfolio of preclinical work performed before drugs are tested in
man, the focus understandably tends to be on the low incidence of ADRs
occurring in clinical studies or when drugs are released into the general
population. This is demonstrated in the high profile case of Rofecoxib (a non-
steroidal anti-inflammatory drug (NSAID), approved in 1999 as Vioxx® for the
treatment of pain), which was withdrawn from the market in 2004 after being
associated with an increased risk of myocardial infarction33-36. A similar story is
true of the lipid-lowering compound cerivastatin (an HMG CoA reductase
inhibitor (statin)), which was withdrawn from the market in 2001 following a
number of fatal events associated with skeletal muscle toxicity in patients
exposed to the drug. Although the effects were proven not to be specific to this
class of drugs, there were considerable consequences on the reputation and
profits of the manufacturer. There were also knock-on effects on patient
25
compliance with this class of compound, due to the fear of possible harmful
effects associated with long-term medication37,38.
Despite the rare situations where preclinical safety testing of a novel compound
has failed to predict adverse effects observed post-marketing, there are many
successes. In fact, statins and NSAIDs are generally considered to have a good
safety profile, as indicated by the number of drugs in these classes that are
currently prescribed clinically for the treatment of obesity or inflammation,
respectively. Current strategies for dealing with possible adverse effects include
the close monitoring of the status of skeletal muscle (for patients on lipid-
lowering drugs) or for gastrointestinal effects sometimes associated with
NSAIDs. Additionally, improvements in product labelling are sometimes
requested in order to guard against complications that may arise from the drug
being administered to more susceptible sections of the population. Improved
methods of monitoring adverse events, combined with an improved
understanding of the mechanisms behind them, would allow investigators to
monitor the effects of new molecules better during drug development39-44.
ADRs are manifested in a number of ways and the manner in which preclinical
safety assessment departments are organised within pharmaceutical
companies reflects this. Reproductive toxicology studies, for example, examine
whether drug treatment to a pregnant mother could result in birth defects,
malformations or even the death of an embryo. The need for such routine
testing was established following the tragedy of Thalidomide which was
marketed in 1957 as an anti-emetic but withdrawn from the market 4 years later
after it was associated with severe birth defects in newborns45. Another
important part of safety assessment is the screening for damage to
deoxyribonucleic acid (DNA) which can result in a genetic mutation or, in some
cases, uncontrolled cell replication and subsequent carcinogenesis. The
potential of compounds to cause mutations in DNA is usually screened using
tests such as the Ames test or the comet assay46-49.
26
1.2.1 Target organ toxicity
ADRs are often considered in terms of their target or off-target effects. So called
“recurrent toxicity” issues are those that are characteristic of certain classes or
types of compounds and that may occur in studies involving these compounds.
Although the effects seen with some of these compounds may not be totally
unexpected, in most cases, the exact mechanism by which toxicity is caused
may be unknown or may be complicated. This leaves chemists to either attempt
to design out unwanted features of a molecule, whilst retaining its efficacy, or to
consider using lower doses of the compound for an alternative indication.
The direct effect on a specific susceptible organ may also be due to several
other factors, including the position of the organ. This is true of the gastro-
intestinal tract with which a drug may have an increased amount of contact,
when taken orally50,51. Other drugs may have a chemical structure that is
associated with a particular toxicity and this is demonstrated in the case of
cationic amphiphilic drugs (CADs) with which an abnormal accumulation of
phospholipids can occur within cells of the target organ. Due to the high levels
of phospholipids contained within surfactant in the lungs, this is often a target
organ (in the form of lung-congestion), as seen with amiodarone (an anti-
arrhythmic CAD-like drug for the treatment of irregular heart beat)52-54. The liver
is often a site for toxic effects due to its detoxification role, and hepatotoxicity is
most commonly the complication associated with drug attrition55-57.
1.2.2 Skeletal muscle toxicity
Skeletal muscle disorders can occur due to a wide range of factors including the
misuse of muscles, congenital disease, inflammatory, metabolic or endocrine
disorders. Symptoms vary with the degree of myopathy ranging from mild
muscle pain (myalgia), cramp or weakness to much more serious effects, such
as rhabdomyolysis. This serious, and sometimes fatal condition, is
characterised by the breakdown of skeletal muscle tissue and the release of
proteins found in high concentration in skeletal muscle (such as myoglobin) into
the circulation. These proteins migrate to the kidney where they can result in
27
renal failure due to the harmful effects of these molecules to structures in the
kidney. Once detected, there are numerous procedures for the treatment of
skeletal muscle related disorders, such as resting the affected muscle groups,
more contemporary treatments such as intramuscular stimulation or even
surgery58-62.
As well as the causes of myopathy already described, myopathy can also be
drug-induced and usually results in an interruption of the nervous stimuli or a
direct effect on the muscle fibre such as trauma or ischemia, which results in
necrosis or apoptosis and subsequent cell death. Drug-induced myopathies are
perhaps one of the more preventable types of skeletal muscle injury and are a
concern to pharmaceutical companies, researchers and medical professionals
endeavouring to understand the underlying causes of these effects, which are
most frequently associated with drugs in the lipid-lowering class of compounds.
When used carefully, lipid-lowering drugs can significantly reduce the risk of
myocardial infarction due to reduced levels of circulating lipids, altered
thrombogenesis and beneficial effects on inflammatory responses. The effects
of these compounds are mainly considered to be beneficial but there are some
concerns regarding incidences of skeletal muscle toxicity seen in clinical studies
involving some compounds in this class. Although the clinical incidence of
myopathy is relatively low, the increasing number of patients being treated
chronically for hypercholesterolemia with drugs such as the aforementioned
statins and peroxisome proliferator-activated receptor (PPAR) agonists mean
that the associated myopathy is a cause for concern. Even more worrying is the
fact that combination therapy is common with such compounds and often leads
to an increased risk of myopathy and a dose-related increase in the incidence of
the more serious effects on skeletal muscle already described63-65.
28
PPAR agonists are a large class of nuclear receptor proteins that act as
transcription factors to modulate the expression of genes involved in lipid
metabolism via the activation of specific receptors. These receptors (α, δ and γ)
have different roles in the regulation of pathways of mammalian metabolism
which include fatty acid oxidation and lipogenesis. The diverse physiological
functions of these receptors are based partly on their different tissue
distribution. PPAR-α is highly expressed in liver, heart, intestine and proximal
tubule cells of the kidney, PPAR-δ is expressed ubiquitously (but most
abundantly in skeletal muscle) and PPAR-γ is expressed predominantly in
adipose tissue and in the immune system. PPAR-α promotes fatty acid
oxidation under conditions of lipid catabolism such as fasting, while PPAR-γ
promotes lipogenesis in adipose tissue. Less is known about the function of
PPAR-δ, although this receptor is believed to alter the circulating lipid profile,
lipid handling in the macrophages and skeletal muscle metabolism66-69.
Peroxisome proliferators have been shown to cause an increase in the number
and size of peroxisomes in the liver, kidney and heart in rodents and are
believed to induce hepatocarcinogenesis in these species following long-term
exposure, although there is less evidence for this clinically. As muscle is a
major site for PPAR expression (especially PPAR-δ), another concern with this
class of compounds is skeletal muscle toxicity seen with these compounds and
a concerted effort is being made to understand the nature of the skeletal muscle
changes associated with PPAR agonists. The fact that a significant amount of
research is directed towards the use of these compounds as high-affinity
specific PPAR receptor agonists for the treatment of hyperglycaemia,
hyperlipidaemia and other metabolic diseases suggests that it is important to
understand whether the muscle fibre degeneration associated with these
compounds is due to altered lipid utilisation, muscle type switching or direct
toxicity70-78.
29
Statins inhibit the action of 3-hydroxy-3-methylglutamyl-coenzyme-A (HMG
CoA) reductase during a rate-limiting step in cholesterol biosynthesis. As a
result, circulating cholesterol levels are decreased leading to an improved
circulating lipid profile in patients. The resulting improvement in the prognosis of
patients with regards to reduced risk of coronary heart disease and stroke is,
however, overshadowed by the risk of muscle disorders and much work tends
to be focused on monitoring the adverse effects of treatment rather than the
beneficial effects79-85.
Phenylenediamines, such as 2,3,5,6 tetramethyl-p-phenylenediamine (TMPD;
also known as Würster’s blue), are a class of compounds that specifically
induce skeletal muscle toxicity through the generation of unusually unstable
radical cations (Wϋrster salts), superoxide and hydroxyl radicals, and hydrogen
peroxide. Free radicals, such as those described, are normally prevented from
doing damage by an endogenous antioxidant system (which includes
glutathione, glutathione reductase and superoxide dismutase). Cytotoxicity
usually occurs as a result of glutathione depletion and the formation of mixed
disulphides, eventually causing damage to cells through oxidative stress due to
the rate of consumption of these protective molecules exceeding their
synthesis. TMPD has been used in toxicology experiments in vivo and in vitro
as an experimental myotoxin due to the predictable and specific nature of the
toxicity induced in skeletal muscle86-90.
30
1.2.3 Assessment of myopathy
Besides the direct visualisation of cells or tissue by light or electron microscopy,
there are few reliable methods of assessing skeletal muscle toxicity in vivo
(Figure 1). Many routinely used markers for the assessment of skeletal muscle
toxicity are not ideally suited for use in a clinical setting because they are
usually non-specific to the site of injury, insensitive to small changes in the
status of skeletal muscle or their measurement requires a relatively invasive
sample collection procedure. The fact that such measurements have limited
value in the assessment of the early stages of myopathy presents a problem for
researchers evaluating the safety of NCEs. The safety of patients involved in
clinical trials is of paramount importance and, as such, more useful biomarkers
are required that may provide additional information on the mechanisms by
which myotoxicity is caused.
Figure 1: A haematoxylin and eosin stained section of skeletal muscle The effects of drug-treatment are commonly assessed in toxicology experiments by
histopathological examination of tissue sections. A transverse section of skeletal
muscle tissue taken from the upper hind limb of a female rat treated with 40μmol/kg
TMPD for 7 days is shown. Treatment with the experimental myotoxin caused single
cell necrosis in affected animals (necrotic fibres indicated by the arrows). Magnification
x40.
31
Effects on skeletal muscle are routinely evaluated in toxicology studies by
measuring the circulating levels of enzymes known to be in high concentration
in skeletal muscle tissue, such as aspartate aminotransferase (AST), creatine
kinase (CK), lactate dehydrogenase (LDH) and aldolase (ALD). AST is a mainly
mitochondrial enzyme, although it is also found in the cytosol of cells. Its wide
distribution in various organs (especially the heart) means that the origins of an
increase in AST usually have to be confirmed using other biochemical markers.
LDH is an enzyme found in a variety of tissues, including liver, heart and
skeletal muscle. The enzyme is tetrameric and is composed of four subunits of
two molecules. Various combinations of these molecules result in 5 different
isoenzymes (LDH1-LDH5), and LDH5 is the isoenzyme found in the highest
concentrations in skeletal muscle. ALD is a cytosolic enzyme that is found at
particularly high levels in skeletal muscle tissue and released into the circulation
following injury to skeletal muscle. However, like the other markers discussed, it
is not specific to skeletal muscle, and is also present in other tissues, such as
kidney, liver and intestine91-93.
CK is the most commonly used, and perhaps most useful, routinely used
marker of muscle injury and is a dimeric enzyme made up of two subunits (M
and B). The three isoforms; CK-MB, CK-BB and CK-MM are in highest
concentration in cardiac muscle, brain and skeletal muscle, respectively92,93.
The measurement of CK in toxicology studies can be problematic because it
has a short half-life in the circulation and is believed to degrade rapidly in the
medium when enzyme leakage from cells is measured during in vitro
experiments. In addition, this marker lacks specificity for skeletal muscle unless
the specific isoenzymes previously mentioned are measured88. More recently,
reportedly more specific markers of skeletal muscle toxicity such as
parvalbumin, skeletal troponin I (sTnI), myosin light chain 3 (Myl3) and fatty acid
binding protein 3 (FABP3) have been proposed, but many of these are yet to be
validated and established as useful biomarkers of myopathy. Issues such as the
sensitivity and specificity of these markers to low levels of injury to skeletal
muscle have yet to be fully assessed91,94-100.
32
1.2.4 Models used in toxicology
Preclinical animal studies are a legal requirement in order to demonstrate the
safety and efficacy of new compounds prior to testing in man. However, there is
a desire to replace current methods with more humane methods (where
possible), to refine the methods used in order to minimise the suffering of study
animals and also to reduce the number of animals used in studies (also known
as the 3Rs). The aim is to find alternatives to using animals where possible and,
whilst it is generally accepted that they cannot completely negate the
requirement for animal testing in drug development, ex vivo or in vitro models
are often utilised in an attempt to model certain aspects of toxicology seen in
vivo101-106. The rat L6 skeletal muscle cell line (originally isolated from primary
cultures of Rattus norvegicus thigh muscle) is often used in vitro to model the
effects of compounds on skeletal muscle tissue107-112.
Myoblasts are immature, undifferentiated cells that retain the ability in vitro to
divide and differentiate into more specialised skeletal muscle cells.
Differentiation into skeletal muscle cells (myogenesis) is regulated by a family of
myogenic regulatory factors (MRFs) that activate the expression of specific
skeletal muscle genes. Transcription factors such as MyoD and Myf5 are
expressed early in development before myogenin stimulates the fusion of
myoblasts into myotubes. MRF4 is highly expressed in myofibres when cells
withdraw from the cell cycle. The conversion of myoblasts to myotubes
correlates with the in vivo situation when repair follows damage to skeletal
muscle tissue. Satellite cells normally remain quiescent amid other cells, but
play a role in repair by proliferating and fusing together during the differentiation
process to form multinucleated myotubes, thus replacing or repairing tissue lost
through disease or injury (Figure 2) 113-117.
33
Rat L6 myoblasts myotubes Myofibre
Figure 2: The differentiation of myoblasts into myotubes Rat L6 myoblasts retain the ability to proliferate in vitro and can be differentiated into
myotubes when cell culture conditions are modified. Differentiation of mononuclear
myoblasts into myotubes is confirmed by the elongation of the cells due to the fusion of
two or more myoblasts into one multinucleated cell. This in vitro system correlates with
in vivo muscle, in that satellite cells are normally present in a quiescent state, but are
activated when damage is done to skeletal muscle tissue. The subsequent proliferation
of myoblasts and their fusion into myotubes facilitates the repair of damaged skeletal
muscle tissue.
1.3 Tools used in biomarker discovery
Current guidelines for safety testing involve the use of well-established
disciplines such as histopathology and clinical chemistry which, although
generally considered fit for purpose, sometimes lack the predictivity and
sensitivity required for the safety assessment of NCEs118-121. Biological samples
such as urine, blood or more complex tissues can be analysed using relatively
novel techniques compared to those applied routinely in safety assessment
screening. The data obtained can then be interpreted using a variety of
statistical methods in order to correlate the condition of a particular patient with
specific changes or with a pattern of changes.
The need to find better ways to monitor or assess drug effects, and therefore
streamline drug development and reduce the rate of attrition of much needed
drug therapies, has been recognised by the FDA in their Critical Path Initiative.
This initiative recognises the slowdown in the registration of new drugs, and the
case has been made for R&D companies to invest in and utilise new
technologies and biomarkers when testing the safety and efficacy of new
34
compounds. At the forefront of the technologies employed in biomarker
discovery are genomics, metabolic profiling and proteomics; broadly classified
as toxicogenomics122-124.
Biomarker discovery can be broadly classified into two main areas, “closed” or
“open” profiling. Measurements can be performed in a directed or closed
manner in which known molecules are specifically measured in order to
correlate the levels of a particular parameter or set of endpoints with a clinical
observation or range of observations. This approach is often applied when there
is some prior knowledge of the effects or targets of the compound. In addition,
there are usually some measures already available with which decisions on the
prognosis of a patient or the direction of drug development can be correlated
with.
An open-screening approach involves the comparison of a population of
unaffected or control subjects or cells with an identical population that have
been treated with relevant doses or concentrations of a test compound.
Molecules or patterns of molecules that may be indicative of an effect resulting
from the administration of the compound can then be discovered. Differences
observed in the levels of putative biomarkers between the different populations
that correlate well with treatment are likely to be related to treatment. Following
validation, these biomarkers may prove useful in preclinical studies and may
also translate to the clinic125-127.
Efforts are currently being made to improve the tools or markers used in safety
assessment so that they are not simply descriptive of the effects of drugs, but
are able to reliably, and reproducibly, predict these effects. Biomarkers should
be specific to the target, sensitive to low levels of injury and should also reflect
reversibility when a drug has been removed and where recovery is evident.
Biomarkers should also be easily measured and, preferably, quantifiable in
samples that do not require invasive sampling of the patient. A rapid change in
its circulatory levels after an adverse effect and a long half-life in the circulation
35
are also advantageous so that the window of opportunity in which it can be
measured is sufficiently long to prevent important changes being missed. The
bridging of preclinical biomarkers to clinical studies is also a desirable feature
so that the effects of new compounds in the clinic can be reliably predicted in
preclinical studies128,129.
Biomarkers can also be classified into several types. For example, biomarkers
of toxicity can be used to indicate an existing or potential effect of a drug, whilst
disease biomarkers are reflective of the presence or absence of disease.
Mechanistic biomarkers provide information on the way in which the effect of an
external stimulus may have arisen. A biomarker can be any one of these types
or a combination of more than one type but, ultimately, surrogate endpoints are
desirable that, as previously stated, can act as a substitute for a clinical
endpoint130,131.
1.3.1 Metabolic profiling
Metabolic profiling studies examine changes in the expression of small
molecules (typically <1000Da) in response to stimuli such as drugs, disease or
the environment. It is one of a number of technical or disciplinary approaches to
the discovery of biomarkers and, in its application, nuclear magnetic resonance
(NMR) spectroscopy is often supplemented by liquid chromatography-mass
spectrometry (LC-MS). NMR or LC-MS spectra are usually considered in
combination with genomic and proteomic data in order to fully understand the
changes observed and discover metabolite profiles that may be characteristic of
perturbations to normal biochemical pathways. In toxicology studies, data
generated following the analysis of samples from control and treated subjects is
compared in order to discover differentially expressed metabolites that may be
indicative of a drug effect132-138.
36
Duchenne muscular dystrophy (DMD) is caused by a mutation in the gene that
codes for dystrophin, which results in an inability to express the dystrophin
protein. This condition is characterised by muscle degeneration and
subsequently a reduced life expectancy. Diagnosis of this congenital condition
can be performed in prenatal tests or by examining DNA but it is usually
conducted by taking a muscle biopsy and performing specific
immunohistochemical staining for dystrophin. Specific profiles for DMD have
been revealed in dystrophic cardiac muscle, skeletal muscle, cerebral cortex
and cerebellum by Griffin et al139 using NMR profiling. Previous to this study,
taurine had also been proposed as a biomarker of DMD by McIntosh et al140
and these studies demonstrate the utility of metabolic profiling as a tool for
biomarker discovery.
Metabolic profiling is often conducted using urine samples due to the value of
this sample as a repository for metabolites, and the non-invasive collection of
urine samples for metabolic profiling studies is a feature of this discipline. A
large amount of useful data is usually obtained from these studies and it
provides an extensive picture of the effects of drugs on an organism. However,
the equipment used for these investigations is extremely expensive and
requires extensive laboratory resources. As with most biomarker discovery
tools, skilled scientists are required to operate the equipment and also to
analyse the complex data produced141,142.
1.3.2 Transcriptomics
Genes are often differentially expressed in response to different stimuli and
transcriptomic profiling is the study of changes in relative messenger
ribonucleic acid (mRNA) levels in response to these stimuli. Such changes can
be linked to a possible outcome of drug treatment. The work of Hirata et al143
demonstrates the value of this approach to biomarker discovery and, in this
particular study, osteopontin was proposed as a biomarker of muscle
regeneration in mice following cardiotoxin treatment.
37
Transcriptomic profiling is a sensitive technique that is capable of detecting
early changes that may have been missed using other techniques. This
discipline is perhaps more suited to the discovery of mechanistic biomarkers as
changes seen tend to reflect the response of the organism to the stimulus
rather than the ultimate effect. This can be a problem though as specific
changes in mRNA expression may not be translated into the functional protein
product partly due to the possibility of alternative splicing of genes or post-
translational modifications to proteins. In addition, the technique is often
conducted using expensive DNA microarrays (e.g. Affymetrix® GeneChip
microarrays or Illumina BeadChip™ arrays) or quantitative polymerase chain
reaction (Q-PCR) reagents, and this can be restrictive when conducting large
studies144-148.
1.4 Proteomics
Proteomics is a powerful analytical tool that can be used to measure and
characterise proteins in a wide variety of biological samples. It can be used to
elucidate mechanisms of toxicity and to discover novel and, hopefully, predictive
markers of disease, toxicity or efficacy149,150. Biomarkers discovered using
proteomics are more likely to be useful in their application due to the fact that
the functional protein product is measured directly and not usually subject to
any structural modifications that may occur when, for instance, genes are
translated145,151-153. Alternative splicing of genes, post-translational modifications
or protein degradation can occur during the complex processing of genes to
active proteins. Therefore measuring the functional protein reduces the
possibility of these events affecting the predictive value of markers154-158.
Proteomics has been successfully used for biomarker discovery in several
disease and toxicity studies investigating different forms of cancer159-167,
hepatotoxicity168-173, pulmonary hypertension174-176, myocardial infarction177,
nephrotoxicity178-181 and skeletal muscle toxicity95,182-184.
38
Protein profiling experiments usually consist of two main steps; protein
separation followed by identification. The separation is necessary to allow the
visualisation of proteins within a complex mixture as individual constituents. The
identification can be performed by simply obtaining a mass:charge (m/z) ratio or
can be more complicated, involving discovering the identity of interesting
proteins using MS and database searching. These steps are sometimes
followed by confirming the response of proteins using an antibody-based method
which may be able to measure proteins more specifically185-187.
Top-down proteomics investigations typically examine intact proteins without
the need to generate peptides enzymatically (or chemically) prior to MS
analysis. The term middle-up has also been used to describe the analysis of
long peptides. Gel-based techniques (e.g. one-dimensional or two-dimensional
electrophoresis) or MS-based techniques such as matrix-assisted laser
desorption/ionization-time of flight-mass spectrometry (MALDI-TOF-MS) can be
used for differential protein expression investigations in these studies.
Sometimes, proteins are then separated or isolated and peptides generated
using enzymes in order to facilitate the identification of proteins. The top-down
approach benefits from there being a greater likelihood of examining a large
proportion of the sequence of molecules. Also, post-translational modifications
are more easily detected in this way (Figure 3)188-190.
Using a bottom-up proteomic approach, a complex mixture of proteins in a
biological sample is subjected to enzymatic digestion in order to generate
peptides. The resulting peptides can then be separated using techniques such
as LC prior to MS analysis. By comparing the intensity of the peptides in
samples from control and treated subjects or by examining the ratio of tags or
labels incorporated into the peptides, protein biomarkers can then be
discovered. Proteins can also be identified using tandem MS (MS/MS) analysis
of the peptides, combined with database searching. The main advantage of this
method is that peptides are easier to detect than proteins, being smaller in size
and therefore within the upper mass limit of most high resolution mass
spectrometers (Figure 3)191,192.
39
Top-
dow
nB
otto
m-u
p
Complex mixture of proteins
(e.g. plasma, whole cell
homogenate)
Separate ‘all’proteins
Digest with trypsin
Peptide MS analysis
Identify ‘all’peptides and
proteins
Profile protein composition
Isolate protein
Separate ‘all’peptides
Digest with trypsin
Peptide MS analysis
2D-gel
Protein band intensity
MALDI-TOF-MS
Ratio of tags
Database matching
Biomarkers
Tandem MS
Peptide ion intensity/ ratio of
tags
Single or multidimensional liquid chromatography
SELDI-TOF-MS
Protein ion intensity
Key:TechniqueQuantification
Figure 3: Strategies for MS-based protein biomarker discovery The top-down approach to proteomic investigations is typically used to study the
differential expression of intact proteins using techniques such as electrophoresis or
MALDI-TOF-MS. The identification of the molecules can subsequently be performed
(following the generation of peptides by enzymatic digestion) using MS in combination
with database searching. Sometimes specific tags are used to relatively quantify
peptides or proteins. More novel proteomic techniques, such as surface-enhanced
laser desorption/ionization-TOF-MS (SELDI-TOF-MS), allow the separation and
characterisation of proteins prior to MS analysis. The bottom-up approach involves the
analysis of proteolytic peptide mixtures, which can be separated and characterised
using MS for the discovery of protein biomarkers during differential peptide or protein
analysis.
40
1.4.1 Enzymatic digestion of proteins
Protein identification has traditionally been performed using techniques such as
N-terminal sequencing (Edman sequencing), immunoblotting or even by
comparing the electrophoretic migration of unknown and known proteins
(standards)193,194. As previously discussed, the proteolytic cleavage of proteins
at specific amino acid residues combined with MS is another option. Protein
databases containing information on the theoretical masses of peptides
generated from proteins can then be interrogated using the peptide masses
determined experimentally in order to identify the proteins of interest. Only a few
peptides are required for the identification of proteins, but it is easier to identify
larger proteins as they are likely to generate more peptides following digestion,
therefore increasing the likelihood of detecting peptides195-197.
Enzymes, by their nature, are very specific in their interactions, and
endoproteases such as trypsin and chymotrypsin break specific peptides bonds
within proteins prior to MS analysis. For example, trypsin hydrolyses the C-
terminal of peptides to arginine or lysine residues (unless they are followed by
proline) whilst the specificity of chymotrypsin is broader and occurs at the C-
terminal to phenylalanine, tryptophan or tyrosine residues (again, unless
followed by proline).
In order to digest proteins (after a gel destaining step, if an in-gel digestion is
performed), they are usually denatured using reagents such as urea or sodium
dodecyl sulphate (SDS), sometimes with a reducing agent such as dithiothreitol.
This procedure causes the proteins to unwind and disrupt the disulphide
linkages therefore allowing the protease to access the unfolded protein. After
denaturing, the sample is incubated with an enzyme such as trypsin for a few
hours or overnight at 37°C so that proteolytic cleavage of the proteins can occur
(the time of incubation depending on the resistance of the proteins to cleavage).
As cysteine can be readily reduced or oxidised, the alkylation of the resulting
thiol groups using iodoacetamide is usually performed in order to transform the
cysteine groups to a more stable form (S-carboxyamidomethylcysteine). When
41
proteins are separated by electrophoresis prior to trypsin digestion, the
reduction and alkylation steps previously described can be performed prior to
the proteins being denatured in the SDS contained within the gels194,198-201.
Trypsin tends to be the enzyme of choice due to its predictable and specific
action, the simplicity of the digestion procedure and the fact that its use usually
results in peptides that are of a suitable size to be analysed by MS. It is also
readily available, cheap, available as a pure preparation (without contamination
by chymotrypsin) and also available in a form that is less prone to the autolytic
action of the enzyme on itself202-205.
1.4.2 Protein or peptide separation
Protein or peptide separation is a critical stage in protein expression studies.
Techniques such as two-dimensional polyacrylamide gel electrophoresis (2DE)
are frequently used to separate proteins206,207. This method is theoretically
capable of separating thousands of different proteins from individual samples.
In its application, a complex protein mixture is loaded onto an immobilised pH
gradient (IPG) strip before a voltage is applied. The proteins migrate through
the gel until they reach their isoelectric point (the point at which their charge is
the same as the surrounding pH) due to the pH gradient from the top to the
bottom of the IPG strip.
The strip is then applied to a polyacrylamide gel that provides a supporting
medium through which the proteins can migrate. The strong negative charge
inherent with SDS within the gel effectively makes all the proteins the same
charge and facilitates the migration of the proteins towards the anode when a
voltage is applied across the gel. Protein separation occurs according to size as
smaller proteins move faster through the gel than larger ones. Following
separation, proteins can then be visualised by staining using Coomasie Blue,
silver or SYPRO® ruby staining (depending on the sensitivity of detection
required).
42
Although it can be time-consuming and resource demanding, if it is not
automated, 2DE is a reliable and data-rich method. A gel-based approach to the
separation of proteins is useful in biomarker discovery, as it is not biased
towards a particular class of molecules on the basis on their physiochemical
properties. However, this method of separation can sometimes fail to reveal,
very low molecular weight (Mw) proteins, low abundance proteins or proteins
that do not stain well. Hydrophobic or very basic proteins are also sometimes
difficult to visualise in these gels208-210. Sometimes, one-dimensional gel
electrophoresis (1DE) is used as an alternative separation step and is simpler
and quicker to conduct although the separation of proteins is not as extensive
as 2DE. It can, however, sometimes reveal proteins that are not shown by
2DE211.
In-gel methods of digesting proteins have made it easier to first separate
proteins using electrophoresis so that peptides generated from different
sections of the gel can be introduced into the mass spectrometer in stages
using LC. This, in turn, allows a more complete analysis of all the peptides
generated from the sample(s) and the presence of SDS in the gels means that
proteins are already denatured prior to trypsin digestion.
Generated peptides are subsequently separated by LC and the individual
peptides are then analysed by MS. During LC, a liquid mobile phase is passed
over a stationary phase in the form of a column composed of chromatographic
packing (e.g. C18 for reverse-phase separations or porous beads used for ion-
exclusion chromatography). Interactions between the molecular constituents of
the samples and the stationary phase are determined by their physicochemical
properties and this determines their retention times and therefore the time at
which they are eluted from the column. LC is very amenable to automation and
therefore it is frequently used in combination with other techniques such as
MS212,213.
43
1.4.3 Mass spectrometry
MS is used to obtain m/z information on peptides, proteins or other molecules
of interest and can be conducted in a number of ways, usually classified based
on the ionisation and mass analysis methods employed. The principles behind
MS ionisation remain the same though and samples are usually introduced via
an inlet and ionised in the source. Mass analysis of the resulting ions then
occurs and the quantity of the ions detected is recorded before analysis is
performed on the collected data (Figure 4). The most popular methods for
ionisation of molecules involve the use of MALDI or electrospray ionisation
(ESI) time of flight (TOF) MS (or methods analogous to these).
Mass analyser(quadrupole, time of flight,
ion-trap)
Ion source(electrospray, MALDI,
SELDI)
Inlet(direct probe, liquid
chromatography)
Detector
Data analysis
Mass analyser(quadrupole, time of flight,
ion-trap)
Ion source(electrospray, MALDI,
SELDI)
Inlet(direct probe, liquid
chromatography)
Detector
Data analysis
Figure 4: Schematic of the basic components of a mass spectrometer Samples are introduced into the mass spectrometer via an inlet. MS analysis requires
the introduction of charged and dried molecules into the mass analyser using either a
direct probe or LC. Commonly utilised ionisation methods include electrospray
ionisation or matrix-assisted laser-desorption/ionization. Analytes can then be
characterised in terms of their m/z in the mass analyser using techniques such as time
of flight or ion trap MS. Data produced can then be analysed using a number of data-
handling tools and statistical methods.
44
1.4.3.1 Matrix-assisted laser desorption/ionization
Prior to mass analysis, ions must be produced and there are various methods
of ionising molecules for this purpose. MALDI is a common method for ionising
proteins and is the method of choice for large biomolecules, as other ionisation
techniques are less suitable for this purpose. In MALDI, the sample is mixed
with an energy-absorbing molecule (also known as the matrix) which is usually
a weak acid, capable of donating one or more protons to the sample molecules
(e.g. sinapinic acid or α-cyano-4-hydroxycinnamic acid (CHCA)). The resulting
crystalline protein-matrix complex is then irradiated by a 337nm nitrogen laser
during MS analysis. The matrix is vapourised by the laser and also imparts
laser energy to the molecules in the complex whilst also protecting the analytes
from any potentially harmful effects of the laser. This process facilitates the
desorption and ionisation of molecules from the surface of the MALDI target
prior to ions entering the mass analyser section of the mass spectrometer
(Figure 5).
Laser
Sample stage
To mass analyzer
Co-crystallised protein/matrix complex
+
+
+
+ + +
+
+
Matrix ionsPeptide ions
+
++
+
+ +
Figure 5: Matrix-assisted laser desorption/ionization During the process of MALDI, sample is applied to the target on the sample stage and
mixed with an energy-absorbing molecule (matrix). The matrix (usually a weak acid)
co-crystallises with the molecules within the sample to form a crystalline complex and
donates protons to the molecules within the sample. Upon irradiation with a 337nm
nitrogen laser, molecules are ionised and a voltage is applied to the sample stage,
forcing the ions away from the sample stage and into the mass analyser section of the
mass spectrometer.
45
MALDI is commonly used for analysing proteins although, in its application,
issues with reproducibility can arise. When matrix is applied manually to the
MALDI target, it can sometimes be applied unevenly. This, combined with the
fact that the laser usually only irradiates a part of the MALDI target in a single
sweep or scan, and that there can be pulse to pulse variation in the laser, can
contribute variations in the signal and this can affect the overall reproducibility
of spectra. As a result, multiple scans are necessary and an average of the
multiple scans is then taken in order to allow an accurate representation of the
whole target area. Despite this, the technique has the advantage of requiring
only a small amount of protein in the analysis. It is also amenable to the
analysis of large molecules with a theoretical upper range of approximately
400kDa depending on the mass analyser used and is relatively straightforward
in its application214-216.
1.4.3.2 Electrospray ionisation
ESI is another soft ionisation method and it produces intact charged and dried
analytes, which then have the capacity to travel or “fly” in the vacuum of the
mass analyser. During ESI, the application of a high voltage between the
microcapillary and the MS inlet allows sample to be introduced into the mass
spectrometer as a fine spray via the microcapillary. The voltage applied results
in the formation of an aerosol containing charged droplets and ions of the same
polarity (depending on whether the mass spectrometer is run in positive or
negative mode) are drawn out towards the counter electrode along the
capillary. When the excess charge at the tip of the capillary overcomes surface
tension, a droplet is formed which travels through the first vacuum stage of the
mass spectrometer. The use of an inert heated drying gas (e.g. nitrogen) in this
region aids the evaporation of solvent from the droplets during this journey and
the droplet size is reduced resulting in the formation of a multiply charged,
desolvated ions217-220.
46
The formation of ions occurs via evaporation of the volatile buffers and solvents
used or via fission caused by electric repulsion between like charges of the
analytes. More specifically, the ion evaporation model (IEM) and the charged
residue model (CRM) have been proposed as the mechanisms by which the
formation of charged molecules takes place. In the IEM, it is proposed that, as
the droplet shrinks, the analyte achieves sufficient energy to transfer into its
gaseous phase. Alternatively, the CRM suggests that all the solvent evaporates
leaving a bare gas phase ion which can then be analysed in the mass analyser
section of the mass spectrometer221,222.
+
+ +
+++
++++++
++++
++++++
++++
++
+
+++ + +
++
Optionalsheath liquid &
drying gas
Capillary column effluent
Charged dropletsDesolvated ions
Entrance to mass spectrometer
Heated desolvatedregion
+2kV to +5kV
++
++++
++++
++ ++
++ ++
++++
++++++
++++++++++
++++
++++++
++++++++++
++++
++++
++
+++ + +
++
Optionalsheath liquid &
drying gas
Capillary column effluent
Charged dropletsDesolvated ions
Entrance to mass spectrometer
Heated desolvatedregion
+2kV to +5kV
++
++++
++++
++
Figure 6: Electrospray ionisation During ESI, sample is introduced to the mass spectrometer via a microcapillary. The
application of a potential difference results in charged ions that are drawn along the
capillary towards a counter electrode in the volatile effluent. The use of a heated drying
gas (e.g. nitrogen) facilitates the desolvation of the ions during the production of
multiply charged ions.
47
1.4.3.3 Time of flight mass analysis
The simplest form of mass analysis is conducted is a TOF mass analyser.
Molecules are ionised at the source before an accelerating potential is applied
to the sample target and facilitates the travel of the charged ions towards the
detector. The flight of the ions takes place in a field free region (vacuum) and
the time taken for ions to reach the detector in the TOF tube is directly
proportional to their mass and charge (Figure 7). Small ions reach the detector
more quickly than larger ones and this travel time is also dependent on the
charge of the molecules. This information allows the construction of a mass
spectrum using the m/z ratio plotted against the intensity of each ion detected.
Sample target
Laser
Accelerationregion
Detector
Reflectron
Ion flight path in field free region
+25kV +0kV
Figure 7: Time of flight mass spectrometry A TOF mass spectrometer is shown with a MALDI source. Ions are produced from the
sample target and accelerated into the mass analyser when a potential difference is
applied in this region. Ions then travel through the vacuum in the TOF tube towards the
detector, which records the time of arrival and the relative quantity of the ions. The time
taken for the ions to reach the detector is proportional to the mass (m) and charge (z)
of the ions and this is represented in a mass spectrum, which shows m/z on the x-axis
and relative ion intensity of the y-axis. The incorporation of a reflectron in some
instruments increases the path length for the ions and thus improves the resolution of
ions from each other although it can lead to a loss in sensitivity, as more ions can be
lost during the extended time of flight.
48
The TOF tube is commonly combined with a quadrupole in which the radio
frequency (rf) and direct current (dc) voltages can be modified. In this way, ions
are transported through an electric field generated by four metal rods and these
rods can either allow all ions to pass through in full scan mode or can be
modified in selected ion monitoring (SIM) mode in order to stabilise the path of
the ions passing through the rods whilst destabilising unwanted ions. Ions with
an unstable trajectory remain undetected as they are discarded before they
reach the detector. Ions with a stable path through the rods, are detected and
quantified (Figure 8).
++
-
-
Source slit Exit slit
To detector
Quadrupole rods
Ions withvariable path
Ions with stable path
Ions in
Figure 8: The path of ions through a quadrupole Following the production of ions in the source, they are transported into the mass
spectrometer through the source slit. The quadrupole rods can be adjusted to allow all
ions to pass through to the detector. Alternatively, their radio frequency and direct
current voltages can be adjusted such that the passage of certain molecules is
favoured and they have a stable path through the quadrupole and towards the detector.
Other ions travel through with an unstable trajectory and are discarded before they
reach the detector.
Whereas, the techniques described can provide m/z information on molecules
of interest, they are unable to provide structural information, as they are soft
ionisation techniques that are incapable of breaking up molecules. This is
usually performed post-source in a collision cell where ions can be collided with
inert gases such as nitrogen, argon or helium during MS/MS analysis (termed
collision-induced dissociation (CID)). When ions are collided in this way, their
49
kinetic energy is internalised and contributes to the production of ion fragments,
which can then be measured precisely in the mass analyser (Figure 9). Peptide
cleavage usually occurs along the backbone of peptides and the fragment ion
type produced following the cleavage of peptides can be described more
specifically. If it occurs at the peptide amide bond and the amino terminal
retains the charge of the ion they are termed b ions whilst, if the charge is
retained at the carboxy terminal, they are termed y ions223-226.
Multiple reaction monitoring (MRM) is usually performed using triple-quadrupole
instruments (QQQ). In such instruments, two quadrupoles and a collision cell
are combined in series (quadrupole-collision cell-quadrupole). The first
quadrupole scans for a specific parent ion, which is then fragmented by CID in
the collision cell before the specific fragments are detected and quantified in the
final quadrupole.
+
+ ++
++
++
++
++
++++++++++
++++++++++
++
++
++ ++
+++
+ ++++
Pusher Detector
Reflectron
Hexapole collision cellHexapole lens
Quadrupole mass filter
Electrosprayion source
+
+ ++
++
++
++
++
++++++++++
++++++++++
++
++
++ ++
+++
+ ++++
Pusher Detector
Reflectron
Hexapole collision cellHexapole lens
Quadrupole mass filter
Electrosprayion source
Figure 9: Hybrid quadrupole-time of flight mass spectrometer The hybrid mass spectrometer consists of a source (usually electrospray) in which
molecules are ionised prior to entering the mass spectrometer. After entry into the
mass spectrometer, they are focused by lenses to pass through the quadrupole mass
filter where specific molecules can be directed towards the collision cell. In this area,
molecules are fragmented by CID and the resulting fragments produced are then
pushed into the time of flight tube where their m/z can be accurately determined.
50
1.4.3.4 Ion trap mass analysis
Ion trap instruments operate differently to other mass spectrometers in that ions
can be formed directly in the trap as well as being characterised within the
same area. Two endcap electrodes (at the entrance and exit) and two ring
electrodes provide rf and dc oscillations in order to create electric fields. Once
ions are trapped within the electric field, they can be selected and scanned
individually from the trap for mass analysis. Alternatively, they can be excited
by the application of a potential difference whilst, at the same time, an inert gas
is introduced into the trap. The subsequent collision of the ions with the collision
gas results in fragmentation of the ion (in MS/MS mode) and this can be
repeated several times (MSn)227,228.
The low mass range and the slow scanning speed of the ion trap have been
identified as being disadvantages of this type of mass spectrometer. Also, there
may be a bias towards multiply charged ions that have more kinetic energy and
therefore more chance of colliding with the inert gases within the collision cell.
This can also be considered an advantage as it effectively extends the mass
range (m/z) of peptides that can be detected. The ability to perform different
forms of MS in the same analyser, combined with the improved sensitivity over
other instruments, is also an advantage.
Ring Electrode
Ring Electrode
Entrance EndcapElectrode
Exit Endcap Electrode
Figure 10: Schematic of an ion trap In an ion trap mass spectrometer, ions are retained within the trap formed by the rf and
dc voltage applied to the endcap electrodes and the ring electrodes. Ions can then be
sequentially selected from the trap and detected. In MS/MS mode, a small amount of
energy can also be applied to ions in order to excite them before being dissociating
using an inert gas within the trap by collision-induced dissociation. In this way,
structural information on analytes of interest can be obtained.
51
1.4.4 Surface-enhanced laser desorption/ionization
The demand for high-throughput proteomics has meant that more novel and
perhaps less resource-demanding technologies for screening proteins based
around MALDI, such as SELDI-TOF-MS, have been developed. In the case of
SELDI-TOF-MS, proteins are first captured using classical chromatographic
chemistries on the active areas of ProteinChip® arrays. A range of ProteinChip®
arrays are available, each designed to selectively retain or enrich specific
subsets of proteins or peptides from complex biological samples, based on the
physicochemical properties of the proteins within the sample, and also
dependent on the assay conditions specified by the user. The technique
reduces the sample preparation steps prior to MS analysis and also allows the
removal of salts and other reagents that can interfere with the process of
ionisation229-233.
ProteinChips such as CM10 (weak cation exchange), Q10 (strong anion
exchange), H50 (hydrophobic) or immobilised metal affinity capture (IMAC 30)
arrays are commonly used for differential protein expression mapping
experiments. Using these arrays, the subset of proteins retained on the array
surface can be enriched by altering assay conditions such as the pH, the salt
concentration or buffers used for binding or washing the arrays during the
protein capture process. After subsets of proteins are separated from the more
complex mixture of proteins in the sample, matrix is applied to enable the
desorption and ionisation of proteins from the ProteinChip® surface for
measurement by TOF-MS, as described for MALDI (Figure 11).
SELDI has been applied to a wide variety of sample types in differential protein
expression mapping experiments. In such experiments, researchers have
aimed to accurately predict or describe toxicological events using either a single
protein biomarker or a panel of biomarkers. The potential of SELDI has been
demonstrated in the identification of biomarkers in disease or toxicity areas
such as ovarian cancer163,234, prostate cancer165, breast cancer161,235,236,
nephritis237, pulmonary hypertension174 and myopathy95,238.
52
T
Acceleratingpotential
SELDI ProteinChip® array
Time-of-flight tube
---Vacuum---
Rel
ativ
e io
n in
tens
ity
m/z
+ ++ ++
+
Det
ecto
r
Mass spectrum
Lenses
Laser source
Figure 11: Schematic of SELDI-TOF-MS The system incorporates a TOF tube in which molecules which are ionised and
desorbed from the ProteinChip® by a laser (λ=337nm) travel towards the detector. The
time taken for the molecules to reach the detector is dependent on their size and
charge. Smaller molecules reach the detector more quickly than larger proteins and
this information is used to construct a mass spectrum displaying m/z on the x-axis and
relative ion intensity on the y-axis is produced. The system is similar to MALDI-TOF-
MS except that the chemically modified surfaces of the target allow the enrichment and
subsequent analysis of specific subsets of molecules prior to MS analysis239.
53
1.4.5 Quantitative proteomics
A number of MS-based methods have been applied to the relative quantification
of different analytes for the purpose of differential protein or peptide expression.
These can either make use of specific labels incorporated into the molecules or
can be label-free.
1.4.5.1 Absolute or relative quantitation
The use of a chemically labelled isotope-coded affinity tag (ICAT) allows
samples to be differentially labelled with two forms of ICAT reagent. Prior to
analysis, samples are tagged with either a light ICAT reagent (typically a biotin
affinity group with a linker that is labelled with hydrogen) or a heavy ICAT
reagent (typically a biotin affinity group with a linker that is labelled with
deuterium). Following purification of the peptides, the resulting differences in the
relative amounts of specific peptides (generated after enzyme digestion) can
then be exploited for differential protein expression mapping using MS240,241.
Isobaric tag for relative and absolute quantitation (ITRAQ) is another MS based
technique used for quantitative protein expression profiling. The N-terminus of
peptides generated by enzymatic digestion of proteins are labelled using
chemical tags, before comparing the levels of group-specific tags in control and
treated or diseased samples by LC-MS/MS. Following the identification and
relative quantitation of the different peptides using their fragmentation patterns,
differential peptide-expression can be performed, since fragmentation of the
chemical tags also results in the formation of reporter ions that can be used to
provide information on the relative concentrations of the peptides242-245.
In the case of stable isotope labelling with amino acids in cell culture (SILAC),
cells grown in culture are provided with normal amino acids or amino acids
labelled with non-radioactive heavy isotopes (grouped according to treatment).
Differences between labelled and non-labelled proteins can be determined by
MS, and therefore, changes in the levels of proteins within each sample (and
most likely due to treatment) can be distinguished246-248.
54
Using standards, absolute quantitation can be achieved using these methods.
In addition, these methods offer high sensitivity, high protein coverage, and are
amenable to high-throughput proteomic analysis. However, they suffer from the
requirement for the expensive tags used which are usually patented and only
available commercially249,250,251.
1.4.5.2 Multidimensional protein identification technology
In shotgun experiments, whole samples are subjected to enzymatic digestion in
order to generate peptides. Changes in the entire peptide profile are then
examined by MS in the search for treatment related changes in protein
expression. Multidimensional protein identification technology (MudPIT) utilises
a similar whole protein digestion methodology. In this case, the protein sample
is digested and the constituent peptides are separated using two on-column LC
steps and peptides are scanned in the mass spectrometer as they leave the
second column. Peptides with a known m/z can then be identified by MS/MS
and database searching. A cation exchange column is usually used followed by
a reversed-phase column and the use of two columns allows a greater amount
of separation to be achieved than using one column252-255.
The major advantage of this technique is that it is sensitive to low abundance
proteins, mainly due to the use of sensitive MS equipment (e.g. ion trap).
However, it is sometimes difficult to resolve the nature of interesting proteins as
the analysis is not conducted using intact proteins. For this reason, it is
sometimes preferable to separate the proteins prior to the application of this
technique213,256-262.
55
1.4.5.3 GeLC-MS
Sometimes proteins within the samples are separated using 1DE, before
dividing the gel into an appropriate number of sections. This results in the
separation of individual samples from each other and also divides each sample
into several different Mw regions, which can then processed separately.
Peptides from individual sections of the gel are then generated by enzymatic
digestion and the specific masses of the generated peptides can then be used
to identify the proteins using LC-MS and protein database searching. Using this
GeLC-MS approach, differential peptide analysis and protein identification is
performed simultaneously254,263,264.
Both MudPIT and GeLC-MS can collectively be considered as label-free
quantitative proteomics (LFQP) when used without labels or tags and have
proved useful in the discovery of biomarkers of lung cancer265, ovarian
cancer266 and breast cancer236. The combination of LC with MS is favourable in
that it is amenable to automation and, therefore, ideal as a method of
introducing samples into the mass spectrometer. When a vast number of ions
are detected, limitations on the sampling time or in the ability of the detector to
record information can lead to ion-suppression whereby the ionisation or
detection of molecules is adversely affected by other constituents that co-elute
with the molecules of interest. The phased introduction of peptides into the
mass spectrometer reduces the number of co-eluting peptides and therefore
increases the proportion of the proteome that can potentially be examined from
samples.
However, a disadvantage of this approach is that small proteins do not usually
generate a sufficient number of peptides to facilitate a confident identification of
the proteins. Also, the relative quantitation of peptides relies on a high degree of
accuracy when preparing samples. For example, sample loading must be equal
when 1DE is performed and gels must also be cut precisely so that accurate
comparisons can be made between different regions of the gels186,267,268.
56
1.5 Summary
In this study, a variety of proteomic techniques were used to detect, measure
and characterise protein expression changes in cultured skeletal muscle cells
following the exposure of the cells to compounds shown in vivo to have
myotoxic potential. SELDI-TOF-MS was used alongside LFQP due to the
documented success of these techniques in the searching for protein
biomarkers of toxicity, combined with their potential to discover biomarkers of
myopathy in myoblasts and myotubes treated with low concentrations of the test
compounds.
1.6 Hypothesis
“Proteomic profiling can be used to identify alterations in protein expression that
are predictive of compound-induced skeletal muscle toxicity using an in vitro cell
culture system.”
1.7 Project aims
• To compare protein expression in control and treated rat L6 myoblasts and
myotubes.
• To identify changes in the protein expression profiles, and in the levels of
specific protein ions, resulting from the administration of potentially myotoxic
compounds.
• To purify and identify potential biomarkers discovered to be useful in an in
vitro rat skeletal muscle cell system.
•
Chapter 2: Materials and methods
57
Chapter 2
2. Materials and methods
Materials
Rat L6 myoblasts were obtained (cryopreserved) from American Type Cell
culture (Virginia, USA). Heat-inactivated horse serum was obtained from
Invitrogen Ltd. (Paisley, UK). 4-[2-(3-Fluoro-4-trifluoromethyl-phenyl)-4-
methyl-thiazol-5-ylmethylsulfanyl]-2-methyl-phenoxy}-acetic acid (GWδ),
which is a highly selective PPAR-δ agonist, was kindly supplied by
GlaxoSmithKline R&D, Stevenage, UK.
Alanine aminotransferase (ALT), AST, CK and LDH kits were obtained from
Bayer (Newbury, UK). The ALD kit used in these investigations was supplied
by Randox laboratories (County Antrim, UK). sTnI, FABP3, parvalbumin-α
and Myl3 measurement kits were obtained from Meso Scale Discovery
(Maryland, USA).
NP20 (normal phase), H50, CM10, IMAC 30 ProteinChip® arrays and
sinapinic acid energy-absorbing molecule (EAM) were purchased from
Ciphergen Biosciences (Fremont, USA).
Precast NuPage® 10% bis(2-hydroxyethyl)-imino-tris(hydroxymethyl)-
methane (Bis-Tris) gels, SeeBlue® Plus 2 pre-stained Mw marker, 2-(N-
morpholino)ethane sulphonic acid SDS running buffer, lithium dodecyl
sulphate (LDS) sample buffer and NuPage antioxidant were obtained from
Invitrogen Ltd. Instant Blue™ was obtained from Novexin Ltd. (Cambridge,
UK). Sequencing grade modified porcine trypsin was obtained from
Promega, UK. HPLC grade acetonitrile, formic acid and methanol were
supplied by VWR international Ltd., (Lutterworth, UK). Zorbax 300-SB-C18
Chapter 2: Materials and methods
58
traps (5 x 0.3mm2) were obtained from Agilent Scientific (Wokingham, UK)
and C18 PepMap columns (75μM internal diameter, 15cm length) were
obtained from LC Packings, (Cumberly, UK).
For immunoblotting, polyclonal Rho GDP dissociation inhibitor (GDI) alpha
(Rho GDI) and myosin light chain polypeptide 2 (Myl2) primary antibodies,
anti-rabbit immunoglobulin G (IgG), horseradish peroxidase linked secondary
antibody and biotinylated protein ladder were obtained from New England
Biolabs (Cambridge, UK). Invitrolon polyvinylidene fluoride (PVDF)
membranes, thin filter blotting paper and thick blotting paper were obtained
from Invitrogen. Kodak D19 developing solution and unifix fixing solution
were obtained from Agar Scientific (Stansted, UK) and hyperfilm ECL 18x24
was obtained from Amersham Pharmacia (Chalfont St Giles, UK).
Dulbecco’s modified Eagle’s medium (DMEM), 2,3,5,6 tetramethyl-p-
phenylene diamine (TMPD), dimethyl sulfoxide (DMSO),
penicillin/streptomycin, horse serum, foetal bovine serum (FBS), 3-(4,5-
dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT), trifluoroacetic
acid (TFA), iodoacetamide, ammonium bicarbonate, acetic acid, Triton® X-
100, transfer buffer, Tween® and all other reagents were obtained from
Sigma-Aldrich Co. Ltd. (Poole, UK).
Chapter 2: Materials and methods
59
Methods
2.1 Cell culture and treatment
Skeletal muscle growth medium (SKGM) was prepared by supplementing
DMEM with 10% FBS, 2mM glutamine, penicillin (100 u/ml penicillin G
sodium) and streptomycin (100μg/ml streptomycin sulphate). Skeletal muscle
differentiation medium (SKDM) was prepared by supplementing DMEM with
2% heat-inactivated horse serum, 2mM glutamine and penicillin and
streptomycin.
Prior to seeding, a vial of rat L6 myoblasts containing 1x106 cells was
removed from storage in the vapour phase of liquid nitrogen (-150°C). The
vial was then allowed to stand at room temperature (RT) for 1min before
placing the vial (without total submersion) in a water bath at 37°C, until
partially thawed. The contents of the vial were then slowly transferred to a
sterile centrifuge tube containing pre-warmed SKGM. The cell density of
myoblasts was then adjusted to 1x104 cells/ml by the addition of an
appropriate amount of SKGM. This mixture was then transferred into a
suitable number of collagen-coated flasks or plates and allowed to proliferate
by incubating the cells at 37°C in a humidified atmosphere of 5% CO2 and
95% air.
Myoblasts were maintained as an adherent culture and growth medium was
replenished every 3-4 days until at least 80% confluence was achieved. Cells
were then sub-cultured at a ratio of 1:3-1:5 as the myoblastic, proliferative
properties of cells diminish if the cells are allowed to become over-crowded.
SKGM was aspirated from the flask before rinsing the cells with Ca2+ and
Mg2+ free Dulbecco’s phosphate buffered saline (D-PBS). Trypsin/EDTA
(0.25%) was then added to the flask(s), which were then incubated for at
least 5min at 37°C to detach cells from the surface of the growth vessel. The
added trypsin was subsequently inactivated by adding a small amount of
SKGM (containing serum) to the flask. After counting the cells using a
Chapter 2: Materials and methods
60
haemocytometer, cells were subcultured into the cell culture vessels until
sufficient numbers of cells were available for experiments. For experiments
using myotubes, myoblasts were differentiated by replacing SKGM with
SKDM, following a wash with D-PBS. Medium was replaced every 3-4 days
until a sufficient proportion (approximately 80%) of the cell population could
be distinguished as myotubes, as observed by LM (usually 10-14 days).
Differentiation into myotubes was apparent by an attenuation of the
proliferative properties of the cells, along with clear morphological changes
such as the increased presence of multinucleated cells (myotubes).
When cells were treated, a stock solution of the test compound was prepared
in neat DMSO before preparing the highest concentration by diluting the
stock solution in each experiment. Cells were then treated with a range of
concentrations of the test compounds for a period of 24h.
2.2 Cell morphology
Cells were examined by transmission electron microscopy (EM) and also by
light microscopy (LM) to check cells for signs of drug-induced damage.
These results were compared with other measures of cytotoxicity such as
total intracellular protein and MTT reduction (described in the following
sections).
Experiments were terminated by aspirating growth or differentiation medium
from cells and then washing the cells twice with D-PBS. To prepare cells for
EM, cells were washed twice (using Ca2+ and Mg2+ free D-PBS) to remove
any residual cells or growth medium. Cells were then fixed in 4%
formaldehyde/1% glutaraldehyde (v/v) for at least 1h and then centrifuged
into a pellet at 14000 x g for 1min for short-term storage prior to analysis.
Cells were processed for EM examinations by washing them five times in a
pH 7.4 sodium phosphate buffer before being fixed in 1% buffered osmium
Chapter 2: Materials and methods
61
tetroxide for 1h and processed into Agar 100 resin. Ultra-thin sections (60-
90nm) were then prepared, stained with 5% uranyl acetate (v/v) and 2.5%
lead citrate (v/v) for 1h and then examined using a Hitachi H7500
transmission electron microscope, operated at 60-80kV. Representative
digital images were captured using the AMT Advantage Camera System
(v5.42.515).
EM examinations were qualitative assessments of general morphological
features and cells were screened for any ultrastructural abnormalities likely
to be indicative of an adverse effect of drug-treatment. The operation of the
electron microscope during EM examinations was kindly performed by Paul
McGill (Pathology, GlaxoSmithKline R&D, Ware, UK).
2.3 Total intracellular protein
Samples were prepared for the measurement of total intracellular protein by
washing the cells as previously described for cell morphology examinations.
Protein extraction buffer (containing 9M urea, 2% CHAPS and 1% DTT) was
then added to each well before the cell culture plate was incubated on a
horizontal shaker for at least 30min at 2-8°C. Cell lysates were then stored at
-80°C until analysis.
Total intracellular protein was then assessed using the bicinchoninic acid
(BCA) protein assay according to the manufacturer’s instructions269. Results
obtained were compared to the absorbances obtained at 595nm following
the analysis of samples from a standard curve (bovine serum albumin) on
the same plate.
Chapter 2: Materials and methods
62
2.4 MTT reduction
Samples for the MTT reduction assay were analysed directly in separate
plates as the MTT assay eventually results in damage to the cells under
investigation.
Two hours before the end of each 24h experiment, 200µl of 5mg/ml MTT
reagent was added to 2ml of growth or differentiation medium in each well of
a 24-well cell culture plate before incubating the plate at 37°C, 5% CO2 and
95% humidity for a further 2h. During this period, MTT reagent was taken up
and reduced by the mitochondria of viable cells to form an insoluble
formazan product. This accumulated within the cell, as it was unable to pass
through the plasma membrane. The growth medium was then removed from
the wells of the plate and the remaining formazan product was solubilised by
adding 2ml of isopropanol to each well and incubating for 30min. The relative
absorbance of the cell lysates was then compared at 550nm270.
2.5 Biochemical analyses
Growth or differentiation medium was prepared for the evaluation of enzyme
leakage by collecting the medium from the cell culture plates and centrifuging
it at 3000 x g for 5min to ensure that samples were not contaminated with
cells. The supernatant was then collected for analysis.
Cell lysates were prepared for biochemical analysis by initially washing the
remaining cells free of growth or differentiation medium (as described
previously for total intracellular protein and EM measurements). Cells were
then lysed by adding 0.1% Triton® X-100 (v/v) and the plate incubated for
30min at 2-8°C with agitation in order to release the intracellular contents of
the cells. The lysates were then stored at -80°C until analysis.
The Advia 1650 (Bayer, Newbury, UK) is a discrete automated chemistry
analyser capable of running a wide variety of spectrophotometric assays in
Chapter 2: Materials and methods
63
serum, urine and in vitro samples and was used to analyse the levels of the
leaked and intracellular levels of ALT, AST, CK, LDH and ALD.
Leaked and intracellular levels of sTnI, FABP3, parvalbumin-α and Myl3
were also assessed. These assays were performed using Meso Scale
Discovery technology, which uses MultiArray™ plates in combination with
electrochemiluminescence detection. Proprietary antibodies are attached to
the surface of the wells of MultiArray™ plates and are able to capture
specific analytes of interest. The use of a non-radioactive light emitting tag
on the antibodies facilitates the quantitation of markers of interest at 620nm
upon electrical stimulation of the target. The wells of the plates also
incorporate carbon electrodes which enhance the electrochemiluminescence
signal when stimulated97,271,272.
The operation of the Advia 1650 clinical chemistry analyser for the
measurement of ALT, AST, CK, LDH, ALD was kindly conducted by Aubrey
Swain (Clinical Pathology, GlaxoSmithKline R&D, Welwyn, UK) and the
operation of the Meso Scale Discovery MultiArray™ plate reader was
performed by Ian Roman (Clinical Pathology, GlaxoSmithKline R&D,
Welwyn, UK).
2.5.1 Alanine aminotransferase
The kit for the measurement of ALT contains two reagents and, during the
analysis, 20μl of sample is initially added to 80μl of a first reagent (containing
L-alanine). The measurement of ALT involves a coupled reaction which is
initiated by the addition of 16μl of a second reagent (containing α-
oxoglutarate) and ALT within the sample catalyses the conversion of L-
alanine and 2-oxoglutarate to pyruvic acid which in turn is involved in the
oxidation of NADH to NAD+. The rate of the resulting absorbance decrease is
monitored at 340nm over 5min and is directly proportional to the activity of
ALT in the sample. The method is validated by the manufacturer as being
linear between 0-1100U/L273-275.
Chapter 2: Materials and methods
64
α-ketoglutarate + L-alanine L-glutamate + Pyruvate ALT
LDH
Pyruvate + NADH + H+ Lactate + NAD+
2.5.2 Aspartate aminotransferase
The sample volumes and assay conditions for this assay are as described for
ALT except that the first reagent contains L-aspartate instead of L-alanine. In
the first reaction, AST catalyses the reaction between α-ketoglutarate and L-
aspartate and L-glutamate and oxaloacetate are formed. In the second
reaction, malate dehydrogenase (MDH) catalyses the oxidation of NADH to
NAD+ and the rate of the resulting decrease in the absorbance of the
reaction mixture (monitored at 340nm) is directly proportional to the AST
activity in the sample. The linear range of the assay is between 0-
1000U/L275,276.
α-ketoglutarate + L-aspartate L-glutamate + Oxaloacetate AST
MDH
Oxaloacetate + NADH + H+ Malate + NAD+
2.5.3 Creatine kinase
In this assay, 80μl of the reagent (containing creatine phosphate, adenosine
diphosphate (ADP), NADP and glucose) is reacted with 8μl of sample and
incubated for 5min. The CK-catalysed reaction between creatine phosphate
and ADP results in the formation of creatine and adenosine triphosphate
(ATP). The ATP derived is, in turn, reacted with glucose and CK activity is
determined by measuring the rate of the increase in the absorbance at
340nm when NADP+ is reduced to NADPH as a result of this reaction277,278.
Chapter 2: Materials and methods
65
ATP + glucose ADP + glucose-6-phosphate
D-glucose-6-phosphate + NADP+ 6-Phosphogluconate + NADPH + H+
Creatine phosphate + ADP Creatine + ATP CK
Hexokinase
Glucose-6-phosphate dehydrogenase
2.5.4 Lactate dehydrogenase
LDH activity is determined by exploiting the LDH-catalysed reduction of
pyruvate to lactate. In the assay, 14μl of sample is reacted with 80μl of the
working reagent (containing pyruvate and NADH) and the rate of the
decrease in the absorbance at 340nm resulting from the oxidation of NADH
to NAD+ is directly proportional to the activity of LDH279.
LDH
Pyruvate + NADH + H+ Lactate + NAD+
2.5.5 Aldolase
In measuring ALD, 20μl sample is initially incubated with 250μl of a reagent
containing fructose-1,6-diphosphate which is converted into glyceraldehyde-
3-phosphate (GAP) and dihydroxyacetone phosphate (DAP) in the presence
of ALD. Triosphosphate isomerase (TIM) is added along with
glycerolphosphate dehydrogenase (GDH) and NADH (in 6μl of a second
reagent) and this converts DAP to glycerol-1-phosphate. The simultaneous
oxidation of NADH to NAD+ is used to assess the levels of ALD by
monitoring the decrease in absorbance at 340nm280,281.
GAP DAP
DAP + NADH + H+ Glycerol-1-phosphate + NAD+
Fructose-1,6-diphosphate GAP + DAP ALD
TIM
GDH
Chapter 2: Materials and methods
66
2.5.6 Meso Scale discovery markers
In measuring the levels of sTnI, Myl3, FABP3 and parvalbumin-α, 25μl of
assay diluent was initially added to the assay plates and incubated for 30min
at RT prior to the addition of 25μl of the sample or calibration standard. This
was followed by a 2h incubation at RT before washing each array with 0.1%
PBS-tween. The relevant specific detection antibody (25μl) was then added
to each well of the plate before another 2h incubation at RT, which was then
followed by another 0.1% PBS-tween wash of the plate. Read buffer (150μl)
was then added before measuring the absorbance of each plate at 620nm.
2.6 SELDI-TOF-MS analysis
2.6.1 ProteinChip® preparation
Samples were analysed by SELDI-TOF-MS on NP20, H50, CM10 and IMAC
30 ProteinChip® arrays in order to determine the protein expression profiles
of skeletal muscle cells under different assay conditions (Figure 12).
Active spotProteinChip® array Hydrophobic coating
Figure 12: A SELDI-TOF-MS ProteinChip® array Samples were applied to the active spot of the ProteinChip® array and incubated for
a short period. This facilitated the interaction between proteins within the sample
and the chemistry on the ProteinChip® surface (based on their physicochemical
characteristics). Non-specifically bound proteins were washed away before
specifically retained proteins were co-crystallised with a solution of EAM prior to MS
analysis.
Chapter 2: Materials and methods
67
H50 ProteinChip® arrays were prepared for SELDI-TOF-MS analysis by
washing each array in 50% acetonitrile:water (v/v) for 5min in centrifuge
tubes (on a roller-mixer). The arrays were then allowed to air-dry for at least
30min before use. IMAC 30 ProteinChips were supplied in a metal-free form
and these were prepared for use by adding 150μl of 100mM copper sulphate
(loading buffer) to each active spot of the array and incubating it with
horizontal shaking for 10min. Loading buffer was then removed before
washing the ProteinChips for 1min with deionised water to remove excess
buffer. Other ProteinChip® arrays did not require the same preparation steps
prior to SELDI-TOF-MS analysis.
All ProteinChip® arrays were then prepared for sample addition by placing
them into a bioprocessor assembly, which allows a sample volume of up to
300μl to be applied to each spot of the array. The binding buffer (150μl in
each well) was then added and the bioprocessor was incubated on a
horizontal shaker for 5min, before again removing the buffer. This wash was
repeated once more before samples were diluted at least 10-fold in the
binding buffer and 1µg total protein of the cell lysates was applied, in
duplicate, to each spot of the ProteinChip® array. After a 60min incubation
period, samples were removed from the arrays and non-specific binding of
proteins was reduced by performing three 5min washes of the array surface
using 150μl of the washing buffer in each well. This was followed by a short
(1min) wash using deionised water to remove any excess buffer that might
interfere with MS analysis (Table 1).
Chapter 2: Materials and methods
68
Table 1: Binding and washing buffers used for ProteinChip® preparation Binding and washing buffers used during SELDI-TOF-MS analysis of control and
treated skeletal muscle cells are shown for each array. Buffers also contained 0.1%
Triton® X-100 to reduce non-specific interactions between proteins in the sample
and the ProteinChip® surface.
ProteinChip®
arrayAssay
conditions Binding/washing buffer
NP20 H20 H20H50 10% acetonitrile 10% acetonitrile, 0.1% TFA
CM10 pH 4 100mM sodium acetate, pH 4CM10 pH 6 100mM sodium acetate, pH 6CM10 pH 8 100mM sodium acetate, pH 8
IMAC 30 Copper 100mM sodium acetate, pH 4
2.6.2 ProteinChip® mass spectrometry
Prior to MS analysis, ProteinChip® arrays were allowed to air-dry before the
addition of 0.5μl of EAM (saturated sinapinic acid solution in 50%
acetonitrile:water (v/v) containing 0.5% TFA) to each spot of the array.
Another 0.5μl of EAM was applied to each spot of the array at least 5min
after the first application before allowing the arrays to air-dry again. The
arrays were then inserted into the Ciphergen ProteinChip® Biology System
IIc (PBS IIc) for MS analysis. ProteinChips were analysed according to an
automatic data collection protocol. The ProteinChip® reader was focused on
a mass of 12kDa, optimised from 2-20kDa, set to a high mass of 100kDa.
Transients were averaged over 50% of the target area in a linear sweep (with
the mass deflector set at 1.5kDa) to generate each spectrum.
Chapter 2: Materials and methods
69
2.6.3 SELDI-TOF-MS data pre-processing and analysis
Spectra collated following the MS analysis of ProteinChip® arrays were
normalised by total ion current (TIC) to ensure that minor irregularities in the
ionisation of protein ions (resulting from inconsistencies in procedures such
as the addition of the EAM) could be accounted for post-analysis. TIC
normalisation ensures that the area under the curve for all spectra is
constant for all spectra obtained.
Spectra were also externally calibrated for mass accuracy using a protein
standard mixture containing bovine insulin (5733.6 Da) and equine
myoglobin (16951.5 Da). This ensured that the Biomarker Wizard (BMW)
software supplied by Ciphergen was able to correctly identify SELDI-TOF-
MS protein ions with the same m/z (within the specified tolerance of 0.3%) as
a cluster whose levels could then be compared to each other in subsequent
statistical analyses (Figure 13). The use of the BMW software uses an ion
detection algorithm which identifies peaks in each spectrum based on the
signal to noise ratio (S/N) and the local valley depth of SELDI-TOF-MS
protein ions. This includes an initial first pass peak detection, followed by a
more sensitive, directed, second pass detection. The appearance of an ion is
then considered in all spectra generated from different samples within a
defined experiment, and clusters of corresponding ions that occur in spectra
produced from different samples are identified and compared. The relative
ion intensity of a particular ion or set of ions is then reported for all the
spectra as an average m/z value for the cluster of corresponding ions from
the various spectra analysed (Figure 13).
Chapter 2: Materials and methods
70
5000 10000 15000 20000
m/z
Cluster 1 Cluster 3Cluster 2
Spectrum 1
Spectrum 2
Spectrum 3
Spectrum 4
Spectrum 5
Spectrum 6
Rel
ativ
e io
n in
tens
ity
5000 10000 15000 20000
m/z
Cluster 1 Cluster 3Cluster 2
Spectrum 1
Spectrum 2
Spectrum 3
Spectrum 4
Spectrum 5
Spectrum 6
Rel
ativ
e io
n in
tens
ity
Figure 13: Cluster generation using the Biomarker Wizard in Ciphergen Peaks software Skeletal muscle cells were grown and treated with TMPD or GWδ (section 2.1). Cell
homogenates were analysed using SELDI-TOF-MS (section 2.6) before peak
detection was performed. A representative diagram is shown to demonstrate cluster
generation in 6 SELDI-TOF-MS spectra using the BMW. The use of the BMW
allows a consistent set (or cluster) of protein ions to be selected in experiments
(based on user defined settings) so that the levels of a protein ion at the same m/z
can be compared across selected spectra. In this example, the height of the missing
peak in cluster 3 (spectrum 6) is reported at the average m/z of the other peaks in
the cluster.
Chapter 2: Materials and methods
71
The BMW was set so that ions were initially detected if they had a S/N ratio
of at least 5 and the second pass was set at a S/N of 2 (Figure 14). This was
performed so that the effect of chemical noise from the EAM or electrical
noise from the internal machinery of the mass spectrometer was minimised
in the spectra. Ions were clustered with a mass tolerance of 0.3% between a
mass range of 2.5 and 20kDa, as the detection of ions below this range
tends to be of limited use due to the influence of ions arising from the use of
sinapinic acid as an EAM. In addition, the limited numbers of ions usually
detected above this range tend to be broad and have low S/N ratios
compared to other ions. In spectra where an ion within a cluster was missing,
an estimated ion intensity was reported. This can cause problems with
statistical analysis of the data as this sometimes leads to the inclusion of
peaks that have a relative ion intensity that is, in reality, below the sensitivity
of the instrument. In order to address this, ions with a relative ion intensity of
less than 0.5 (the average noise level calculated for spectra generated in
these experiments) were reported as 0.5 (Figure 15).
Figure 14: Biomarker Wizard detection settings Skeletal muscle cell homogenates were analysed using SELDI-TOF-MS and protein
ions were detected in the resulting spectra. The BMW detects peaks in two phases.
The initial phase determined peaks with a S/N ratio of 5, while the second pass
detected smaller peaks at those peak locations using greater sensitivity (lower S/N
of 2). Peaks were recognised as a cluster if the difference was within 0.3% of the
average m/z.
Chapter 2: Materials and methods
72
0
1
2
3
3029.3+H
3673.2+H3706.6+H
4122.9+H4283.8+H
4777.6+HRel
ativ
e io
n in
tens
ity
m/z
x3 Magnification
0
1
2
3
3673.2+H3706.6+H
4066 +H
4122.9+H
4199 +H
4283.8+H4777.6+HR
elat
ive
ion
inte
nsity
m/z
x3 Magnification
Figure 15: Noise calculation in Ciphergen Peaks software An example of the noise calculation is shown above with the dotted line
representing the noise setting for this SELDI-TOF-MS spectrum. Ions below this
level (as shown in the magnified region in the red circle) were entered as a relative
ion intensity of 0.5. This was the average noise setting at the maximum ion height
for the spectra in this experiment and this defined the limit of sensitivity for the
measurements.
2.6.4 SELDI-TOF-MS statistical analysis
Average values for all protein ions within SELDI-TOF-MS spectra were
calculated for the replicates of each sample. A statistical analysis was then
performed in order to identify ions whose relative levels were significantly
different between treatment groups (Student’s t-test, p<0.05). This procedure
was performed in repeat experiments in order to focus on changes that were
consistent in the repeat experiments of a training dataset. A third repeat
experiment functioned as a test-dataset in which changes observed in the
training dataset were validated.
Statistical analysis was performed using Microsoft Excel (Microsoft
Corporation, Washington, USA), Statistica 8.0 (Statsoft Inc, Oklahoma,
USA), GraphPad Prism® (GraphPad software, California, USA) and SIMCA-
P+ v11.5 (Umetrics, Umeå, Sweden). Database searching for the
identification of responsive protein ions was performed using tools (e.g.
TagIdent) which were available via the Expert Protein Analysis System
(ExPASy) web-based server (www.expasy.org).
Chapter 2: Materials and methods
73
2.7 Label-free quantitative proteomics
LFQP differential protein expression profiling experiments were conducted
using a combination of 1DE and LC-MS/MS. Skeletal muscle cells treated
with the same concentrations of TMPD and GWδ as used in SELDI-TOF-MS
protein-profiling investigations (25 and 50μM, respectively) were analysed
using this approach. Samples were prefractionated by 1DE in order to
reduce their complexity before subjecting individual gel fractions to
enzymatic digestion and measuring the relative expression of the generated
peptides using LC-MS/MS.
2.7.1 One-dimensional sodium dodecyl sulphate polyacrylamide gel-electrophoresis
Samples were prepared by denaturing 5μg total protein in lithium dodecyl
sulphate (LDS) buffer and then reducing the sample in 5mM dithiothreitol, at
70°C for 10min. Samples were then cooled to RT for 10min before alkylating
with 50mM iodoacetamide for 20min. Proteins in the samples were then
separated by 1DE using 10% Bis-Tris gels. Following electrophoresis, gels
were stained with Instant Blue™ for 15min before rinsing off excess stain with
deionised water. Gels were then examined so that differences in the intensity
of protein bands likely to have resulted from drug treatment (as indicated by
their consistency within the control or treated groups) could be identified.
2.7.2 In-gel protein digestion
Samples were prepared for LC-MS/MS analysis by cutting the gels
longitudinally into individual lanes for the different samples (6 control and 6
TMPD-treated samples) and also transversely into 11 Mw regions (Figure
16). Stain was then removed from the stained gel pieces by incubating each
gel piece in 1ml 100mM ammonium bicarbonate in 50% acetonitrile until the
gel pieces were transparent (the solution being replenished as necessary to
achieve this). Individual gel pieces were then washed in deionised water
before dehydrating in acetonitrile for at least 30min. Following this,
Chapter 2: Materials and methods
74
acetonitrile was removed before drying the gel pieces in a vacuum
centrifuge. In-gel protein digestion was then performed by the addition of
300μl of 50mM ammonium bicarbonate containing 1ng/μl porcine trypsin and
incubating the sample at 37°C for at least 16h. After this incubation period,
peptides were extracted by the addition of 300µl 0.1% formic acid in 2%
acetonitrile:water [v/v] and mixed for 30min. An aliquot of this sample (400μl)
was removed, freeze-dried and stored until MS analysis.
2.7.3 LC-MS/MS analysis
The lyophilised sample was reconstituted in 60μl of water containing 0.1%
TFA immediately prior to mass spectrometric analysis. Peptides were applied
(in a 10μl injection) onto a Zorbax 300-SB-C18 trap (5 x 0.3mm2) and
separated by nano-LC using 75μm internal diameter, 15cm C18 Pepmap
columns. A flow rate of 250nl/min was used over a linear gradient of 0-45%
acetonitrile containing 0.1% formic acid (v/v) for 70min. Two traps and
columns were run in parallel in order to increase sample throughput although
the run order was adjusted such that samples for comparison to each other
were analysed on the same traps and columns. After LC analysis, eluted
peptides were introduced by electro-nanospray into an LTQ linear ion trap
mass spectrometer for peptide measurement and identification using the
settings detailed in Table 2.
Chapter 2: Materials and methods
75
a) b)
c) d)
Figure 16: Sectioning of one-dimensional electrophoresis gels Skeletal muscle cells were cultured before they were treated with TMPD for 24h
(section 2.1). Proteins within the samples were separated using 1-dimensional
electrophoresis (section 2.7.1) before the gels were (a) divided longitudinally into
lanes to separate individual samples and (b) transversely into Mw regions (c) guided
by Mw markers in order to fractionate the gels prior to subjecting individual excised
sections (d) to proteolytic digestion using porcine trypsin.
Chapter 2: Materials and methods
76
Table 2: Ion trap mass spectrometer settings used for the analysis of skeletal muscle cells Skeletal muscle cells were cultured and treated with TMPD (section 2.1). An LTQ
ion trap mass spectrometer was used to analyse the peptides generated following
1DE analysis of skeletal muscle cells (section 2.7.1) using the settings described
below.
Ion source Nanospray ionisationCapillary temperature (°C) 180Source voltage (kV) 1.4Capillary voltage (V) 3Data acquisition time (min) 70Optimised m/z range 400.0-1600.0
Data dependent settings Minimum signal required 1000 MS isolation width (Da) 2 MS normalised collision energy 35 Activation Q (v) 0.25 Activation time (ms) 30
Please note:
a) A small capillary voltage applied adds a small amount of internal energy, which can assist in the fragmentation of ions within the trap.
b) Minimum signal required refers to the minimum intensity of ions selected for MS/MS. c) MS isolation width is the error permitted for the selection of ions from the trap. d) A normalised collision energy accounts for the fact that ions at a higher m/z range require more collision
energy for fragmentation. This relationship between the m/z and the fragmentation is linear and predictable and is corrected for by this normalisation.
e) The activation Q is the radio frequency used during ion fragmentation. 2.7.4 LS-MS/MS data analysis
Bioworks Browser v3.2 (Thermo Fisher Scientific, Massachusetts, USA) was
used in combination with SEQUEST to search for protein identification
matches within a rat database of proteins and contaminants (Rattus
norvegicus RefSeq obtained from the NCBI ftp site, updated on 06/02/2008)
which contains 36612 entries282. Searches were made for all tryptic peptides
(assuming that the enzyme cleaved at both ends and also allowing for up to
two missed cleavages). Cysteine residues were treated as a fixed
modification (carboxyamidomethylation) and no variations in the other
residues were considered. The mass tolerance for the precursor and
fragment ions was set at 2.0 and 1.0 atomic mass units (amu), respectively
and chromatography data was limited to between 15 and 45min, as data
outside these times only contained a minimal number of peptides.
Chapter 2: Materials and methods
77
Using Bioworks Browser, unified search results file (srf) files were generated
for all of the proteins measured and used to examine and compare normal
protein expression in myoblasts and myotubes. This data was also examined
to investigate changes in protein expression that may have been induced by
drug administration on a global level using all the proteins expressed in the
skeletal muscle cells.
For a more focused differential peptide expression analysis, data was
imported into Decyder™ MS software version 2.0 (GE Healthcare, Chalfont
St. Giles, UK) where peptide ions were detected using the settings described
in Table 3. Base ion chromatograms were time-aligned in the PepMatch
portion of the Decyder™ MS software (so that chromatography retention
times were consistent between different LC runs) before peptide matching
was performed using the settings also shown in Table 3.
Table 3: Detection settings used in Decyder™ MS software Peptides were detected in the PepDetect module of Decyder™ MS software using
the settings shown so that an optimal S/N was achieved. Subsequent peptide
matching between chromatographic runs was performed using the PepMatch
module of the Decyder™ MS software during differential peptide expression
analysis.
Detection settings
Crop retention time 15-45Typical peak width (min) 0.8LC peak shape tolerance (%) 20m/z shift tolerance (amu) 0.6m/s shape tolerance (%) 5Remove peptides below S/N 1.1Remove peptides of unspecified charge below S/N 1.1Include charge states 1-3
Matching settings
Tolerance for retention time matching (min) 0.5Tolerance for m/z matching (Da) 0.5
Chapter 2: Materials and methods
78
Following the detection and matching of peptides, the levels of each peptide
detected were then assessed and compared between control (n=6) and
TMPD-treated (n=6) samples. The aim was to identify proteins that showed
consistent and reproducible differences in their levels in response to
treatment with TMPD (Student’s t-test; p<0.05).
Proteins were only considered as putative biomarkers if all the peptides
contributing to its identification showed a consistent response to treatment
(i.e. in their relative levels in samples and also in terms of being increased or
decreased). Whilst a peptide may have been detected at sufficient levels to
contribute to the identification of a protein, it was recognised that it might not
necessarily be above the limit of quantitation. As such, peptides were only
used for the quantitation of proteins if they were detected in at least half of
the samples included (6/12) to ensure that unreliable data was not used for
quantitation. The cross-correlation (Xcorr) value obtained from SEQUEST
gives an indication of how close the peptide fragment mass spectrum
obtained matches the theoretical spectrum predicted for that peptide and a
score of >2 usually indicates a good match283-285. In considering proteins as
possible biomarkers, the Xcorr value was also taken into account, as was the
correlation between the Mw region of the 1D-gel in which it was found and
the calculated Mw.
Chapter 2: Materials and methods
79
Retention time (min)
m/z
ControlControl TreatedTreated
Peptide detection
Differential peptide analysis
ControlControl TreatedTreated
(a) (b)
(c)
(d)
Retention time (min)
m/z
K.LKGADPEDVITGAFK.V
Figure 17: Schematic of quantitative Decyder™ MS differential peptide analysis (a) All the peptides detected were initially represented in a gel-view plot where the
intensity of individual spots represented the levels at which individual peptides were
detected. (b) Peptides were then detected using user-defined settings (Table 3).
Blue boxes represent peptides selected for comparison and brown crosses show
where MS/MS data was available for the identification of peptides. (c) The intensity
of selected peptides could then be compared to determine differences in peptide
expression between control and treated samples. (d) The peptides could also be
visualised using a 3D plot. One of the peptides used in the quantitation of myosin
light chain 2 in myotubes is shown here.
Chapter 2: Materials and methods
80
2.7.5 Immunoblotting
Compound-induced changes in the expression of some of the protein
biomarkers discovered by LC-MS/MS analysis were confirmed by Western
blotting. Cell lysates were analysed by 1DE (10μg total protein per
sample/lane; all other procedures and conditions as described in section
2.7.1). Samples were analysed along with prestained Mw markers and a
biotinylated protein ladder in order to accurately estimate protein Mw and the
transfer of proteins to the PVDF membrane.
Prior to the transfer of proteins from the gel to the PVDF membrane, thick
blotting paper was immersed until saturated in transfer buffer. A 0.45μM
(pore size) PVDF membrane was then fully immersed in methanol for 1min
followed by another 1min immersion in deionised water, before a final 5min
immersion in transfer buffer. Thin filter paper was also immersed in transfer
buffer prior to preparing a sandwich of thick blotting paper/thin filter
paper/PVDF membrane/gel/thin filter paper/thick blotting paper with the
PVDF membrane nearest the positive electrode onto the positive electrode of
a semi-dry blotter (Figure 18). Excess air was then removed from the blot by
rolling with a test tube before placing the negative electrode on top of the
sandwich and allowing the transfer of proteins to the membrane by running
the semi-dry blotter at 120mA for 1h.
+
-
+
-Gel
PVDF membrane
Thin filter paper
Thick blotting paper
Figure 18: Set up of PVDF membrane/gel sandwich for protein transfer Following 1DE, gels were set up for the transfer of separated proteins to a PVDF
membrane. Thick blotting paper and filter paper were also incorporated into a
sandwich in between the negative and positive electrode of a semi-dry blotter before
transferring proteins from the gel to the PVDF membrane by applying a charge of
120mA for 1 hour.
Chapter 2: Materials and methods
81
Non-specific binding sites were blocked by incubating the PVDF membrane
with blocking solution (5% BSA [w/v] in 0.1% PBS/Tween) for 1h at RT with
agitation. The membrane was then washed using 0.1% PBS/Tween (3 x
5min washes). The primary antibodies were then diluted appropriately
(1:1000 for RhoGDI and 1:500 for Myl2 in blocking solution) and the
membrane incubated in the antibody solution at RT with agitation (1h for
RhoGDI and 4h for Myl2). Following this, the membrane was washed again
(3x5min with 0.1% PBS/Tween) before incubating in a solution containing the
horseradish peroxidase (HRP)-linked secondary antibody and antibiotin HRP
(at 1:2000 and 1:1000 dilution, respectively). Final washes in (3 x 5min) in
0.1% PBS/Tween were then performed prior to antibody detection.
In order to detect bound proteins, the PVDF membrane was incubated (with
agitation) with freshly prepared Lumiglo® peroxidase chemiluminescence
substrate solution for 1min before draining any excess reagent from the
membrane and placing it inside a plastic transparent wallet within a
developing cassette. Under red light conditions, hyperfilm was exposed to
the membrane for an appropriate exposure time such that films were not
over or under exposed (approximately 15s). The film was subsequently
developed by immersing the film in developer solution for 5min, followed by a
short 30s wash in tap water. This was then followed by a 5min incubation in
fixing solution followed by a 1min wash in tap water. Films were then air-
dried before examining the expression of specifically bound proteins.
Quantitative assessments of the levels of the proteins measured specifically
by immunoblotting were made using ImageJ picture processing and analysis
software available from http://rsbweb.nih.gov/ij/.
Chapter 3: Results
82
Chapter 3
3. Results
The value of an in vitro system for assessing the myotoxic potential of novel
compounds was investigated using cultured rat L6 myoblasts and myotubes
differentiated from myoblasts. The effectiveness of different strategies for
testing the effects of test compounds on skeletal muscle was examined and
treatment-induced changes in the levels of proteins expressed in skeletal
muscle cells were assessed in order to identify possible biomarkers of
myopathy.
3.1 Establishment of the cell culture system
After seeding, myoblasts remained in the medium as unattached, spherical cells
in suspension within the growth medium (Figure 19a). After this, cells gradually
attached to the surface of collagen-coated cell culture plates and non-attached
cells were removed when the medium was replenished after 2h (Figure 19b).
Over the next 22h, cells increased to 10-15% coverage of the cell culture plate
(Figure 19c). When observed at 24h, myoblasts were spindle-shaped and were
also shown (by EM) to be present as individual cells with a single nucleus
(Figure 19d). They also displayed features of an immature, undifferentiated cell
such as an abundance of ribosomes attached to rough endoplasmic reticulum
(RER). There were also very few mitochondria and a euchromatic (often folded)
nucleus containing a prominent nucleolus and many free (unattached)
ribosomes present within the cell, indicative of an actively proliferating cell
involved in protein synthesis (Figure 19e).
Chapter 3: Results
83
a) b) c)
x100
d) e)
x8000 x30000
Nucleus
Nucleus
RER
x100 x100
Figure 19: Establishment of the rat L6 myblast cell culture system Rat L6 myoblasts were resuscitated before a) seeding the cells as a suspension culture
(section 2.1). Cultures were then prepared for morphological examinations and cell
morphology was examined using LM and EM (section 2.2). b) During a 2h period, the
majority of myoblasts attached to the surface of collagen-coated plates, appearing as
individual spindle-shaped cells. c) Over the next 24h, cells gradually increased to 10-
15% plate coverage due to proliferation. d) Myoblasts were characterised by a
euchromatic nucleus containing a prominent nucleolus (∗) and many ribosomes, but
few other organelles within the cytoplasm. These cells also interdigitated with adjacent
cells without fusing to each other. e) As shown in the area defined by the dashed box in
(d), the cytoplasm of myoblasts also contained an abundance of ribosomes, few
mitochondria (arrows) and profiles of RER dilated with medium electron dense
material. (Magnification shown).
Chapter 3: Results
84
After 10 days, successful myotube formation was evidenced as approximately
80% of the cells present showed features typical of mature differentiated
skeletal muscle cells (Figure 20a). LM revealed the presence of elongated
myotubes and these were shown (when examined by EM) to be multinucleated,
having been formed by the fusion of two or more myoblasts (Figure 20b). The
differentiation of myoblasts into myotubes was further confirmed by the
presence of myofilaments, although they were disorganised in their orientation
(Figure 20c). Within cultures of myotubes, a proportion of cells (approximately
5-10%) were present whose appearance was more typical of the myoblasts
shown in Figure 19.
Chapter 3: Results
85
x8000 x30000x100
Nucleus
Nucleusa) b) c)
Figure 20: Morphology of rat L6 myotubes treated with vehicle control alone Myoblasts were initially cultured and then differentiated into myotubes over the course of 10-14 days (section 2.1). Cells were then
prepared for EM examinations (section 2.2) and (a) myotubes were distinguished by the presence of elongated cells, which were formed
by the fusion of myoblasts. b) The multinucleate nature of myotubes was apparent in this portion of the myotube and (c) the differentiated
status of the cell was further confirmed by an abundance of myofilaments characteristic of skeletal muscle, as indicated by the arrows in
the area defined by the dashed box in (b). (magnification shown).
Chapter 3: Results
86
Intracellular levels of ALD, ALT, AST, CK, LDH, sTnI, Myl3, FABP3 and
parvalbumin-α were compared between myoblasts and myotubes in order to
examine the influence of successfully myotube formation on their levels.
Significant changes in the intracellular concentrations of ALD, ALT, AST, CK,
LDH and sTnI were observed between myoblasts and myotubes (Student’s t-
test, p<0.05). ALD and AST were increased 2-fold whilst ALT was increased 3-
fold. These changes were accompanied by a 1.4-fold reduction in LDH. The
most marked changes were in CK and sTnI, which were, increased 12 and 11-
fold, respectively, in myotubes when compared to myoblasts. Increases in the
levels of these proteins were due to myoblasts maturing into adult skeletal
muscle cells (Figure 21).
0
5
10
15
ALD ALT AST CK LD sTnI Myl3 FABP3 Parvalb.
Myo
tube
/myo
blas
trat
io
***
**
**
*
*
0
5
10
15
ALD ALT AST CK LD sTnI Myl3 FABP3 Parvalb.
Myo
tube
/myo
blas
trat
io
***
**
**
*
*
Figure 21: Baseline biochemical levels in skeletal muscle cells Skeletal muscle cells were cultured (section 2.1) and sampled for biochemical analysis
(section 2.5). The myotube/myoblast ratio was calculated as the fold-change of
individual myotube cultures (n=3) compared to the mean of myoblast cultures (n=3).
Significant increases in the intracellular levels of ALD, ALT, AST, CK and sTnI were
observed when myotubes were compared to myoblasts (Student’s t-test, p<0.05). The
dotted line represents the level at which there was no difference in the intracellular
levels of the proteins (myotube/myoblast ratio=1). The changes in the levels were due
to the maturation of myoblasts into myotubes and the largest increases were seen in
CK and sTnI (12 and 11-fold, respectively). Where visible, error bars show standard
error of the mean (SEM) and significant changes are denoted by *p<0.05, **p<0.01.
Chapter 3: Results
87
Myoblasts were grown after being resuscitated from storage at -150°C in the
vapour phase of liquid nitrogen and displayed features characteristic of an
actively proliferating skeletal muscle cell. Following differentiation, these cells
showed very different morphological characteristics that were more typical of
skeletal muscle tissue in vivo. In addition, the intracellular levels of the
biochemical parameters measured were compared and the maturation of the
myoblasts into myotubes was corroborated by significant increases in the
intracellular concentrations of ALD, ALT, AST, CK and sTnI. These
observations confirmed that cell culture conditions could be modified so that
cells were encouraged to fuse together during differentiation and form mature
myotubes.
3.2 Effect of the test compounds on skeletal muscle cells
The effects of treating skeletal muscle cells for 24h with of a range of
concentrations (up to 100μM) of an experimental myotoxin (TMPD) or a
potentially myotoxic drug (GWδ) were investigated by examining changes in the
morphology of cells. Changes in the intracellular protein content were also
examined along with changes in the ability of cells to reduce a tetrazolium salt
to a formazan dye (MTT assay).
3.2.1 Effect of test compounds on myoblasts
Myoblasts treated with ≤25μM TMPD were indistinguishable from those treated
with vehicle alone in terms of size, shape and general appearance. Cell death
following treatment with 50 and 100μM TMPD was apparent by a reduction in
the number of intact cells due to a loss of cell-membrane integrity (Figure 22).
At these concentrations of TMPD, a small proportion of the myoblasts were
necrotic or unusually dense, degenerate cells that, on occasions, had ridge-like
folds of the membrane. Some myoblasts showed additional degenerative
ultrastructural changes such as swollen mitochondria with disrupted cristae,
increased numbers of neutral lipid droplets or evidence of cytoplasmic
vacuolation (Figure 23).
Chapter 3: Results
88
Vehicle control 6.25μM TMPD 12.5μM TMPD
25μM TMPD 50μM TMPD 100μM TMPD
Figure 22: Effect of a range of concentrations of TMPD on rat L6 myoblasts following 24 hours of TMPD treatment. Myoblasts were cultured before being treated with the test compound (section 2.1). Morphological examinations (section 2.2) showed the
typical spindle-shaped appearance of rat L6 myoblasts. The number of intact cells was also reduced to approximately 50% and 20% when
cells were treated with 50 and 100μM TMPD, respectively. (Magnification x200).
Chapter 3: Results
89
a) b)
X 30000 X 30000
Figure 23: Degenerative ultrastructural changes in myoblasts caused by TMPD treatment. Myoblasts were cultured before they were treated with TMPD (section 2.1) and
prepared for EM examination (section 2.2). a) Myoblasts treated with ≥ 50μM TMPD
contained vacuoles of variable size (one indicated by the arrow), a number of lipid
droplets (L) and slightly swollen mitochondria (M) with disorganised cristae. These
features were not observed in cells treated with vehicle alone. b) Interdigitations of
individual myoblasts with neighbouring cells were absent, whilst single nuclei contained
marginated chromatin and a fragmented nucleolus. Several swollen, degenerate
mitochondria and some lipid droplets were also present in the cytoplasm of affected
cells. (Magnification shown).
Total intracellular protein content was used as a measure of cell viability, as
total protein levels decrease when cell death occurs and cells are detached
from the surface of the cell culture plate and subsequently removed when
growth medium is replenished. TMPD had no significant effect on the total
intracellular protein content of myoblast cultures treated with ≤25μM TMPD
(Student’s t-test, p<0.05) and significant reductions were only observed when
myoblasts were treated with 50 and 100μM TMPD (Figure 24a). The effect of
GWδ on myoblasts was also investigated and a change in total intracellular
protein content was only observed in myoblasts treated with 100μM GWδ
(Figure 24b).
Chapter 3: Results
90
MTT reduction was used to assess the ability of the mitochondria of healthy
cells to reduce a yellow tetrazolium salt (MTT) to an insoluble purple formazan
dye, which is assumed to be proportional to the number of viable cells
remaining in the cell culture. Significant changes in the ability of myoblasts
treated with >25μM TMPD to reduce MTT confirmed that TMPD was cytotoxic
at 50 and 100μM (Figure 24a). GWδ was not toxic to myoblasts at
concentrations of ≤50μM and the toxic effect of this compound at 100μM was
confirmed by a significant change in MTT reduction (Figure 24b).
0
0.2
0.4
0.6
0.8
1
1.2
1.4
1.6
0 10 20 30 40 50 60 70 80 90 100
[TMPD] μM
Fold
cha
nge
over
con
trol m
ean Total protein
MTT reduction***
***
**
*
a)***
*
***
***
0
0.2
0.4
0.6
0.8
1
1.2
1.4
1.6
0 10 20 30 40 50 60 70 80 90 100
[GWδ] μM
Fold
cha
nge
over
con
trol m
ean Total protein
MTT reduction
b)
** **** **
***
**
Figure 24: Effect of a range of concentrations of TMPD and GWδ on
myoblasts Myoblasts were cultured and treated with TMPD or GWδ (section 2.1). a) Intracellular
protein content was assessed (section 2.3) and was decreased significantly in
myoblasts treated with 50 and 100μM TMPD (n=8 controls and 8 treated) and (b) in
myoblasts treated with 100μM GWδ (Student’s t-test, p<0.05). This was supported by
significant reductions in MTT reduction (performed as described in section 2.4) at the
same concentrations of both test compounds. Increases in intracellular protein content
and MTT reduction may reflect an increase in cell proliferation at low concentrations of
the test compounds. Significant changes compared to controls are denoted by *p<0.05,
**p<0.01, ***p<0.001. Due to small variance, errors bars are not always visible on
these graphs. Error bars show ± SEM.
Chapter 3: Results
91
Reductions in the intracellular levels of protein, the ability of myoblasts to
reduce MTT and changes in the morphology of myoblasts at concentrations of
50 and 100μM together demonstrated that TMPD was cytotoxic to myoblasts at
these concentrations. Similar changes in total intracellular protein, MTT
reduction and cell morphology were indicative of the toxic effect of GWδ at a
higher concentration of 100μM.
3.2.2 Effect on test compounds myotubes
Myotubes treated with 6.25, 12.5 and 25μM TMPD were identical in appearance
to cells treated with vehicle alone. Cytotoxicity was only evident in cells treated
with 50 and 100μM TMPD and was demonstrated by similar degenerative
ultrastructural changes to those described for myoblasts treated with the same
concentrations of TMPD. At these concentrations of TMPD, necrotic cells and
ridge-like folds in the plasma membrane were observed. It was also noted that,
of the viable cells remaining, a higher proportion contained a single nucleus, as
was the case with myoblasts (Figure 25). This indicated that a higher proportion
(approximately 15%) of myoblasts either failed to differentiate into myotubes or
may have been dedifferentiated into myoblasts. These changes were
accompanied by a decrease in the number of intact cells when viewed by LM
(Figure 26).
Chapter 3: Results
92
a) b)
X 8000 X 30000
Nucleus
Figure 25: Degenerative ultrastructural changes in myotubes Myotubes were differentiated from cultured myoblasts (section 2.1). Cells were then
prepared and examined qualitatively by EM (section 2.2). a) Of the viable cells
remaining, myotubes treated with ≥50μM TMPD had an elongated shape typical of
myotubes, although some of the cells (approximately 15%) contained a single nucleus
within the cell and, in this way, resembled myoblasts (Figure 19). b) In the area defined
by the dotted box in (a) there were many vacuoles of variable size (derived from
swollen rough endoplasmic reticulum (arrows)) and some lipid droplets (L). Some
areas of disrupted myofilaments (*) that were contained within the cytoplasm of
affected myotubes are also shown. These degenerative changes were absent from
myotubes treated with vehicle control alone.
Chapter 3: Results
93
Vehicle control 6.25μM TMPD 12.5μM TMPD
25μM TMPD 50μM TMPD 100μM TMPD
Figure 26: Effect of a range of concentrations of TMPD on rat L6 myoblasts following 24 hours of TMPD treatment. Myotubes were differentiated from myoblasts before treating them with a range of concentrations of TMPD (section 2.1). Myotubes were
then examined by light microscopy (section 2.2) and, in contrast to myoblasts, they were elongated due to the fusion of multiple myoblasts
during cell differentiation. The number of intact cells was reduced to approximately 80% and 30% plate coverage when cells were treated
with 50 and 100μM TMPD, respectively, due to drug-induced reductions in cell viability. (Magnification x200).
Chapter 3: Results
94
The effect of TMPD on myotubes was also assessed by measuring total
intracellular protein content. Significant reductions in the intracellular levels of
protein at concentrations of 50 and 100μM demonstrated that TMPD was
cytotoxic to myotubes at these concentrations (Student’s t-test, p<0.05) (Figure
27a). The effects of GWδ were also assessed by measuring intracellular protein
content, and this compound was toxic to myotubes at a concentration of 100μM
(Figure 27b).
Significant decreases in the ability of myotubes to reduce MTT again confirmed
that TMPD was cytotoxic at 50μM and 100μM (Figure 27a). Similar changes in
MTT reduction were again indicative of the toxic effect of GWδ at a higher
concentration of 100μM (Figure 27b), as it was for myoblasts.
0
0.2
0.4
0.6
0.8
1
1.2
1.4
1.6
0 10 20 30 40 50 60 70 80 90 100
[TMPD] μM
Fold
cha
nge
over
con
trol m
ean
Total proteinMTT reduction
a)***
*****
***
***
***
***
0
0.2
0.4
0.6
0.8
1
1.2
1.4
1.6
0 10 20 30 40 50 60 70 80 90 100
[GWδ] μM
Fold
cha
nge
over
con
trol m
ean
Total proteinMTT reduction
b)
*
***
***
Figure 27: Effect of a range of concentrations of TMPD and GWδ on
myotubes Myotubes were differentiated from cultured myoblasts and treated with TMPD or GWδ
(section 2.1). a) Intracellular protein content was assessed (section 2.3) and was
measured at significantly lower levels in myotubes treated with 50 and 100μM TMPD
(n=8 controls and 8 treated) or (b) 100μM GWδ. TMPD and GWδ were toxic to
myotubes at the same concentrations observed for myoblasts as demonstrated by
significant decreases in MTT reduction at the same concentrations of the test
compounds (Student’s t-test, ≤0.05). Significant changes compared to controls are
denoted by *p<0.05, **p<0.01 or ***p<0.001. Error bars show SEM although, due to
small variance, they are not always visible on these graphs.
Chapter 3: Results
95
These results demonstrated that the in vitro system established using
myoblasts (section 3.1) was sensitive to treatment with either an experimental
myotoxin (TMPD) or a potentially myotoxic drug (GWδ). Furthermore, the
assessment of cell morphology, total intracellular protein content and MTT
reduction showed that TMPD and GWδ realised their myotoxic potential at
concentrations of 50 and 100μM, respectively. One of the aims of this work was
to discover changes in the levels of proteins that were likely to be predictive of
drug effects rather than descriptive. As such, it was decided to focus further
biomarker discovery investigations on changes observed in skeletal muscle
cells treated with concentrations of the test compounds that were not overtly
cytotoxic (≤25μM TMPD and ≤50μM GWδ).
Chapter 3: Results
96
3.3 Measurement of candidate biomarkers to discover protein biomarkers of myopathy
These investigations were conducted in order to determine whether a selection
of candidate biomarkers could detect drug-induced changes in the function of
skeletal muscle cells at lower concentrations of the test compounds than the
methods already described (total intracellular protein content, MTT reduction
and cell morphology). These markers were a combination of established
markers commonly used for the assessment of skeletal muscle toxicity in vivo
(CK, LDH, ALT, AST, ALD)91,92 and markers that have more recently been
proposed as biomarkers of myopathy (sTnI, Myl3, FABP3 and parvalbumin-
α)95,286,287.
3.3.1 Effect of drug treatment on the leakage of candidate markers
Circulatory levels of markers known to be in high concentration in skeletal
muscle are commonly used to determine the status of the tissue in clinical
studies or toxicology experiments. Therefore, the effect of a range of
concentrations of TMPD (up to 100μM) on the levels of the aforementioned
candidate markers in medium taken from treated cells was investigated.
In medium taken from myoblasts, the only marker that demonstrated a dose-
related increase in its levels was AST which was significantly increased at 50
and 100μM TMPD (Student’s t-test, p<0.05). sTnI and parvalbumin-α were not
detected in the medium taken from myoblasts (Figure 28).
In myotubes, both AST and sTnI were significantly increased in the medium
from cells treated with 50 and 100μM TMPD. In addition, CK and FABP3 were
also increased in myotubes treated with 100μM TMPD. Myl3 and parvalbumin-α
were not detected in the medium taken from myotubes (Figure 29).
Chapter 3: Results
97
ALD leakage
0 25 50 75 100
0.000
0.001
0.002
0.003
0.004
[TMPD]μM
Enzy
me
activ
ity (u
/l)
ALT leakage
0 25 50 75 100
0.0000
0.0025
0.0050
0.0075
0.0100
[TMPD]μM
Enzy
me
activ
ity (u
/l)
AST leakage
0 25 50 75 100
0.00
0.01
0.02
0.03
0.04
[TMPD]μM
Enzy
me
activ
ity (u
/l)
CK leakage
0 25 50 75 100
0.0000
0.0025
0.0050
0.0075
0.0100
[TMPD]μM
Enz
yme
activ
ity (u
/l)
LD leakage
0 25 50 75 100
0.00
0.01
0.02
0.03
0.04
0.05
0.06
[TMPD]μM
Enz
yme
activ
ity (u
/l)
**
Parvalbumin-α leakage
0 25 50 75 100
0.0000
0.0025
0.0050
0.0075
0.0100
[TMPD]μM
Con
cent
ratio
n (n
g/m
l)
sTnI leakage
0 25 50 75 100
0.0000
0.0025
0.0050
0.0075
0.0100
[TMPD]μM
Con
cent
ratio
n (n
g/m
l)
Myl3 leakage
0 25 50 75 100
0.0000
0.0001
0.0002
0.0003
0.0004
0.0005
[TMPD]μM
Con
cent
ratio
n (n
g/m
l)
FABP3 leakage
0 25 50 75 100
0.000
0.001
0.002
0.003
[TMPD]μM
Con
cent
ratio
n (n
g/m
l)
LDH
Figure 28: TMPD-induced changes in leakage into myoblast conditioned medium Myoblasts were grown and treated with up to 100μM TMPD (section 2.1) before
measuring the levels of ALD, ALT, AST, CK, LDH, sTnI, Myl3, FABP3 and
parvalbumin-α leaked into the medium (section 2.5). Significant dose-related increases
in the levels of AST leaked into the medium were observed in cells treated with 50 and
100μM TMPD, although cell death was already evident in cells treated with this
concentration (section 3.2). sTnI and parvalbumin-α were not detected in the medium
taken from myoblasts. Error bars show SEM deviation (where visible) and significant
changes are denoted by *p<0.05 (Student’s t-test).
Chapter 3: Results
98
sTnI leakage
0 25 50 75 100
0.000
0.005
0.010
0.015
[TMPD]μM
Con
cent
ratio
n (n
g/m
l)
ALD leakage
0 25 50 75 100
0.0000
0.0005
0.0010
0.0015
0.0020
[TMPD]μM
Con
cent
ratio
n (n
g/m
l)
ALT leakage
0 25 50 75 100
0.000
0.001
0.002
0.003
[TMPD]μM
Con
cent
ratio
n (n
g/m
l)
AST leakage
0 25 50 75 100
0.000
0.005
0.010
0.015
0.020
[TMPD]μM
Con
cent
ratio
n (n
g/m
l)
CK leakage
0 25 50 75 100
0.0000
0.0005
0.0010
0.0015
[TMPD]μM
Con
cent
ratio
n (n
g/m
l)
LD leakage
0 25 50 75 100
0.0000
0.0025
0.0050
0.0075
0.0100
[TMPD]μM
Con
cent
ratio
n (n
g/m
l)
******
**
Myl3 leakage
0 25 50 75 100
0.0000
0.0025
0.0050
0.0075
0.0100
[TMPD]μM
Con
cent
ratio
n (n
g/m
l)
FABP3 leakage
0 25 50 75 100
0.0000
0.0001
0.0002
0.0003
0.0004
0.0005
[TMPD]μM
Con
cent
ratio
n (n
g/m
l) **Parvalbumin-α leakage
0 25 50 75 100
0.0000
0.0025
0.0050
0.0075
0.0100
[TMPD]μM
Con
cent
ratio
n (n
g/m
l)
*
LDH
Figure 29: TMPD-induced changes in leakage into myotube conditioned medium Myotubes were differentiated from myoblasts and treated with up to 100μM TMPD
(section 2.1) before measuring the levels of candidate biomarkers leaked into the
medium (section 2.5). Significant increases in the leakage of AST and sTnI were
observed in cells treated with 50 and 100μM. These changes were accompanied by
increases in CK and FABP3 in medium taken from cells treated with 100μM TMPD.
Myl3 and parvalbumin-α were not detected in the medium taken from myotubes. Error
bars show SEM deviation and significant changes are denoted by *p<0.05, **p<0.01,
***p<0.001 (Student’s t-test).
Chapter 3: Results
99
Leaked levels of AST (and sTnI in the case of myotubes) proved to be most
sensitive for detecting drug-induced changes in skeletal muscle cells.
Treatment-related increases in the leaked levels of these markers were
observed when cells were treated with 50 or 100μM TMPD, although these
changes were seen at concentrations already shown to be cytotoxic to
myoblasts and myotubes (section 3.2). There were also increases in the levels
of CK and FABP3 in myotubes at the highest concentration of TMPD tested
(100μM). Overall, there was no additional value in measuring the leakage of
candidate markers in the medium compared to examining changes in cell
morphology or measuring intracellular protein content or MTT reduction.
3.3.2 Drug-induced changes in the intracellular levels of candidate markers
The effect of a range of concentrations of TMPD (up to 100μM) on the
intracellular levels of ALD, ALT, AST, CK, LDH, sTnI, Myl3, FABP3 and
parvalbumin-α was also assessed.
Reductions in the levels of ALT, AST, CK, LDH and sTnI related to the cytotoxic
effect of the test compound were only observed when myoblasts were treated
with 100μM TMPD (Student’s t-test, p<0.05) (Figure 30).
The levels of AST and LDH were decreased in myotubes treated with 100μM
TMPD. The most responsive markers to the treatment of myotubes with TMPD
were ALD and CK, which were reduced following treatment with ≥12.5μM
TMPD (Figure 31).
Chapter 3: Results
100
Intracellular ALT
0 25 50 75 100
0.000
0.002
0.004
0.006
0.008
[TMPD]μM
Enzy
me
activ
ity (u
/l)
Intracellular ALD
0 25 50 75 100
0.00
0.05
0.10
0.15
[TMPD]μM
Enzy
me
activ
ity (u
/l)
Intracellular AST
0 25 50 75 100
0.000
0.025
0.050
0.075
0.100
[TMPD]μM
Enzy
me
activ
ity (u
/l)
Intracellular CK
0 25 50 75 100
0.00
0.01
0.02
0.03
0.04
0.05
[TMPD]μM
Enzy
me
activ
ity (u
/l)
Intracellular LD
0 25 50 75 100
0.0
0.5
1.0
1.5
2.0
[TMPD]μM
Enzy
me
activ
ity (u
/l)
* **
****
**
***
*
Intracellular sTnI
0 25 50 75 100
0.0000
0.0005
0.0010
0.0015
[TMPD]μM
Con
cent
ratio
n (n
g/m
l)
Intracellular Myl3
0 25 50 75 100
0.00000
0.00005
0.00010
0.00015
[TMPD]μM
Con
cent
ratio
n (n
g/m
l)
Intracellular FABP3
0 25 50 75 100
0.0000
0.0005
0.0010
0.0015
[TMPD]μM
Con
cent
ratio
n (n
g/m
l)
Intracellular parvalbumin-α
0 25 50 75 100
0.000
0.002
0.004
0.006
[TMPD]μM
Con
cent
ratio
n (n
g/m
l)
*
*
LDH
Figure 30: TMPD-induced changes in intracellular enzymes in myoblasts Myoblasts were grown and treated with up to 100μM TMPD (section 2.1) before the
intracellular levels of ALD, ALT, AST, CK, LDH, sTnI, Myl3, FABP3 and parvalbumin-α
were assessed (section 2.5). Significant reductions in the levels of ALT, AST, CK, LDH
and sTnI were only observed in cells treated with 100μM TMPD. These changes were
related to the toxic effects of drug-treatment. Error bars show SEM (where visible) and
significant changes are denoted by *p<0.05, **p<0.01, ***p<0.001 (Student’s t-test).
Chapter 3: Results
101
Intracellular ALD
0 25 50 75 1000.00
0.05
0.10
0.15
0.20
0.25
[TMPD]μM
Enz
yme
activ
ity (u
/l)
Intracellular ALT
0 25 50 75 1000.000
0.005
0.010
0.015
[TMPD]μM
Enz
yme
activ
ity (u
/l)
Intracellular AST
0 25 50 75 1000.000
0.025
0.050
0.075
0.100
[TMPD]μM
Enz
yme
activ
ity (u
/l)
Intracellular CK
0 25 50 75 1000.0
0.1
0.2
0.3
0.4
0.5
0.6
[TMPD]μM
Enzy
me
activ
ity (u
/l)
Intracellular LD
0 25 50 75 1000.00
0.25
0.50
0.75
1.00
[TMPD]μM
Enz
yme
activ
ity (u
/l)
Intracellular sTnI
0 25 50 75 1000.000
0.025
0.050
0.075
0.100
[TMPD]μM
Con
cent
ratio
n (n
g/m
l)
Intracellular Myl3
0 25 50 75 1000.0000
0.0001
0.0002
0.0003
0.0004
[TMPD]μM
Con
cent
ratio
n (n
g/m
l)
Intracellular FABP3
0 25 50 75 1000.00
0.01
0.02
0.03
0.04
[TMPD]μM
Con
cent
ratio
n (n
g/m
l)
Intracellular parvalbumin-α
0 25 50 75 1000.000
0.005
0.010
0.015
[TMPD]μM
Con
cent
ratio
n (n
g/m
l)
*** *** ***
**
**
** **
LDH
Figure 31: TMPD-induced changes in intracellular enzymes in myotubes Myotubes were obtained by differentiating myoblasts and these cells were treated with
up to 100μM TMPD (section 2.1). Intracellular enzyme levels were assessed (section
2.5) and significant dose-related reductions in the intracellular levels of AST and LDH
were only observed in cells treated with 100μM TMPD (Student’s t-test, p<0.05). ALD
and CK were more sensitive to treatment and were reduced following treatment with
lower concentrations of TMPD (≥12.5μM TMPD). Error bars show SEM deviation and
significant changes are denoted by *p<0.05, **p<0.01, ***p<0.001.
Chapter 3: Results
102
In the assessment of the intracellular levels of the candidate biomarkers
measured, ALD and CK were the most responsive markers to the effects of
TMPD in myotubes and were decreased when cells were treated with ≥12.5μM
TMPD. These markers were responsive at lower concentrations of the test
compound than intracellular protein content, MTT reduction and cell morphology
assessments (section 3.2) and could serve as sensitive biomarkers of myopathy
in vitro.
3.4 Differential protein expression profiling using SELDI-TOF-MS
Results described in section 3.2 demonstrated that TMPD was cytotoxic at a
concentration of 50μM in both myoblasts and myotubes and that GWδ was toxic
to both cell types at a concentration of 100μM. SELDI-TOF-MS protein profiling
was subsequently used in an attempt to discover protein ions whose levels
changed when skeletal muscle cells were treated with the test compounds.
Concentrations of the test compounds were used that were not overtly cytotoxic
with a view to discovering putative biomarkers of myopathy in myoblasts and
myotubes.
A number of different ProteinChip® arrays and assay conditions were used in
order to enrich specific subsets of proteins on the array surfaces, based on their
physicochemical characteristics (Figure 32). By using a number of different
surfaces and assay conditions, an investigation of a greater proportion of the
proteome was possible than might have been achieved using a more limited set
of assay conditions.
Chapter 3: Results
103
5000 10000 15000 200005000 10000 15000 20000
-505
101520
-50
5
10
15
-505
101520
-505
101520
-505
101520
0
5
10
15
5000 10000 15000 20000
-505
101520
-50
5
10
15
-505
101520
-505
101520
-505
101520
-5
0
5
10
15
5000 10000 15000 20000 5000 10000 15000 20000
NP20 myoblasts
H50 myoblasts
CM10 (pH4) myoblasts
CM10 (pH6) myoblasts
CM10 (pH8) myoblasts
IMAC myoblasts
NP20 myotubes
H50 myotubes
CM10 (pH4) myotubes
CM10 (pH6) myotubes
CM10 (pH8) myotubes
IMAC myotubes
m/z m/z
Rel
ativ
e io
n in
tens
ity
Rel
ativ
e io
n in
tens
ity
Figure 32: Representative SELDI-TOF-MS spectra of skeletal muscle cells from different ProteinChip® arrays Myoblasts and myotubes were cultured (section 2.1) and prepared for SELDI-TOF-MS analysis as described in section 2.6. Spectra
generated following SELDI-TOF-MS analysis of the cell homogenates on normal phase (NP20), hydrophobic (H50), weak cation
exchange (CM10) and immobilised metal affinity capture (IMAC 30) loaded with copper are shown. Protein expression analysis was
conducted on a variety of ProteinChip® array surfaces (and using different assay conditions) in order to modify the proteins retained from
the samples on each array. The m/z is shown on the x-axis and the relative ion intensity on the y-axis.
Chapter 3: Results
104
3.4.1 Optimisation of SELDI-TOF-MS protein ion detection settings
Prior to differential protein expression analysis, consideration was given to
suitable BMW settings for detecting clusters of protein ions (as described in
section 2.6.3). The aim was to achieve BMW settings such that a maximal
number of protein ions could be detected with the minimal inclusion of electrical
noise from the mass spectrometer or chemical noise from reagents used during
ProteinChip® array preparation288-290.
Using SELDI-TOF-MS data generated following the analysis of myoblasts on
NP20 ProteinChip® arrays (n=6 control, 6 TMPD-treated analysed in duplicate;
total 24 spectra), BMW settings were adjusted so that ions were detected using
a range of S/N settings between 10 and 1 during the first phase of cluster
generation (Figure 14). This range was selected as a S/N setting of 1 tends to
include many ions (possibly including noise), whilst a setting of 10 may exclude
some useful ions due to its high stringency. The S/N setting in the second pass
had no significant effect on the number of ions detected as this phase simply
supplements the first phase of detection. Therefore, a default setting of 2 was
used for this phase of detection.
A suitable S/N setting of 5 was selected for the BMW based on the deduction
that the large drop in the number of ions detected from 814 (at a BMW S/N
setting of 1) to 54 ions (when the setting was at 5) was due to the exclusion of
noise (Figure 33). An example of the ions detected in myoblasts is shown in
Figure 34.
Chapter 3: Results
105
0 1 2 3 4 5 6 7 8 9 10
0
200
400
600
800
1000
Biomarker Wizard setting (S/N)
No.
of i
ons
dete
cted
S/N setting No. ions
Decrease from previous
setting
1 814 -2 224 5903 104 1204 62 425 54 86 47 77 39 88 33 69 29 4
10 23 6
Figure 33: Influence of Biomarker Wizard detection settings on protein ion detection Myoblasts were cultured and treated with the TMPD or GWδ (section 2.1) before
analysing cell homogenates using SELDI-TOF-MS (section 2.6). In order to determine
suitable BMW detection settings, they were adjusted so that the S/N of ions included in
clusters ranged from 1-10. A S/N setting of 5 was used for the BMW in subsequent
experiments, as a setting of 10 was too stringent, leading to the exclusion of a high
proportion of ions whilst a setting of 1 resulted in the inclusion of 814 ions, which
incorporated spectral noise. S/N = signal to noise ratio.
5000 10000 15000 20000
0
2
4
m/z
Rel
ativ
e io
n in
tens
ity
Figure 34: Peaks detected in SELDI-TOF-MS spectra using selected Biomarker Wizard detection settings Myoblasts were cultured and treated with the test compounds (section 2.1) before
analysing cell homogenates by SELDI-TOF-MS (section 2.6). Suitable detection
settings (S/N of ions detected = 5 in the first pass detection) for the BMW were
obtained as shown in Figure 33 and 54 ions detected on NP20 ProteinChip® arrays in
this experiment are highlighted ( ).
Chapter 3: Results
106
3.4.2 Discovery of protein profiles that discriminate between control and treated muscle cells
SELDI-TOF-MS analysis of 6 control and 6 treated samples on NP20, H50,
CM10 (at pH 4, 6 and 8) and IMAC 30 ProteinChip® arrays was considered as a
“SELDI-TOF-MS experiment run” during these investigations. For each SELDI-
TOF-MS experiment run, a different culture of skeletal muscle cells was
resuscitated from storage and treated with the test compounds (as described in
section 2.1). Cell homogenates were then analysed on NP20, H50, CM10 (at
pH 4, 6 and 8) and IMAC 30 ProteinChip® arrays, as described in section 2.6 to
form the dataset for a SELDI-TOF-MS experiment run.
Principal component analysis (PCA) of data generated in an initial SELDI-TOF-
MS experiment run established that the treatment of myoblasts with TMPD
(Figure 35a) and GWδ (Figure 35b) resulted in changes in the levels of some
protein ions as indicated by the separation between control and treated
myoblasts in the scores plots. Furthermore, these changes were observed in
myoblasts treated with concentrations of the test compounds at which no
adverse effects of treatment were detected by examining cell morphology or by
using measures such as MTT reduction or total intracellular protein content.
PCA of data generated following the SELDI-TOF-MS analysis of myotubes was
also conducted in order to determine whether there were any changes in the
protein ion profiles of TMPD or GWδ-treated myotubes. For both TMPD and
GWδ-treated myotubes, there was natural separation (in PC2) between control
and treated samples. This demonstrated that, as with myoblasts, treating cells
with the test compounds resulted in changes to the protein ion profiles in
myotubes (Figure 36).
Chapter 3: Results
107
ControlTreatedControlTreateda) Scores plot (TMPD myoblasts) Control
TreatedControlTreated
b) Scores plot (GWδ myoblasts)
Principal component 1 (PC1) PC1
PC
2
-20
-10
0
10
20
-40 -30 -20 -10 0 10 20 30 40
-20
-10
0
10
20
-40 -30 -20 -10 0 10 20 30 40
-20
-10
0
10
20
-30 -20 -10 0 10 20 30
-20
-10
0
10
20
-30 -20 -10 0 10 20 30
PC
2
Figure 35: Principal components analysis comparing control and treated myoblasts following SELDI-TOF-MS analysis Myoblasts were cultured (section 2.1) and data generated following SELDI-TOF-MS
analysis of the cell lysates (section 2.6) was evaluated using PCA (including all 592
ions detected in the analysis). a) Separation between control and TMPD-treated
myoblasts is shown in the scores plot and the variation describing the separation due
to TMPD treatment in myoblasts (22%) is mainly in principal component 2 (PC2). b)
GWδ-treatment also resulted in differences between the protein ion profiles of control
and treated myoblasts, which was accounted for mainly in PC1 (17% of the variation).
Chapter 3: Results
108
ControlTreatedControlTreateda) Scores plot (TMPD myotubes) Control
TreatedControlTreated
b) Scores plot (GWδ myotubes)
Principal component 1 (PC1) PC1
PC
2
PC
2
-20
-10
0
10
20
-30 -20 -10 0 10 20 30
-20
-10
0
10
20
-30 -20 -10 0 10 20 30
-20
-10
0
10
20
-40 -30 -20 -10 0 10 20 30
-20
-10
0
10
20
-40 -30 -20 -10 0 10 20 30
Figure 36: Principal components analysis comparing control and treated myotubes following SELDI-TOF-MS analysis Cultured myotubes were collected (section 2.1) and data generated following SELDI-
TOF-MS analysis (section 2.6) was evaluated using PCA (including all 331 ions
detected in the analysis). a) Separation between control and TMPD-treated myotubes
is shown in the scores plot and the variation describing the separation due to TMPD
treatment in myotubes (18%) is mainly in principal component 2 (PC2). b) GWδ-
treatment also resulted in differences between the protein ion profiles of control and
treated myotubes, which was accounted for mainly in PC2 (15% of the variation).
These results demonstrated that the treatment of myotubes with TMPD and GWδ
resulted in changes to the protein ion profiles.
Chapter 3: Results
109
PCA was unsupervised and was performed in order to determine whether there
was any natural separation between control and treated skeletal muscle cells.
The natural separation between control and treated skeletal muscle cells shown
by PCA served as evidence that treating the cells with TMPD or GWδ resulted
in changes to the protein expression profile in the cells. Partial least squares
discriminant analysis (PLS-DA) was subsequently conducted and this
supervised method of data analysis was performed in an attempt to determine
the protein ions responsible for the separation between control and treated
myoblasts previously shown by PCA. This analysis, however, resulted in the
identification of over 85 protein ions (out of a total of 592 ions) in TMPD-treated
myoblasts that had a variable importance of projection (VIP) coefficient >1
(Figure 37a and b). The VIP coefficient is a measure that summarises the
importance of each variable in the model. A VIP coefficient of >1 suggests that
an ion was an important variable in describing the separation between control
and treated samples291,292. Myoblasts treated with GWδ had over 95 ions that
had VIP coefficients >1 and that were therefore important in discriminating
between control and treated samples (Figure 37c and d).
As with myoblasts, PLS-DA was conducted in order to determine the main ions
that were changed in response to treatment with the test compounds. Again,
this was difficult to determine using this method, mainly due to the high number
of ions (331) included in the analysis. Although the scores plot showed clear
separation between control and TMPD-treated myotubes (Figure 38a and b),
there were 87 ions that had a VIP coefficient of >1. In GWδ-treated myotubes,
there was again clear separation between control and treated cells but the 97
ions with VIP coefficients of >1 again made it difficult to determine the most
important ions (Figure 38c and d).
Chapter 3: Results
110
ControlTreatedControlTreatedc) Scores plot (GWδ myoblasts) x
yd) Loadings plot (GWδ myoblasts)
ControlTreatedControlTreateda) Scores plot (TMPD myoblasts) x
yb) Loadings plot (TMPD myoblasts)
Principal component 1 (PC1)
PC
2
Principal component 1 (PC1)
PC
2
-20
-10
0
10
20
-20 -10 0 10 20-20
-10
0
10
20
-20 -10 0 10 20 -0.2
-0.1
-0.0
0.1
0.2
-0.1 0.0 0.1
w*c
[2]
NP20 2506
NP20 3001
NP20 3012NP20 3038
NP20 3264NP20 3289
NP20 3743NP20 3815NP20 3830
NP20 3874NP20 3979
NP20 4020 NP20 4341NP20 4354NP20 4570NP20 4705NP20 4719NP20 4748
NP20 4912
NP20 5663NP20 5674NP20 5747
NP20 5761
NP20 6001NP20 6119NP20 6163NP20 6191
NP20 6259NP20 6306
NP20 6420
NP20 6588
NP20 6658
NP20 6925NP20 7026
NP20 7181NP20 7494
NP20 7507NP20 8500
NP20 9097NP20 9113NP20 9123NP20 9137NP20 9161NP20 9349
NP20 9825NP20 10270
NP20 10790NP20 12350
NP20 13840NP20 14050
NP20 15060NP20 16030NP20 17810NP20 19090H50 2586
H50 2611H50 2613H50 2615H50 2618H50 2642
H50 2672H50 2688
H50 2693
H50 2703
H50 2731H50 2754H50 2787H50 2805
H50 2894
H50 2971
H50 2982
H50 3000
H50 3010H50 3036
H50 3081H50 3094H50 3106H50 3116H50 3134H50 3186H50 3204
H50 3217
H50 3227
H50 3249H50 3261H50 3289
H50 3345H50 3376
H50 3441
H50 3469H50 3510H50 3586H50 3606
H50 3625
H50 3657H50 3679
H50 3705H50 3940
H50 3967
H50 3989H50 4005H50 4016
H50 4052
H50 4110H50 4136
H50 4193H50 4221
H50 4238H50 4267H50 4297H50 4368H50 4525H50 4628H50 4716
H50 4745
H50 4843
H50 4898
H50 4922
H50 4970H50 5021
H50 5062
H50 5221H50 5273
H50 5294
H50 5314
H50 5375
H50 5523
H50 5579
H50 5635
H50 5789
H50 5825H50 5854
H50 5946
H50 5993H50 6020H50 6076
H50 6275H50 6377
H50 6538H50 6783H50 7027
H50 7094H50 7280H50 8036H50 8543H50 8586H50 8747H50 9041H50 9144
H50 10050H50 10250
H50 10730
H50 11050H50 11150
H50 11300H50 11350
H50 11710
H50 12350H50 12710H50 14030
H50 14300
H50 15030H50 15400
H50 16860CM10 pH4 2523
CM10 pH4 2542CM10 pH4 2680
CM10 pH4 2701
CM10 pH4 2722CM10 pH4 2882CM10 pH4 2995
CM10 pH4 3042CM10 pH4 3149
CM10 pH4 3178CM10 pH4 3358
CM10 pH4 3513
CM10 pH4 3584CM10 pH4 3660CM10 pH4 3675
CM10 pH4 3747
CM10 pH4 3882
CM10 pH4 3925CM10 pH4 3979
CM10 pH4 4019CM10 pH4 4076CM10 pH4 4096
CM10 pH4 4116
CM10 pH4 4203CM10 pH4 4236
CM10 pH4 4279CM10 pH4 4337CM10 pH4 4353CM10 pH4 4424
CM10 pH4 4519CM10 pH4 4571CM10 pH4 4704CM10 pH4 4717CM10 pH4 4748CM10 pH4 4761CM10 pH4 4786CM10 pH4 4832
CM10 pH4 4866CM10 pH4 4910
CM10 pH4 4935CM10 pH4 4959CM10 pH4 4976
CM10 pH4 5002
CM10 pH4 5127
CM10 pH4 5165CM10 pH4 5216
CM10 pH4 5292
CM10 pH4 5374CM10 pH4 5492
CM10 pH4 5524CM10 pH4 5623
CM10 pH4 5656
CM10 pH4 5670CM10 pH4 5743CM10 pH4 5757
CM10 pH4 5978
CM10 pH4 5997CM10 pH4 6076CM10 pH4 6116CM10 pH4 6142CM10 pH4 6159
CM10 pH4 6187
CM10 pH4 6229CM10 pH4 6257
CM10 pH4 6271
CM10 pH4 6303
CM10 pH4 6315CM10 pH4 6345CM10 pH4 6415
CM10 pH4 6508
CM10 pH4 6581
CM10 pH4 6640
CM10 pH4 6871
CM10 pH4 6912CM10 pH4 6950
CM10 pH4 6994
CM10 pH4 7033CM10 pH4 7060
CM10 pH4 7139CM10 pH4 7179
CM10 pH4 7217CM10 pH4 7503CM10 pH4 8177CM10 pH4 8436CM10 pH4 8549CM10 pH4 8602
CM10 pH4 8666
CM10 pH4 8890CM10 pH4 9090CM10 pH4 9109
CM10 pH4 9135CM10 pH4 9150CM10 pH4 9176CM10 pH4 9348
CM10 pH4 9389CM10 pH4 9592
CM10 pH4 9967
CM10 pH4 10070CM10 pH4 10250
CM10 pH4 10960CM10 pH4 11020
CM10 pH4 11290
CM10 pH4 11700
CM10 pH4 12340
CM10 pH4 13820CM10 pH4 14020
CM10 pH4 14770CM10 pH4 14950CM10 pH4 16850CM10 pH4 17200CM10 pH4 17790CM10 pH4 18460
CM10 pH6 2523CM10 pH6 2544
CM10 pH6 2566
CM10 pH6 2592CM10 pH6 2633
CM10 pH6 2657CM10 pH6 2677
CM10 pH6 2702
CM10 pH6 2722CM10 pH6 2768CM10 pH6 2792
CM10 pH6 2835CM10 pH6 2922
CM10 pH6 2971
CM10 pH6 2997
CM10 pH6 3042
CM10 pH6 3133CM10 pH6 3150CM10 pH6 3174
CM10 pH6 3209CM10 pH6 3283CM10 pH6 3330
CM10 pH6 3361
CM10 pH6 3404
CM10 pH6 3437CM10 pH6 3496
CM10 pH6 3553CM10 pH6 3585CM10 pH6 3658
CM10 pH6 3814
CM10 pH6 3836
CM10 pH6 3882
CM10 pH6 3905CM10 pH6 3924CM10 pH6 3978CM10 pH6 4021CM10 pH6 4078
CM10 pH6 4110CM10 pH6 4223CM10 pH6 4245
CM10 pH6 4337
CM10 pH6 4352
CM10 pH6 4380
CM10 pH6 4566
CM10 pH6 4703
CM10 pH6 4717
CM10 pH6 4747
CM10 pH6 4762CM10 pH6 4789
CM10 pH6 4911
CM10 pH6 4943CM10 pH6 5095CM10 pH6 5126
CM10 pH6 5293
CM10 pH6 5373CM10 pH6 5621
CM10 pH6 5655CM10 pH6 5672
CM10 pH6 5742CM10 pH6 5758
CM10 pH6 5980CM10 pH6 5998CM10 pH6 6076CM10 pH6 6116CM10 pH6 6160
CM10 pH6 6186
CM10 pH6 6230
CM10 pH6 6259
CM10 pH6 6302
CM10 pH6 6345CM10 pH6 6373CM10 pH6 6416
CM10 pH6 6510CM10 pH6 6582
CM10 pH6 6642CM10 pH6 6686
CM10 pH6 6709CM10 pH6 6736CM10 pH6 6910CM10 pH6 7003CM10 pH6 7035
CM10 pH6 7062
CM10 pH6 7135
CM10 pH6 7177
CM10 pH6 7222
CM10 pH6 7487
CM10 pH6 8161CM10 pH6 8539CM10 pH6 8661
CM10 pH6 9093CM10 pH6 9122
CM10 pH6 9136
CM10 pH6 9149CM10 pH6 9343CM10 pH6 9390
CM10 pH6 11000
CM10 pH6 11290
CM10 pH6 11700CM10 pH6 12340CM10 pH6 13810CM10 pH6 14010
CM10 pH6 14760CM10 pH6 14940CM10 pH6 16960CM10 pH6 17190CM10 pH6 17780CM10 pH6 18440CM10 pH8 2522
CM10 pH8 2546CM10 pH8 2611
CM10 pH8 2660
CM10 pH8 2680
CM10 pH8 2702CM10 pH8 2724
CM10 pH8 2746
CM10 pH8 2767CM10 pH8 2874CM10 pH8 2924
CM10 pH8 2970CM10 pH8 2992CM10 pH8 3007
CM10 pH8 3036CM10 pH8 3077CM10 pH8 3150
CM10 pH8 3210
CM10 pH8 3283
CM10 pH8 3360
CM10 pH8 3439CM10 pH8 3501CM10 pH8 3616CM10 pH8 3659
CM10 pH8 3742
CM10 pH8 3814CM10 pH8 3827
CM10 pH8 3881
CM10 pH8 3925
CM10 pH8 3980
CM10 pH8 4021
CM10 pH8 4080
CM10 pH8 4097CM10 pH8 4115CM10 pH8 4224CM10 pH8 4245
CM10 pH8 4338
CM10 pH8 4354CM10 pH8 4543CM10 pH8 4562
CM10 pH8 4704CM10 pH8 4747CM10 pH8 4762CM10 pH8 4789CM10 pH8 4815
CM10 pH8 4913CM10 pH8 4946
CM10 pH8 5216CM10 pH8 5294CM10 pH8 5374CM10 pH8 5522CM10 pH8 5621
CM10 pH8 5657CM10 pH8 5673CM10 pH8 5744
CM10 pH8 5760 CM10 pH8 5979CM10 pH8 5999
CM10 pH8 6056
CM10 pH8 6077CM10 pH8 6118
CM10 pH8 6143CM10 pH8 6161
CM10 pH8 6187CM10 pH8 6229
CM10 pH8 6260CM10 pH8 6303
CM10 pH8 6346
CM10 pH8 6374CM10 pH8 6416
CM10 pH8 6511
CM10 pH8 6541CM10 pH8 6582
CM10 pH8 6643CM10 pH8 6687
CM10 pH8 6907CM10 pH8 6951
CM10 pH8 7005
CM10 pH8 7037CM10 pH8 7179
CM10 pH8 7488CM10 pH8 7503
CM10 pH8 8190
CM10 pH8 8494
CM10 pH8 8540CM10 pH8 8650
CM10 pH8 9078CM10 pH8 9093
CM10 pH8 9121CM10 pH8 9135CM10 pH8 9341
CM10 pH8 9390CM10 pH8 9599
CM10 pH8 10250CM10 pH8 10960CM10 pH8 11300CM10 pH8 11670CM10 pH8 12280CM10 pH8 12340
CM10 pH8 13810CM10 pH8 14010CM10 pH8 14950CM10 pH8 15990CM10 pH8 16970CM10 pH8 17760CM10 pH8 18440IMAC 2511
IMAC 2538
IMAC 2551IMAC 2561IMAC 2578IMAC 2594IMAC 2616
IMAC 2658
IMAC 2682IMAC 2692IMAC 2700
IMAC 2716IMAC 2736IMAC 2756
IMAC 2854IMAC 2878IMAC 2920
IMAC 2959IMAC 2990IMAC 3012
IMAC 3028IMAC 3047
IMAC 3065IMAC 3077IMAC 3084IMAC 3100IMAC 3155
IMAC 3177IMAC 3205IMAC 3356IMAC 3397
IMAC 3493IMAC 3615
IMAC 3658
IMAC 3795IMAC 3824IMAC 3882
IMAC 3925
IMAC 4058IMAC 4115IMAC 4225
IMAC 4567
IMAC 4588
IMAC 4705IMAC 4746
IMAC 4843
IMAC 4909IMAC 4939IMAC 4995IMAC 5211IMAC 5289IMAC 5426
IMAC 5523IMAC 5620IMAC 5790
IMAC 5979
IMAC 5998
IMAC 6051
IMAC 6116
IMAC 6160
IMAC 6186
IMAC 6230
IMAC 6259
IMAC 6302
IMAC 6344
IMAC 6415
IMAC 6579
IMAC 6622IMAC 6684
IMAC 6709IMAC 6737
IMAC 6873
IMAC 6928IMAC 6952IMAC 6991
IMAC 7034IMAC 7062
IMAC 7177IMAC 7222
IMAC 7493IMAC 8194
IMAC 8557
IMAC 8600
IMAC 9119IMAC 9138
IMAC 9166
IMAC 9389IMAC 9595IMAC 9792IMAC 9941IMAC 9978IMAC 10060
IMAC 10240IMAC 11010
IMAC 11300
IMAC 11700IMAC 13830
IMAC 14040IMAC 14760IMAC 16860IMAC 16960IMAC 17800IMAC 18450
$M3.DA1
$M3.DA2
-0.2
-0.1
-0.0
0.1
0.2
-0.1 0.0 0.1
w*c
[2]
NP20 2506
NP20 3001
NP20 3012NP20 3038
NP20 3264NP20 3289
NP20 3743NP20 3815NP20 3830
NP20 3874NP20 3979
NP20 4020 NP20 4341NP20 4354NP20 4570NP20 4705NP20 4719NP20 4748
NP20 4912
NP20 5663NP20 5674NP20 5747
NP20 5761
NP20 6001NP20 6119NP20 6163NP20 6191
NP20 6259NP20 6306
NP20 6420
NP20 6588
NP20 6658
NP20 6925NP20 7026
NP20 7181NP20 7494
NP20 7507NP20 8500
NP20 9097NP20 9113NP20 9123NP20 9137NP20 9161NP20 9349
NP20 9825NP20 10270
NP20 10790NP20 12350
NP20 13840NP20 14050
NP20 15060NP20 16030NP20 17810NP20 19090H50 2586
H50 2611H50 2613H50 2615H50 2618H50 2642
H50 2672H50 2688
H50 2693
H50 2703
H50 2731H50 2754H50 2787H50 2805
H50 2894
H50 2971
H50 2982
H50 3000
H50 3010H50 3036
H50 3081H50 3094H50 3106H50 3116H50 3134H50 3186H50 3204
H50 3217
H50 3227
H50 3249H50 3261H50 3289
H50 3345H50 3376
H50 3441
H50 3469H50 3510H50 3586H50 3606
H50 3625
H50 3657H50 3679
H50 3705H50 3940
H50 3967
H50 3989H50 4005H50 4016
H50 4052
H50 4110H50 4136
H50 4193H50 4221
H50 4238H50 4267H50 4297H50 4368H50 4525H50 4628H50 4716
H50 4745
H50 4843
H50 4898
H50 4922
H50 4970H50 5021
H50 5062
H50 5221H50 5273
H50 5294
H50 5314
H50 5375
H50 5523
H50 5579
H50 5635
H50 5789
H50 5825H50 5854
H50 5946
H50 5993H50 6020H50 6076
H50 6275H50 6377
H50 6538H50 6783H50 7027
H50 7094H50 7280H50 8036H50 8543H50 8586H50 8747H50 9041H50 9144
H50 10050H50 10250
H50 10730
H50 11050H50 11150
H50 11300H50 11350
H50 11710
H50 12350H50 12710H50 14030
H50 14300
H50 15030H50 15400
H50 16860CM10 pH4 2523
CM10 pH4 2542CM10 pH4 2680
CM10 pH4 2701
CM10 pH4 2722CM10 pH4 2882CM10 pH4 2995
CM10 pH4 3042CM10 pH4 3149
CM10 pH4 3178CM10 pH4 3358
CM10 pH4 3513
CM10 pH4 3584CM10 pH4 3660CM10 pH4 3675
CM10 pH4 3747
CM10 pH4 3882
CM10 pH4 3925CM10 pH4 3979
CM10 pH4 4019CM10 pH4 4076CM10 pH4 4096
CM10 pH4 4116
CM10 pH4 4203CM10 pH4 4236
CM10 pH4 4279CM10 pH4 4337CM10 pH4 4353CM10 pH4 4424
CM10 pH4 4519CM10 pH4 4571CM10 pH4 4704CM10 pH4 4717CM10 pH4 4748CM10 pH4 4761CM10 pH4 4786CM10 pH4 4832
CM10 pH4 4866CM10 pH4 4910
CM10 pH4 4935CM10 pH4 4959CM10 pH4 4976
CM10 pH4 5002
CM10 pH4 5127
CM10 pH4 5165CM10 pH4 5216
CM10 pH4 5292
CM10 pH4 5374CM10 pH4 5492
CM10 pH4 5524CM10 pH4 5623
CM10 pH4 5656
CM10 pH4 5670CM10 pH4 5743CM10 pH4 5757
CM10 pH4 5978
CM10 pH4 5997CM10 pH4 6076CM10 pH4 6116CM10 pH4 6142CM10 pH4 6159
CM10 pH4 6187
CM10 pH4 6229CM10 pH4 6257
CM10 pH4 6271
CM10 pH4 6303
CM10 pH4 6315CM10 pH4 6345CM10 pH4 6415
CM10 pH4 6508
CM10 pH4 6581
CM10 pH4 6640
CM10 pH4 6871
CM10 pH4 6912CM10 pH4 6950
CM10 pH4 6994
CM10 pH4 7033CM10 pH4 7060
CM10 pH4 7139CM10 pH4 7179
CM10 pH4 7217CM10 pH4 7503CM10 pH4 8177CM10 pH4 8436CM10 pH4 8549CM10 pH4 8602
CM10 pH4 8666
CM10 pH4 8890CM10 pH4 9090CM10 pH4 9109
CM10 pH4 9135CM10 pH4 9150CM10 pH4 9176CM10 pH4 9348
CM10 pH4 9389CM10 pH4 9592
CM10 pH4 9967
CM10 pH4 10070CM10 pH4 10250
CM10 pH4 10960CM10 pH4 11020
CM10 pH4 11290
CM10 pH4 11700
CM10 pH4 12340
CM10 pH4 13820CM10 pH4 14020
CM10 pH4 14770CM10 pH4 14950CM10 pH4 16850CM10 pH4 17200CM10 pH4 17790CM10 pH4 18460
CM10 pH6 2523CM10 pH6 2544
CM10 pH6 2566
CM10 pH6 2592CM10 pH6 2633
CM10 pH6 2657CM10 pH6 2677
CM10 pH6 2702
CM10 pH6 2722CM10 pH6 2768CM10 pH6 2792
CM10 pH6 2835CM10 pH6 2922
CM10 pH6 2971
CM10 pH6 2997
CM10 pH6 3042
CM10 pH6 3133CM10 pH6 3150CM10 pH6 3174
CM10 pH6 3209CM10 pH6 3283CM10 pH6 3330
CM10 pH6 3361
CM10 pH6 3404
CM10 pH6 3437CM10 pH6 3496
CM10 pH6 3553CM10 pH6 3585CM10 pH6 3658
CM10 pH6 3814
CM10 pH6 3836
CM10 pH6 3882
CM10 pH6 3905CM10 pH6 3924CM10 pH6 3978CM10 pH6 4021CM10 pH6 4078
CM10 pH6 4110CM10 pH6 4223CM10 pH6 4245
CM10 pH6 4337
CM10 pH6 4352
CM10 pH6 4380
CM10 pH6 4566
CM10 pH6 4703
CM10 pH6 4717
CM10 pH6 4747
CM10 pH6 4762CM10 pH6 4789
CM10 pH6 4911
CM10 pH6 4943CM10 pH6 5095CM10 pH6 5126
CM10 pH6 5293
CM10 pH6 5373CM10 pH6 5621
CM10 pH6 5655CM10 pH6 5672
CM10 pH6 5742CM10 pH6 5758
CM10 pH6 5980CM10 pH6 5998CM10 pH6 6076CM10 pH6 6116CM10 pH6 6160
CM10 pH6 6186
CM10 pH6 6230
CM10 pH6 6259
CM10 pH6 6302
CM10 pH6 6345CM10 pH6 6373CM10 pH6 6416
CM10 pH6 6510CM10 pH6 6582
CM10 pH6 6642CM10 pH6 6686
CM10 pH6 6709CM10 pH6 6736CM10 pH6 6910CM10 pH6 7003CM10 pH6 7035
CM10 pH6 7062
CM10 pH6 7135
CM10 pH6 7177
CM10 pH6 7222
CM10 pH6 7487
CM10 pH6 8161CM10 pH6 8539CM10 pH6 8661
CM10 pH6 9093CM10 pH6 9122
CM10 pH6 9136
CM10 pH6 9149CM10 pH6 9343CM10 pH6 9390
CM10 pH6 11000
CM10 pH6 11290
CM10 pH6 11700CM10 pH6 12340CM10 pH6 13810CM10 pH6 14010
CM10 pH6 14760CM10 pH6 14940CM10 pH6 16960CM10 pH6 17190CM10 pH6 17780CM10 pH6 18440CM10 pH8 2522
CM10 pH8 2546CM10 pH8 2611
CM10 pH8 2660
CM10 pH8 2680
CM10 pH8 2702CM10 pH8 2724
CM10 pH8 2746
CM10 pH8 2767CM10 pH8 2874CM10 pH8 2924
CM10 pH8 2970CM10 pH8 2992CM10 pH8 3007
CM10 pH8 3036CM10 pH8 3077CM10 pH8 3150
CM10 pH8 3210
CM10 pH8 3283
CM10 pH8 3360
CM10 pH8 3439CM10 pH8 3501CM10 pH8 3616CM10 pH8 3659
CM10 pH8 3742
CM10 pH8 3814CM10 pH8 3827
CM10 pH8 3881
CM10 pH8 3925
CM10 pH8 3980
CM10 pH8 4021
CM10 pH8 4080
CM10 pH8 4097CM10 pH8 4115CM10 pH8 4224CM10 pH8 4245
CM10 pH8 4338
CM10 pH8 4354CM10 pH8 4543CM10 pH8 4562
CM10 pH8 4704CM10 pH8 4747CM10 pH8 4762CM10 pH8 4789CM10 pH8 4815
CM10 pH8 4913CM10 pH8 4946
CM10 pH8 5216CM10 pH8 5294CM10 pH8 5374CM10 pH8 5522CM10 pH8 5621
CM10 pH8 5657CM10 pH8 5673CM10 pH8 5744
CM10 pH8 5760 CM10 pH8 5979CM10 pH8 5999
CM10 pH8 6056
CM10 pH8 6077CM10 pH8 6118
CM10 pH8 6143CM10 pH8 6161
CM10 pH8 6187CM10 pH8 6229
CM10 pH8 6260CM10 pH8 6303
CM10 pH8 6346
CM10 pH8 6374CM10 pH8 6416
CM10 pH8 6511
CM10 pH8 6541CM10 pH8 6582
CM10 pH8 6643CM10 pH8 6687
CM10 pH8 6907CM10 pH8 6951
CM10 pH8 7005
CM10 pH8 7037CM10 pH8 7179
CM10 pH8 7488CM10 pH8 7503
CM10 pH8 8190
CM10 pH8 8494
CM10 pH8 8540CM10 pH8 8650
CM10 pH8 9078CM10 pH8 9093
CM10 pH8 9121CM10 pH8 9135CM10 pH8 9341
CM10 pH8 9390CM10 pH8 9599
CM10 pH8 10250CM10 pH8 10960CM10 pH8 11300CM10 pH8 11670CM10 pH8 12280CM10 pH8 12340
CM10 pH8 13810CM10 pH8 14010CM10 pH8 14950CM10 pH8 15990CM10 pH8 16970CM10 pH8 17760CM10 pH8 18440IMAC 2511
IMAC 2538
IMAC 2551IMAC 2561IMAC 2578IMAC 2594IMAC 2616
IMAC 2658
IMAC 2682IMAC 2692IMAC 2700
IMAC 2716IMAC 2736IMAC 2756
IMAC 2854IMAC 2878IMAC 2920
IMAC 2959IMAC 2990IMAC 3012
IMAC 3028IMAC 3047
IMAC 3065IMAC 3077IMAC 3084IMAC 3100IMAC 3155
IMAC 3177IMAC 3205IMAC 3356IMAC 3397
IMAC 3493IMAC 3615
IMAC 3658
IMAC 3795IMAC 3824IMAC 3882
IMAC 3925
IMAC 4058IMAC 4115IMAC 4225
IMAC 4567
IMAC 4588
IMAC 4705IMAC 4746
IMAC 4843
IMAC 4909IMAC 4939IMAC 4995IMAC 5211IMAC 5289IMAC 5426
IMAC 5523IMAC 5620IMAC 5790
IMAC 5979
IMAC 5998
IMAC 6051
IMAC 6116
IMAC 6160
IMAC 6186
IMAC 6230
IMAC 6259
IMAC 6302
IMAC 6344
IMAC 6415
IMAC 6579
IMAC 6622IMAC 6684
IMAC 6709IMAC 6737
IMAC 6873
IMAC 6928IMAC 6952IMAC 6991
IMAC 7034IMAC 7062
IMAC 7177IMAC 7222
IMAC 7493IMAC 8194
IMAC 8557
IMAC 8600
IMAC 9119IMAC 9138
IMAC 9166
IMAC 9389IMAC 9595IMAC 9792IMAC 9941IMAC 9978IMAC 10060
IMAC 10240IMAC 11010
IMAC 11300
IMAC 11700IMAC 13830
IMAC 14040IMAC 14760IMAC 16860IMAC 16960IMAC 17800IMAC 18450
$M3.DA1
$M3.DA2
-20
-10
0
10
20
-40 -30 -20 -10 0 10 20 30 40
-20
-10
0
10
20
-40 -30 -20 -10 0 10 20 30 40 -0.2
-0.1
-0.0
-0.1 0.0 0.1
w*c
[2]
NP20 2506
NP20 3001
NP20 3012NP20 3038NP20 3264NP20 3289
NP20 3743
NP20 3815NP20 3830NP20 3874NP20 3979NP20 4020NP20 4341NP20 4354
NP20 4570
NP20 4705NP20 4719NP20 4748
NP20 4912NP20 5663
NP20 5674NP20 5747
NP20 5761
NP20 6001NP20 6119NP20 6163NP20 6191NP20 6259
NP20 6306
NP20 6420
NP20 6588
NP20 6658
NP20 6925
NP20 7026NP20 7181
NP20 7494
NP20 7507NP20 8500
NP20 9097
NP20 9113
NP20 9123NP20 9137
NP20 9161NP20 9349
NP20 9825
NP20 10270NP20 10790NP20 12350
NP20 13840
NP20 14050NP20 15060NP20 16030
NP20 17810NP20 19090H50 2586
H50 2611
H50 2613
H50 2615H50 2618
H50 2642H50 2672
H50 2688H50 2693
H50 2703
H50 2731H50 2754H50 2787H50 2805H50 2894
H50 2971
H50 2982
H50 3000
H50 3010
H50 3036
H50 3081
H50 3094
H50 3106
H50 3116
H50 3134H50 3186
H50 3204
H50 3217H50 3227H50 3249H50 3261
H50 3289
H50 3345H50 3376H50 3441H50 3469H50 3510
H50 3586H50 3606H50 3625H50 3657
H50 3679H50 3705
H50 3940H50 3967
H50 3989
H50 4005H50 4016
H50 4052
H50 4110H50 4136
H50 4193H50 4221
H50 4238H50 4267H50 4297
H50 4368H50 4525H50 4628H50 4716
H50 4745H50 4843H50 4898H50 4922
H50 4970H50 5021
H50 5062H50 5221H50 5273
H50 5294H50 5314H50 5375H50 5523H50 5579
H50 5635
H50 5789
H50 5825H50 5854H50 5946H50 5993H50 6020 H50 6076H50 6275H50 6377
H50 6538
H50 6783
H50 7027H50 7094
H50 7280H50 8036
H50 8543H50 8586
H50 8747H50 9041
H50 9144H50 10050
H50 10250
H50 10730
H50 11050H50 11150
H50 11300H50 11350
H50 11710
H50 12350
H50 12710
H50 14030
H50 14300H50 15030H50 15400
H50 16860CM10 pH4 2523
CM10 pH4 2542CM10 pH4 2680
CM10 pH4 2701
CM10 pH4 2722CM10 pH4 2882
CM10 pH4 2995
CM10 pH4 3042
CM10 pH4 3149
CM10 pH4 3178
CM10 pH4 3358CM10 pH4 3513CM10 pH4 3584
CM10 pH4 3660CM10 pH4 3675
CM10 pH4 3747
CM10 pH4 3882CM10 pH4 3925
CM10 pH4 3979
CM10 pH4 4019
CM10 pH4 4076CM10 pH4 4096
CM10 pH4 4116CM10 pH4 4203
CM10 pH4 4236CM10 pH4 4279
CM10 pH4 4337CM10 pH4 4353
CM10 pH4 4424
CM10 pH4 4519CM10 pH4 4571CM10 pH4 4704
CM10 pH4 4717
CM10 pH4 4748CM10 pH4 4761CM10 pH4 4786
CM10 pH4 4832
CM10 pH4 4866
CM10 pH4 4910
CM10 pH4 4935CM10 pH4 4959
CM10 pH4 4976
CM10 pH4 5002
CM10 pH4 5127
CM10 pH4 5165
CM10 pH4 5216
CM10 pH4 5292
CM10 pH4 5374
CM10 pH4 5492CM10 pH4 5524
CM10 pH4 5623
CM10 pH4 5656CM10 pH4 5670
CM10 pH4 5743
CM10 pH4 5757
CM10 pH4 5978CM10 pH4 5997
CM10 pH4 6076
CM10 pH4 6116
CM10 pH4 6142CM10 pH4 6159
CM10 pH4 6187
CM10 pH4 6229CM10 pH4 6257CM10 pH4 6271
CM10 pH4 6303
CM10 pH4 6315CM10 pH4 6345
CM10 pH4 6415
CM10 pH4 6508
CM10 pH4 6581
CM10 pH4 6640
CM10 pH4 6871
CM10 pH4 6912
CM10 pH4 6950CM10 pH4 6994
CM10 pH4 7033CM10 pH4 7060CM10 pH4 7139
CM10 pH4 7179CM10 pH4 7217
CM10 pH4 7503CM10 pH4 8177
CM10 pH4 8436
CM10 pH4 8549
CM10 pH4 8602
CM10 pH4 8666
CM10 pH4 8890
CM10 pH4 9090CM10 pH4 9109
CM10 pH4 9135
CM10 pH4 9150CM10 pH4 9176
CM10 pH4 9348
CM10 pH4 9389CM10 pH4 9592
CM10 pH4 9967CM10 pH4 10070
CM10 pH4 10250 CM10 pH4 10960CM10 pH4 11020
CM10 pH4 11290
CM10 pH4 11700
CM10 pH4 12340CM10 pH4 13820
CM10 pH4 14020CM10 pH4 14770
CM10 pH4 14950
CM10 pH4 16850CM10 pH4 17200CM10 pH4 17790CM10 pH4 18460
CM10 pH6 2523
CM10 pH6 2544
CM10 pH6 2566
CM10 pH6 2592
CM10 pH6 2633
CM10 pH6 2657
CM10 pH6 2677
CM10 pH6 2702CM10 pH6 2722
CM10 pH6 2768CM10 pH6 2792
CM10 pH6 2835
CM10 pH6 2922
CM10 pH6 2971CM10 pH6 2997
CM10 pH6 3042
CM10 pH6 3133CM10 pH6 3150CM10 pH6 3174
CM10 pH6 3209
CM10 pH6 3283
CM10 pH6 3330CM10 pH6 3361
CM10 pH6 3404
CM10 pH6 3437
CM10 pH6 3496CM10 pH6 3553
CM10 pH6 3585
CM10 pH6 3658
CM10 pH6 3814
CM10 pH6 3836CM10 pH6 3882CM10 pH6 3905
CM10 pH6 3924
CM10 pH6 3978CM10 pH6 4021CM10 pH6 4078
CM10 pH6 4110CM10 pH6 4223
CM10 pH6 4245
CM10 pH6 4337
CM10 pH6 4352
CM10 pH6 4380
CM10 pH6 4566
CM10 pH6 4703CM10 pH6 4717
CM10 pH6 4747
CM10 pH6 4762CM10 pH6 4789
CM10 pH6 4911
CM10 pH6 4943
CM10 pH6 5095CM10 pH6 5126
CM10 pH6 5293
CM10 pH6 5373
CM10 pH6 5621
CM10 pH6 5655CM10 pH6 5672CM10 pH6 5742CM10 pH6 5758
CM10 pH6 5980
CM10 pH6 5998CM10 pH6 6076CM10 pH6 6116
CM10 pH6 6160
CM10 pH6 6186CM10 pH6 6230CM10 pH6 6259CM10 pH6 6302
CM10 pH6 6345
CM10 pH6 6373
CM10 pH6 6416
CM10 pH6 6510CM10 pH6 6582
CM10 pH6 6642CM10 pH6 6686
CM10 pH6 6709CM10 pH6 6736
CM10 pH6 6910CM10 pH6 7003
CM10 pH6 7035
CM10 pH6 7062
CM10 pH6 7135CM10 pH6 7177CM10 pH6 7222CM10 pH6 7487
CM10 pH6 8161
CM10 pH6 8539
CM10 pH6 8661
CM10 pH6 9093
CM10 pH6 9122
CM10 pH6 9136
CM10 pH6 9149
CM10 pH6 9343CM10 pH6 9390
CM10 pH6 11000CM10 pH6 11290CM10 pH6 11700
CM10 pH6 12340
CM10 pH6 13810
CM10 pH6 14010
CM10 pH6 14760CM10 pH6 14940CM10 pH6 16960CM10 pH6 17190
CM10 pH6 17780CM10 pH6 18440
CM10 pH8 2522CM10 pH8 2546
CM10 pH8 2611CM10 pH8 2660CM10 pH8 2680
CM10 pH8 2702CM10 pH8 2724CM10 pH8 2746CM10 pH8 2767
CM10 pH8 2874
CM10 pH8 2924
CM10 pH8 2970
CM10 pH8 2992CM10 pH8 3007
CM10 pH8 3036CM10 pH8 3077
CM10 pH8 3150CM10 pH8 3210
CM10 pH8 3283
CM10 pH8 3360
CM10 pH8 3439
CM10 pH8 3501
CM10 pH8 3616CM10 pH8 3659
CM10 pH8 3742
CM10 pH8 3814CM10 pH8 3827 CM10 pH8 3881CM10 pH8 3925
CM10 pH8 3980CM10 pH8 4021
CM10 pH8 4080
CM10 pH8 4097
CM10 pH8 4115
CM10 pH8 4224
CM10 pH8 4245CM10 pH8 4338
CM10 pH8 4354
CM10 pH8 4543
CM10 pH8 4562
CM10 pH8 4704CM10 pH8 4747
CM10 pH8 4762
CM10 pH8 4789
CM10 pH8 4815
CM10 pH8 4913CM10 pH8 4946
CM10 pH8 5216
CM10 pH8 5294CM10 pH8 5374
CM10 pH8 5522
CM10 pH8 5621
CM10 pH8 5657CM10 pH8 5673
CM10 pH8 5744CM10 pH8 5760CM10 pH8 5979
CM10 pH8 5999
CM10 pH8 6056
CM10 pH8 6077CM10 pH8 6118
CM10 pH8 6143
CM10 pH8 6161CM10 pH8 6187
CM10 pH8 6229CM10 pH8 6260
CM10 pH8 6303
CM10 pH8 6346
CM10 pH8 6374
CM10 pH8 6416
CM10 pH8 6511CM10 pH8 6541
CM10 pH8 6582CM10 pH8 6643
CM10 pH8 6687CM10 pH8 6907
CM10 pH8 6951
CM10 pH8 7005CM10 pH8 7037
CM10 pH8 7179CM10 pH8 7488
CM10 pH8 7503
CM10 pH8 8190
CM10 pH8 8494
CM10 pH8 8540
CM10 pH8 8650
CM10 pH8 9078
CM10 pH8 9093
CM10 pH8 9121
CM10 pH8 9135CM10 pH8 9341
CM10 pH8 9390
CM10 pH8 9599
CM10 pH8 10250
CM10 pH8 10960CM10 pH8 11300
CM10 pH8 11670
CM10 pH8 12280CM10 pH8 12340
CM10 pH8 13810CM10 pH8 14010CM10 pH8 14950
CM10 pH8 15990CM10 pH8 16970CM10 pH8 17760CM10 pH8 18440
IMAC 2511
IMAC 2538IMAC 2551IMAC 2561IMAC 2578
IMAC 2594
IMAC 2616
IMAC 2658
IMAC 2682IMAC 2692IMAC 2700
IMAC 2716
IMAC 2736
IMAC 2756
IMAC 2854IMAC 2878
IMAC 2920IMAC 2959IMAC 2990IMAC 3012IMAC 3028
IMAC 3047IMAC 3065IMAC 3077IMAC 3084IMAC 3100IMAC 3155IMAC 3177IMAC 3205IMAC 3356IMAC 3397IMAC 3493IMAC 3615IMAC 3658IMAC 3795
IMAC 3824
IMAC 3882IMAC 3925
IMAC 4058IMAC 4115
IMAC 4225IMAC 4567IMAC 4588IMAC 4705IMAC 4746
IMAC 4843IMAC 4909IMAC 4939
IMAC 4995IMAC 5211IMAC 5289IMAC 5426
IMAC 5523IMAC 5620
IMAC 5790
IMAC 5979IMAC 5998IMAC 6051
IMAC 6116IMAC 6160IMAC 6186IMAC 6230IMAC 6259
IMAC 6302IMAC 6344IMAC 6415IMAC 6579IMAC 6622IMAC 6684IMAC 6709IMAC 6737IMAC 6873IMAC 6928IMAC 6952IMAC 6991
IMAC 7034IMAC 7062
IMAC 7177IMAC 7222
IMAC 7493IMAC 8194
IMAC 8557IMAC 8600
IMAC 9119IMAC 9138IMAC 9166
IMAC 9389IMAC 9595IMAC 9792IMAC 9941IMAC 9978
IMAC 10060
IMAC 10240IMAC 11010
IMAC 11300IMAC 11700IMAC 13830IMAC 14040IMAC 14760IMAC 16860IMAC 16960IMAC 17800
IMAC 18450
$M4.DA1
$M4.DA2
-0.2
-0.1
-0.0
-0.1 0.0 0.1
w*c
[2]
NP20 2506
NP20 3001
NP20 3012NP20 3038NP20 3264NP20 3289
NP20 3743
NP20 3815NP20 3830NP20 3874NP20 3979NP20 4020NP20 4341NP20 4354
NP20 4570
NP20 4705NP20 4719NP20 4748
NP20 4912NP20 5663
NP20 5674NP20 5747
NP20 5761
NP20 6001NP20 6119NP20 6163NP20 6191NP20 6259
NP20 6306
NP20 6420
NP20 6588
NP20 6658
NP20 6925
NP20 7026NP20 7181
NP20 7494
NP20 7507NP20 8500
NP20 9097
NP20 9113
NP20 9123NP20 9137
NP20 9161NP20 9349
NP20 9825
NP20 10270NP20 10790NP20 12350
NP20 13840
NP20 14050NP20 15060NP20 16030
NP20 17810NP20 19090H50 2586
H50 2611
H50 2613
H50 2615H50 2618
H50 2642H50 2672
H50 2688H50 2693
H50 2703
H50 2731H50 2754H50 2787H50 2805H50 2894
H50 2971
H50 2982
H50 3000
H50 3010
H50 3036
H50 3081
H50 3094
H50 3106
H50 3116
H50 3134H50 3186
H50 3204
H50 3217H50 3227H50 3249H50 3261
H50 3289
H50 3345H50 3376H50 3441H50 3469H50 3510
H50 3586H50 3606H50 3625H50 3657
H50 3679H50 3705
H50 3940H50 3967
H50 3989
H50 4005H50 4016
H50 4052
H50 4110H50 4136
H50 4193H50 4221
H50 4238H50 4267H50 4297
H50 4368H50 4525H50 4628H50 4716
H50 4745H50 4843H50 4898H50 4922
H50 4970H50 5021
H50 5062H50 5221H50 5273
H50 5294H50 5314H50 5375H50 5523H50 5579
H50 5635
H50 5789
H50 5825H50 5854H50 5946H50 5993H50 6020 H50 6076H50 6275H50 6377
H50 6538
H50 6783
H50 7027H50 7094
H50 7280H50 8036
H50 8543H50 8586
H50 8747H50 9041
H50 9144H50 10050
H50 10250
H50 10730
H50 11050H50 11150
H50 11300H50 11350
H50 11710
H50 12350
H50 12710
H50 14030
H50 14300H50 15030H50 15400
H50 16860CM10 pH4 2523
CM10 pH4 2542CM10 pH4 2680
CM10 pH4 2701
CM10 pH4 2722CM10 pH4 2882
CM10 pH4 2995
CM10 pH4 3042
CM10 pH4 3149
CM10 pH4 3178
CM10 pH4 3358CM10 pH4 3513CM10 pH4 3584
CM10 pH4 3660CM10 pH4 3675
CM10 pH4 3747
CM10 pH4 3882CM10 pH4 3925
CM10 pH4 3979
CM10 pH4 4019
CM10 pH4 4076CM10 pH4 4096
CM10 pH4 4116CM10 pH4 4203
CM10 pH4 4236CM10 pH4 4279
CM10 pH4 4337CM10 pH4 4353
CM10 pH4 4424
CM10 pH4 4519CM10 pH4 4571CM10 pH4 4704
CM10 pH4 4717
CM10 pH4 4748CM10 pH4 4761CM10 pH4 4786
CM10 pH4 4832
CM10 pH4 4866
CM10 pH4 4910
CM10 pH4 4935CM10 pH4 4959
CM10 pH4 4976
CM10 pH4 5002
CM10 pH4 5127
CM10 pH4 5165
CM10 pH4 5216
CM10 pH4 5292
CM10 pH4 5374
CM10 pH4 5492CM10 pH4 5524
CM10 pH4 5623
CM10 pH4 5656CM10 pH4 5670
CM10 pH4 5743
CM10 pH4 5757
CM10 pH4 5978CM10 pH4 5997
CM10 pH4 6076
CM10 pH4 6116
CM10 pH4 6142CM10 pH4 6159
CM10 pH4 6187
CM10 pH4 6229CM10 pH4 6257CM10 pH4 6271
CM10 pH4 6303
CM10 pH4 6315CM10 pH4 6345
CM10 pH4 6415
CM10 pH4 6508
CM10 pH4 6581
CM10 pH4 6640
CM10 pH4 6871
CM10 pH4 6912
CM10 pH4 6950CM10 pH4 6994
CM10 pH4 7033CM10 pH4 7060CM10 pH4 7139
CM10 pH4 7179CM10 pH4 7217
CM10 pH4 7503CM10 pH4 8177
CM10 pH4 8436
CM10 pH4 8549
CM10 pH4 8602
CM10 pH4 8666
CM10 pH4 8890
CM10 pH4 9090CM10 pH4 9109
CM10 pH4 9135
CM10 pH4 9150CM10 pH4 9176
CM10 pH4 9348
CM10 pH4 9389CM10 pH4 9592
CM10 pH4 9967CM10 pH4 10070
CM10 pH4 10250 CM10 pH4 10960CM10 pH4 11020
CM10 pH4 11290
CM10 pH4 11700
CM10 pH4 12340CM10 pH4 13820
CM10 pH4 14020CM10 pH4 14770
CM10 pH4 14950
CM10 pH4 16850CM10 pH4 17200CM10 pH4 17790CM10 pH4 18460
CM10 pH6 2523
CM10 pH6 2544
CM10 pH6 2566
CM10 pH6 2592
CM10 pH6 2633
CM10 pH6 2657
CM10 pH6 2677
CM10 pH6 2702CM10 pH6 2722
CM10 pH6 2768CM10 pH6 2792
CM10 pH6 2835
CM10 pH6 2922
CM10 pH6 2971CM10 pH6 2997
CM10 pH6 3042
CM10 pH6 3133CM10 pH6 3150CM10 pH6 3174
CM10 pH6 3209
CM10 pH6 3283
CM10 pH6 3330CM10 pH6 3361
CM10 pH6 3404
CM10 pH6 3437
CM10 pH6 3496CM10 pH6 3553
CM10 pH6 3585
CM10 pH6 3658
CM10 pH6 3814
CM10 pH6 3836CM10 pH6 3882CM10 pH6 3905
CM10 pH6 3924
CM10 pH6 3978CM10 pH6 4021CM10 pH6 4078
CM10 pH6 4110CM10 pH6 4223
CM10 pH6 4245
CM10 pH6 4337
CM10 pH6 4352
CM10 pH6 4380
CM10 pH6 4566
CM10 pH6 4703CM10 pH6 4717
CM10 pH6 4747
CM10 pH6 4762CM10 pH6 4789
CM10 pH6 4911
CM10 pH6 4943
CM10 pH6 5095CM10 pH6 5126
CM10 pH6 5293
CM10 pH6 5373
CM10 pH6 5621
CM10 pH6 5655CM10 pH6 5672CM10 pH6 5742CM10 pH6 5758
CM10 pH6 5980
CM10 pH6 5998CM10 pH6 6076CM10 pH6 6116
CM10 pH6 6160
CM10 pH6 6186CM10 pH6 6230CM10 pH6 6259CM10 pH6 6302
CM10 pH6 6345
CM10 pH6 6373
CM10 pH6 6416
CM10 pH6 6510CM10 pH6 6582
CM10 pH6 6642CM10 pH6 6686
CM10 pH6 6709CM10 pH6 6736
CM10 pH6 6910CM10 pH6 7003
CM10 pH6 7035
CM10 pH6 7062
CM10 pH6 7135CM10 pH6 7177CM10 pH6 7222CM10 pH6 7487
CM10 pH6 8161
CM10 pH6 8539
CM10 pH6 8661
CM10 pH6 9093
CM10 pH6 9122
CM10 pH6 9136
CM10 pH6 9149
CM10 pH6 9343CM10 pH6 9390
CM10 pH6 11000CM10 pH6 11290CM10 pH6 11700
CM10 pH6 12340
CM10 pH6 13810
CM10 pH6 14010
CM10 pH6 14760CM10 pH6 14940CM10 pH6 16960CM10 pH6 17190
CM10 pH6 17780CM10 pH6 18440
CM10 pH8 2522CM10 pH8 2546
CM10 pH8 2611CM10 pH8 2660CM10 pH8 2680
CM10 pH8 2702CM10 pH8 2724CM10 pH8 2746CM10 pH8 2767
CM10 pH8 2874
CM10 pH8 2924
CM10 pH8 2970
CM10 pH8 2992CM10 pH8 3007
CM10 pH8 3036CM10 pH8 3077
CM10 pH8 3150CM10 pH8 3210
CM10 pH8 3283
CM10 pH8 3360
CM10 pH8 3439
CM10 pH8 3501
CM10 pH8 3616CM10 pH8 3659
CM10 pH8 3742
CM10 pH8 3814CM10 pH8 3827 CM10 pH8 3881CM10 pH8 3925
CM10 pH8 3980CM10 pH8 4021
CM10 pH8 4080
CM10 pH8 4097
CM10 pH8 4115
CM10 pH8 4224
CM10 pH8 4245CM10 pH8 4338
CM10 pH8 4354
CM10 pH8 4543
CM10 pH8 4562
CM10 pH8 4704CM10 pH8 4747
CM10 pH8 4762
CM10 pH8 4789
CM10 pH8 4815
CM10 pH8 4913CM10 pH8 4946
CM10 pH8 5216
CM10 pH8 5294CM10 pH8 5374
CM10 pH8 5522
CM10 pH8 5621
CM10 pH8 5657CM10 pH8 5673
CM10 pH8 5744CM10 pH8 5760CM10 pH8 5979
CM10 pH8 5999
CM10 pH8 6056
CM10 pH8 6077CM10 pH8 6118
CM10 pH8 6143
CM10 pH8 6161CM10 pH8 6187
CM10 pH8 6229CM10 pH8 6260
CM10 pH8 6303
CM10 pH8 6346
CM10 pH8 6374
CM10 pH8 6416
CM10 pH8 6511CM10 pH8 6541
CM10 pH8 6582CM10 pH8 6643
CM10 pH8 6687CM10 pH8 6907
CM10 pH8 6951
CM10 pH8 7005CM10 pH8 7037
CM10 pH8 7179CM10 pH8 7488
CM10 pH8 7503
CM10 pH8 8190
CM10 pH8 8494
CM10 pH8 8540
CM10 pH8 8650
CM10 pH8 9078
CM10 pH8 9093
CM10 pH8 9121
CM10 pH8 9135CM10 pH8 9341
CM10 pH8 9390
CM10 pH8 9599
CM10 pH8 10250
CM10 pH8 10960CM10 pH8 11300
CM10 pH8 11670
CM10 pH8 12280CM10 pH8 12340
CM10 pH8 13810CM10 pH8 14010CM10 pH8 14950
CM10 pH8 15990CM10 pH8 16970CM10 pH8 17760CM10 pH8 18440
IMAC 2511
IMAC 2538IMAC 2551IMAC 2561IMAC 2578
IMAC 2594
IMAC 2616
IMAC 2658
IMAC 2682IMAC 2692IMAC 2700
IMAC 2716
IMAC 2736
IMAC 2756
IMAC 2854IMAC 2878
IMAC 2920IMAC 2959IMAC 2990IMAC 3012IMAC 3028
IMAC 3047IMAC 3065IMAC 3077IMAC 3084IMAC 3100IMAC 3155IMAC 3177IMAC 3205IMAC 3356IMAC 3397IMAC 3493IMAC 3615IMAC 3658IMAC 3795
IMAC 3824
IMAC 3882IMAC 3925
IMAC 4058IMAC 4115
IMAC 4225IMAC 4567IMAC 4588IMAC 4705IMAC 4746
IMAC 4843IMAC 4909IMAC 4939
IMAC 4995IMAC 5211IMAC 5289IMAC 5426
IMAC 5523IMAC 5620
IMAC 5790
IMAC 5979IMAC 5998IMAC 6051
IMAC 6116IMAC 6160IMAC 6186IMAC 6230IMAC 6259
IMAC 6302IMAC 6344IMAC 6415IMAC 6579IMAC 6622IMAC 6684IMAC 6709IMAC 6737IMAC 6873IMAC 6928IMAC 6952IMAC 6991
IMAC 7034IMAC 7062
IMAC 7177IMAC 7222
IMAC 7493IMAC 8194
IMAC 8557IMAC 8600
IMAC 9119IMAC 9138IMAC 9166
IMAC 9389IMAC 9595IMAC 9792IMAC 9941IMAC 9978
IMAC 10060
IMAC 10240IMAC 11010
IMAC 11300IMAC 11700IMAC 13830IMAC 14040IMAC 14760IMAC 16860IMAC 16960IMAC 17800
IMAC 18450
$M4.DA1
$M4.DA2
Figure 37: Partial least squares discriminant analysis (PLS-DA) comparing control and treated myoblasts Cultured control and treated myoblasts (section 2.1) were collected and spectra
generated following SELDI-TOF-MS analysis (section 2.6) were evaluated using PLS-
DA. a) Separation between control and TMPD-treated myoblasts is shown in the
scores plot following the supervised analysis. b) The loadings plot is shown for
illustrative purposes and 85 ions had a VIP coefficient of >1 which suggests that the
variables were important in the model. c) Separation between control and GWδ-treated
myoblasts is shown in the scores plot. b) Again, the loadings plot is shown for
illustrative purposes and, in this case, 95 ions had a VIP coefficient of >1. This data
demonstrated the difficulty in discovering the most influential ions when all 592 peaks
detected on ProteinChip® arrays were included in the analysis.
Chapter 3: Results
111
ControlTreatedControlTreatedc) Scores plot (TMPD myotubes) x
yd) Loadings plot (TMPD myotubes)
ControlTreatedControlTreateda) Scores plot (TMPD myotubes) x
yb) Loadings plot (TMPD myotubes)
Principal component 1 (PC1)
PC
2
Principal component 1 (PC1)
PC
2
-20
-10
0
10
20
-30 -20 -10 0 10 20
-20
-10
0
10
20
-30 -20 -10 0 10 20
-0.1
0.0
0.1
0.2
-0.1 0.0 0.1 0.2
w*c
[2]
NP20 3002NP20 3012NP20 3037NP20 3263NP20 3814NP20 3829
NP20 4020
NP20 4340NP20 4356NP20 5753
NP20 6305
NP20 6588
NP20 7027
NP20 9138NP20 9825
NP20 10280
NP20 11340
NP20 12360
NP20 13850NP20 14050
NP20 15070NP20 16040NP20 17830NP20 19090H50 2504H50 2572H50 2599
H50 2615H50 2643H50 2678H50 2703H50 2970
H50 2999
H50 3009
H50 3035H50 3094H50 3106H50 3133
H50 3187H50 3204
H50 3218H50 3248H50 3260H50 3290
H50 3345H50 3357H50 3388H50 3509H50 3600H50 3624H50 3654H50 3702H50 3940 H50 3987H50 4004
H50 4015
H50 4049H50 4111H50 4146
H50 4179H50 4194H50 4207H50 4221H50 4238H50 4263
H50 4300H50 4503H50 4628H50 4902H50 4966
H50 5019H50 5064H50 5116
H50 5156H50 5212H50 5269H50 5526H50 5633H50 5851
H50 5946
H50 6024H50 6076H50 6109H50 6271H50 7028
H50 7123H50 7279H50 8033H50 8121H50 9036H50 9779H50 10040H50 10270H50 11040H50 11300H50 11700H50 12710H50 13840H50 14030H50 16820H50 18910
CM10 pH4 2537CM10 pH4 2609CM10 pH4 2657
CM10 pH4 2700
CM10 pH4 2717CM10 pH4 2761
CM10 pH4 2862CM10 pH4 2922
CM10 pH4 3044
CM10 pH4 3182
CM10 pH4 3363CM10 pH4 3437
CM10 pH4 3481
CM10 pH4 3584CM10 pH4 4110
CM10 pH4 4337
CM10 pH4 4354
CM10 pH4 4932
CM10 pH4 4959
CM10 pH4 4975
CM10 pH4 5002
CM10 pH4 5138CM10 pH4 5165
CM10 pH4 5298
CM10 pH4 5392
CM10 pH4 5537
CM10 pH4 5655CM10 pH4 5743
CM10 pH4 6230
CM10 pH4 6272CM10 pH4 6579
CM10 pH4 6914
CM10 pH4 7486
CM10 pH4 8176
CM10 pH4 8600CM10 pH4 8641
CM10 pH4 9165
CM10 pH4 9972
CM10 pH4 10250
CM10 pH4 11010CM10 pH4 11290CM10 pH4 11700
CM10 pH4 12340
CM10 pH4 13820CM10 pH4 14020CM10 pH4 14760
CM10 pH4 15800CM10 pH4 17800CM10 pH4 18490CM10 pH4 18920
CM10 pH6 2522CM10 pH6 2640CM10 pH6 2700
CM10 pH6 2723
CM10 pH6 2870
CM10 pH6 2883
CM10 pH6 2921CM10 pH6 2988
CM10 pH6 3044
CM10 pH6 3158
CM10 pH6 3178
CM10 pH6 3358CM10 pH6 3401
CM10 pH6 3494
CM10 pH6 3658
CM10 pH6 3742CM10 pH6 3813CM10 pH6 3977
CM10 pH6 4081
CM10 pH6 4131
CM10 pH6 4225CM10 pH6 4253
CM10 pH6 4337CM10 pH6 4353
CM10 pH6 4379
CM10 pH6 4542
CM10 pH6 4746
CM10 pH6 4815CM10 pH6 4911CM10 pH6 5141
CM10 pH6 5375CM10 pH6 5532
CM10 pH6 5655CM10 pH6 5671CM10 pH6 5742
CM10 pH6 5758
CM10 pH6 5978
CM10 pH6 6270CM10 pH6 6313
CM10 pH6 6578CM10 pH6 6642CM10 pH6 6685CM10 pH6 6907
CM10 pH6 7015
CM10 pH6 7486CM10 pH6 7502
CM10 pH6 8158CM10 pH6 8467CM10 pH6 8535
CM10 pH6 8888
CM10 pH6 9126
CM10 pH6 9389CM10 pH6 9596
CM10 pH6 9967
CM10 pH6 10260
CM10 pH6 11290
CM10 pH6 12340
CM10 pH6 13810CM10 pH6 14010CM10 pH6 17780CM10 pH6 18450
CM10 pH8 2522
CM10 pH8 2546
CM10 pH8 2588CM10 pH8 2662
CM10 pH8 2701CM10 pH8 2723CM10 pH8 2809
CM10 pH8 2869
CM10 pH8 2882CM10 pH8 2927
CM10 pH8 2969
CM10 pH8 2987
CM10 pH8 3043
CM10 pH8 3077CM10 pH8 3209CM10 pH8 3364
CM10 pH8 3382
CM10 pH8 3438
CM10 pH8 3491
CM10 pH8 3659
CM10 pH8 3741CM10 pH8 3820
CM10 pH8 3978
CM10 pH8 4081
CM10 pH8 4096
CM10 pH8 4338
CM10 pH8 4354
CM10 pH8 4379
CM10 pH8 4433CM10 pH8 4450
CM10 pH8 4542
CM10 pH8 4747CM10 pH8 5283CM10 pH8 5376
CM10 pH8 5657CM10 pH8 5673
CM10 pH8 5744 CM10 pH8 5760
CM10 pH8 5788
CM10 pH8 5978
CM10 pH8 6175
CM10 pH8 6231CM10 pH8 6272CM10 pH8 6314
CM10 pH8 6580
CM10 pH8 6643CM10 pH8 6686
CM10 pH8 6913
CM10 pH8 7017CM10 pH8 7488
CM10 pH8 7502
CM10 pH8 8179
CM10 pH8 8469
CM10 pH8 9120
CM10 pH8 9158CM10 pH8 9392CM10 pH8 9596
CM10 pH8 10260CM10 pH8 11300
CM10 pH8 12330CM10 pH8 13820
CM10 pH8 14020
CM10 pH8 15690CM10 pH8 16960CM10 pH8 17620
CM10 pH8 17780
IMAC 2552IMAC 2646
IMAC 2657
IMAC 2695
IMAC 2718
IMAC 2923
IMAC 2985IMAC 3006
IMAC 3042IMAC 3157IMAC 3177
IMAC 3366IMAC 3437
IMAC 3808IMAC 3823
IMAC 3974IMAC 4130
IMAC 4253
IMAC 4401
IMAC 4595
IMAC 4842
IMAC 4955
IMAC 5080IMAC 5274IMAC 5298
IMAC 5536
IMAC 5579
IMAC 5653IMAC 5739
IMAC 6228IMAC 6269IMAC 6312IMAC 6578
IMAC 6829
IMAC 6921IMAC 7493
IMAC 8194IMAC 8602IMAC 8643
IMAC 9157
IMAC 9390
IMAC 9711IMAC 9974
IMAC 10240IMAC 11020
IMAC 11290
IMAC 11700
IMAC 12340
IMAC 12700IMAC 13830IMAC 14020
IMAC 14760
IMAC 17810$M3.DA1
$M3.DA2
-0.1
0.0
0.1
0.2
-0.1 0.0 0.1 0.2
w*c
[2]
NP20 3002NP20 3012NP20 3037NP20 3263NP20 3814NP20 3829
NP20 4020
NP20 4340NP20 4356NP20 5753
NP20 6305
NP20 6588
NP20 7027
NP20 9138NP20 9825
NP20 10280
NP20 11340
NP20 12360
NP20 13850NP20 14050
NP20 15070NP20 16040NP20 17830NP20 19090H50 2504H50 2572H50 2599
H50 2615H50 2643H50 2678H50 2703H50 2970
H50 2999
H50 3009
H50 3035H50 3094H50 3106H50 3133
H50 3187H50 3204
H50 3218H50 3248H50 3260H50 3290
H50 3345H50 3357H50 3388H50 3509H50 3600H50 3624H50 3654H50 3702H50 3940 H50 3987H50 4004
H50 4015
H50 4049H50 4111H50 4146
H50 4179H50 4194H50 4207H50 4221H50 4238H50 4263
H50 4300H50 4503H50 4628H50 4902H50 4966
H50 5019H50 5064H50 5116
H50 5156H50 5212H50 5269H50 5526H50 5633H50 5851
H50 5946
H50 6024H50 6076H50 6109H50 6271H50 7028
H50 7123H50 7279H50 8033H50 8121H50 9036H50 9779H50 10040H50 10270H50 11040H50 11300H50 11700H50 12710H50 13840H50 14030H50 16820H50 18910
CM10 pH4 2537CM10 pH4 2609CM10 pH4 2657
CM10 pH4 2700
CM10 pH4 2717CM10 pH4 2761
CM10 pH4 2862CM10 pH4 2922
CM10 pH4 3044
CM10 pH4 3182
CM10 pH4 3363CM10 pH4 3437
CM10 pH4 3481
CM10 pH4 3584CM10 pH4 4110
CM10 pH4 4337
CM10 pH4 4354
CM10 pH4 4932
CM10 pH4 4959
CM10 pH4 4975
CM10 pH4 5002
CM10 pH4 5138CM10 pH4 5165
CM10 pH4 5298
CM10 pH4 5392
CM10 pH4 5537
CM10 pH4 5655CM10 pH4 5743
CM10 pH4 6230
CM10 pH4 6272CM10 pH4 6579
CM10 pH4 6914
CM10 pH4 7486
CM10 pH4 8176
CM10 pH4 8600CM10 pH4 8641
CM10 pH4 9165
CM10 pH4 9972
CM10 pH4 10250
CM10 pH4 11010CM10 pH4 11290CM10 pH4 11700
CM10 pH4 12340
CM10 pH4 13820CM10 pH4 14020CM10 pH4 14760
CM10 pH4 15800CM10 pH4 17800CM10 pH4 18490CM10 pH4 18920
CM10 pH6 2522CM10 pH6 2640CM10 pH6 2700
CM10 pH6 2723
CM10 pH6 2870
CM10 pH6 2883
CM10 pH6 2921CM10 pH6 2988
CM10 pH6 3044
CM10 pH6 3158
CM10 pH6 3178
CM10 pH6 3358CM10 pH6 3401
CM10 pH6 3494
CM10 pH6 3658
CM10 pH6 3742CM10 pH6 3813CM10 pH6 3977
CM10 pH6 4081
CM10 pH6 4131
CM10 pH6 4225CM10 pH6 4253
CM10 pH6 4337CM10 pH6 4353
CM10 pH6 4379
CM10 pH6 4542
CM10 pH6 4746
CM10 pH6 4815CM10 pH6 4911CM10 pH6 5141
CM10 pH6 5375CM10 pH6 5532
CM10 pH6 5655CM10 pH6 5671CM10 pH6 5742
CM10 pH6 5758
CM10 pH6 5978
CM10 pH6 6270CM10 pH6 6313
CM10 pH6 6578CM10 pH6 6642CM10 pH6 6685CM10 pH6 6907
CM10 pH6 7015
CM10 pH6 7486CM10 pH6 7502
CM10 pH6 8158CM10 pH6 8467CM10 pH6 8535
CM10 pH6 8888
CM10 pH6 9126
CM10 pH6 9389CM10 pH6 9596
CM10 pH6 9967
CM10 pH6 10260
CM10 pH6 11290
CM10 pH6 12340
CM10 pH6 13810CM10 pH6 14010CM10 pH6 17780CM10 pH6 18450
CM10 pH8 2522
CM10 pH8 2546
CM10 pH8 2588CM10 pH8 2662
CM10 pH8 2701CM10 pH8 2723CM10 pH8 2809
CM10 pH8 2869
CM10 pH8 2882CM10 pH8 2927
CM10 pH8 2969
CM10 pH8 2987
CM10 pH8 3043
CM10 pH8 3077CM10 pH8 3209CM10 pH8 3364
CM10 pH8 3382
CM10 pH8 3438
CM10 pH8 3491
CM10 pH8 3659
CM10 pH8 3741CM10 pH8 3820
CM10 pH8 3978
CM10 pH8 4081
CM10 pH8 4096
CM10 pH8 4338
CM10 pH8 4354
CM10 pH8 4379
CM10 pH8 4433CM10 pH8 4450
CM10 pH8 4542
CM10 pH8 4747CM10 pH8 5283CM10 pH8 5376
CM10 pH8 5657CM10 pH8 5673
CM10 pH8 5744 CM10 pH8 5760
CM10 pH8 5788
CM10 pH8 5978
CM10 pH8 6175
CM10 pH8 6231CM10 pH8 6272CM10 pH8 6314
CM10 pH8 6580
CM10 pH8 6643CM10 pH8 6686
CM10 pH8 6913
CM10 pH8 7017CM10 pH8 7488
CM10 pH8 7502
CM10 pH8 8179
CM10 pH8 8469
CM10 pH8 9120
CM10 pH8 9158CM10 pH8 9392CM10 pH8 9596
CM10 pH8 10260CM10 pH8 11300
CM10 pH8 12330CM10 pH8 13820
CM10 pH8 14020
CM10 pH8 15690CM10 pH8 16960CM10 pH8 17620
CM10 pH8 17780
IMAC 2552IMAC 2646
IMAC 2657
IMAC 2695
IMAC 2718
IMAC 2923
IMAC 2985IMAC 3006
IMAC 3042IMAC 3157IMAC 3177
IMAC 3366IMAC 3437
IMAC 3808IMAC 3823
IMAC 3974IMAC 4130
IMAC 4253
IMAC 4401
IMAC 4595
IMAC 4842
IMAC 4955
IMAC 5080IMAC 5274IMAC 5298
IMAC 5536
IMAC 5579
IMAC 5653IMAC 5739
IMAC 6228IMAC 6269IMAC 6312IMAC 6578
IMAC 6829
IMAC 6921IMAC 7493
IMAC 8194IMAC 8602IMAC 8643
IMAC 9157
IMAC 9390
IMAC 9711IMAC 9974
IMAC 10240IMAC 11020
IMAC 11290
IMAC 11700
IMAC 12340
IMAC 12700IMAC 13830IMAC 14020
IMAC 14760
IMAC 17810$M3.DA1
$M3.DA2
-20
-10
0
10
20
-20 -10 0 10 20
-20
-10
0
10
20
-20 -10 0 10 20
-0.1
0.0
0.1
0.2
-0.2 -0.1 -0.0 0.1
w*c
[2]
NP20 3002
NP20 3012
NP20 3037NP20 3263
NP20 3814NP20 3829
NP20 4020
NP20 4340
NP20 4356
NP20 5753
NP20 6305
NP20 6588
NP20 7027
NP20 9138
NP20 9825
NP20 10280
NP20 11340
NP20 12360NP20 13850
NP20 14050NP20 15070
NP20 16040NP20 17830NP20 19090
H50 2504H50 2572H50 2599H50 2615
H50 2643
H50 2678H50 2703
H50 2970
H50 2999
H50 3009
H50 3035
H50 3094
H50 3106
H50 3133
H50 3187
H50 3204H50 3218H50 3248
H50 3260H50 3290H50 3345H50 3357H50 3388
H50 3509H50 3600
H50 3624
H50 3654H50 3702H50 3940
H50 3987
H50 4004H50 4015
H50 4049
H50 4111
H50 4146H50 4179
H50 4194
H50 4207H50 4221
H50 4238H50 4263
H50 4300
H50 4503H50 4628H50 4902
H50 4966
H50 5019H50 5064
H50 5116H50 5156H50 5212
H50 5269H50 5526
H50 5633
H50 5851H50 5946
H50 6024
H50 6076
H50 6109
H50 6271H50 7028H50 7123
H50 7279
H50 8033
H50 8121
H50 9036H50 9779H50 10040H50 10270
H50 11040H50 11300
H50 11700
H50 12710H50 13840H50 14030H50 16820H50 18910
CM10 pH4 2537CM10 pH4 2609
CM10 pH4 2657
CM10 pH4 2700
CM10 pH4 2717
CM10 pH4 2761CM10 pH4 2862CM10 pH4 2922
CM10 pH4 3044
CM10 pH4 3182CM10 pH4 3363CM10 pH4 3437
CM10 pH4 3481CM10 pH4 3584
CM10 pH4 4110
CM10 pH4 4337CM10 pH4 4354
CM10 pH4 4932CM10 pH4 4959
CM10 pH4 4975
CM10 pH4 5002
CM10 pH4 5138
CM10 pH4 5165
CM10 pH4 5298
CM10 pH4 5392
CM10 pH4 5537
CM10 pH4 5655CM10 pH4 5743
CM10 pH4 6230
CM10 pH4 6272
CM10 pH4 6579
CM10 pH4 6914
CM10 pH4 7486CM10 pH4 8176 CM10 pH4 8600
CM10 pH4 8641
CM10 pH4 9165CM10 pH4 9972
CM10 pH4 10250
CM10 pH4 11010
CM10 pH4 11290
CM10 pH4 11700
CM10 pH4 12340CM10 pH4 13820
CM10 pH4 14020
CM10 pH4 14760CM10 pH4 15800
CM10 pH4 17800
CM10 pH4 18490CM10 pH4 18920
CM10 pH6 2522
CM10 pH6 2640
CM10 pH6 2700CM10 pH6 2723CM10 pH6 2870CM10 pH6 2883
CM10 pH6 2921
CM10 pH6 2988
CM10 pH6 3044
CM10 pH6 3158
CM10 pH6 3178
CM10 pH6 3358CM10 pH6 3401CM10 pH6 3494
CM10 pH6 3658
CM10 pH6 3742CM10 pH6 3813
CM10 pH6 3977
CM10 pH6 4081
CM10 pH6 4131
CM10 pH6 4225
CM10 pH6 4253
CM10 pH6 4337
CM10 pH6 4353
CM10 pH6 4379CM10 pH6 4542
CM10 pH6 4746
CM10 pH6 4815
CM10 pH6 4911
CM10 pH6 5141
CM10 pH6 5375
CM10 pH6 5532
CM10 pH6 5655CM10 pH6 5671
CM10 pH6 5742
CM10 pH6 5758
CM10 pH6 5978CM10 pH6 6270CM10 pH6 6313CM10 pH6 6578
CM10 pH6 6642CM10 pH6 6685
CM10 pH6 6907
CM10 pH6 7015
CM10 pH6 7486
CM10 pH6 7502CM10 pH6 8158
CM10 pH6 8467
CM10 pH6 8535
CM10 pH6 8888CM10 pH6 9126
CM10 pH6 9389
CM10 pH6 9596
CM10 pH6 9967
CM10 pH6 10260
CM10 pH6 11290CM10 pH6 12340
CM10 pH6 13810
CM10 pH6 14010
CM10 pH6 17780
CM10 pH6 18450CM10 pH8 2522
CM10 pH8 2546CM10 pH8 2588
CM10 pH8 2662
CM10 pH8 2701
CM10 pH8 2723
CM10 pH8 2809
CM10 pH8 2869
CM10 pH8 2882
CM10 pH8 2927
CM10 pH8 2969
CM10 pH8 2987
CM10 pH8 3043
CM10 pH8 3077
CM10 pH8 3209
CM10 pH8 3364
CM10 pH8 3382
CM10 pH8 3438
CM10 pH8 3491
CM10 pH8 3659
CM10 pH8 3741
CM10 pH8 3820CM10 pH8 3978
CM10 pH8 4081
CM10 pH8 4096
CM10 pH8 4338
CM10 pH8 4354
CM10 pH8 4379
CM10 pH8 4433CM10 pH8 4450
CM10 pH8 4542CM10 pH8 4747
CM10 pH8 5283
CM10 pH8 5376
CM10 pH8 5657
CM10 pH8 5673CM10 pH8 5744
CM10 pH8 5760
CM10 pH8 5788
CM10 pH8 5978
CM10 pH8 6175CM10 pH8 6231
CM10 pH8 6272
CM10 pH8 6314CM10 pH8 6580
CM10 pH8 6643
CM10 pH8 6686
CM10 pH8 6913CM10 pH8 7017
CM10 pH8 7488
CM10 pH8 7502
CM10 pH8 8179
CM10 pH8 8469
CM10 pH8 9120CM10 pH8 9158CM10 pH8 9392
CM10 pH8 9596
CM10 pH8 10260
CM10 pH8 11300
CM10 pH8 12330CM10 pH8 13820
CM10 pH8 14020
CM10 pH8 15690CM10 pH8 16960CM10 pH8 17620
CM10 pH8 17780IMAC 2552IMAC 2646
IMAC 2657
IMAC 2695
IMAC 2718
IMAC 2923IMAC 2985IMAC 3006IMAC 3042IMAC 3157IMAC 3177IMAC 3366IMAC 3437IMAC 3808IMAC 3823IMAC 3974IMAC 4130IMAC 4253IMAC 4401
IMAC 4595IMAC 4842
IMAC 4955IMAC 5080
IMAC 5274
IMAC 5298
IMAC 5536
IMAC 5579IMAC 5653
IMAC 5739IMAC 6228IMAC 6269IMAC 6312
IMAC 6578
IMAC 6829
IMAC 6921
IMAC 7493IMAC 8194
IMAC 8602
IMAC 8643
IMAC 9157IMAC 9390
IMAC 9711IMAC 9974
IMAC 10240IMAC 11020IMAC 11290
IMAC 11700
IMAC 12340
IMAC 12700IMAC 13830
IMAC 14020
IMAC 14760IMAC 17810
$M4.DA1
$M4.DA2
-0.1
0.0
0.1
0.2
-0.2 -0.1 -0.0 0.1
w*c
[2]
NP20 3002
NP20 3012
NP20 3037NP20 3263
NP20 3814NP20 3829
NP20 4020
NP20 4340
NP20 4356
NP20 5753
NP20 6305
NP20 6588
NP20 7027
NP20 9138
NP20 9825
NP20 10280
NP20 11340
NP20 12360NP20 13850
NP20 14050NP20 15070
NP20 16040NP20 17830NP20 19090
H50 2504H50 2572H50 2599H50 2615
H50 2643
H50 2678H50 2703
H50 2970
H50 2999
H50 3009
H50 3035
H50 3094
H50 3106
H50 3133
H50 3187
H50 3204H50 3218H50 3248
H50 3260H50 3290H50 3345H50 3357H50 3388
H50 3509H50 3600
H50 3624
H50 3654H50 3702H50 3940
H50 3987
H50 4004H50 4015
H50 4049
H50 4111
H50 4146H50 4179
H50 4194
H50 4207H50 4221
H50 4238H50 4263
H50 4300
H50 4503H50 4628H50 4902
H50 4966
H50 5019H50 5064
H50 5116H50 5156H50 5212
H50 5269H50 5526
H50 5633
H50 5851H50 5946
H50 6024
H50 6076
H50 6109
H50 6271H50 7028H50 7123
H50 7279
H50 8033
H50 8121
H50 9036H50 9779H50 10040H50 10270
H50 11040H50 11300
H50 11700
H50 12710H50 13840H50 14030H50 16820H50 18910
CM10 pH4 2537CM10 pH4 2609
CM10 pH4 2657
CM10 pH4 2700
CM10 pH4 2717
CM10 pH4 2761CM10 pH4 2862CM10 pH4 2922
CM10 pH4 3044
CM10 pH4 3182CM10 pH4 3363CM10 pH4 3437
CM10 pH4 3481CM10 pH4 3584
CM10 pH4 4110
CM10 pH4 4337CM10 pH4 4354
CM10 pH4 4932CM10 pH4 4959
CM10 pH4 4975
CM10 pH4 5002
CM10 pH4 5138
CM10 pH4 5165
CM10 pH4 5298
CM10 pH4 5392
CM10 pH4 5537
CM10 pH4 5655CM10 pH4 5743
CM10 pH4 6230
CM10 pH4 6272
CM10 pH4 6579
CM10 pH4 6914
CM10 pH4 7486CM10 pH4 8176 CM10 pH4 8600
CM10 pH4 8641
CM10 pH4 9165CM10 pH4 9972
CM10 pH4 10250
CM10 pH4 11010
CM10 pH4 11290
CM10 pH4 11700
CM10 pH4 12340CM10 pH4 13820
CM10 pH4 14020
CM10 pH4 14760CM10 pH4 15800
CM10 pH4 17800
CM10 pH4 18490CM10 pH4 18920
CM10 pH6 2522
CM10 pH6 2640
CM10 pH6 2700CM10 pH6 2723CM10 pH6 2870CM10 pH6 2883
CM10 pH6 2921
CM10 pH6 2988
CM10 pH6 3044
CM10 pH6 3158
CM10 pH6 3178
CM10 pH6 3358CM10 pH6 3401CM10 pH6 3494
CM10 pH6 3658
CM10 pH6 3742CM10 pH6 3813
CM10 pH6 3977
CM10 pH6 4081
CM10 pH6 4131
CM10 pH6 4225
CM10 pH6 4253
CM10 pH6 4337
CM10 pH6 4353
CM10 pH6 4379CM10 pH6 4542
CM10 pH6 4746
CM10 pH6 4815
CM10 pH6 4911
CM10 pH6 5141
CM10 pH6 5375
CM10 pH6 5532
CM10 pH6 5655CM10 pH6 5671
CM10 pH6 5742
CM10 pH6 5758
CM10 pH6 5978CM10 pH6 6270CM10 pH6 6313CM10 pH6 6578
CM10 pH6 6642CM10 pH6 6685
CM10 pH6 6907
CM10 pH6 7015
CM10 pH6 7486
CM10 pH6 7502CM10 pH6 8158
CM10 pH6 8467
CM10 pH6 8535
CM10 pH6 8888CM10 pH6 9126
CM10 pH6 9389
CM10 pH6 9596
CM10 pH6 9967
CM10 pH6 10260
CM10 pH6 11290CM10 pH6 12340
CM10 pH6 13810
CM10 pH6 14010
CM10 pH6 17780
CM10 pH6 18450CM10 pH8 2522
CM10 pH8 2546CM10 pH8 2588
CM10 pH8 2662
CM10 pH8 2701
CM10 pH8 2723
CM10 pH8 2809
CM10 pH8 2869
CM10 pH8 2882
CM10 pH8 2927
CM10 pH8 2969
CM10 pH8 2987
CM10 pH8 3043
CM10 pH8 3077
CM10 pH8 3209
CM10 pH8 3364
CM10 pH8 3382
CM10 pH8 3438
CM10 pH8 3491
CM10 pH8 3659
CM10 pH8 3741
CM10 pH8 3820CM10 pH8 3978
CM10 pH8 4081
CM10 pH8 4096
CM10 pH8 4338
CM10 pH8 4354
CM10 pH8 4379
CM10 pH8 4433CM10 pH8 4450
CM10 pH8 4542CM10 pH8 4747
CM10 pH8 5283
CM10 pH8 5376
CM10 pH8 5657
CM10 pH8 5673CM10 pH8 5744
CM10 pH8 5760
CM10 pH8 5788
CM10 pH8 5978
CM10 pH8 6175CM10 pH8 6231
CM10 pH8 6272
CM10 pH8 6314CM10 pH8 6580
CM10 pH8 6643
CM10 pH8 6686
CM10 pH8 6913CM10 pH8 7017
CM10 pH8 7488
CM10 pH8 7502
CM10 pH8 8179
CM10 pH8 8469
CM10 pH8 9120CM10 pH8 9158CM10 pH8 9392
CM10 pH8 9596
CM10 pH8 10260
CM10 pH8 11300
CM10 pH8 12330CM10 pH8 13820
CM10 pH8 14020
CM10 pH8 15690CM10 pH8 16960CM10 pH8 17620
CM10 pH8 17780IMAC 2552IMAC 2646
IMAC 2657
IMAC 2695
IMAC 2718
IMAC 2923IMAC 2985IMAC 3006IMAC 3042IMAC 3157IMAC 3177IMAC 3366IMAC 3437IMAC 3808IMAC 3823IMAC 3974IMAC 4130IMAC 4253IMAC 4401
IMAC 4595IMAC 4842
IMAC 4955IMAC 5080
IMAC 5274
IMAC 5298
IMAC 5536
IMAC 5579IMAC 5653
IMAC 5739IMAC 6228IMAC 6269IMAC 6312
IMAC 6578
IMAC 6829
IMAC 6921
IMAC 7493IMAC 8194
IMAC 8602
IMAC 8643
IMAC 9157IMAC 9390
IMAC 9711IMAC 9974
IMAC 10240IMAC 11020IMAC 11290
IMAC 11700
IMAC 12340
IMAC 12700IMAC 13830
IMAC 14020
IMAC 14760IMAC 17810
$M4.DA1
$M4.DA2
Figure 38: Partial least squares discriminant analysis (PLS-DA) comparing control and treated myotubes Cultured control and treated myotubes (section 2.1) were collected and spectra
generated following SELDI-TOF-MS analysis (section 2.6) were evaluated using PLS-
DA. a) Separation between control and TMPD-treated myotubes is shown in the scores
plot following the supervised analysis. b) The loadings plot is shown for illustrative
purposes only and 87 ions had a VIP coefficient of >1 which suggests that the
variables were important in the model. c) Separation between control and GWδ-treated
myotubes is shown in the scores plot. b) Again, the loadings plot is shown for
illustrative purposes and, in this case, 97 ions had a VIP coefficient of >1. When all 331
peaks detected on ProteinChip® arrays were included in the analysis, it was difficult to
determine the most influential ions.
Chapter 3: Results
112
These results demonstrated that the treatment of myoblasts and myotubes with
TMPD or GWδ resulted in changes in their pattern of protein expression. This
was demonstrated by PCA of data generated following the SELDI-TOF-MS
analysis of skeletal muscle cells on a variety of ProteinChip® arrays. Although
PLS-DA showed clear separation between control and treated samples, it was
not an appropriate method for determining the main ions that changed in
response to treatment, due to the high number of variables entered into the
analyses. The loadings plots generated resulted in the discovery of a high
number of ions that could be considered as important in the models and the
inclusion of all the ions detected may have lead to the inclusion of some
unsuitable protein ions. Because of this, it was decided to focus on the levels of
specific ions and determine whether individual ions where responsive to
treatment using a univariate approach, rather than the multivariate one
described here.
3.4.3 Selection of TMPD or GWδ responsive protein ions in skeletal
muscle cells
Changes in the expression of specific protein ions were examined in order to
determine whether their levels were changed significantly in response to
treatment with either TMPD or GWδ (Student’s t-test, p<0.05). SELDI-TOF-MS
data from the same SELDI-TOF-MS experiment run described in section 3.4.2
was examined and consideration was also given to the possibility of false
discoveries, where a number of false positives may have occurred by chance.
In order to address this, a correction for multiple testing was applied in the form
of a Benjamini-Hochberg false discovery rate (BH-FDR)293. Another commonly
used method for correcting for an expected number of false discoveries was
also applied in the form of a Bonferroni correction294. Ions that were changed in
response to treatment here were validated in an independent repeat SELDI-
TOF-MS experiment run.
Chapter 3: Results
113
When changes in the levels of protein ions in myoblasts were examined, these
methods resulted in the exclusion of a large proportion of the responsive protein
ions initially discovered. Despite this, 3 protein ions (1 ion at m/z 5743 on CM10
arrays at pH 8 and 2 ions at m/z 8557 and 9389 on IMAC arrays) were
significantly changed in response to TMPD treatment following correction with
the milder form of correction (BH-FDR). The ion at m/z 8557 was also included
as being responsive to TMPD treatment following the application of the
Bonferroni correction (Table 4). None of the ions were included when GWδ
treatment was considered. However, in the repeat SELDI-TOF-MS experiment
run (No. 2), only the levels of the ion at m/z 5743 were significantly altered in
response to TMPD treatment although this same ion was ultimately excluded by
both the BH-FDR and Bonferroni corrections.
Table 4: Number of responsive proteins ion discovered in myoblasts Myoblasts were cultured and treated with TMPD or GWδ for 24h (section 2.1) prior to
SELDI-TOF-MS analysis (section 2.6). The number of proteins ions whose levels were
significantly altered in response to treatment is shown along with the number of
responsive proteins ions discovered after correction for the BH-FDR or using a
Bonferroni correction (Students t-test, p<0.05). The number of selected protein ions
was vastly reduced when the corrections for multiple testing were applied suggesting
that these corrections may have been too harsh in this case. Use of the BH-FDR
correction resulted in the inclusion of 3 ions at m/z 5743, 8557 and 9389 for TMPD-
treated myoblasts whilst a Bonferroni correction resulted in the inclusion of just 1 ion at
m/z 8557.
TMPD GWδ TMPD GWδ TMPD GWδ
NP20 10 8 0 0 0 0 54H50 9 6 0 0 0 0 109
CM10 (pH 4) 15 11 0 0 0 0 112CM10 (pH6) 17 14 0 0 0 0 107CM10 (pH8) 18 19 1 0 0 0 106
IMAC 30 11 13 2 0 1 0 103
ProteinChip array
Number of responsive protein ionsNumber of
ions detectedNo correction BH-FDR Bonferroni
Chapter 3: Results
114
When changes in the levels of protein ions measured in myotubes were
examined following correction for multiple testing, a higher number of protein
ions remained than was observed in myoblasts. As with myoblasts, the BH-FDR
correction resulted in the inclusion of a higher number of ions than the
Bonferroni correction (Table 5).
Table 5: Number of responsive proteins ion discovered in myotubes Myotubes were cultured and treated with TMPD or GWδ for 24h (section 2.1) before
SELDI-TOF-MS analysis was conducted (section 2.6). The number of responsive
protein ions is shown before and after correction for multiple testing using the BH-FDR
or using the Bonferroni correction. In most cases, the number of responsive protein
ions was vastly reduced as these correction methods apply a threshold to the p value
obtained.
TMPD GWδ TMPD GWδ TMPD GWδ
NP20 4 5 2 2 2 1 24H50 29 8 15 12 3 2 77
CM10 (pH 4) 1 1 1 1 1 1 50CM10 (pH 6) 8 6 1 0 1 0 61CM10 (pH 8) 18 17 1 1 1 1 66
IMAC 30 30 25 21 19 2 2 53
ProteinChip array
Number of responsive protein ions Number of ions
detectedNo correction BH-FDR Bonferroni
More protein ions were included after the corrections for multiple testing were
applied than there were with myoblasts. Following a BH-FDR correction in the
initial SELDI-TOF-MS experiment run (No. 1), a total of 41 ions were shown as
being responsive to treatment with TMPD in myotubes (35 ions for GWδ). When
the Bonferroni correction was applied to the same data, the number of
responsive ions was reduced to only 10 ions (7 for GWδ). Of these, only the
levels of ions at m/z 5298 (on CM10 arrays at pH 4) and 6588 (on NP20 arrays)
were still changed significantly in the repeat SELDI-TOF-MS experiment run
experiment (No. 2). However, the ion at m/z 6588 was excluded by the
Bonferroni correction in all cases despite that its levels were significantly
changed in response to treatment (Student’s t-test, p<0.05).
Chapter 3: Results
115
The BH-FDR and the Bonferroni correction are sometimes considered to be
severe methods for accounting for false discovery and this appeared to be the
case for myoblasts295. Only 1 ion at m/z 5743 resulted that was changed in
response to the treatment of myoblasts with TMPD in an initial SELDI-TOF-MS
experiment run (No. 1). However, even this ion was eventually excluded after
the correction thresholds were applied in an independent SELDI-TOF-MS
experiment run (No. 2).
When SELDI-TOF-MS data from myotubes was examined, a higher number of
protein ions were included after the correction thresholds were applied in the
initial SELDI-TOF-MS experiment run (No. 1) than there were for myoblasts.
However, a high number of those ions were not included when tested in an
independent repeat SELDI-TOF-MS experiment run. Despite this, an ion at m/z
5298 was still included.
The risks associated with accepting the changes from a single experiment were
also recognised during these investigations and it was therefore decided to
address the possibility of false discovery by repeating experiments and looking
for consistent changes in the levels of protein ions in different, independent
SELDI-TOF-MS experiment runs.
Chapter 3: Results
116
3.4.4 Reproducibility of SELDI-TOF-MS protein ions in muscle cells
Prior to differential protein expression experiments, an investigation into SELDI-
TOF-MS reproducibility was conducted. Data from two repeat SELDI-TOF-MS
experiment runs (No.s 1 and 2) were considered as a training dataset whilst the
findings from these experiments were validated in a third independent SELDI-
TOF-MS experiment run (test dataset, No. 3). An investigation into the
variability of the ions detected using the biomarker settings described in section
2.6.3 was conducted in order to gain an appreciation for the intra and inter-
experimental variability of the SELDI-TOF-MS data generated on CM10
ProteinChip® arrays (at pH 4) in the training dataset (SELDI-TOF-MS
experiment run (No.3).
In myoblasts, the average coefficient of variation in the levels of the 112 protein
ions detected was 27% in the first SELDI-TOF-MS experiment run. In the repeat
SELDI-TOF-MS experiment run of the training dataset, the coefficient of
variation was 25%. The levels of 12 of the 112 protein ions measured by
SELDI-TOF-MS were significantly different between the two repeat experiments
(Student’s t-test, p<0.05) (Figure 39). In myotubes, the average CV% for the 50
ions measured in the first experiment run was 24% and was 17% in the second
experiment run. The levels of 4 of the 50 ions measured were significantly
different between repeat experiment runs (Figure 40).
These results demonstrated that whilst some of the ions detected by SELDI-
TOF-MS demonstrated a degree of intra and inter-experimental variability, the
majority of the ions were reproducible within acceptable limits as determined by
their CV% and significance testing between the repeat experiments. Also the
reproducibility appears to be comparable with that obtained by other SELDI-
TOF-MS users296-298. The source of the variation may be difficult to pinpoint
without additional investigations but these results highlighted the need to
examine changes observed within experiment runs before examining changes
across different repeat SELDI-TOF-MS experiment runs.
Chapter 3: Results
117
Figure 39: Intra and inter-experiment reproducibility of SELDI-TOF-MS protein ions in myoblasts Myoblasts were cultured (section 2.1) and untreated myoblasts were analysed using SELDI-TOF-MS on CM10 ProteinChip® arrays (pH
4). The CV% for all of the ions detected is shown and the red line is set at 30% which is the approximate CV% commonly observed in
SELDI-TOF-MS experiments by other users. The levels of the ions detected were compared between 2 repeat SELDI-TOF-MS
experiment runs in order to evaluate inter-experiment reproducibility. The filled bars (12/112) represent the ions whose levels varied
significantly between the 2 repeat experiments and significant changes are denoted by *p<0.05, **p<0.01, p<0.001(Student’s t-test). Error
bars represent SEM where visible.
0
10
20
30
40
50
60
70
80
90
100
2523
2542
2680
2701
2722
2882
2995
3042
3149
3178
3358
3513
3584
3660
3675
3747
3882
3925
3979
4019
4076
4096
4116
4203
4236
4279
4337
4353
4424
4519
4571
4704
4717
4748
4761
4786
4832
4866
4910
4935
4959
4976
5002
5127
5165
5216
5292
5374
5492
5524
5623
5656
5670
5743
5757
5978
5997
6076
6116
6142
6159
6187
6229
6257
6271
6303
6315
6345
6415
6508
6581
6640
6871
6912
6950
6994
7033
7060
7139
7179
7217
7503
8177
8436
8549
8602
8666
8890
9090
9109
9135
9150
9176
9348
9389
9592
9967
1006
610
254
1096
511
020
1129
411
704
1234
013
821
1401
814
765
1494
716
852
1719
817
790
1846
3
m/z
CV%
SELDI-TOF-MS experiment 1
Chapter 3: Results
118
Figure 39 continued
0
10
20
30
40
50
60
70
80
90
100
2523
2542
2680
2701
2722
2882
2995
3042
3149
3178
3358
3513
3584
3660
3675
3747
3882
3925
3979
4019
4076
4096
4116
4203
4236
4279
4337
4353
4424
4519
4571
4704
4717
4748
4761
4786
4832
4866
4910
4935
4959
4976
5002
5127
5165
5216
5292
5374
5492
5524
5623
5656
5670
5743
5757
5978
5997
6076
6116
6142
6159
6187
6229
6257
6271
6303
6315
6345
6415
6508
6581
6640
6871
6912
6950
6994
7033
7060
7139
7179
7217
7503
8177
8436
8549
8602
8666
8890
9090
9109
9135
9150
9176
9348
9389
9592
9967
1006
610
254
1096
511
020
1129
411
704
1234
013
821
1401
814
765
1494
716
852
1719
817
790
1846
3
m/z
CV%
*
*
**
*
** **
*
**
*
**
***
*
***
SELDI-TOF-MS experiment 2
Chapter 3: Results
119
Figure 40: Intra and inter-experimental reproducibility of SELDI-TOF-MS protein ions in myotubes Myotubes were cultured (section 2.1) and untreated myotubes were analysed using SELDI-TOF-MS on CM10 ProteinChip® arrays (pH 4).
The CV% for all of the ions detected is shown and the red line is set at 30% which is the approximate CV% commonly observed in SELDI-
TOF-MS experiments by other users. The levels of the ions detected were compared between 2 repeat SELDI-TOF-MS experiments in
order to evaluate inter-experiment reproducibility. The filled bars (4/50) represent the ions whose levels varied significantly between the 2
repeat experiments and significant changes are denoted by *p<0.05, **p<0.01 (Student’s t-test). Error bars represent SEM where visible.
0
10
20
30
40
50
60
70
80
90
100
2537
2609
2657
2700
2717
2761
2862
2922
3044
3182
3363
3437
3481
3584
4110
4337
4354
4932
4959
4975
5002
5138
5165
5298
5392
5537
5655
5743
6230
6272
6579
6914
7486
8176
8600
8641
9165
9972
1025
0
1101
4
1129
5
1169
7
1233
6
1382
4
1402
1
1476
2
1580
0
1780
2
1849
0
1891
6
m/z
CV%
SELDI-TOF-MS experiment 1
Chapter 3: Results
120
Figure 40 continued
0
10
20
30
40
50
60
70
80
90
10025
37
2609
2657
2700
2717
2761
2862
2922
3044
3182
3363
3437
3481
3584
4110
4337
4354
4932
4959
4975
5002
5138
5165
5298
5392
5537
5655
5743
6230
6272
6579
6914
7486
8176
8600
8641
9165
9972
1025
0
1101
4
1129
5
1169
7
1233
6
1382
4
1402
1
1476
2
1580
0
1780
2
1849
0
1891
6
m/z
CV%
**
**
*
SELDI-TOF-MS experiment 2
Chapter 3: Results
121
3.4.5 Validation of TMPD and GWδ responsive protein ions in skeletal
muscle cells
Differential protein expression was investigated further in the two repeat SELDI-
TOF-MS experiment runs of the training dataset (No.s 1 and 2). In order to
ensure that the same ions were compared between the different defined
experiments, SELDI-TOF-MS data from both experiment series was combined
before clusters of protein ions were detected using the BMW (as described in
section 2.6.3). Differentially expressed ions were considered for further
statistical analyses if the ion was consistently expressed at significantly different
levels between control and treated samples in both repeat experiment runs of
the training dataset (Student’s t-test, p<0.05). This consideration for the
consistency of the response of individual protein ions effectively reduced the p
value for selected protein ions to less than 0.0025 (i.e. p=0.052).
A total of 592 protein ions were detected in myoblasts on 6 different types of
ProteinChip® array using the BMW settings described in section 2.6.3. The
treatment of myoblasts with 25μM TMPD resulted in a consistent increase in the
levels of 3 protein ions at m/z 5743, 6910 and 7177 along with a decrease in the
levels of 7 ions at m/z, 6306, 6420, 9097, 9113, 9123, 9137 and 9161 in the two
repeat experiments of the training dataset. These changes were detected on a
combination of NP20 and CM10 (pH 4 and 6) ProteinChip® arrays (Table 6 and
Figure 41a).
Treating myoblasts with 50μM GWδ resulted in a significant increase in the
levels of 6 ions at m/z 4910, 4913, 5655, 7177, 7222 and 13814. These
changes were accompanied by a decrease in the levels of 7 protein ions at m/z,
6306, 6420, 9097, 9113, 9123, 9137 and 11708. The changes associated with
GWδ treatment were detected on a combination of NP20, H50 and CM10 (pH4,
6 and 8) ProteinChip® arrays. Also, the levels of the ions at m/z 6306, 6420,
7177, 9097, 9113, 9123 and 9137 were changed in response to treatment with
both TMPD and GWδ (Table 6 and Figure 41b).
Chapter 3: Results
122
Table 6: Summary of protein expression changes in myoblasts Myoblasts were cultured, treated with the test compounds (section 2.1) and analysed
using SELDI-TOF-MS (section 2.6). The total number of protein ions detected (section
2.6.3) on each type of ProteinChip® array and in 2 repeat SELDI-TOF-MS experiments
that formed the training dataset is shown. Cell homogenates from control myoblasts
were compared with TMPD or GWδ-treated samples. Spectra from both experiments
were combined before ion detection was performed simultaneously for both
experiments in order to ensure the generation of consistent clusters of protein ions
using the BMW. The number of ions that were responsive to treatment in each of the
repeat experiments and those that were consistently responsive to treatment are
shown (Student’s t-test, p<0.05).
TMPD GWδ TMPD GWδ TMPD GWδ
NP20 10 8 29 26 7 6 54H50 9 6 52 47 0 1 109
CM10 (pH 4) 15 11 32 30 1 1 112CM10 (pH6) 17 14 35 29 2 4 107CM10 (pH8) 18 19 21 22 0 1 106
IMAC 30 11 13 2 2 0 0 103
ProteinChip array Experiment 1 Experiment 2 Combined
Number of responsive protein ions Total number of ions
detected
Chapter 3: Results
123
-1.0
-0.5
0.0
0.5
1.0
1.5
CM
10 p
H4
5743
CM
10 p
H6
6910
NP
20 9
161
CM
10 p
H4
4910
CM
10 p
H8
4913
CM
10 p
H6
5655
CM
10 p
H6
7222
H50
117
08
CM
10 p
H6
1381
4
NP
20 6
306
NP
20 6
420
CM
10 p
H6
7177
NP
20 9
097
NP
20 9
113
NP
20 9
123
NP
20 9
137
m/z
Rel
ativ
e le
vel
TMPD responders GWδ responders Common responders
+1.0
+1.5
+0.5
-0.5
-1.0
a)
-1.0
-0.5
0.0
0.5
1.0
1.5
CM
10 p
H4
5743
CM
10 p
H6
6910
NP
20 9
161
CM
10 p
H4
4910
CM
10 p
H8
4913
CM
10 p
H6
5655
CM
10 p
H6
7222
H50
117
08
CM
10 p
H6
1381
4
NP
20 6
306
NP
20 6
420
CM
10 p
H6
7177
NP
20 9
097
NP
20 9
113
NP
20 9
123
NP
20 9
137
m/z
b)
Rel
ativ
e le
vel
+1.0
+1.5
+0.5
-0.5
-1.0
TMPD responders GWδ responders Common responders
Figure 41: Responsive SELDI-TOF-MS protein ions in myoblasts Rat L6 myoblasts were cultured and treated with 25μM TMPD or 50μM GWδ for 24h
(section 2.1). Samples were then analysed on NP20, CM10 (at pH 4, 6 and 8) and
IMAC 30 ProteinChip® arrays (section 2.6). Protein ions whose levels were consistently
changed in response to treatment with a) TMPD or b) GWδ in two repeat SELDI-TOF-
MS experiment runs are shown by solid bars (Student’s t-test, p<0.05). Ions whose
levels changed in response to treatment with TMPD are grouped on the left, those that
changed following GWδ treatment are grouped in the middle section and those ions
that changed following the treatment of myoblasts with both test compounds are
grouped on the right. Error bars show ± SEM.
Chapter 3: Results
124
Proteins whose ion levels were consistently changed in myotubes in response
to treatment with TMPD or GWδ were identified using a similar approach to that
used for myoblasts. Again, SELDI-TOF-MS data generated in the 2 repeat
SELDI-TOF-MS experiment runs of a training dataset was examined to discover
protein ions whose levels were consistently changed in response to treatment
(Student’s t-test, p<0.05).
In myotubes, 331 protein ions were detected using the different ProteinChip®
arrays and assay conditions previously described for myoblasts (section 2.6).
The treatment of myotubes with 25µM TMPD resulted in a significant increase
(in the training dataset) in the levels of 4 protein ions at m/z 3158, 3978, 5298,
and 6305 along with a decrease in the levels of 6 ions at m/z 4815, 5579, 6312,
6578, 6588 and 12357. These ions were detected on a combination of NP20,
IMAC and CM10 (pH4 and 6) ProteinChip® arrays. Treating myoblasts with
50μM GWδ resulted in an increase in the levels of 4 ions at m/z 3158, 5298,
6305 and 9126 and a decrease in the levels of 3 ions at m/z 5753, 6588 and
12357. The levels of the ions at m/z 3158, 5298, 6305, 6588 and 12357 were
responsive to treatment with both test compounds (Table 7 and Figure 42).
Chapter 3: Results
125
Table 7: Summary of protein expression changes in myotubes Myotubes were grown (section 2.1) and analysed using SELDI-TOF-MS (section 2.6).
Cell homogenates from control myotubes were compared with TMPD or GWδ-treated
samples and the total number of protein ions detected (section 2.6.3) on each type of
ProteinChip® array in the training dataset is shown. Spectra were combined and protein
ions were detected simultaneously for both SELDI-TOF-MS experiments in order to
ensure the generation of consistent clusters of protein ions using the BMW and those
ions that were responsive in each of the repeat experiments are shown (Student’s t-
test, p<0.05).
TMPD GWδ TMPD GWδ TMPD GWδ
NP20 4 5 9 8 3 4 24H50 29 25 12 8 0 0 77
CM10 (pH 4) 1 1 4 7 1 1 50CM10 (pH6) 8 6 16 10 2 2 61CM10 (pH8) 18 17 11 9 1 0 66
IMAC 30 30 25 4 2 3 0 53
Total number of ions
detected
ProteinChip array Experiment 1 Experiment 2 Combined
Number of responsive protein ions
Chapter 3: Results
126
-1.0
-0.5
0.0
0.5
1.0
1.5
2.0
2.5
CM
10 p
H8
3978
CM
10 p
H6
4815
IMA
C 5
579
IMA
C 6
312
IMA
C 6
578
NP
20 5
753
CM
10 p
H6
9126
CM
10 p
H6
3158
CM
10 p
H4
5298
NP
20 6
305
NP
20 6
588
NP
20 1
2357
m/z
Rel
ativ
e le
vel
+1.0
+1.5
+0.5
-0.5
-1.0
TMPD responders GWδ responders Common responders
+2.0
+2.5
a)
-1.0
-0.5
0.0
0.5
1.0
1.5
2.0
2.5
CM
10 p
H8
3978
CM
10 p
H6
4815
IMA
C 5
579
IMA
C 6
312
IMA
C 6
578
NP
20 5
753
CM
10 p
H6
9126
CM
10 p
H6
3158
CM
10 p
H4
5298
NP
20 6
305
NP
20 6
588
NP
20 1
2357
m/z
b)
Rel
ativ
e le
vel
+1.0
+1.5
+0.5
-0.5
-1.0
+2.0
+2.5
TMPD responders GWδ responders Common responders
Figure 42: Responsive SELDI-TOF-MS protein ions in myotubes Rat L6 myotubes were differentiated from cultured myoblasts and treated with TMPD or
GWδ for 24h (section 2.1). Samples were then analysed on NP20, CM10 (at pH 4, 6
and 8) and IMAC 30 ProteinChip® arrays (section 2.6). Ions whose levels were
consistently changed in response to treating myotubes with a) TMPD or b) GWδ are
shown by solid bars (Student’s t-test, p<0.05). Ions whose levels were changed
following treatment with TMPD are grouped on the left, those changed following GWδ
treatment are grouped in the middle section and those ions that changed following
treatment with both test compounds are grouped on the right. Error bars show ± SEM.
Chapter 3: Results
127
The levels of protein ions that were consistently responsive to treatment with
the test compounds were examined in an independent repeat third SELDI-TOF-
MS experiment run (No. 3) in order to determine the most consistent of the
changes observed in the training dataset. The levels of 7 out of the 10 protein
ions originally found (at m/z 9161, 6306, 7177, 9097, 9113, 9123 and 9137)
were significantly altered in response to the treatment of myoblasts with TMPD
as they were in the two SELDI-TOF-MS experiment runs that formed the
training dataset. However, the levels of the other 3 ions originally tested were
not significantly changed in the test dataset (Figure 43). This included an ion at
m/z 5743, which had also been selected as being responsive to treatment with
TMPD after the BH-FDR and Bonferroni corrections, but was excluded by the
corrections in the repeat SELDI-TOF-MS experiment run No. 2 (section 3.4.2).
A similar story was true for myoblasts treated with GWδ in that ions at m/z 6306,
7177, 9097, 9113, 9123, 9137 and 9161 were significantly changed in response
to treatment, as they were in the training dataset but the levels 7 of the 13 ions
examined were not changed in the test dataset (Figure 44). Representative
SELDI-TOF-MS putative biomarkers in myoblasts are shown in Figure 45.
Protein ions in myotubes whose levels were consistently changed treated with
TMPD or GWδ were also examined in the myotubes test dataset to see whether
they were still responsive in the independent test dataset. The levels of 3
protein ions at m/z 4815, 3185 and 5298 were changed significantly in response
to treatment with TMPD although the levels of 7 other ions tested were not
changed significantly when tested (Figure 46) (Student’s t-test, p>0.05).
Amongst these, an ion at m/z 5298 was also selected in myotubes when the
corrections for multiple testing were applied (BH-FDR and Bonferroni) and this
may be a more robust marker than others found during these analyses. In the
case of GWδ, the levels of the 7 ions tested were not significantly different to
those in control myotubes (Figure 47). Representative SELDI-TOF-MS putative
biomarkers in myotubes are shown in Figure 48.
Chapter 3: Results
128
CM
10 p
H4
5743
CM
10 p
H6
6910
NP
20 9
161
NP
20 6
306
NP
20 6
420
CM
10 p
H6
7177
NP
20 9
097
NP
20 9
113
NP
20 9
123
NP
20 9
137
0
1
2
3
4
5
6
7
8
9
Rel
ativ
e io
n in
tens
ity
TMPD responders Common responders
**
**
**
** ** ****
ControlTreatedControlTreatedControlTreated
Figure 43: Levels of TMPD-responsive protein ions in myoblasts Myoblasts were grown and treated with 25μM TMPD (section 2.1) before SELDI-TOF-
MS analysis on different ProteinChip® arrays (section 2.6). Proteins that were
responsive to treatment in two repeat experiments of the training dataset were tested in
an independent third dataset to examine their robustness as putative biomarkers of
myopathy. All of the proteins examined showed similar responses to treatment with the
test compound although significant differences were only observed in the levels of
protein ions examined at m/z 6306, 7177, 9097, 9113, 9123, 9137 and 9161 (Student’s
t-test, p<0.05). Where visible, bars show mean of the control or treated group and
significant changes are denoted by **p<0.01.
Chapter 3: Results
129
CM
10 p
H4
4910
CM
10 p
H8
4912
CM
10 p
H6
5655
CM
10 p
H6
7222
H50
117
08
CM
10 p
H6
1381
4
NP2
0 63
06
NP2
0 64
20
CM
10 p
H6
7177
NP2
0 90
97
NP2
0 91
13
NP2
0 91
23
NP2
0 91
37
0
1
2
3
4
5
6
7
8
Rel
ativ
e io
n in
tens
ity
ControlTreatedControlTreatedControlTreated
*
*
***
****
GWδ responders Common responders
Figure 44: Levels of GWδ-responsive protein ions in myoblasts
Myoblasts were grown and treated (section 2.1) with 50μM GWδ prior to SELDI-TOF-
MS analysis (section 2.6). The levels of proteins that were responsive to treatment in
the training dataset were examined in an independent test dataset. All protein ions
examined showed similar responses to treatment with GWδ but significant differences
were only observed in the levels of protein ions at m/z 6306, 7222, 9097, 9113, 9123
and 9137 (Student’s t-test, p<0.05). Where visible, the bars show the mean of the
control or treated group and significant changes are denoted by *p<0.05 and **p<0.01.
Chapter 3: Results
130
5000 6000 7000 8000
0
2
4
6 6306+H
m/z
5000 10000 15000 20000
-505
101520
-505
1015
-505
101520
-505
101520
-505
101520
0
5
10
15
5000 10000 15000 20000
m/z
Rel
ativ
e io
n in
tens
ityNP20
H50
CM10 (pH4)
CM10 (pH6)
CM10 (pH8)
IMAC-30
NP20
6000 6500 7000 7500 8000 85000
5
15
m/z
7177+H
CM10 (pH6)
10
9000 9500 100000
2
4
m/z
NP2012 34
5
9097+H19113+H29123+H39137+H49161+H5
7222+H
Figure 45: Representative SELDI-TOF-MS spectra showing putative biomarkers in TMPD or GWδ-treated myoblasts
Rat L6 myoblasts were cultured and treated with 25μM TMPD or 50μM GWδ for 24h (section 2.1). SELDI-TOF-MS analysis was
conducted (section 2.6) and protein ions at m/z 6306, 7177, 7222, 9097, 9113, 9123, 9137 and 9161 (detected on NP20 and CM10
ProteinChip® arrays) were consistently responsive to treatment with TMPD or GWδ in both the training and test datasets. The m/z is
shown on the x-axis and the relative ion intensity is displayed on the y-axis.
Chapter 3: Results
131
CM
10 p
H8
3978
CM
10 p
H6
4815
IMAC
557
9
IMAC
631
2
IMAC
657
8
CM
10 p
H6
3158
CM
10 p
H4
5298
NP2
0 63
05
NP2
0 65
88
NP2
0 12
357
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
4.5
5.0
5.5
6.0
6.5
Rel
ativ
e io
n in
tens
ity
ControlTreatedControlTreatedControlTreated
TMPD responders Common responders
*
*
**
Figure 46: Levels of TMPD-responsive protein ions in myotubes Myotubes were cultured and treated (section 2.1) with 25μM TMPD prior to SELDI-TOF-
MS analysis (section 2.6). The levels of proteins that were responsive to treatment in the
training dataset were examined in a test dataset. All protein ions examined showed similar
responses to those observed in the training dataset but significant differences in their
levels were only observed in ions at m/z 3158, 4815 and 5298 (Student’s t-test, p<0.05).
Where visible, bars show the mean of the control or treated group and significant changes
are denoted by *p<0.05 and **p<0.01.
Chapter 3: Results
132
NP2
0 57
53
CM
10 p
H6
9126
CM
10 p
H6
3158
CM
10 p
H4
5298
NP2
0 63
05
NP2
0 65
88
NP2
0 12
357
0.0
2.5
5.0
7.5
Rel
ativ
e io
n in
tens
ity
ControlTreatedControlTreatedControlTreated
GWδ responders Common responders
Figure 47: Levels of GWδ-responsive protein ions in myotubes
Myotubes were cultured and treated with 50μM GWδ (section 2.1) prior to SELDI-TOF-MS
analysis (section 2.6). Proteins that were responsive to treatment in the training dataset
were validated in the test dataset. All protein ions examined showed similar responses to
treatment with GWδ but, in this case, none of the changes were statistically significant
(Student’s t-test, p>0.05). Where visible, bars show the mean of the control or treated
group.
Chapter 3: Results
133
5000 5100 5200 5300 5500
0
2.5
5
7.5
5400
m/z
5298+H
4600 4800 5000 5200
0
0.5
1
1.5
2
m/z
4815+H
5000 10000 15000 20000
-505
101520
-505
1015
-505
101520
-505
101520
-505
101520
-505
1015
5000 10000 15000 20000
NP20
H50
CM10 (pH4)
CM10 (pH6)
CM10 (pH8)
IMAC-30
m/z
Rel
ativ
e io
n in
tens
ity
CM10 (pH4)
CM10 (pH6)
3158+H
2500 3000 3500 4000
-1
0
1
2
3
4CM10 (pH6)
Figure 48: Representative SELDI-TOF-MS spectra showing putative biomarkers of myopathy in myotubes Rat L6 myotubes were cultured and treated with TMPD or GWδ for 24h (section 2.1). SELDI-TOF-MS analysis was conducted (section
2.6) and protein ions at m/z 3158, 4815 and 5298 (detected on CM10 ProteinChip® arrays) were consistently responsive to treatment
with TMPD in repeat experiments. The m/z is displayed on the x-axis and the relative ion intensity is displayed on the y-axis.
Chapter 3: Results
134
A number of approaches to discovering the most robust markers of TMPD or
GWδ treatment were attempted in these SELDI-TOF-MS investigations. The
need for multiple approaches to the analysis of this data perhaps demonstrated
the difficulty in determining the most useful protein ions. SELDI-TOF-MS was
useful in discovering protein ions that were consistently responsive to the
effects of treatment with the test compounds although the changes observed
were relatively small changes and were difficult to distinguish amongst the
background variation of their levels. When the levels of potentially useful protein
ions were evaluated in the test dataset, they were either increased or
decreased in response to treatment as previously observed some of the
changes were not statistically significant when investigated in independent
repeat SELDI-TOF-MS experiments.
3.4.6 Selection of responsive protein ion combinations in skeletal muscle cells
In the previous section, protein ions that were responsive to treatment with
25μM TMPD or 50μM GWδ (Figure 41 and Figure 42) were discovered.
However, not all of the protein ions suspected to be useful based on initial
SELDI-TOF-MS experiments were verified successfully when examined in the
test dataset. These ions still responded moderately to treatment and it was
believed that, despite the changes in the levels of some of the protein ions
being statistically insignificant (Student’s t-test, p>0.05), a combination of
protein ions may have increased the ability to discriminate between control and
treated cells when compared to the individual ions.
In order to investigate the use of multiple protein ions, those ions whose levels
were changed in response to TMPD or GWδ treatment in the 2 repeat SELDI-
TOF-MS experiment runs (No.s 1 and 2) were entered into a forward stepwise
regression analysis (p to enter of 0.05, p to remove of 0.06). Those that were
responsive to treatment with TMPD were used in the regression analysis for
TMPD-treated cells and those ions whose levels were changed in response to
GWδ treatment were used for GWδ-treated cells. Regression analysis was then
Chapter 3: Results
135
used in order to discover protein ions whose levels correlated well with
treatment but that also contributed independently to the model (had a low
covariance with each other). In a forward stepwise regression analysis, each
variable is evaluated for its effectiveness in distinguishing between control and
treated samples. If the variable assessed does not improve the model, it is
removed and the utility of another variable is assessed in the next step of the
analysis. This is continued until only variables that contribute significantly to the
model remain299. This analysis was repeated using data from the test dataset
(SELDI-TOF-MS experiment run No. 3) in order to validate findings from the
training dataset (SELDI-TOF-MS experiment runs 1 and 2).
When considering myoblasts, forward stepwise regression analysis of SELDI-
TOF-MS data from the training dataset resulted in the inclusion of 3 protein ions
at m/z 5743, 6420 and 7177 in a model that described the differences between
control myoblasts and those treated with TMPD. When regression analysis was
repeated using data from the test dataset, the ions at m/z 6420 and 7177 were
again selected as being useful when considered together. These same two
protein ions were also selected by stepwise regression analysis when SELDI-
TOF-MS data from control and GWδ-treated myoblasts was examined (in both
the training and test dataset) (Table 8). Therefore, in combination, these 2
proteins ions were consistently useful in discriminating between control and
treated myoblasts.
Chapter 3: Results
136
Table 8: Summary of forward stepwise regression of responsive protein ions in control and treated myoblasts Myoblasts were cultured and treated with TMPD or GWδ (section 2.1). Protein ions that
were responsive to treatment were observed in spectra generated following SELDI-
TOF-MS analysis of cell homogenates (section 2.6) in 2 repeat experiments and a
forward stepwise regression analysis of these ions was performed (p to enter of 0.05, p
to remove of 0.06). A repeat regression analysis using SELDI-TOF-MS data from an
independent test dataset allowed selected ions to be validated and ions at m/z 6420
and 7177 were useful for characterising control and TMPD or GWδ-treated myoblasts.
Training dataset Test dataset
p value p value
1 6420 NP20 0.002 <0.001
2 7177 CM10 (pH 6) 0.016 <0.001
3 5743 CM10 (pH 4) 0.041 0.72
1 7177 CM10 (pH 6) <0.001 0.001
2 6420 NP20 <0.001 <0.001
ProteinChip array
TMPD
GWδ
Test compound Step m/z
Correlation between the protein ions resulting from regression analysis was also
considered in order to ensure that the ions selected contributed independently
to the discrimination between control and treated myoblasts. The correlation
coefficients for ions shown in Table 8 were examined and the variables at m/z
6420 and 7177 had low correlation coefficients when compared to each other
(between -0.3 and -0.1 for myoblasts treated with TMPD or <0.1 when GWδ-
treated myoblasts were considered). This suggested that these protein ions
contributed independently to the model describing the differences between
control and treated myoblasts300,301.
Chapter 3: Results
137
Forward stepwise regression analysis was also applied to SELDI-TOF-MS data
from the training dataset of myotubes and resulted in the inclusion of 5 protein
ions at m/z 3158, 3978, 4815, 5298 and 6578 that correlated well with TMPD
treatment. The ions at m/z 5298 and 4815 were consistently useful in the model
when data from the test dataset was considered in a repeat regression analysis.
However, when considering control and GWδ-treated myotubes, ions at m/z
5298 and 6305 that were initially selected by regression analysis were not
effective at discriminating between control and treated myotubes when tested
using an independent dataset (Table 9).
Table 9: Summary of forward stepwise regression of responsive protein ions in control and treated myotubes
Myotubes were cultured and treated with TMPD or GWδ (section 2.1). Protein ions
within the samples were observed in spectra following SELDI-TOF-MS analysis
(section 2.6). A forward stepwise regression analysis of responsive protein ions from
the training dataset determined the ions that contributed significantly to the
discrimination between control and treated myotubes (Student’s t-test, p<0.05). A
repeat regression analysis using SELDI-TOF-MS data from an independent test
dataset allowed selected ions to be validated and ions at m/z 5298 and 4815 resulted
from this analysis that were useful for TMPD-treated myotubes.
Training dataset Test dataset
p value p value
1 5298 CM10 pH 4 <0.001 <0.001
2 4815 CM10 pH 6 <0.001 0.003
3 3978 CM10 pH 8 <0.001 0.649
4 3158 CM10 pH 6 <0.001 0.804
5 6578 IMAC 30 <0.001 0.995
1 5298 CM10 pH 4 <0.001 0.087
2 6305 NP20 0.020 0.072
Test compound Step m/z ProteinChip
array
GWδ
TMPD
Chapter 3: Results
138
The correlation coefficients for protein ions selected by regression analysis and
validated successfully in the test dataset were examined for ions (at m/z 5298
and 4815) selected following the analysis of SELDI-TOF-MS data from control
and TMPD or GWδ-treated myotubes. The correlation coefficients were low for
these variables (r is between -0.3 and -0.1 for TMPD-treated cells or <0.1, when
examining GWδ treatment) and this demonstrated that the ions selected
contributed independently to the discrimination between control and TMPD-
treated myotubes.
From the previous forward regression analysis, ions at m/z 6420 and 7177 were
shown to contribute significantly and independently to the discrimination
between control and TMPD or GWδ-treated myoblasts (Student’s t-test,
p<0.05). When these ions were considered individually, they were unable to
discriminate completely between control and treated myoblasts (Figure 43 and
Figure 44). When these ions were considered together, their discriminatory
power was greater and they were able to separate control and TMPD-treated
(Figure 49a) or control and GWδ-treated myoblasts (Figure 49b) completely.
When the ions at m/z 5298 and 4815 were examined individually their levels
changed significantly in response to TMPD-treatment (Student’s t-test, p<0.05)
(Figure 46). Complete separation between control and TMPD-treated myotubes
was possible when they were used together although it was not possible when
the ions were considered separately (Figure 50a). Ions at 6305 and 5298 were
not changed significantly when examined individually in GWδ-treated myotubes
in the test dataset (Figure 47). Even the combined power of these ions was not
sufficient to discriminate completely between control and GWδ-treated
myotubes although only 2/6 treated samples were classified incorrectly (Figure
50b).
Chapter 3: Results
139
m/z 7177
m/z
642
0
3.0 3.5 4.0 4.5 5.0 5.5 6.0 6.5 7.0 7.5 8.00.4
0.6
0.8
1.0
1.2
1.4
1.6
1.8
2.0
2.2
2.4
2.6
2.8
m/z 7177
m/z
642
0
3 4 5 6 7 8 90.4
0.6
0.8
1.0
1.2
1.4
1.6
1.8
2.0
2.2
2.4
2.6
2.8
3.0a) b)
Control Treated
TMPD myoblasts GWδ myoblasts
Figure 49: Two-dimensional scatterplots of TMPD and GWδ responsive
protein ions selected by regression analysis in myoblasts Myoblasts were cultured and treated with TMPD or GWδ (section 2.1). Following
SELDI-TOF-MS analysis (Section 2.6), protein ions at m/z 6420 and 7177 were
selected by forward stepwise regression analysis for their combined ability to
discriminate between a) control and TMPD-treated myoblasts and b) control and GWδ-
treated myoblasts. The broken line in the scatterplots shows the discrimination
between control and treated myoblasts, based on the combined intensities of the ions
shown.
Chapter 3: Results
140
m/z 5298
m/z
481
5
2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.00.5
0.6
0.7
0.8
0.9
1.0
1.1
m/z 5298
m/z
630
5
2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.0 6.5 7.0 7.50.4
0.5
0.6
0.7
0.8
0.9
1.0
1.1
Control Treated
a) b)TMPD myotubes GWδ myotubes
Figure 50: Two-dimensional scatterplots of TMPD and GWδ responsive
protein ions selected by regression analysis in myotubes Myotubes were cultured and treated with TMPD or GWδ (section 2.1). Following
SELDI-TOF-MS analysis (Section 2.6), protein ions at m/z 5298 and 4815 were
selected by forward stepwise regression analysis for their combined ability to
discriminate between a) control and TMPD-treated myotubes. Ions at 6305 and 5298
were tested for their combined power to discriminate between b) control and GWδ-
treated myotubes and they only misclassified 2/6 treated myotubes. The broken line in
the scatterplots shows the discrimination between control and treated myotubes, based
on the combined intensities of the ions shown.
Chapter 3: Results
141
These results demonstrated that whilst some of the protein ions selected by
univariate analysis may not have been validated successfully in the test dataset,
there was the possibility of combining protein ions in order to increase their
discriminatory power. This was the case with TMPD and GWδ-treated
myoblasts and also with TMPD-treated myotubes and complete separation
between control and treated samples was possible using a combination of
responsive protein ions even where it had previously not been possible with
individual ions. Despite the increased power of protein ion combinations, 2/6
GWδ-treated myotubes were still misclassified as control samples (false
negatives).
The levels of ions at m/z 6420 and 9137 that were seen to be useful in
discriminating between control and treated myoblasts were assessed in cells
treated with lower concentrations of the test compounds. However, neither the
levels of the ion at m/z 6420 nor the ion at m/z 9137 were responsive to
treatment with <25μM TMPD or <50μM GWδ (Figure 51).
An ion at m/z 5298 was changed significantly when examined individually in the
test dataset (Student’s t-test, p<0.05) and it was also selected by regression
analysis. The levels of this ion in myotubes treated with lower concentrations of
TMPD and GWδ than were used previously were also investigated but they
were not changed in response to treatment with <25μM TMPD or <50μM GWδ
(Figure 52).
Chapter 3: Results
142
0
0.5
1
1.5
2
2.5
VC 6.25 12.5 25 50
[Drug] μM
Rel
ativ
e io
n in
tens
ity
m/z 6420
*
TMPD GWδ
0
1
2
3
4
5
6
7
VC 6.25 12.5 25 50
[Drug] μM
Rel
ativ
e io
n in
tens
ity
m/z 9137
** **
TMPD GWδ
Figure 51: Effect of TMPD or GWδ on rat L6 myoblasts Myoblasts were grown and tested with a range of concentrations of up to 25μM TMPD
or up to 50μM GWδ (section 2.1) before SELDI-TOF-MS analysis of cell homogenates
(section 2.6). Changes in the levels of putative protein ion biomarkers of myopathy at
m/z 6420 and 9137 in myoblasts treated with lower concentrations of the test
compounds than previously used were examined. These ions were not responsive to
the treatment of myoblasts with <25μM TMPD or <50μM GWδ. VC = vehicle control.
Error bars show ± SEM. Significant changes are denoted by *p <0.05, **p<0.01.
Chapter 3: Results
143
0
1
2
3
4
5
6
VC 6.25 12.5 25 50
[Drug] μM
Rel
ativ
e io
n in
tens
ity
m/z 5298
**
**
TMPD GWδ
Figure 52: Effect of TMPD or GWδ on rat L6 myotubes Myotubes were differentiated from myoblasts and tested with a range of concentrations
of up to 25μM TMPD or up to 50μM GWδ (section 2.1) prior to SELDI-TOF-MS analysis
(section 2.6). Changes in the levels of a protein ion at m/z 5298 in myotubes treated
with lower concentrations of the test compounds than previously used were examined.
This ion was not responsive to the treatment of myotubes with <25μM TMPD or <50μM
GWδ. VC = vehicle control. Error bars show ± SEM. Significant changes are denoted
by **p<0.01.
3.4.7 Identification of SELDI-TOF-MS putative biomarkers of myopathy
Attempts were made to identify the protein ions discovered as putative
biomarkers of myopathy in previous sections. The ExPASy server (provided by
the Swiss institute of Bioinformatics and linked to the UniProt consortium
knowledgebase) was interrogated using the Mw of putative biomarkers
discovered by SELDI-TOF-MS. The charge (z) of the protein ions detected by
SELDI-TOF-MS was taken as equal to one and the database was also
interrogated with z equal to 1, 2 or 3 to account for multiply charged ions.
Protein dimerisation was also accounted for and an error of 0.1% (the expected
error for a SELDI-TOF-MS external calibration) in the interrogated mass was
accounted for.
Chapter 3: Results
144
The masses of the ions discovered by SELDI-TOF-MS were also checked
against those of proteins ions identified by LC-MS/MS (described in detail in
section 3.5). The advantage of this method of verification was that the peptides
and proteins included in the search were measured directly in this study and
were not selected solely on the basis of a Mw match for the responsive ions
discovered in myoblasts (Table 10-Table 12) or those discovered in myotubes
(Table 13-Table 15).
Table 10: Identification of SELDI-TOF-MS biomarkers of myopathy in myoblasts (TMPD responders) Myoblasts were cultured and treated with 25μM TMPD for 24h (section 2.1). Protein
ions at m/z 5743, 6910 and 9181 were responsive to the treatment of myoblasts with
TMPD when measured using SELDI-TOF-MS (section 2.6). Mw information available
on the ExPASy database was interrogated using the TagIdent tool in order to discover
possible matches for the ions detected on ProteinChip® arrays. LC-MS/MS data
generated following the analysis of myoblasts cell homogenates was also interrogated
via Bioworks Browser (section 3.5). Although several possible identifications were
available for each of the ions when TagIdent was used, no possible identifications were
provided from the LC-MS/MS data.
TMP
D
Gw
d
Com
mon
Protein Accession code Calculated Mw (Da) Charge (z)
Cytochrome c, somatic NP_036971 11474 1a
Vesicle-associated membrane protein 3 NP_476438 11480 1a
Vesicle-associated membrane protein 5 NP_446007 11495 1a
Octadecaneuropeptide NP_114054 1912 3
39S ribosomal protein L41, mitochondrial NP_001013444 13818 1a
Trefoil factor 1 NP_476470 6912 1
60S ribosomal protein L41 NP_620783 3456 2
Protein B-Myc NP_001013181 18317 1a
60S ribosomal protein L21 NP_445782 18335 1a
UDP-N-acetylglucosamine transferase subunit ALG13 homolog NP_001013973 18329 1a
Corticotropin P01194 4582 2
TagIdent UniProtKB/Swiss-Prot search
Average relative level SEM
-0.7 0.04
0.26
0.09CM10 pH 6
CM10 pH 4
-
- -
-
NP20
0.8
0.46910
5743
9161
SELDI-TOF-MS m/z (z=1)
ProteinChip® array
-
-
Responsive to treatment with
a) Charge (z) = 1 but the possibility of dimerisation was considered
Chapter 3: Results
145
Table 11: Identification of SELDI-TOF-MS putative biomarkers of myopathy in
myoblasts (GWδ responders)
Myoblasts were grown and treated with 50μM GWδ for 24h (section 2.1). The protein
ions shown were responsive to treatment when measured using SELDI-TOF-MS
(section 2.6). The TagIdent tool (available via the ExPASy server) was used to assign
possible matches for the ions detected on ProteinChip® arrays. LC-MS/MS data was
also cross-referenced via Bioworks Browser (section 3.5). No possible identifications
were provided by TagIdent for the ions at m/z 4910 and 4913. Also, no identifications
were provided by interrogating LC-MS/MS data.
TMP
D
Gw
d
Com
mon
Protein Accession code Calculated Mw (Da) Charge (z)
4910 CM10 pH 4 - - 0.7 0.22 - - - -
4913 CM10 pH 8 - - 0.6 0.17 - - - -
Ig lambda-2 chain C region P20767 11318 1a
Vesicle-associated membrane protein 8 NP_114015 11320 1a
7222 CM10 pH 6 - - 0.4 0.04 Apelin-31 NP_113800 3607 2
Neighbor of COX4 NP_001012165 23406 1a
Coiled-coil domain-containing protein 153 NP_001013975 23407 1a
Membrane-associated progesterone receptor component 2 NP_001008375 23403 1a
Regulator of G-protein signaling 18 NP_001040549 27645 1a
N-terminal acetyltransferase complex ARD1 subunit homolog B NP_001019913 27623 1a
Pyridoxal phosphate phosphatase PHOSPHO2 NP_001007643 27627 1a
Monocyte to macrophage differentiation protein NP_001007674 27649 1a
39S ribosomal protein L41, mitochondrial NP_001013444 13818 1
Trefoil factor 1 NP_476470 6912 2
TagIdent UniProtKB/Swiss-Prot search
SELDI-TOF-MS m/z (z=1)
ProteinChip® array
Responsive to treatment with
SEM
5655 CM10 pH 6 -
Average relative level
- - 0.180.6
- 0.05-0.6-11708
CM10 pH 613814
H50
0.5 0.13-
a) Charge (z) = 1 but the possibility of dimerisation was considered
Chapter 3: Results
146
Table 12: Identification of SELDI-TOF-MS putative biomarkers of myopathy in myoblasts (common responders) Myoblasts were grown and treated with 25μM TMPD or 50μM GWδ (section 2.1). The
protein ions shown were responsive to treatment with both compounds when measured
using SELDI-TOF-MS (section 2.6). TagIdent was used to assign possible matches for
the protein ions. The probability that the assignment of the ion at m/z 9123 as 40S
ribosomal protein S21 (NP_112372) was correct was increased as it was identified
directly in myoblasts using LC-MS/MS (section 3.5) and also showed a similar
response to treatment with TMPD (relative level in TMPD-treated myoblasts = -0.3).
TMP
D
Gw
d
Com
mon
Protein Accession code Calculated Mw (Da) Charge (z)
Tubulin-specific chaperone A NP_001013263 12613 1a
Pituitary adenylate cyclase-activating polypeptide 27 NP_058685.1 3149 1
Somatostatin-28 NP_036791.1 3151 1
Nuclear transition protein 2 NP_058753.2 12849 1a
Cystatin-12 NP_714956.1 12829 1a
WAP four-disulfide core domain protein 12 NP_001008866.1 6425 1
Serine protease inhibitor Kazal-type 7 Q6IE51 6418 1
Melanotropin beta P01194 2141 3
Potassium voltage-gated channel subfamily E member 2 NP_598287.1 14357 1a
Transmembrane protein 100 NP_001017479.1 14352 1a
Atriopeptin-2 NP_036744.1 2389 3
Melanin-concentrating hormone NP_036757.1 2389 3
Protein ARMET P0C5H9 18207 1a
Epithelial membrane protein 3 NP_110474 18198 1a
Secretin NP_073161 3028 3
TagIdent UniProtKB/Swiss-Prot search
SELDI-TOF-MS m/z (z=1)
ProteinChip® array
Responsive to treatment with
NP20 -
-
--
-
-
9097
-
NP20 -
0.080.5
-0.5 0.04
-0.6 0.04
Average relative level SEM
-0.7 0.03
CM10 pH 6
6306
6420
7177
NP20
a) Charge (z) = 1 but the possibility of dimerisation was considered
Chapter 3: Results
147
Table 12 continued
TMP
D
Gw
d
Com
mon
Protein Accession code Calculated Mw (Da) Charge (z)
Gremlin-1 NP_062155 18239 1a
Protein ARMET P0C5H9 18207 1a
Alpha-S2-casein-like B NP_775129 18244 1a
Small muscular protein NP_445847 9121 1
C-terminal peptide NP_036758 3041 3
Gremlin-1 NP_062155 18239 1a
Interleukin-18 NP_062038 18261 1a
Alpha-S2-casein-like B NP_775129 18244 1a
40S ribosomal protein S21 NP_112373 9127 1
Small muscular protein NP_445847 9121 1
C-terminal peptide NP_036758 3041 3
Protein max NP_071546 18272 1a
Interleukin-18 NP_062038 18261 1a
T-cell surface glycoprotein CD3 gamma chain NP_001071114 18285 1a
40S ribosomal protein S21 NP_112373 9127 1
QRF-amide NP_937843 4572 2
9123
9137 -
0.03-0.7-
Responsive to treatment with
-
0.03-0.7
NP20
NP20 - 0.03-0.7
-NP20
TagIdent UniProtKB/Swiss-Prot search
-
Average relative level SEMSELDI-TOF-MS
m/z (z=1)ProteinChip®
array
9113
a) Charge (z) = 1 but the possibility of dimerisation was considered
Chapter 3: Results
148
Table 13: Identification of SELDI-TOF-MS putative biomarkers of myopathy in myotubes (TMPD responders) Myotubes were differentiated from myoblasts and treated with 25μM TMPD for 24h
(section 2.1). The protein ions shown were responsive to treatment when measured
using SELDI-TOF-MS (section 2.6). Mw information available on the ExPASy server
was interrogated using the TagIdent tool in order to discover possible matches for the
ions detected on ProteinChip® arrays. LC-MS/MS data generated following 1DE
analysis of myotubes was also interrogated via Bioworks Browser (section 3.5). No
identification possibilities were available for ions at m/z 3978 or 4815 although the ion
at m/z 6578 was a possible match with transcription elongation factor B (SIII),
polypeptide 2 (NP_ 112391) which was detected and identified by LC-MS/MS.
TMP
D
Gw
d
Com
mon
Protein Accession code Calculated Mw (Da) Charge (Z)
3978 CM10 pH 8 - - 0.95 0.18 - - - -
4815 CM10 pH 6 - - -0.4 0.04 - - - -
5579 IMAC-30 - - -0.2 0.05 Beta-defensin 11 NP_001032594.1 5581 1
6312 IMAC-30 - - -0.3 0.04 Tubulin-specific chaperone A NP_001013263.1 12613 1a
Protein JTB NP_062086.1 13160 1a
Bone morphogenetic protein 4 NP_036959.2 13155 1a6578 0.04-0.2--IMAC-30
Average relative level SEM
TagIdent UniProtKB/Swiss-Prot search
SELDI-TOF-MS m/z (z=1)
ProteinChip® array
Responsive to treatment with
a) Charge (z) = 1 but the possibility of dimerisation was considered
Chapter 3: Results
149
Table 14: Identification of SELDI-TOF-MS putative biomarkers of
myopathy in myotubes (GWδ responders)
Myotubes were cultured and treated with 50μM GWδ (section 2.1). The protein ions
shown were responsive to treatment when measured using SELDI-TOF-MS (section
2.6). TagIdent was used to assign possible matches for the protein ions based on Mw
information (with a tolerance of 0.1%, which is typical of SELDI-TOF-MS mass
accuracy). No identification possibilities were provided when LC-MS/MS data was
interrogated.
TMP
D
Gw
d
Com
mon
Protein Accession code Calculated Mw (Da) Charge (Z)
60S acidic ribosomal protein P1 NP_001007605.1 11498 1a
Vesicle-associated membrane protein 5 NP_446007.1 11495 1a
Gremlin-1 NP_062155.1 18239 1a
Interleukin-18 NP_062038.1 18261 1a
Alpha-S2-casein-like B NP_775129.1 18244 1a
40S ribosomal protein S21 NP_112373.1 9127 1
Small muscular protein NP_445847.1 9121 1
C-terminal peptide NP_036758.1 3041 3
SELDI-TOF-MS m/z (z=1)
ProteinChip® array
Responsive to treatment with
-9126
5753 0.0799
-
-
0.421.3
Average relative level SEM
CM10 pH 6
TagIdent UniProtKB/Swiss-Prot search
-0.5NP20 -
a) Charge (z) = 1 but the possibility of dimerisation was considered
Chapter 3: Results
150
Table 15: Identification of SELDI-TOF-MS putative biomarkers of myopathy in myotubes (common responders) Myotubes were differentiated from myoblasts and treated with 25μM TMPD or 50μM
GWδ (section 2.1). The protein ions shown were responsive to treatment when
measured using SELDI-TOF-MS (section 2.6) and TagIdent was used to assign
possible matches based on Mw information. No identification possibilities were
provided when LC-MS/MS data was interrogated. TM
PD
Gw
d
Com
mon
Protein Accession code Calculated Mw (Da) Charge (Z)
3158 CM10 pH 6 - - 1.3 0.23 - - - -
NADH-ubiquinone oxidoreductase NP_062096.1 10598 1
Endothelin-3 NP_001071118.1 2647 2
Tubulin-specific chaperone A NP_001013263.1 12613 1a
Pituitary adenylate cyclase-activating polypeptide 27 NP_058685.1 3149 2
Somatostatin-28 NP_036791.1 3151 2
Transcription elongation factor B polypeptide 2 NP_112391.1 13170 1a
Prolactin release-inhibiting factor 1 NP_036899.1 6585 1
Synaptogyrin-2 NP_446005.1 24711 1a
Regulator of G-protein signaling 19 NP_067693.1 24738 1a
Phosphatidylcholine transfer protein NP_058921.1 24734 1a
Neuron-specific vesicular protein calcyon NP_620270.1 24718 1a
60S ribosomal protein L10a NP_112327.1 24700 1a
Acyl-protein thioesterase 1 NP_037138.1 24709 1a
RPA-interacting protein NP_001028232.1 24710 1a
Sentrin-specific protease 8 NP_001012355.1 24728 1a
Transmembrane protein 204 NP_001009620.1 24701 1a
Follistatin-related protein 3 NP_446081.1 24719 1a
Fibroblast growth factor-binding protein 1 NP_072125.1 24696 1a
Macrophage migration inhibitory factor NP_112313.1 12346 1
Oligosaccharyl transferase subunit DAD1 NP_620265.1 12366 1
60S ribosomal protein L40 NP_113875.1 6182 2
SELDI-TOF-MS m/z (z=1)
ProteinChip® array
Responsive to treatment with
Average relative level SEM
- 0.080.8
0.01-0.4
0.35
0.07
1.3
-0.4
-
NP20
- -
--
NP20
NP20
-
-
12357
6588
6305
5298
TagIdent UniProtKB/Swiss-Prot search
CM10 pH 4
a) Charge (z) = 1 but the possibility of dimerisation was considered
Chapter 3: Results
151
The levels of several protein ions changed in response to the treatment of
skeletal muscle cells with either TMPD or GWδ. Although the changes seen by
SELDI-TOF-MS were useful in discriminating between control and treated cells,
this technique does not routinely provide the identifications of the proteins
measured on ProteinChip® arrays. The attempts made to conclusively identify
the responsive protein ions based on Mw information and using the TagIdent
tool available via the ExPASy server led to the inclusion of several possibilities.
The interrogation of LC-MS/MS data generated following the analysis of skeletal
muscle cells treated in the same way provided another possible means for
verifying the identity of responsive proteins that had been measured in the
samples. Furthermore, their response to treatment could be compared to that
observed by SELDI-TOF-MS.
For most cases, this method of discovering the identity of the protein ions
discovered was not useful, as the mass accuracy provided by SELDI-TOF-MS
does not allow the options for the identity of the proteins to be narrowed down
sufficiently to facilitate a conclusive identification of the proteins. The only
possible exception was for an ion that was responsive to the treatment of
myoblasts with TMPD and GWδ (at m/z 9123). The possibility that the identity of
this ion was 40S ribosomal protein S21 (NP_112373) was increased as the
protein was also measured by LS-MS/MS. Also comparing the response of the
protein as measured by SELDI-TOF-MS and LC-MS/MS increased the
likelihood of a correct identification although it is still not conclusive. Despite the
lack of a definite identification, the changes observed using SELDI-TOF-MS
could still be considered as part of an anonymous screening approach. An
approach to biomarker discovery that would facilitate the simultaneous
identification of the markers of interest alongside the quantitation would be
advantageous.
Chapter 3: Results
152
3.5 Discovery of protein biomarkers of myopathy using label-free quantitative proteomics
An anonymous approach to the discovery of compound-induced changes in the
levels of proteins in skeletal muscle cells was described in section 3.4. The
identity of the ions was not readily available and the dynamic range of the
observations was restricted to between 2 and 20kDa due to limitations in
SELDI-TOF-MS. A label-free approach to the identification of biomarkers of
compound effects was also attempted using skeletal muscle cells treated with a
concentration of TMPD that was not seen to be overtly cytotoxic to skeletal
muscle cells (25μM).
Replicate cell lysate proteins were prefractionated by size using 1DE and the
resultant polyacrylamide gel was sectioned before generating peptides through
the enzymatic digestion of the proteins in each Mw section of the gel. The
resulting peptide ions were then detected, quantified and identified using LC-
MS/MS before treatment-induced changes in the relative levels of some of the
putative protein biomarkers were confirmed using immunoblotting.
3.5.1 1D gel-electrophoresis of rat L6 muscle cells
Samples were separated by 1DE along with Mw markers ranging from 3-
198kDa, and gels were divided into sections as described in section 2.7.1
(Figure 53). The low Mw region (<3kDa) was excluded from further analyses as
proteins in this region generated an insufficient number of peptides to facilitate a
reliable identification (>2 peptides).
There were no treatment-related changes in the intensities of the protein bands
on 1D gels following the analysis of myoblasts (Figure 54) or myotubes (Figure
55). Although this meant that clear changes in specific areas of the gel could
not be focused on in future experiments, this method provided an effective way
of separating proteins within the samples.
Chapter 3: Results
153
M C C C C C C T T T T TM
198
98
62
49
38
28
17
14
6
3
C C C C C C T T T T TMw(kDa)
98
62
49
38
28
17
14
6
3
Slice 11
Slice 10
Slice 9
Slice 7
Slice 6
Slice 5Slice 4Slice 3
Slice 2
Slice 1
Slice 8
TM C C C C C C T T T T TM
198
98
62
49
38
28
17
14
6
3
C C C C C C T T T T TMw(kDa)
98
62
49
38
28
17
14
6
3
Slice 11
Slice 10
Slice 9
Slice 7
Slice 6
Slice 5Slice 4Slice 3
Slice 2
Slice 1
Slice 8Slice 8
T
Figure 53: Fractionation of 1DE gels Skeletal muscle cells were cultured and treated with 25μM TMPD (section 2.1). 1DE
was performed (section 2.7.1) and the resulting gels were cut transversely into 11 Mw
regions and longitudinally into lanes in order to separate samples (along the dotted
lines). In each gel, 6 control samples and 6 TMPD-treated samples were analysed,
resulting in 132 individual sections in which in-gel protein digestion was subsequently
performed using trypsin in order to generate peptides. M= Mw markers, C = control
samples and T = TMPD-treated samples.
Chapter 3: Results
154
M C C C C C C T T T T TMC C C C C C T T T T T
Slice 11
Slice 10
Slice 9
Slice 8
Slice 7
Slice 6
Slice 5Slice 4Slice 3Slice 2
Slice 1198
98
62
49
38
28
17
14
6
3
Mw(kDa)
98
62
49
38
28
17
14
6
3
TM C C C C C C T T T T TMC C C C C C T T T T T
Slice 11
Slice 10
Slice 9
Slice 8
Slice 7
Slice 6
Slice 5Slice 4Slice 3Slice 2
Slice 1198
98
62
49
38
28
17
14
6
3
Mw(kDa)
98
62
49
38
28
17
14
6
3
T
Figure 54: 1D gel of rat L6 myoblasts treated with vehicle control or 25µM TMPD Myoblasts were cultured and treated with 25μM TMPD (section 2.1). In each gel,
proteins in 6 control samples (C) and 6 TMPD-treated samples (T) were separated
(section 2.7.1) and the resulting gels were divided as shown in Figure 53. M= Mw
markers, C = control samples and T = TMPD-treated samples. Mw markers are shown
(M) and the positions where the gels were sliced are shown on the right.
Chapter 3: Results
155
M C C C C C C T T T T T TMC T T T T T T
198
98
62
49
38
28
17
14
6
3
Mw(kDa)
98
62
49
38
28
17
14
6
3Slice 11
Slice 10
Slice 9
Slice 8
Slice 7
Slice 6
Slice 5Slice 4Slice 3Slice 2
Slice 1
M C C C C C C T T T T T TMC T T T T T T
198
98
62
49
38
28
17
14
6
3
Mw(kDa)
98
62
49
38
28
17
14
6
3
198
98
62
49
38
28
17
14
6
3
Mw(kDa)
98
62
49
38
28
17
14
6
3Slice 11
Slice 10
Slice 9
Slice 8
Slice 7
Slice 6
Slice 5Slice 4Slice 3Slice 2
Slice 1
Figure 55: 1D gel of rat L6 myotubes treated with vehicle control or 25µM TMPD Myotubes were differentiated from myoblasts and treated with 25μM TMPD (section
2.1). Proteins in 6 control samples (C) and 6 TMPD-treated samples (T) were
separated (section 2.7.1) and the resulting gels were divided as shown in Figure 53.
M= Mw markers, C = control samples and T = TMPD-treated samples. Mw markers are
shown (M) and the positions where the gels were sliced are shown on the right.
Chapter 3: Results
156
3.5.2 Liquid chromatography-tandem mass spectrometry analysis of rat muscle cell homogenates
In the next stage of LFQP analysis, the peptides generated by enzymatic
digestion of the 1D gel sections were detected, measured and identified
simultaneously using a nano-LC system coupled with an LTQ-mass
spectrometer for automated LC-MS/MS analysis (section 2.7.3). This was
performed in order to ascertain whether treating cells with a low concentration
of TMPD (25μM) resulted in any changes in the levels of the peptide ions
measured in myoblasts and myotubes.
The expression of peptides was initially examined using search results files (srf)
generated using Bioworks Browser (as described in section 2.7.3). Proteins in
each Mw region were identified by at least two unique peptides during their
identification. In addition, the p value for the protein detected was <0.05 in order
to ensure confidence in the protein identification and proteins and peptides that
did not meet these criteria were excluded. In general, proteins were detected
within the Mw region that correlated with their calculated m/z although some
error was allowed for unexpected electrophoretic migration of proteins within the
gel. All the proteins detected reliably in control and treated cells are considered
here and the effects of treatment on the expression of proteins in considered in
subsequent sections. The proteins detected and identified in myoblasts are
detailed in Table 16. Proteins identified in myotubes are described in Table 17.
In total 2100 peptides were detected and 1314 (63%) of these contributed to the
identification of proteins in myoblasts. Out of the total number of peptides
detected, 1664 (79%) of the total number of peptides detected contributed to
the identification of proteins in myotubes. In myoblasts (and when considering
both control and treated cells), a total of 217 proteins were identified ranging
from 8249 to 918957 Da in size and 269 proteins were identified in myotubes
(ranging from 5763 to 918957). Of these proteins, 155 were common to both
myoblasts and myotubes (Figure 56).
Chapter 3: Results
157
62(436)
114(786)
155(878)
62(436)
114(786)
155(878)
Figure 56: Number of proteins identified in myoblasts and myotubes Myoblasts and myotubes were cultured (section 2.1) and the proteins in skeletal
muscle cell homogenates were separated by 1DE (section 2.7.1). Peptides were
generated from the different sections of the 1D gel by trypsin digestion and these were
detected, measured and identified by LC-MS/MS (sections 2.7.2 and 2.7.3). A total of
331 proteins were identified in myoblasts and myotubes. Of these proteins, 62 (19%)
and 114 (34%) were unique to myoblasts and myotubes, respectively. 155 (47%) of the
proteins identified were common to both the immature and differentiated form of the rat
L6 skeletal muscle cells. In identifying these proteins, a total of 2100 peptides were
measured (1314 in myoblasts and 1664 in myotubes).
Chapter 3: Results
158
Table 16: Proteins identified in rat L6 myoblasts Myoblasts were cultured (section 2.1) and cell homogenates were analysed by 1DE (section 2.7.1). Separated skeletal muscle cell samples were
divided into Mw regions in the resulting gel and peptides were generated by trypsin digestion. Peptides detected, measured and identified by LC-
MS/MS are shown (section 2.7.3). Only proteins identified reliably by two or more peptides are included.
Protein NCBI accession code Mw (Da) p value
Number of peptides used for
identification
Number of gel sections
detected in
MW regions detected in (No. peptides in each region)
Detected in control
myoblasts
Detected in treated
myoblasts
Also detected in myotubes
Mw range 6-14kDa
ATP synthase, H+ transporting, mitochondrial F0 complex, subunit E
NP_536729.1 8249 1.1E-04 2 1 3-6 (1)
D-dopachrome tautomerase NP 077045.1 13125 1.9E-04 2 2 6-14 (2)Heat shock 10 kDa protein 1 NP 037098.1 10869 7.4E-11 6 12 6-14 (12)Prothymosin alpha NP 068508.1 12375 9.0E-10 3 2 14-17 (2)Ribosomal protein S21 NP 112373.1 9122 9.2E-05 3 2 6-14 (2)S100 calcium binding protein A10 NP 112376.1 11067 8.7E-05 2 1 6-14 (1)S100 calcium binding protein A6 NP 445937.1 10028 6.4E-04 2 1 3-6 (1)S100 calcium-binding protein A4 NP 036750.1 11769 3.5E-06 6 15 3-6 (3), 6-14 (12)Thioredoxin NP 446252.1 11666 6.7E-08 5 12 6-14 (12)
Mw range 14-28kDa
Calmodulin 3 NP 036650.1 16827 6.8E-10 2 5 14-17 (5)Cytochrome c oxidase, subunit Va NP 665726.1 16119 1.5E-05 2 2 6-14 (2)Hemoglobin, epsilon 2 NP 001019976.1 16378 1.4E-04 2 1 75-98 (1)Histone cluster 1, H2aa NP 068611.1 14275 1.1E-08 4 18 14-17 (6), 3-6 (2), 6-14 (10)Lectin, galactoside-binding, soluble, 1 NP 063969.1 14847 1.5E-09 6 12 6-14 (12)Profilin 1 NP 071956.2 14948 1.1E-09 5 5 6-14 (5)Ribosomal protein S14 NP 073163.1 16249 7.5E-08 4 3 14-17 (3)Ribosomal protein S19 NP 001032423.1 16076 6.6E-05 6 5 14-17 (5)Superoxide dismutase 1, soluble NP 058746.1 15902 3.0E-09 4 4 14-17 (4)
Mw range 28-38kDa
Acidic (leucine-rich) nuclear phosphoprotein 32 family, member B
NP_571986.1 31042 5.4E-07 2 1 28-38 (1)
ADP-ribosylation factor 5 NP 077063.1 20517 2.5E-05 2 1 14-17 (1)Aldo-keto reductase family 1, member B10 (aldose reductase)
NP_001013102.1 35976 5.5E-04 2 1 28-38 (1)
Aldo-keto reductase family 1, member B8 NP 775159.1 36168 7.7E-04 3 2 28-38 (2)Aldo-keto reductase family 1, member C-like 2 NP 001008343.1 34478 1.2E-04 2 4 28-38 (4)Annexin A4 NP 077069.3 35871 9.7E-05 4 2 28-38 (2)
Chapter 3: Results
159
Table 16 continued
Protein NCBI accession code Mw (Da) p value
Number of peptides used for
identification
Number of gel sections
detected in
MW regions detected in (No. peptides in each region)
Detected in control
myoblasts
Detected in treated
myoblasts
Also detected in myotubes
Mw range 28-38kDa continued
Annexin A5 NP 037264.1 35722 2.2E-09 13 12 28-38 (12)Calcyclin binding protein NP 001004208.1 26525 5.7E-04 2 1 28-38 (1)Calumenin isoform b NP 001029070.1 37124 6.3E-10 3 11 38-49 (11)Chloride intracellular channel 1 NP 001002807.1 26964 6.6E-14 6 8 28-38 (8)Chloride intracellular channel 4 (mitochondrial) NP 114006.1 28615 1.8E-09 3 3 28-38 (3)Cofilin 1 NP 058843.1 18521 1.0E-06 3 5 17-28 (5)Diaphorase (NADH/NADPH) NP 058696.2 30927 1.4E-10 9 12 28-38 (12)F-actin capping protein beta subunit NP 001005903.1 30609 2.8E-05 2 1 28-38 (1)Glutaredoxin 3 NP 116003.2 37825 3.4E-04 2 1 98-198 (1)Glutathione S-transferase omega 1 NP 001007603.1 27665 1.4E-06 2 1 28-38 (1)Guanine nucleotide binding protein, beta polypeptide 2-like 1
NP_570090.1 35073 5.3E-09 4 8 28-38 (8)
Heterogeneous nuclear ribonucleoprotein A1 NP 058944.1 34191 1.2E-11 9 9 28-38 (5), 38-49 (4)Heterogeneous nuclear ribonucleoprotein C (C1/C2) NP 001020804.1 32837 1.3E-04 3 2 38-49 (2)Laminin receptor 1 NP 058834.1 32803 8.9E-15 8 20 28-38 (8), 38-49 (12)L-lactate dehydrogenase B NP 036727.1 36589 5.7E-10 4 8 28-38 (8)Malate dehydrogenase 1, NAD (soluble) NP 150238.1 36460 5.7E-09 5 6 28-38 (6)Malate dehydrogenase, mitochondrial NP 112413.2 35661 4.9E-11 9 11 28-38 (11)Myoglobin NP 067599.1 17146 4.1E-04 2 1 14-17 (1)Non-metastatic cells 2, protein (NM23B) expressed in NP 114021.2 17272 6.6E-05 4 6 14-17 (6)NudC domain containing 2 NP 001009621.1 17663 2.4E-05 2 1 14-17 (1)PDZ and LIM domain 1 NP 059061.1 35503 6.9E-13 5 12 38-49 (12)Peptidylprolyl isomerase A (cyclophilin A) NP 058797.1 17863 3.4E-10 7 11 14-17 (11)Peptidylprolyl isomerase B NP 071981.1 22788 1.6E-09 2 3 14-17 (1), 17-28 (2)Peroxiredoxin 1 NP 476455.1 22095 1.1E-05 4 12 14-17 (1), 17-28 (11)Peroxiredoxin 2 NP 058865.1 21770 4.9E-05 4 9 14-17 (1), 17-28 (8)Peroxiredoxin 4 NP 445964.1 30988 4.6E-04 2 2 17-28 (2)Peroxiredoxin 5 precursor NP 446062.1 22165 3.7E-07 3 1 14-17 (1)Phosphatidylethanolamine binding protein NP 058932.1 20788 4.6E-11 2 2 17-28 (2)Phosphoglycerate mutase 2 (muscle) NP 059024.1 28737 1.6E-08 3 5 28-38 (5)Prohibitin 2 NP 001013053.1 33292 1.6E-08 7 7 28-38 (7)Proliferating cell nuclear antigen NP 071776.1 28730 2.9E-12 6 9 28-38 (9)Proteasome (prosome, macropain) subunit, alpha type 4 NP 058977.1 29479 1.0E-06 3 7 28-38 (7)Proteasome (prosome, macropain) subunit, alpha type 7 NP 001008218.1 27839 3.5E-09 5 5 28-38 (5)RAN, member RAS oncogene family NP 445891.1 24408 2.4E-08 4 11 17-28 (6), 28-38 (5)Rho GDP dissociation inhibitor (GDI) alpha NP 001007006.1 23393 4.7E-12 5 14 17-28 (6), 28-38 (8)Ribosomal protein L11 NP 001020910.1 19012 5.8E-06 3 3 17-28 (3)Ribosomal protein S3 NP 001009239.1 26657 4.6E-12 13 11 6-14 (1), 28-38 (10)
Chapter 3: Results
160
Table 16 continued
Protein NCBI accession code Mw (Da) p value
Number of peptides used for
identification
Number of gel sections
detected in
MW regions detected in (No. peptides in each region)
Detected in control
myoblasts
Detected in treated
myoblasts
Also detected in myotubes
Mw range 28-38kDa continued
Secreted protein, acidic, cysteine-rich (osteonectin) NP 036788.1 34274 6.3E-10 2 1 38-49 (1)Solute carrier family 25, member 5 NP 476443.1 32880 2.2E-07 3 2 98-198 (1), 28-38 (1)Splicing factor, arginine/serine-rich 10 NP 476460.1 33646 2.4E-06 2 1 38-49 (1)Splicing factor, arginine/serine-rich 2 NP 001009720.1 25461 1.3E-04 2 6 28-38 (6)Splicing factor, arginine/serine-rich 7 NP 001034124.1 17880 3.5E-10 3 3 28-38 (3)Stathmin 1 NP 058862.1 17278 5.2E-05 3 3 14-17 (3)Transgelin NP 113737.1 22588 3.3E-07 6 12 14-17 (1), 17-28 (11)Transgelin 2 NP 001013145.1 22379 2.4E-06 2 5 14-17 (1), 17-28 (4)Tropomyosin 1, alpha isoform c NP 001029242.1 32828 2.9E-07 8 9 28-38 (4), 38-49 (5)Tropomyosin 1, alpha isoform h NP 001029246.1 28679 2.1E-09 13 20 28-38 (12), 38-49 (8)Tropomyosin 2 NP 001019516.1 32938 4.4E-08 11 20 28-38 (9), 38-49 (11)Tropomyosin 3, gamma isoform 1 NP 476556.2 28692 1.6E-08 10 12 28-38 (12)Tropomyosin 3, gamma isoform 2 NP 775134.1 29017 1.6E-06 2 6 28-38 (6)Tropomyosin 4 NP 036810.1 28492 4.3E-11 14 12 28-38 (12)Tyrosine 3-monooxgenase/tryptophan 5-monooxgenase activation protein, beta polypeptide
NP_062250.1 28037 4.6E-09 3 3 28-38 (3)
Tyrosine 3-monooxygenase/tryptophan 5-monooxygenase activation protein
NP_113791.1 29103 2.6E-05 5 6 28-38 (6)
Tyrosine 3-monooxygenase/tryptophan 5-monooxygenase activation protein, eta polypeptide
NP_037184.1 28194 5.6E-04 4 2 28-38 (2)
Tyrosine 3-monooxygenase/tryptophan 5-monooxygenase activation protein, gamma polypeptide
NP_062249.1 28285 1.4E-11 4 9 28-38 (9)
Tyrosine 3-monooxygenase/tryptophan 5-monooxygenase activation protein, theta polypeptide
NP_037185.1 27761 5.1E-10 3 6 28-38 (6)
Tyrosine 3-monooxygenase/tryptophan 5-monooxygenase activation protein, zeta polypeptide
NP_037143.2 27754 1.4E-12 8 13 17-28 (1), 28-38 (12)
Ubiquitin-conjugating enzyme E2N NP 446380.1 17113 1.6E-06 2 2 14-17 (2)Voltage-dependent anion channel 1 NP 112643.1 30737 1.3E-06 3 2 28-38 (2)Voltage-dependent anion channel 2 NP 112644.1 31726 3.4E-06 4 7 28-38 (7)V-set domain containing T cell activation inhibitor 1 NP 001019415.1 30644 3.4E-04 2 1 38-49 (1)
Mw range 38-49kDa
Actin, alpha 1, skeletal muscle NP_062085.1 42024 9.0E-13 17 60 >198 (10), 75-98 (10), 98-198 (12), 49-62 (1), 28-38 (6), 38-49
(12), 14-17 (2), 6-14 (7)Aldolase A-like 1 NP 001013965.1 39467 1.1E-08 8 12 38-49 (12)
Chapter 3: Results
161
Table 16 continued
Protein NCBI accession code Mw (Da) p value
Number of peptides used for
identification
Number of gel sections
detected in
MW regions detected in (No. peptides in each region)
Detected in control
myoblasts
Detected in treated
myoblasts
Also detected in myotubes
Mw range 38-49kDa continued
Annexin A1 NP 037036.1 38805 2.3E-12 20 13 28-38 (1), 38-49 (12)Annexin A2 NP_063970.1 38654 3.1E-11 20 27 >198 (2), 6-14 (2), 28-38 (11), 38-
49 (12)ARP2 actin-related protein 2 homolog NP 001009268.1 44705 6.2E-06 2 1 38-49 (1)Calreticulin NP 071794.1 47966 1.8E-07 4 4 49-62 (4)Eukaryotic translation initiation factor 4A2 NP 001008336.1 46373 3.4E-10 5 12 38-49 (12)Fructose-bisphosphate aldolase A NP 036627.1 39327 1.0E-30 4 12 38-49 (12)Glutamate oxaloacetate transaminase 2 NP 037309.1 47284 4.1E-06 2 1 38-49 (1)Heterogeneous nuclear ribonucleoprotein A3 isoform a NP 937765.1 39628 4.1E-07 5 3 28-38 (1), 38-49 (2)Heterogeneous nuclear ribonucleoprotein F NP 001032362.1 45701 8.9E-10 4 3 38-49 (3)Keratin 13 NP_001004021.1 47699 6.2E-07 5 8 >198 (3), 14-17 (1), 75-98 (1), 98-
198 (1), 3-6 (2)Keratin 15 NP_001004022.1 48840 2.7E-06 4 10 >198 (2), 75-98 (1), 62-75 (1), 38-
49 (1), 14-17 (1), 3-6 (3), 6-14 (1)Keratin 19 NP 955792.1 44609 2.3E-04 3 4 3-6 (3), 6-14 (1)Nucleosome assembly protein 1-like 4 NP 001012170.1 43890 6.0E-07 2 1 49-62 (1)Phosphoglycerate kinase 2 NP 001012130.1 44981 4.6E-11 3 6 38-49 (6)Poly(rC) binding protein 2 NP 001013241.1 38556 4.1E-04 3 3 38-49 (3)Protein disulfide isomerase-associated 6 NP 001004442.1 48730 2.1E-10 6 12 38-49 (12)RAB3A interacting protein (rabin3)-like 1 NP 599238.1 41881 4.4E-04 2 1 62-75 (1)Serine (or cysteine) peptidase inhibitor, clade H, member NP 058869.1 46488 2.1E-13 10 19 75-98 (7), 38-49 (12)
Mw range 49-62kDa
Aldehyde dehydrogenase 3A1 NP 114178.1 50307 1.6E-09 12 21 49-62 (9), 38-49 (12)Apoptosis antagonizing transcription factor NP 446172.1 59461 8.2E-04 2 1 75-98 (1)ATP synthase, H+ transporting, mitochondrial F1 complex, alpha subunit 1, cardiac muscle
NP_075581.1 59717 2.1E-09 8 14 49-62 (10), 28-38 (4)
Caldesmon 1 NP 037278.1 60548 2.0E-11 9 12 75-98 (12)Cathepsin F NP 001029282.1 51796 4.1E-05 2 2 75-98 (2)Chaperonin containing TCP1, subunit 2 (beta) NP 001005905.1 57422 1.6E-06 4 6 49-62 (6)Chaperonin containing TCP1, subunit 3 (gamma) NP 954522.1 60608 3.7E-06 5 6 62-75 (6)Chaperonin containing TCP1, subunit 5 (epsilon) NP 001004078.1 59499 9.3E-07 4 2 62-75 (2)Chaperonin containing Tcp1, subunit 6A (zeta 1) NP 001028856.1 57981 4.8E-08 8 9 49-62 (9)Chaperonin subunit 4 (delta) NP 877966.1 58063 3.5E-07 4 8 49-62 (8)Cortactin isoform B NP 068640.2 56907 8.4E-05 3 2 75-98 (2)Dihydropyrimidinase-like 3 NP 037066.1 61928 8.1E-09 4 11 75-98 (11)Eukaryotic translation elongation factor 1 alpha 2 NP_036792.2 50422 9.4E-07 7 27 >198 (2), 98-198 (2), 49-62 (7),
28-38 (4), 38-49 (12)Glucose-6-phosphate dehydrogenase NP 058702.1 59338 4.3E-06 2 1 49-62 (1)
Chapter 3: Results
162
Table 16 continued
Protein NCBI accession code Mw (Da) p value
Number of peptides used for
identification
Number of gel sections
detected in
MW regions detected in (No. peptides in each region)
Detected in control
myoblasts
Detected in treated
myoblasts
Also detected in myotubes
Mw range 49-62kDa continued
Glutamate dehydrogenase 1 NP 036702.1 61377 6.2E-10 6 9 49-62 (9)Heterogeneous nuclear ribonucleoprotein K NP 476482.1 50944 3.7E-11 5 10 49-62 (10)HLA-B associated transcript 1 NP 579834.2 49004 4.1E-04 2 2 38-49 (2)Keratin 10 NP_001008804.1 56471 2.4E-10 10 52 >198 (10), 75-98 (2), 98-198 (6),
49-62 (5), 62-75 (1), 28-38 (5), 38-49 (1), 14-17 (6), 3-6 (10), 6-
Keratin 14 NP 001008751.1 52651 2.6E-06 2 1 >198 (1)Keratin 4 NP_001008806.1 57631 1.1E-04 5 24 >198 (3), 75-98 (1), 49-62 (3), 28-
38 (6), 38-49 (3), 14-17 (1), 17-28 (2), 3-6 (3), 6-14 (2)
Keratin 42 NP_001008816.1 50182 1.6E-06 7 22 >198 (3), 75-98 (1), 98-198 (1), 49-62 (2), 38-49 (2), 14-17 (3), 3-
6 (8), 6-14 (2)Keratin 5 NP 899162.1 61889 6.4E-06 7 3 14-17 (1), 98-198 (1), 38-49 (1)Keratin 77 NP_001008807.1 57220 1.0E-08 7 120 >198 (11), 75-98 (11), 98-198
(12), 49-62 (12), 62-75 (11), 28-38 (11), 38-49 (11), 14-17 (9), 17-
28 (9) 3-6 (12) 6-14 (11)Mitochondrial aldehyde dehydrogenase 2 NP 115792.1 56453 9.9E-10 2 1 49-62 (1)Mitochondrial ATP synthase beta subunit NP 599191.1 56319 2.2E-10 7 12 49-62 (11), 38-49 (1)PCTAIRE protein kinase 1 isoform b NP 112339.1 52490 1.6E-05 2 1 49-62 (1)Phenylalanyl-tRNA synthetase 2, mitochondrial NP 001013157.1 55130 2.8E-04 2 1 28-38 (1)Prolyl 4-hydroxylase, beta polypeptide NP 037130.1 56829 5.1E-11 19 12 49-62 (12)Prosaposin NP 037145.1 61084 5.1E-04 2 1 6-14 (1)Protein disulfide-isomerase A3 NP 059015.1 56554 2.0E-08 17 12 49-62 (12)Pyruvate kinase, muscle NP 445749.1 57781 2.1E-11 15 14 49-62 (12), 62-75 (2)Ribonuclease/angiogenin inhibitor 1 NP 620805.1 49873 1.1E-08 3 3 38-49 (3)Solute carrier family 3, member 2 NP 062156.1 58036 3.5E-06 2 1 75-98 (1)Tubulin, beta 5 NP 775125.1 49639 4.5E-10 2 18 49-62 (6), 38-49 (12)Vimentin NP_112402.1 53700 4.9E-12 32 48 3-6 (12), 6-14 (12), 49-62 (12),
38-49 (12)
Mw range 62-70kDa
Alpha-fetoprotein NP 036625.1 68341 3.1E-04 2 2 38-49 (2)Calnexin NP 742005.1 67213 1.6E-13 3 12 75-98 (12)Eukaryotic translation initiation factor 4B NP 001008325.1 69023 5.1E-07 6 6 75-98 (6)Ezrin NP 062230.1 69348 1.5E-08 3 1 75-98 (1)
Chapter 3: Results
163
Table 16 continued
Protein NCBI accession code Mw (Da) p value
Number of peptides used for
identification
Number of gel sections
detected in
MW regions detected in (No. peptides in each region)
Detected in control
myoblasts
Detected in treated
myoblasts
Also detected in myotubes
Mw range 62-70kDa continued
Fragile X mental retardation 1 NP 434691.1 65591 4.8E-04 2 1 62-75 (1)Keratin 2 NP_001008899.1 69085 9.9E-10 7 16 >198 (5), 98-198 (2), 49-62 (1),
62-75 (1), 38-49 (2), 14-17 (2), 3-6 (1), 6-14 (2)
Keratin 75 NP 001008828.1 62130 2.3E-05 3 2 >198 (2)
Mw range 70-98kDa
Moesin NP 110490.1 67697 1.9E-04 7 4 75-98 (4)Radixin NP 001005889.2 68501 2.9E-09 8 16 75-98 (12), 62-75 (4)WD repeat domain 1 NP 001014157.1 66140 8.5E-06 3 3 62-75 (3)Cd44 molecule NP 037056.2 85865 5.9E-05 2 1 62-75 (1)Dynein cytoplasmic 1 intermediate chain 2 NP 446332.1 71134 3.9E-04 2 1 75-98 (1)Eukaryotic translation elongation factor 1 delta NP 001013122.1 72083 3.4E-09 5 11 28-38 (11)Eukaryotic translation elongation factor 2 NP 058941.1 95223 1.6E-13 20 17 75-98 (12), 98-198 (5)F-box protein 11 NP 853662.1 93983 7.1E-05 2 1 38-49 (1)Gelsolin NP 001004080.1 86014 2.4E-07 4 6 75-98 (6)Heat shock 105kDa/110kDa protein 1 NP 001011901.1 96357 5.2E-06 2 5 75-98 (5)Heat shock 70kD protein 1-like NP 997711.1 70505 2.1E-11 7 21 75-98 (9), 62-75 (12)Heat shock protein 4 NP 705893.1 93997 2.1E-06 4 2 75-98 (2)Heat shock protein 5 NP 037215.1 72303 1.2E-10 17 21 75-98 (10), 62-75 (11)Heat shock protein 8 NP 077327.1 70827 4.7E-11 12 21 75-98 (9), 62-75 (12)Heat shock protein 90, alpha (cytosolic), class A member 1
NP_786937.1 84762 5.2E-13 17 12 75-98 (12)
Lamin A isoform C2 NP 001002016.1 71856 5.3E-07 17 19 75-98 (12), 49-62 (2), 62-75 (5)Neural cell adhesion molecule 1 NP 113709.1 94599 1.4E-07 4 9 98-198 (9)Neurolysin (metallopeptidase M3 family) NP 446422.2 80231 8.2E-04 2 1 38-49 (1)Nuclear RNA export factor 1 NP 067590.1 70318 9.9E-05 2 1 6-14 (1)Poly(A) binding protein, cytoplasmic 1 NP 599180.1 70656 3.0E-05 4 4 75-98 (4)Protein disulfide isomerase-associated 4 NP 446301.1 72761 1.2E-05 4 8 75-98 (8)RAD17 homolog NP 001019949.1 77233 4.0E-04 2 1 >198 (1)Ribonucleotide reductase M1 NP 001013254.1 90236 9.7E-04 2 1 75-98 (1)Signal transducer and activator of transcription 5A NP 058760.1 90776 7.5E-04 3 1 >198 (1)TNF receptor-associated protein 1 NP 001034090.1 80411 1.1E-08 4 5 75-98 (5)Transferrin NP 001013128.1 76346 2.5E-04 2 2 75-98 (2)Transient receptor potential cation channel, subfamily C, member 1
NP_446010.1 87561 9.3E-04 2 1 75-98 (1)
Transketolase NP 072114.1 71141 2.1E-06 4 1 62-75 (1)Valosin-containing protein NP 446316.1 89293 1.8E-08 14 12 75-98 (12)
Chapter 3: Results
164
Table 16 continued
Protein NCBI accession code Mw (Da) p value
Number of peptides used for
identification
Number of gel sections
detected in
MW regions detected in (No. peptides in each region)
Detected in control
myoblasts
Detected in treated
myoblasts
Also detected in myotubes
Mw range 98-198kDa continued
Actinin, alpha 1 NP 112267.1 102896 3.2E-12 23 15 75-98 (12), 98-198 (3)Aldehyde oxidase 3 NP 001008527.1 146656 1.1E-05 2 1 38-49 (1)Alpha actinin 4 NP 113863.2 104849 2.8E-10 21 12 75-98 (12)ATP-binding cassette, sub-family C (CFTR/MRP), member 2
NP_036965.1 173273 4.1E-04 2 1 49-62 (1)
Clathrin, heavy chain (Hc) NP 062172.1 191476 8.5E-08 17 10 98-198 (10)Coatomer protein complex, subunit beta 1 NP 542959.1 106943 6.0E-05 2 1 75-98 (1)Coatomer protein complex, subunit beta 2 (beta prime) NP 068533.1 102486 2.7E-06 3 5 75-98 (4), 98-198 (1)Collagen, type I, alpha 2 NP 445808.1 129486 8.7E-13 18 12 98-198 (12)Collagen, type II, alpha 1 NP 037061.1 134488 1.5E-05 3 1 98-198 (1)Collagen, type III, alpha 1 NP 114474.1 138851 1.6E-09 9 6 98-198 (6)Cullin-associated and neddylation-dissociated 1 NP 446456.1 136275 1.7E-06 9 6 98-198 (6)Exportin 1, CRM1 homolog NP 445942.1 122959 1.2E-04 2 1 28-38 (1)Fibronectin type III domain containing 1 NP 001033704.1 188873 1.0E-04 2 1 3-6 (1)Glutamyl-prolyl-tRNA synthetase NP 001019409.1 166675 3.7E-12 3 2 98-198 (2)Golgi apparatus protein 1 NP 058907.1 133469 6.7E-04 3 1 98-198 (1)Inter-alpha trypsin inhibitor, heavy chain 3 NP 059047.1 99036 5.7E-04 2 1 98-198 (1)Kinesin family member 5B NP 476550.1 109463 1.8E-06 2 1 98-198 (1)Leucine-rich PPR-motif containing NP 001008519.1 156553 5.1E-08 2 1 98-198 (1)Leucyl-tRNA synthetase NP 001009637.1 134192 9.1E-05 3 2 98-198 (2)Microtubule-associated protein 4 NP 001019449.1 110233 5.7E-07 5 4 98-198 (4)Phosphoinositide-3-kinase, class 3 NP 075247.1 101470 2.2E-04 3 1 49-62 (1)Pregnancy-zone protein NP 665722.1 167053 3.2E-04 2 1 >198 (1)Rho-associated coiled-coil containing protein kinase 1 NP 112360.1 159526 8.9E-04 2 1 98-198 (1)SEC31 homolog A NP 148981.1 135266 5.7E-07 5 5 98-198 (5)Staphylococcal nuclease domain containing 1 NP 073185.2 101889 9.6E-09 8 7 75-98 (7)TAO kinase 3 NP 001019425.1 105408 1.9E-04 2 1 28-38 (1)Thrombospondin 1 NP 001013080.1 129588 1.0E-05 3 3 98-198 (3)Ubiquitin-activating enzyme E1 NP 001014102.1 117713 6.1E-06 2 2 75-98 (2)
Chapter 3: Results
165
Table 16 continued
Protein NCBI accession code Mw (Da) p value
Number of peptides used for
identification
Number of gel sections
detected in
MW regions detected in (No. peptides in each region)
Detected in control
myoblasts
Detected in treated
myoblasts
Also detected in myotubes
Mw range >198kDa
A-kinase anchor protein 6 NP 072140.1 254192 3.4E-04 2 1 75-98 (1)Alpha-spectrin 2 NP 741984.2 284418 4.3E-08 13 5 >198 (5)ATP-binding cassette, sub-family A (ABC1), member 2 NP 077372.1 270753 4.1E-04 2 1 75-98 (1)Cadherin, EGF LAG seven-pass G-type receptor 3 (flamingo homolog, Drosophila)
NP_112610.1 359128 4.0E-04 2 1 38-49 (1)
Fatty acid synthase NP 059028.1 272477 2.9E-09 8 13 >198 (10), 98-198 (3)HLA-B associated transcript 2 NP 997627.1 228905 6.9E-04 2 1 >198 (1)Inositol 1,4,5-triphosphate receptor, type 2 NP 112308.1 306861 6.8E-04 2 1 62-75 (1)Kalirin, RhoGEF kinase NP 114451.1 336374 3.3E-04 2 1 14-17 (1)Multiple PDZ domain protein NP 062069.1 218455 6.7E-04 2 1 6-14 (1)Myosin, heavy chain 10, non-muscle NP 113708.1 228823 1.5E-07 5 9 98-198 (9)Myosin, heavy chain 9, non-muscle NP 037326.1 226196 7.2E-12 42 22 >198 (10), 98-198 (12)Nestin NP 037119.1 198625 8.2E-12 8 7 >198 (7)Piccolo isoform 1 NP 064483.1 552380 1.6E-04 2 1 >198 (1)Plectin 1 NP 071796.1 533214 1.5E-12 118 15 >198 (12), 98-198 (3)Spectrin repeat containing, nuclear envelope 1 NP 001025080.1 918957 1.8E-04 2 1 >198 (1)Spectrin, beta, non-erythrocytic 1 NP 001013148.1 273415 8.3E-05 6 3 >198 (2), 28-38 (1)
Chapter 3: Results
166
Table 17: Proteins identified in rat L6 myotubes Myotubes were cultured (section 2.1) and cell homogenates were analysed by 1DE (section 2.7.1). Separated skeletal muscle cell samples were then
divided into Mw regions in the resulting gel and peptides were generated by trypsin digestion. Peptides detected, measured and identified by LC-
MS/MS are shown here (section 2.7.3). Only proteins identified reliably by two or more peptides are included.
Protein NCBI accession code Mw (Da) p value
Number of peptides used for
identification
MW regions detected in (No. peptides in each region)
Detected in control
myotubes
Detected in treated
myotubes
Also detected in myoblasts
Mw range 3-6kDa
ATP synthase, H+ transporting, mitochondrial F1 complex, epsilon subunit
NP_620799.1 5763 1.8E-04 2 3-6 (4)
Mw range 6-14kDa
ATP synthase, H+ transporting, mitochondrial F0 complex, subunit E
NP_536729.1 8249 5.0E-06 2 6-14 (5), 3-6 (5)
D-dopachrome tautomerase NP 077045.1 13125 6.0E-06 3 6-14 (3)Heat shock 10 kDa protein 1 NP 037098.1 10869 2.5E-11 6 6-14 (12)Prothymosin alpha NP 068508.1 12375 7.0E-10 5 14-17 (12)S100 calcium-binding protein A4 NP 036750.1 11769 8.3E-06 4 6-14 (12), 3-6 (9)Thioredoxin NP 446252.1 11666 2.5E-07 6 6-14 (11)Vesicle-associated membrane protein 2 NP 036795.1 12683 6.9E-06 3 6-14 (1), 14-17 (1)
Mw range 14-17kDa
Calmodulin 3 NP 036650.1 16827 2.6E-13 3 14-17 (11)Epididymal secretory protein E1 NP 775141.1 16353 4.3E-04 2 14-17 (1)Eukaryotic translation initiation factor 5A NP 001028853.1 16821 2.6E-05 3 14-17 (2)Histone cluster 1, H2aa NP_068611.1 14275 1.5E-08 5 >198 (1), 17-28 (6), 28-38 (2), 6-
14 (11), 14-17 (12), 3-6 (2)Lectin, galactoside-binding, soluble, 1 NP 063969.1 14847 2.1E-09 6 6-14 (12)Phosphoprotein enriched in astrocytes 15A NP 001013249.1 15031 6.0E-04 2 14-17 (1)Profilin 1 NP 071956.2 14948 1.3E-10 5 6-14 (7)Profilin 2 NP 110500.1 16104 3.6E-09 2 6-14 (1)Ribosomal protein S14 NP 073163.1 16249 1.8E-09 4 14-17 (12)Ribosomal protein S19 NP 001032423.1 16076 8.0E-05 6 14-17 (10)Ribosomal protein S23 NP 511172.1 15798 4.4E-05 3 14-17 (5)Superoxide dismutase 1, soluble NP 058746.1 15902 3.7E-10 5 14-17 (12)X-linked eukaryotic translation initiation factor 1A NP 001008773.2 16502 8.3E-04 2 14-17 (1)
Chapter 3: Results
167
Table 17 continued
Protein NCBI accession code Mw (Da) p value
Number of peptides used for
identification
MW regions detected in (No. peptides in each region)
Detected in control
myotubes
Detected in treated
myotubes
Also detected in myoblasts
Mw range 17-28kDa
Adenine phosphoribosyl transferase NP 001013079.1 19533 4.6E-05 2 17-28 (2)Adenylate kinase 1 NP 077325.2 21570 1.3E-10 3 17-28 (11)Adenylate kinase 3 NP 037350.1 25422 1.5E-06 7 17-28 (12)ADP-ribosylation factor 5 NP 077063.1 20517 1.1E-05 2 14-17 (10)ATP synthase, H+ transporting, mitochondrial F1 complex, delta subunit precursor
NP_620806.1 17584 4.1E-08 2 14-17 (4)
Chloride intracellular channel 1 NP 001002807.1 26964 1.2E-13 3 28-38 (2)Cofilin 1 NP 058843.1 18521 5.2E-13 4 17-28 (12)Crystallin, alpha B NP 037067.1 20076 1.5E-04 3 6-14 (4)Cysteine and glycine-rich protein 1 NP 058844.1 20600 2.0E-07 3 17-28 (5)Cysteine and glycine-rich protein 2 NP 803174.2 20912 1.2E-10 3 17-28 (9)Cytidine monophosphate (UMP-CMP) kinase 1, cytosolic
NP_001020826.1 25816 1.0E-04 3 17-28 (1)
Cytochrome c oxidase subunit IV isoform 1 NP 058898.1 19502 1.1E-06 2 6-14 (4)Glutathione S-transferase omega 1 NP 001007603.1 27665 3.4E-07 2 28-38 (1)Mitochondrial ATP synthase, O subunit NP 620238.1 23383 5.1E-10 5 17-28 (5)Myosin light chain, phosphorylatable, fast skeletal muscle
NP_036737.1 18957 5.4E-12 12 17-28 (2), 14-17 (12)
Non-metastatic cells 2, protein (NM23B) expressed in
NP_114021.2 17272 1.8E-07 4 14-17 (11)
Parkinson disease protein 7 NP 476484.1 19961 2.0E-05 4 17-28 (4)Peptidylprolyl isomerase A (cyclophilin A) NP 058797.1 17863 1.5E-10 6 14-17 (12)Peptidylprolyl isomerase B NP 071981.1 22788 1.7E-09 5 17-28 (12)Peroxiredoxin 1 NP 476455.1 22095 6.5E-08 6 17-28 (12)Peroxiredoxin 2 NP 058865.1 21770 1.3E-06 4 17-28 (11)Peroxiredoxin 5 precursor NP 446062.1 22165 1.8E-11 4 14-17 (12)Peroxiredoxin 6 NP 446028.1 24803 5.6E-06 4 17-28 (3)Phosphatidylethanolamine binding protein NP 058932.1 20788 4.1E-14 4 17-28 (12)Proteasome (prosome, macropain) subunit, alpha type 5
NP_058978.1 26374 1.4E-05 2 17-28 (2)
Proteasome (prosome, macropain) subunit, alpha type 7
NP_001008218.1 27839 3.0E-08 6 28-38 (7)
Proteasome (prosome, macropain) subunit, beta type 1
NP_446042.1 26462 1.0E-05 3 17-28 (4)
Proteasome (prosome, macropain) subunit, beta type 6
NP_476440.2 25273 7.5E-08 2 17-28 (12)
Proteasome beta 3 subunit NP 058981.1 22949 8.7E-08 2 17-28 (2)RAB11a, member RAS oncogene family NP 112414.1 24378 1.6E-08 3 17-28 (5)RAB7A, member RAS oncogene family NP 076440.1 23489 4.8E-10 7 17-28 (11)RAN, member RAS oncogene family NP 445891.1 24408 7.6E-09 5 17-28 (4), 28-38 (6)
Chapter 3: Results
168
Table 17 continued
Protein NCBI accession code Mw (Da) p value
Number of peptides used for
identification
MW regions detected in (No. peptides in each region)
Detected in control
myotubes
Detected in treated
myotubes
Also detected in myoblasts
Mw range 17-28kDa continued
Rho GDP dissociation inhibitor (GDI) alpha NP 001007006.1 23393 4.9E-12 4 17-28 (8), 28-38 (10)Ribosomal protein L11 NP 001020910.1 19012 6.1E-10 5 17-28 (12)Ribosomal protein S3 NP 001009239.1 26657 3.4E-08 7 28-38 (8), 6-14 (1), >198 (1)Splicing factor, arginine/serine-rich 2 NP 001009720.1 25461 1.2E-04 2 28-38 (7)Splicing factor, arginine/serine-rich 7 NP 001034124.1 17880 1.0E-06 2 28-38 (1)Stathmin 1 NP 058862.1 17278 4.4E-06 3 14-17 (12)Stathmin-like 2 NP 445892.1 20743 1.4E-05 2 14-17 (3)Superoxide dismutase 2 NP 058747.1 24659 1.6E-07 6 17-28 (12)Transgelin NP 113737.1 22588 8.1E-07 10 17-28 (12)Transgelin 2 NP 001013145.1 22379 3.8E-05 3 17-28 (8)Transmembrane emp24 protein transport domain containing 9
NP_001009703.1 27011 2.9E-05 2 17-28 (1)
Troponin C type 1 (slow) NP 001029277.1 18409 6.6E-10 4 14-17 (10)Trypsin 10 precursor NP 001004097.1 26109 3.2E-06 2 49-62 (1), 62-75 (2)Tyrosine 3-monooxygenase/tryptophan 5-monooxygenase activation protein, theta polypeptide
NP_037185.1 27761 1.1E-09 2 28-38 (6)
Tyrosine 3-monooxygenase/tryptophan 5-monooxygenase activation protein, zeta polypeptide
NP_037143.2 27754 2.6E-13 9 17-28 (1), 28-38 (12)
Ubiquitin-conjugating enzyme E2N NP 446380.1 17113 1.4E-06 3 14-17 (11)
Mw range 28-38kDa
Aldo-keto reductase family 1, member A1 (aldehyde reductase)
NP_112262.1 36483 2.6E-07 5 28-38 (5)
Aldo-keto reductase family 1, member C-like 2 NP 001008343.1 34478 2.6E-04 3 28-38 (2)Annexin A4 NP 077069.3 35871 5.3E-09 6 28-38 (5)Annexin A5 NP 037264.1 35722 1.9E-07 13 17-28 (3), 28-38 (12)Calponin 3, acidic NP 062232.1 36412 2.8E-04 2 38-49 (1)Calumenin isoform b NP 001029070.1 37124 1.6E-08 2 38-49 (10)Chloride intracellular channel 4 (mitochondrial) NP 114006.1 28615 1.9E-05 2 28-38 (1)Diaphorase (NADH/NADPH) NP 058696.2 30927 1.0E-09 8 28-38 (11)Dimethylarginine dimethylaminohydrolase 2 NP 997697.1 29669 5.0E-04 2 28-38 (1)Electron-transfer-flavoprotein, alpha polypeptide NP 001009668.1 34929 5.2E-04 3 28-38 (3)Endoplasmic reticulum protein 29 NP 446413.1 28557 4.1E-07 5 17-28 (2), 28-38 (3)F-actin capping protein beta subunit NP 001005903.1 30609 1.1E-05 4 28-38 (4)Guanine nucleotide binding protein, beta polypeptide 2-like 1
NP_570090.1 35073 5.4E-09 3 28-38 (11)
Heterogeneous nuclear ribonucleoprotein A1 NP 058944.1 34191 1.7E-11 8 38-49 (4), 28-38 (8)Laminin receptor 1 NP 058834.1 32803 1.1E-15 6 38-49 (8), 28-38 (5)
Chapter 3: Results
169
Table 17 continued
Protein NCBI accession code Mw (Da) p value
Number of peptides used for
identification
MW regions detected in (No. peptides in each region)
Detected in control
myotubes
Detected in treated
myotubes
Also detected in myoblasts
Mw range 28-38kDa continued
L-lactate dehydrogenase B NP 036727.1 36589 3.0E-09 4 28-38 (6)Malate dehydrogenase 1, NAD (soluble) NP 150238.1 36460 1.2E-09 8 28-38 (10)Malate dehydrogenase, mitochondrial NP 112413.2 35661 1.0E-30 11 28-38 (11)PDZ and LIM domain 1 NP 059061.1 35503 1.4E-10 3 38-49 (8)Peroxiredoxin 3 NP 071985.1 28303 7.1E-07 4 17-28 (8)Peroxiredoxin 4 NP 445964.1 30988 1.5E-06 3 17-28 (12)Phosphoglycerate mutase 2 (muscle) NP 059024.1 28737 1.5E-08 4 28-38 (10)Prohibitin 2 NP 001013053.1 33292 4.2E-09 6 28-38 (8)Proliferating cell nuclear antigen NP 071776.1 28730 9.8E-11 3 28-38 (4)Protease (prosome, macropain) 28 subunit, alpha NP 058960.2 28617 2.4E-05 3 28-38 (5)Proteasome (prosome, macropain) subunit, alpha type 3
NP_058976.1 28401 3.3E-06 2 28-38 (1)
Proteasome (prosome, macropain) subunit, alpha type 4
NP_058977.1 29479 1.5E-07 2 28-38 (9)
Ribosomal protein S6 NP 058856.1 28663 2.5E-04 3 3-6 (3)Secreted protein, acidic, cysteine-rich (osteonectin)
NP_036788.1 34274 1.2E-11 3 38-49 (4)
Solute carrier family 25, member 5 NP 476443.1 32880 4.8E-05 2 >198 (1)Tropomyosin 1, alpha isoform c NP 001029242.1 32828 1.4E-08 9 38-49 (9), 28-38 (11)Tropomyosin 1, alpha isoform h NP 001029246.1 28679 4.3E-10 14 38-49 (9), 28-38 (12)Tropomyosin 2 NP 001019516.1 32938 1.7E-08 11 38-49 (11), 28-38 (7)Tropomyosin 3, gamma isoform 1 NP 476556.2 28692 2.4E-08 10 28-38 (10)Tropomyosin 4 NP 036810.1 28492 9.0E-11 11 28-38 (11)Troponin T type 2 (cardiac) NP 036808.1 35709 2.2E-07 2 38-49 (1)Troponin T type 3 (skeletal, fast) NP 113720.1 30338 6.8E-07 2 28-38 (3)Tyrosine 3-monooxgenase/tryptophan 5-monooxgenase activation protein, beta polypeptide
NP_062250.1 28037 1.9E-09 3 28-38 (6)
Tyrosine 3-monooxygenase/tryptophan 5-monooxygenase activation protein
NP_113791.1 29103 4.4E-09 9 28-38 (11)
Tyrosine 3-monooxygenase/tryptophan 5-monooxygenase activation protein, eta polypeptide
NP_037184.1 28194 7.0E-05 3 28-38 (3)
Tyrosine 3-monooxygenase/tryptophan 5-monooxygenase activation protein, gamma polypeptide
NP_062249.1 28285 5.2E-11 4 28-38 (11)
Voltage-dependent anion channel 1 NP 112643.1 30737 8.7E-09 3 28-38 (10)Voltage-dependent anion channel 2 NP 112644.1 31726 1.0E-07 5 28-38 (10)
Chapter 3: Results
170
Table 17 continued
Protein NCBI accession code Mw (Da) p value
Number of peptides used for
identification
MW regions detected in (No. peptides in each region)
Detected in control
myotubes
Detected in treated
myotubes
Also detected in myoblasts
Mw range 38-49kDa
Actin, alpha 1, skeletal muscle NP_062085.1 42024 4.9E-13 18 98-198 (11), >198 (5), 75-98 (11), 38-49 (12), 49-62 (4), 17-28 (8), 28-38 (11), 6-14 (12), 14-17
Aldolase A-like 1 NP 001013965.1 39467 9.8E-09 6 38-49 (12)Annexin A1 NP 037036.1 38805 1.0E-13 18 38-49 (12), 28-38 (11)Annexin A2 NP 063970.1 38654 1.5E-10 17 38-49 (12), 28-38 (11), >198 (2)Dynactin 2 NP 001004239.1 44121 4.9E-04 2 38-49 (1)Eukaryotic translation initiation factor 4A2 NP 001008336.1 46373 4.8E-08 3 38-49 (3)Farnesyl diphosphate synthase (farnesyl pyrophosphate synthetase, dimethylallyltranstransferase, gera
NP_114028.1 40778 9.3E-10 2 38-49 (1)
Fructose-bisphosphate aldolase A NP 036627.1 39327 1.0E-30 5 38-49 (8)Glutamate oxaloacetate transaminase 2 NP 037309.1 47284 2.5E-07 4 38-49 (4)Heterogeneous nuclear ribonucleoprotein F NP 001032362.1 45701 3.2E-09 3 38-49 (7)Integrin-linked kinase-associated serine/threonine phosphatase 2C
NP_072128.1 42718 6.7E-04 2 49-62 (1)
Isocitrate dehydrogenase 3 (NAD+) alpha NP 446090.1 39588 8.3E-07 2 38-49 (1), 28-38 (3)Keratin 13 NP 001004021.1 47699 1.4E-05 2 98-198 (1), 14-17 (1)Keratin 15 NP_001004022.1 48840 1.9E-06 3 17-28 (1), 3-6 (2), 6-14 (1), >198
(1), 14-17 (1)Lysosomal-associated membrane protein 1 NP 036989.1 43941 4.9E-04 2 75-98 (1)Phosphoglycerate kinase 2 NP 001012130.1 44981 2.2E-10 3 38-49 (2)Protein disulfide isomerase-associated 6 NP 001004442.1 48730 3.5E-12 6 38-49 (12)Pyruvate dehydrogenase (lipoamide) beta NP 001007621.1 38957 1.3E-10 2 28-38 (1)Septin 2 NP 476489.1 41566 1.2E-04 2 38-49 (1)Serine (or cysteine) peptidase inhibitor, clade H, member 1
NP_058869.1 46488 8.5E-14 8 38-49 (12), 49-62 (2), 6-14 (2)
Serum deprivation response NP 001007713.1 46358 1.7E-11 2 49-62 (2)Ubiquinol cytochrome c reductase core protein 2 NP 001006971.1 48366 1.5E-04 2 38-49 (1)
Mw range 49-62kDa
Aldehyde dehydrogenase 3A1 NP_114178.1 50307 2.9E-11 18 38-49 (12), 49-62 (11), 17-28 (1), 75-98 (1)
Aldehyde dehydrogenase 9A1 NP 071609.2 54015 2.0E-06 5 49-62 (7)Ankyrin repeat, SAM and basic leucine zipper domain containing 1
NP_570106.1 53107 4.6E-05 2 14-17 (1)
Annexin A11 NP 001011918.1 54126 4.7E-08 2 49-62 (3)ATP synthase, H+ transporting, mitochondrial F1 complex, alpha subunit 1, cardiac muscle
NP_075581.1 59717 4.4E-12 14 49-62 (12), 28-38 (10)
Chapter 3: Results
171
Table 17 continued
Protein NCBI accession code Mw (Da) p value
Number of peptides used for
identification
MW regions detected in (No. peptides in each region)
Detected in control
myotubes
Detected in treated
myotubes
Also detected in myoblasts
Mw range 49-62kDa continued
Caldesmon 1 NP 037278.1 60548 8.9E-12 18 62-75 (12)Chaperonin containing TCP1, subunit 2 (beta) NP 001005905.1 57422 8.9E-10 8 49-62 (9)Chaperonin containing TCP1, subunit 3 (gamma) NP 954522.1 60608 3.8E-06 6 49-62 (10)Chaperonin containing TCP1, subunit 5 (epsilon) NP 001004078.1 59499 8.2E-07 6 49-62 (7)Chaperonin containing Tcp1, subunit 6A (zeta 1) NP 001028856.1 57981 2.4E-12 7 49-62 (6)Chaperonin subunit 4 (delta) NP 877966.1 58063 2.7E-06 6 49-62 (3)Dihydropyrimidinase-like 3 NP 037066.1 61928 3.6E-07 6 62-75 (11)Dipeptidylpeptidase 7 NP 114179.1 55079 2.3E-06 3 49-62 (4)Dynein, cytoplasmic 1 light intermediate chain 2 NP 112288.1 54711 4.5E-05 2 49-62 (1)EH-domain containing 4 NP 647540.1 61429 1.6E-04 2 49-62 (1)Enabled homolog NP 001012150.1 58792 7.8E-05 3 62-75 (4)Eukaryotic translation elongation factor 1 alpha 2 NP 036792.2 50422 2.9E-07 7 38-49 (12), 49-62 (1), 98-198 (1)Glutamate dehydrogenase 1 NP 036702.1 61377 2.5E-10 9 49-62 (12)Heterogeneous nuclear ribonucleoprotein K NP 476482.1 50944 1.4E-11 7 49-62 (11)Katanin p60 subunit A-like 1 NP 001006957.1 55166 2.2E-04 2 38-49 (1)Keratin 10 NP_001008804.1 56471 2.0E-07 7 98-198 (3), >198 (5), 75-98 (2),
49-62 (1), 17-28 (2), 28-38 (2), 6-14 (3), 14-17 (2), 3-6 (12)
Keratin 14 NP 001008751.1 52651 1.3E-08 3 6-14 (1), >198 (1), 3-6 (1)Keratin 4 NP_001008806.1 57631 2.9E-05 4 98-198 (1), 75-98 (2), 17-28 (2),
28-38 (2), 6-14 (6), 14-17 (1), 3-6 Keratin 42 NP_001008816.1 50182 8.6E-06 4 6-14 (2), 3-6 (2), 98-198 (2),
>198 (6), 75-98 (1)Keratin 5 NP 899162.1 61889 3.6E-05 3 3-6 (2)Keratin 77 NP_001008807.1 57220 3.7E-09 6 98-198 (11), >198 (10), 62-75
(9), 75-98 (7), 38-49 (11), 49-62 (10), 17-28 (9), 28-38 (10), 6-14
(12) 14-17 (12) 3-6 (12)Mitochondrial aldehyde dehydrogenase 2 NP 115792.1 56453 3.9E-11 7 49-62 (7)Mitochondrial ATP synthase beta subunit NP 599191.1 56319 1.0E-10 10 38-49 (6), 49-62 (11)Myosin binding protein H NP 114001.1 52624 7.1E-05 2 49-62 (2)PC4 and SFRS1 interacting protein 1 NP 786941.1 59603 3.8E-04 2 38-49 (1)Plasma glutamate carboxypeptidase NP 113828.1 51937 6.3E-09 3 49-62 (9)Procollagen-proline, 2-oxoglutarate 4-dioxygenase (proline 4-hydroxylase), alpha polypeptide I
NP_742059.2 60860 4.1E-04 3 49-62 (2)
Prolyl 4-hydroxylase, beta polypeptide NP 037130.1 56829 4.4E-15 22 49-62 (12), 6-14 (1)Prosaposin NP 037145.1 61084 6.1E-07 3 6-14 (11), 3-6 (1)Protein disulfide-isomerase A3 NP 059015.1 56554 1.1E-09 17 49-62 (12)
Chapter 3: Results
172
Table 17 continued
Protein NCBI accession code Mw (Da) p value
Number of peptides used for
identification
MW regions detected in (No. peptides in each region)
Detected in control
myotubes
Detected in treated
myotubes
Also detected in myoblasts
Mw range 49-62kDa continued
Pyruvate kinase, muscle NP 445749.1 57781 1.0E-30 19 49-62 (12)T-complex 1 NP 036802.1 60322 4.8E-12 4 49-62 (2)Trimethyllysine hydroxylase, epsilon NP 596878.2 50884 5.9E-04 2 49-62 (1)Tubulin, beta 5 NP 775125.1 49639 5.6E-10 3 38-49 (11), 49-62 (4)Tubulin, beta 6 NP 001020846.1 50027 2.6E-04 3 49-62 (2)Vimentin NP_112402.1 53700 1.9E-13 38 >198 (3), 75-98 (5), 38-49 (12),
49-62 (12), 6-14 (11), 3-6 (12)
Mw range 62-70kDa
Alpha isoform of regulatory subunit A, protein phosphatase 2
NP_476481.1 65281 1.5E-04 2 49-62 (1)
Calnexin NP 742005.1 67213 5.9E-12 4 62-75 (10), 75-98 (4)Eukaryotic translation initiation factor 4B NP 001008325.1 69023 2.0E-09 6 75-98 (8)FK506 binding protein 9 NP 001007647.1 63087 4.7E-05 2 62-75 (1)Glucose phosphate isomerase NP 997475.1 62787 1.1E-06 2 49-62 (2)Inner membrane protein, mitochondrial NP 001030100.1 67135 3.2E-09 3 75-98 (4)Kelch repeat and BTB (POZ) domain containing 10
NP_476539.1 68170 1.0E-06 4 62-75 (3)
Keratin 2 NP_001008899.1 69085 4.4E-09 3 28-38 (1), 6-14 (1), >198 (2), 62-75 (1), 3-6 (1)
Lamin B1 NP 446357.1 66566 2.3E-07 9 62-75 (9)Matrix metalloproteinase 14 (membrane-inserted) NP 112318.1 66038 4.3E-05 2 38-49 (1)Moesin NP 110490.1 67697 2.1E-11 5 62-75 (9)PDZ and LIM domain 5 NP 445778.1 63161 1.3E-04 2 62-75 (4)Radixin NP 001005889.2 68501 7.8E-08 10 28-38 (1), 62-75 (11), 75-98 (1)REC8 homolog NP 001011916.1 67832 8.9E-04 2 75-98 (1)Spermatogenesis associated 18 NP 955406.2 63844 9.0E-04 2 6-14 (1)WD repeat domain 1 NP 001014157.1 66140 1.2E-14 4 49-62 (2)X-prolyl aminopeptidase (aminopeptidase P) 1, soluble
NP_571988.1 69613 6.0E-06 3 62-75 (6)
Chapter 3: Results
173
Table 17 continued
Protein NCBI accession code Mw (Da) p value
Number of peptides used for
identification
MW regions detected in (No. peptides in each region)
Detected in control
myotubes
Detected in treated
myotubes
Also detected in myoblasts
Mw range 70-98kDa
Aconitase 2, mitochondrial NP 077374.2 85380 3.3E-11 11 75-98 (11)Acyl-Coenzyme A dehydrogenase, very long chain
NP_037023.1 70705 7.1E-04 2 62-75 (1)
ATP-binding cassette, sub-family B, member 7, mitochondrial precursor
NP_997683.1 82506 7.1E-04 2 28-38 (2)
Cadherin 13 NP 620244.1 78037 6.7E-05 5 75-98 (3)Cd44 molecule NP 037056.2 85865 2.8E-06 2 62-75 (2)Coiled-coil domain containing 33 NP 001014113.1 87710 2.2E-04 2 98-198 (1)DEAD (Asp-Glu-Ala-Asp) box polypeptide 1 NP 445866.1 82445 4.7E-04 2 75-98 (1)Eukaryotic translation elongation factor 1 delta NP 001013122.1 72083 8.1E-09 4 28-38 (11)Eukaryotic translation elongation factor 2 NP 058941.1 95223 1.0E-14 18 98-198 (1), 75-98 (11)Gelsolin NP 001004080.1 86014 1.1E-10 11 75-98 (10)Heat shock 105kDa/110kDa protein 1 NP 001011901.1 96357 5.7E-06 3 75-98 (2)Heat shock 70kD protein 1-like NP 997711.1 70505 1.8E-11 8 62-75 (12)Heat shock protein 4 NP 705893.1 93997 3.2E-08 5 75-98 (5)Heat shock protein 5 NP 037215.1 72303 1.1E-16 31 62-75 (12)Heat shock protein 8 NP 077327.1 70827 1.1E-14 17 62-75 (12)Heat shock protein 90, alpha (cytosolic), class A member 1
NP_786937.1 84762 7.2E-14 19 75-98 (11)
KH-type splicing regulatory protein (FUSE binding protein 2)
NP_598286.1 74181 1.2E-07 3 75-98 (1)
Lamin A isoform C2 NP_001002016.1 71856 1.3E-13 33 49-62 (10), 98-198 (1), 62-75 (12), 75-98 (8)
Major vault protein NP 073206.2 95739 3.8E-04 2 75-98 (1)Neural cell adhesion molecule 1 NP 113709.1 94599 1.3E-07 12 98-198 (11), 75-98 (1)Nexilin (F actin binding protein) isoform b NP 631977.1 71899 1.7E-05 4 75-98 (2)Nucleolar protein 10 NP 001014098.1 80188 6.6E-04 3 49-62 (1), 3-6 (1)Plasticity related gene 1 NP 001001508.1 83309 3.8E-04 2 >198 (1)Poly(A) binding protein, cytoplasmic 1 NP 599180.1 70656 5.5E-09 9 62-75 (11)Procollagen-lysine 1, 2-oxoglutarate 5-dioxygenase 1
NP_446279.1 83560 1.6E-07 2 75-98 (1)
Protein disulfide isomerase-associated 4 NP 446301.1 72761 1.1E-07 5 62-75 (8), 75-98 (9)Protein phosphatase 1, regulatory subunit 10 NP 075240.1 92771 4.9E-04 2 75-98 (1)Signal transducer and activator of transcription 3 NP 036879.1 87983 9.2E-05 2 14-17 (2)Squamous cell carcinoma antigen recognized by T cells
NP_113784.1 90984 9.3E-04 2 62-75 (1)
TNF receptor-associated protein 1 NP 001034090.1 80411 4.4E-09 2 75-98 (5)Transducin (beta)-like 3 NP 001008278.1 88316 3.0E-04 3 62-75 (1)Transketolase NP 072114.1 71141 2.1E-11 10 62-75 (10)Valosin-containing protein NP 446316.1 89293 2.8E-11 21 75-98 (11)Zinc finger protein 111 NP 579857.1 85367 8.1E-05 2 38-49 (1)
Chapter 3: Results
174
Table 17 continued
Protein NCBI accession code Mw (Da) p value
Number of peptides used for
identification
MW regions detected in (No. peptides in each region)
Detected in control
myotubes
Detected in treated
myotubes
Also detected in myoblasts
Mw range 98-198kDa
Actinin, alpha 1 NP 112267.1 102896 3.5E-12 30 98-198 (7), 75-98 (11)Adaptor-related protein complex 1, beta 1 subunit NP 058973.1 104522 7.8E-04 2 75-98 (1)Alpha actinin 4 NP 113863.2 104849 2.0E-11 24 98-198 (3), 75-98 (11)Clathrin, heavy chain (Hc) NP 062172.1 191476 1.9E-09 20 98-198 (11)Coatomer protein complex, subunit beta 2 (beta prime)
NP_068533.1 102486 2.2E-11 3 >198 (2)
Collagen, type I, alpha 2 NP 445808.1 129486 2.9E-09 16 98-198 (7)Collagen, type III, alpha 1 NP 114474.1 138851 1.2E-11 11 98-198 (7)Cullin-associated and neddylation-dissociated 1 NP 446456.1 136275 1.2E-06 4 98-198 (2)Dynactin 1 NP 077044.1 141842 4.0E-04 2 98-198 (1)Golgi apparatus protein 1 NP 058907.1 133469 1.7E-04 2 98-198 (1)Integrin alpha 7 NP 110469.1 124117 1.5E-04 2 75-98 (1)Kinesin family member 5B NP 476550.1 109463 1.3E-09 3 75-98 (3)
Mw range 98-198kDa continued
Lethal giant larvae homolog 1 NP 690057.1 112424 3.0E-06 2 75-98 (1)Leucine-rich PPR-motif containing NP 001008519.1 156553 7.4E-07 4 98-198 (3)LIM domain 7 NP 001001515.1 195450 1.0E-09 4 98-198 (4)Methylenetetrahydrofolate dehydrogenase 1 NP 071953.1 100932 9.3E-04 2 75-98 (1)Microtubule-associated protein 4 NP 001019449.1 110233 2.1E-06 9 98-198 (9)Na+/K+ -ATPase alpha 1 subunit NP 036636.1 112982 2.1E-06 3 98-198 (2)Proteasome (prosome, macropain) 26S subunit, non-ATPase, 2
NP_001026809.1 100124 3.2E-08 3 75-98 (2)
SEC31 homolog A NP 148981.1 135266 1.8E-08 5 98-198 (3)Staphylococcal nuclease domain containing 1 NP 073185.2 101889 2.6E-06 6 75-98 (6)Thrombospondin 1 NP 001013080.1 129588 5.0E-05 3 98-198 (3)Thyroid hormone receptor associated protein 3 NP 001009693.1 108188 1.0E-04 2 3-6 (1)Ubiquitin-activating enzyme E1 NP 001014102.1 117713 2.6E-10 12 98-198 (2), 75-98 (11)
Chapter 3: Results
175
Table 17 continued
Protein NCBI accession code Mw (Da) p value
Number of peptides used for
identification
MW regions detected in (No. peptides in each region)
Detected in control
myotubes
Detected in treated
myotubes
Also detected in myoblasts
Mw range >198kDa
Acetyl-coenzyme A carboxylase alpha NP 071529.1 265023 3.0E-04 2 62-75 (1)Alpha-spectrin 2 NP 741984.2 284418 9.3E-12 17 49-62 (1), >198 (5)Citron (rho-interacting, serine/threonine kinase 21)
NP_001025082.1 235161 4.2E-04 4 38-49 (1)
Fatty acid synthase NP 059028.1 272477 5.6E-09 9 >198 (5)Insulin-like growth factor 2 receptor NP 036888.1 273220 3.7E-05 2 >198 (1)Kalirin, RhoGEF kinase NP 114451.1 336374 8.2E-04 2 3-6 (1)Myeloid/lymphoid or mixed-lineage leukemia (trithorax homolog, Drosophila); translocated to, 4
NP_037349.1 207548 7.6E-04 2 28-38 (1)
Myosin, heavy chain 10, non-muscle NP 113708.1 228823 2.6E-05 3 >198 (2)Myosin, heavy chain 3, skeletal muscle, embryonic
NP_036736.1 223718 1.0E-30 62 98-198 (12), >198 (5)
Myosin, heavy chain 9, non-muscle NP 037326.1 226196 2.2E-13 33 98-198 (10), >198 (5)Nestin NP 037119.1 198625 1.2E-11 25 >198 (5)Phosphatidylinositol 4-kinase a NP 071637.1 231169 6.6E-04 2 >198 (1)Phosphodiesterase 4D interacting protein NP 071777.1 261879 7.3E-04 2 75-98 (1)Plectin 1 NP 071796.1 533214 3.4E-12 78 17-28 (4), 98-198 (8), >198 (5)Sodium channel, voltage-gated, type III, alpha NP 037251.1 221241 2.4E-05 2 38-49 (1)Spectrin repeat containing, nuclear envelope 1 NP 001025080.1 918957 3.1E-04 6 38-49 (2), 98-198 (1)Spectrin, beta, non-erythrocytic 1 NP 001013148.1 273415 9.1E-11 6 >198 (1)Sperm flagellar 2 NP 072142.1 200840 3.8E-04 2 38-49 (1)ZUBR1 NP 001034115.1 573472 6.7E-04 2 62-75 (1)
Chapter 3: Results
176
3.5.2.1 Effect of TMPD-treatment on biological processes in myoblasts and myotubes
Protein expression was investigated on a global level (using all the proteins
identified using the srf files in Bioworks Browser). The nature of proteins
expressed in control myoblasts was examined in order to appreciate the
ongoing biological processes in untreated myoblasts. A list of proteins unique to
control myoblasts was interrogated using the Panther classification tool
available at www.pantherdb.org. Proteins present in treated myoblasts and
those in control or treated myotubes were not considered at this stage of the
analysis.
From the analysis of control myoblasts, it was revealed that myoblasts were
mainly involved in the biological processes of maintaining their shape and
structure and also their motility (as evidenced by 45 out of 180 proteins
examined in this analysis being involved in these functions). Myoblasts were
also heavily involved in the essential process of protein metabolism and
modification, which could involve protein synthesis, modification, or
degradation, most likely as part of the growth and cell turnover processes that
are typical of most proliferating cells (Figure 57).
In myotubes, the main biological processes that the cells were involved in were
the same as described for myoblasts. In this case, however, protein metabolism
and modification were the predominant processes as shown by 57 of 228
proteins considered being involved. In the case of myotubes, 42/228 proteins
were involved in maintaining cell structure and motility (Figure 58).
Chapter 3: Results
177
4542
27
21 20 2017
1513 13
10 10 97 6
4 4 4 3 3 3 2 2
0
5
10
15
20
25
30
35
40
45
50
Cel
l stru
ctur
e an
d m
otili
ty
Pro
tein
met
abol
ism
and
mod
ifica
tion
Nuc
leos
ide,
nuc
leot
ide
and
nucl
eic
acid
met
abol
ism
Sig
nal t
rans
duct
ion
Dev
elop
men
tal p
roce
sses
Imm
unity
and
def
ense
Intra
cellu
lar p
rote
in tr
affic
Cel
l cyc
le
Tran
spor
t
Bio
logi
cal p
roce
ss u
ncla
ssifi
ed
Mus
cle
cont
ract
ion
Car
bohy
drat
e m
etab
olis
m
Pro
tein
targ
etin
g an
d lo
caliz
atio
n
Apo
ptos
is
Cel
l adh
esio
n
Neu
rona
l act
iviti
es
Oth
er m
etab
olis
m
Cel
l pro
lifer
atio
n an
d di
ffere
ntia
tion
Lipi
d, fa
tty a
cid
and
ster
oid
met
abol
ism
Ele
ctro
n tra
nspo
rt
Am
ino
acid
met
abol
ism
Sul
fur m
etab
olis
m
Hom
eost
asis
Biological process
No.
of p
rote
ins
invo
lved
Figure 57: Normal biological processes in rat L6 myoblasts Myoblasts were grown (section 2.1) and the proteome of the cells was investigated using LC-MS/MS (section 2.7) in order to understand the normal
biological processes ongoing within the cells. Out of 180 proteins investigated, 45 proteins were involved in maintaining the shape and structure of the
cell, whilst 42 were involved in protein metabolism and modification. A total of 23 biological processes were identified that are natural to this cell type,
although it was not possible to classify the function of all of the proteins examined (biological process unclassified). Biological processes for which >1
protein could be associated with involved were included.
Chapter 3: Results
178
57
42
28 2725 25
22 22
1715
1311
8 8 7 6 6 6 5 4 3 3 2 2
0
10
20
30
40
50
60
Pro
tein
met
abol
ism
and
mod
ifica
tion
Cel
l stru
ctur
e an
d m
otili
ty
Nuc
leos
ide,
nuc
leot
ide
and
nucl
eic
acid
met
abol
ism
Imm
unity
and
def
ense
Sig
nal t
rans
duct
ion
Dev
elop
men
tal p
roce
sses
Intra
cellu
lar p
rote
in tr
affic
Bio
logi
cal p
roce
ss u
ncla
ssifi
ed
Mus
cle
cont
ract
ion
Cel
l cyc
le
Car
bohy
drat
e m
etab
olis
m
Tran
spor
t
Cel
l pro
lifer
atio
n an
d di
ffere
ntia
tion
Pro
tein
targ
etin
g an
d lo
caliz
atio
n
Lipi
d, fa
tty a
cid
and
ster
oid
met
abol
ism
Oth
er m
etab
olis
m
Cel
l adh
esio
n
Ele
ctro
n tra
nspo
rt
Apop
tosi
s
Neu
rona
l act
iviti
es
Coe
nzym
e an
d pr
osth
etic
gro
up m
etab
olis
m
Am
ino
acid
met
abol
ism
Onc
ogen
esis
Hom
eost
asis
Biological process
No.
of p
rote
ins
invo
lved
Figure 58: Normal biological processes in rat L6 myotubes Myotubes were differentiated from myoblasts (section 2.1) and their constituent proteins were measured using LC-MS/MS (section 2.7) in order to
understand the normal biological processes characteristic of this cell type. Out of 228 proteins examined, 57 proteins were involved in protein
metabolism and modification whilst 42 were involved in maintaining the shape and structure of the cell. A total of 24 biological processes were identified
that are natural to this cell type although the function of some of the proteins was not clear (biological process unclassified). Biological processes with
which <2 proteins could be associated with were not included in the analysis.
Chapter 3: Results
179
The biological processes ongoing in myoblasts and myotubes were compared
directly in order to investigate the differences in the crucial biological processes
in both cell types. The number of proteins involved in each of the biological
processes already described were compared so that it could be determined
whether cell differentiation resulted in (or from) changes in some of the
processes already described.
In the classification of the biological processes affected when myoblasts were
matured into myotubes, biological processes such as lipid and fatty acid
metabolism, cell proliferation and differentiation, electron transport and muscle
contraction were increased in myotubes when compared to myoblasts (Figure
59). Lipid and fatty acid metabolism is a common route for energy production in
many biological systems302-304. In addition, skeletal muscle is an organ rich in
mitochondria and this is a major site for fatty acid metabolism via the β-
oxidation of fatty acids and also via the electron transport chain (which was
another upregulated biological process in myotubes)305,306.
Cell proliferation and differentiation were classified together and it is likely that
the increase in this biological process in myotubes was mainly due to an
increase in the expression of proteins involved in the differentiation of myoblasts
into myotubes. The differentiation of myoblasts into myotubes was further
evidenced by the increase in the number of proteins involved in muscle
contraction, which is a function more specific to mature skeletal muscle cells.
This ties in with the presence of myofilaments in myotubes, as observed by EM
(section 3.1, Figure 20). Apoptosis was downregulated in myotubes and this
may be a consequence of a reduced amount of cell proliferation, which is more
typical of undifferentiated myoblasts, and therefore a reduced turnover of cells
in the corresponding cultures307.
Chapter 3: Results
180
-0.5
0.0
0.5
1.0
1.5
Lipi
d, fa
tty a
cid
and
ster
oid
met
abol
ism
Cel
l pro
lifer
atio
n an
d di
ffere
ntia
tion
Ele
ctro
n tra
nspo
rt
Mus
cle
cont
ract
ion
Bio
logi
cal p
roce
ss u
ncla
ssifi
ed
Oth
er m
etab
olis
m
Pro
tein
met
abol
ism
and
mod
ifica
tion
Imm
unity
and
def
ense
Car
bohy
drat
e m
etab
olis
m
Intra
cellu
lar p
rote
in tr
affic
Dev
elop
men
tal p
roce
sses
Sig
nal t
rans
duct
ion
Nuc
leos
ide,
nuc
leot
ide
and
nucl
eic
acid
met
abol
ism
Am
ino
acid
met
abol
ism
Cel
l adh
esio
n
Cel
l cyc
le
Hom
eost
asis
Neu
rona
l act
iviti
es
Cel
l stru
ctur
e an
d m
otili
ty
Pro
tein
targ
etin
g an
d lo
caliz
atio
n
Tran
spor
t
Apo
ptos
is
Biological process
Rel
ativ
e le
vel
Figure 59: Comparison of the normal biological process in myoblasts and myotubes Myoblasts and myotubes were cultured (section 2.1) and their protein expression was evaluated by LC-MS/MS (section 2.7) and the biological
processes ongoing in myoblasts and myotubes were compared. The number of proteins involved in each of the biological processes (in myotubes) is
shown relative to the number involved in myoblasts and processes such as lipid and fatty-acid metabolism, electron transport, cell differentiation and
muscle contraction were upregulated. These processes could be considered to be typical of mature, mitochondria-rich skeletal muscle requiring energy
for muscle contraction.
Chapter 3: Results
181
The effect of TMPD administration on the overall protein expression in
myoblasts was also investigated by examining proteins that were uniquely
expressed in either control myoblasts or in TMPD-treated myoblasts. For the
most part, proteins that were reduced or absent in myoblasts as a result of
TMPD-treatment were replaced by other proteins that were involved in the
same biological processes (e.g. cell adhesion, cell structure and motility or
immunity and defence).
Changes in the expression of cell cycle proteins and those involved in apoptosis
or lipid, fatty acid and steroid metabolism may have been a consequence of a
proliferative response to drug treatment whereby programmed cell death was
decreased and more cells were encouraged to enter the cell cycle in order to
replace damaged or unhealthy cells. The increase in energy production (as
evidenced by increased oxidation of fatty acids) may reflect an increased
energy requirement in myoblasts treated low concentrations of TMPD (Table
18).
Chapter 3: Results
182
Table 18: Biological processes affected in TMPD-treated myoblasts Protein expression in cultured myoblasts (section 2.1) was investigated using LC-
MS/MS (section 2.7) in order to investigate TMPD-induced changes in the number of
proteins involved in particular biological processes. The biological processes appeared
to be unaffected but changes in apoptosis, lipid metabolism and cell cycle proteins
resulted from TMPD treatment.
Biological process No. of proteins Biological process No. of
proteins
Apoptosis 3 ReducedBiological process unclassified 7 Biological process unclassified 2 Reduced
Blood circulation and gas exchange 2 IncreasedCarbohydrate metabolism 2 ReducedCell adhesion 2 Cell adhesion 2 UnchangedCell cycle 3 Cell cycle 7 Increased
Cell proliferation and differentiation 2 IncreasedCell structure and motility 6 Cell structure and motility 6 UnchangedDevelopmental processes 2 Developmental processes 2 UnchangedImmunity and defense 5 Immunity and defense 4 ReducedIntracellular protein traffic 5 Intracellular protein traffic 3 Reduced
Lipid, fatty acid and steroid metabolism 3 IncreasedNeuronal activities 2 ReducedNucleoside, nucleotide and nucleic acid metabolism 11
Nucleoside, nucleotide and nucleic acid metabolism 3 ReducedOther metabolism 3 Increased
Protein metabolism and modification 8 Protein metabolism and modification 9 IncreasedProtein targeting and localization 2 Protein targeting and localization 2 UnchangedSignal transduction 6 Signal transduction 7 IncreasedTransport 4 Transport 7 Increased
Proteins present in control myoblast only Proteins present in treated myoblast onlyResponse to
treatment
The effect of TMPD administration on the overall protein expression in
myotubes was also investigated. As with myoblasts, some biological processes
were unaffected and homeostasis was maintained by proteins that were
reduced as a result of treatment being replaced by other proteins (e.g.
intracellular protein traffic). Increases in cell cycle proteins and those involved
with cell structure and motility may again reflect increased cell proliferation to
replace or repair damaged cells. With respect to this, the biological pathways
behind the lack of a change in cell proliferation proteins and the decrease in
those involved in protein metabolism may reflect the fact that the cell was able
to cope adequately with the change in biological functions (Table 19).
Chapter 3: Results
183
Table 19: Biological processes affected in TMPD-treated myotubes Protein expression in cultured myotubes (section 2.1) was investigated by LC-MS/MS
(section 2.7) in order to investigate changes in the number of proteins involved in
specific biological processes. Changes in cell cycle and cell structure proteins may be
a consequence of increased cell proliferation in response to TMPD treatment.
Biological process No, of proteins Biological process No, of
proteins
Biological process unclassified 9 Biological process unclassified 4 ReducedCarbohydrate metabolism 2 Reduced
Cell adhesion 2 IncreasedCell cycle 2 Cell cycle 8 IncreasedCell proliferation and differentiation 2 Cell proliferation and differentiation 2 UnchangedCell structure and motility 3 Cell structure and motility 9 IncreasedDevelopmental processes 4 Developmental processes 3 ReducedElectron transport 2 ReducedImmunity and defense 4 ReducedIntracellular protein traffic 6 Intracellular protein traffic 7 IncreasedLipid, fatty acid and steroid metabolism 3 Lipid, fatty acid and steroid metabolism 5 IncreasedMuscle contraction 3 ReducedNeuronal activities 2 Neuronal activities 2 UnchangedNucleoside, nucleotide and nucleic acid metabolism 6 Nucleoside, nucleotide and nucleic acid
metabolism 4 Reduced
Other metabolism 2 IncreasedProtein metabolism and modification 13 Protein metabolism and modification 7 Reduced
Protein targeting and localization 2 IncreasedSignal transduction 2 Signal transduction 9 IncreasedTransport 2 Transport 6 Increased
Response to treatment
Proteins present in control myotube only Proteins present in treated myotube only
This approach to analysing this data was useful in representing an overall
impression of the process in which both myoblasts and myotubes were involved
with. Both myoblasts and myotubes are derived from the same cell and
therefore closely related. This is reflected in the cells appearing to be involved
in similar normal biological processes. In addition, the myotubes cultures
contained a proportion of myoblasts (as described in section 3.1) and this may
also have contributed to some of the similarities in the biological processes
described for each cell type.
This semi-quantitative approach to the analysis of all the LC-MS/MS data
generated allowed a general appreciation of the overall changes in myoblasts
and myotubes but a more detailed analysis looking into the levels of expression
of particular proteins was required. The use of Decyder™ differential analysis
Chapter 3: Results
184
software was used to specifically detect, match and analyse protein levels in
control and treated skeletal muscle cells and is described in the next section.
3.5.3 Biomarker discovery in TMPD-treated skeletal muscle cells using LC-MS/MS
Decyder™ MS software was used in combination with Bioworks Browser to
compare the relative levels of peptides from different samples (based on
retention time and m/z). The use of the aforementioned software allowed the
comparison of the relative intensities of peptides between control and treated
samples.
3.5.3.1 Optimisation of Decyder™ MS peptide detection settings
Following LC-MS/MS analysis of skeletal muscle cells, the PepDetect module of
Decyder™ MS software was used to detect a consistent number of peptides
across the spectra generated (as described in section 2.7.4). Chemical or
electrical noise can interfere with the interpretation of MS data, and so different
detection settings were investigated so that a maximal number of ions could be
included without including excessive background noise.
Using data generated following the LC-MS/MS analysis of peptides generated
from the >198kDa Mw region of a control myoblasts sample in a 1D gel,
PepDetect settings were adjusted so that ions were detected using a range of
S/N settings between 0.1 and 2 during the detection. This range was selected
as using a S/N setting of 0.1 may tend to include too many ions (including
spectral noise) due to the filtering criteria being relatively relaxed. Alternatively,
a setting of 2 may exclude some useful ions due to its high stringency for the
acceptance of ions.
A suitable S/N (to exclude) setting of 1.1 was selected for the PepDetect
module of the Decyder™ MS software as this appeared to be the setting at
which a maximal number of peptides could be detected before the setting was
changed to 1.2 and the stringency of the setting became exclusive of some ions
Chapter 3: Results
185
that may prove useful. The setting at 1.1 reduces the risk of including peptides
that may be unreliable (Figure 60).
0
20
40
60
80
100
120
140
160
180
0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0
Remove peptides below S/N
No.
pep
tides
det
ecte
d
Figure 60: Optimisation of detection settings in the PepDetect module of Decyder™ MS software Myoblasts were cultured (section 2.1) and proteins in the sample were separated by
1DE (section 2.7.1). Peptides were then generated from sections of the 1D gel and
detected by LC-MS/MS (sections 2.7.2 and 2.7.3). In order to determine suitable
settings for peptide detection in the PepDetect module of Decyder™ MS software,
settings were adjusted so that the S/N of ions excluded ranged from 0.1 to 2. A S/N
setting of 1.1 was used in subsequent experiments, as a setting of 2 was too stringent,
leading to the possible exclusion of some potentially useful ions whilst a setting of 0.1
may lead to the inclusion of ions that were affected by chemical or electrical noise. S/N
= signal to noise ratio.
3.5.3.2 Comparison of enzyme activity and protein levels
In order to verify the TMPD-induced changes observed biochemically, changes
observed using the assays described in section 3.3 were compared with
changes in their protein levels as measured by LC-MS/MS. However, this
comparison was only possible for CK, ALD and parvalbumin in myoblasts
(Table 20) and only for CK, ALT, AST, ALD, sTnI and Myl3 in myotubes (Table
21). In doing this, only the peptides used in the quantification of ALD (in
myoblasts) and for ALD and AST (in myotubes) were based on reliable
identifications of the protein (based on 2 or more peptides with a p value of
Chapter 3: Results
186
<0.05) and these proteins are also listed in Table 16. The remaining
comparisons were either based on proteins identified using only 1 peptide or
peptides in which the confidence in the identification was unacceptable
(p>0.05). As such, these comparisons are only a guide and may serve as
further confirmation of the changes observed using biochemical assays.
In comparing changes in the biochemical levels of markers in myoblasts with
the protein levels measured by LC-MS/MS, only 1 peptide was significantly
reduced in its levels. However, a 2 fold decrease in the levels of a peptide used
to quantify CK (K.SQEEYPDLSK.H) was not backed up by changes in the other
peptide used in its measurement (-.MPFGNTHNKFK.L), despite the fact that
biochemical measure of CK did demonstrate that this marker was decreased
following treatment with 25μM TMPD. As a result of the conflicting changes in
peptides from the same protein, this measurement cannot be relied on. Also,
the peptides used were not identified with a high level of confidence (p>0.05)
making it possible that even the peptides used in the quantitation may have
been incorrectly identified (Table 20).
Chapter 3: Results
187
Table 20: Comparison of biochemical levels with protein levels in myoblasts Changes in the levels of biochemical markers in myoblasts treated with TMPD (Figure
30) were compared with changes observed in the corresponding proteins using data
generated by LC-MS/MS. Only the comparisons in the levels of ALD are based on
reliable identifications (based on >1 peptide with a p value of <0.05). Only a change in
the level of one of the peptides used to measure CK was observed (-
.MPFGNTHNKFK.L) and this was not backed up by the changes seen in the other
peptide used to quantify CK (K.SQEEYPDLSK.H). No significant reduction in the levels
of CK were observed biochemically in cells treated with 25μM TMPD (Student’s t-test,
p<0.05).
Description NCBI MW Peptide used for quantitation
No. peptides used in
quantitation (control/treated)
Fold change compared to
control
Muscle creatine kinase NP_036662.1 42992 -.MPFGNTHNKFK.L 4/4 1.4K.SQEEYPDLSK.H 3/3 0.5
L-lactate dehydrogenase A NP_058721.1 36427 Insufficient peptides for quantitation - -
Glutamic-pyruvate transaminase (alanine aminotransferase)
NP_112301.1 55074 Insufficient peptides for quantitation - -
Glutamic-oxaloacetic transaminase 1, soluble (aspartate aminotransferase)
NP_036703.1 46299 Insufficient peptides for quantitation - -
Fructose-bisphosphate aldolase A NP_036627.1 39327 K.ELADIAHR.I 5/5 1.0K.ADDGRPFPQVIK.S 6/4 0.8
Troponin I type 1 (skeletal, slow) NP_058880.1 21711 Insufficient peptides for quantitation -
Parvalbumin NP_071944.1 11918 K.GFSSDARDLSAK.E 6/6 0.1Myosin, light chain 3, alkali; ventricular, skeletal, slow
NP_036738.1 22142 Insufficient peptides for quantitation - -
Fatty acid binding protein 3, muscle and heart
NP_077076.1 14766 Insufficient peptides for quantitation - -
No significant changes were observed by LC-MS/MS in the levels of the
candidate markers in myotubes (Student’s t-test, p<0.05). This reflects what
was observed when the intracellular levels of each marker was measured more
specifically using the biochemical assays. The comparisons made here
between the biochemical levels and the measured proteins levels perhaps
suggest that there was increased sensitivity with the biochemical assays when
measuring these candidate biomarkers (Table 21).
Chapter 3: Results
188
Table 21: Comparison of biochemical levels with protein levels in myotubes Changes in the levels of biochemical markers in myotubes treated with TMPD were
compared with changes observed in the corresponding proteins using data generated
by LC-MS/MS. Only the comparisons in the levels of AST and ALD are based on
reliable identifications (based on >1 peptide with a p value of <0.05). None of the
changes observed were significant, and this reflects what was observed biochemically
despite the fact that both CK and ALD were observed to be reduced in cells treated
with 25μM TMPD when measured more specifically (Student’s t-test, p<0.05).
Description NCBI MW Peptide used for quantitation
No. peptides used in
quantitation (control/treated)
Fold change compared to
control
Muscle creatine kinase NP_036662.1 42992 K.GGDDLDPNYVLSSR.V 5/4 0.9L-lactate dehydrogenase A NP_058721.1 36427 Insufficient peptides for quantitation - -
Glutamic-pyruvate transaminase (alanine aminotransferase)
NP_112301.1 55074 K.LMSVRLCPPVPGQALMDMVVSPPTPSEPSFK.Q
5/3 0.8
Glutamic-oxaloacetic transaminase 1, soluble (aspartate aminotransferase)
NP_036703.1 46299 R.IAATILTSPDLR.K 3/3 0.8
Fructose-bisphosphate aldolase A NP_036627.1 39327 K.ADDGRPFPQVIK.S 6/6 1.3K.ELADIAHR.I 3/4 1.3
Troponin I type 1 (skeletal, slow) NP_058880.1 21711 R.KNVEAMSGMEGR.K 2/2 1.6Parvalbumin NP_071944.1 11918 Insufficient peptides for quantitation - -
Myosin, light chain 3, alkali; ventricular, skeletal, slow
NP_036738.1 22142 K.ITYGQCGDVLR.A 6/6 1.3
Fatty acid binding protein 3, muscle and heart
NP_077076.1 14766 Insufficient peptides for quantitation - -
Comparing the activity of the candidate biomarkers measured in section 3.3
with the protein levels measured by LC-MS/MS was not useful as it would
appear as if the majority of the parameters measured biochemically (and more
specifically) were either more suitable for detection or measurement using the
specific assays described earlier. It is also possible that the effect of treatment
was on the activity of the enzymes measured and not their proteins levels.
Here, it was not possible to verify the changes observed biochemically using
LC-MS/MS data.
Chapter 3: Results
189
3.5.3.3 Selection of putative biomarkers of TMPD-induced myopathy in myoblasts
At least 2 peptides with a p value <0.05 were required to ensure confidence in
the conclusive identification of proteins (as described in section 3.5.2). Whilst it
was acceptable for the peptides used in the identification to be detected in only
one sample, this would not provide a sufficient number of samples from which a
reliable quantitation of the protein could be made. As such, in a first pass filter
of Decyder™ MS software output, the identification of proteins was considered
separately from the measurement of peptides.
The levels of proteins were only compared if the constituent peptides were
observed on at least 50% (6/12) of the total number of observable occasions.
This meant that a sufficient number of peptides could be measured to facilitate
a reliable assessment of TMPD-induced changes and significant changes in the
levels of the peptides could then be identified (Student’s t-test, p<0.05).
Consideration was also given to consistency in the response of all the peptides
chosen for measurement to TMPD-treatment across different samples, in terms
of the magnitude of change and also in terms of whether the peptides were
increased or decreased.
The levels of individual peptides were also examined in order to ensure the
accuracy of peptide selection by the PepDetect module of the Decyder™ MS
software. This was performed so that peptides were not selected erroneously
and also so that peptide matches were not overlooked due to slight shifts in the
retention time or in the measured m/z of peptides. In such instances, peptides
that should have been matched were manually selected, deselected or their
charge assignment corrected manually in order to accurately reflect the true
nature of the measured peptides.
Using this relatively strict approach to putative biomarker selection, 8 proteins (4
up-regulated and 4 down-regulated) ranging in Mw from 15-919kDa were
differentially expressed in myoblasts as a result of treating the cells with TMPD.
The magnitude of the changes ranged from 1.2 to 5.6-fold. The responsive
Chapter 3: Results
190
proteins were diverse in their biological functions and classified as being
involved in biological processes such as cell structure and motility (2),
metabolism (1), protein metabolism and modification (3) and signal transduction
(2) (Table 22).
Chapter 3: Results
191
Table 22: Identified myoblasts cell proteins that showed differential expression following treatment with TMPD Myoblasts were cultured and treated with TMPD for 24h (section 2.1). Proteins within the samples were then separated by 1DE (section
2.7.1) before the resulting gel was divided into sections. Peptides generated from these sections were analysed by LC-MS/MS in order to
determine any TMPD-induced changes in the expression of proteins in myoblasts. Proteins (8) were shown to differ significantly in their
expression as a result of TMPD-treatment (Student’s t-test, p<0.05). Of these proteins, the levels of 4 were increased and 4 were
decreased.
Proteins Peptides used for quantitation
No. ions measured in control/treated
Fold change Probability
Overall fold change ± SEM
(CV)
Cell structure and motility
Keratin 2 (KRT2) NP_001008899.1 Cytoplasm 69a) 98-198 7b) R.GFSSGSAVVSGGSR.R (+2) 4.36 5/5 +1.6 0.009 +1.6
Keratin 19 (KRT19) NP_955792.1 Cytoplasm 45a) 14-17 3b) R.LAADDFR.T (+1) 2.22 6/4 +5.5 0.04 +5.6 ±0.1
R.LASYLDKVR.A (+2) 3.32 6/4 +5.7 0.004 (3%)
Metabolism
Aldehyde dehydrogenase family 3, member A1 (ALDH3A1)
NP_114178.1 Cytoplasm 50 38-49 12b) K.SLLNEEAHK.A (+2) 2.33 6/6 -1.5 0.04 -1.5 ±0.1
R.EKPLALYVFSNNEK.V (+3) 4.42 6/6 -1.4 0.02 (11%)
Subcellular location
Comparison between control and TMPD-treated
Name (gene name) NCBI accession code
Mean Xcorr
Peptide (charge)Mw (kDa)
Mw region (kDa)
identified in
No. of unique peptide ions
used in identification
a) There was a discrepancy between the calculated Mw and the region of the gel in which the protein was detected. It is possible that a post-translational modification to the protein may have
occurred or that the protein may have been fragmented or digested prior to detection.
b) More peptides were used in the identification of these proteins but some of the peptides were below the level of quantitation. Therefore, only peptides suitable for measurement were used
in determining the levels of the protein.
Chapter 3: Results
192
Table 22 continued Proteins Peptides used for quantitation
No. ions measured in control/treated
Fold change Probability
Overall fold change ± SEM
(CV)
Protein metabolism and modification
Serine [or cysteine] proteinase inhibitor, clade H (SERPINH1)
NP_058869.1 Extracellular Space 47 38-49 10b) K.HLAGLGLTEAIDK.N (+2) 3.04 6/5 -1.7 0.0009 -1.8 ±0.2
K.KPVEAAAPGTAEK.L (+3) 4.12 5/5 -2.0 0.03 (13%)
Eukaryotic translation initiation factor 4B (EIF4B)
NP_001008325.1 Cytoplasm 69a) 6-14 6b) K.DETKVDGVSTTK.G (+2) 2.21 6/6 -1.4 0.03 -1.4
Spectrin repeat containing, nuclear envelope 1 (SYNE1)
NP_001025080.1 Nucleus 919a) 75-98 2b) K.LSEFVVTK.I (+2) 2.25 6/6 +1.2 0.02 +1.2
Signal transduction
Rho GDP dissociation inhibitor [GDI] alpha (ARHGDIA)
NP_001007006.1 Cytoplasm 23 28-38 5b) R.VAVSADPNVPNVIVTR.L (+2) 4.75 6/6 +2.0 0.03 +2.1 ±0.1
K.SIQEIQELDKDDESLRK.Y (+2) 4.36 4/5 +2.2 0.04 (7%)
Lectin, galactose binding, soluble 1 (LGALS1)
NP_063969.1 Extracellular Space 15 6-14 6b) K.VRGELAPDAK.S (+2) 2.69 6/6 -2.0 0.004 -1.7 ±0.2
K.SFVLNLGK.D (+2) 2.7 5/6 -1.4 0.01 (17%)
K.DSNNLCLHFNPR.F (+3) 3.61 5/5 -1.7 0.03
Mean Xcorr
Comparison between control and TMPD-treatedMw region
(kDa) identified in
No. of unique peptide ions
used in identification
Peptide (charge)Name (gene name) NCBI accession code Subcellular location Mw
(kDa)
a) There was a discrepancy between the calculated Mw and the region of the gel in which the protein was detected. It is possible that a post-translational modification to the protein may have
occurred or that the protein may have been fragmented or digested prior to detection.
b) More peptides were used in the identification of these proteins but some of the peptides were below the level of quantitation. Therefore, only peptides suitable for measurement were used
in determining the levels of the protein.
Chapter 3: Results
193
After finding a reliable profile of proteins whose levels were changed following
the treatment of myoblasts with TMPD (Table 22), data generated by LC-
MS/MS was examined further in order to reveal a single protein that might be
suitable for further investigation of the protein levels using an alternative method
to LFQP. In doing this, the certainty in the identification (as evidenced by the
number of peptides used in the identification and the mean Xcorr value) and the
magnitude of change was taken into consideration.
Rho GDP dissociation inhibitor (GDI) alpha (NP_001007006.1) was selected as
the calculated Mw (23kDa) correlated well with the Mw region in which the
protein was measured (28-38kDa). Two peptides were used to quantify the
protein, and the observation that levels of the peptides in individual samples
correlated well with each other, increased the certainty in the measurement of
the protein levels. In addition, an overall 2.1-fold increase in the levels of the
protein was observed in response to TMPD treatment (Figure 61). An example
of the mass spectra obtained is shown in Figure 62.
Control Treated0
25000
50000
75000
100000
125000
R.VAVSADPNVPNVIVTR.L
Inte
nsity
6
5
41 32
2
5
3 41
6
Control Treated0
25000
50000
75000
100000
125000
R.VAVSADPNVPNVIVTR.L
Inte
nsity
Control Treated0
25000
50000
75000
100000
125000
R.VAVSADPNVPNVIVTR.L
Inte
nsity
6
5
41 32
2
5
3 41
6
Control Treated0
25000
50000
75000
100000
125000
K.SIQEIQELDKDDESLRK.Y
Inte
nsity
6
5
12
6
2
54
3
Control Treated0
25000
50000
75000
100000
125000
K.SIQEIQELDKDDESLRK.Y
Inte
nsity
Control Treated0
25000
50000
75000
100000
125000
K.SIQEIQELDKDDESLRK.Y
Inte
nsity
6
5
12
6
2
54
3
n=6
n=6
n=4
n=5
Figure 61: Relative levels of peptides used to quantify Rho GDP dissociation factor Inhibitor (GDI) alpha Myoblasts were grown and treated with 25μM TMPD for 24h (section 2.1). Proteins in
cell homogenates were separated by 1DE and peptides were generated from sections
of the 1D gel (section 2.7). The levels of Rho GDI were increased significantly (2.1-fold)
in response to treatment (Student’s t-test, p<0.05). The individual sample identities are
displayed on the graph to show the correlation between the different peptide levels in
each individual sample.
Chapter 3: Results
194
300 400 500 600 700 800 900 1000 1100 1200 1300 1400 1500 1600 1700 1800m/z
0
5
10
15
20
25
30
35
40
45
50
55
60
65
70
75
80
85
90
95
100
Relative Abu
ndan
ce
yy7+1
798.4
yy10+1
1108.5
bb 9+1
853.3
yy11+1
1223.4
bb 6+1
543.2
yy13+2
691.5bb 13
+1
1276.3bb 12+1
1163.3yy4
+1
488.4yy3+1
375.3bb 14
+1
1375.3yy12
+1
1294.4yy8+1
897.5yy13
+1
1381.6yy10
+2
555.0yy2+1
276.1yy9
+1
1011.4bb 15
+1
1476.4yy11
+2
612.5
bb 4+1
357.2
bb 10+1
950.3yy14
+1
1480.6
a)
b)
VAVSADPNVPNVIVTR
Pos AA b-ions y-ions Rev pos
1 V 100.08 - 162 A 171.11 1551.85 153 V 270.18 1480.81 144 S 357.21 1381.74 135 A 428.25 1294.71 126 D 543.28 1223.67 117 P 640.33 1108.65 108 N 754.37 1011.59 99 V 853.44 897.55 810 P 950.49 798.48 711 N 1064.54 701.43 612 V 1163.61 587.39 513 I 1276.69 488.32 414 V 1375.76 375.24 315 T 1476.81 276.17 216 R - 175.12 1
m/z
Rel
ative
abun
danc
e
Figure 62: Example spectrum obtained following the MS/MS analysis of a peptide from Rho GDI Myoblasts were cultured and treated with 25μM TMPD (section 2.1). a) In combination, the 2 peptides used in the quantitation of Rho GDI covered 20%
of the total protein and b) The example MS/MS spectrum shown is from 1 of 2 peptides used to quantify the protein. Identified b-ions and y-ions in the
spectrum and the corresponding table are coloured blue and red, respectively, and this spectrum was typical of the spectra used to sequence this
peptide. Pos, position; AA, Amino acid; Rev pos, reverse sequence position.
Chapter 3: Results
195
TMPD-induced increases in the level of Rho GDI observed by LC-MS/MS were
confirmed by immunoblotting in order to verify the changes seen. Homogenates
from myoblasts treated with a range of concentrations of TMPD (0-25μM) were
probed specifically for Rho GDI myoblasts using a polyclonal antibody raised in
rabbits. Rho GDI was significantly increased 1.4 and 1.6 times in myoblasts
treated with 12.5 and 25μM TMPD, respectively (Figure 63).
60708090
100110120130140150160
-5 0 5 10 15 20 25
Control 6.25μMTMPD
12.5μMTMPD
25μMTMPD
[TMPD] (μM)
% c
hang
e ov
erco
ntro
l mea
n
**
Figure 63: TMPD-induced changes in the levels of Rho GDI in myoblasts Myoblasts were cultured and treated with a range of concentrations of TMPD (section
2.1). TMPD-induced increases in the levels of Rho GDI originally observed by LC-
MS/MS were confirmed by immunoblotting using a polyclonal antibody measuring Rho
GDI (section 2.7.5). Increases of 1.4 and 1.6-fold in this protein were observed
following treatment with 12.5μM and 25μM TMPD, respectively. Error bars show ±
SEM and significant changes are denoted by *p<0.05, **p<0.01.
Chapter 3: Results
196
The effect of other potential myotoxins on the levels of Rho GDI in myoblasts
was also investigated. Treating myoblasts with 6.25, 12.5, 25 and 50μM GWδ
had no significant effect on the levels of Rho GDI treated although it did
increase (Students t-test; p<0.05). Myoblasts treated with pravastatin, which is
another potentially myotoxic lipid-lowering compound were increased 2.6 and
1.9-fold when cells were treated with 50 and 100μM of the compound (Figure
64).
40
60
80
100
120
140
160
180
-5 0 5 10 15 20 25 30 35 40 45 50
Control 12.5μMGWδ
25μMGWδ
50μMGWδ
Control25μM
Pravastatin50μM
Pravastatin100μM
Pravastatin
[GWδ] (μM)
% c
hang
e ov
erco
ntro
l mea
n
% c
hang
e ov
erco
ntro
l mea
n
40
80
120
160
200
240
280
320
360*
**
0 25 50 75 100[Pravastatin] (μM)
40
60
80
100
120
140
160
180
-5 0 5 10 15 20 25 30 35 40 45 50
Control 12.5μMGWδ
25μMGWδ
50μMGWδ
Control25μM
Pravastatin50μM
Pravastatin100μM
Pravastatin
[GWδ] (μM)
% c
hang
e ov
erco
ntro
l mea
n
% c
hang
e ov
erco
ntro
l mea
n
40
80
120
160
200
240
280
320
360*
**
0 25 50 75 1000 25 50 75 100[Pravastatin] (μM)
Figure 64: Rho GDI levels in myoblasts treated with GWδ or pravastatin
Myoblasts were cultured and treated with a range of concentrations of GWδ or
pravastatin (section 2.1). The levels of Rho GDI were investigated following the
treatment of myoblasts with these test compounds in order to assess whether
increases in the levels of Rho GDI were replicated using other potentially myotoxic
compounds (section 2.7.5). Treating myoblasts with GWδ had no significant effect on
the levels of Rho GDI but a 2.6 and 1.9-fold increase was observed when cells were
treated with 50 and 100μM, respectively, of pravastatin (Student’s t-test, p<0.05). Error
bars show ± SEM and significant changes are denoted by *p<0.05, **p<0.01.
Chapter 3: Results
197
3.5.3.4 Selection of putative biomarkers of TMPD-induced myopathy in myotubes
Using the same approach described for myoblasts in section 3.5.3.3, differential
expression analysis of peptides generated from control and TMPD-treated rat
L6 myotubes revealed that the levels of 10 proteins were differentially
expressed in response to drug-treatment (Student’s t-test, p<0.05). Out of these
proteins, 9 were measured at decreased levels in treated cells when compared
to controls and 1 protein was up-regulated. The magnitude of the changes
ranged from 1.3 to 2.6-fold. The proteins were classified according to their
biological function and they included proteins that affected biological processes
such as the cell cycle (3), electron transport (3), muscle contraction (1) and
protein metabolism and modification (3) (Table 23).
Chapter 3: Results
198
Table 23: Identified myotubes cell proteins that showed differential expression following treatment with TMPD Myotubes were differentiated from myoblasts and treated with TMPD (section 2.1). Proteins within the samples were then separated by
1DE before analysing peptides generated from different sections of the resulting gel by LC-MS/MS (section 2.7). Proteins (10) were
shown to differ significantly in their expression as a result of TMPD-treatment (Student’s t-test, p<0.05). Of these proteins, the level of 1
proteins was increased and 9 were decreased.
Proteins Peptides used for quantitation
No. ions measured in control/treated
Fold change Probability
Overall fold change ± SEM
(CV)
Cell cycle
Caldesmon 1 (CALD1) NP_037278.1 Cytoplasm 60 62-75 18b) R.LEQYTNAIEGTK.A (+2) 4.24 6/6 -1.3 0.04 -1.3
Proteasome 26S subunit, non-ATPase, 2 (PSMD2)
NP_001026809.1 Cytoplasm 100a) 14-17 3b) R.ELDIMEPK.V (+2) 2.14 5/4 -1.7 0.006 -1.7
WD repeat domain 1 (WDR1) NP_001014157.1 Extracellular Space 66 49-62 4b) K.YAPSGFYIASGDISGK.L (+2) 4.96 6/6 +2.6 0.001 +2.6
Electron transport
ATP synthase, H+ transporting, mitochondrial F1 c (ATP5D)
NP_620806.1 Cytoplasm 18 14-17 2 K.AQSELSGAADEAAR.A (+2) 5.07 6/6 -1.7 0.004 -1.5 ±0.1
R.IEANEALVK.A (+2) 2.65 5/6 -1.4 0.04 (11%)
Cytochrome c oxidase subunit IV isoform 1 (COX4I1)
NP_058898.1 Cytoplasm 19 6-14 2b) K.VNPIQGFSAK.W (+2) 2.78 6/6 -1.3 0.03 -1.3
Nucleoside diphosphate kinase B (NME2) NP_114021.2 17 14-17 4b) R.GDFCIQVGR.N (+2) 3.33 6/6 -2.0 0.003 -1.9 ±0.1
R.NIIHGSDSVESAEK.E (+2) 3.85 6/6 -1.7 0.04 (10%)
K.DRPFFPGLVK.Y (+2) 2.7 6/5 -2.0 0.01
Mean Xcorr
Comparison between control and TMPD-treatedNo. of unique peptide ions
used in identification
Mw region (kDa)
identified inPeptide (charge)Name (gene name) NCBI accession
codeMw
(kDa)Subcellular location
a) There was a discrepancy between the calculated Mw and the region of the gel in which the protein was detected. It is possible that a post-translational modification to the protein may have
occurred or that the protein may have been fragmented or digested prior to detection.
b) More peptides were used in the identification of these proteins but some of the peptides were below the level of quantitation. Therefore, only peptides suitable for measurement were used
in determining the levels of the protein.
Chapter 3: Results
199
Table 23 continued.
Proteins Peptides used for quantitation
No. ions measured in control/treated
Fold change Probability
Overall fold change ± SEM
(CV)
Muscle contraction
Myosin, light polypeptide 2 (MYL2) NP_036737.1 Cytoplasm 19 14-17 12b) K.LKGADPEDVITGAFK.V (+3) 4.63 6/6 -2.5 0.02 -2.0 ±0.1
K.GADPEDVITGAFK.V (+2) 4.55 6/6 -2.0 0.03 (15%)
K.NICYVITHGDAKDQE.- (+3) 4.94 6/6 -2.0 0.006
R.FSQEEIK.N (+2) 2.54 6/5 -2.0 0.01
K.EAFTVIDQNR.D (+2) 4.37 6/6 -1.7 0.08
Protein metabolism and modification
Ribosomal protein S14 (RPS14) NP_073163.1 Cytoplasm 16 14-17 4b) K.TPGPGAQSALR.A (+2) 3.11 6/6 -1.7 0.005 -1.7
R.IEDVTPIPSDSTR.R (+2) 4.19 6/5 -1.7 0.004
Ribosomal protein S19 (RPS19) NP_001032423.1 Cytoplasm 16 14-17 6b) K.DVNQQEFVR.A (+2) 3.49 5/4 -1.7 0.009 -1.8 ±0.2
R.VLQALEGLK.M (+1) 2.69 3/5 -2.0 0.001 (13%)
Ubiquitin-conjugating enzyme E2N (UBE2N)
NP_446380.1 Cytoplasm 17 14-17 3b) R.LLAEPVPGIK.A (+2) 2.46 6/6 -1.4 0.04 -1.4
Name (gene name) NCBI accession code Subcellular location Mw
(kDa) Peptide (charge)Mw region
(kDa) identified in
No. of unique peptide ions
used in identification
Mean Xcorr
Comparison between control and TMPD-treated
a) There was a discrepancy between the calculated Mw and the region of the gel in which the protein was detected. It is possible that a post-translational modification to the protein may have
occurred or that the protein may have been fragmented or digested prior to detection.
b) More peptides were used in the identification of these proteins but some of the peptides were below the level of quantitation. Therefore, only peptides suitable for measurement were used
in determining the levels of the protein.
Chapter 3: Results
200
In myotubes, the levels of myosin light chain polypeptide 2 (NP_036737.1) were
decreased 2-fold in response to treatment with 25μM TMPD. There was also a
reasonable degree of correlation between the calculated Mw (19kDa) and the
gel slice in which the protein was measured (14-17kDa). In quantifying its
levels, 5 peptides were used which increased the certainty that its levels were
measured accurately. The correlation between the levels of the peptides used in
the measurement of the protein in different samples was also good (Figure 65).
In order to confirm the changes in the levels of myosin light chain 2 (Myl2)
observed by LC-MS/MS, homogenates from myotubes treated with a range of
concentrations of TMPD (0-25μM) were probed specifically for Myl2 using a
rabbit polyclonal antibody to the protein. Myl2 was significantly decreased 2.4-
fold in cells treated with as low as 6.25μM TMPD (Student’s t-test, p<0.05). The
levels of Myl2 were also reduced from control levels (2.3 and 2.4-fold) when
myotubes were treated with 12.5μM and 25μM TMPD, respectively (Figure 67).
An example of the mass spectra obtained is shown in Figure 66.
Chapter 3: Results
201
Control Treated0
100000
200000
300000
400000
500000
K.LKGADPEDVITGAFK.V
Inte
nsity
6
5
34 21 6 5 21
34
Control Treated0
100000
200000
300000
400000
500000
K.LKGADPEDVITGAFK.V
Inte
nsity
Control Treated0
100000
200000
300000
400000
500000
K.LKGADPEDVITGAFK.V
Inte
nsity
6
5
34 21 6 5 21
34
Control Treated0
100000
200000
300000
K.GADPEDVITGAFK.V
Inte
nsity
6
5
3 421
5
6
4
13 2
Control Treated0
100000
200000
300000
K.GADPEDVITGAFK.V
Inte
nsity
6
5
3 421
5
6
4
13 2
Control Treated0
100000
200000
300000
K.NICYVITHGDAKDQE.-
Inte
nsity 5
3
6
1
24
53
6
1
2 4
Control Treated0
100000
200000
300000
K.NICYVITHGDAKDQE.-
Inte
nsity
Control Treated0
100000
200000
300000
K.NICYVITHGDAKDQE.-
Inte
nsity 5
3
6
1
24
53
6
1
2 4
Control Treated0
25000
50000
75000
100000
125000
R.FSQEEIK.N
Inte
nsity
6
2
45
31
24
5
31
Control Treated0
25000
50000
75000
100000
125000
R.FSQEEIK.N
Inte
nsity
Control Treated0
25000
50000
75000
100000
125000
R.FSQEEIK.N
Inte
nsity
6
2
45
31
24
5
31
Control Treated0
50000
100000
150000
200000
250000
K.EAFTVIDQNR.D
Inte
nsity
6
5
41 23
5
1 36 2
4
Control Treated0
50000
100000
150000
200000
250000
K.EAFTVIDQNR.D
Inte
nsity
Control Treated0
50000
100000
150000
200000
250000
K.EAFTVIDQNR.D
Inte
nsity
6
5
41 23
5
1 36 2
4
n=6
n=6
n=6
n=6
n=6
n=6
n=6
n=5
n=6
n=6
Figure 65: Relative levels of peptides used to quantify myosin light chain polypeptide 2 in myotubes Myotubes were differentiated from myoblasts and treated with 25μM TMPD for 24h
(section 2.1). Proteins in myotubes were separated by 1DE and peptides were
generated from fractions of the gel (section 2.7). The levels of Myl2 were decreased
significantly (2-fold) in response to treatment (Student’s t-test, p<0.05). 5 different
peptides contributed to the quantitation of the protein and the identity of each sample is
displayed on the graph to show the correlation between the 5 different peptide levels in
each individual sample.
Chapter 3: Results
202
400 600 800 1000 1200 1400 1600 1800m/z
0
5
10
15
20
25
30
35
40
45
50
55
60
65
70
75
80
85
90
95
100
Relative Abu
ndan
ce
yy10+1
1076.4
bb 5+1
485.2
yy13+1
1319.3
yy7+1
735.3 bb 8+1
826.3yy11
+1
1191.3yy14+2
724.5 bb 10+1
1038.3yy5
+1
523.2
yy6+1
636.3
bb 11+1
1139.4yy8
+1
850.3
bb 9+1
925.2
yy4+1
422.2bb 14
+2
707.9bb 14
+1
1414.5bb 2
+1
242.2
bb 4+1
370.2 yy14+1
1447.9
a)
b)
LKGADPEDVITGAFK
Pos AA b-ions y-ions Rev pos
1 L 114.09 - 152 K 242.19 1447.74 143 G 299.21 1319.65 134 A 370.24 1262.63 125 D 485.27 1191.59 116 P 582.32 1076.56 107 E 711.37 979.51 98 D 826.39 850.47 89 V 925.46 735.44 710 I 1038.55 636.37 611 T 1139.59 523.29 512 G 1196.62 422.24 413 A 1267.65 365.22 314 F 1414.72 294.18 215 K - 147.11 1
m/z
Rel
ativ
e ab
unda
nce
Figure 66: Example spectrum obtained following the MS/MS analysis of a peptide from Myl2
Myotubes were cultured and treated with 25μM TMPD (section 2.1). a) In combination, the 5 peptides used in the quantitation of Myl2 covered 34% of
the total protein. a) The example MS/MS spectrum shown is from 1 of 5 peptides used to quantify Myl2. Identified b-ions and y-ions in the spectrum and
the corresponding table are coloured blue and red, respectively, and this spectrum was typical of the spectra used to sequence the peptide.
Chapter 3: Results
203
102030405060708090
100110120
-5 0 5 10 15 20 25
* ** **
Control 6.25μMTMPD
12.5μMTMPD
25μMTMPD
[TMPD] (μM)
% c
hang
e ov
erco
ntro
l mea
n
Figure 67: TMPD-induced changes in the levels of myosin light chain 2 in myotubes Myotubes were differentiated from myoblasts and treated with a range of
concentrations of TMPD (section 2.1). Reductions in the levels of Myl2 originally
observed by LC-MS/MS were confirmed by immunoblotting using a rabbit polyclonal
antibody to Myl2 (section 2.7.5). The sensitivity of the myotubes to TMPD was
demonstrated by a 2.4-fold reduction in the levels of Myl2 when cells were treated with
a low concentration of the compound (6.25μM). Significant reductions of 2.3 and 2.4-
fold in the levels of this protein compared to control samples were also observed
following treatment with 12.5μM and 25μM TMPD, respectively (Student’s t-test;
p<0.05). Error bars show ± SEM and significant changes are denoted by *p<0.05,
**p<0.01.
Chapter 3: Results
204
The levels of Myl2 in myotubes treated with a range of concentrations of other
potential myotoxins (GWδ and pravastatin) were also investigated. There were
no statistically significant changes in the levels of this protein in cells treated
with 12.5, 25 and 50μM GWδ (Students t-test; p<0.05). This was also true of
myotubes treated with 25, 50 or 100μM pravastatin (Figure 68).
50
70
90
110
130
150
170
190
210
-5 0 5 10 15 20 25 30 35 40 45 50
Control 12.5μMGWδ
25μMGWδ
50μMGWδ
Control 25μMPravastatin
50μMPravastatin
100μMPravastatin
[GWδ] (μM)
% c
hang
e ov
erco
ntro
l mea
n
% c
hang
e ov
erco
ntro
l mea
n20
40
60
80
100
120
140
160
[Pravastatin] (μM)
0 25 50 75 10050
70
90
110
130
150
170
190
210
-5 0 5 10 15 20 25 30 35 40 45 50
Control 12.5μMGWδ
25μMGWδ
50μMGWδ
Control 25μMPravastatin
50μMPravastatin
100μMPravastatin
[GWδ] (μM)
% c
hang
e ov
erco
ntro
l mea
n
% c
hang
e ov
erco
ntro
l mea
n20
40
60
80
100
120
140
160
[Pravastatin] (μM)
0 25 50 75 1000 25 50 75 100
Figure 68: Myosin light chain 2 levels in myotubes following treatment
with GWδ or Pravastatin
Myotubes were grown and treated with a range of concentrations of GWδ or
pravastatin (section 2.1). The levels of Myl2 were investigated following the treatment
of myotubes with these test compounds in order to assess whether TMPD-induced
decreases in the levels of Myl2 were replicated using other potentially myotoxic
compounds (section 2.7.5). Myotubes were insensitive to the effects of both test
compounds as demonstrated by the absences of statistically significant changes in the
levels of Myl2 following treatment with GWδ or pravastatin. Error bars show ± SEM
where visible.
Chapter 3: Results
205
The use of Decyder™ MS software allowed the semi-automated detection,
identification, matching and measurement of peptides during these
investigations. This, in turn, facilitated a more focused investigation into TMPD-
induced changes in the levels of peptides and proteins from a large dataset
generated by LC-MS/MS that perhaps was not possible otherwise. However,
the output from Decyder™ MS analysis is highly dependent on the settings used
during its application. Due to the wide dynamic range of expression and the
varied nature of the peptides measured, it was difficult to define settings that
were suitable for all the peptides present. As such, it was decided to essentially
over-detect peptides during the detection stage performed in the PepDetect
module of the software and risk introducing peptides that might have been
affected by electrical or chemical noise into the analysis. Some of this noise
was then removed by using stringent criteria for the matching of peptides and
also during the selection of putative biomarkers.
A number of possible biomarkers of skeletal muscle toxicity were revealed and
the treatment-induced changes in the levels of Rho GDI and Myl2 in myoblasts
and myotubes, respectively, were confirmed by immunoblotting. In this way, any
doubt in the authenticity of any changes revealed by LC-MS/MS were removed
as they were confirmed when the proteins were measured more specifically.
Technically, the use of LC-MS/MS is a more resource-demanding approach to
biomarker discovery than using candidate biomarker screening (section 3.3) or
the anonymous screening of protein ions (section 3.4). However, it results in
useful data that can be interrogated to discover compound-induced changes in
the levels of proteins and biomarkers of toxicity. The conclusive identification of
each peptide and protein measured also allows the findings to be readily put
into biological context. As a result of knowing the proteins that are responsive to
treatment, the changes observed can then be measured more specifically using
a more established or less technically demanding technique such as
immunoblotting.
Chapter 4: Discussion
206
Chapter 4
4. Discussion
Evidence of skeletal muscle toxicity can occur in preclinical and clinical studies
and can lead to the withdrawal of compounds, especially those belonging to the
lipid-lowering class of drugs such as PPAR agonists or statins. Despite the
incidence of myopathy being low in these studies, concerns over the safety of
patients have led pharmaceutical companies and other sponsors to take this
recurrent toxicology issue seriously. In addition, negative public perception and
the threat of litigation are issues that regulatory authorities and research
companies strive to address. For these reasons, skeletal muscle is a tissue that
is usually monitored carefully when therapies for dyslipidaemia are being
developed for clinical use63,68,85.
CK is a biochemical marker that is often used to monitor the effects of drug-
treatment on skeletal muscle but, in most cases, it is not sensitive or specific
enough and is only indicative of late-stage damage rather than being predictive
of myopathy. Other biochemical markers of myopathy have, so far, performed
little better and more novel biomarkers have yet to be fully validated for this
purpose308,309. It was hypothesised that proteomics could be used to discover
more useful biomarkers of myopathy using TMPD (an experimental myotoxin)
and a potentially myotoxic PPAR-δ agonist (GWδ). Testing skeletal muscle cells
with non-cytotoxic concentrations of the test compounds in vitro could also
provide a suitable test system for discovering novel biomarkers of myopathy. In
addition, it was proposed that this in vitro system might provide a way of
screening novel compounds for their myopathic potential, prior to the selection
of candidate drugs for development.
In this study, the search for protein biomarkers of myopathy was conducted
using a number of different strategies. A candidate biomarker approach was
Chapter 4: Discussion
207
used in which the value of markers associated with skeletal muscle was
determined. MS based techniques were also used and this included an
anonymous screening approach in which SELDI-TOF-MS was used to assess
treatment related changes in the levels of protein ions in skeletal muscle cells
treated with potentially myotoxic compounds. Biomarker discovery was also
attempted using another MS based technique, which involved the combination
of 1DE with LC-MS/MS.
4.1 The use of rat L6 skeletal muscle cells
The 3Rs (reduction, refinement and replacement) initiative encourages research
institutions to use alternatives to animals, where possible and relevant, to
assess the safety and efficacy of their compounds. In vitro alternatives are often
sought due to ethical considerations surrounding the use of animals in drug
research and the advantages of cost savings in terms of animal costs and
animal husbandry101,310,311. In addition, there is sometimes a need to screen
new molecules at the start of the drug discovery process in order to select
molecules with the least potential of causing adverse effects in the latter stages
of development. Along with reducing the costs of the large amounts of drug
required for animal studies, this can reduce the costs of drug development by
ensuring that compounds with a high liability are removed from the process thus
reducing late-stage drug attrition due to unexpected findings312-316.
Rat L6 skeletal muscle cells were used to determine the effects of test
compounds selected for their ability to induce adverse effects on skeletal
muscle in vivo and in vitro86,88,89,317. The rat L6 cell line closely mimics many
aspects of skeletal muscle (both structurally and biochemically). Proliferating
myoblasts are similar to myogenic satellite cells which, in vivo, are involved in
the growth, repair and maintenance of skeletal muscle tissue318-320. In culture,
once they were attached to surface of the collagen-coated plate, they were
shown to proliferate when grown in high serum conditions. EM examinations
demonstrated that they existed as mononucleated single cells containing many
Chapter 4: Discussion
208
ribosomes and very few mitochondria, suggesting that the cells were
undifferentiated, proliferating cells heavily involved in protein synthesis321.
Myotubes were successfully differentiated from myoblasts after approximately
10 days, as evidenced by the presence of cells with multiple nuclei. These cells
were also elongated having been formed by the fusion of multiple myoblasts. It
was also noted that, although the majority of cells in myotubes cultures were
myogenic (>80%), there were still a proportion of myoblasts present
(approximately 5-10%). This should be considered a desirable feature of the
system making it more similar to skeletal muscle tissue. In vivo, immature
muscle cells reside adjacent to mature cells in a quiescent state and are
activated in the event of injury, functioning to repair or replace damaged skeletal
muscle tissue107,109,322,323.
The presence of myofilaments in myotubes provided further evidence of their
differentiation from myoblasts. Myotubes also showed characteristic striations
formed by light and thick bands (of myosin and actin, respectively), although the
myofilaments were disorganised. During sample collection, cells were washed
free of growth or differentiation medium and centrifuged in order to form a pellet
so that the cells could be fixed in formaldehyde/gluteraldehyde for EM
examinations. This procedure may have disrupted some of the fine structure
present in the cells and subsequently the characteristic light and dark bands
formed by the thin (actin) and thick (myosin) filaments that form the
characteristic striations seen in muscle tissue in vivo (both of which are involved
in the ATP driven procedure of muscle contraction under the influence of Ca2+).
It is also possible that myofilaments were disorganised as the cells were non-
contracting in culture, in the absence of a suitable stimulus for contraction324,325.
An initial assessment of the relative levels of the enzymes in myoblasts and
myotubes showed that ALT, AST, CK, ALD and sTnI were detected at
significantly higher levels in myotubes when compared to myoblasts. These
higher basal levels of the markers in myotubes were due to the differentiated or
Chapter 4: Discussion
209
more “muscle-like” status of the mature cells. Previous researchers have used
CK as a marker of differentiation in myotubes although the use of the other
markers for this purpose has not been widely reported322,326-331. CK is the most
commonly used marker in studies involving skeletal muscle and was the most
responsive biochemical marker to the differentiation of myoblasts to myotubes.
skeletal troponin I (sTnI) is a less well characterised marker for assessing the
status of skeletal muscle and was also increased when myoblasts were
differentiated88,92,93.
Myoblasts were simpler to cultivate in vitro, as myotubes required the additional,
time-consuming step of cell-differentiation. As such, myoblasts may be more
suitable in a high-throughput screen for testing the myotoxic potential of NCEs.
The length of time required for the differentiation of the myotubes might be
considered to be a potential barrier to the use of myotubes in a high-throughput
routine screen, although the growth and differentiation characteristics are very
predictable and experiments could be planned accordingly so that this need not
be a bottleneck in the process. Despite this, both systems were responsive to
treatment with the test compounds and could help in elucidating the mechanism
of toxicity of NCEs.
4.2 The assessment of drug-induced cytotoxicity
TMPD was selected for its ability to cause skeletal muscle toxicity, along with a
highly selective PPAR-δ agonist (GWδ) for which incidents of myopathy have
been reported86-90,332-336. Methods commonly used for the assessment of
cytotoxicity in vitro (such as the visualisation of cells by microscopy, total
intracellular protein and MTT reduction) were used to determine concentrations
of the test compounds at which cellular function was compromised or at which
the membrane integrity of myoblasts or myotubes was adversely affected337,338.
These methods were used also to select non-overtly cytotoxic concentrations of
TMPD or GWδ with which to conduct differential protein expression
investigations.
Chapter 4: Discussion
210
Ultrastructural changes in cells (observed by EM and LM) were indicative of
compound-induced effects. The levels at which these changes were observed
also correlated well with the concentrations at which significant reductions in
total intracellular protein and changes in the ability of the mitochondria of cells
to convert a tetrazolium salt (MTT) to a formazan dye were observed. These
cells displayed swollen mitochondria, disorganised cristae and the presence of
vacuoles or lipid droplets within the cells. However, these alterations in the
morphology and function of the cells were only evident at concentrations at
which adverse effects had already been observed by relatively unsophisticated
methods such as LM. This demonstrated the lack of sensitivity of these
commonly used methods for assessing the effects of test compounds and the
need for more sensitive methods or biomarkers for assessing myopathy.
During EM examinations, it was also noticed that there was a higher proportion
of cells that displayed features typical of myoblasts within myotube cultures
treated with toxic concentrations of TMPD than were observed in untreated
cultures. This observation may correspond to the situation in vivo where
quiescent satellite cells are activated after injury to skeletal muscle and re-enter
the cell cycle. Cells then divide and replicate before differentiating into
myotubes and fusing with other myotubes to replace damaged skeletal muscle
tissue. This indicated the potential to use markers of muscle differentiation as
biomarkers of effect319,339,340.
Taken together, these results demonstrated that both in vitro systems (using
myoblasts and myotubes) were responsive to treatment and therefore suitable
for biomarker discovery experiments. These results also allowed the selection
of appropriate concentrations of the test compounds with which to treat skeletal
muscle cells in differential protein expression experiments. In order to detect
changes that were predictive of drug effects, subsequent experiments were
performed on cells treated with concentrations of the test compounds that were
not overtly cytotoxic.
Chapter 4: Discussion
211
4.3 The candidate biomarker approach to biomarker discovery
CK is found in several tissues but is detected at especially high levels in skeletal
muscle, brain and heart tissue. An increase in the circulatory level of CK is
usually used in toxicology experiments to indicate irreversible damage to the
cell membrane of skeletal muscle cells332,341-343. It is often used in combination
with other enzymes that are present in high concentrations in skeletal muscle
(e.g. ALT, AST, LDH and ALD) that can help to clarify the origins of any
changes in enzyme levels. Changes in the circulatory levels of these markers
have been used extensively to monitor adverse effects on skeletal muscle in
vivo. In spite of their routine use, it is widely recognised that these enzymes
sometimes lack sensitivity to low levels of damage and also lack specificity for
the site of injury unless specific isoenzymes such as CK-MM (specific to
skeletal muscle) are measured95,98,286,344-348. As such, other more novel markers
such as sTnI, Myl3, FABP3 and parvalbumin-α have been proposed and their
usefulness was also assessed in this work.
Enzyme leakage from cells into the circulation is commonly used for the
assessment of damage to organs and therefore the leakage of enzymes from
skeletal muscle cells in vitro was assessed. However, these markers were
generally not of any additional value in these investigations. AST leaked from
the cells of myoblasts and myotubes and the leakage of sTnI from myotubes
were the most useful markers of TMPD-induced cytotoxicity in these
investigations although these changes were observed at concentrations already
known to be cytotoxic (as demonstrated by changes in microscopy, intracellular
protein content and MTT reduction).
Intracellular decreases in the levels of ALT, AST, CK and LDH in myoblasts
treated with 100μM TMPD and an increase in the leaked levels of AST in
myoblasts treated with 50μM TMPD demonstrated that there was no additional
value in measuring candidate biomarkers in myoblasts. It was also noted that
the levels of sTnI, Myl3, FABP3 and parvalbumin-α were, for the most part,
Chapter 4: Discussion
212
below the limit of detection for myoblasts due to the comparatively low levels
detected in undifferentiated cells.
In myotubes, the intracellular levels of ALD and CK were increased in cells
treated with ≥ 6.25μM TMPD. All other changes were, again, only observed at
levels already shown to be cytotoxic to myotubes. As with myoblasts, there
appeared to be little additional value in measuring candidate biomarkers when
compared to the more commonly used measures of cytotoxicity such as
intracellular protein content and MTT reduction. The sensitivity of ALD and CK
to the treatment of myotubes with TMPD was unexpected (and previously
unreported) but may be useful for the in vitro assessment of the myotoxic
potential of NCEs in myotubes.
In interpreting the results from biochemical analyses, it was not possible to
correlate changes in the intracellular levels or markers with leaked levels
because some of the candidate biomarkers were detected in cells at low
concentrations (especially myoblasts). For biochemical analyses, cells were
lysed into 200μl of extraction buffer but 2ml of medium was collected, resulting
in a greater dilution of the leaked levels of the markers. This resulted in a
greater chance of their levels being below or near the limit of detection. It is also
recognised that CK, for instance, is broken down or loses activity in the medium
making it difficult to correlate intracellular and leaked levels of the marker88. In
addition, it should be noted that the assays used for these investigations are
normally aimed at human diagnostics and therefore at much higher
concentrations of the markers than were evident in these experiments.
4.4 SELDI-TOF-MS as a tool for biomarker discovery
SELDI-TOF-MS is based on the same principles as MALDI-TOF-MS and
enriches specific subsets of proteins from samples on the chromatographic
surfaces of the ProteinChip® arrays used. By effectively limiting the proportion
of the proteome viewed in a single analysis, a staggered analysis of the proteins
in a sample can be performed. This is a similar principle to using LC to
Chapter 4: Discussion
213
introduce samples into a mass spectrometer, and usually leads to a reduction in
ion-suppression, which might occur when highly abundant proteins are present
or many proteins co-elute from an HPLC column during LC-MS analyses.
SELDI-TOF-MS can also be technically less demanding than some of the other
MS based techniques and has the additional benefit of generating a large
quantity of data in a high-throughput manner164,231,233,234,349,350.
Protein profiling investigations were aimed at identifying compound-induced
differences in protein expression when control cells and treated cells were
compared. Control cells were compared with those treated with a concentration
of the test compounds at which cytotoxicity was not evident, as monitored by
more established methods of screening cell status. At these concentrations,
putative biomarkers were likely to be predictive of treatment effects and
indicative of the mechanism by which cytotoxicity would eventually occur with
increased exposure of the cells to the test compounds.
Separation between control and treated myoblasts and myotubes following PCA
analysis of SELDI-TOF-MS data demonstrated that both TMPD and GWδ-
treatment caused changes in the levels of protein ions that could be revealed
using this technique. PCA is usually the first stage of such analyses and natural
separation between control and treated groups usually indicates that PLS-DA is
a valid next step in the analysis. PLS-DA is commonly used to force the
separation between groups so that the variables responsible for the separation
can be established. However, the inclusion of a large number of variables can
lead to complicated loadings plots in which it is difficult to interpret which are the
most important variables in the analysis. The use of VIP plots can help in this
interpretation but, in these investigations, too many ions resulted. The use of
false discovery rates was also attempted but the methods used (BH-FDR and
the Bonferroni correction) are widely recognised to be too conservative and this
proved to be the case in these investigations295.
Chapter 4: Discussion
214
It was recognised during these SELDI-TOF-MS investigations that the
reproducibility of the compound-induced changes observed was a critical factor
that determines the usefulness of any biomarkers discovered. Issues with
reproducibility in this technique originate from a number of sources such as the
preparation of ProteinChip® arrays or pulse-to-pulse variation in the laser
source of the mass spectrometer. However, these are issues that are inherent
in MS (and other techniques) and not all of them (apart from the ProteinChip®
arrays) are unique to SELDI-TOF-MS297,351-353. Some of these variations arise
during the application of EAM solution or from residual buffer solutions left on
ProteinChip® arrays during their preparation (chemical noise). Electrical noise
generally emanates from the internal machinery of the mass spectrometer and,
along with chemical noise, can also affect mass spectra. This is managed
during the processing of SELDI-TOF-MS data using the BMW incorporated into
Ciphergen Peaks software. The BMW not only allows the clustering of protein
ions that can be compared to each other in differential protein expression
experiments, but it also ensures that appropriate peaks are selected for analysis
with the minimal inclusion of peaks that might have been adversely affected by
noise. The possibility of selecting variable ions was reduced by using
appropriate settings for the BMW (section 3.4.1)288-290.
In these investigations, issues with reproducibility were also addressed by
repeating experiments. The initial findings from the statistical analysis of two
repeat experiments of a training dataset were tested in an independent, smaller
test dataset. Markers that responded consistently to treatment with the test
compounds were then considered as putative biomarkers. SELDI-TOF-MS
revealed a protein expression profile consisting of 16 protein ions that were able
to discriminate between control and TMPD-treated myoblasts in a training
dataset, although not all of these ions were validated successfully in the test
dataset. In myotubes, a similar profile of 12 protein ions was discovered whose
levels changed in response to treatment with the same test compounds. Some
researchers have proposed that the conclusive identification of proteins may not
be necessary and that protein fingerprints that are capable of distinguishing
Chapter 4: Discussion
215
between control and treated samples may be sufficient162,164,234,354,355. Although
a combination of biomarkers could potentially be useful in the prediction of
disease or toxicity, a specific combination of a number of variables may be
difficult to reproduce in independent repeat experiments. It is therefore more
difficult to produce fully validated assays. There is also a possibility of
statistically over-fitting the data when a complex protein profile is considered in
this way.
It is often preferable to discover the most reproducible changes and to
conclusively identify them, so that the response of the marker observed using
MS can be validated using another technique356,357. Despite this, SELDI-TOF-
MS demonstrated that ions at m/z 5743 and 5298 might have been the most
robust markers in myoblasts and myotubes, respectively, due to them being
highlighted following corrections for multiple testing and also in repeat
experiments. The fact that some of the ions previously shown in the profile were
reproducible in training dataset, but not in the test dataset, highlights the risk of
using a specific profile or “fingerprint”. This demonstrated that whilst some of
the ions previously selected in the training dataset showed reasonable intra-
experimental reproducibility, they were variable when considered inter-
experimentally.
Following the discovery of proteins ions that proved useful in the distinguishing
between control and treated samples, proteins were identified on the basis on
Mw information and their response to drug treatment. Data generated by LC-
MS/MS was used in combination with the TagIdent protein identification tool
available via the ExPASy proteomics database, which searches the UniProt
consortium knowledgebase for possible protein matches based on m/z
information, and a user defined mass accuracy (0.1% of the measured m/z).
The response of proteins to drug treatment, as measured by LC-MS/MS, was
also compared to the response of the SELDI-TOF-MS protein ions in an attempt
to identify proteins conclusively, but was of little additional value in this exercise.
It was not possible to conclusively identify the protein ions of interest although
Chapter 4: Discussion
216
these methods of searching gave an idea of some possible identities for
unknown proteins. Protein ions found using SELDI-TOF-MS ultimately remained
unidentified, making it difficult to consider the effect of the changes biologically.
The difficulty in obtaining conclusive identifications using SELDI-TOF-MS is, in
part, due to the difficulty in isolating proteins that may have been retained on
the chip surface. Furthermore, the relatively low resolution of the Ciphergen
PBS IIc mass spectrometer usually prevents the identification of interesting
proteins based on an accurate Mw assignment. This is combined with the
possibility that protein ions detected at certain molecular weights could actually
be multiple charges of the protein of interest358,359.
Other approaches to the identification of proteins discovered by SELDI-TOF-MS
include prefractionation of the sample to allow isolation of the specific proteins
in order to allow specific proteins of interest to be characterised more easily.
The use of magnetic beads designed to mimic the chemistries available on the
ProteinChip® surface would also facilitate the isolation of interesting proteins
from the beads, perhaps after they have been characterised on SELDI-TOF-MS
ProteinChip® arrays. Antibody arrays designed to specifically probe for
candidate biomarkers could also be potentially useful360-365. The approach of
using ProteinChip® arrays as a target in higher resolution mass
spectrophotometers has also been attempted and may increase the possibility
of a conclusive identification366,367. Despite some limitations, SELDI-TOF-MS is
still a useful tool for investigating protein expression, but the lack of a conclusive
identification of responsive proteins means that it is difficult to fully validate
putative biomarkers and, therefore, using them preclinically or clinically could be
questionable.
Chapter 4: Discussion
217
4.5 The label-free approach to biomarker discovery
LC-MS/MS was combined with 1DE during the LFQP analysis of control and
treated skeletal muscle cells. The prefractionation of samples using 1DE
allowed the complexity of samples to be reduced prior to enzymatically
digesting the proteins and generating peptides, which could then be measured
by LC-MS/MS. This in turn allowed more peptides to be detected (and therefore
more proteins) with the increased possibility of discovering protein biomarkers
of treatment effect. At this sample preparation stage, some researchers have
been able to focus subsequent analyses on specific areas of the gel, if clear
differences in protein expression were observed236. In this study, no changes in
the protein expression profile were visible by 1DE so it was not possible to
focus on any specific areas of the 1D-gel.
In this study, 6 replicate control and 6 replicate TMPD-treated samples
effectively became 132 samples when each lane of the gel was divided into 11
Mw regions. Whilst dividing the gel into multiple sections reduced the
complexity of the samples, it resulted in an increased number of samples and
therefore an increase in the complexity of the data analysis. In addition, for a
reliable comparison of expressed protein levels to be made between control and
drug-treated samples, the accuracy of gel slicing was essential during sample
processing.
LC-MS/MS analysis of the peptides generated from the gel pieces was a time-
consuming process with each sample taking 70min to analyse. In practice, two
traps and columns were run in parallel in order to increase sample throughput
and samples that were to be compared to each other were analysed on the
same column in order to minimise any variation in the retention times of eluted
peptides. Although the data used was truncated to between 15 and 45 min (as
the excluded regions contained a minimal number of peptides), MS analysis
resulted in a large dataset containing 2100 peptides, which contributed to the
identification of 217 proteins in myoblasts and 269 proteins in myotubes.
Chapter 4: Discussion
218
Intracellular levels of candidate biomarkers measured biochemically were
compared to the levels of the corresponding proteins measured by LC-MS/MS.
This analysis proved not to be useful at this level and this was mainly due to the
fact that most of the peptides used in these comparison were not reliably
identified (based on 1 peptide with a p value>0.05). In some instances, the
peptides used may have been suitable for the identification of the protein but
were not expressed at sufficiently high levels to facilitate a reliable quantitation
of their levels. It should be noted, however, that this exercise essentially meant
comparing specifically measured and sensitive enzyme activity with the protein
levels of these markers.
Myoblasts and myotubes were involved in very similar cellular processes when
the global protein expression was considered and this is perhaps not surprising
as 155 of the total number of proteins identified in skeletal muscle cells were
common to both cell types. In a comparison between the normal biological
processes, myotubes were shown to be highly involved in processes such as
lipid, fatty acid and steroid metabolism, cell differentiation and muscle
contraction, which are more likely to occur in skeletal muscle in vivo.
Conversely, the process of apoptosis was comparatively higher in myoblasts
when compared to myotubes and it would appear as if this is a common finding
in vivo. In vivo macrophages have frequently been found in close proximity with
developing muscle tissue and the process of cellular proliferation was apparent
bearing in mind the nature of the cells used in this work307,339,368. The remaining
processes were those that are typical of a highly metabolic tissue with a high
energy requirement, such as skeletal muscle.
A profile of 8 proteins whose levels were responsive to TMPD treatment was
revealed when the peptides levels were examined more specifically using
Decyder™ MS software (4 increased and 4 decreased). In myotubes, 10
proteins were discovered that were similarly responsive to treatment with TMPD
(9 decreased and 1 increased in its levels). In contrast to the anonymous
Chapter 4: Discussion
219
approach using SELDI-TOF-MS, this method proved to be advantageous in that
nearly all of the peptides detected were routinely identified during MS analysis.
This allowed changes in the levels of the proteins to be considered in terms of
the affected cellular processes for both myoblasts and myotubes.
The use of Decyder™ MS software allowed the automation of some aspects of
the complex data analysis procedures involved with such a large dataset. In the
PepDetect module of the software peptides were detected automatically (based
on user settings) although peptide detection was rigorously checked for any
inconsistencies so that occasional errors in feature selection could be corrected
manually. This was critical to the matching of peptides, as errors here would
lead to errors in data analysis. Peptide matching was performed in the
PepMatch portion of the Decyder™ MS software and allowed peptides to be
considered together for proteins and also allowed their response to drug-
treatment to be assessed.
A number of proteins associated with cell proliferation, growth, structure and
intracellular signalling were responsive to TMPD. In particular, many appeared
to be directly involved with the reorganisation of the actin cytoskeleton, which
plays an important role in many essential biological processes within the cell.
Rho GDI is an important protein involved in signal transduction processes within
the cell, including the reorganisation of the actin cytoskeleton369,370. This occurs
via the interaction between Rho GDI and the family of small guanosine
triphosphatases (GTPases) and guanosine diphosphatases (GDPases). GTP
and GDP are key molecular switches of many cellular processes and they are,
in turn, influenced by the binding of Rho GDI which was increased in this
study371-374. Rho GDI has also been associated with the nuclear hormone
receptor estrogen receptor β (ESR2)375 which was also increased, although it
was not included in Table 16 as the confidence in the identification was
insufficient (based on <2 peptides). The levels of several structural proteins
such as Keratin 2 and 19 were also increased in myoblasts as a response to
Chapter 4: Discussion
220
treatment with TMPD and these proteins form part of the intermediate filament
and are therefore involved with the actin cytoskeleton376-378.
Some of the responsive proteins were also involved with myoblast
differentiation. Increases in the levels of Rho GDI have been linked with a
negative effect on cell differentiation in skeletal muscle cells, regulating the
expression of myogenic regulatory factors such as myogenin and myocyte
enhancing factor 2 (MEF2)379. Nuclear envelope spectrin repeat protein 1 (also
known as nesprin 1) is a protein that is believed to maintain the structural
integrity of the nucleus during myoblast differentiation and an increase in its
levels may reflect an effect on cell differentiation380. Lectin galactose binding
soluble 1 is another protein closely linked with cell differentiation and decreases
in its levels support the suggestion that cell differentiation was discouraged
(perhaps in preference to cell proliferation) in response to treatment with
TMPD381-384. Lectin has also been associated with apoptosis385 and, in
conjunction with decreases in the levels of serine [or cysteine] proteinase
inhibitor (which is involved in proteolysis386-388) this may provide additional
evidence for an increase in cell proliferation.
The majority of the proteins that were differentially expressed in treated
myotubes were reduced in their levels, which might suggest an adverse effect
of TMPD treatment. These included proteins involved with metabolism and
transport, intracellular signalling, protein synthesis, and cell structure. The levels
of ubiquitin-conjugating enzyme E2N (UBE2N) were decreased and this protein
is responsible for binding to and targeting other proteins for degradation. It is
also known to act in conjunction with proteasome 26S subunit, non-ATPase, 2
(PSMD2) in defining the fate of cells and this was also decreased389. Together,
these results may suggest that protein synthesis was increased although the
reasons for reductions in the levels of cytoplasmic ribosomal proteins
(ribosomal protein s14 and ribosomal protein s19), which are important proteins
involved in protein biosynthesis, are not yet clear but may well reflect a transient
change in their levels. EM examinations in these investigations (section 3.1)
Chapter 4: Discussion
221
demonstrated that myotubes were actually a mixture of differentiated skeletal
muscle cells (approximately 80%) with a small population of myoblasts
(approximately 10%) and this may be reflected in the sometimes complex
cellular responses of “myotubes” to treatment with TMPD.
Myl2 is an important contractile protein and reductions in its levels may reflect a
TMPD-induced change from the mature form of skeletal muscle cells to the
undifferentiated proliferative form in order to address the potentially harmful
effects of treatment with the test compound390-392. ATP synthase and
cytochrome c oxidase are electron transport proteins, found in the mitochondria
of mature skeletal muscle cells, and these also were reduced which might
support a cell type switch to the proliferative immature form of the cells.
Amongst the changes observed in both myoblasts and myotubes, increases in
the levels of Rho GDI in myoblasts and a reduction in the level of Myl2 were the
most consistent changes and their response to drug-treatment was confirmed
by immunoblotting. An increase in Rho GDI was also observed following the
treatment of myoblasts with similar concentrations of GWδ and pravastatin.
Although, Myl2 was not decreased following the treatment of myotubes with
GWδ, there were decreases in its level observed following treatment with
pravastatin. It is possible that GWδ did not cause structural changes to
myotubes (as might be evidenced by changes in Myl2).
Despite being technically more demanding and having a much lower throughput
than SELDI-TOF-MS, the vast amount of information generated by the powerful
combination of electrophoresis, LC and MS/MS is a definite advantage during
biomarker discovery. As discussed previously, the conclusive identification of
proteins of interest is a key aspect of biomarker validation as it allows biological
understanding to be applied to the findings. Despite the advantage of being able
to generate a large amount of data quickly and relatively easily, the lack of
conclusive identifications of the proteins observed is a major shortfall of the
SELDI-TOF-MS approach. Although attempts were made to identify the
Chapter 4: Discussion
222
responsive proteins seen by SELDI-TOF-MS, the mass accuracy of the
technique does not allow for conclusive identification of the proteins. In addition,
the limited mass range of SELDI-TOF-MS (<20kDa) meant that large proteins
were not observed using this technique. However, the technique was shown in
this study to be capable of discovering protein biomarkers of myopathy in the
form of responsive protein ions. Other researchers have shown that
identification of these putative biomarkers is possible through the use of on-chip
or off-chip fractionation, modification of existing LC-MS/MS technology to
integrate with ProteinChip® arrays or via the use of antibody-based
ProteinChip® arrays174,367.
LFQP provides an alternative method for discovering protein biomarkers that
has advantages over the labelling techniques such as ITRAQ or ICAT, in that
the use of expensive reagents or reagents with a limited availability is avoided.
However, the vast amount of data produced can be difficult and time-consuming
to analyse. It should be noted that the identifications provided by LFPQ should
be considered with caution. Confidence in the identifications routinely provided
LC-MS/MS (in combination with database searching) should be considered
based on the number of peptides and the confidence in the identification (as
indicated by the p value or the mean Xcorr value)197. In this work, although all
proteins were considered, the most reproducible changes were focused on,
based on protein identifications from multiple unique peptides (Table 22 and
Table 23). It should be pointed out, however, that the proteins discovered in
these investigations are likely to require further validation prior to use as
biomarkers of myopathy in preclinical or clinical studies.
Chapter 4: Discussion
223
4.6 Summary
During these investigations, a number of proteins were discovered to be
responsive to the treatment of skeletal muscle cells with TMPD or GWδ. During
these studies, TMPD was used in all experiments and results obtained using
this compound allowed a comparison to be made between all the different
approaches to biomarker discovery used in these investigations (Figure 69).
Methods of assessing compound effects on cells, such as measuring total
intracellular protein, MTT reduction or the visualisation of cells using microscopy
were useful in determining the cytototoxic potential of the test compounds in the
test system. They were therefore useful for defining the concentrations at which
biomarker discovery investigations could be conducted. However, these
methods were only able to describe overt cytotoxicity caused by the test
compounds and, when lower concentrations of the test compounds were used,
they were unable to predict the effects at higher concentrations of the
compounds. In general, candidate markers provided no additional value over
the previously mentioned cytotoxicity markers although intracellular levels of
ALD and CK proved to be sensitive markers of TMPD-induced myopathy in
myotubes.
SELDI-TOF-MS was also used to search for protein biomarkers of myopathy
and the levels of a number of protein ions were changed in response to
treatment with 25μM TMPD in both myoblasts and myotubes. Although the
putative biomarkers discovered using this approach were more sensitive to
treatment than total intracellular protein, MTT reduction, microscopy and the
candidate markers, they were not responsive to <25μM TMPD. In addition, it
was not possible to obtain conclusive identifications for the proteins discovered,
making it difficult to fully understand the biological nature of the changes.
Background variation in the levels of some of the SELDI-TOF-MS ions meant
that, as is often the case with MS experiments, the analysis of a large number
of samples was required in order to generate statistical confidence in the
Chapter 4: Discussion
224
robustness of the changes found. This often means that experiments are
designed in order to overcome potential issues with reproducibility, which
sometimes limits the number of different types of experiment that can be
performed.
A similar limitation to the design of experiments was true of the LFQP approach
in that, although sensitivity was improved using an ion-trap mass spectrometer
and the reproducibility was as would be expected for MS experiments, the
analysis was low-throughput. Despite this, the combination of electrophoresis,
MS/MS and immunoblotting was still the most useful approach to the discovery
of biomarkers as it facilitated the discovery, identification and quantitation of
sensitive protein biomarkers of myopathy in both myoblasts and myotubes.
Chapter 4: Discussion
225
a)Endpoint 6.25 12.5 25 50 100
Total intracellular proteinMTT reductionLight microscopyElectron microscopy
ALT leakageAST leakageCK leakageLDH leakageAldolase leakagesTnI leakageMyl3 leakageFABP3 leakageParvalbumin-α leakage
Intracellular ALT Intracellular AST Intracellular CK Intracellular LDH Intracellular Aldolase Intracellular sTnI Intracellular Myl3 Intracellular FABP3 Intracellular Parvalbumin-α
m/z 5743m/z 6306m/z 6420m/z 6910m/z 7177m/z 9097m/z 9113m/z 9123m/z 9137m/z 9161m/z 6420 and 7177 (combined)
Keratin 2Keratin 19Aldehyde dehydrogenaseSerine proteinase inhibitorEukaryotic tranlsation factor 4BSpectrin repeat containing, nuclear envelopeRho GDP dissociation inhibitor (GDI) alphaLectin, galactose binding
[TMPD] (μM)
Candidate biomarkers (leakage)
Viability endpoints
Label-free quantitative proteomics
Candidate biomarkers (intracellular)
SELDI-TOF-MS anonymous screening
b)Endpoint 6.25 12.5 25 50 100
Total intracellular proteinMTT reductionLight microscopyElectron microscopy
ALT leakageAST leakageCK leakageLDH leakageAldolase leakagesTnI leakageMyl3 leakageFABP3 leakageParvalbumin-α leakage
Intracellular ALT Intracellular AST Intracellular CK Intracellular LDH Intracellular Aldolase Intracellular sTnI Intracellular Myl3 Intracellular FABP3 Intracellular Parvalbumin-α
m/z 3158 m/z 3978 m/z 4815 m/z 5298 m/z 5579 m/z 6305 m/z 6312 m/z 6578 m/z 6588 m/z 12357 m/z 4815 and 5298 (combined)
CaldesmonProteasome 26s subunitWD repeat domain 1ATP synthase H+ transportingCytochrome c oxidase subunit IVNucleoside diphosphate kinase BMyosin light polypeptide 2Ribosomal protein S14Ribosomal protein S19Ubiquitin-conjugating enzyme E2N
Label-free quantitative proteomics
[TMPD] (μM)
Candidate biomarkers (leakage)
Candidate biomarkers (intracellular)
SELDI-TOF-MS anonymous screening
Viability endpoints
Chapter 4: Discussion
226
Figure 69: Summary of protein changes in myoblasts and myotubes Myoblasts and myotubes were cultured and treated with a range of concentrations of
TMPD (section 2.1). A number of approaches to discovering protein biomarkers of
myopathy were attempted and their effectiveness in a) myoblasts and b) myotubes is
summarised here. TMPD was overtly cytotoxic to myoblasts and myotubes at 50μM
and filled boxes represent the concentration at which the levels of markers where
changed significantly in response to treatment (Student’s t-test, p<0.05). Grey boxes
represent the concentrations at which the markers were unresponsive to TMPD-
treatment when tested, and the hatched boxes represent where Rho GDI and Myl2
were measured by immunoblotting. The assessment of endpoints commonly used to
assess cell viability were only changed significantly at concentrations in which overt
cytotoxicity was already evident (Student’s t-test, p<0.05) and intracellular levels of CK
and ALD (in myotubes) were the only candidate biomarkers that were responsive at
low concentrations of TMPD. A number of SELDI-TOF-MS protein ions and proteins
discovered using a combination of electrophoresis and MS/MS were discovered to be
responsive to low concentrations of TMPD. Following the discovery of responsive
proteins by LFQP, changes in Rho GDI and Myl2 were confirmed using immunoblotting
and found to be the most sensitive biomarkers of myopathy.
Chapter 5: Conclusions
227
Chapter 5
5. Conclusions
This work has demonstrated that an in vitro skeletal muscle cell system, using
myoblasts or myotubes, can be established that is responsive to the test
compounds included in this study (TMPD and GWδ). A candidate biomarker
approach to biomarker discovery was attempted but was of little additional value
to routinely used biochemical markers of cytotoxicity. Intracellular levels of ALD
and CK proved to be useful markers of myopathy but other markers tested
were, in general, non-responsive to the treatment of cells with low
concentrations of the test compounds used.
SELDI-TOF-MS was useful in discovering protein ions whose levels were
changed in response to treatment, and these proteins may have been indicative
of early changes in the status of skeletal muscle cells. Although SELDI-TOF-MS
was amenable for high-throughput analysis, it was limited to the low Mw mass
range, although this could also be considered useful due to the fact that other
techniques sometimes neglect this mass range. Interesting proteins were not
readily identified using this SELDI-TOF-MS and this made it difficult to fully
understand the changes observed. Despite the label-free approach to
biomarker identification being limited by its throughput, it proved to be a
valuable method for identifying protein biomarkers of myopathy. Unlike SELDI-
TOF-MS, the conclusive identification of putative biomarkers was routinely
achieved using this technique and this allowed the changes observed to be
considered in their proper biological context.
Chapter 5: Conclusions
228
These studies demonstrated that proteomics is a powerful tool that can be used
to search for biomarkers or patterns of biomarkers of toxicity in an in vitro
system utilising skeletal muscle cells. The use of such biomarkers could allow
researchers to closely monitor the effects of test compounds on experimental
animals in preclinical studies and thus prevent the late stage attrition of
compounds in drug development. In clinical studies or for patients on long-term
treatment for dyslipidaemia, the use of effective biomarkers of myopathy may
permit the safety monitoring of patients for muscle disorders, so that treatment
can be withdrawn or modified before permanent damage can occur.
229
References
1. Howland, R.H. How are drugs approved? Part 2: Ethical foundations of clinical research. Journal of Psychosocial Nursing and Mental Health Services 46, 15-20, 2008.
2. Vitiello, B. and Jensen, P.S. Medication development and testing in children and adolescents: Current problems, future directions. Archives of General Psychiatry 54, 871-876, 1997.
3. Semaan, S., Santibanez, S., Garfein, R.S., Heckathorn, D.D. and Des Jarlais, D.C. Ethical and regulatory considerations in HIV prevention studies employing respondent-driven sampling. International Journal of Drug Policy 20, 14-27, 2009.
4. Macrae, D. Conducting clinical trials in pediatrics. Critical Care Medicine 37, S136-S139, 2009.
5. Kapogiannis, B.G. and Mattison, D.R. Adolescents in clinical trials. Clinical Pharmacology and Therapeutics 84, 655-659, 2008.
6. Halpern, S.D., Karlawish, J.H.T. and Berlin, J.A. The continuing unethical conduct of underpowered clinical trials. Journal of the American Medical Association 288, 358-362, 2002.
7. Otto, S., Furlan, R.L.E. and Sanders, J.K.M. Dynamic combinatorial chemistry. Drug Discovery Today 7, 117-125, 2002.
8. DiMasi, J.A., Hansen, R.W. and Grabowski, H.G. The price of innovation: New estimates of drug development costs. Journal of Health Economics 22, 151-185, 2003.
9. DiMasi, J.A., Grabowski, H.G. and Vernon, J. R&D costs and returns by therapeutic category. Drug Information Journal 38, 211-223, 2004.
10. DiMasi, J.A. Trends in drug development costs, times, and risks. Drug Information Journal 29, 375-384, 1995.
11. Grabowski, H., Vernon, J. and DiMasi, J.A. Returns on research and development for 1990s new drug introductions. PharmacoEconomics 20, 11-29, 2002.
12. Gordon, E.M., Barrett, R.W., Dower, W.J., Fodor, S.P.A. and Gallop, M.A. Applications of combinatorial technologies to drug discovery. 2. Combinatorial organic synthesis, library screening strategies, and future directions. Journal of Medicinal Chemistry 37, 1385-1401, 1994.
230
13. Balkenhohl, F., Von dem Bussche-Hünnefeld, C., Lansky, A. and Zechel, C. Combinatorial synthesis of small organic molecules. Angewandte Chemie (International Edition in English) 35, 2288-2337, 1996.
14. Berndt, E.R., Gottschalk, A.H.B., Philipson, T.J. and Strobeck, M.W. Industry funding of the FDA: Effects of PDUFA on approval times and withdrawal rates. Nature Reviews Drug Discovery 4, 545-554, 2005.
15. Fung, M., Thornton, A., Mybeck, K., Wu, J.H.H., Hornbuckle, K. and Muniz, E. Evaluation of the characteristics of safety withdrawal of prescription drugs from worldwide pharmaceutical markets - 1960 to 1999. Drug Information Journal 35, 293-317, 2001.
16. Tylenda, C.A. The nightmare of FDA clearance/approval to market: Perception or reality? Technology and Health Care 4, 259-267, 1996.
17. Ratliff, A. Biotechnology and pharmaceutical R&D and licensing trends: You pays your money and takes your chances. Journal of Commercial Biotechnology 10, 54-59, 2003.
18. Price, J.M. The liabilities and consequences of medical device development. Journal of Biomedical Materials Research 21, 35-58, 1987.
19. Kahan, S.E., Seftel, A.D. and Resnick, M.I. Sildenafil and the Internet. Journal of Urology 163, 919-923, 2000.
20. Kelly, W.N. Potential risks and prevention, part 1: Fatal adverse drug events. American Journal of Health-System Pharmacy 58, 1317-1324, 2001.
21. Kelly, W.N. Potential risks and prevention, part 2: Drug-induced permanent disabilities. American Journal of Health-System Pharmacy 58, 1325-1329, 2001.
22. Howland, R.H. How are drugs approved? Part 1: The evolution of the food and drug administration. Journal of Psychosocial Nursing and Mental Health Services 46, 15-19, 2008.
23. Saviola, J. The FDA's role in medical device clinical studies of human subjects. Journal of Neural Engineering 2, S1-S4, 2005.
24. Shah, R.B. and Khan, M.A. The evolution of FDA's role in ensuring product quality. Pharmaceutical Technology 31, 52-58, 2007.
25. Howland, R.H. How are drugs approved? Part 3. The stages of drug development. Journal of Psychosocial Nursing and Mental Health Services 46, 17-20, 2008.
231
26. Lipsky, M.S. and Sharp, L.K. From idea to market: The drug approval process. Journal of the American Board of Family Practice 14, 362-367, 2001.
27. Li, A.P. A comprehensive approach for drug safety assessment. Chemico-Biological Interactions 150, 27-33, 2004.
28. Kinter, L.B. and Valentin, J.P. Safety pharmacology and risk assessment. Fundamental and Clinical Pharmacology 16, 175-182, 2002.
29. Mayne, J.T., Ku, W.W. and Kennedy, S.P. Informed toxicity assessment in drug discovery: Systems-based toxicology. Current Opinion in Drug Discovery and Development 9, 75-83, 2006.
30. Pilling, A.M. The role of toxicologic pathologist in the preclinical safety evaluation of biotechnology-derived pharmaceuticals. Toxicologic Pathology 27, 678-688, 1999.
31. Redfern, W.S., Wakefield, I.D., Prior, H., Pollard, C.E., Hammond, T.G. and Valentin, J.P. Safety pharmacology - A progressive approach. Fundamental and Clinical Pharmacology 16, 161-173, 2002.
32. Williams, P.D. The role of pharmacological profiling in safety assessment. Regulatory Toxicology and Pharmacology 12, 238-252, 1990.
33. Brophy, J.M. Celecoxib and cardiovascular risks. Expert Opinion on Drug Safety 4, 1005-1015, 2005.
34. Edwards, I.R. What are the real lessons from Vioxx®? Drug Safety 28, 651-658, 2005.
35. Davies, N.M. and Jamali, F. COX-2 selective inhibitors cardiac toxicity: Getting to the heart of the matter. Journal of Pharmacy and Pharmaceutical Sciences 7, 332-336, 2004.
36. Antman, E.M., Bennett, J.S., Daugherty, A., Furberg, C., Roberts, H. and Taubert, K.A. Use of nonsteroidal antiinflammatory drugs: An update for clinicians: A scientific statement from the American Heart Association. Circulation 115, 1634-1642, 2007.
37. Black, D.M. Issues that led to the withdrawal of cerivastatin. Current Atherosclerosis reports 5, 10, 2003.
38. Griffin, J.P. The withdrawal of Baycol (cerivastatin). Adverse Drug Reactions and Toxicological Reviews 20, 177-180, 2001.
39. Bellosta, S., Ferri, N., Bernini, F., Paoletti, R. and Corsini, A. Non-lipid-related effects of statins. Annals of Medicine 32, 164-176, 2000.
232
40. Bersano, A., Ballabio, E., Lanfranconi, S., Mazzucco, S., Candelise, L. and Monaco, S. Statins and stroke. Current Medicinal Chemistry 15, 2380-2392, 2008.
41. Delanty, N. and Vaughan, C.J. Vascular effects of statins in stroke. Stroke 28, 2315-2320, 1997.
42. Heeschen, C., Hamm, C.W., Laufs, U., Snapinn, S., Böhm, M. and White, H.D. Withdrawal of statins increases event rates in patients with acute coronary syndromes. Circulation 105, 1446-1452, 2002.
43. Lanier, A.B. Use of nonsteroidal anti-inflammatory drugs following exercise-induced muscle injury. Sports Medicine 33, 177-185, 2003.
44. Liao, J.K. and Laufs, U. Pleiotropic effects of statins. Annual Review of Pharmacology and Toxicology 45, 89-118, 2005.
45. Diggle, G.E. Thalidomide: 40 years on. International Journal of Clinical Practice 55, 627-631, 2001.
46. Ames, B.N., Lee, F.D. and Durston, W.E. An improved bacterial test system for the detection and classification of mutagens and carcinogens. Proceedings of the National Academy of Sciences of the United States of America 70, 782-786, 1973.
47. Mortelmans, K. and Zeiger, E. The Ames Salmonella/microsome mutagenicity assay. Mutation Research - Fundamental and Molecular Mechanisms of Mutagenesis 455, 29-60, 2000.
48. Fairbairn, D.W., Olive, P.L. and O'Neill, K.L. The comet assay: A comprehensive review. Mutation Research - Reviews in Genetic Toxicology 339, 37-59, 1995.
49. Tice, R.R., Agurell, E., Anderson, D., Burlinson, B., Hartmann, A., Kobayashi, H., Miyamae, Y., Rojas, E., Ryu, J.C. and Sasaki, Y.F. Single cell gel/comet assay: Guidelines for in vitro and in vivo genetic toxicology testing. Environmental and Molecular Mutagenesis 35, 206-221, 2000.
50. Ono, M., Watabe, S., Nagase, M. and Asami, K. Acute toxicity of ibuprofen by rectal administration in rats and mice. Comparison with oral route. Pharmacometrics 24, 415-430, 1982.
51. Meerum Terwogt, J.M., Schellens, J.H.M., Bokkel Huinink, W.W. and Beijnen, J.H. Clinical pharmacology of anticancer agents in relation to formulations and administration routes. Cancer Treatment Reviews 25, 83-101, 1999.
52. Halliwell, W.H. Cationic amphiphilic drug-induced phospholipidosis. Toxicologic Pathology 25, 53-60, 1997.
233
53. Nonoyama, T. and Fukuda, R. Drug-induced phospholipidosis - Pathological aspects and its prediction. Journal of Toxicologic Pathology 21, 9-24, 2008.
54. Reasor, M.J., Hastings, K.L. and Ulrich, R.G. Drug-induced phospholipidosis: Issues and future directions. Expert Opinion on Drug Safety 5, 567-583, 2006.
55. Takikawa, H. Recent status of drug-induced liver injury. Hepatology Research 39, 1-6, 2009.
56. Kaplowitz, N. Idiosyncratic drug hepatotoxicity. Nature Reviews Drug Discovery 4, 489-499, 2005.
57. Kaplowitz, N. Drug-Induced Liver Injury. Clinical Infectious Diseases 38, S44-S48, 2004.
58. Allison, R.C. and Bedsole, D.L. The other medical causes of rhabdomyolysis. American Journal of Medical Sciences 326, 79-88, 2003.
59. Chatzizisis, Y.S., Misirli, G., Hatzitolios, A.I. and Giannoglou, G.D. The syndrome of rhabdomyolysis: Complications and treatment. European Journal of Internal Medicine 19, 568-574, 2008.
60. Jamal, S.M., Eisenberg, M.J. and Christopoulos, S. Rhabdomyolysis associated with hydroxymethylglutaryl-coenzyme a reductase inhibitors. American Heart Journal 147, 956-965, 2004.
61. Omar, M.A., Wilson, J.P. and Cox, T.S. Rhabdomyolysis and HMG-CoA reductase inhibitors. Annals of Pharmacotherapy 35, 1096-1107, 2001.
62. Lima, R.S., Silva Junior, G.B., Liborio, A.B. and Daher, E.F. Acute kidney injury due to rhabdomyolysis. Saudi journal of kidney diseases and transplantation 19, 721-729, 2008.
63. Evans, M. and Rees, A. Effects of HMG-CoA reductase inhibitors on skeletal muscle: Are all statins the same? Drug Safety 25, 649-663, 2002.
64. Baer, A.N. and Wortmann, R.L. Myotoxicity associated with lipid-lowering drugs. Current Opinion in Rheumatology 19, 67-73, 2007.
65. Wortmann, R.L. Lipid-lowering agents and myopathy. Current Opinion in Rheumatology 14, 643-647, 2002.
66. Devchand, P.R., Ijpenberg, A., Devesvergne, B. and Wahli, W. PPARs: Nuclear receptors for fatty acids, eicosanoids, and xenobiotics. Advances in Experimental Medicine and Biology 469, 231-236, 2000.
234
67. Kersten, S. and Wahli, W. Peroxisome proliferator activated receptor agonists. EXS 89, 141-151, 2000.
68. Kersten, S., Desvergne, B. and Wahli, W. Roles of PPARS in health and disease. Nature 405, 421-424, 2000.
69. Gervois, P., Torra, I.P., Fruchart, J.C. and Staels, B. Regulation of lipid and lipoprotein metabolism by PPAR activators. Clinical Chemistry and Laboratory Medicine 38, 3-11, 2000.
70. Gilde, A.J., Fruchart, J.C. and Staels, B. Peroxisome Proliferator-Activated Receptors at the Crossroads of Obesity, Diabetes, and Cardiovascular Disease. Journal of the American College of Cardiology 48, A24-A32, 2006.
71. Pourcet, B., Fruchart, J.C., Staels, B. and Glineur, C. Selective PPAR modulators, dual and pan PPAR agonists: Multimodal drugs for the treatment of Type 2 diabetes and atherosclerosis. Expert Opinion on Emerging Drugs 11, 379-401, 2006.
72. Staels, B., Dallongeville, J., Auwerx, J., Schoonjans, K., Leitersdorf, E. and Fruchart, J.C. Mechanism of action of fibrates on lipid and lipoprotein metabolism. Circulation 98, 2088-2093, 1998.
73. Staels, B., Maes, M. and Zambon, A. Fibrates and future PPARα agonists in the treatment of cardiovascular disease. Nature Clinical Practice Cardiovascular Medicine 5, 542-553, 2008.
74. Muoio, D.M., Way, J.M., Tanner, C.J., Winegar, D.A., Kliewer, S.A., Houmard, J.A., Kraus, W.E. and Lynis Dohm, G. Peroxisome proliferator-activated receptor-α regulates fatty acid utilization in primary human skeletal muscle cells. Diabetes 51, 901-909, 2002.
75. Kim, D.J., Bility, M.T., Billin, A.N., Willson, T.M., Gonzalez, F.J. and Peters, J.M. PPARβ/δ selectively induces differentiation and inhibits cell proliferation. Cell Death and Differentiation 13, 53-60, 2006.
76. Oliver, J., Shenk, J.L., Snaith, M.R., Russell, C.S., Plunket, K.D., Bodkin, N.L., Lewis, M.C., Winegar, D.A., Sznaidman, M.L., Lambert, M.H., Xu, H.E., Sternbach, D.D., Kliewer, S.A., Hansen, B.C. and Willson, T.M. A selective peroxisome proliferator-activated receptor δ agonist promotes reverse cholesterol transport. Proceedings of the National Academy of Sciences of the United States of America 98, 5306-5311, 2001.
77. Shearer, B.G., Patel, H.S., Billin, A.N., Way, J.M., Winegar, D.A., Lambert, M.H., Xu, R.X., Leesnitzer, L.M., Merrihew, R.V., Huet, S. and Willson, T.M. Discovery of a novel class of PPARδ partial agonists. Bioorganic and Medicinal Chemistry Letters 18, 5018-5022, 2008.
235
78. Billin, A.N. PPAR-β/δ agonists for type 2 diabetes and dyslipidemia: An adopted orphan still looking for a home. Expert Opinion on Investigational Drugs 17, 1465-1471, 2008.
79. Armitage, J. The safety of statins in clinical practice. Lancet 370, 1781-1790, 2007.
80. Bellosta, S., Paoletti, R. and Corsini, A. Safety of statins: Focus on clinical pharmacokinetics and drug interactions. Circulation 109, III50-III57, 2004.
81. Bolego, C., Baetta, R., Bellosta, S., Corsini, A. and Paoletti, R. Safety considerations for statins. Current Opinion in Lipidology 13, 637-644, 2002.
82. Brown, W.V. Safety of statins. Current Opinion in Lipidology 19, 558-562, 2008.
83. Endres, M. Statins and stroke. Journal of Cerebral Blood Flow and Metabolism 25, 1093-1110, 2005.
84. Sinzinger, H., Wolfram, R. and Peskar, B.A. Muscular side effects of statins. Journal of Cardiovascular Pharmacology 40, 163-171, 2002.
85. Hodel, C. Myopathy and rhabdomyolysis with lipid-lowering drugs. Toxicology Letters 128, 159-168, 2002.
86. Munday, R., Manns, E., Fowke, E.A. and Hoggard, G.K. Muscle necrosis by N-methylated p-phenylenediamines in rats: Structure-activity relationships and correlation with free-radical production in vitro. Toxicology 57, 303-314, 1989.
87. Munday, R., Manns, E., Fowke, E.A. and Hoggard, G.K. Structure-activity relationships in the myotoxicity of ring-methylated p-phenylenediamines in rats and correlation with autoxidation rates in vitro. Chemico-Biological Interactions 76, 31-45, 1990.
88. Harauchi, T. and Hirata, M. Effects of p-phenylenediamines and adriamycin on primary culture of rat skeletal muscle cells. Toxicology Letters 66, 35-46, 1993.
89. Blair, J.A., Waterfield, C.J., Lightfoot, R.M., Rowlands, A., York, M.J. and Turton, J.A. The skeletal muscle-specific toxicity of 2,3,5,6 tetramethyl-p-phenylenediamine (TMPD): some preliminary studies. Human and Experimental Toxicology 16, 403, 1997.
90. Draper, R.P., Waterfield, C.J., York, M.J. and Timbrell, J.A. Studies on the muscle toxicant 2,3,5,6-tetramethyl p-phenylenediamine: Effects on various biomarkers including urinary creatine and taurine. Archives of Toxicology 69, 111-117, 1994.
236
91. Bohlmeyer, T.J., Wu, A.H.B. and Perryman, M.B. Evaluation of laboratory tests as a guide to diagnosis and therapy of myositis. Rheumatic Disease Clinics of North America 20, 845-856, 1994.
92. Lott, J.A. and Landesman, P.W. The enzymology of skeletal muscle disorders. Critical Reviews in Clinical Laboratory Sciences 20, 153-190, 1984.
93. Lott, J.A. and Stang, J.M. Differential diagnosis of patients with abnormal serum creatine kinase isoenzymes. Clinics in Laboratory Medicine 9, 627-642, 1989.
94. Walker, D. Serum chemical biomarkers of cardiac injury for nonclinical safety testing. Toxicologic Pathology 34, 94-104, 2006.
95. Dare, T.O., Davies, H.A., Turton, J.A., Lomas, L., Williams, T.C. and York, M.J. Application of surface-enhanced laser desorption/ionization technology to the detection and identification of urinary parvalbumin-α: A biomarker of compound-induced skeletal muscle toxicity in the rat. Electrophoresis 23, 3241-3251, 2002.
96. Margaritis, I., Tessier, F., Verdera, F., Bermon, S. and Marconnet, P. Muscle enzyme release does not predict muscle function impairment after triathlon. Journal of Sports Medicine and Physical Fitness 39, 133-139, 1999.
97. Collings, F.B. and Vaidya, V.S. Novel technologies for the discovery and quantitation of biomarkers of toxicity. Toxicology 245, 167-174, 2008.
98. Crisman, T.S., Claffey, K.P. and Saouaf, R. Measurement of rat heart fatty acid binding protein by ELISA. Tissue distribution, development changes and subcellular distribution. Journal of Molecular and Cellular Cardiology 19, 423-431, 1987.
99. Simpson, J.A., Labugger, R., Collier, C., Brison, R.J., Iscoe, S. and Van Eyk, J.E. Fast and slow skeletal troponin I in serum from patients with various skeletal muscle disorders: A pilot study. Clinical Chemistry 51, 966-972, 2005.
100. Celio, M.R. and Heizmann, C.W. Calcium-binding protein parvalbumin is associated with fast contracting muscle fibres. Nature 297, 504-506, 1982.
101. Balls, M. Replacement of animal procedures: Alternatives in research, education and testing. Laboratory Animals 28, 193-211, 1994.
102. Pearson, R.M. In-vitro techniques: Can they replace animal testing? Human Reproduction 1, 559-560, 1986.
237
103. Peters, T.S. Do preclinical testing strategies help predict human hepatotoxic potentials? Toxicologic Pathology 33, 146-154, 2005.
104. Richmond, J. Refinement, reduction, and replacement of animal use for regulatory testing: Future improvements and implementation within the regulatory framework. ILAR Journal 43, S63-S68, 2002.
105. Rowan, A.N. and Goldberg, A.M. Perspectives on alternatives to current animal testing techniques in preclinical toxicology. Annual Review of Pharmacology and Toxicology 25, 225-247, 1985.
106. Atterwill, C.K. and Wing, M.G. In vitro preclinical lead optimisation technologies (PLOTs) in pharmaceutical development. Toxicology Letters 127, 143-151, 2002.
107. Richler, C. and Yaffe, D. The in vitro cultivation and differentiation capacities of myogenic cell lines. Developmental Biology 23, 1-22, 1970.
108. Yaffe, D. and Saxel, O. Serial passaging and differentiation of myogenic cells isolated from dystrophic mouse muscle. Nature 270, 725-727, 1977.
109. Yaffe, D. Retention of differentiation potentialities during prolonged cultivation of myogenic cells. Proceedings of the National Academy of Sciences of the United States of America 61, 477-483, 1968.
110. Yaffe, D. and Saxel, O. A myogenic cell line with altered serum requirements for differentiation. Differentiation 7, 159-166, 1977.
111. LaBarge, M.A. and Blau, H.M. Biological progression from adult bone marrow to mononucleate muscle stem cell to multinucleate muscle fiber in response to injury. Cell 111, 589-601, 2002.
112. Kato, I., Harihara, A. and Mizushima, Y. An in vitro model for assessing muscle irritation of antibiotics using rat primary cultured skeletal muscle fibers. Toxicology and Applied Pharmacology 117, 194-199, 1992.
113. Brockes, J.P. and Kumar, A. Comparative aspects of animal regeneration. Annual Review of Cell and Developmental Biology 24, 525-549, 2008.
114. Duckmanton, A., Kumar, A., Chang, Y.T. and Brockes, J.P. A single-cell analysis of myogenic dedifferentiation induced by small molecules. Chemistry and Biology 12, 1117-1126, 2005.
115. Godwin, J.W. and Brockes, J.P. Regeneration, tissue injury and the immune response. Journal of Anatomy 209, 423-432, 2006.
116. Lo, D.C., Allen, F. and Brockes, J.P. Reversal of muscle differentiation during urodele limb regeneration. Proceedings of the National Academy of Sciences of the United States of America 90, 7230-7234, 1993.
238
117. Velloso, C.P., Kumar, A., Tanaka, E.M. and Brockes, J.P. Generation of mononucleate cells from post-mitotic myotubes proceeds in the absence of cell cycle progression. Differentiation 66, 239-246, 2000.
118. Nathwani, R.A., Pais, S., Reynolds, T.B. and Kaplowitz, N. Serum alanine aminotransferase in skeletal muscle diseases. Hepatology 41, 380-382, 2005.
119. Targoff, I.N. Laboratory testing in the diagnosis and management of idiopathic inflammatory myopathies. Rheumatic Disease Clinics of North America 28, 859-890, 2002.
120. Rider, L.G. and Miller, F.W. Laboratory evaluation of the inflammatory myopathies. Clinical and Diagnostic Laboratory Immunology 2, 1-9, 1995.
121. Dufour, D.R. Creatine kinase:aspartate aminotransferase activity ratio as an indicator of the source of an increased creatine kinase activity. Clinical Chemistry 34, 2506-2510, 1988.
122. Nuwaysir, E.F., Bittner, M., Trent, J., Barrett, J.C. and Afshari, C.A. Microarrays and toxicology: The advent of toxicogenomics. Molecular Carcinogenesis 24, 153-159, 1999.
123. Tomer, K.B., Dubi, J., Devereux, T.R., Lida, M., Sills, R., Cunningham, M., Paules, R., Bruno, M., Hartis, J., Sieber, S., Wetmore, B. and Merrick, B.A. Toxicogenomics/proteomics of non-genotoxic carcinogens. Proceedings 50th ASMS Conference on Mass Spectrometry and Allied Topics, 263-264, 2002.
124. Merrick, B.A. and Bruno, M.E. Genomic and proteomic profiling for biomarkers and signature profiles of toxicity. Current Opinion in Molecular Therapeutics 6, 600-607, 2004.
125. Roussel, E. and Gingras, M.C. High-throughput gene expression profiling - A work-in-progress with great potential for proteomics. Current Opinion in Drug Discovery and Development 9, 332-338, 2006.
126. Metsis, A., Andersson, U., Baurén, G., Ernfors, P., Lönnerberg, P., Montelius, A., Oldin, M., Pihlak, A. and Linnarsson, S. Whole-genome expression profiling through fragment display and combinatorial gene identification. Nucleic Acids Research 32, 2004.
127. Scheel, J., von Brevern, M.C., Hörlein, A., Fischer, A., Schneider, A. and Bach, A. Yellow pages to the transcriptome. Pharmacogenomics 3, 791-807, 2002.
128. Ilyin, S.E., Belkowski, S.M. and Plata-Salamán, C.R. Biomarker discovery and validation: Technologies and integrative approaches. Trends in Biotechnology 22, 411-416, 2004.
239
129. Timbrell, J.A. Biomarkers in toxicology. Toxicology 129, 1-12, 1998.
130. Zolg, J.W. and Langen, H. How industry is approaching the search for new diagnostic markers and biomarkers. Molecular and Cellular Proteomics 3, 345-354, 2004.
131. Atkinson, A.J., Colburn, W.A., DeGruttola, V.G., DeMets, D.L., Downing, G.J., Hoth, D.F., Oates, J.A., Peck, C.C., Schooley, R.T., Spilker, B.A., Woodcock, J. and Zeger, S.L. Biomarkers and surrogate endpoints: Preferred definitions and conceptual framework. Clinical Pharmacology and Therapeutics 69, 89-95, 2001.
132. Clarke, C.J. and Haselden, J.N. Metabolic profiling as a tool for understanding mechanisms of toxicity. Toxicologic Pathology 36, 140-147, 2008.
133. Coen, M., Holmes, E., Lindon, J.C. and Nicholson, J.K. NMR-based metabolic profiling and metabonomic approaches to problems in molecular toxicology. Chemical Research in Toxicology 21, 9-27, 2008.
134. Holmes, E., Wilson, I.D. and Nicholson, J.K. Metabolic Phenotyping in Health and Disease. Cell 134, 714-717, 2008.
135. Lindon, J.C. and Nicholson, J.K. Analytical technologies for metabonomics and metabolomics, and multi-omic information recovery. TrAC - Trends in Analytical Chemistry 27, 194-204, 2008.
136. Xu, E.Y., Schaefer, W.H. and Xu, Q. Metabolomics in pharmaceutical research and development: Metabolites, mechanisms and pathways. Current Opinion in Drug Discovery and Development 12, 40-52, 2009.
137. Wishart, D.S., Knox, C., Guo, A.C., Eisner, R., Young, N., Gautam, B., Hau, D.D., Psychogios, N., Dong, E., Bouatra, S., Mandal, R., Sinelnikov, I., Xia, J., Jia, L., Cruz, J.A., Lim, E., Sobsey, C.A., Shrivastava, S., Huang, P., Liu, P., Fang, L., Peng, J., Fradette, R., Cheng, D., Tzur, D., Clements, M., Lewis, A., de souza, A., Zuniga, A., Dawe, M., Xiong, Y., Clive, D., Greiner, R., Nazyrova, A., Shaykhutdinov, R., Li, L., Vogel, H.J. and Forsythei, I. HMDB: A knowledgebase for the human metabolome. Nucleic Acids Research 37, D603-D610, 2009.
138. Vangala, S. and Tonelli, A. Biomarkers, metabonomics, and drug development: Can inborn errors of metabolism help in understanding drug toxicity? AAPS Journal 9, E284-E297, 2007.
139. Griffin, J.L., Williams, H.J., Sang, E., Clarke, K., Rae, C. and Nicholson, J.K. Metabolic profiling of genetic disorders: A multitissue 1H nuclear magnetic resonance spectroscopic and pattern recognition study into dystrophic tissue. Analytical Biochemistry 293, 16-21, 2001.
240
140. McIntosh, L., Granberg, K.E., Brière, K.M. and Anderson, J.E. Nuclear magnetic resonance spectroscopy study of muscle growth, mdx dystrophy and glucocorticoid treatments: Correlation with repair. NMR in Biomedicine 11, 1-10, 1998.
141. Lindon, J.C., Holmes, E., Bollard, M.E., Stanley, E.G. and Nicholson, J.K. Metabonomics technologies and their applications in physiological monitoring, drug safety assessment and disease diagnosis. Biomarkers 9, 1-31, 2004.
142. Shaw, M. Novel biomarker platforms in toxicology. Drug Discovery Today: Technologies 3, 95-102, 2006.
143. Hirata, A., Masuda, S., Tamura, T., Kai, K., Ojima, K., Fukase, A., Motoyoshi, K., Kamakura, K., Miyagoe-Suzuki, Y. and Takeda, S. Expression profiling of cytokines and related genes in regenerating skeletal muscle after cardiotoxin injection: A role for osteopontin. American Journal of Pathology 163, 203-215, 2003.
144. Huby, R. and Tugwood, J.D. Gene expression profiling for pharmaceutical safety assessment. Expert Opinion on Drug Metabolism & Toxicology 1, 247-260, 2005.
145. Tugwood, J.D., Hollins, L.E. and Cockerill, M.J. Genomics and the search for novel biomarkers in toxicology. Biomarkers 8, 79-92, 2003.
146. Lord, P.G., Nie, A. and McMillian, M. The evolution of gene expression studies in drug safety assessment. Toxicology Mechanisms and Methods 16, 51-58, 2006.
147. Orr, M.S. Toxicogenomics and cross-species biomarker discovery: Applications in drug discovery and safety assessment. Toxicology Mechanisms and Methods 16, 79-87, 2006.
148. Gant, T.W. Novel and future applications of microarrays in toxicological research. Expert Opinion on Drug Metabolism and Toxicology 3, 599-608, 2007.
149. Hortin, G.L., Jortani, S.A., Ritchie, J., Valdes, J. and Chan, D.W. Proteomics: A new diagnostic frontier. Clinical Chemistry 52, 1218-1222, 2006.
150. Chan, D.W. Clinical proteomics: My view of the current issues. Clinical Proteomics 2, 1-4, 2006.
151. Anderson, L. and Seilhamer, J. A comparison of selected mRNA and protein abundances in human liver. Electrophoresis 18, 533-537, 1997.
241
152. Anderson, N.L. and Anderson, N.G. The human plasma proteome: history, character, and diagnostic prospects. Molecular & Cellular Proteomics : MCP 1, 845-867, 2002.
153. Pandey, A. and Mann, M. Proteomics to study genes and genomes. Nature 405, 837-846, 2000.
154. Tress, M.L., Bodenmiller, B., Aebersold, R. and Valencia, A. Proteomics studies confirm the presence of alternative protein isoforms on a large scale. Genome Biology 9, R162, 2008.
155. Mathesius, U., Imin, N., Natera, S.H. and Rolfe, B.G. Proteomics as a functional genomics tool. Methods in molecular biology 236, 395-414, 2003.
156. Lin, D., Tabb, D.L. and Yates III, J.R. Large-scale protein identification using mass spectrometry. Biochimica et Biophysica Acta - Proteins and Proteomics 1646, 1-10, 2003.
157. Mann, M., Hendrickson, R.C. and Pandey, A. Analysis of proteins and proteomes by mass spectrometry. Annual Review of Biochemistry 70, 437-473, 2001.
158. Ansong, C., Purvine, S.O., Adkins, J.N., Lipton, M.S. and Smith, R.D. Proteogenomics: Needs and roles to be filled by proteomics in genome annotation. Briefings in Functional Genomics and Proteomics 7, 50-62, 2008.
159. Grizzle, W.E., Adam, B.L., Bigbee, W.L., Conrads, T.P., Carroll, C., Feng, Z., Izbicka, E., Jendoubi, M., Johnsey, D., Kagan, J., Leach, R.J., McCarthy, D.B., Semmes, O.J., Srivastava, S., Srivastava, S., Thompson, I.M., Thornquist, M.D., Verma, M., Zhang, Z. and Zou, Z. Serum protein expression profiling for cancer detection: Validation of a SELDI-based approach for prostate cancer. Disease Markers 19, 185-195, 2003.
160. Gast, M.C.W., Schellens, J.H.M. and Beijnen, J.H. Clinical proteomics in breast cancer: a review. Breast Cancer Research and Treatment 116, 17-29, 2009.
161. Wulfkuhle, J.D., McLean, K.C., Paweletz, C.P., Sgroi, D.C., Trock, B.J., Steeg, P.S. and Petricoin III, E.F. New approaches to proteomic analysis of breast cancer. Proteomics 1, 1205-1215, 2001.
162. Zhang, J.Z., Zheng, Y.H., Feng, K. and Zou, D.W. Application of SELDI protein fingerprinting in early detection of cancer. World Chinese Journal of Digestology 12, 2773-2777, 2004.
242
163. Loo, L.H., Quinn, J., Cordingley, H., Roberts, S., Hrebien, L. and Kam, M. Classification of SELDI-ToF mass spectra of ovarian cancer serum samples using a proteomic pattern recognizer. Bioengineering, Proceedings of the Northeast Conference, 130-131, 2003.
164. Petricoin, E.F. and Liotta, L.A. SELDI-TOF-based serum proteomic pattern diagnostics for early detection of cancer. Current Opinion in Biotechnology 15, 24-30, 2004.
165. Petricoin, E.F., Ornstein, D.K. and Liotta, L.A. Clinical proteomics: Applications for prostate cancer biomarker discovery and detection. Urologic Oncology: Seminars and Original Investigations 22, 322-328, 2004.
166. Solassol, J., Jacot, W., Lhermitte, L., Boulle, N., Maudelonde, T. and Mangé, A. Clinical proteomics and mass spectrometry profiling for cancer detection. Expert Review of Proteomics 3, 311-320, 2006.
167. Bañez, L.L., Srivastava, S. and Moul, J.W. Proteomics in prostate cancer. Current Opinion in Urology 15, 151-156, 2005.
168. Man, W.J., White, I.R., Bryant, D., Bugelski, P., Camilleri, P., Cutler, P., Heald, G., Lord, P.G., Wood, J. and Kramer, K. Protein expression analysis of drug-mediated hepatotoxicity in the Sprague-Dawley rat. Proteomics 2, 1577-1585, 2002.
169. Yamamoto, T., Kikkawa, R., Yamada, H. and Horii, I. Investigation of proteomic biomarkers in in vivo hepatotoxicity study of rat liver: Toxicity differentiation in hepatotoxicants. Journal of Toxicological Sciences 31, 49-60, 2006.
170. Fountoulakis, M., Berndt, P., Boelsterli, U.A., Crameri, F., Winter, M., Albertini, S. and Suter, L. Two-dimensional database of mouse liver proteins: Changes in hepatic protein levels following treatment with acetaminophen or its nontoxic regioisomer 3-acetamidophenol. Electrophoresis 21, 2148-2161, 2000.
171. Gao, J., Garulacan, L.A., Storm, S.M., Hefta, S.A., Opiteck, G.J., Lin, J.H., Moulin, F. and Dambach, D.M. Identification of in vitro protein biomarkers of idiosyncratic liver toxicity. Toxicology in Vitro 18, 533-541, 2004.
172. Newsholme, S.J., Maleeff, B.F., Steiner, S., Anderson, N.L. and Schwartz, L.W. Two-dimensional electrophoresis of liver proteins: Characterization of a drug-induced hepatomegaly in rats. Electrophoresis 21, 2122-2128, 2000.
243
173. Teck, Y.L., Chon, K.L., Salto-Tellez, M. and Chung, M.C.M. A proteomic analysis of thioacetamide-induced hepatotoxicity and cirrhosis in rat livers. Proteomics 4, 3960-3974, 2004.
174. Abdul-Salam, V.B., Paul, G.A., Ali, J.O., Gibbs, S.R., Rahman, D., Taylor, G.W., Wilkins, M.R. and Edwards, R.J. Identification of plasma protein biomarkers associated with idiopathic pulmonary arterial hypertension. Proteomics 6, 2286-2294, 2006.
175. Yu, M., Wang, X.X., Zhang, F.R., Shang, Y.P., Du, Y.X., Chen, H.J. and Chen, J.Z. Proteomic analysis of the serum in patients with idiopathic pulmonary arterial hypertension. Journal of Zhejiang University Science B 8, 221-227, 2007.
176. Kwapiszewska, G., Wygrecka, M., Marsh, L.M., Schmitt, S., Trösser, R., Wilhelm, J., Helmus, K., Eul, B., Zakrzewicz, A., Ghofrani, H.A., Schermuly, R.T., Bohle, R.M., Grimminger, F., Seeger, W., Eickelberg, O., Fink, L. and Weissmann, N. Fhl-1, a new key protein in pulmonary hypertension. Circulation 118, 1183-1194, 2008.
177. Zhang, G., Zhou, B., Zheng, Y., Feng, K., Rao, L., Zhang, J., Xin, J., Zhang, B. and Zhang, L. Time course proteomic profile of rat acute myocardial infarction by SELDI-TOF MS analysis. International Journal of Cardiology 131, 225-233, 2009.
178. Thongboonkerd, V. Proteomics in nephrology: Current status and future directions. American Journal of Nephrology 24, 360-378, 2004.
179. Cutler, P., Bell, D.J., Birrell, H.C., Connelly, J.C., Connor, S.C., Holmes, E., Mitchell, B.C., Monté, S.Y., Neville, B.A., Pickford, R., Polley, S., Schneider, K. and Skehel, J.M. An integrated proteomic approach to studying glomerular nephrotoxicity. Electrophoresis 20, 3647-3658, 1999.
180. Bandara, L.R., Kelley, M.D., Lock, E.A. and Kennedy, S. A potential biomarker of kidney damage identified by proteomics: Preliminary findings. Biomarkers 8, 272-286, 2003.
181. Thongboonkerd, V. Proteomic analysis of renal diseases: Unraveling the pathophysiology and biomarker discovery. Expert Review of Proteomics 2, 349-366, 2005.
182. Emery, P.W. and Preedy, V.R. Measuring muscle protein turnover in vivo: What can 3-methylhistidine production tell us? Clinical Science 104, 557-558, 2003.
183. Piec, I., Listrat, A., Alliot, J., Chambon, C., Taylor, R.G. and Bechet, D. Differential proteome analysis of aging in rat skeletal muscle. FASEB Journal 19, 1143-1145, 2005.
244
184. Fernandez-Solà, J., Preedy, V.R., Lang, C.H., Gonzalez-Reimers, E., Arno, M., Lin, J.C.I., Wiseman, H., Zhou, S., Emery, P.W., Nakahara, T., Hashimoto, K., Hirano, M., Santolaria-Fernández, F., González-Hernández, T., Fatjó, F., Sacanella, E., Estruch, R., Nicolás, J.M. and Urbano-Márquez, A. Molecular and cellular events in alcohol-induced muscle disease. Alcoholism: Clinical and Experimental Research 31, 1953-1962, 2007.
185. Merrick, B.A. The human proteome organization (HUPO) and environmental health. Environmental Health Perspectives 111, 1-5, 2003.
186. Issaq, H.J. The role of separation science in proteomics research. Electrophoresis 22, 3629-3638, 2001.
187. Domon, B. and Aebersold, R. Mass spectrometry and protein analysis. Science 312, 212-217, 2006.
188. Scherperel, G. and Reid, G.E. Emerging methods in proteomics: Top-down protein characterization by multistage tandem mass spectrometry. Analyst 132, 500-506, 2007.
189. Han, X., Aslanian, A. and Yates III, J.R. Mass spectrometry for proteomics. Current Opinion in Chemical Biology 12, 483-490, 2008.
190. Chen, G. and Pramanik, B.N. LC-MS for protein characterization: Current capabilities and future trends. Expert Review of Proteomics 5, 435-444, 2008.
191. Wysocki, V.H., Resing, K.A., Zhang, Q. and Cheng, G. Mass spectrometry of peptides and proteins. Methods 35, 211-222, 2005.
192. Kettman, J.R., Frey, J.R. and Lefkovits, I. Proteome, transcriptome and genome: Top down or bottom up analysis? Biomolecular Engineering 18, 207-212, 2001.
193. Perkins, D.N., Pappin, D.J.C., Creasy, D.M. and Cottrell, J.S. Probability-based protein identification by searching sequence databases using mass spectrometry data. Electrophoresis 20, 3551-3567, 1999.
194. Gevaert, K. and Vandekerckhove, J. Protein identification methods in proteomics. Electrophoresis 21, 1145-1154, 2000.
195. James, P., Quadroni, M., Carafoli, E. and Gonnet, G. Protein identification by mass profile fingerprinting. Biochemical and Biophysical Research Communications 195, 58-64, 1993.
196. Hunt, D.F., Yates III, J.R. and Shabanowitz, J. Protein sequencing by tandem mass spectrometry. Proceedings of the National Academy of Sciences of the United States of America 83, 6233-6237, 1986.
245
197. Wilkins, M.R., Appel, R.D., Van Eyk, J.E., Chung, M.C.M., Görg, A., Hecker, M., Huber, L.A., Langen, H., Link, A.J., Paik, Y.K., Patterson, S.D., Pennington, S.R., Rabilloud, T., Simpson, R.J., Weiss, W. and Dunn, M.J. Guidelines for the next 10 years of proteomics. Proteomics 6, 4-8, 2006.
198. Shevchenko, A., Wilm, M., Vorm, O. and Mann, M. Mass spectrometric sequencing of proteins from silver-stained polyacrylamide gels. Analytical Chemistry 68, 850-858, 1996.
199. Rosenfeld, J., Capdevielle, J., Guillemot, J.C. and Ferrara, P. In-gel digestion of proteins for internal sequence analysis after one- or two-dimensional gel electrophoresis. Analytical Biochemistry 203, 173-179, 1992.
200. Issaq, H.J., Chan, K.C., Janini, G.M., Conrads, T.P. and Veenstra, T.D. Multidimensional separation of peptides for effective proteomic analysis. Journal of Chromatography B: Analytical Technologies in the Biomedical and Life Sciences 817, 35-47, 2005.
201. Jenö, P., Mini, T., Moes, S., Hintermann, E. and Horst, M. Internal sequences from proteins digested in polyacrylamide gels. Analytical Biochemistry 224, 75-82, 1995.
202. Kiser, J.Z., Post, M., Wang, B. and Miyagi, M. Streptomyces erythraeus trypsin for proteomics applications. Journal of Proteome Research 8, 1810-1817, 2009.
203. Dong, M.W. Tryptic mapping by reversed phase liquid chromatography. Advances in Chromatography 32, 21-51, 1992.
204. Ingram, V. Sequence methods. Methods in Enzymology 6, 831-848, 1963.
205. Keil-Dlouhá, V., Zylber, N., Imhoff, J.M., Tong, N.T. and Keil, B. Proteolytic activity of pseudotrypsin. FEBS Letters 16, 291-295, 1971.
206. Weber, K. and Osborn, M. The reliability of molecular weight determinations by dodecyl sulfate-polyacrylamide gel electrophoresis. Journal of Biological Chemistry 244, 4406-4412, 1969.
207. O'Farrell, P.H. High resolution two dimensional electrophoresis of proteins. Journal of Biological Chemistry 250, 4007-4021, 1975.
208. Witzmann, F.A. and Richardson, M.R. Two-dimensional gels for toxicological drug discovery applications. Expert Opinion on Drug Metabolism and Toxicology 2, 103-111, 2006.
246
209. Beranova-Giorgianni, S. Proteome analysis by two-dimensional gel electrophoresis and mass spectrometry: Strengths and limitations. TrAC - Trends in Analytical Chemistry 22, 273-281, 2003.
210. Lilley, K.S., Razzaq, A. and Dupree, P. Two-dimensional gel electrophoresis: Recent advances in sample preparation, detection and quantitation. Current Opinion in Chemical Biology 6, 46-50, 2002.
211. Wu, S., Tang, X.T., Siems, W.F. and Bruce, J.E. A hybrid LC-Gel-MS method for proteomics research and its application to protease functional pathway mapping. Journal of Chromatography B: Analytical Technologies in the Biomedical and Life Sciences 822, 98-111, 2005.
212. Sandra, K., Moshir, M., D'hondt, F., Verleysen, K., Kas, K. and Sandra, P. Highly efficient peptide separations in proteomics. Part 1. Unidimensional high performance liquid chromatography. Journal of Chromatography B: Analytical Technologies in the Biomedical and Life Sciences 866, 48-63, 2008.
213. Stoll, D.R., Li, X., Wang, X., Carr, P.W., Porter, S.E.G. and Rutan, S.C. Fast, comprehensive two-dimensional liquid chromatography. Journal of Chromatography A 1168, 3-43, 2007.
214. Sickmann, A., Mreyen, M. and Meyer, H.E. Mass spectrometry--a key technology in proteome research. Advances in Biochemical Engineering/Biotechnology 83, 141-176, 2003.
215. Gusev, A.I. Interfacing matrix-assisted laser desorption/ionization mass spectrometry with column and planar separations. Fresenius' Journal of Analytical Chemistry 366, 691-700, 2000.
216. Kowalski, P. and Stoerker, J. Accelerating discoveries in the proteome and genome with MALDI TOF MS. Pharmacogenomics 1, 359-366, 2000.
217. Fenn, J.B., Mann, M., Meng, C.K., Wong, S.F. and Whitehouse, C.M. Electrospray ionization-principles and practice. Mass Spectrometry Reviews 9, 37-70, 1990.
218. Fenn, J.B., Rosell, J. and Meng, C.K. In electrospray ionization, how much pull does an ion need to escape its droplet prison? Journal of the American Society for Mass Spectrometry 8, 1147-1157, 1997.
219. Kebarle, P. and Tang, L. From ions in solution to ions in the gas phase: The mechanism of electrospray mass spectrometry. Analytical Chemistry 65, 1993.
220. Kebarle, P. and Peschke, M. On the mechanisms by which the charged droplets produced by electrospray lead to gas phase ions. Analytica Chimica Acta 406, 11-35, 2000.
247
221. Iribarne, J.V. and Thomson, B.A. On the evaporation of small ions from charged droplets. The Journal of Chemical Physics 64, 2287-2294, 1976.
222. Rohner, T.C., Lion, N. and Girault, H.H. Electrochemical and theoretical aspects of electrospray ionisation. Physical Chemistry Chemical Physics 6, 3056-3068, 2004.
223. Graves, P.R. and Haystead, T.A.J. Molecular biologist's guide to proteomics. Microbiology and Molecular Biology Reviews 66, 39-63, 2002.
224. Hunter, T.C., Andon, N.L., Koller, A., Yates III, J.R. and Haynes, P.A. The functional proteomics toolbox: Methods and applications. Journal of Chromatography B: Analytical Technologies in the Biomedical and Life Sciences 782, 165-181, 2002.
225. Ong, S.E. and Mann, M. Mass spectrometry-based proteomics turns quantitative. Nature Chemical Biology 1, 252-262, 2005.
226. Von Neuhoff, N., Pich, A. and Milos, P.M. Mass spectrometry-based methods for biomarker detection and analysis. Drug Discovery Today: Technologies 2, 361-367, 2005.
227. Qian, W.J., Jacobs, J.M., Liu, T., Camp II, D.G. and Smith, R.D. Advances and challenges in liquid chromatography-mass spectrometry-based proteomics profiling for clinical applications. Molecular and Cellular Proteomics 5, 1727-1744, 2006.
228. Schaeffer-Reiss, C. A brief summary of the different types of mass spectrometers used in proteomics. Methods in Molecular Biology 484, 3-16, 2008.
229. Paweletz, C.P., Gillespie, J.W., Ornstein, D.K., Simone, N.L., Brown, M.R., Cole, K.A., Wang, Q.H., Huang, J., Hu, N., Yip, T.T., Rich, W.E., Kohn, E.C., Linehan, W.M., Weber, T., Taylor, P., Emmert-Buck, M.R., Liotta, L.A. and Petricoin III, E.F. Rapid protein display profiling of cancer progression directly from human tissue using a protein biochip. Drug Development Research 49, 34-42, 2000.
230. Fung, E., Diamond, D., Simonsesn, A.H. and Weinberger, S.R. The use of SELDI ProteinChip array technology in renal disease research. Methods in Molecular Medicine 86, 295-312, 2003.
231. Merchant, M. and Weinberger, S.R. Recent advancements in surface-enhanced laser desorption/ionization-time of flight-mass spectrometry. Electrophoresis 21, 1164-1177, 2000.
248
232. Weinberger, S.R., Boschetti, E., Santambien, P. and Brenac, V. Surface-enhanced laser desorption-ionization retentate chromatography™ mass spectrometry (SELDI-RC-MS): A new method for rapid development of process chromatography conditions. Journal of Chromatography B: Analytical Technologies in the Biomedical and Life Sciences 782, 307-316, 2002.
233. Weinberger, S.R., Morris, T.S. and Pawlak, M. Recent trends in protein biochip technology. Pharmacogenomics 1, 395-416, 2000.
234. Petricoin III, E.F., Ardekani, A.M., Hitt, B.A., Levine, P.J., Fusaro, V.A., Steinberg, S.M., Mills, G.B., Simone, C., Fishman, D.A., Kohn, E.C. and Liotta, L.A. Use of proteomic patterns in serum to identify ovarian cancer. Lancet 359, 572-577, 2002.
235. Clarke, C.H., Buckley, J.A. and Fung, E.T. SELDI-TOF-MS proteomics of breast cancer. Clinical chemistry and laboratory medicine : CCLM / FESCC 43, 1314-1320, 2005.
236. Zhu, Z., Boobis, A.R. and Edwards, R.J. Identification of estrogen-responsive proteins in MCF-7 human breast cancer cells using label-free quantitative proteomics. Proteomics 8, 1987-2005, 2008.
237. Mosley, K., Tam, F.W.K., Edwards, R.J., Crozier, J., Pusey, C.D. and Lightstone, L. Urinary proteomic profiles distinguish between active and inactive lupus nephritis. Rheumatology 45, 1497-1504, 2006.
238. Le Bihan, M.C., Hou, Y., Harris, N., Tarelli, E. and Coulton, G.R. Proteomic analysis of fast and slow muscles from normal and kyphoscoliotic mice using protein arrays, 2-DE and MS. Proteomics 6, 4646-4661, 2006.
239. Issaq, H.J., Veenstra, T.D., Conrads, T.P. and Felschow, D. The SELDI-TOF MS approach to proteomics: Protein profiling and biomarker identification. Biochemical and Biophysical Research Communications 292, 587-592, 2002.
240. Gygi, S.P., Rist, B., Gerber, S.A., Turecek, F., Gelb, M.H. and Aebersold, R. Quantitative analysis of complex protein mixtures using isotope-coded affinity tags. Nature Biotechnology 17, 994-999, 1999.
241. Haynes, P.A. and Yates III, J.R. Proteome profiling-pitfalls and progress. Comparative and Functional Genomics 17, 81-87, 2000.
242. Unwin, R.D., Pierce, A., Watson, R.B., Sternberg, D.W. and Whetton, A.D. Quantitative proteomic analysis using isobaric protein tags enables rapid comparison of changes in transcript and protein levels in transformed cells. Molecular and Cellular Proteomics 4, 924-935, 2005.
249
243. Chong, P.K., Gan, C.S., Pham, T.K. and Wright, P.C. Isobaric tags for relative and absolute quantitation (iTRAQ) reproducibility: Implication of multiple injections. Journal of Proteome Research 5, 1232-1240, 2006.
244. Chee, S.G., Poh, K.C., Trong, K.P. and Wright, P.C. Technical, experimental, and biological variations in isobaric tags for relative and absolute quantitation (iTRAQ). Journal of Proteome Research 6, 821-827, 2007.
245. Wiese, S., Reidegeld, K.A., Meyer, H.E. and Warscheid, B. Protein labeling by iTRAQ: A new tool for quantitative mass spectrometry in proteome research. Proteomics 7, 340-350, 2007.
246. Ong, S.E., Blagoev, B., Kratchmarova, I., Kristensen, D.B., Steen, H., Pandey, A. and Mann, M. Stable isotope labeling by amino acids in cell culture, SILAC, as a simple and accurate approach to expression proteomics. Molecular & Cellular Proteomics 1, 376-386, 2002.
247. Blagoev, B., Kratchmarova, I., Ong, S.E., Nielsen, M., Foster, L.J. and Mann, M. A proteomics strategy to elucidate functional protein-protein interactions applied to EGF signaling. Nature Biotechnology 21, 315-318, 2003.
248. Ong, S.E., Foster, L.J. and Mann, M. Mass spectrometric-based approaches in quantitative proteomics. Methods 29, 124-130, 2003.
249. Chaerkady, R. and Pandey, A. Quantitative proteomics for identification of cancer biomarkers. Proteomics - Clinical Applications 1, 1080-1089, 2007.
250. Sun, S., Lee, N.P.Y., Poon, R.T.P., Fan, S.T., He, Q.Y., Lau, G.K. and Luk, J.M. Oncoproteomics of hepatocellular carcinoma: From cancer markers' discovery to functional pathways. Liver International 27, 1021-1038, 2007.
251. Moseley, M.A. Current trends in differential expression proteomics: isotopically coded tags. Trends in Biotechnology 19, 2001.
252. Kislinger, T. and Emili, A. Going global: Protein expression profiling using shotgun mass spectrometry. Current Opinion in Molecular Therapeutics 5, 285-293, 2003.
253. Nicholas, B., Skipp, P., Mould, R., Rennard, S., Davies, D.E., O'Connor, C.D. and Djukanovic, R. Shotgun proteomic analysis of human-induced sputum. Proteomics 6, 4390-4401, 2006.
254. Old, W.M., Meyer-Arendt, K., Aveline-Wolf, L., Pierce, K.G., Mendoza, A., Sevinsky, J.R., Resing, K.A. and Ahn, N.G. Comparison of label-free methods for quantifying human proteins by shotgun proteomics. Molecular and Cellular Proteomics 4, 1487-1502, 2005.
250
255. Purvine, S., Eppel, J.T., Yi, E.C. and Goodlett, D.R. Shotgun collision-induced dissociation of peptides using a time of flight mass analyzer. Proteomics 3, 847-850, 2003.
256. Kislinger, T. and Emili, A. Multidimensional protein identification technology: Current status and future prospects. Expert Review of Proteomics 2, 27-39, 2005.
257. Delahunty, C.M. and Yates III, J.R. MudPIT: Multidimensional protein identification technology. BioTechniques 43, 563-569, 2007.
258. Paoletti, A.C., Zybailov, B. and Washburn, M.P. Principles and applications of multidimensional protein identification technology. Expert Review of Proteomics 1, 275-282, 2004.
259. McDonald, W.H. and Yates III, J.R. Shotgun proteomics and biomarker discovery. Disease Markers 18, 99-105, 2002.
260. Wolters, D.A., Washburn, M.P. and Yates III, J.R. An automated multidimensional protein identification technology for shotgun proteomics. Analytical Chemistry 73, 5683-5690, 2001.
261. Washburn, M.P., Wolters, D. and Yates, J.R. Large-scale analysis of the yeast proteome by multidimensional protein identification technology. Nature Biotechnology 19, 242-247, 2001.
262. Mauri, P. and Scigelova, M. Multidimensional protein identification technology for clinical proteomic analysis. Clinical Chemistry and Laboratory Medicine 47, 636-645, 2009.
263. Kristjansdottir, K., Wolfgeher, D., Lucius, N., Angulo, D.S. and Kron, S.J. Phosphoprotein profiling by PA-GeLC-MS/MS. Journal of Proteome Research 7, 2812-2824, 2008.
264. Graham, R.L.J., Sharma, M.K., Ternan, N.G., Weatherly, D.B., Tarleton, R.L. and McMullan, G. A semi-quantitative GeLC-MS analysis of temporal proteome expression in the emerging nosocomial pathogen Ochrobactrum anthropi. Genome Biology 8, 2007.
265. Pan, J., Chen, H.Q., Sun, Y.H., Zhang, J.H. and Luo, X.Y. Comparative proteomic analysis of non-small-cell lung cancer and normal controls using serum label-free quantitative shotgun technology. Lung 186, 255-261, 2008.
266. Fitzpatrick, D.P.G., You, J.S., Bernis, K.J., Wery, J.P., Ludwig, J.R. and Wang, M. Searching for potential biomarkers of cisplatin resistance in human ovarian cancer using a label-free LC/MS-based protein quantification method. Proteomics - Clinical Applications 1, 246-263, 2007.
251
267. Issaq, H.J. and Veenstra, T.D. Two-dimensional polyacrylamide gel electrophoresis (2D-PAGE): Advances and perspectives. BioTechniques 44, 697-700, 2008.
268. Merril, C.R. Gel-staining techniques. Methods in Enzymology 182, 477-488, 1990.
269. Smith, P.K., Krohn, R.I. and Hermanson, G.T. Measurement of protein using bicinchoninic acid. Analytical Biochemistry 150, 76-85, 1985.
270. Mosmann, T. Rapid colorimetric assay for cellular growth and survival: Application to proliferation and cytotoxicity assays. Journal of Immunological Methods 65, 55-63, 1983.
271. Chowdhury, F., Williams, A. and Johnson, P. Validation and comparison of two multiplex technologies, Luminex® and Mesoscale Discovery, for human cytokine profiling. Journal of Immunological Methods 340, 55-64, 2009.
272. Dieterle, F. and Marrer, E. New technologies around biomarkers and their interplay with drug development. Analytical and Bioanalytical Chemistry 390, 141-154, 2008.
273. Bergmeyer, H.U. and Horder, M. IFCC methods for the measurement of catalytic concentrations of enzymes. III. IFCC method for alanine aminotransferase (L-alanine: 2-oxoglutarate aminotransferase, EC 2.6.1.2). Clinica Chimica Acta 105, 147F-154F , 1980.
274. Tietz, N. Clinical guide to laboratory tests (4th ed.), WB Saunders Company, London, 20-21, 1995.
275. Bergmeyer, H.U. and Horder, M. IFCC methods for the measurement of catalytic concentration of enzymes. III. IFCC method for alanine aminotransferase. Journal of Clinical Chemistry and Clinical Biochemistry 18, 521-534, 1980.
276. Tietz, N. Clinical guide to laboratory tests (4th ed.), WB Saunders Company, London, 76-77, 1995.
277. Szasz, G., Gruber, W. and Bernt, E. Creatine kinase in serum: I. Determination of optimum reaction conditions. Clinical Chemistry 22, 650-656, 1976.
278. Tietz, N. Clinical guide to laboratory tests (4th ed.), WB Saunders Company, London, 180-181, 1995.
279. Tietz, N. Clinical guide to laboratory tests (4th ed.), WB Saunders Company, London, 384-385, 1995.
252
280. Harjanne, A. The kinetic measurement of serum aldolase. Clinica Chimica Acta 92, 311-313, 1979.
281. Feissli, S., Forster, G., Laudahn, G., Schmidt, E. and Schmidt, F.W. Normal-Werte und Alterung von Hauptketten-Enzymen im Serum. Klinische Wochenschrift 44, 390-396, 1966.
282. Pruitt, K.D., Tatusova, T. and Maglott, D.R. NCBI reference sequences (RefSeq): A curated non-redundant sequence database of genomes, transcripts and proteins. Nucleic Acids Research 35, D61-D65, 2007.
283. Eng, J.K., McCormack, A.L. and Yates III, J.R. An approach to correlate tandem mass spectral data of peptides with amino acid sequences in a protein database. Journal of the American Society for Mass Spectrometry 5, 976-989, 1994.
284. Anderson, D.C., Li, W., Payan, D.G. and Noble, W.S. A new algorithm for the evaluation of shotgun peptide sequencing in proteomics: Support vector machine classification of peptide MS/MS spectra and SEQUEST scores. Journal of Proteome Research 2, 137-146, 2003.
285. MacCoss, M.J., Wu, C.C. and Yates III, J.R. Probability based validation of protein identifications using a modified SEQUEST algorithm. Analytical Chemistry 74, 5593-5599, 2002.
286. Sorichter, S., Mair, J., Koller, A., Gebert, W., Rama, D., Calzolari, C., Artner-Dworzak, E. and Puschendorf, B. Skeletal troponin I as a marker of exercise-induced muscle damage. Journal of Applied Physiology 83, 1076-1082, 1997.
287. Amat, A.M., Corrales, J.A.M., Serrano, F.R., Boulaiz, H., Salazar, J.C.P., Contreras, F.H., Perez, O.C., Delgado, E.C., Martín, I. and Jimenez, A.A. Role of α-actin in muscle damage of injured athletes in comparison with traditional markers. British Journal of Sports Medicine 41, 442-446, 2007.
288. Busch, K.L. Chemical noise in mass spectrometry: Part I. Spectroscopy 17, 32-37, 2002.
289. Krutchinsky, A.N. and Chait, B.T. On the nature of the chemical noise in MALDI mass spectra. Journal of the American Society for Mass Spectrometry 13, 129-134, 2002.
290. Shin, H., Mutlu, M., Koomen, J.M. and Markey, M.K. Parametric power spectral density analysis of noise from instrumentation in MALDI TOF mass spectrometry. Cancer Informatics 3, 219-230, 2007.
291. Han, S.H. and Kim, J. A comparison of screening methods: Selecting important design variables for modeling product usability. International Journal of Industrial Ergonomics 32, 189-198, 2003.
253
292. Kettaneh, N. and Wold, S. Analyzing complicated data sets by PCA and PLS. Annual Quality Congress Transactions, 70-72, 1998.
293. Hochberg, Y. and Benjamini, Y. More powerful procedures for multiple significance testing. Statistics in Medicine 9, 811-818, 1990.
294. Curtin, F. and Schulz, P. Multiple correlations and Bonferroni's correction. Biological Psychiatry 44, 775-777, 1998.
295. Benjamini, Y. and Hochberg, Y. On the adaptive control of the false discovery rate in multiple testing with independent statistics. Journal of Educational and Behavioral Statistics 25, 60-83, 2000.
296. Baggerly, K.A., Morris, J.S. and Coombes, K.R. Reproducibility of SELDI-TOF protein patterns in serum: Comparing datasets from different experiments. Bioinformatics 20, 777-785, 2004.
297. Poon, T.C.W. Opportunities and limitations of SELDI-TOF-MS in biomedical research: Practical advices. Expert Review of Proteomics 4, 51-65, 2007.
298. Semmes, O.J., Feng, Z., Adam, B.L., Banez, L.L., Bigbee, W.L., Campos, D., Cazares, L.H., Chan, D.W., Grizzle, W.E., Izbicka, E., Kagan, J., Malik, G., McLerran, D., Moul, J.W., Partin, A., Prasanna, P., Rosenzweig, J., Sokoll, L.J., Srivastava, S., Srivastava, S., Thompson, I., Welsh, M.J., White, N., Winget, M., Yasui, Y., Zhang, Z. and Zhu, L. Evaluation of serum protein profiling by surface-enhanced laser desorption/ionization time-of-flight mass spectrometry for the detection of prostate cancer: I. Assessment of platform reproducibility. Clinical Chemistry 51, 102-112, 2005.
299. Hocking, R.R. The analysis and selection of variables in linear regression. Biometrics 32, 1-49, 1976.
300. Cohen, J. Statistical power analysis for the behavioral sciences (2nd ed.), Lawrence Erlbaum, New Jersey, 109-139, 1988.
301. Cohen, J. A power primer. Psychological Bulletin 112, 155-159, 1992.
302. Spriet, L.L. and Watt, M.J. Regulatory mechanisms in the interaction between carbohydrate and lipid oxidation during exercise. Acta Physiologica Scandinavica 178, 443-452, 2003.
303. Kiens, B. and Roepstorff, C. Utilization of long-chain fatty acids in human skeletal muscle during exercise. Acta Physiologica Scandinavica 178, 391-396, 2003.
304. Saltin, B., Astrand, P.O., Crawford, M.A. and Saltin, B. Free fatty acids and exercise. American Journal of Clinical Nutrition 57, 752S-758S, 1993.
254
305. Rasmussen, B.B. and Wolfe, R.R. Regulation of fatty acid oxidation in skeletal muscle. Annual Review of Nutrition 19, 463-484, 1999.
306. Jeukendrup, A.E., Saris, W.H.M. and Wagenmakers, A.J.M. Fat metabolism during exercise: A review. Part I: Fatty acid mobilization and muscle metabolism. International Journal of Sports Medicine 19, 231-244, 1998.
307. Abood, E.A. and Jones, M.M. Macrophages in developing mammalian skeletal muscle: Evidence for muscle fibre death as a normal developmental event. Acta Anatomica 140, 201-212, 1991.
308. Phillips, P.S., Haas, R.H., Bannykh, S., Hathaway, S., Gray, N.L., Kimura, B.J., Vladutiu, G.D. and England, J.D.F. Statin-associated myopathy with normal creatine kinase levels. Annals of Internal Medicine 137, 581-585, 2002.
309. Thompson, P.D., Clarkson, P. and Karas, R.H. Statin-Associated Myopathy. Journal of the American Medical Association 289, 1681-1690, 2003.
310. Fenwick, N.P. and Fraser, D. The three Rs in the pharmaceutical industry: Perspectives of scientists and regulators. Animal Welfare 14, 367-377, 2005.
311. Fentem, J. and Balls, M. In vitro alternatives to toxicity testing in animals. Chemistry and Industry 6, 207-211, 1992.
312. Kramer, J.A., Sagartz, J.E. and Morris, D.L. The application of discovery toxicology and pathology towards the design of safer pharmaceutical lead candidates. Nature Reviews Drug Discovery 6, 636-649, 2007.
313. Kramer, J.A., Sagartz, J.E. and Morris, D.L. The application of discovery toxicology and pathology towards the design of safer pharmaceutical lead candidates. Nature Reviews Drug Discovery 6, 636-649, 2007.
314. Niles, A.L., Moravec, R.A. and Riss, T.L. Update on in vitro cytotoxicity assays for drug development. Expert Opinion on Drug Discovery 3, 655-669, 2008.
315. Xia, M., Huang, R., Witt, K.L., Southall, N., Fostel, J., Cho, M.H., Jadhav, A., Smith, C.S., Inglese, J., Portier, C.J., Tice, R.R. and Austin, C.P. Compound cytotoxicity profiling using quantitative high-throughput screening. Environmental Health Perspectives 116, 284-291, 2008.
316. Venkatakrishnan, K., von Moltke, L.L., Obach, R.S. and Greenblatt, D.J. Drug metabolism and drug interactions: Application and clinical value of in vitro models. Current Drug Metabolism 4, 423-459, 2003.
255
317. Blair, J.A., Waterfield, C.J., Lightfoot, R.M., Rowlands, A., York, M.J. and Turton, J.A. The skeletal muscle-specific toxicity of 2,3,5,6 tetramethyl-p-phenylenediamine (TMPD): some preliminary studies. Human and Experimental Toxicology 16, 403, 1997.
318. Anderson, J.E. The satellite cell as a companion in skeletal muscle plasticity: Currency, conveyance, clue, connector and colander. Journal of Experimental Biology 209, 2276-2292, 2006.
319. Maier, A., Zhou, Z. and Bornemann, A. The expression profile of myogenic transcription factors in satellite cells from denervated rat muscle. Brain Pathology 12, 170-177, 2002.
320. Qin, Z., Zhu, W., Xu, G., Wu, L. and Zhang, D. Culture of skeletal muscle satellite cells in vitro. Medical Journal of Wuhan University 28, 492-495+528, 2007.
321. Zahradka, P., Larson, D.E. and Sells, B.H. Regulation of ribosome biogenesis in differentiated rat myotubes. Molecular and Cellular Biochemistry 104, 189-194, 1991.
322. Portiér, G.L., Benders, A.A.G.M., Oosterhof, A., Veerkamp, J.H. and Van Kuppevelt, T.H. Differentiation markers of mouse C2C12 and rat L6 myogenic cell lines and the effect of the differentiation medium. In Vitro Cellular and Developmental Biology - Animal 35, 219-227, 1999.
323. Van Der Ven, P.F.M., Schaart, G., Jap, P.H.K., Sengers, R.C.A., Stadhouders, A.M. and Ramaekers, F.C.S. Differentiation of human skeletal muscle cells in culture: Maturation as indicated by titin and desmin striation. Cell and Tissue Research 270, 189-198, 1992.
324. Yamasaki, K.I., Hayashi, H., Nishiyama, K., Kobayashi, H., Uto, S., Kondo, H., Hashimoto, S. and Fujisato, T. Control of myotube contraction using electrical pulse stimulation for bio-actuator. Journal of Artificial Organs 12, 131-137, 2009.
325. Kawahara, Y., Yamaoka, K., Iwata, M., Fujimura, M., Kajiume, T., Magaki, T., Takeda, M., Ide, T., Kataoka, K., Asashima, M. and Yuge, L. Novel electrical stimulation sets the cultured myoblast contractile function to 'on'. Pathobiology 73, 288-294, 2007.
326. Lawson, M.A. and Purslow, P.P. Differentiation of myoblasts in serum-free media: Effects of modified media are cell line-specific. Cells Tissues Organs 167, 130-137, 2000.
327. El Abida, K., Duvallet, A., Thieulart, L., Rieu, M. and Beaudry, M. Lactate transport during differentiation of skeletal muscle cells: Evidence for a specific carrier in L6 myotubes. Acta Physiologica Scandinavica 144, 469-471, 1992.
256
328. Benders, A.A.G.M., Van Kuppevelt, T.H.M.S., Oosterhof, A. and Veerkamp, J.H. The biochemical and structural maturation of human skeletal muscle cells in culture: The effect of the serum substitute ultroser G. Experimental Cell Research 195, 284-294, 1991.
329. Crerar, M.M., Leather, R., David, E. and Pearson, M.L. Myogenic differentiation of L6 rat myoblasts: Evidence for pleiotropic effects on myogenesis by RNA polymerase II mutations to α-amanitin resistance. Molecular and Cellular Biology 3, 946-955, 1983.
330. Mandel, J.L. and Pearson, M.L. Insulin stimulates myogenesis in a rat myoblast line. Nature 251, 618-620, 1974.
331. Dym, H. and Yaffe, D. Expression of creatine kinase isoenzymes in myogenic cell lines. Developmental Biology 68, 592-599, 1979.
332. Casey, W.M., Brodie, T., Yoon, L., Ni, H., Jordan, H.L. and Cariello, N.F. Correlation analysis of gene expression and clinical chemistry to identify biomarkers of skeletal myopathy in mice treated with PPAR agonist GW610742X. Biomarkers 13, 364-376, 2008.
333. Florentin, M., Liberopoulos, E.N., Mikhailidis, D.P. and Elisaf, M.S. Fibrate-associated adverse effects beyond muscle and liver toxicity. Current Pharmaceutical Design 14, 574-587, 2008.
334. Fredenrich, A. and Grimaldi, P.A. PPAR delta: An uncompletely known nuclear receptor. Diabetes and Metabolism 31, 23-27, 2005.
335. Johnson, T.E., Zhang, X., Shi, S. and Umbenhauer, D.R. Statins and PPARα agonists induce myotoxicity in differentiated rat skeletal muscle cultures but do not exhibit synergy with co-treatment. Toxicology and Applied Pharmacology 208, 210-221, 2005.
336. Rubenstrunk, A., Hanf, R., Hum, D.W., Fruchart, J.C. and Staels, B. Safety issues and prospects for future generations of PPAR modulators. Biochimica et Biophysica Acta - Molecular and Cell Biology of Lipids 1771, 1065-1081, 2007.
337. Barile, F.A., Arjun, S. and Hopkinson, D. In vitro cytotoxicity testing: Biological and statistical significance. Toxicology in Vitro 7, 111-116, 1993.
338. Riddell, R.J., Clothier, R.H. and Balls, M. An evaluation of three in vitro cytotoxicity assays. Food and Chemical Toxicology 24, 469-471, 1986.
339. Borisov, A.B., Dedkov, E.I. and Carlson, B.M. Interrelations of myogenic response, progressive atrophy of muscle fibers, and cell death in denervated skeletal muscle. Anatomical Record 264, 203-218, 2001.
257
340. Seale, P., Ishibashi, J., Scimè, A. and Rudnicki, M.A. Pax7 is necessary and sufficient for the myogenic specification of CD45+:Sca1+ stem cells from injured muscle. PLoS Biology 2, 0664-0672, 2004.
341. Itagaki, M., Takaguri, A., Kano, S., Kaneta, S., Ichihara, K. and Satoh, K. Possible mechanisms underlying statin-induced skeletal muscle toxicity in L6 fibroblasts and in rats. Journal of Pharmacological Sciences 109, 94-101, 2009.
342. Seachrist, J.L., Loi, C.M., Evans, M.G., Criswell, K.A. and Rothwell, C.E. Roles of exercise and pharmacokinetics in cerivastatin-induced skeletal muscle toxicity. Toxicological Sciences 88, 551-561, 2005.
343. Reijneveld, J.C., Koot, R.W., Bredman, J.J., Joles, J.A. and Bär, P.R. Differential effects of 3-hydroxy-3-methylglutaryl-coenzyme A reductase inhibitors on the development of myopathy in young rats. Pediatric Research 39, 1028-1035, 1996.
344. Pritt, M.I., Hall, D.G., Recknor, J., Credille, K.M., Brown, D.D., Yumibe, N.P., Schultze, A.E. and Watson, D.E. Fabp3 as a biomarker of skeletal muscle toxicity in the rat: Comparison with conventional biomarkers. Toxicological Sciences 103, 382-396, 2008.
345. Guerrero, M., Guiu-Comadevall, M., Cadefau, J.A., Parra, J., Balius, R., Estruch, A., Rodas, G., Bedini, J.L. and Cussó, R. Fast and slow myosins as markers of muscle injury. British Journal of Sports Medicine 42, 581-584, 2008.
346. Simpson, J.A., Labugger, R., Collier, C., Brison, R.J., Iscoe, S. and Van Eyk, J.E. Fast and slow skeletal troponin I in serum from patients with various skeletal muscle disorders: A pilot study. Clinical Chemistry 51, 966-972, 2005.
347. Zhen, E.Y., Berna, M.J., Jin, Z., Pritt, M.L., Watson, D.E., Ackermann, B.L. and Hale, J.E. Quantification of heart fatty acid binding protein as a biomarker for drug-induced cardiac and musculoskeletal necroses. Proteomics - Clinical Applications 1, 661-671, 2007.
348. Procajlo, A., Zbanyszek, M., Sobiech, P., Stopyra, A. and Rajski, K. Troponin - A new marker in the diagnostics of muscle diseases in animals. Polish Journal of Veterinary Sciences 6, 297-299, 2003.
349. Fung, E.T., Wright, G.L. and Dalmasso, E.A. Proteomic strategies for biomarker identification: Progress and challenges. Current Opinion in Molecular Therapeutics 2, 643-650, 2000.
350. Rubin, R.B. and Merchant, M. A rapid protein profiling system that speeds study of cancer and other diseases. American clinical laboratory 19, 28-29, 2000.
258
351. Whelan, L.C., Power, K.A.R., McDowell, D.T., Kennedy, J. and Gallagher, W.M. Applications of SELDI-MS technology in oncology. Journal of Cellular and Molecular Medicine 12, 1535-1547, 2008.
352. Aebersold, R. and Mann, M. Mass spectrometry-based proteomics. Nature 422, 198-207, 2003.
353. Hardin, E.D. and Vestal, M.L. Laser ionization mass spectrometry of nonvolatile samples. Analytical Chemistry 53, 1492-1497, 1981.
354. Agranoff, D., Stich, A., Abel, P. and Krishna, S. Proteomic fingerprinting for the diagnosis of human African trypanosomiasis. Trends in Parasitology 21, 154-157, 2005.
355. Wang, L., Zheng, W., Mu, L. and Zhang, S.Z. Identifying biomarkers of endometriosis using serum protein fingerprinting and artificial neural networks. International Journal of Gynecology and Obstetrics 101, 253-258, 2008.
356. Ahmed, F.E. Application of MALDI/SELDI mass spectrometry to cancer biomarker discovery and validation. Current Proteomics 5, 224-252, 2008.
357. Ahmed, F.E. Application of MALDI/SELDI mass spectrometry to cancer biomarker discovery and validation. Current Proteomics 5, 224-252, 2008.
358. Kiehntopf, M., Siegmund, R. and Deufel, T. Use of SELDI-TOF mass spectrometry for identification of new biomarkers: Potential and limitations. Clinical Chemistry and Laboratory Medicine 45, 1435-1449, 2007.
359. Fang, M., Boobis, A.R. and Edwards, R.J. Searching for novel biomarkers of centrally and peripehrally-acting neurotoxicants, using surface-enhanced laser desorption/ionisation-time-of-flight mass spectrometry (SELDI-TOF MS). Food and Chemical Toxicology 45, 2126-2137, 2007.
360. Zhu, Y., Valdes, J. and Jortani, S.A. Application of bioaffinity mass spectrometry for analysis of ligands. Therapeutic Drug Monitoring 27, 694-699, 2005.
361. Brenac, V., Ravault, V., Santambien, P. and Boschetti, E. Capture of a monoclonal antibody and prediction of separation conditions using a synthetic multimodal ligand attached on chips and beads. Journal of Chromatography B: Analytical Technologies in the Biomedical and Life Sciences 818, 61-66, 2005.
259
362. Chapman, K. The ProteinChip Biomarker System from Ciphergen Biosystems: a novel proteomics platform for rapid biomarker discovery and validation. Biochemical Society Transactions 30, 82-87, 2002.
363. Austen, B.M., Frears, E.R. and Davies, H. The use of Seldi proteinchip™ arrays to monitor production of alzheimer's β-amyloid in transfected cells. Journal of Peptide Science 6, 459-469, 2000.
364. Davies, H., Lomas, L., Austen, B. and Holmes, G. Profiling of amyloid β peptide variants using SELDI ProteinChip® arrays. BioTechniques 27, 1258-1261, 1999.
365. Sihlbom, C., Kanmert, I., Von Bahr, H. and Davidsson, P. Evaluation of the combination of bead technology with SELDI-TOF-MS and 2-D DIGE for detection of plasma proteins. Journal of Proteome Research 7, 4191-4198, 2008.
366. Kwapiszewska, G., Meyer, M., Bogumil, R., Bohle, R.M., Seeger, W., Weissmann, N. and Fink, L. Identification of proteins in laser-microdissected small cell numbers by SELDI-TOF and Tandem MS. BMC Biotechnology 4, 1-10, 2004.
367. Peng, J., Stanley, A.J., Cairns, D., Selby, P.J. and Banks, R.E. Using the protein chip interface with quadrupole time-of-flight mass spectrometry to directly identify peaks in SELDI profiles - Initial evaluation using low molecular weight serum peaks. Proteomics 9, 492-498, 2009.
368. Borisov, A.B. and Carlson, B.M. Cell death in denervated skeletal muscle is distinct from classical apoptosis. Anatomical Record 258, 305-318, 2000.
369. Takai, Y., Kaibuchi, K., Sasaki, T., Tanaka, K., Shirataki, H. and Nakanishi, H. Rho small G protein and cytoskeletal control. Princess Takamatsu symposia 24, 338-350, 1994.
370. Nobes, C. and Hall, A. Regulation and function of the Rho subfamily of small GTPases. Current Opinion in Genetics and Development 4, 77-81, 1994.
371. Hall, A. Rho GTPases and the actin cytoskeleton. Science 279, 509-514, 1998.
372. DerMardirossian, C. and Bokoch, G.M. GDIs: Central regulatory molecules in Rho GTPase activation. Trends in Cell Biology 15, 356-363, 2005.
373. Dovas, A. and Couchman, J.R. Rho GDI: Multiple functions in the regulation of Rho family GTPase activities. Biochemical Journal 390, 1-9, 2005.
260
374. Olofsson, B. Rho guanine dissociation inhibitors: Pivotal molecules in cellular signalling. Cellular Signalling 11, 545-554, 1999.
375. El Marzouk, S., Schultz-Norton, J.R., Likhite, V.S., McLeod, I.X., Yates, J.R. and Nardulli, A.M. Rho GDP dissociation inhibitor α interacts with estrogen receptor α and influences estrogen responsiveness. Journal of Molecular Endocrinology 39, 249-259, 2007.
376. Tang, D.D. and Anfinogenova, Y. Physiologic properties and regulation of the actin cytoskeleton in vascular smooth muscle. Journal of Cardiovascular Pharmacology and Therapeutics 13, 130-140, 2008.
377. Gerthoffer, W.T. Mechanisms of vascular smooth muscle cell migration. Circulation Research 100, 607-621, 2007.
378. Murphy, G.A., Solski, P.A., Jillian, S.A., Pérez de la Ossa, P., D'Eustachio, P., Der, C.J. and Rush, M.G. Cellular functions of TC10, a Rho family GTPase: Regulation of morphology, signal transduction and cell growth. Oncogene 18, 3831-3845, 1999.
379. Takano, H., Komuro, I., Oka, T., Shiojima, I., Hiroi, Y., Mizuno, T. and Yazaki, Y. The Rho family G proteins play a critical role in muscle differentiation. Molecular and Cellular Biology 18, 1580-1589, 1998.
380. Zhang, Q., Skepper, J.N., Yang, F., Davies, J.D., Hegyi, L., Roberts, R.G., Weissberg, P.L., Ellis, J.A. and Shanahan, C.M. Nesprins: A novel family of spectrin-repeat-containing proteins that localize to the nuclear membrane in multiple tissues. Journal of Cell Science 114, 4485-4498, 2001.
381. Qiu, H., Zhao, S., Yu, M., Fan, B. and Liu, B. Expression patterns and subcellular localization of porcine (Sus Scrofa) lectin, galactose-binding, soluble 1 gene. Acta Biochimica et Biophysica Sinica 40, 85-90, 2008.
382. Chan, J., O'Donoghue, K., Gavina, M., Torrente, Y., Kennea, N., Mehmet, H., Stewart, H., Watt, D.J., Morgan, J.E. and Fisk, N.M. Galectin-1 induces skeletal muscle differentiation in human fetal mesenchymal stem cells and increases muscle regeneration. Stem Cells 24, 1879-1891, 2006.
383. Kami, K. and Senba, E. Galectin-1 is a novel factor that regulates myotube growth in regenerating skeletal muscles. Current Drug Targets 6, 395-405, 2005.
384. Watt, D.J., Jones, G.E. and Goldring, K. The involvement of galectin-1 in skeletal muscle determination, differentiation and regeneration. Glycoconjugate Journal 19, 615-619, 2002.
261
385. Martinez, V.G., Pelizzari, E.H., Díaz, E.S., Cigorraga, S.B., Lustig, L., Denduchis, B., Wolfenstein-Todel, C. and Iglesias, M.M. Galectic-1, a cell adhesion modulator, induces apoptosis of rat Leydig cells in vitro. Glycobiology 14, 127-137, 2004.
386. Liu, B., Xu, X.C., Cheng, Y., Huang, J., Liu, Y.H., Liu, Z., Min, M.W., Bian, H.J., Chen, J. and Bao, J.K. Apoptosis-inducing effect and structural basis of Polygonatum cyrtonema lectin and chemical modification properties on its mannose-binding sites. Journal of Biochemistry and Molecular Biology 41, 369-375, 2008.
387. Galban, V.D., Evangelista, E.A., Migliorini, R.H. and Kettelhut, I.C. Role of ubiquitin-proteasome-dependent proteolytic process in degradation of muscle protein from diabetic rabbits. Molecular and Cellular Biochemistry 225, 35-41, 2001.
388. Fang, C.H., Wang, J.J., Hobler, S., Li, B.G., Fischer, J.E. and Hasselgren, P.O. Proteasome blockers inhibit protein breakdown in skeletal muscle after burn injury in rats. Clinical Science 95, 225-233, 1998.
389. Taillandier, D., Combaret, L., Pouch, M.N., Samuels, S.E., Béchet, D. and Attaix, D. The role of ubiquitin-proteasome-dependent proteolysis in the remodelling of skeletal muscle. Proceedings of the Nutrition Society 63, 357-361, 2004.
390. Wang, Y., Szczesna-Cordary, D., Craig, R., Diaz-Perez, Z., Guzman, G., Miller, T. and Potter, J.D. Fast skeletal muscle regulatory light chain is required for fast and slow skeletal muscle development. FASEB Journal 21, 2205-2214, 2007.
391. Zhi, G., Ryder, J.W., Huang, J., Ding, P., Chen, Y., Zhao, Y., Kamm, K.E. and Stull, J.T. Myosin light chain kinase and myosin phosphorylation effect frequency-dependent potentiation of skeletal muscle contraction. Proceedings of the National Academy of Sciences of the United States of America 102, 17519-17524, 2005.
392. Bicer, S. and Reiser, P.J. Myosin light chain isoform expression among single mammalian skeletal muscle fibers: Species variations. Journal of Muscle Research and Cell Motility 25, 623-633, 2004.
