Hindawi Publishing CorporationISRN ObesityVolume 2013 Article ID 161345 9 pageshttpdxdoiorg1011552013161345

Research ArticleDysregulated Alternative Splicing Pattern of PKC120575 duringDifferentiation of Human Preadipocytes Represents DistinctDifferences between Lean and Obese Adipocytes

Gay Carter1 Andreacute Apostolatos2 Rekha Patel2 Abhishek Mathur1 Denise Cooper12

Michel Murr13 and Niketa A Patel12

1 James A Haley Veteransrsquo Hospital Research Service VAR 151 13000 Bruce B Downs Boulevard Tampa FL 33612 USA2Department of Molecular Medicine Morsani College of Medicine University of South Florida Tampa FL 33612 USA3Department of Surgery Morsani College of Medicine University of South Florida Tampa FL 33612 USA

Correspondence should be addressed to Niketa A Patel npatelhealthusfedu

Received 15 January 2013 Accepted 5 March 2013

Academic Editors H Gordish-Dressman U J Magalang and S Straube

Copyright copy 2013 Gay Carter et al This is an open access article distributed under the Creative Commons Attribution Licensewhich permits unrestricted use distribution and reproduction in any medium provided the original work is properly cited

Obesity and its comorbidities affect millions of people Here we demonstrate that human preadipocytes are susceptible toprogrammed cell death (apoptosis) while mature adipocytes are resistant to apoptosis The molecular mechanisms underlying thephenotype of apoptosis-resistant adipocytes are lesser known To study the role of apoptosis and definemolecular differences in thedevelopmental process of adipogenesis human preadipocytes were differentiated in vitro to mature adipocytes Many genes in theapoptosis pathway are alternatively spliced Our data demonstrates that during differentiation PKC120575 Bclx and Caspase9 switchto their prosurvival splice variants along with an increase in Bcl2 expression when the cells terminally differentiate into matureadipocytes Next we determined the expression pattern of these genes in obesity Our data indicated high expression of PKC120575VIIIin adipose tissue of obese patient in different depots We demonstrate a shift in the in vitro expression of these splice variants indifferentiating preadipocytes derived from obese patients along with a decrease in adipogenesis markers Hence the programmedsplicing of antiapoptotic proteins is a pivotal switch in differentiation that commits adipocytes to a prosurvival pathway Theexpression pattern of these genes is dysregulated in obesity and may contribute to adipose tissue dysfunction

1 Introduction

The human proteome is changing constantly in responseto hormones age and developmental stage or disease Thegenetic makeup of the body indicates about 25000 genesresponsible for close to 100000 proteins in a given proteomeAlternative splicing is a quintessentialmechanism to generateproteins with distinct functions from the same gene Alter-native splicing occurs in more than 90 of genes and is apowerful step in gene expression to diversify the genomicrepertoire

Genetic environmental and cultural factors contributeto the onset of obesity In order to develop a therapeuticagent to combat obesity it is essential to understand the

molecularmechanisms underlying adipogenesis Differentia-tion of preadipocytes to mature adipocytes is usually studiedin 3T3L1 and 3T3F442A murine preadipocyte cell lines asthey reproduce adipogenesis in vitro including expressionof adipogenic genes and morphological changes Howeverbeyond the obvious species differences preadipocytes frommouse and humans show differences as shown by gene-centric analysis of adipogenesis marker genes such as PPAR120574and CEBP120572 120573 and 120575 [1] It is also known that unlike murineadipocytes human preadipocytes do not require clonalexpansion to differentiate into adipocytes in vitro Hencewe sought to elucidate the gene expression patterns duringhuman adipogenesis which dictate the mature adipocytephenotype Previous studies on adipogenesis have focused

2 ISRN Obesity

on transcriptional control in adipocyte differentiation Addi-tional mechanisms of gene expression which are required tomaintain the adipocyte phenotype are unknown

Cell death is achieved by several mechanisms includingnecrosis (sudden cell death fatally affecting neighboringcells) apoptosis (programmed cell death) and autophagy(large scale degradation via lysosomes) It is now establishedthat loss of adipose tissue mass in rodents and humansis achieved by inducing apoptosis [2ndash6] Caspases directlyand indirectly orchestrate the morphologic changes of thecell during apoptosis Caspases have been characterized intoupstream initiators (caspase 2 8 9 10) and downstreamexecutioners (caspase 3 6 7) [7] Bcl-2 protein family hasabout 17 members and the decision of a cell to undergoapoptosis is governed by a complex network of interac-tions between these Bcl-2members caspases SmacDIABLO(second mitochondria-derived activator of caspase) andX-linked inhibitor of apoptosis (XIAP) The prosurvivalsubfamily comprises of Bcl-2 Bcl-xL Bcl-w Mcl-1 A1and human Bcl-B The release of SmacDIABLO from themitochondria is regulated by antiapoptotic Bcl-2 proteinsmdashBcl-2 and Bcl-xL Hence the decision to undergo apoptosis isa cascade of events and proteins that work in concert PKC120575is a signaling kinase affecting downstream apoptotic path-ways Our laboratory has demonstrated that the alternativelyspliced products of human PKC120575 have distinct functionsin apoptosis PKC120575I promotes apoptosis while PKC120575VIII isantiapoptotic [8]

Earlier studies in mouse models indicated thatpreadipocytes can undergo programmed cell death(apoptosis) Here we demonstrate that human preadipocytesare susceptible to apoptosis while mature adipocytesare resistant to apoptosis Many genes in the apoptosispathway are alternatively spliced We sought to determinethe alternative splicing patterns of genes in the apoptosispathway which promote the antiapoptotic nature of themature adipocytes This contributes to the inability of anindividual to maintain weight loss as diet and exercise mayreduce the size of the fat cells but the cells do not die andcan easily gain back the fat mass The excess energy (input gtexpenditure) is stored in the white adipose tissue (WAT)which is also an important endocrine regulator

2 Materials and Methods

21 Human Preadipocytes The cells were purchased as cry-opreserved preadipocytes from ZenBio (Research TrianglePark NC USA) Since these commercial cell lines andtheir information are publicly available they are exemptfrom human subjects research determination from IRB inaccordance with HHS regulation at 45 CFR Part 46 Thelean preadipocytes originated from subcutaneous adiposetissue of a healthy donor of 26 years of age undergoingelective surgery with a body mass index of 213 who wasnot diabetic and not a smoker The obese preadipocytesoriginated from subcutaneous adipose tissue of a 40-year-old donor undergoing elective surgery with a body massindex of 546 who was not diabetic and not a smoker

The preadipocyte cell line (prescreen by ZenBio) is sortedand characterized such that it represents only preadipocytesand is free from stem cells or differentiating adipocytes orcontaminating endothelial cells Both the preadipocytes aretested in culture to differentiate into mature adipocytes andshow accumulation of lipid express aP2 respond to lipolyticagents and secrete adiponectin and leptin At the start ofall experiments cells are grown to confluency such thatall cells are synchronized and then differentiated The cellswere cultured according to the manufacturerrsquos instructionsBriefly cryopreserved preadipocytes were passaged withpreadipocyte medium (PM-1 DMEMHamrsquos F-12 mediumHEPES FBS penicillin streptomycin amphotericin B Zen-Bio) and then plated 40625 cellscm2 with PM-1 Cells werefed every other day with PM-1 until confluent To inducedifferentiation PM-1 medium was replaced with differentia-tionmedium (DM2Zen-Bio) including biotin pantothenatehuman insulin dexamethasone isobutylmethylxanthine anda PPAR120574 agonist (days 0ndash7) After 7 days DM-2 mediumwas removed and cells were incubated for additional 7days with adipocyte medium (AM1 Zen-Bio days 7ndash14)which included PM-1 biotin pantothenate human insulinand dexamethasone By day 14 cells contained large lipiddroplets and were considered mature adipocytes Cells weremaintained at 37∘C in a humidified 5 CO

2atmosphere

22 Adipose Lysates We obtained protein lysates from DrsMurr and Cooper USF and VAMC Tampa The proteinlysates were harvested from adipose tissues in Dr Cooperrsquoslab and originated from subcutaneous and omental adiposedepots from an obese patient The tissues were obtainedunder consent from a deidentified patient age 45 nondi-abetic BMI 56 undergoing Roux-en-y gastroenterostomysurgery The fat collection protocols were approved by theInstitutional Review Board of the University of South FloridaEthics Committee protocol number 108360

23 Quantitative Real-Time RT-PCR Total RNA was iso-lated from differentiated lean and obese preadipocytes usingRNAbee according to the manufacturerrsquos protocol (TelTestInc) 2 120583g RNAwas reverse transcribed usingQiagenrsquos RT kit2 120583L of cDNA was amplified by real-time quantitative PCRusing Syber (SYBR)Greenwith anABIPRISM7900 sequencedetection system (PE Applied Biosystems) to quantify theabsolute levels of the transcripts in the samples 120573-actin wasused as the endogenous control Two sets of primer pairsfor each transcript FOXO1 SIRT1 PPAR120574 and adiponectinwere designed to have an annealing temperature of sim60∘CPrimers for PKC120575VIII are described in our earlier publication[9] These primers were initially tested using cDNA fromhuman preadipocytes in an RT-PCR reaction using Taqpolymerase to give distinct products corresponding to therespective transcripts Next the optimal primer concentra-tion was determined from a range of 50ndash900 nM The finalconcentration of 300 nM was selected to ensure efficiencyand specificity for its target based on the dissociation curvethat showed a single sharp peak indicating that the primersamplify one specific target For absolute quantification a

ISRN Obesity 3

standard curve was generated for each gene in every assayTo do so 100ndash04 ng of RNA was reverse transcribed asdescribed above The resulting cDNA was used to obtain astandard curve correlating the amounts with the thresholdcycle number (Ct values) A linear relationship (1199032 gt096) was obtained for each gene Real-time PCR was thenperformed on samples and standards in triplicate The platesetup also included a standard series no template controlno RNA control no reverse transcriptase control and noamplification control The dissociation curve was analyzedfor each sample Absolute quantification ofmRNAexpressionlevels for individual transcriptswas calculated bynormalizingthe values to GAPDH The results were analyzed with atwo-tailed Studentrsquos 119905-test using PRISM4 statistical analysissoftware (GraphPad) A level of 119875 lt 005 was consideredstatistically significant Significance was determined afterthree or more experiments

24 Western Blot Analysis Protein lysates were obtainedfrom the cells using lysis buffer containing proteaseinhibitors Protein lysates were also harvested from the snap-frozen adipose tissues by homogenization and sonicationin the lysis buffer The lysates (40 120583g) were separated bySDS-PAGE with 10 gels electrophoretically transferredto nitrocellulose membranes blocked with Tris-bufferedsaline containing 01 Tween 20 and 5 nonfat dried milkwashed and incubated with anti-PKC120575 or antibody specificfor PKC120575VIII (Patel lab [8]) anti-PPAR120574 anti-adiponectinanti-Bcl2 anti-caspase 9 (Cell Signaling) anti-XIAP(AnaSpec) anti-Bcl-xL (Santa Cruz Biotechnology) and120573-actin (Sigma) After incubation with antirabbit IgG-HRP enhanced chemiluminescence (Pierce) was used fordetection

25 Adipocyte Size Adipocytes were differentiated in100mm plate and their sizes determined on day 14 (ieof mature adipocytes) Maximal diameter of 10 adjacentadipocytes from 6 different fields was calculated using NikonEclipse inverted microscope and NIS elements advanceresearch image tool software The data was transferred toExcel to calculate the mean diameter and standard deviationThis was repeated in three separate experiments to ensurereproducibility

26 Apoptosis Assay Human preadipocytes were culturedon 60mm dishes as described in methods For apoptosisassays cells were serum starved for 48 hours Media werecollected and cells were washed one time with HBSS andthen trypsinized for 50 minutes Five ml complete mediawas added to neutralize the trypsin Media and washes werepooled and centrifuged at 1200 RPMS for five minutes Cellswere washed one time with PBS and one time with bindingbuffer and then incubated for 15min with 50 120583L AV-FITCand 50 120583L PI in 100 uL binding buffer (BD PharmagenSan Diego CA USA) at room temperature in the dark400120583L binding buffer was added and cells were analyzedby flow cytometry within one hour Annexin V-FITC and

PI fluorescence were measured using an Accuri C6 flowcytometer

27 Statistical Analysis Analyses were performed usingPRISM software and analyzed using either two-tailed Stu-dentrsquos 119905-test or two-way ANOVA lowast119875 lt 005 was signifi-cant lowastlowastlowast119875 lt 00001 was highly significant Analysis wasperformed within group and between groups

3 Results

31 Human Preadipocytes Are Susceptible to ApoptosisHuman preadipocytes were commercially obtained fromZenBio such that a pure population of preadipocytesdevoid of contaminating stem cells or adipocytes or othernonadipocyte lineage cells were used in our experimentsThecells were cultured as described inmethods Cells were serumstarved for 48 hours to induce apoptosis and assayed by flowcytometry on days 0 and 10 FITC Annexin V was used toquantitatively assess cells undergoing apoptosis along withpropidium iodide to enable detection of percentage of cellsundergoing either early apoptosis or late apoptosis AnnexinV binds to phosphatidylserine which is displayed on the cellmembrane of apoptotic cells and PI will stain only dead ordamaged cellsThe results (Figure 1) demonstrate that on day0 preadipocytes which are serum starved undergo increasedapoptosis while the mature adipocytes (day 10) are moreresistant to apoptosis upon serum starvation

32 Differentiation of Human Preadipocytes DemonstratesSwitch in Apoptosis Genes Since the adipocytes demon-strated a switch in the susceptibility to apoptosis duringdifferentiation we sought to elucidate the expression patternsof apoptotic genes during differentiation We evaluated thealternative splicing of genes during differentiation with afocus on apoptotic genes We used gene-level exon arrayfrom Affymetrix (experiments performed by MicroarrayCore Facility Moffitt Cancer Institute and analyzed withExon array analyzer software to sort specific biochemicalpathways and genes) to identify a group of transcripts that arepredominantly alternatively spliced and expressed We useddays 0 6 and 8 of differentiating human preadipocytes (datanot shown) The splicing pattern of genes of the apoptosispathway that showed a dramatic change between days 0 and6 of adipogenesis is explained below

33 Alternative Splicing of PKC120575 Protein kinase C delta(PKC120575) is a serinethreonine kinase which plays a central rolein apoptosis Previously we identified a new splice variantof human PKC120575 PKC120575VIII (Figure 2(a)) Sequencing andcomputational analysis of the PKC120575VIII sequence indicatedthat this human splice variant is generated by utilization ofan alternative downstream 51015840 splice site of PKC120575 pre-mRNAexon 10 Our data indicated that PKC120575I promotes apoptosiswhile PKC120575VIII promotes cell survival [8]

34 Alternative Splicing of Bcl-x Bcl-x is a dominant regu-lator of programmed cell death in mammalian cells Bcl-x is

4 ISRN Obesity

lowastlowastlowast

lowastlowastlowast

60

50

40

30

20

10

0

Ann

exin

PI

Day 0 Day 10

ControlSerum deprived

lowastlowastlowast

Figure 1 Human lean preadipocytes were differentiated in vitroCells were serum deprived for 48 h on day 0 (preadipocytes) andon day 10 (mature adipocytes) AnnexinPI was analyzed by flowcytometry Experiments were repeated 5 times Statistical analysisperformed by two-way ANOVA 119875 gt 075 ns not significant withingroup lowastlowastlowast119875 lt 00001 highly significant between control and serum-deprived lowastlowastlowast119875 lt 00001 highly significant between days 0 and 10serum-deprived

alternatively spliced to Bcl-xL Bcl-xS and Bcl-x120573 Bcl-xL isgenerated via alternate 51015840 splice site selection on exon 2 ofBcl-x pre-mRNA and includes the BH1 and BH2 domains(Figure 2(b)) Bcl-x120573 is the least studied and has no knownrole in apoptosis The long form Bcl-xL inhibits apoptosiswhile the short isoform Bcl-xS promotes apoptosis Bcl2 andBcl-xL dimerize to initiate the cell survival pathways

35 Alternative Splicing of Caspase 9 Caspase 9 an initiatorcaspase is a ubiquitously expressed protease It representsa pivotal signaling protein in the apoptotic cascade Inhibi-tion of Caspase 9 by a chemical inhibitor mollugin affectsadipogenesis [10 11] It is alternatively spliced to caspase 9a(apoptotic) and caspase 9b (antiapoptosis) via inclusion ofthe cassette exons 3 4 5 and 6 in caspase 9a (Figure 2(c))Caspase 9a activates caspase 3 which cleaves substratesthereby mediating apoptosis

To verify our results from the exon array we carriedout a detailed analysis of the differentiation of humanpreadipocytes from days 0 to 10 as they differentiate intomature adipocytes as described Our data indicates thatbetween days 2 and 4 of differentiation a marked shift isobserved in the expression of genes involved in prosurvivalpathways (Figure 2(d)) Day 6 of differentiation is markedby the splicing of antiapoptotic proteins that advances the

mature adipocytes to be resistant to apoptosis On day 0cells expressed the apoptotic splice variants PKC120575I caspase9a and lower levels of Bcl-xL By day 4 cells startedexpressing the antiapoptotic proteins Bcl2 and splice variantsPKC120575VIII caspase 9b and Bcl-xL with a concurrent decreasein proapoptotic proteins This represents a critical switchin adipocyte differentiation modulated by the splicing ofapoptosis genes

36 Expression of PKC120575 Splice Variants in Obese PatientsOur above data was obtained in normal lean differentiatingpreadipocytes Next we sought to determine whether thesplicing pattern and expression of these genes was affectedby the metabolic state Hence we sought to measure thelevels of PKC120575 from obese patients We obtained proteinlysates from subcutaneous and omental adipose tissues froman obese patient undergoing Roux-en-y gastroenterostomysurgery (from Drs Murr and Cooper Tampa VAMC) Thesamples were from an obese patient closely matching theBMI and other criteria to the preadipocytes obtained fromZenBio to allow for a better comparisonThe results indicateda sharp increase in PKC120575VIII expression in the obese patientcompared to PKC120575I (Figure 3(a)) It is important to note thatthese adipose tissue samples represent a heterogeneous cellpopulation containing predominantlymature adipocytes andalso differentiating preadipocytes as well as other stem cellswhich may or may not have the adipocyte lineage

Next we obtained preadipocytes from the subcutaneousdepot of obese patients (ZenBio) and differentiated them invitro We chose to further study the preadipocyte differentia-tion of obese subcutaneous fraction because our lean cell linewas derived from the subcutaneous fractionThe cell line wasselected to match the adipose tissue lysate (above) such thatwe could evaluate differentiation of obese preadipocytes invitro It also allowed us to compare the expression patterns ofthe genes between the lean and obese during differentiationWe first evaluated apoptosis and susceptibility to serumdeprivation on days 0 and 10 by annexin V and PI flowcytometry Our results (Figures 3(b) and 3(c)) indicatedthat obese preadipocytes are more resistant to apoptosiscompared to lean on day 0 (UR quadrant with both AVand PI lean 307 obese 167) Day 10 mature adipocytesdemonstrate much lower susceptibility to serum-deprivedapoptosis (lean 187 obese 104)

To determine the splicing pattern of differentiating obesepreadipocytes we performed Western blot analysis Weobserved that these cells expressed PKC120575VIII Bcl-xL Bcl2and caspase 9 on day 0 that is in the preadipocyte stage Weobserved increased expression of PKC120575VIII as determined bythe PKC120575 splice variant ratio observed on day 0 (Figure 4)This data suggests that the early expression of these genesin the adipocytes promotes their survival mechanisms andthereby provides resistance to cell death

37 Adipogenesis in Obesity Is Dysregulated The lean andobese human preadipocytes were differentiated for 0 to14 days and adipogenesis was observed by staining of thelipid droplets by Oil Red O The lean cells showed lipid

ISRN Obesity 5

PKC120575VIIImRNA E9 E10 E11

Cell survivalantiapoptosis

PKC120575pre-mRNA

E9 E10 E11

SS II

SS I

PKC120575ImRNA

E9 E10 E11 Cell apoptosis

(a)

Bcl-xLmRNA E1 E2 E3

Cell survivalantiapoptosis

Bcl-xpre-mRNA

E1 E2 E3SS II

SS I

Bcl-xSmRNA

E1 E2 E3 Cell apoptosis

(b)

Caspase 9amRNA Cell apoptosis

Caspase 9Pre-mRNA

Exon 2 Exons 3 4 5 6Cassette Exon 7

Caspase 9bmRNA

Cell survivalantiapoptosis

(c)

Day

PKC120575VIIIPKC120575I

Bcl-2

Bcl-xL

Caspase 9aCaspase 9b

0 2 4 6 8 10 0 2 4 6 8 10

TNF120572

XIAP

Adiponectin

PPAR

120573-actin

60

50

40

30

20

10

0Day 0 Day 2 Day 4 Day 6 Day 8 Day 10

120574

PKC120575

exon

inclu

sion

()

(d)

Figure 2 (a) Schematic of alternative splicing of PKC120575 pre-mRNA generating PKC120575I mRNA and PKC120575VIII mRNA via alternative 51015840 splicesite selection E exon SS splice site (b) Schematic of alternative splicing of Bcl-x pre-mRNA generating Bcl-xL mRNA and Bcl-xS mRNAvia alternative 51015840 splice site selection E exon SS splice site (c) Schematic of alternative splicing of caspase 9 pre-mRNA generating caspase9a mRNA and caspase 9b mRNA via cassette exon inclusion (d) Western blot analysis of differentiating lean preadipocytes from days 0 to 10using antibodies as indicated in the figure Between days 2 and 4 a marked shift is observed in the splicing pattern of survival proteins Thisperiodmarks terminal differentiation of adipocytesThe blots are representative of 3 experiments performed individually with similar resultsGraph represents percent PKC120575 exon inclusion calculated as PKC120575VIII(PKC120575VIII + PKC120575I) times 100 and is representative of four experimentsperformed separately

accumulation beginning at day 4 We observed that obesepreadipocyteswere slower to differentiate but exhibited largerlipid droplets by day 10 We measured and compared themature adipocyte size on day 14 between lean and obesesamples (Figure 5) Our results indicate a dramatic increasein adipocyte size in obese samples Further the expressionof adiponectin and PPAR120574 was dysregulated in differen-tiating obese preadipocytes (Figure 4) which may explainthe aberrant differentiation and adipogenesis observed in

obese patients These results indicated that the adipogenesisprogram was dysregulated in obese adipocytes

To evaluate this we performed real-time qPCR analysisto measure the expression of adipogenesis marker genesFOXO SIRT1 adiponectin and PPAR120574 with a concurrentmeasure of PKC120575VIII expression Total RNA was isolatedon day 2 of differentiation of lean and obese preadipocytesOur results (Figure 6) indicated that PKC120575VIII expressionwas increased dramatically in obese samples along with a

6 ISRN Obesity

E01 Unstained controlGate [no gating]

16777215

10000000

5000000

0

SSC-

A

0 5000000 10000000 16777

FSC-A

Day 10

1072

106

105

104

103

102

PI-A

103 104 105 106 1072

Annexin V-A

F04 Lean-controlGate (P1 in all)

1072

106

105

104

103

102

PI-A

103 104 105 106 1072

Annexin V-A

1072

106

105

104

103

102

PI-A

103 104 105 106 1072

Annexin V-A

1034 104 105 106 1072

Annexin V-A

1072

106

105

104

1027

PI-A

1034 104 105 106 1072

Annexin V-A

G06 Lean-serum deprived

H05 Obese-controlGate (P1 in all)

H07 Obese-serum deprivedGate (P1 in all)

E04 Lean-controlGate (P1 in all)

E06 Lean-serum deprivedGate (P1 in all)

E12 Obese-controlGate (P1 in all)

F05 Obese-serum deprivedGate (P1 in all)

Gate (P1 in all)

1072

106

105

104

1027PI

-A

1072

106

105

104

103

102

PI-A

103 104 105 106 1072

Annexin V-A

1034 104 105 106 1072

Annexin V-A

1072

106

105

104

1027

PI-A

1034 104 105 106 1072

Annexin V-A

1072

106

105

104

1027

PI-A

F01 Unstained controlGate [no gating]

16777215

10000000

5000000

0

SSC-

A0 5000000 10000000 16777

FSC-A

Day 0

lowastlowastlowast

lowastlowastlowast lowastlowastlowast

35

30

25

20

15

10

5

0

Ann

exin

VP

I

Leancontrol

Leanserum

deprivedObesecontrol

Obeseserum

deprived

Day 0Day 10

lowastlowastlowastlowastlowastlowast

(a)

(b)

(c)

1 2

PKC120575VIIIPKC120575I

120573-actin

80

60

40

20

0Omental Subcutaneous

PKC120575

exon

inclu

sion

()

Figure 3 (a) Western blot analysis of adipose tissue lysates obtained from obese depots from (1) omental (2) subcutaneous Theblots are representative of experiments repeated thrice from the lysates Graph represents percent PKC120575 exon inclusion calculated asPKC120575VIII(PKC120575VIII + PKC120575I)times 100 and is representative of three experiments (b)Human lean and obese preadipocytes were differentiatedin vitro Cells were serum deprived for 48 h on day 0 (preadipocytes) and on day 10 (mature adipocytes) Annexin VPI staining was analyzedby flow cytometryThe first plot demonstrates the gating of the adipocyte populationThe remaining plots shown are lean control lean-serumdeprived obese control and obese serum-deprived and are divided into four quadrants based on Annexin V (AV) and propidium iodidestaining (PI) staining Q1-LL represents viable (unstained) cells Q1-LR represents AV staining only (early apoptosis) Q1-UR represents cellsstainedwith bothAVandPI (later apoptosis) Q1-UL represents cells stainedwith PI only (end-stage apoptosis) Data shown are representativeof five different experiments (c) Graph represents AVPI and represents 5 experiments Statistical analysis performed by two-way ANOVA119875 gt 075 ns not significant within group lowastlowastlowast119875 lt 00001 highly significant between control and serum-deprived lowastlowastlowast119875 lt 00001 highlysignificant between lean and obese lowastlowastlowast119875 lt 00001 highly significant between days 0 and 10 serum-deprived

ISRN Obesity 7

0 2 4 6 8 10Day 0 2 4 6 8 10

PKC120575VIIIPKC120575I

Bcl-2

Bcl-xL

Caspase 9a

Caspase 9b

Glut4

Adiponectin

PPAR120574

120573-actin

60

50

40

30

20

10

0Day 0 Day 2 Day 4 Day 6 Day 8 Day 10

PKC120575

exon

inclu

sion

()

Figure 4 Western blot analysis of differentiating subcutaneousobese preadipocytes from days 0 to 10 using antibodies as indicatedThe blots are representative of four experiments performed indi-vidually with similar results Graph represents percent PKC120575 exoninclusion calculated as PKC120575VIII(PKC120575VIII + PKC120575I) times 100 and isrepresentative of four experiments performed separately

Lean Obese

(a)

40000

35000

30000

25000

20000

15000

10000

5000

0Lean Obese

Mea

n ad

ipoc

yte s

ize (120583

m)

(b)

Figure 5 Human lean or obese preadipocytes were differentiated invitro (a) Adipocytes on day 14 were stained for lipid content withOil Red O and (b) adipocyte size measured (methods) Experimentswere repeated 4 times with similar result

25

2

15

1

05

0

mRN

A ex

pres

sion

lowast

lowast

lowast

lowast

lowast

SIRT1 FOXO1 Adiponectin

LeanObese

PKC120575VIII PPAR120574

Figure 6 Lean and obese human preadipocytes were differentiatedin vitro Total RNA was collected on day 2 Real Time SYBR qPCRanalysis of PKC120575VIII SIRT1 FOXO1 adiponectin and PPAR120574was performed in triplicate and repeated 3 times in separateexperimentsThe absolute mRNA expression transcripts are shownlowast119875 lt 005 (by two-tailed Studentrsquos 119905-test) significant between leanand obese for each mRNA

marked decrease in FOXO SIRT1 adiponectin and PPAR120574expression compared to lean samples

4 Discussion

Obesity is an epidemic facing the world with the USAleading the statistics on increasing number of overweight andobese population Adipose tissue is an important endocrineregulator of energy homeostasis and glucose metabolismand the excess energy is stored in the white adipose tissueThe onset of adipogenesis and the transcriptional factorsthat initiate it have been studied in depth However withour current knowledge of the genome it is evident thatthere are several other mechanisms in play which regulatethe adipocyte phenotype during differentiation Apoptosis isrequired for fat cell turnover throughout our life Howeverfor weight loss in obese patients it may be beneficial toincrease adipocyte apoptosis along with therapies targetinghyperplasia and hypertrophy in adipocytes To our knowl-edge this is the first demonstration of alternative splicingevents during adipogenesis which contribute to the matureadipocyte phenotype Our results indicate a distinct splicingpattern between lean and obese patients

Apoptosis in preadipocytes and adipocytes was previ-ously observed in human cell lines However the under-lying mechanisms were not described Alternative splicinga posttranscriptional event is a powerful regulator of geneexpression We have identified key genes which are alter-natively spliced that dramatically switch their splice variantexpression during the progression of adipogenesis This mayexplain why preadipocytes are more susceptible to apoptoticagents while the mature adipocytes are resistant to it Protein

8 ISRN Obesity

kinase C delta (PKC120575) is a crucial signaling kinase affectingdownstream apoptotic pathways [8 12] Our data indicatesincreased expression of PKC120575VIII along with an early switchbetween the splice variants associated with apoptosis duringadipogenesis process of preadipocytes derived from obesepatients

PKC120575 is a serinethreonine kinase which plays a centralrole in apoptosis PKC120575 has dual functions as a mediator ofapoptosis and as an antiapoptosis effector Its splice variantsfunction as a switch that determines cell survival and fateThis could explain the opposing effects of PKC120575 on cellularapoptosis cited in the literature [13ndash15] Genetic switchesbased on alternative splicing are important for many cellularand developmental processes PKC120575I promotes apoptosiswhile PKC120575VIII promotes survival We have shown thatPKC120575II (the mouse splice variant) and PKC120575VIII functionas prosurvival proteins [8 12] the functions of other PKC120575splice variants are not yet established Our previous studiesin neurons have demonstrated that the prosurvival variant ofPKC120575 increases the expression of Bcl2 and Bcl-xL which areknown to increase cellular survival [16]

Obesity is associated with a chronic low inflamma-tory state due to the changes in secretory functions ofthe adipocytes Enlarged adipocytes (hypertrophy) lead toincreased adipocyte-derived free fatty acids which stimulatemacrophages In severely obese patientsmacrophages invadethe adipose tissue and release chemokines leading to necrosisand autophagy of adipocytes [17] These complications addto obese-associated metabolic syndrome diabetes insulinresistance and hepatic steatosis These forms of cell deathare distinct from apoptosis (programmed cell death) inadipocytes Apoptosis is shown to be substantially greaterin omental depots compared to subcutaneous depots inhealthy individuals [18] Here we show that obese samplesare resistant to apoptosis and express the antiapoptosis genesfourfold greater than in lean samples This may explain theincreased difficulty to lose weight and maintain the weightloss in overweight and obese individuals Larger adipocytesare an indicator of type II diabetes independent of insulinresistance [19] Previous studies showed that PPAR120574 andadiponectin expression was lower in obesity and type IIdiabetes [20 21] Our data indicated that obese patients havedysregulated adipogenesis program This may be modulatedby the signaling kinase PKC120575VIII which is overexpressedin obese individuals Our data also indicated dysregulatedalternative splicing of apoptosis genes in obese differentiatingcells This aberrant molecular profile may be a causativeeffect leading to obesity or may be a consequence of obesityand its comorbidities We are evaluating this further in ourlaboratory Here we have defined the differences in geneexpression between lean and obese adipocytes

In conclusion it may be possible to modulate alternativesplicing during adipogenesis for therapeutic treatments ofobesity Our results suggest a novel role for a key signalingkinase PKC120575VIII in the process of adipogenesis Expressionlevels of PKC120575VIII splice variant are increased in obesity andmay be linked to the cause or consequence of dysregulatedadipogenesis in obese patients Since new adipocytes arerequired for the maintenance and the endocrine function of

adipocytes balancing the gene expression of apoptotic genesmay be a better approach rather than complete inhibition ofadipogenesis

Conflict of Interests

The authors have no conflict of interests to declare

Authorsrsquo Contribution

GCarter andAApostolatos contributed equally to the paper

Acknowledgments

This work was supported by the Department of VeteransAffairs Medical Research Grant (NAP)The contents do notrepresent the views of the Department of Veterans Affairs ortheUnited States GovernmentThe authors thankDr AndreaMoor for help with microscopic images and adipocyte sizedetermination They thank Sean Yoder from Moffitt arraycore for array analysis

References

[1] S Rodrıguez-Acebes N Palacios J I Botella-Carretero et alldquoGene expression profiling of subcutaneous adipose tissue inmorbid obesity using a focused microarray distinct expressionof cell-cycle- and differentiation-related genesrdquo BMC MedicalGenomics vol 3 article 61 2010

[2] M Guzey D Jukic J Arlotti M Acquafondata R Dhir andR H Getzenberg ldquoIncreased apoptosis of periprostatic adiposetissue in VDR null micerdquo Journal of Cellular Biochemistry vol93 no 1 pp 133ndash141 2004

[3] P Fischer-Posovszky H Tornqvist K M Debatin and MWabitsch ldquoInhibition of death-receptor mediated apoptosis inhuman adipocytes by the insulin-like growth factor I (IGF-I)IGF-I receptor autocrine circuitrdquo Endocrinology vol 145 no4 pp 1849ndash1859 2004

[4] O Ishiko T Sumi H Yoshida Y Hyun and S Ogita ldquoCom-parison of expression of apoptosis regulatory proteins in theadipose tissue of tumor-bearing and diet-restricted rabbitsrdquoInternational Journal of Molecular Medicine vol 8 no 5 pp543ndash547 2001

[5] M A Della-Fera Y H Choi D L Hartzell J Duan MHamrick and C A Baile ldquoSensitivity of obob mice to Leptin-induced adipose tissue apoptosisrdquo Obesity Research vol 13 no9 pp 1540ndash1547 2005

[6] D H Kim S C Woods and R J Seeley ldquoPeptide designedto elicit apoptosis in adipose tissue endothelium reduces foodintake and body weightrdquo Diabetes vol 59 no 4 pp 907ndash9152010

[7] W C Earnshaw L M Martins and S H Kaufmann ldquoMam-malian caspases structure activation substrates and functionsduring apoptosisrdquo Annual Review of Biochemistry vol 68 pp383ndash424 1999

[8] K Jiang A H Apostolatos T Ghansah et al ldquoIdentificationof a novel antiapoptotic human protein kinase C 120575 isoformPKC120575VIII in NT2 cellsrdquo Biochemistry vol 47 no 2 pp 787ndash797 2008

ISRN Obesity 9

[9] H Apostolatos A Apostolatos T Vickers et al ldquoVitamin Ametabolite all-trans-retinoic acid mediates alternative splicingof protein kinase C 120575VIII (PKC120575VIII) isoform via splicingfactor SC35rdquo Journal of Biological Chemistry vol 285 no 34pp 25987ndash25995 2010

[10] D Y Jun C R Han M S Choi M A Bae M H Woo and YH Kim ldquoEffect of mollugin on apoptosis and adipogenesis of3T3-L1 preadipocytesrdquo Phytotherapy Research vol 25 no 5 pp724ndash731 2011

[11] S M Kim H S Park D Y Jun et al ldquoMollugin induces apop-tosis in human Jurkat T cells through endoplasmic reticulumstress-mediated activation of JNK and caspase-12 and subse-quent activation of mitochondria-dependent caspase cascaderegulated by Bcl-xLrdquo Toxicology and Applied Pharmacology vol241 no 2 pp 210ndash220 2009

[12] N A Patel S S Song and D R Cooper ldquoPKC120575 alternativelyspliced isoforms modulate cellular apoptosis in retinoic acid-induced differentiation of human NT2 cells and mouse embry-onic stem cellsrdquo Gene Expression vol 13 no 2 pp 73ndash84 2005

[13] J J Peluso A Pappalardo and G Fernandez ldquoBasic fibrob-last growth factor maintains calcium homeostasis and gran-ulosa cell viability by stimulating calcium efflux via a PKC120575-dependent pathwayrdquo Endocrinology vol 142 no 10 pp 4203ndash4211 2001

[14] L E Kilpatrick J Y Lee K M Haines D E Campbell KE Sullivan and H M Korchak ldquoA role for PKC-120575 and PI3-kinase in TNF-120572-mediated antiapoptotic signaling in thehuman neutrophilrdquo The American Journal of Physiology vol283 no 1 pp C48ndashC57 2002

[15] A Zrachia M Dobroslav M Blass et al ldquoInfection of gliomacells with Sindbis virus induces selective activation and tyrosinephosphorylation of protein kinase C 120575 implications for sindbisvirus-induced apoptosisrdquo Journal of Biological Chemistry vol277 no 26 pp 23693ndash23701 2002

[16] A Apostolatos S Song S Acosta et al ldquoInsulin promotesneuronal survival via the alternatively spliced protein kinaseCdeltaII isoformrdquo Journal of Biological Chemistry vol 287 pp9299ndash9310 2012

[17] K J Strissel Z Stancheva H Miyoshi et al ldquoAdipocytedeath adipose tissue remodeling and obesity complicationsrdquoDiabetes vol 56 no 12 pp 2910ndash2918 2007

[18] D Papineau A Gagnon and A Sorisky ldquoApoptosis of humanabdominal preadipocytes before and after differentiation intoadipocytes in culturerdquo Metabolism vol 52 no 8 pp 987ndash9922003

[19] C Weyer J E Foley C Bogardus P A Tataranni and RE Pratley ldquoEnlarged subcutaneous abdominal adipocyte sizebut not obesity itself predicts type II diabetes independent ofinsulin resistancerdquo Diabetologia vol 43 no 12 pp 1498ndash15062000

[20] L B Tanko A Siddiq C Lecoeur et al ldquoACDCadiponectinand PPAR-120574 gene polymorphisms implications for features ofobesityrdquo Obesity Research vol 13 no 12 pp 2113ndash2121 2005

[21] N Ouchi S Kihara T Funahashi Y Matsuzawa and KWalshldquoObesity adiponectin and vascular inflammatory diseaserdquo Cur-rent Opinion in Lipidology vol 14 no 6 pp 561ndash566 2003

2 ISRN Obesity

on transcriptional control in adipocyte differentiation Addi-tional mechanisms of gene expression which are required tomaintain the adipocyte phenotype are unknown

Cell death is achieved by several mechanisms includingnecrosis (sudden cell death fatally affecting neighboringcells) apoptosis (programmed cell death) and autophagy(large scale degradation via lysosomes) It is now establishedthat loss of adipose tissue mass in rodents and humansis achieved by inducing apoptosis [2ndash6] Caspases directlyand indirectly orchestrate the morphologic changes of thecell during apoptosis Caspases have been characterized intoupstream initiators (caspase 2 8 9 10) and downstreamexecutioners (caspase 3 6 7) [7] Bcl-2 protein family hasabout 17 members and the decision of a cell to undergoapoptosis is governed by a complex network of interac-tions between these Bcl-2members caspases SmacDIABLO(second mitochondria-derived activator of caspase) andX-linked inhibitor of apoptosis (XIAP) The prosurvivalsubfamily comprises of Bcl-2 Bcl-xL Bcl-w Mcl-1 A1and human Bcl-B The release of SmacDIABLO from themitochondria is regulated by antiapoptotic Bcl-2 proteinsmdashBcl-2 and Bcl-xL Hence the decision to undergo apoptosis isa cascade of events and proteins that work in concert PKC120575is a signaling kinase affecting downstream apoptotic path-ways Our laboratory has demonstrated that the alternativelyspliced products of human PKC120575 have distinct functionsin apoptosis PKC120575I promotes apoptosis while PKC120575VIII isantiapoptotic [8]

Earlier studies in mouse models indicated thatpreadipocytes can undergo programmed cell death(apoptosis) Here we demonstrate that human preadipocytesare susceptible to apoptosis while mature adipocytesare resistant to apoptosis Many genes in the apoptosispathway are alternatively spliced We sought to determinethe alternative splicing patterns of genes in the apoptosispathway which promote the antiapoptotic nature of themature adipocytes This contributes to the inability of anindividual to maintain weight loss as diet and exercise mayreduce the size of the fat cells but the cells do not die andcan easily gain back the fat mass The excess energy (input gtexpenditure) is stored in the white adipose tissue (WAT)which is also an important endocrine regulator

2 Materials and Methods

21 Human Preadipocytes The cells were purchased as cry-opreserved preadipocytes from ZenBio (Research TrianglePark NC USA) Since these commercial cell lines andtheir information are publicly available they are exemptfrom human subjects research determination from IRB inaccordance with HHS regulation at 45 CFR Part 46 Thelean preadipocytes originated from subcutaneous adiposetissue of a healthy donor of 26 years of age undergoingelective surgery with a body mass index of 213 who wasnot diabetic and not a smoker The obese preadipocytesoriginated from subcutaneous adipose tissue of a 40-year-old donor undergoing elective surgery with a body massindex of 546 who was not diabetic and not a smoker

The preadipocyte cell line (prescreen by ZenBio) is sortedand characterized such that it represents only preadipocytesand is free from stem cells or differentiating adipocytes orcontaminating endothelial cells Both the preadipocytes aretested in culture to differentiate into mature adipocytes andshow accumulation of lipid express aP2 respond to lipolyticagents and secrete adiponectin and leptin At the start ofall experiments cells are grown to confluency such thatall cells are synchronized and then differentiated The cellswere cultured according to the manufacturerrsquos instructionsBriefly cryopreserved preadipocytes were passaged withpreadipocyte medium (PM-1 DMEMHamrsquos F-12 mediumHEPES FBS penicillin streptomycin amphotericin B Zen-Bio) and then plated 40625 cellscm2 with PM-1 Cells werefed every other day with PM-1 until confluent To inducedifferentiation PM-1 medium was replaced with differentia-tionmedium (DM2Zen-Bio) including biotin pantothenatehuman insulin dexamethasone isobutylmethylxanthine anda PPAR120574 agonist (days 0ndash7) After 7 days DM-2 mediumwas removed and cells were incubated for additional 7days with adipocyte medium (AM1 Zen-Bio days 7ndash14)which included PM-1 biotin pantothenate human insulinand dexamethasone By day 14 cells contained large lipiddroplets and were considered mature adipocytes Cells weremaintained at 37∘C in a humidified 5 CO

2atmosphere

22 Adipose Lysates We obtained protein lysates from DrsMurr and Cooper USF and VAMC Tampa The proteinlysates were harvested from adipose tissues in Dr Cooperrsquoslab and originated from subcutaneous and omental adiposedepots from an obese patient The tissues were obtainedunder consent from a deidentified patient age 45 nondi-abetic BMI 56 undergoing Roux-en-y gastroenterostomysurgery The fat collection protocols were approved by theInstitutional Review Board of the University of South FloridaEthics Committee protocol number 108360

23 Quantitative Real-Time RT-PCR Total RNA was iso-lated from differentiated lean and obese preadipocytes usingRNAbee according to the manufacturerrsquos protocol (TelTestInc) 2 120583g RNAwas reverse transcribed usingQiagenrsquos RT kit2 120583L of cDNA was amplified by real-time quantitative PCRusing Syber (SYBR)Greenwith anABIPRISM7900 sequencedetection system (PE Applied Biosystems) to quantify theabsolute levels of the transcripts in the samples 120573-actin wasused as the endogenous control Two sets of primer pairsfor each transcript FOXO1 SIRT1 PPAR120574 and adiponectinwere designed to have an annealing temperature of sim60∘CPrimers for PKC120575VIII are described in our earlier publication[9] These primers were initially tested using cDNA fromhuman preadipocytes in an RT-PCR reaction using Taqpolymerase to give distinct products corresponding to therespective transcripts Next the optimal primer concentra-tion was determined from a range of 50ndash900 nM The finalconcentration of 300 nM was selected to ensure efficiencyand specificity for its target based on the dissociation curvethat showed a single sharp peak indicating that the primersamplify one specific target For absolute quantification a

ISRN Obesity 3

standard curve was generated for each gene in every assayTo do so 100ndash04 ng of RNA was reverse transcribed asdescribed above The resulting cDNA was used to obtain astandard curve correlating the amounts with the thresholdcycle number (Ct values) A linear relationship (1199032 gt096) was obtained for each gene Real-time PCR was thenperformed on samples and standards in triplicate The platesetup also included a standard series no template controlno RNA control no reverse transcriptase control and noamplification control The dissociation curve was analyzedfor each sample Absolute quantification ofmRNAexpressionlevels for individual transcriptswas calculated bynormalizingthe values to GAPDH The results were analyzed with atwo-tailed Studentrsquos 119905-test using PRISM4 statistical analysissoftware (GraphPad) A level of 119875 lt 005 was consideredstatistically significant Significance was determined afterthree or more experiments

24 Western Blot Analysis Protein lysates were obtainedfrom the cells using lysis buffer containing proteaseinhibitors Protein lysates were also harvested from the snap-frozen adipose tissues by homogenization and sonicationin the lysis buffer The lysates (40 120583g) were separated bySDS-PAGE with 10 gels electrophoretically transferredto nitrocellulose membranes blocked with Tris-bufferedsaline containing 01 Tween 20 and 5 nonfat dried milkwashed and incubated with anti-PKC120575 or antibody specificfor PKC120575VIII (Patel lab [8]) anti-PPAR120574 anti-adiponectinanti-Bcl2 anti-caspase 9 (Cell Signaling) anti-XIAP(AnaSpec) anti-Bcl-xL (Santa Cruz Biotechnology) and120573-actin (Sigma) After incubation with antirabbit IgG-HRP enhanced chemiluminescence (Pierce) was used fordetection

25 Adipocyte Size Adipocytes were differentiated in100mm plate and their sizes determined on day 14 (ieof mature adipocytes) Maximal diameter of 10 adjacentadipocytes from 6 different fields was calculated using NikonEclipse inverted microscope and NIS elements advanceresearch image tool software The data was transferred toExcel to calculate the mean diameter and standard deviationThis was repeated in three separate experiments to ensurereproducibility

26 Apoptosis Assay Human preadipocytes were culturedon 60mm dishes as described in methods For apoptosisassays cells were serum starved for 48 hours Media werecollected and cells were washed one time with HBSS andthen trypsinized for 50 minutes Five ml complete mediawas added to neutralize the trypsin Media and washes werepooled and centrifuged at 1200 RPMS for five minutes Cellswere washed one time with PBS and one time with bindingbuffer and then incubated for 15min with 50 120583L AV-FITCand 50 120583L PI in 100 uL binding buffer (BD PharmagenSan Diego CA USA) at room temperature in the dark400120583L binding buffer was added and cells were analyzedby flow cytometry within one hour Annexin V-FITC and

PI fluorescence were measured using an Accuri C6 flowcytometer

27 Statistical Analysis Analyses were performed usingPRISM software and analyzed using either two-tailed Stu-dentrsquos 119905-test or two-way ANOVA lowast119875 lt 005 was signifi-cant lowastlowastlowast119875 lt 00001 was highly significant Analysis wasperformed within group and between groups

3 Results

31 Human Preadipocytes Are Susceptible to ApoptosisHuman preadipocytes were commercially obtained fromZenBio such that a pure population of preadipocytesdevoid of contaminating stem cells or adipocytes or othernonadipocyte lineage cells were used in our experimentsThecells were cultured as described inmethods Cells were serumstarved for 48 hours to induce apoptosis and assayed by flowcytometry on days 0 and 10 FITC Annexin V was used toquantitatively assess cells undergoing apoptosis along withpropidium iodide to enable detection of percentage of cellsundergoing either early apoptosis or late apoptosis AnnexinV binds to phosphatidylserine which is displayed on the cellmembrane of apoptotic cells and PI will stain only dead ordamaged cellsThe results (Figure 1) demonstrate that on day0 preadipocytes which are serum starved undergo increasedapoptosis while the mature adipocytes (day 10) are moreresistant to apoptosis upon serum starvation

32 Differentiation of Human Preadipocytes DemonstratesSwitch in Apoptosis Genes Since the adipocytes demon-strated a switch in the susceptibility to apoptosis duringdifferentiation we sought to elucidate the expression patternsof apoptotic genes during differentiation We evaluated thealternative splicing of genes during differentiation with afocus on apoptotic genes We used gene-level exon arrayfrom Affymetrix (experiments performed by MicroarrayCore Facility Moffitt Cancer Institute and analyzed withExon array analyzer software to sort specific biochemicalpathways and genes) to identify a group of transcripts that arepredominantly alternatively spliced and expressed We useddays 0 6 and 8 of differentiating human preadipocytes (datanot shown) The splicing pattern of genes of the apoptosispathway that showed a dramatic change between days 0 and6 of adipogenesis is explained below

33 Alternative Splicing of PKC120575 Protein kinase C delta(PKC120575) is a serinethreonine kinase which plays a central rolein apoptosis Previously we identified a new splice variantof human PKC120575 PKC120575VIII (Figure 2(a)) Sequencing andcomputational analysis of the PKC120575VIII sequence indicatedthat this human splice variant is generated by utilization ofan alternative downstream 51015840 splice site of PKC120575 pre-mRNAexon 10 Our data indicated that PKC120575I promotes apoptosiswhile PKC120575VIII promotes cell survival [8]

34 Alternative Splicing of Bcl-x Bcl-x is a dominant regu-lator of programmed cell death in mammalian cells Bcl-x is

4 ISRN Obesity

lowastlowastlowast

lowastlowastlowast

60

50

40

30

20

10

0

Ann

exin

PI

Day 0 Day 10

ControlSerum deprived

lowastlowastlowast

Figure 1 Human lean preadipocytes were differentiated in vitroCells were serum deprived for 48 h on day 0 (preadipocytes) andon day 10 (mature adipocytes) AnnexinPI was analyzed by flowcytometry Experiments were repeated 5 times Statistical analysisperformed by two-way ANOVA 119875 gt 075 ns not significant withingroup lowastlowastlowast119875 lt 00001 highly significant between control and serum-deprived lowastlowastlowast119875 lt 00001 highly significant between days 0 and 10serum-deprived

alternatively spliced to Bcl-xL Bcl-xS and Bcl-x120573 Bcl-xL isgenerated via alternate 51015840 splice site selection on exon 2 ofBcl-x pre-mRNA and includes the BH1 and BH2 domains(Figure 2(b)) Bcl-x120573 is the least studied and has no knownrole in apoptosis The long form Bcl-xL inhibits apoptosiswhile the short isoform Bcl-xS promotes apoptosis Bcl2 andBcl-xL dimerize to initiate the cell survival pathways

35 Alternative Splicing of Caspase 9 Caspase 9 an initiatorcaspase is a ubiquitously expressed protease It representsa pivotal signaling protein in the apoptotic cascade Inhibi-tion of Caspase 9 by a chemical inhibitor mollugin affectsadipogenesis [10 11] It is alternatively spliced to caspase 9a(apoptotic) and caspase 9b (antiapoptosis) via inclusion ofthe cassette exons 3 4 5 and 6 in caspase 9a (Figure 2(c))Caspase 9a activates caspase 3 which cleaves substratesthereby mediating apoptosis

To verify our results from the exon array we carriedout a detailed analysis of the differentiation of humanpreadipocytes from days 0 to 10 as they differentiate intomature adipocytes as described Our data indicates thatbetween days 2 and 4 of differentiation a marked shift isobserved in the expression of genes involved in prosurvivalpathways (Figure 2(d)) Day 6 of differentiation is markedby the splicing of antiapoptotic proteins that advances the

mature adipocytes to be resistant to apoptosis On day 0cells expressed the apoptotic splice variants PKC120575I caspase9a and lower levels of Bcl-xL By day 4 cells startedexpressing the antiapoptotic proteins Bcl2 and splice variantsPKC120575VIII caspase 9b and Bcl-xL with a concurrent decreasein proapoptotic proteins This represents a critical switchin adipocyte differentiation modulated by the splicing ofapoptosis genes

36 Expression of PKC120575 Splice Variants in Obese PatientsOur above data was obtained in normal lean differentiatingpreadipocytes Next we sought to determine whether thesplicing pattern and expression of these genes was affectedby the metabolic state Hence we sought to measure thelevels of PKC120575 from obese patients We obtained proteinlysates from subcutaneous and omental adipose tissues froman obese patient undergoing Roux-en-y gastroenterostomysurgery (from Drs Murr and Cooper Tampa VAMC) Thesamples were from an obese patient closely matching theBMI and other criteria to the preadipocytes obtained fromZenBio to allow for a better comparisonThe results indicateda sharp increase in PKC120575VIII expression in the obese patientcompared to PKC120575I (Figure 3(a)) It is important to note thatthese adipose tissue samples represent a heterogeneous cellpopulation containing predominantlymature adipocytes andalso differentiating preadipocytes as well as other stem cellswhich may or may not have the adipocyte lineage

Next we obtained preadipocytes from the subcutaneousdepot of obese patients (ZenBio) and differentiated them invitro We chose to further study the preadipocyte differentia-tion of obese subcutaneous fraction because our lean cell linewas derived from the subcutaneous fractionThe cell line wasselected to match the adipose tissue lysate (above) such thatwe could evaluate differentiation of obese preadipocytes invitro It also allowed us to compare the expression patterns ofthe genes between the lean and obese during differentiationWe first evaluated apoptosis and susceptibility to serumdeprivation on days 0 and 10 by annexin V and PI flowcytometry Our results (Figures 3(b) and 3(c)) indicatedthat obese preadipocytes are more resistant to apoptosiscompared to lean on day 0 (UR quadrant with both AVand PI lean 307 obese 167) Day 10 mature adipocytesdemonstrate much lower susceptibility to serum-deprivedapoptosis (lean 187 obese 104)

To determine the splicing pattern of differentiating obesepreadipocytes we performed Western blot analysis Weobserved that these cells expressed PKC120575VIII Bcl-xL Bcl2and caspase 9 on day 0 that is in the preadipocyte stage Weobserved increased expression of PKC120575VIII as determined bythe PKC120575 splice variant ratio observed on day 0 (Figure 4)This data suggests that the early expression of these genesin the adipocytes promotes their survival mechanisms andthereby provides resistance to cell death

37 Adipogenesis in Obesity Is Dysregulated The lean andobese human preadipocytes were differentiated for 0 to14 days and adipogenesis was observed by staining of thelipid droplets by Oil Red O The lean cells showed lipid

ISRN Obesity 5

PKC120575VIIImRNA E9 E10 E11

Cell survivalantiapoptosis

PKC120575pre-mRNA

E9 E10 E11

SS II

SS I

PKC120575ImRNA

E9 E10 E11 Cell apoptosis

(a)

Bcl-xLmRNA E1 E2 E3

Cell survivalantiapoptosis

Bcl-xpre-mRNA

E1 E2 E3SS II

SS I

Bcl-xSmRNA

E1 E2 E3 Cell apoptosis

(b)

Caspase 9amRNA Cell apoptosis

Caspase 9Pre-mRNA

Exon 2 Exons 3 4 5 6Cassette Exon 7

Caspase 9bmRNA

Cell survivalantiapoptosis

(c)

Day

PKC120575VIIIPKC120575I

Bcl-2

Bcl-xL

Caspase 9aCaspase 9b

0 2 4 6 8 10 0 2 4 6 8 10

TNF120572

XIAP

Adiponectin

PPAR

120573-actin

60

50

40

30

20

10

0Day 0 Day 2 Day 4 Day 6 Day 8 Day 10

120574

PKC120575

exon

inclu

sion

()

(d)

Figure 2 (a) Schematic of alternative splicing of PKC120575 pre-mRNA generating PKC120575I mRNA and PKC120575VIII mRNA via alternative 51015840 splicesite selection E exon SS splice site (b) Schematic of alternative splicing of Bcl-x pre-mRNA generating Bcl-xL mRNA and Bcl-xS mRNAvia alternative 51015840 splice site selection E exon SS splice site (c) Schematic of alternative splicing of caspase 9 pre-mRNA generating caspase9a mRNA and caspase 9b mRNA via cassette exon inclusion (d) Western blot analysis of differentiating lean preadipocytes from days 0 to 10using antibodies as indicated in the figure Between days 2 and 4 a marked shift is observed in the splicing pattern of survival proteins Thisperiodmarks terminal differentiation of adipocytesThe blots are representative of 3 experiments performed individually with similar resultsGraph represents percent PKC120575 exon inclusion calculated as PKC120575VIII(PKC120575VIII + PKC120575I) times 100 and is representative of four experimentsperformed separately

accumulation beginning at day 4 We observed that obesepreadipocyteswere slower to differentiate but exhibited largerlipid droplets by day 10 We measured and compared themature adipocyte size on day 14 between lean and obesesamples (Figure 5) Our results indicate a dramatic increasein adipocyte size in obese samples Further the expressionof adiponectin and PPAR120574 was dysregulated in differen-tiating obese preadipocytes (Figure 4) which may explainthe aberrant differentiation and adipogenesis observed in

obese patients These results indicated that the adipogenesisprogram was dysregulated in obese adipocytes

To evaluate this we performed real-time qPCR analysisto measure the expression of adipogenesis marker genesFOXO SIRT1 adiponectin and PPAR120574 with a concurrentmeasure of PKC120575VIII expression Total RNA was isolatedon day 2 of differentiation of lean and obese preadipocytesOur results (Figure 6) indicated that PKC120575VIII expressionwas increased dramatically in obese samples along with a

6 ISRN Obesity

E01 Unstained controlGate [no gating]

16777215

10000000

5000000

0

SSC-

A

0 5000000 10000000 16777

FSC-A

Day 10

1072

106

105

104

103

102

PI-A

103 104 105 106 1072

Annexin V-A

F04 Lean-controlGate (P1 in all)

1072

106

105

104

103

102

PI-A

103 104 105 106 1072

Annexin V-A

1072

106

105

104

103

102

PI-A

103 104 105 106 1072

Annexin V-A

1034 104 105 106 1072

Annexin V-A

1072

106

105

104

1027

PI-A

1034 104 105 106 1072

Annexin V-A

G06 Lean-serum deprived

H05 Obese-controlGate (P1 in all)

H07 Obese-serum deprivedGate (P1 in all)

E04 Lean-controlGate (P1 in all)

E06 Lean-serum deprivedGate (P1 in all)

E12 Obese-controlGate (P1 in all)

F05 Obese-serum deprivedGate (P1 in all)

Gate (P1 in all)

1072

106

105

104

1027PI

-A

1072

106

105

104

103

102

PI-A

103 104 105 106 1072

Annexin V-A

1034 104 105 106 1072

Annexin V-A

1072

106

105

104

1027

PI-A

1034 104 105 106 1072

Annexin V-A

1072

106

105

104

1027

PI-A

F01 Unstained controlGate [no gating]

16777215

10000000

5000000

0

SSC-

A0 5000000 10000000 16777

FSC-A

Day 0

lowastlowastlowast

lowastlowastlowast lowastlowastlowast

35

30

25

20

15

10

5

0

Ann

exin

VP

I

Leancontrol

Leanserum

deprivedObesecontrol

Obeseserum

deprived

Day 0Day 10

lowastlowastlowastlowastlowastlowast

(a)

(b)

(c)

1 2

PKC120575VIIIPKC120575I

120573-actin

80

60

40

20

0Omental Subcutaneous

PKC120575

exon

inclu

sion

()

Figure 3 (a) Western blot analysis of adipose tissue lysates obtained from obese depots from (1) omental (2) subcutaneous Theblots are representative of experiments repeated thrice from the lysates Graph represents percent PKC120575 exon inclusion calculated asPKC120575VIII(PKC120575VIII + PKC120575I)times 100 and is representative of three experiments (b)Human lean and obese preadipocytes were differentiatedin vitro Cells were serum deprived for 48 h on day 0 (preadipocytes) and on day 10 (mature adipocytes) Annexin VPI staining was analyzedby flow cytometryThe first plot demonstrates the gating of the adipocyte populationThe remaining plots shown are lean control lean-serumdeprived obese control and obese serum-deprived and are divided into four quadrants based on Annexin V (AV) and propidium iodidestaining (PI) staining Q1-LL represents viable (unstained) cells Q1-LR represents AV staining only (early apoptosis) Q1-UR represents cellsstainedwith bothAVandPI (later apoptosis) Q1-UL represents cells stainedwith PI only (end-stage apoptosis) Data shown are representativeof five different experiments (c) Graph represents AVPI and represents 5 experiments Statistical analysis performed by two-way ANOVA119875 gt 075 ns not significant within group lowastlowastlowast119875 lt 00001 highly significant between control and serum-deprived lowastlowastlowast119875 lt 00001 highlysignificant between lean and obese lowastlowastlowast119875 lt 00001 highly significant between days 0 and 10 serum-deprived

ISRN Obesity 7

0 2 4 6 8 10Day 0 2 4 6 8 10

PKC120575VIIIPKC120575I

Bcl-2

Bcl-xL

Caspase 9a

Caspase 9b

Glut4

Adiponectin

PPAR120574

120573-actin

60

50

40

30

20

10

0Day 0 Day 2 Day 4 Day 6 Day 8 Day 10

PKC120575

exon

inclu

sion

()

Figure 4 Western blot analysis of differentiating subcutaneousobese preadipocytes from days 0 to 10 using antibodies as indicatedThe blots are representative of four experiments performed indi-vidually with similar results Graph represents percent PKC120575 exoninclusion calculated as PKC120575VIII(PKC120575VIII + PKC120575I) times 100 and isrepresentative of four experiments performed separately

Lean Obese

(a)

40000

35000

30000

25000

20000

15000

10000

5000

0Lean Obese

Mea

n ad

ipoc

yte s

ize (120583

m)

(b)

Figure 5 Human lean or obese preadipocytes were differentiated invitro (a) Adipocytes on day 14 were stained for lipid content withOil Red O and (b) adipocyte size measured (methods) Experimentswere repeated 4 times with similar result

25

2

15

1

05

0

mRN

A ex

pres

sion

lowast

lowast

lowast

lowast

lowast

SIRT1 FOXO1 Adiponectin

LeanObese

PKC120575VIII PPAR120574

Figure 6 Lean and obese human preadipocytes were differentiatedin vitro Total RNA was collected on day 2 Real Time SYBR qPCRanalysis of PKC120575VIII SIRT1 FOXO1 adiponectin and PPAR120574was performed in triplicate and repeated 3 times in separateexperimentsThe absolute mRNA expression transcripts are shownlowast119875 lt 005 (by two-tailed Studentrsquos 119905-test) significant between leanand obese for each mRNA

marked decrease in FOXO SIRT1 adiponectin and PPAR120574expression compared to lean samples

4 Discussion

Obesity is an epidemic facing the world with the USAleading the statistics on increasing number of overweight andobese population Adipose tissue is an important endocrineregulator of energy homeostasis and glucose metabolismand the excess energy is stored in the white adipose tissueThe onset of adipogenesis and the transcriptional factorsthat initiate it have been studied in depth However withour current knowledge of the genome it is evident thatthere are several other mechanisms in play which regulatethe adipocyte phenotype during differentiation Apoptosis isrequired for fat cell turnover throughout our life Howeverfor weight loss in obese patients it may be beneficial toincrease adipocyte apoptosis along with therapies targetinghyperplasia and hypertrophy in adipocytes To our knowl-edge this is the first demonstration of alternative splicingevents during adipogenesis which contribute to the matureadipocyte phenotype Our results indicate a distinct splicingpattern between lean and obese patients

Apoptosis in preadipocytes and adipocytes was previ-ously observed in human cell lines However the under-lying mechanisms were not described Alternative splicinga posttranscriptional event is a powerful regulator of geneexpression We have identified key genes which are alter-natively spliced that dramatically switch their splice variantexpression during the progression of adipogenesis This mayexplain why preadipocytes are more susceptible to apoptoticagents while the mature adipocytes are resistant to it Protein

8 ISRN Obesity

kinase C delta (PKC120575) is a crucial signaling kinase affectingdownstream apoptotic pathways [8 12] Our data indicatesincreased expression of PKC120575VIII along with an early switchbetween the splice variants associated with apoptosis duringadipogenesis process of preadipocytes derived from obesepatients

PKC120575 is a serinethreonine kinase which plays a centralrole in apoptosis PKC120575 has dual functions as a mediator ofapoptosis and as an antiapoptosis effector Its splice variantsfunction as a switch that determines cell survival and fateThis could explain the opposing effects of PKC120575 on cellularapoptosis cited in the literature [13ndash15] Genetic switchesbased on alternative splicing are important for many cellularand developmental processes PKC120575I promotes apoptosiswhile PKC120575VIII promotes survival We have shown thatPKC120575II (the mouse splice variant) and PKC120575VIII functionas prosurvival proteins [8 12] the functions of other PKC120575splice variants are not yet established Our previous studiesin neurons have demonstrated that the prosurvival variant ofPKC120575 increases the expression of Bcl2 and Bcl-xL which areknown to increase cellular survival [16]

Obesity is associated with a chronic low inflamma-tory state due to the changes in secretory functions ofthe adipocytes Enlarged adipocytes (hypertrophy) lead toincreased adipocyte-derived free fatty acids which stimulatemacrophages In severely obese patientsmacrophages invadethe adipose tissue and release chemokines leading to necrosisand autophagy of adipocytes [17] These complications addto obese-associated metabolic syndrome diabetes insulinresistance and hepatic steatosis These forms of cell deathare distinct from apoptosis (programmed cell death) inadipocytes Apoptosis is shown to be substantially greaterin omental depots compared to subcutaneous depots inhealthy individuals [18] Here we show that obese samplesare resistant to apoptosis and express the antiapoptosis genesfourfold greater than in lean samples This may explain theincreased difficulty to lose weight and maintain the weightloss in overweight and obese individuals Larger adipocytesare an indicator of type II diabetes independent of insulinresistance [19] Previous studies showed that PPAR120574 andadiponectin expression was lower in obesity and type IIdiabetes [20 21] Our data indicated that obese patients havedysregulated adipogenesis program This may be modulatedby the signaling kinase PKC120575VIII which is overexpressedin obese individuals Our data also indicated dysregulatedalternative splicing of apoptosis genes in obese differentiatingcells This aberrant molecular profile may be a causativeeffect leading to obesity or may be a consequence of obesityand its comorbidities We are evaluating this further in ourlaboratory Here we have defined the differences in geneexpression between lean and obese adipocytes

In conclusion it may be possible to modulate alternativesplicing during adipogenesis for therapeutic treatments ofobesity Our results suggest a novel role for a key signalingkinase PKC120575VIII in the process of adipogenesis Expressionlevels of PKC120575VIII splice variant are increased in obesity andmay be linked to the cause or consequence of dysregulatedadipogenesis in obese patients Since new adipocytes arerequired for the maintenance and the endocrine function of

adipocytes balancing the gene expression of apoptotic genesmay be a better approach rather than complete inhibition ofadipogenesis

Conflict of Interests

The authors have no conflict of interests to declare

Authorsrsquo Contribution

GCarter andAApostolatos contributed equally to the paper

Acknowledgments

This work was supported by the Department of VeteransAffairs Medical Research Grant (NAP)The contents do notrepresent the views of the Department of Veterans Affairs ortheUnited States GovernmentThe authors thankDr AndreaMoor for help with microscopic images and adipocyte sizedetermination They thank Sean Yoder from Moffitt arraycore for array analysis

References

[1] S Rodrıguez-Acebes N Palacios J I Botella-Carretero et alldquoGene expression profiling of subcutaneous adipose tissue inmorbid obesity using a focused microarray distinct expressionof cell-cycle- and differentiation-related genesrdquo BMC MedicalGenomics vol 3 article 61 2010

[2] M Guzey D Jukic J Arlotti M Acquafondata R Dhir andR H Getzenberg ldquoIncreased apoptosis of periprostatic adiposetissue in VDR null micerdquo Journal of Cellular Biochemistry vol93 no 1 pp 133ndash141 2004

[3] P Fischer-Posovszky H Tornqvist K M Debatin and MWabitsch ldquoInhibition of death-receptor mediated apoptosis inhuman adipocytes by the insulin-like growth factor I (IGF-I)IGF-I receptor autocrine circuitrdquo Endocrinology vol 145 no4 pp 1849ndash1859 2004

[4] O Ishiko T Sumi H Yoshida Y Hyun and S Ogita ldquoCom-parison of expression of apoptosis regulatory proteins in theadipose tissue of tumor-bearing and diet-restricted rabbitsrdquoInternational Journal of Molecular Medicine vol 8 no 5 pp543ndash547 2001

[5] M A Della-Fera Y H Choi D L Hartzell J Duan MHamrick and C A Baile ldquoSensitivity of obob mice to Leptin-induced adipose tissue apoptosisrdquo Obesity Research vol 13 no9 pp 1540ndash1547 2005

[6] D H Kim S C Woods and R J Seeley ldquoPeptide designedto elicit apoptosis in adipose tissue endothelium reduces foodintake and body weightrdquo Diabetes vol 59 no 4 pp 907ndash9152010

[7] W C Earnshaw L M Martins and S H Kaufmann ldquoMam-malian caspases structure activation substrates and functionsduring apoptosisrdquo Annual Review of Biochemistry vol 68 pp383ndash424 1999

[8] K Jiang A H Apostolatos T Ghansah et al ldquoIdentificationof a novel antiapoptotic human protein kinase C 120575 isoformPKC120575VIII in NT2 cellsrdquo Biochemistry vol 47 no 2 pp 787ndash797 2008

ISRN Obesity 9

[9] H Apostolatos A Apostolatos T Vickers et al ldquoVitamin Ametabolite all-trans-retinoic acid mediates alternative splicingof protein kinase C 120575VIII (PKC120575VIII) isoform via splicingfactor SC35rdquo Journal of Biological Chemistry vol 285 no 34pp 25987ndash25995 2010

[10] D Y Jun C R Han M S Choi M A Bae M H Woo and YH Kim ldquoEffect of mollugin on apoptosis and adipogenesis of3T3-L1 preadipocytesrdquo Phytotherapy Research vol 25 no 5 pp724ndash731 2011

[11] S M Kim H S Park D Y Jun et al ldquoMollugin induces apop-tosis in human Jurkat T cells through endoplasmic reticulumstress-mediated activation of JNK and caspase-12 and subse-quent activation of mitochondria-dependent caspase cascaderegulated by Bcl-xLrdquo Toxicology and Applied Pharmacology vol241 no 2 pp 210ndash220 2009

[12] N A Patel S S Song and D R Cooper ldquoPKC120575 alternativelyspliced isoforms modulate cellular apoptosis in retinoic acid-induced differentiation of human NT2 cells and mouse embry-onic stem cellsrdquo Gene Expression vol 13 no 2 pp 73ndash84 2005

[13] J J Peluso A Pappalardo and G Fernandez ldquoBasic fibrob-last growth factor maintains calcium homeostasis and gran-ulosa cell viability by stimulating calcium efflux via a PKC120575-dependent pathwayrdquo Endocrinology vol 142 no 10 pp 4203ndash4211 2001

[14] L E Kilpatrick J Y Lee K M Haines D E Campbell KE Sullivan and H M Korchak ldquoA role for PKC-120575 and PI3-kinase in TNF-120572-mediated antiapoptotic signaling in thehuman neutrophilrdquo The American Journal of Physiology vol283 no 1 pp C48ndashC57 2002

[15] A Zrachia M Dobroslav M Blass et al ldquoInfection of gliomacells with Sindbis virus induces selective activation and tyrosinephosphorylation of protein kinase C 120575 implications for sindbisvirus-induced apoptosisrdquo Journal of Biological Chemistry vol277 no 26 pp 23693ndash23701 2002

[16] A Apostolatos S Song S Acosta et al ldquoInsulin promotesneuronal survival via the alternatively spliced protein kinaseCdeltaII isoformrdquo Journal of Biological Chemistry vol 287 pp9299ndash9310 2012

[17] K J Strissel Z Stancheva H Miyoshi et al ldquoAdipocytedeath adipose tissue remodeling and obesity complicationsrdquoDiabetes vol 56 no 12 pp 2910ndash2918 2007

[18] D Papineau A Gagnon and A Sorisky ldquoApoptosis of humanabdominal preadipocytes before and after differentiation intoadipocytes in culturerdquo Metabolism vol 52 no 8 pp 987ndash9922003

[19] C Weyer J E Foley C Bogardus P A Tataranni and RE Pratley ldquoEnlarged subcutaneous abdominal adipocyte sizebut not obesity itself predicts type II diabetes independent ofinsulin resistancerdquo Diabetologia vol 43 no 12 pp 1498ndash15062000

[20] L B Tanko A Siddiq C Lecoeur et al ldquoACDCadiponectinand PPAR-120574 gene polymorphisms implications for features ofobesityrdquo Obesity Research vol 13 no 12 pp 2113ndash2121 2005

[21] N Ouchi S Kihara T Funahashi Y Matsuzawa and KWalshldquoObesity adiponectin and vascular inflammatory diseaserdquo Cur-rent Opinion in Lipidology vol 14 no 6 pp 561ndash566 2003

ISRN Obesity 3

standard curve was generated for each gene in every assayTo do so 100ndash04 ng of RNA was reverse transcribed asdescribed above The resulting cDNA was used to obtain astandard curve correlating the amounts with the thresholdcycle number (Ct values) A linear relationship (1199032 gt096) was obtained for each gene Real-time PCR was thenperformed on samples and standards in triplicate The platesetup also included a standard series no template controlno RNA control no reverse transcriptase control and noamplification control The dissociation curve was analyzedfor each sample Absolute quantification ofmRNAexpressionlevels for individual transcriptswas calculated bynormalizingthe values to GAPDH The results were analyzed with atwo-tailed Studentrsquos 119905-test using PRISM4 statistical analysissoftware (GraphPad) A level of 119875 lt 005 was consideredstatistically significant Significance was determined afterthree or more experiments

24 Western Blot Analysis Protein lysates were obtainedfrom the cells using lysis buffer containing proteaseinhibitors Protein lysates were also harvested from the snap-frozen adipose tissues by homogenization and sonicationin the lysis buffer The lysates (40 120583g) were separated bySDS-PAGE with 10 gels electrophoretically transferredto nitrocellulose membranes blocked with Tris-bufferedsaline containing 01 Tween 20 and 5 nonfat dried milkwashed and incubated with anti-PKC120575 or antibody specificfor PKC120575VIII (Patel lab [8]) anti-PPAR120574 anti-adiponectinanti-Bcl2 anti-caspase 9 (Cell Signaling) anti-XIAP(AnaSpec) anti-Bcl-xL (Santa Cruz Biotechnology) and120573-actin (Sigma) After incubation with antirabbit IgG-HRP enhanced chemiluminescence (Pierce) was used fordetection

25 Adipocyte Size Adipocytes were differentiated in100mm plate and their sizes determined on day 14 (ieof mature adipocytes) Maximal diameter of 10 adjacentadipocytes from 6 different fields was calculated using NikonEclipse inverted microscope and NIS elements advanceresearch image tool software The data was transferred toExcel to calculate the mean diameter and standard deviationThis was repeated in three separate experiments to ensurereproducibility

26 Apoptosis Assay Human preadipocytes were culturedon 60mm dishes as described in methods For apoptosisassays cells were serum starved for 48 hours Media werecollected and cells were washed one time with HBSS andthen trypsinized for 50 minutes Five ml complete mediawas added to neutralize the trypsin Media and washes werepooled and centrifuged at 1200 RPMS for five minutes Cellswere washed one time with PBS and one time with bindingbuffer and then incubated for 15min with 50 120583L AV-FITCand 50 120583L PI in 100 uL binding buffer (BD PharmagenSan Diego CA USA) at room temperature in the dark400120583L binding buffer was added and cells were analyzedby flow cytometry within one hour Annexin V-FITC and

PI fluorescence were measured using an Accuri C6 flowcytometer

27 Statistical Analysis Analyses were performed usingPRISM software and analyzed using either two-tailed Stu-dentrsquos 119905-test or two-way ANOVA lowast119875 lt 005 was signifi-cant lowastlowastlowast119875 lt 00001 was highly significant Analysis wasperformed within group and between groups

3 Results

31 Human Preadipocytes Are Susceptible to ApoptosisHuman preadipocytes were commercially obtained fromZenBio such that a pure population of preadipocytesdevoid of contaminating stem cells or adipocytes or othernonadipocyte lineage cells were used in our experimentsThecells were cultured as described inmethods Cells were serumstarved for 48 hours to induce apoptosis and assayed by flowcytometry on days 0 and 10 FITC Annexin V was used toquantitatively assess cells undergoing apoptosis along withpropidium iodide to enable detection of percentage of cellsundergoing either early apoptosis or late apoptosis AnnexinV binds to phosphatidylserine which is displayed on the cellmembrane of apoptotic cells and PI will stain only dead ordamaged cellsThe results (Figure 1) demonstrate that on day0 preadipocytes which are serum starved undergo increasedapoptosis while the mature adipocytes (day 10) are moreresistant to apoptosis upon serum starvation

32 Differentiation of Human Preadipocytes DemonstratesSwitch in Apoptosis Genes Since the adipocytes demon-strated a switch in the susceptibility to apoptosis duringdifferentiation we sought to elucidate the expression patternsof apoptotic genes during differentiation We evaluated thealternative splicing of genes during differentiation with afocus on apoptotic genes We used gene-level exon arrayfrom Affymetrix (experiments performed by MicroarrayCore Facility Moffitt Cancer Institute and analyzed withExon array analyzer software to sort specific biochemicalpathways and genes) to identify a group of transcripts that arepredominantly alternatively spliced and expressed We useddays 0 6 and 8 of differentiating human preadipocytes (datanot shown) The splicing pattern of genes of the apoptosispathway that showed a dramatic change between days 0 and6 of adipogenesis is explained below

33 Alternative Splicing of PKC120575 Protein kinase C delta(PKC120575) is a serinethreonine kinase which plays a central rolein apoptosis Previously we identified a new splice variantof human PKC120575 PKC120575VIII (Figure 2(a)) Sequencing andcomputational analysis of the PKC120575VIII sequence indicatedthat this human splice variant is generated by utilization ofan alternative downstream 51015840 splice site of PKC120575 pre-mRNAexon 10 Our data indicated that PKC120575I promotes apoptosiswhile PKC120575VIII promotes cell survival [8]

34 Alternative Splicing of Bcl-x Bcl-x is a dominant regu-lator of programmed cell death in mammalian cells Bcl-x is

4 ISRN Obesity

lowastlowastlowast

lowastlowastlowast

60

50

40

30

20

10

0

Ann

exin

PI

Day 0 Day 10

ControlSerum deprived

lowastlowastlowast

Figure 1 Human lean preadipocytes were differentiated in vitroCells were serum deprived for 48 h on day 0 (preadipocytes) andon day 10 (mature adipocytes) AnnexinPI was analyzed by flowcytometry Experiments were repeated 5 times Statistical analysisperformed by two-way ANOVA 119875 gt 075 ns not significant withingroup lowastlowastlowast119875 lt 00001 highly significant between control and serum-deprived lowastlowastlowast119875 lt 00001 highly significant between days 0 and 10serum-deprived

alternatively spliced to Bcl-xL Bcl-xS and Bcl-x120573 Bcl-xL isgenerated via alternate 51015840 splice site selection on exon 2 ofBcl-x pre-mRNA and includes the BH1 and BH2 domains(Figure 2(b)) Bcl-x120573 is the least studied and has no knownrole in apoptosis The long form Bcl-xL inhibits apoptosiswhile the short isoform Bcl-xS promotes apoptosis Bcl2 andBcl-xL dimerize to initiate the cell survival pathways

35 Alternative Splicing of Caspase 9 Caspase 9 an initiatorcaspase is a ubiquitously expressed protease It representsa pivotal signaling protein in the apoptotic cascade Inhibi-tion of Caspase 9 by a chemical inhibitor mollugin affectsadipogenesis [10 11] It is alternatively spliced to caspase 9a(apoptotic) and caspase 9b (antiapoptosis) via inclusion ofthe cassette exons 3 4 5 and 6 in caspase 9a (Figure 2(c))Caspase 9a activates caspase 3 which cleaves substratesthereby mediating apoptosis

To verify our results from the exon array we carriedout a detailed analysis of the differentiation of humanpreadipocytes from days 0 to 10 as they differentiate intomature adipocytes as described Our data indicates thatbetween days 2 and 4 of differentiation a marked shift isobserved in the expression of genes involved in prosurvivalpathways (Figure 2(d)) Day 6 of differentiation is markedby the splicing of antiapoptotic proteins that advances the

mature adipocytes to be resistant to apoptosis On day 0cells expressed the apoptotic splice variants PKC120575I caspase9a and lower levels of Bcl-xL By day 4 cells startedexpressing the antiapoptotic proteins Bcl2 and splice variantsPKC120575VIII caspase 9b and Bcl-xL with a concurrent decreasein proapoptotic proteins This represents a critical switchin adipocyte differentiation modulated by the splicing ofapoptosis genes

36 Expression of PKC120575 Splice Variants in Obese PatientsOur above data was obtained in normal lean differentiatingpreadipocytes Next we sought to determine whether thesplicing pattern and expression of these genes was affectedby the metabolic state Hence we sought to measure thelevels of PKC120575 from obese patients We obtained proteinlysates from subcutaneous and omental adipose tissues froman obese patient undergoing Roux-en-y gastroenterostomysurgery (from Drs Murr and Cooper Tampa VAMC) Thesamples were from an obese patient closely matching theBMI and other criteria to the preadipocytes obtained fromZenBio to allow for a better comparisonThe results indicateda sharp increase in PKC120575VIII expression in the obese patientcompared to PKC120575I (Figure 3(a)) It is important to note thatthese adipose tissue samples represent a heterogeneous cellpopulation containing predominantlymature adipocytes andalso differentiating preadipocytes as well as other stem cellswhich may or may not have the adipocyte lineage

Next we obtained preadipocytes from the subcutaneousdepot of obese patients (ZenBio) and differentiated them invitro We chose to further study the preadipocyte differentia-tion of obese subcutaneous fraction because our lean cell linewas derived from the subcutaneous fractionThe cell line wasselected to match the adipose tissue lysate (above) such thatwe could evaluate differentiation of obese preadipocytes invitro It also allowed us to compare the expression patterns ofthe genes between the lean and obese during differentiationWe first evaluated apoptosis and susceptibility to serumdeprivation on days 0 and 10 by annexin V and PI flowcytometry Our results (Figures 3(b) and 3(c)) indicatedthat obese preadipocytes are more resistant to apoptosiscompared to lean on day 0 (UR quadrant with both AVand PI lean 307 obese 167) Day 10 mature adipocytesdemonstrate much lower susceptibility to serum-deprivedapoptosis (lean 187 obese 104)

To determine the splicing pattern of differentiating obesepreadipocytes we performed Western blot analysis Weobserved that these cells expressed PKC120575VIII Bcl-xL Bcl2and caspase 9 on day 0 that is in the preadipocyte stage Weobserved increased expression of PKC120575VIII as determined bythe PKC120575 splice variant ratio observed on day 0 (Figure 4)This data suggests that the early expression of these genesin the adipocytes promotes their survival mechanisms andthereby provides resistance to cell death

37 Adipogenesis in Obesity Is Dysregulated The lean andobese human preadipocytes were differentiated for 0 to14 days and adipogenesis was observed by staining of thelipid droplets by Oil Red O The lean cells showed lipid

ISRN Obesity 5

PKC120575VIIImRNA E9 E10 E11

Cell survivalantiapoptosis

PKC120575pre-mRNA

E9 E10 E11

SS II

SS I

PKC120575ImRNA

E9 E10 E11 Cell apoptosis

(a)

Bcl-xLmRNA E1 E2 E3

Cell survivalantiapoptosis

Bcl-xpre-mRNA

E1 E2 E3SS II

SS I

Bcl-xSmRNA

E1 E2 E3 Cell apoptosis

(b)

Caspase 9amRNA Cell apoptosis

Caspase 9Pre-mRNA

Exon 2 Exons 3 4 5 6Cassette Exon 7

Caspase 9bmRNA

Cell survivalantiapoptosis

(c)

Day

PKC120575VIIIPKC120575I

Bcl-2

Bcl-xL

Caspase 9aCaspase 9b

0 2 4 6 8 10 0 2 4 6 8 10

TNF120572

XIAP

Adiponectin

PPAR

120573-actin

60

50

40

30

20

10

0Day 0 Day 2 Day 4 Day 6 Day 8 Day 10

120574

PKC120575

exon

inclu

sion

()

(d)

Figure 2 (a) Schematic of alternative splicing of PKC120575 pre-mRNA generating PKC120575I mRNA and PKC120575VIII mRNA via alternative 51015840 splicesite selection E exon SS splice site (b) Schematic of alternative splicing of Bcl-x pre-mRNA generating Bcl-xL mRNA and Bcl-xS mRNAvia alternative 51015840 splice site selection E exon SS splice site (c) Schematic of alternative splicing of caspase 9 pre-mRNA generating caspase9a mRNA and caspase 9b mRNA via cassette exon inclusion (d) Western blot analysis of differentiating lean preadipocytes from days 0 to 10using antibodies as indicated in the figure Between days 2 and 4 a marked shift is observed in the splicing pattern of survival proteins Thisperiodmarks terminal differentiation of adipocytesThe blots are representative of 3 experiments performed individually with similar resultsGraph represents percent PKC120575 exon inclusion calculated as PKC120575VIII(PKC120575VIII + PKC120575I) times 100 and is representative of four experimentsperformed separately

accumulation beginning at day 4 We observed that obesepreadipocyteswere slower to differentiate but exhibited largerlipid droplets by day 10 We measured and compared themature adipocyte size on day 14 between lean and obesesamples (Figure 5) Our results indicate a dramatic increasein adipocyte size in obese samples Further the expressionof adiponectin and PPAR120574 was dysregulated in differen-tiating obese preadipocytes (Figure 4) which may explainthe aberrant differentiation and adipogenesis observed in

obese patients These results indicated that the adipogenesisprogram was dysregulated in obese adipocytes

To evaluate this we performed real-time qPCR analysisto measure the expression of adipogenesis marker genesFOXO SIRT1 adiponectin and PPAR120574 with a concurrentmeasure of PKC120575VIII expression Total RNA was isolatedon day 2 of differentiation of lean and obese preadipocytesOur results (Figure 6) indicated that PKC120575VIII expressionwas increased dramatically in obese samples along with a

6 ISRN Obesity

E01 Unstained controlGate [no gating]

16777215

10000000

5000000

0

SSC-

A

0 5000000 10000000 16777

FSC-A

Day 10

1072

106

105

104

103

102

PI-A

103 104 105 106 1072

Annexin V-A

F04 Lean-controlGate (P1 in all)

1072

106

105

104

103

102

PI-A

103 104 105 106 1072

Annexin V-A

1072

106

105

104

103

102

PI-A

103 104 105 106 1072

Annexin V-A

1034 104 105 106 1072

Annexin V-A

1072

106

105

104

1027

PI-A

1034 104 105 106 1072

Annexin V-A

G06 Lean-serum deprived

H05 Obese-controlGate (P1 in all)

H07 Obese-serum deprivedGate (P1 in all)

E04 Lean-controlGate (P1 in all)

E06 Lean-serum deprivedGate (P1 in all)

E12 Obese-controlGate (P1 in all)

F05 Obese-serum deprivedGate (P1 in all)

Gate (P1 in all)

1072

106

105

104

1027PI

-A

1072

106

105

104

103

102

PI-A

103 104 105 106 1072

Annexin V-A

1034 104 105 106 1072

Annexin V-A

1072

106

105

104

1027

PI-A

1034 104 105 106 1072

Annexin V-A

1072

106

105

104

1027

PI-A

F01 Unstained controlGate [no gating]

16777215

10000000

5000000

0

SSC-

A0 5000000 10000000 16777

FSC-A

Day 0

lowastlowastlowast

lowastlowastlowast lowastlowastlowast

35

30

25

20

15

10

5

0

Ann

exin

VP

I

Leancontrol

Leanserum

deprivedObesecontrol

Obeseserum

deprived

Day 0Day 10

lowastlowastlowastlowastlowastlowast

(a)

(b)

(c)

1 2

PKC120575VIIIPKC120575I

120573-actin

80

60

40

20

0Omental Subcutaneous

PKC120575

exon

inclu

sion

()

Figure 3 (a) Western blot analysis of adipose tissue lysates obtained from obese depots from (1) omental (2) subcutaneous Theblots are representative of experiments repeated thrice from the lysates Graph represents percent PKC120575 exon inclusion calculated asPKC120575VIII(PKC120575VIII + PKC120575I)times 100 and is representative of three experiments (b)Human lean and obese preadipocytes were differentiatedin vitro Cells were serum deprived for 48 h on day 0 (preadipocytes) and on day 10 (mature adipocytes) Annexin VPI staining was analyzedby flow cytometryThe first plot demonstrates the gating of the adipocyte populationThe remaining plots shown are lean control lean-serumdeprived obese control and obese serum-deprived and are divided into four quadrants based on Annexin V (AV) and propidium iodidestaining (PI) staining Q1-LL represents viable (unstained) cells Q1-LR represents AV staining only (early apoptosis) Q1-UR represents cellsstainedwith bothAVandPI (later apoptosis) Q1-UL represents cells stainedwith PI only (end-stage apoptosis) Data shown are representativeof five different experiments (c) Graph represents AVPI and represents 5 experiments Statistical analysis performed by two-way ANOVA119875 gt 075 ns not significant within group lowastlowastlowast119875 lt 00001 highly significant between control and serum-deprived lowastlowastlowast119875 lt 00001 highlysignificant between lean and obese lowastlowastlowast119875 lt 00001 highly significant between days 0 and 10 serum-deprived

ISRN Obesity 7

0 2 4 6 8 10Day 0 2 4 6 8 10

PKC120575VIIIPKC120575I

Bcl-2

Bcl-xL

Caspase 9a

Caspase 9b

Glut4

Adiponectin

PPAR120574

120573-actin

60

50

40

30

20

10

0Day 0 Day 2 Day 4 Day 6 Day 8 Day 10

PKC120575

exon

inclu

sion

()

Figure 4 Western blot analysis of differentiating subcutaneousobese preadipocytes from days 0 to 10 using antibodies as indicatedThe blots are representative of four experiments performed indi-vidually with similar results Graph represents percent PKC120575 exoninclusion calculated as PKC120575VIII(PKC120575VIII + PKC120575I) times 100 and isrepresentative of four experiments performed separately

Lean Obese

(a)

40000

35000

30000

25000

20000

15000

10000

5000

0Lean Obese

Mea

n ad

ipoc

yte s

ize (120583

m)

(b)

Figure 5 Human lean or obese preadipocytes were differentiated invitro (a) Adipocytes on day 14 were stained for lipid content withOil Red O and (b) adipocyte size measured (methods) Experimentswere repeated 4 times with similar result

25

2

15

1

05

0

mRN

A ex

pres

sion

lowast

lowast

lowast

lowast

lowast

SIRT1 FOXO1 Adiponectin

LeanObese

PKC120575VIII PPAR120574

Figure 6 Lean and obese human preadipocytes were differentiatedin vitro Total RNA was collected on day 2 Real Time SYBR qPCRanalysis of PKC120575VIII SIRT1 FOXO1 adiponectin and PPAR120574was performed in triplicate and repeated 3 times in separateexperimentsThe absolute mRNA expression transcripts are shownlowast119875 lt 005 (by two-tailed Studentrsquos 119905-test) significant between leanand obese for each mRNA

marked decrease in FOXO SIRT1 adiponectin and PPAR120574expression compared to lean samples

4 Discussion

Obesity is an epidemic facing the world with the USAleading the statistics on increasing number of overweight andobese population Adipose tissue is an important endocrineregulator of energy homeostasis and glucose metabolismand the excess energy is stored in the white adipose tissueThe onset of adipogenesis and the transcriptional factorsthat initiate it have been studied in depth However withour current knowledge of the genome it is evident thatthere are several other mechanisms in play which regulatethe adipocyte phenotype during differentiation Apoptosis isrequired for fat cell turnover throughout our life Howeverfor weight loss in obese patients it may be beneficial toincrease adipocyte apoptosis along with therapies targetinghyperplasia and hypertrophy in adipocytes To our knowl-edge this is the first demonstration of alternative splicingevents during adipogenesis which contribute to the matureadipocyte phenotype Our results indicate a distinct splicingpattern between lean and obese patients

Apoptosis in preadipocytes and adipocytes was previ-ously observed in human cell lines However the under-lying mechanisms were not described Alternative splicinga posttranscriptional event is a powerful regulator of geneexpression We have identified key genes which are alter-natively spliced that dramatically switch their splice variantexpression during the progression of adipogenesis This mayexplain why preadipocytes are more susceptible to apoptoticagents while the mature adipocytes are resistant to it Protein

8 ISRN Obesity

kinase C delta (PKC120575) is a crucial signaling kinase affectingdownstream apoptotic pathways [8 12] Our data indicatesincreased expression of PKC120575VIII along with an early switchbetween the splice variants associated with apoptosis duringadipogenesis process of preadipocytes derived from obesepatients

PKC120575 is a serinethreonine kinase which plays a centralrole in apoptosis PKC120575 has dual functions as a mediator ofapoptosis and as an antiapoptosis effector Its splice variantsfunction as a switch that determines cell survival and fateThis could explain the opposing effects of PKC120575 on cellularapoptosis cited in the literature [13ndash15] Genetic switchesbased on alternative splicing are important for many cellularand developmental processes PKC120575I promotes apoptosiswhile PKC120575VIII promotes survival We have shown thatPKC120575II (the mouse splice variant) and PKC120575VIII functionas prosurvival proteins [8 12] the functions of other PKC120575splice variants are not yet established Our previous studiesin neurons have demonstrated that the prosurvival variant ofPKC120575 increases the expression of Bcl2 and Bcl-xL which areknown to increase cellular survival [16]

Obesity is associated with a chronic low inflamma-tory state due to the changes in secretory functions ofthe adipocytes Enlarged adipocytes (hypertrophy) lead toincreased adipocyte-derived free fatty acids which stimulatemacrophages In severely obese patientsmacrophages invadethe adipose tissue and release chemokines leading to necrosisand autophagy of adipocytes [17] These complications addto obese-associated metabolic syndrome diabetes insulinresistance and hepatic steatosis These forms of cell deathare distinct from apoptosis (programmed cell death) inadipocytes Apoptosis is shown to be substantially greaterin omental depots compared to subcutaneous depots inhealthy individuals [18] Here we show that obese samplesare resistant to apoptosis and express the antiapoptosis genesfourfold greater than in lean samples This may explain theincreased difficulty to lose weight and maintain the weightloss in overweight and obese individuals Larger adipocytesare an indicator of type II diabetes independent of insulinresistance [19] Previous studies showed that PPAR120574 andadiponectin expression was lower in obesity and type IIdiabetes [20 21] Our data indicated that obese patients havedysregulated adipogenesis program This may be modulatedby the signaling kinase PKC120575VIII which is overexpressedin obese individuals Our data also indicated dysregulatedalternative splicing of apoptosis genes in obese differentiatingcells This aberrant molecular profile may be a causativeeffect leading to obesity or may be a consequence of obesityand its comorbidities We are evaluating this further in ourlaboratory Here we have defined the differences in geneexpression between lean and obese adipocytes

In conclusion it may be possible to modulate alternativesplicing during adipogenesis for therapeutic treatments ofobesity Our results suggest a novel role for a key signalingkinase PKC120575VIII in the process of adipogenesis Expressionlevels of PKC120575VIII splice variant are increased in obesity andmay be linked to the cause or consequence of dysregulatedadipogenesis in obese patients Since new adipocytes arerequired for the maintenance and the endocrine function of

adipocytes balancing the gene expression of apoptotic genesmay be a better approach rather than complete inhibition ofadipogenesis

Conflict of Interests

The authors have no conflict of interests to declare

Authorsrsquo Contribution

GCarter andAApostolatos contributed equally to the paper

Acknowledgments

This work was supported by the Department of VeteransAffairs Medical Research Grant (NAP)The contents do notrepresent the views of the Department of Veterans Affairs ortheUnited States GovernmentThe authors thankDr AndreaMoor for help with microscopic images and adipocyte sizedetermination They thank Sean Yoder from Moffitt arraycore for array analysis

References

[1] S Rodrıguez-Acebes N Palacios J I Botella-Carretero et alldquoGene expression profiling of subcutaneous adipose tissue inmorbid obesity using a focused microarray distinct expressionof cell-cycle- and differentiation-related genesrdquo BMC MedicalGenomics vol 3 article 61 2010

[2] M Guzey D Jukic J Arlotti M Acquafondata R Dhir andR H Getzenberg ldquoIncreased apoptosis of periprostatic adiposetissue in VDR null micerdquo Journal of Cellular Biochemistry vol93 no 1 pp 133ndash141 2004

[3] P Fischer-Posovszky H Tornqvist K M Debatin and MWabitsch ldquoInhibition of death-receptor mediated apoptosis inhuman adipocytes by the insulin-like growth factor I (IGF-I)IGF-I receptor autocrine circuitrdquo Endocrinology vol 145 no4 pp 1849ndash1859 2004

[4] O Ishiko T Sumi H Yoshida Y Hyun and S Ogita ldquoCom-parison of expression of apoptosis regulatory proteins in theadipose tissue of tumor-bearing and diet-restricted rabbitsrdquoInternational Journal of Molecular Medicine vol 8 no 5 pp543ndash547 2001

[5] M A Della-Fera Y H Choi D L Hartzell J Duan MHamrick and C A Baile ldquoSensitivity of obob mice to Leptin-induced adipose tissue apoptosisrdquo Obesity Research vol 13 no9 pp 1540ndash1547 2005

[6] D H Kim S C Woods and R J Seeley ldquoPeptide designedto elicit apoptosis in adipose tissue endothelium reduces foodintake and body weightrdquo Diabetes vol 59 no 4 pp 907ndash9152010

[7] W C Earnshaw L M Martins and S H Kaufmann ldquoMam-malian caspases structure activation substrates and functionsduring apoptosisrdquo Annual Review of Biochemistry vol 68 pp383ndash424 1999

[8] K Jiang A H Apostolatos T Ghansah et al ldquoIdentificationof a novel antiapoptotic human protein kinase C 120575 isoformPKC120575VIII in NT2 cellsrdquo Biochemistry vol 47 no 2 pp 787ndash797 2008

ISRN Obesity 9

[9] H Apostolatos A Apostolatos T Vickers et al ldquoVitamin Ametabolite all-trans-retinoic acid mediates alternative splicingof protein kinase C 120575VIII (PKC120575VIII) isoform via splicingfactor SC35rdquo Journal of Biological Chemistry vol 285 no 34pp 25987ndash25995 2010

[10] D Y Jun C R Han M S Choi M A Bae M H Woo and YH Kim ldquoEffect of mollugin on apoptosis and adipogenesis of3T3-L1 preadipocytesrdquo Phytotherapy Research vol 25 no 5 pp724ndash731 2011

[11] S M Kim H S Park D Y Jun et al ldquoMollugin induces apop-tosis in human Jurkat T cells through endoplasmic reticulumstress-mediated activation of JNK and caspase-12 and subse-quent activation of mitochondria-dependent caspase cascaderegulated by Bcl-xLrdquo Toxicology and Applied Pharmacology vol241 no 2 pp 210ndash220 2009

[12] N A Patel S S Song and D R Cooper ldquoPKC120575 alternativelyspliced isoforms modulate cellular apoptosis in retinoic acid-induced differentiation of human NT2 cells and mouse embry-onic stem cellsrdquo Gene Expression vol 13 no 2 pp 73ndash84 2005

[13] J J Peluso A Pappalardo and G Fernandez ldquoBasic fibrob-last growth factor maintains calcium homeostasis and gran-ulosa cell viability by stimulating calcium efflux via a PKC120575-dependent pathwayrdquo Endocrinology vol 142 no 10 pp 4203ndash4211 2001

[14] L E Kilpatrick J Y Lee K M Haines D E Campbell KE Sullivan and H M Korchak ldquoA role for PKC-120575 and PI3-kinase in TNF-120572-mediated antiapoptotic signaling in thehuman neutrophilrdquo The American Journal of Physiology vol283 no 1 pp C48ndashC57 2002

[15] A Zrachia M Dobroslav M Blass et al ldquoInfection of gliomacells with Sindbis virus induces selective activation and tyrosinephosphorylation of protein kinase C 120575 implications for sindbisvirus-induced apoptosisrdquo Journal of Biological Chemistry vol277 no 26 pp 23693ndash23701 2002

[16] A Apostolatos S Song S Acosta et al ldquoInsulin promotesneuronal survival via the alternatively spliced protein kinaseCdeltaII isoformrdquo Journal of Biological Chemistry vol 287 pp9299ndash9310 2012

[17] K J Strissel Z Stancheva H Miyoshi et al ldquoAdipocytedeath adipose tissue remodeling and obesity complicationsrdquoDiabetes vol 56 no 12 pp 2910ndash2918 2007

[18] D Papineau A Gagnon and A Sorisky ldquoApoptosis of humanabdominal preadipocytes before and after differentiation intoadipocytes in culturerdquo Metabolism vol 52 no 8 pp 987ndash9922003

[19] C Weyer J E Foley C Bogardus P A Tataranni and RE Pratley ldquoEnlarged subcutaneous abdominal adipocyte sizebut not obesity itself predicts type II diabetes independent ofinsulin resistancerdquo Diabetologia vol 43 no 12 pp 1498ndash15062000

[20] L B Tanko A Siddiq C Lecoeur et al ldquoACDCadiponectinand PPAR-120574 gene polymorphisms implications for features ofobesityrdquo Obesity Research vol 13 no 12 pp 2113ndash2121 2005

[21] N Ouchi S Kihara T Funahashi Y Matsuzawa and KWalshldquoObesity adiponectin and vascular inflammatory diseaserdquo Cur-rent Opinion in Lipidology vol 14 no 6 pp 561ndash566 2003

4 ISRN Obesity

lowastlowastlowast

lowastlowastlowast

60

50

40

30

20

10

0

Ann

exin

PI

Day 0 Day 10

ControlSerum deprived

lowastlowastlowast

Figure 1 Human lean preadipocytes were differentiated in vitroCells were serum deprived for 48 h on day 0 (preadipocytes) andon day 10 (mature adipocytes) AnnexinPI was analyzed by flowcytometry Experiments were repeated 5 times Statistical analysisperformed by two-way ANOVA 119875 gt 075 ns not significant withingroup lowastlowastlowast119875 lt 00001 highly significant between control and serum-deprived lowastlowastlowast119875 lt 00001 highly significant between days 0 and 10serum-deprived

alternatively spliced to Bcl-xL Bcl-xS and Bcl-x120573 Bcl-xL isgenerated via alternate 51015840 splice site selection on exon 2 ofBcl-x pre-mRNA and includes the BH1 and BH2 domains(Figure 2(b)) Bcl-x120573 is the least studied and has no knownrole in apoptosis The long form Bcl-xL inhibits apoptosiswhile the short isoform Bcl-xS promotes apoptosis Bcl2 andBcl-xL dimerize to initiate the cell survival pathways

35 Alternative Splicing of Caspase 9 Caspase 9 an initiatorcaspase is a ubiquitously expressed protease It representsa pivotal signaling protein in the apoptotic cascade Inhibi-tion of Caspase 9 by a chemical inhibitor mollugin affectsadipogenesis [10 11] It is alternatively spliced to caspase 9a(apoptotic) and caspase 9b (antiapoptosis) via inclusion ofthe cassette exons 3 4 5 and 6 in caspase 9a (Figure 2(c))Caspase 9a activates caspase 3 which cleaves substratesthereby mediating apoptosis

To verify our results from the exon array we carriedout a detailed analysis of the differentiation of humanpreadipocytes from days 0 to 10 as they differentiate intomature adipocytes as described Our data indicates thatbetween days 2 and 4 of differentiation a marked shift isobserved in the expression of genes involved in prosurvivalpathways (Figure 2(d)) Day 6 of differentiation is markedby the splicing of antiapoptotic proteins that advances the

mature adipocytes to be resistant to apoptosis On day 0cells expressed the apoptotic splice variants PKC120575I caspase9a and lower levels of Bcl-xL By day 4 cells startedexpressing the antiapoptotic proteins Bcl2 and splice variantsPKC120575VIII caspase 9b and Bcl-xL with a concurrent decreasein proapoptotic proteins This represents a critical switchin adipocyte differentiation modulated by the splicing ofapoptosis genes

36 Expression of PKC120575 Splice Variants in Obese PatientsOur above data was obtained in normal lean differentiatingpreadipocytes Next we sought to determine whether thesplicing pattern and expression of these genes was affectedby the metabolic state Hence we sought to measure thelevels of PKC120575 from obese patients We obtained proteinlysates from subcutaneous and omental adipose tissues froman obese patient undergoing Roux-en-y gastroenterostomysurgery (from Drs Murr and Cooper Tampa VAMC) Thesamples were from an obese patient closely matching theBMI and other criteria to the preadipocytes obtained fromZenBio to allow for a better comparisonThe results indicateda sharp increase in PKC120575VIII expression in the obese patientcompared to PKC120575I (Figure 3(a)) It is important to note thatthese adipose tissue samples represent a heterogeneous cellpopulation containing predominantlymature adipocytes andalso differentiating preadipocytes as well as other stem cellswhich may or may not have the adipocyte lineage

Next we obtained preadipocytes from the subcutaneousdepot of obese patients (ZenBio) and differentiated them invitro We chose to further study the preadipocyte differentia-tion of obese subcutaneous fraction because our lean cell linewas derived from the subcutaneous fractionThe cell line wasselected to match the adipose tissue lysate (above) such thatwe could evaluate differentiation of obese preadipocytes invitro It also allowed us to compare the expression patterns ofthe genes between the lean and obese during differentiationWe first evaluated apoptosis and susceptibility to serumdeprivation on days 0 and 10 by annexin V and PI flowcytometry Our results (Figures 3(b) and 3(c)) indicatedthat obese preadipocytes are more resistant to apoptosiscompared to lean on day 0 (UR quadrant with both AVand PI lean 307 obese 167) Day 10 mature adipocytesdemonstrate much lower susceptibility to serum-deprivedapoptosis (lean 187 obese 104)

To determine the splicing pattern of differentiating obesepreadipocytes we performed Western blot analysis Weobserved that these cells expressed PKC120575VIII Bcl-xL Bcl2and caspase 9 on day 0 that is in the preadipocyte stage Weobserved increased expression of PKC120575VIII as determined bythe PKC120575 splice variant ratio observed on day 0 (Figure 4)This data suggests that the early expression of these genesin the adipocytes promotes their survival mechanisms andthereby provides resistance to cell death

37 Adipogenesis in Obesity Is Dysregulated The lean andobese human preadipocytes were differentiated for 0 to14 days and adipogenesis was observed by staining of thelipid droplets by Oil Red O The lean cells showed lipid

ISRN Obesity 5

PKC120575VIIImRNA E9 E10 E11

Cell survivalantiapoptosis

PKC120575pre-mRNA

E9 E10 E11

SS II

SS I

PKC120575ImRNA

E9 E10 E11 Cell apoptosis

(a)

Bcl-xLmRNA E1 E2 E3

Cell survivalantiapoptosis

Bcl-xpre-mRNA

E1 E2 E3SS II

SS I

Bcl-xSmRNA

E1 E2 E3 Cell apoptosis

(b)

Caspase 9amRNA Cell apoptosis

Caspase 9Pre-mRNA

Exon 2 Exons 3 4 5 6Cassette Exon 7

Caspase 9bmRNA

Cell survivalantiapoptosis

(c)

Day

PKC120575VIIIPKC120575I

Bcl-2

Bcl-xL

Caspase 9aCaspase 9b

0 2 4 6 8 10 0 2 4 6 8 10

TNF120572

XIAP

Adiponectin

PPAR

120573-actin

60

50

40

30

20

10

0Day 0 Day 2 Day 4 Day 6 Day 8 Day 10

120574

PKC120575

exon

inclu

sion

()

(d)

Figure 2 (a) Schematic of alternative splicing of PKC120575 pre-mRNA generating PKC120575I mRNA and PKC120575VIII mRNA via alternative 51015840 splicesite selection E exon SS splice site (b) Schematic of alternative splicing of Bcl-x pre-mRNA generating Bcl-xL mRNA and Bcl-xS mRNAvia alternative 51015840 splice site selection E exon SS splice site (c) Schematic of alternative splicing of caspase 9 pre-mRNA generating caspase9a mRNA and caspase 9b mRNA via cassette exon inclusion (d) Western blot analysis of differentiating lean preadipocytes from days 0 to 10using antibodies as indicated in the figure Between days 2 and 4 a marked shift is observed in the splicing pattern of survival proteins Thisperiodmarks terminal differentiation of adipocytesThe blots are representative of 3 experiments performed individually with similar resultsGraph represents percent PKC120575 exon inclusion calculated as PKC120575VIII(PKC120575VIII + PKC120575I) times 100 and is representative of four experimentsperformed separately

accumulation beginning at day 4 We observed that obesepreadipocyteswere slower to differentiate but exhibited largerlipid droplets by day 10 We measured and compared themature adipocyte size on day 14 between lean and obesesamples (Figure 5) Our results indicate a dramatic increasein adipocyte size in obese samples Further the expressionof adiponectin and PPAR120574 was dysregulated in differen-tiating obese preadipocytes (Figure 4) which may explainthe aberrant differentiation and adipogenesis observed in

obese patients These results indicated that the adipogenesisprogram was dysregulated in obese adipocytes

To evaluate this we performed real-time qPCR analysisto measure the expression of adipogenesis marker genesFOXO SIRT1 adiponectin and PPAR120574 with a concurrentmeasure of PKC120575VIII expression Total RNA was isolatedon day 2 of differentiation of lean and obese preadipocytesOur results (Figure 6) indicated that PKC120575VIII expressionwas increased dramatically in obese samples along with a

6 ISRN Obesity

E01 Unstained controlGate [no gating]

16777215

10000000

5000000

0

SSC-

A

0 5000000 10000000 16777

FSC-A

Day 10

1072

106

105

104

103

102

PI-A

103 104 105 106 1072

Annexin V-A

F04 Lean-controlGate (P1 in all)

1072

106

105

104

103

102

PI-A

103 104 105 106 1072

Annexin V-A

1072

106

105

104

103

102

PI-A

103 104 105 106 1072

Annexin V-A

1034 104 105 106 1072

Annexin V-A

1072

106

105

104

1027

PI-A

1034 104 105 106 1072

Annexin V-A

G06 Lean-serum deprived

H05 Obese-controlGate (P1 in all)

H07 Obese-serum deprivedGate (P1 in all)

E04 Lean-controlGate (P1 in all)

E06 Lean-serum deprivedGate (P1 in all)

E12 Obese-controlGate (P1 in all)

F05 Obese-serum deprivedGate (P1 in all)

Gate (P1 in all)

1072

106

105

104

1027PI

-A

1072

106

105

104

103

102

PI-A

103 104 105 106 1072

Annexin V-A

1034 104 105 106 1072

Annexin V-A

1072

106

105

104

1027

PI-A

1034 104 105 106 1072

Annexin V-A

1072

106

105

104

1027

PI-A

F01 Unstained controlGate [no gating]

16777215

10000000

5000000

0

SSC-

A0 5000000 10000000 16777

FSC-A

Day 0

lowastlowastlowast

lowastlowastlowast lowastlowastlowast

35

30

25

20

15

10

5

0

Ann

exin

VP

I

Leancontrol

Leanserum

deprivedObesecontrol

Obeseserum

deprived

Day 0Day 10

lowastlowastlowastlowastlowastlowast

(a)

(b)

(c)

1 2

PKC120575VIIIPKC120575I

120573-actin

80

60

40

20

0Omental Subcutaneous

PKC120575

exon

inclu

sion

()

Figure 3 (a) Western blot analysis of adipose tissue lysates obtained from obese depots from (1) omental (2) subcutaneous Theblots are representative of experiments repeated thrice from the lysates Graph represents percent PKC120575 exon inclusion calculated asPKC120575VIII(PKC120575VIII + PKC120575I)times 100 and is representative of three experiments (b)Human lean and obese preadipocytes were differentiatedin vitro Cells were serum deprived for 48 h on day 0 (preadipocytes) and on day 10 (mature adipocytes) Annexin VPI staining was analyzedby flow cytometryThe first plot demonstrates the gating of the adipocyte populationThe remaining plots shown are lean control lean-serumdeprived obese control and obese serum-deprived and are divided into four quadrants based on Annexin V (AV) and propidium iodidestaining (PI) staining Q1-LL represents viable (unstained) cells Q1-LR represents AV staining only (early apoptosis) Q1-UR represents cellsstainedwith bothAVandPI (later apoptosis) Q1-UL represents cells stainedwith PI only (end-stage apoptosis) Data shown are representativeof five different experiments (c) Graph represents AVPI and represents 5 experiments Statistical analysis performed by two-way ANOVA119875 gt 075 ns not significant within group lowastlowastlowast119875 lt 00001 highly significant between control and serum-deprived lowastlowastlowast119875 lt 00001 highlysignificant between lean and obese lowastlowastlowast119875 lt 00001 highly significant between days 0 and 10 serum-deprived

ISRN Obesity 7

0 2 4 6 8 10Day 0 2 4 6 8 10

PKC120575VIIIPKC120575I

Bcl-2

Bcl-xL

Caspase 9a

Caspase 9b

Glut4

Adiponectin

PPAR120574

120573-actin

60

50

40

30

20

10

0Day 0 Day 2 Day 4 Day 6 Day 8 Day 10

PKC120575

exon

inclu

sion

()

Figure 4 Western blot analysis of differentiating subcutaneousobese preadipocytes from days 0 to 10 using antibodies as indicatedThe blots are representative of four experiments performed indi-vidually with similar results Graph represents percent PKC120575 exoninclusion calculated as PKC120575VIII(PKC120575VIII + PKC120575I) times 100 and isrepresentative of four experiments performed separately

Lean Obese

(a)

40000

35000

30000

25000

20000

15000

10000

5000

0Lean Obese

Mea

n ad

ipoc

yte s

ize (120583

m)

(b)

Figure 5 Human lean or obese preadipocytes were differentiated invitro (a) Adipocytes on day 14 were stained for lipid content withOil Red O and (b) adipocyte size measured (methods) Experimentswere repeated 4 times with similar result

25

2

15

1

05

0

mRN

A ex

pres

sion

lowast

lowast

lowast

lowast

lowast

SIRT1 FOXO1 Adiponectin

LeanObese

PKC120575VIII PPAR120574

Figure 6 Lean and obese human preadipocytes were differentiatedin vitro Total RNA was collected on day 2 Real Time SYBR qPCRanalysis of PKC120575VIII SIRT1 FOXO1 adiponectin and PPAR120574was performed in triplicate and repeated 3 times in separateexperimentsThe absolute mRNA expression transcripts are shownlowast119875 lt 005 (by two-tailed Studentrsquos 119905-test) significant between leanand obese for each mRNA

marked decrease in FOXO SIRT1 adiponectin and PPAR120574expression compared to lean samples

4 Discussion

Obesity is an epidemic facing the world with the USAleading the statistics on increasing number of overweight andobese population Adipose tissue is an important endocrineregulator of energy homeostasis and glucose metabolismand the excess energy is stored in the white adipose tissueThe onset of adipogenesis and the transcriptional factorsthat initiate it have been studied in depth However withour current knowledge of the genome it is evident thatthere are several other mechanisms in play which regulatethe adipocyte phenotype during differentiation Apoptosis isrequired for fat cell turnover throughout our life Howeverfor weight loss in obese patients it may be beneficial toincrease adipocyte apoptosis along with therapies targetinghyperplasia and hypertrophy in adipocytes To our knowl-edge this is the first demonstration of alternative splicingevents during adipogenesis which contribute to the matureadipocyte phenotype Our results indicate a distinct splicingpattern between lean and obese patients

Apoptosis in preadipocytes and adipocytes was previ-ously observed in human cell lines However the under-lying mechanisms were not described Alternative splicinga posttranscriptional event is a powerful regulator of geneexpression We have identified key genes which are alter-natively spliced that dramatically switch their splice variantexpression during the progression of adipogenesis This mayexplain why preadipocytes are more susceptible to apoptoticagents while the mature adipocytes are resistant to it Protein

8 ISRN Obesity

kinase C delta (PKC120575) is a crucial signaling kinase affectingdownstream apoptotic pathways [8 12] Our data indicatesincreased expression of PKC120575VIII along with an early switchbetween the splice variants associated with apoptosis duringadipogenesis process of preadipocytes derived from obesepatients

PKC120575 is a serinethreonine kinase which plays a centralrole in apoptosis PKC120575 has dual functions as a mediator ofapoptosis and as an antiapoptosis effector Its splice variantsfunction as a switch that determines cell survival and fateThis could explain the opposing effects of PKC120575 on cellularapoptosis cited in the literature [13ndash15] Genetic switchesbased on alternative splicing are important for many cellularand developmental processes PKC120575I promotes apoptosiswhile PKC120575VIII promotes survival We have shown thatPKC120575II (the mouse splice variant) and PKC120575VIII functionas prosurvival proteins [8 12] the functions of other PKC120575splice variants are not yet established Our previous studiesin neurons have demonstrated that the prosurvival variant ofPKC120575 increases the expression of Bcl2 and Bcl-xL which areknown to increase cellular survival [16]

Obesity is associated with a chronic low inflamma-tory state due to the changes in secretory functions ofthe adipocytes Enlarged adipocytes (hypertrophy) lead toincreased adipocyte-derived free fatty acids which stimulatemacrophages In severely obese patientsmacrophages invadethe adipose tissue and release chemokines leading to necrosisand autophagy of adipocytes [17] These complications addto obese-associated metabolic syndrome diabetes insulinresistance and hepatic steatosis These forms of cell deathare distinct from apoptosis (programmed cell death) inadipocytes Apoptosis is shown to be substantially greaterin omental depots compared to subcutaneous depots inhealthy individuals [18] Here we show that obese samplesare resistant to apoptosis and express the antiapoptosis genesfourfold greater than in lean samples This may explain theincreased difficulty to lose weight and maintain the weightloss in overweight and obese individuals Larger adipocytesare an indicator of type II diabetes independent of insulinresistance [19] Previous studies showed that PPAR120574 andadiponectin expression was lower in obesity and type IIdiabetes [20 21] Our data indicated that obese patients havedysregulated adipogenesis program This may be modulatedby the signaling kinase PKC120575VIII which is overexpressedin obese individuals Our data also indicated dysregulatedalternative splicing of apoptosis genes in obese differentiatingcells This aberrant molecular profile may be a causativeeffect leading to obesity or may be a consequence of obesityand its comorbidities We are evaluating this further in ourlaboratory Here we have defined the differences in geneexpression between lean and obese adipocytes

In conclusion it may be possible to modulate alternativesplicing during adipogenesis for therapeutic treatments ofobesity Our results suggest a novel role for a key signalingkinase PKC120575VIII in the process of adipogenesis Expressionlevels of PKC120575VIII splice variant are increased in obesity andmay be linked to the cause or consequence of dysregulatedadipogenesis in obese patients Since new adipocytes arerequired for the maintenance and the endocrine function of

adipocytes balancing the gene expression of apoptotic genesmay be a better approach rather than complete inhibition ofadipogenesis

Conflict of Interests

The authors have no conflict of interests to declare

Authorsrsquo Contribution

GCarter andAApostolatos contributed equally to the paper

Acknowledgments

This work was supported by the Department of VeteransAffairs Medical Research Grant (NAP)The contents do notrepresent the views of the Department of Veterans Affairs ortheUnited States GovernmentThe authors thankDr AndreaMoor for help with microscopic images and adipocyte sizedetermination They thank Sean Yoder from Moffitt arraycore for array analysis

References

[1] S Rodrıguez-Acebes N Palacios J I Botella-Carretero et alldquoGene expression profiling of subcutaneous adipose tissue inmorbid obesity using a focused microarray distinct expressionof cell-cycle- and differentiation-related genesrdquo BMC MedicalGenomics vol 3 article 61 2010

[2] M Guzey D Jukic J Arlotti M Acquafondata R Dhir andR H Getzenberg ldquoIncreased apoptosis of periprostatic adiposetissue in VDR null micerdquo Journal of Cellular Biochemistry vol93 no 1 pp 133ndash141 2004

[3] P Fischer-Posovszky H Tornqvist K M Debatin and MWabitsch ldquoInhibition of death-receptor mediated apoptosis inhuman adipocytes by the insulin-like growth factor I (IGF-I)IGF-I receptor autocrine circuitrdquo Endocrinology vol 145 no4 pp 1849ndash1859 2004

[4] O Ishiko T Sumi H Yoshida Y Hyun and S Ogita ldquoCom-parison of expression of apoptosis regulatory proteins in theadipose tissue of tumor-bearing and diet-restricted rabbitsrdquoInternational Journal of Molecular Medicine vol 8 no 5 pp543ndash547 2001

[5] M A Della-Fera Y H Choi D L Hartzell J Duan MHamrick and C A Baile ldquoSensitivity of obob mice to Leptin-induced adipose tissue apoptosisrdquo Obesity Research vol 13 no9 pp 1540ndash1547 2005

[6] D H Kim S C Woods and R J Seeley ldquoPeptide designedto elicit apoptosis in adipose tissue endothelium reduces foodintake and body weightrdquo Diabetes vol 59 no 4 pp 907ndash9152010

[7] W C Earnshaw L M Martins and S H Kaufmann ldquoMam-malian caspases structure activation substrates and functionsduring apoptosisrdquo Annual Review of Biochemistry vol 68 pp383ndash424 1999

[8] K Jiang A H Apostolatos T Ghansah et al ldquoIdentificationof a novel antiapoptotic human protein kinase C 120575 isoformPKC120575VIII in NT2 cellsrdquo Biochemistry vol 47 no 2 pp 787ndash797 2008

ISRN Obesity 9

[9] H Apostolatos A Apostolatos T Vickers et al ldquoVitamin Ametabolite all-trans-retinoic acid mediates alternative splicingof protein kinase C 120575VIII (PKC120575VIII) isoform via splicingfactor SC35rdquo Journal of Biological Chemistry vol 285 no 34pp 25987ndash25995 2010

[10] D Y Jun C R Han M S Choi M A Bae M H Woo and YH Kim ldquoEffect of mollugin on apoptosis and adipogenesis of3T3-L1 preadipocytesrdquo Phytotherapy Research vol 25 no 5 pp724ndash731 2011

[11] S M Kim H S Park D Y Jun et al ldquoMollugin induces apop-tosis in human Jurkat T cells through endoplasmic reticulumstress-mediated activation of JNK and caspase-12 and subse-quent activation of mitochondria-dependent caspase cascaderegulated by Bcl-xLrdquo Toxicology and Applied Pharmacology vol241 no 2 pp 210ndash220 2009

[12] N A Patel S S Song and D R Cooper ldquoPKC120575 alternativelyspliced isoforms modulate cellular apoptosis in retinoic acid-induced differentiation of human NT2 cells and mouse embry-onic stem cellsrdquo Gene Expression vol 13 no 2 pp 73ndash84 2005

[13] J J Peluso A Pappalardo and G Fernandez ldquoBasic fibrob-last growth factor maintains calcium homeostasis and gran-ulosa cell viability by stimulating calcium efflux via a PKC120575-dependent pathwayrdquo Endocrinology vol 142 no 10 pp 4203ndash4211 2001

[14] L E Kilpatrick J Y Lee K M Haines D E Campbell KE Sullivan and H M Korchak ldquoA role for PKC-120575 and PI3-kinase in TNF-120572-mediated antiapoptotic signaling in thehuman neutrophilrdquo The American Journal of Physiology vol283 no 1 pp C48ndashC57 2002

[15] A Zrachia M Dobroslav M Blass et al ldquoInfection of gliomacells with Sindbis virus induces selective activation and tyrosinephosphorylation of protein kinase C 120575 implications for sindbisvirus-induced apoptosisrdquo Journal of Biological Chemistry vol277 no 26 pp 23693ndash23701 2002

[16] A Apostolatos S Song S Acosta et al ldquoInsulin promotesneuronal survival via the alternatively spliced protein kinaseCdeltaII isoformrdquo Journal of Biological Chemistry vol 287 pp9299ndash9310 2012

[17] K J Strissel Z Stancheva H Miyoshi et al ldquoAdipocytedeath adipose tissue remodeling and obesity complicationsrdquoDiabetes vol 56 no 12 pp 2910ndash2918 2007

[18] D Papineau A Gagnon and A Sorisky ldquoApoptosis of humanabdominal preadipocytes before and after differentiation intoadipocytes in culturerdquo Metabolism vol 52 no 8 pp 987ndash9922003

[19] C Weyer J E Foley C Bogardus P A Tataranni and RE Pratley ldquoEnlarged subcutaneous abdominal adipocyte sizebut not obesity itself predicts type II diabetes independent ofinsulin resistancerdquo Diabetologia vol 43 no 12 pp 1498ndash15062000

[20] L B Tanko A Siddiq C Lecoeur et al ldquoACDCadiponectinand PPAR-120574 gene polymorphisms implications for features ofobesityrdquo Obesity Research vol 13 no 12 pp 2113ndash2121 2005

[21] N Ouchi S Kihara T Funahashi Y Matsuzawa and KWalshldquoObesity adiponectin and vascular inflammatory diseaserdquo Cur-rent Opinion in Lipidology vol 14 no 6 pp 561ndash566 2003

ISRN Obesity 5

PKC120575VIIImRNA E9 E10 E11

Cell survivalantiapoptosis

PKC120575pre-mRNA

E9 E10 E11

SS II

SS I

PKC120575ImRNA

E9 E10 E11 Cell apoptosis

(a)

Bcl-xLmRNA E1 E2 E3

Cell survivalantiapoptosis

Bcl-xpre-mRNA

E1 E2 E3SS II

SS I

Bcl-xSmRNA

E1 E2 E3 Cell apoptosis

(b)

Caspase 9amRNA Cell apoptosis

Caspase 9Pre-mRNA

Exon 2 Exons 3 4 5 6Cassette Exon 7

Caspase 9bmRNA

Cell survivalantiapoptosis

(c)

Day

PKC120575VIIIPKC120575I

Bcl-2

Bcl-xL

Caspase 9aCaspase 9b

0 2 4 6 8 10 0 2 4 6 8 10

TNF120572

XIAP

Adiponectin

PPAR

120573-actin

60

50

40

30

20

10

0Day 0 Day 2 Day 4 Day 6 Day 8 Day 10

120574

PKC120575

exon

inclu

sion

()

(d)

Figure 2 (a) Schematic of alternative splicing of PKC120575 pre-mRNA generating PKC120575I mRNA and PKC120575VIII mRNA via alternative 51015840 splicesite selection E exon SS splice site (b) Schematic of alternative splicing of Bcl-x pre-mRNA generating Bcl-xL mRNA and Bcl-xS mRNAvia alternative 51015840 splice site selection E exon SS splice site (c) Schematic of alternative splicing of caspase 9 pre-mRNA generating caspase9a mRNA and caspase 9b mRNA via cassette exon inclusion (d) Western blot analysis of differentiating lean preadipocytes from days 0 to 10using antibodies as indicated in the figure Between days 2 and 4 a marked shift is observed in the splicing pattern of survival proteins Thisperiodmarks terminal differentiation of adipocytesThe blots are representative of 3 experiments performed individually with similar resultsGraph represents percent PKC120575 exon inclusion calculated as PKC120575VIII(PKC120575VIII + PKC120575I) times 100 and is representative of four experimentsperformed separately

accumulation beginning at day 4 We observed that obesepreadipocyteswere slower to differentiate but exhibited largerlipid droplets by day 10 We measured and compared themature adipocyte size on day 14 between lean and obesesamples (Figure 5) Our results indicate a dramatic increasein adipocyte size in obese samples Further the expressionof adiponectin and PPAR120574 was dysregulated in differen-tiating obese preadipocytes (Figure 4) which may explainthe aberrant differentiation and adipogenesis observed in

obese patients These results indicated that the adipogenesisprogram was dysregulated in obese adipocytes

To evaluate this we performed real-time qPCR analysisto measure the expression of adipogenesis marker genesFOXO SIRT1 adiponectin and PPAR120574 with a concurrentmeasure of PKC120575VIII expression Total RNA was isolatedon day 2 of differentiation of lean and obese preadipocytesOur results (Figure 6) indicated that PKC120575VIII expressionwas increased dramatically in obese samples along with a

6 ISRN Obesity

E01 Unstained controlGate [no gating]

16777215

10000000

5000000

0

SSC-

A

0 5000000 10000000 16777

FSC-A

Day 10

1072

106

105

104

103

102

PI-A

103 104 105 106 1072

Annexin V-A

F04 Lean-controlGate (P1 in all)

1072

106

105

104

103

102

PI-A

103 104 105 106 1072

Annexin V-A

1072

106

105

104

103

102

PI-A

103 104 105 106 1072

Annexin V-A

1034 104 105 106 1072

Annexin V-A

1072

106

105

104

1027

PI-A

1034 104 105 106 1072

Annexin V-A

G06 Lean-serum deprived

H05 Obese-controlGate (P1 in all)

H07 Obese-serum deprivedGate (P1 in all)

E04 Lean-controlGate (P1 in all)

E06 Lean-serum deprivedGate (P1 in all)

E12 Obese-controlGate (P1 in all)

F05 Obese-serum deprivedGate (P1 in all)

Gate (P1 in all)

1072

106

105

104

1027PI

-A

1072

106

105

104

103

102

PI-A

103 104 105 106 1072

Annexin V-A

1034 104 105 106 1072

Annexin V-A

1072

106

105

104

1027

PI-A

1034 104 105 106 1072

Annexin V-A

1072

106

105

104

1027

PI-A

F01 Unstained controlGate [no gating]

16777215

10000000

5000000

0

SSC-

A0 5000000 10000000 16777

FSC-A

Day 0

lowastlowastlowast

lowastlowastlowast lowastlowastlowast

35

30

25

20

15

10

5

0

Ann

exin

VP

I

Leancontrol

Leanserum

deprivedObesecontrol

Obeseserum

deprived

Day 0Day 10

lowastlowastlowastlowastlowastlowast

(a)

(b)

(c)

1 2

PKC120575VIIIPKC120575I

120573-actin

80

60

40

20

0Omental Subcutaneous

PKC120575

exon

inclu

sion

()

Figure 3 (a) Western blot analysis of adipose tissue lysates obtained from obese depots from (1) omental (2) subcutaneous Theblots are representative of experiments repeated thrice from the lysates Graph represents percent PKC120575 exon inclusion calculated asPKC120575VIII(PKC120575VIII + PKC120575I)times 100 and is representative of three experiments (b)Human lean and obese preadipocytes were differentiatedin vitro Cells were serum deprived for 48 h on day 0 (preadipocytes) and on day 10 (mature adipocytes) Annexin VPI staining was analyzedby flow cytometryThe first plot demonstrates the gating of the adipocyte populationThe remaining plots shown are lean control lean-serumdeprived obese control and obese serum-deprived and are divided into four quadrants based on Annexin V (AV) and propidium iodidestaining (PI) staining Q1-LL represents viable (unstained) cells Q1-LR represents AV staining only (early apoptosis) Q1-UR represents cellsstainedwith bothAVandPI (later apoptosis) Q1-UL represents cells stainedwith PI only (end-stage apoptosis) Data shown are representativeof five different experiments (c) Graph represents AVPI and represents 5 experiments Statistical analysis performed by two-way ANOVA119875 gt 075 ns not significant within group lowastlowastlowast119875 lt 00001 highly significant between control and serum-deprived lowastlowastlowast119875 lt 00001 highlysignificant between lean and obese lowastlowastlowast119875 lt 00001 highly significant between days 0 and 10 serum-deprived

ISRN Obesity 7

0 2 4 6 8 10Day 0 2 4 6 8 10

PKC120575VIIIPKC120575I

Bcl-2

Bcl-xL

Caspase 9a

Caspase 9b

Glut4

Adiponectin

PPAR120574

120573-actin

60

50

40

30

20

10

0Day 0 Day 2 Day 4 Day 6 Day 8 Day 10

PKC120575

exon

inclu

sion

()

Figure 4 Western blot analysis of differentiating subcutaneousobese preadipocytes from days 0 to 10 using antibodies as indicatedThe blots are representative of four experiments performed indi-vidually with similar results Graph represents percent PKC120575 exoninclusion calculated as PKC120575VIII(PKC120575VIII + PKC120575I) times 100 and isrepresentative of four experiments performed separately

Lean Obese

(a)

40000

35000

30000

25000

20000

15000

10000

5000

0Lean Obese

Mea

n ad

ipoc

yte s

ize (120583

m)

(b)

Figure 5 Human lean or obese preadipocytes were differentiated invitro (a) Adipocytes on day 14 were stained for lipid content withOil Red O and (b) adipocyte size measured (methods) Experimentswere repeated 4 times with similar result

25

2

15

1

05

0

mRN

A ex

pres

sion

lowast

lowast

lowast

lowast

lowast

SIRT1 FOXO1 Adiponectin

LeanObese

PKC120575VIII PPAR120574

Figure 6 Lean and obese human preadipocytes were differentiatedin vitro Total RNA was collected on day 2 Real Time SYBR qPCRanalysis of PKC120575VIII SIRT1 FOXO1 adiponectin and PPAR120574was performed in triplicate and repeated 3 times in separateexperimentsThe absolute mRNA expression transcripts are shownlowast119875 lt 005 (by two-tailed Studentrsquos 119905-test) significant between leanand obese for each mRNA

marked decrease in FOXO SIRT1 adiponectin and PPAR120574expression compared to lean samples

4 Discussion

Obesity is an epidemic facing the world with the USAleading the statistics on increasing number of overweight andobese population Adipose tissue is an important endocrineregulator of energy homeostasis and glucose metabolismand the excess energy is stored in the white adipose tissueThe onset of adipogenesis and the transcriptional factorsthat initiate it have been studied in depth However withour current knowledge of the genome it is evident thatthere are several other mechanisms in play which regulatethe adipocyte phenotype during differentiation Apoptosis isrequired for fat cell turnover throughout our life Howeverfor weight loss in obese patients it may be beneficial toincrease adipocyte apoptosis along with therapies targetinghyperplasia and hypertrophy in adipocytes To our knowl-edge this is the first demonstration of alternative splicingevents during adipogenesis which contribute to the matureadipocyte phenotype Our results indicate a distinct splicingpattern between lean and obese patients

Apoptosis in preadipocytes and adipocytes was previ-ously observed in human cell lines However the under-lying mechanisms were not described Alternative splicinga posttranscriptional event is a powerful regulator of geneexpression We have identified key genes which are alter-natively spliced that dramatically switch their splice variantexpression during the progression of adipogenesis This mayexplain why preadipocytes are more susceptible to apoptoticagents while the mature adipocytes are resistant to it Protein

8 ISRN Obesity

kinase C delta (PKC120575) is a crucial signaling kinase affectingdownstream apoptotic pathways [8 12] Our data indicatesincreased expression of PKC120575VIII along with an early switchbetween the splice variants associated with apoptosis duringadipogenesis process of preadipocytes derived from obesepatients

PKC120575 is a serinethreonine kinase which plays a centralrole in apoptosis PKC120575 has dual functions as a mediator ofapoptosis and as an antiapoptosis effector Its splice variantsfunction as a switch that determines cell survival and fateThis could explain the opposing effects of PKC120575 on cellularapoptosis cited in the literature [13ndash15] Genetic switchesbased on alternative splicing are important for many cellularand developmental processes PKC120575I promotes apoptosiswhile PKC120575VIII promotes survival We have shown thatPKC120575II (the mouse splice variant) and PKC120575VIII functionas prosurvival proteins [8 12] the functions of other PKC120575splice variants are not yet established Our previous studiesin neurons have demonstrated that the prosurvival variant ofPKC120575 increases the expression of Bcl2 and Bcl-xL which areknown to increase cellular survival [16]

Obesity is associated with a chronic low inflamma-tory state due to the changes in secretory functions ofthe adipocytes Enlarged adipocytes (hypertrophy) lead toincreased adipocyte-derived free fatty acids which stimulatemacrophages In severely obese patientsmacrophages invadethe adipose tissue and release chemokines leading to necrosisand autophagy of adipocytes [17] These complications addto obese-associated metabolic syndrome diabetes insulinresistance and hepatic steatosis These forms of cell deathare distinct from apoptosis (programmed cell death) inadipocytes Apoptosis is shown to be substantially greaterin omental depots compared to subcutaneous depots inhealthy individuals [18] Here we show that obese samplesare resistant to apoptosis and express the antiapoptosis genesfourfold greater than in lean samples This may explain theincreased difficulty to lose weight and maintain the weightloss in overweight and obese individuals Larger adipocytesare an indicator of type II diabetes independent of insulinresistance [19] Previous studies showed that PPAR120574 andadiponectin expression was lower in obesity and type IIdiabetes [20 21] Our data indicated that obese patients havedysregulated adipogenesis program This may be modulatedby the signaling kinase PKC120575VIII which is overexpressedin obese individuals Our data also indicated dysregulatedalternative splicing of apoptosis genes in obese differentiatingcells This aberrant molecular profile may be a causativeeffect leading to obesity or may be a consequence of obesityand its comorbidities We are evaluating this further in ourlaboratory Here we have defined the differences in geneexpression between lean and obese adipocytes

In conclusion it may be possible to modulate alternativesplicing during adipogenesis for therapeutic treatments ofobesity Our results suggest a novel role for a key signalingkinase PKC120575VIII in the process of adipogenesis Expressionlevels of PKC120575VIII splice variant are increased in obesity andmay be linked to the cause or consequence of dysregulatedadipogenesis in obese patients Since new adipocytes arerequired for the maintenance and the endocrine function of

adipocytes balancing the gene expression of apoptotic genesmay be a better approach rather than complete inhibition ofadipogenesis

Conflict of Interests

The authors have no conflict of interests to declare

Authorsrsquo Contribution

GCarter andAApostolatos contributed equally to the paper

Acknowledgments

This work was supported by the Department of VeteransAffairs Medical Research Grant (NAP)The contents do notrepresent the views of the Department of Veterans Affairs ortheUnited States GovernmentThe authors thankDr AndreaMoor for help with microscopic images and adipocyte sizedetermination They thank Sean Yoder from Moffitt arraycore for array analysis

References

[1] S Rodrıguez-Acebes N Palacios J I Botella-Carretero et alldquoGene expression profiling of subcutaneous adipose tissue inmorbid obesity using a focused microarray distinct expressionof cell-cycle- and differentiation-related genesrdquo BMC MedicalGenomics vol 3 article 61 2010

[2] M Guzey D Jukic J Arlotti M Acquafondata R Dhir andR H Getzenberg ldquoIncreased apoptosis of periprostatic adiposetissue in VDR null micerdquo Journal of Cellular Biochemistry vol93 no 1 pp 133ndash141 2004

[3] P Fischer-Posovszky H Tornqvist K M Debatin and MWabitsch ldquoInhibition of death-receptor mediated apoptosis inhuman adipocytes by the insulin-like growth factor I (IGF-I)IGF-I receptor autocrine circuitrdquo Endocrinology vol 145 no4 pp 1849ndash1859 2004

[4] O Ishiko T Sumi H Yoshida Y Hyun and S Ogita ldquoCom-parison of expression of apoptosis regulatory proteins in theadipose tissue of tumor-bearing and diet-restricted rabbitsrdquoInternational Journal of Molecular Medicine vol 8 no 5 pp543ndash547 2001

[5] M A Della-Fera Y H Choi D L Hartzell J Duan MHamrick and C A Baile ldquoSensitivity of obob mice to Leptin-induced adipose tissue apoptosisrdquo Obesity Research vol 13 no9 pp 1540ndash1547 2005

[6] D H Kim S C Woods and R J Seeley ldquoPeptide designedto elicit apoptosis in adipose tissue endothelium reduces foodintake and body weightrdquo Diabetes vol 59 no 4 pp 907ndash9152010

[7] W C Earnshaw L M Martins and S H Kaufmann ldquoMam-malian caspases structure activation substrates and functionsduring apoptosisrdquo Annual Review of Biochemistry vol 68 pp383ndash424 1999

[8] K Jiang A H Apostolatos T Ghansah et al ldquoIdentificationof a novel antiapoptotic human protein kinase C 120575 isoformPKC120575VIII in NT2 cellsrdquo Biochemistry vol 47 no 2 pp 787ndash797 2008

ISRN Obesity 9

[9] H Apostolatos A Apostolatos T Vickers et al ldquoVitamin Ametabolite all-trans-retinoic acid mediates alternative splicingof protein kinase C 120575VIII (PKC120575VIII) isoform via splicingfactor SC35rdquo Journal of Biological Chemistry vol 285 no 34pp 25987ndash25995 2010

[10] D Y Jun C R Han M S Choi M A Bae M H Woo and YH Kim ldquoEffect of mollugin on apoptosis and adipogenesis of3T3-L1 preadipocytesrdquo Phytotherapy Research vol 25 no 5 pp724ndash731 2011

[11] S M Kim H S Park D Y Jun et al ldquoMollugin induces apop-tosis in human Jurkat T cells through endoplasmic reticulumstress-mediated activation of JNK and caspase-12 and subse-quent activation of mitochondria-dependent caspase cascaderegulated by Bcl-xLrdquo Toxicology and Applied Pharmacology vol241 no 2 pp 210ndash220 2009

[12] N A Patel S S Song and D R Cooper ldquoPKC120575 alternativelyspliced isoforms modulate cellular apoptosis in retinoic acid-induced differentiation of human NT2 cells and mouse embry-onic stem cellsrdquo Gene Expression vol 13 no 2 pp 73ndash84 2005

[13] J J Peluso A Pappalardo and G Fernandez ldquoBasic fibrob-last growth factor maintains calcium homeostasis and gran-ulosa cell viability by stimulating calcium efflux via a PKC120575-dependent pathwayrdquo Endocrinology vol 142 no 10 pp 4203ndash4211 2001

[14] L E Kilpatrick J Y Lee K M Haines D E Campbell KE Sullivan and H M Korchak ldquoA role for PKC-120575 and PI3-kinase in TNF-120572-mediated antiapoptotic signaling in thehuman neutrophilrdquo The American Journal of Physiology vol283 no 1 pp C48ndashC57 2002

[15] A Zrachia M Dobroslav M Blass et al ldquoInfection of gliomacells with Sindbis virus induces selective activation and tyrosinephosphorylation of protein kinase C 120575 implications for sindbisvirus-induced apoptosisrdquo Journal of Biological Chemistry vol277 no 26 pp 23693ndash23701 2002

[16] A Apostolatos S Song S Acosta et al ldquoInsulin promotesneuronal survival via the alternatively spliced protein kinaseCdeltaII isoformrdquo Journal of Biological Chemistry vol 287 pp9299ndash9310 2012

[17] K J Strissel Z Stancheva H Miyoshi et al ldquoAdipocytedeath adipose tissue remodeling and obesity complicationsrdquoDiabetes vol 56 no 12 pp 2910ndash2918 2007

[18] D Papineau A Gagnon and A Sorisky ldquoApoptosis of humanabdominal preadipocytes before and after differentiation intoadipocytes in culturerdquo Metabolism vol 52 no 8 pp 987ndash9922003

[19] C Weyer J E Foley C Bogardus P A Tataranni and RE Pratley ldquoEnlarged subcutaneous abdominal adipocyte sizebut not obesity itself predicts type II diabetes independent ofinsulin resistancerdquo Diabetologia vol 43 no 12 pp 1498ndash15062000

[20] L B Tanko A Siddiq C Lecoeur et al ldquoACDCadiponectinand PPAR-120574 gene polymorphisms implications for features ofobesityrdquo Obesity Research vol 13 no 12 pp 2113ndash2121 2005

[21] N Ouchi S Kihara T Funahashi Y Matsuzawa and KWalshldquoObesity adiponectin and vascular inflammatory diseaserdquo Cur-rent Opinion in Lipidology vol 14 no 6 pp 561ndash566 2003

6 ISRN Obesity

E01 Unstained controlGate [no gating]

16777215

10000000

5000000

0

SSC-

A

0 5000000 10000000 16777

FSC-A

Day 10

1072

106

105

104

103

102

PI-A

103 104 105 106 1072

Annexin V-A

F04 Lean-controlGate (P1 in all)

1072

106

105

104

103

102

PI-A

103 104 105 106 1072

Annexin V-A

1072

106

105

104

103

102

PI-A

103 104 105 106 1072

Annexin V-A

1034 104 105 106 1072

Annexin V-A

1072

106

105

104

1027

PI-A

1034 104 105 106 1072

Annexin V-A

G06 Lean-serum deprived

H05 Obese-controlGate (P1 in all)

H07 Obese-serum deprivedGate (P1 in all)

E04 Lean-controlGate (P1 in all)

E06 Lean-serum deprivedGate (P1 in all)

E12 Obese-controlGate (P1 in all)

F05 Obese-serum deprivedGate (P1 in all)

Gate (P1 in all)

1072

106

105

104

1027PI

-A

1072

106

105

104

103

102

PI-A

103 104 105 106 1072

Annexin V-A

1034 104 105 106 1072

Annexin V-A

1072

106

105

104

1027

PI-A

1034 104 105 106 1072

Annexin V-A

1072

106

105

104

1027

PI-A

F01 Unstained controlGate [no gating]

16777215

10000000

5000000

0

SSC-

A0 5000000 10000000 16777

FSC-A

Day 0

lowastlowastlowast

lowastlowastlowast lowastlowastlowast

35

30

25

20

15

10

5

0

Ann

exin

VP

I

Leancontrol

Leanserum

deprivedObesecontrol

Obeseserum

deprived

Day 0Day 10

lowastlowastlowastlowastlowastlowast

(a)

(b)

(c)

1 2

PKC120575VIIIPKC120575I

120573-actin

80

60

40

20

0Omental Subcutaneous

PKC120575

exon

inclu

sion

()

Figure 3 (a) Western blot analysis of adipose tissue lysates obtained from obese depots from (1) omental (2) subcutaneous Theblots are representative of experiments repeated thrice from the lysates Graph represents percent PKC120575 exon inclusion calculated asPKC120575VIII(PKC120575VIII + PKC120575I)times 100 and is representative of three experiments (b)Human lean and obese preadipocytes were differentiatedin vitro Cells were serum deprived for 48 h on day 0 (preadipocytes) and on day 10 (mature adipocytes) Annexin VPI staining was analyzedby flow cytometryThe first plot demonstrates the gating of the adipocyte populationThe remaining plots shown are lean control lean-serumdeprived obese control and obese serum-deprived and are divided into four quadrants based on Annexin V (AV) and propidium iodidestaining (PI) staining Q1-LL represents viable (unstained) cells Q1-LR represents AV staining only (early apoptosis) Q1-UR represents cellsstainedwith bothAVandPI (later apoptosis) Q1-UL represents cells stainedwith PI only (end-stage apoptosis) Data shown are representativeof five different experiments (c) Graph represents AVPI and represents 5 experiments Statistical analysis performed by two-way ANOVA119875 gt 075 ns not significant within group lowastlowastlowast119875 lt 00001 highly significant between control and serum-deprived lowastlowastlowast119875 lt 00001 highlysignificant between lean and obese lowastlowastlowast119875 lt 00001 highly significant between days 0 and 10 serum-deprived

ISRN Obesity 7

0 2 4 6 8 10Day 0 2 4 6 8 10

PKC120575VIIIPKC120575I

Bcl-2

Bcl-xL

Caspase 9a

Caspase 9b

Glut4

Adiponectin

PPAR120574

120573-actin

60

50

40

30

20

10

0Day 0 Day 2 Day 4 Day 6 Day 8 Day 10

PKC120575

exon

inclu

sion

()

Figure 4 Western blot analysis of differentiating subcutaneousobese preadipocytes from days 0 to 10 using antibodies as indicatedThe blots are representative of four experiments performed indi-vidually with similar results Graph represents percent PKC120575 exoninclusion calculated as PKC120575VIII(PKC120575VIII + PKC120575I) times 100 and isrepresentative of four experiments performed separately

Lean Obese

(a)

40000

35000

30000

25000

20000

15000

10000

5000

0Lean Obese

Mea

n ad

ipoc

yte s

ize (120583

m)

(b)

Figure 5 Human lean or obese preadipocytes were differentiated invitro (a) Adipocytes on day 14 were stained for lipid content withOil Red O and (b) adipocyte size measured (methods) Experimentswere repeated 4 times with similar result

25

2

15

1

05

0

mRN

A ex

pres

sion

lowast

lowast

lowast

lowast

lowast

SIRT1 FOXO1 Adiponectin

LeanObese

PKC120575VIII PPAR120574

Figure 6 Lean and obese human preadipocytes were differentiatedin vitro Total RNA was collected on day 2 Real Time SYBR qPCRanalysis of PKC120575VIII SIRT1 FOXO1 adiponectin and PPAR120574was performed in triplicate and repeated 3 times in separateexperimentsThe absolute mRNA expression transcripts are shownlowast119875 lt 005 (by two-tailed Studentrsquos 119905-test) significant between leanand obese for each mRNA

marked decrease in FOXO SIRT1 adiponectin and PPAR120574expression compared to lean samples

4 Discussion

Obesity is an epidemic facing the world with the USAleading the statistics on increasing number of overweight andobese population Adipose tissue is an important endocrineregulator of energy homeostasis and glucose metabolismand the excess energy is stored in the white adipose tissueThe onset of adipogenesis and the transcriptional factorsthat initiate it have been studied in depth However withour current knowledge of the genome it is evident thatthere are several other mechanisms in play which regulatethe adipocyte phenotype during differentiation Apoptosis isrequired for fat cell turnover throughout our life Howeverfor weight loss in obese patients it may be beneficial toincrease adipocyte apoptosis along with therapies targetinghyperplasia and hypertrophy in adipocytes To our knowl-edge this is the first demonstration of alternative splicingevents during adipogenesis which contribute to the matureadipocyte phenotype Our results indicate a distinct splicingpattern between lean and obese patients

Apoptosis in preadipocytes and adipocytes was previ-ously observed in human cell lines However the under-lying mechanisms were not described Alternative splicinga posttranscriptional event is a powerful regulator of geneexpression We have identified key genes which are alter-natively spliced that dramatically switch their splice variantexpression during the progression of adipogenesis This mayexplain why preadipocytes are more susceptible to apoptoticagents while the mature adipocytes are resistant to it Protein

8 ISRN Obesity

kinase C delta (PKC120575) is a crucial signaling kinase affectingdownstream apoptotic pathways [8 12] Our data indicatesincreased expression of PKC120575VIII along with an early switchbetween the splice variants associated with apoptosis duringadipogenesis process of preadipocytes derived from obesepatients

PKC120575 is a serinethreonine kinase which plays a centralrole in apoptosis PKC120575 has dual functions as a mediator ofapoptosis and as an antiapoptosis effector Its splice variantsfunction as a switch that determines cell survival and fateThis could explain the opposing effects of PKC120575 on cellularapoptosis cited in the literature [13ndash15] Genetic switchesbased on alternative splicing are important for many cellularand developmental processes PKC120575I promotes apoptosiswhile PKC120575VIII promotes survival We have shown thatPKC120575II (the mouse splice variant) and PKC120575VIII functionas prosurvival proteins [8 12] the functions of other PKC120575splice variants are not yet established Our previous studiesin neurons have demonstrated that the prosurvival variant ofPKC120575 increases the expression of Bcl2 and Bcl-xL which areknown to increase cellular survival [16]

Obesity is associated with a chronic low inflamma-tory state due to the changes in secretory functions ofthe adipocytes Enlarged adipocytes (hypertrophy) lead toincreased adipocyte-derived free fatty acids which stimulatemacrophages In severely obese patientsmacrophages invadethe adipose tissue and release chemokines leading to necrosisand autophagy of adipocytes [17] These complications addto obese-associated metabolic syndrome diabetes insulinresistance and hepatic steatosis These forms of cell deathare distinct from apoptosis (programmed cell death) inadipocytes Apoptosis is shown to be substantially greaterin omental depots compared to subcutaneous depots inhealthy individuals [18] Here we show that obese samplesare resistant to apoptosis and express the antiapoptosis genesfourfold greater than in lean samples This may explain theincreased difficulty to lose weight and maintain the weightloss in overweight and obese individuals Larger adipocytesare an indicator of type II diabetes independent of insulinresistance [19] Previous studies showed that PPAR120574 andadiponectin expression was lower in obesity and type IIdiabetes [20 21] Our data indicated that obese patients havedysregulated adipogenesis program This may be modulatedby the signaling kinase PKC120575VIII which is overexpressedin obese individuals Our data also indicated dysregulatedalternative splicing of apoptosis genes in obese differentiatingcells This aberrant molecular profile may be a causativeeffect leading to obesity or may be a consequence of obesityand its comorbidities We are evaluating this further in ourlaboratory Here we have defined the differences in geneexpression between lean and obese adipocytes

In conclusion it may be possible to modulate alternativesplicing during adipogenesis for therapeutic treatments ofobesity Our results suggest a novel role for a key signalingkinase PKC120575VIII in the process of adipogenesis Expressionlevels of PKC120575VIII splice variant are increased in obesity andmay be linked to the cause or consequence of dysregulatedadipogenesis in obese patients Since new adipocytes arerequired for the maintenance and the endocrine function of

adipocytes balancing the gene expression of apoptotic genesmay be a better approach rather than complete inhibition ofadipogenesis

Conflict of Interests

The authors have no conflict of interests to declare

Authorsrsquo Contribution

GCarter andAApostolatos contributed equally to the paper

Acknowledgments

This work was supported by the Department of VeteransAffairs Medical Research Grant (NAP)The contents do notrepresent the views of the Department of Veterans Affairs ortheUnited States GovernmentThe authors thankDr AndreaMoor for help with microscopic images and adipocyte sizedetermination They thank Sean Yoder from Moffitt arraycore for array analysis

References

[1] S Rodrıguez-Acebes N Palacios J I Botella-Carretero et alldquoGene expression profiling of subcutaneous adipose tissue inmorbid obesity using a focused microarray distinct expressionof cell-cycle- and differentiation-related genesrdquo BMC MedicalGenomics vol 3 article 61 2010

[2] M Guzey D Jukic J Arlotti M Acquafondata R Dhir andR H Getzenberg ldquoIncreased apoptosis of periprostatic adiposetissue in VDR null micerdquo Journal of Cellular Biochemistry vol93 no 1 pp 133ndash141 2004

[3] P Fischer-Posovszky H Tornqvist K M Debatin and MWabitsch ldquoInhibition of death-receptor mediated apoptosis inhuman adipocytes by the insulin-like growth factor I (IGF-I)IGF-I receptor autocrine circuitrdquo Endocrinology vol 145 no4 pp 1849ndash1859 2004

[4] O Ishiko T Sumi H Yoshida Y Hyun and S Ogita ldquoCom-parison of expression of apoptosis regulatory proteins in theadipose tissue of tumor-bearing and diet-restricted rabbitsrdquoInternational Journal of Molecular Medicine vol 8 no 5 pp543ndash547 2001

[5] M A Della-Fera Y H Choi D L Hartzell J Duan MHamrick and C A Baile ldquoSensitivity of obob mice to Leptin-induced adipose tissue apoptosisrdquo Obesity Research vol 13 no9 pp 1540ndash1547 2005

[6] D H Kim S C Woods and R J Seeley ldquoPeptide designedto elicit apoptosis in adipose tissue endothelium reduces foodintake and body weightrdquo Diabetes vol 59 no 4 pp 907ndash9152010

[7] W C Earnshaw L M Martins and S H Kaufmann ldquoMam-malian caspases structure activation substrates and functionsduring apoptosisrdquo Annual Review of Biochemistry vol 68 pp383ndash424 1999

[8] K Jiang A H Apostolatos T Ghansah et al ldquoIdentificationof a novel antiapoptotic human protein kinase C 120575 isoformPKC120575VIII in NT2 cellsrdquo Biochemistry vol 47 no 2 pp 787ndash797 2008

ISRN Obesity 9

[9] H Apostolatos A Apostolatos T Vickers et al ldquoVitamin Ametabolite all-trans-retinoic acid mediates alternative splicingof protein kinase C 120575VIII (PKC120575VIII) isoform via splicingfactor SC35rdquo Journal of Biological Chemistry vol 285 no 34pp 25987ndash25995 2010

[10] D Y Jun C R Han M S Choi M A Bae M H Woo and YH Kim ldquoEffect of mollugin on apoptosis and adipogenesis of3T3-L1 preadipocytesrdquo Phytotherapy Research vol 25 no 5 pp724ndash731 2011

[11] S M Kim H S Park D Y Jun et al ldquoMollugin induces apop-tosis in human Jurkat T cells through endoplasmic reticulumstress-mediated activation of JNK and caspase-12 and subse-quent activation of mitochondria-dependent caspase cascaderegulated by Bcl-xLrdquo Toxicology and Applied Pharmacology vol241 no 2 pp 210ndash220 2009

[12] N A Patel S S Song and D R Cooper ldquoPKC120575 alternativelyspliced isoforms modulate cellular apoptosis in retinoic acid-induced differentiation of human NT2 cells and mouse embry-onic stem cellsrdquo Gene Expression vol 13 no 2 pp 73ndash84 2005

[13] J J Peluso A Pappalardo and G Fernandez ldquoBasic fibrob-last growth factor maintains calcium homeostasis and gran-ulosa cell viability by stimulating calcium efflux via a PKC120575-dependent pathwayrdquo Endocrinology vol 142 no 10 pp 4203ndash4211 2001

[14] L E Kilpatrick J Y Lee K M Haines D E Campbell KE Sullivan and H M Korchak ldquoA role for PKC-120575 and PI3-kinase in TNF-120572-mediated antiapoptotic signaling in thehuman neutrophilrdquo The American Journal of Physiology vol283 no 1 pp C48ndashC57 2002

[15] A Zrachia M Dobroslav M Blass et al ldquoInfection of gliomacells with Sindbis virus induces selective activation and tyrosinephosphorylation of protein kinase C 120575 implications for sindbisvirus-induced apoptosisrdquo Journal of Biological Chemistry vol277 no 26 pp 23693ndash23701 2002

[16] A Apostolatos S Song S Acosta et al ldquoInsulin promotesneuronal survival via the alternatively spliced protein kinaseCdeltaII isoformrdquo Journal of Biological Chemistry vol 287 pp9299ndash9310 2012

[17] K J Strissel Z Stancheva H Miyoshi et al ldquoAdipocytedeath adipose tissue remodeling and obesity complicationsrdquoDiabetes vol 56 no 12 pp 2910ndash2918 2007

[18] D Papineau A Gagnon and A Sorisky ldquoApoptosis of humanabdominal preadipocytes before and after differentiation intoadipocytes in culturerdquo Metabolism vol 52 no 8 pp 987ndash9922003

[19] C Weyer J E Foley C Bogardus P A Tataranni and RE Pratley ldquoEnlarged subcutaneous abdominal adipocyte sizebut not obesity itself predicts type II diabetes independent ofinsulin resistancerdquo Diabetologia vol 43 no 12 pp 1498ndash15062000

[20] L B Tanko A Siddiq C Lecoeur et al ldquoACDCadiponectinand PPAR-120574 gene polymorphisms implications for features ofobesityrdquo Obesity Research vol 13 no 12 pp 2113ndash2121 2005

[21] N Ouchi S Kihara T Funahashi Y Matsuzawa and KWalshldquoObesity adiponectin and vascular inflammatory diseaserdquo Cur-rent Opinion in Lipidology vol 14 no 6 pp 561ndash566 2003

ISRN Obesity 7

0 2 4 6 8 10Day 0 2 4 6 8 10

PKC120575VIIIPKC120575I

Bcl-2

Bcl-xL

Caspase 9a

Caspase 9b

Glut4

Adiponectin

PPAR120574

120573-actin

60

50

40

30

20

10

0Day 0 Day 2 Day 4 Day 6 Day 8 Day 10

PKC120575

exon

inclu

sion

()

Figure 4 Western blot analysis of differentiating subcutaneousobese preadipocytes from days 0 to 10 using antibodies as indicatedThe blots are representative of four experiments performed indi-vidually with similar results Graph represents percent PKC120575 exoninclusion calculated as PKC120575VIII(PKC120575VIII + PKC120575I) times 100 and isrepresentative of four experiments performed separately

Lean Obese

(a)

40000

35000

30000

25000

20000

15000

10000

5000

0Lean Obese

Mea

n ad

ipoc

yte s

ize (120583

m)

(b)

Figure 5 Human lean or obese preadipocytes were differentiated invitro (a) Adipocytes on day 14 were stained for lipid content withOil Red O and (b) adipocyte size measured (methods) Experimentswere repeated 4 times with similar result

25

2

15

1

05

0

mRN

A ex

pres

sion

lowast

lowast

lowast

lowast

lowast

SIRT1 FOXO1 Adiponectin

LeanObese

PKC120575VIII PPAR120574

Figure 6 Lean and obese human preadipocytes were differentiatedin vitro Total RNA was collected on day 2 Real Time SYBR qPCRanalysis of PKC120575VIII SIRT1 FOXO1 adiponectin and PPAR120574was performed in triplicate and repeated 3 times in separateexperimentsThe absolute mRNA expression transcripts are shownlowast119875 lt 005 (by two-tailed Studentrsquos 119905-test) significant between leanand obese for each mRNA

marked decrease in FOXO SIRT1 adiponectin and PPAR120574expression compared to lean samples

4 Discussion

Obesity is an epidemic facing the world with the USAleading the statistics on increasing number of overweight andobese population Adipose tissue is an important endocrineregulator of energy homeostasis and glucose metabolismand the excess energy is stored in the white adipose tissueThe onset of adipogenesis and the transcriptional factorsthat initiate it have been studied in depth However withour current knowledge of the genome it is evident thatthere are several other mechanisms in play which regulatethe adipocyte phenotype during differentiation Apoptosis isrequired for fat cell turnover throughout our life Howeverfor weight loss in obese patients it may be beneficial toincrease adipocyte apoptosis along with therapies targetinghyperplasia and hypertrophy in adipocytes To our knowl-edge this is the first demonstration of alternative splicingevents during adipogenesis which contribute to the matureadipocyte phenotype Our results indicate a distinct splicingpattern between lean and obese patients

Apoptosis in preadipocytes and adipocytes was previ-ously observed in human cell lines However the under-lying mechanisms were not described Alternative splicinga posttranscriptional event is a powerful regulator of geneexpression We have identified key genes which are alter-natively spliced that dramatically switch their splice variantexpression during the progression of adipogenesis This mayexplain why preadipocytes are more susceptible to apoptoticagents while the mature adipocytes are resistant to it Protein

8 ISRN Obesity

kinase C delta (PKC120575) is a crucial signaling kinase affectingdownstream apoptotic pathways [8 12] Our data indicatesincreased expression of PKC120575VIII along with an early switchbetween the splice variants associated with apoptosis duringadipogenesis process of preadipocytes derived from obesepatients

PKC120575 is a serinethreonine kinase which plays a centralrole in apoptosis PKC120575 has dual functions as a mediator ofapoptosis and as an antiapoptosis effector Its splice variantsfunction as a switch that determines cell survival and fateThis could explain the opposing effects of PKC120575 on cellularapoptosis cited in the literature [13ndash15] Genetic switchesbased on alternative splicing are important for many cellularand developmental processes PKC120575I promotes apoptosiswhile PKC120575VIII promotes survival We have shown thatPKC120575II (the mouse splice variant) and PKC120575VIII functionas prosurvival proteins [8 12] the functions of other PKC120575splice variants are not yet established Our previous studiesin neurons have demonstrated that the prosurvival variant ofPKC120575 increases the expression of Bcl2 and Bcl-xL which areknown to increase cellular survival [16]

Obesity is associated with a chronic low inflamma-tory state due to the changes in secretory functions ofthe adipocytes Enlarged adipocytes (hypertrophy) lead toincreased adipocyte-derived free fatty acids which stimulatemacrophages In severely obese patientsmacrophages invadethe adipose tissue and release chemokines leading to necrosisand autophagy of adipocytes [17] These complications addto obese-associated metabolic syndrome diabetes insulinresistance and hepatic steatosis These forms of cell deathare distinct from apoptosis (programmed cell death) inadipocytes Apoptosis is shown to be substantially greaterin omental depots compared to subcutaneous depots inhealthy individuals [18] Here we show that obese samplesare resistant to apoptosis and express the antiapoptosis genesfourfold greater than in lean samples This may explain theincreased difficulty to lose weight and maintain the weightloss in overweight and obese individuals Larger adipocytesare an indicator of type II diabetes independent of insulinresistance [19] Previous studies showed that PPAR120574 andadiponectin expression was lower in obesity and type IIdiabetes [20 21] Our data indicated that obese patients havedysregulated adipogenesis program This may be modulatedby the signaling kinase PKC120575VIII which is overexpressedin obese individuals Our data also indicated dysregulatedalternative splicing of apoptosis genes in obese differentiatingcells This aberrant molecular profile may be a causativeeffect leading to obesity or may be a consequence of obesityand its comorbidities We are evaluating this further in ourlaboratory Here we have defined the differences in geneexpression between lean and obese adipocytes

In conclusion it may be possible to modulate alternativesplicing during adipogenesis for therapeutic treatments ofobesity Our results suggest a novel role for a key signalingkinase PKC120575VIII in the process of adipogenesis Expressionlevels of PKC120575VIII splice variant are increased in obesity andmay be linked to the cause or consequence of dysregulatedadipogenesis in obese patients Since new adipocytes arerequired for the maintenance and the endocrine function of

adipocytes balancing the gene expression of apoptotic genesmay be a better approach rather than complete inhibition ofadipogenesis

Conflict of Interests

The authors have no conflict of interests to declare

Authorsrsquo Contribution

GCarter andAApostolatos contributed equally to the paper

Acknowledgments

This work was supported by the Department of VeteransAffairs Medical Research Grant (NAP)The contents do notrepresent the views of the Department of Veterans Affairs ortheUnited States GovernmentThe authors thankDr AndreaMoor for help with microscopic images and adipocyte sizedetermination They thank Sean Yoder from Moffitt arraycore for array analysis

References

[1] S Rodrıguez-Acebes N Palacios J I Botella-Carretero et alldquoGene expression profiling of subcutaneous adipose tissue inmorbid obesity using a focused microarray distinct expressionof cell-cycle- and differentiation-related genesrdquo BMC MedicalGenomics vol 3 article 61 2010

[2] M Guzey D Jukic J Arlotti M Acquafondata R Dhir andR H Getzenberg ldquoIncreased apoptosis of periprostatic adiposetissue in VDR null micerdquo Journal of Cellular Biochemistry vol93 no 1 pp 133ndash141 2004

[3] P Fischer-Posovszky H Tornqvist K M Debatin and MWabitsch ldquoInhibition of death-receptor mediated apoptosis inhuman adipocytes by the insulin-like growth factor I (IGF-I)IGF-I receptor autocrine circuitrdquo Endocrinology vol 145 no4 pp 1849ndash1859 2004

[4] O Ishiko T Sumi H Yoshida Y Hyun and S Ogita ldquoCom-parison of expression of apoptosis regulatory proteins in theadipose tissue of tumor-bearing and diet-restricted rabbitsrdquoInternational Journal of Molecular Medicine vol 8 no 5 pp543ndash547 2001

[5] M A Della-Fera Y H Choi D L Hartzell J Duan MHamrick and C A Baile ldquoSensitivity of obob mice to Leptin-induced adipose tissue apoptosisrdquo Obesity Research vol 13 no9 pp 1540ndash1547 2005

[6] D H Kim S C Woods and R J Seeley ldquoPeptide designedto elicit apoptosis in adipose tissue endothelium reduces foodintake and body weightrdquo Diabetes vol 59 no 4 pp 907ndash9152010

[7] W C Earnshaw L M Martins and S H Kaufmann ldquoMam-malian caspases structure activation substrates and functionsduring apoptosisrdquo Annual Review of Biochemistry vol 68 pp383ndash424 1999

[8] K Jiang A H Apostolatos T Ghansah et al ldquoIdentificationof a novel antiapoptotic human protein kinase C 120575 isoformPKC120575VIII in NT2 cellsrdquo Biochemistry vol 47 no 2 pp 787ndash797 2008

ISRN Obesity 9

[9] H Apostolatos A Apostolatos T Vickers et al ldquoVitamin Ametabolite all-trans-retinoic acid mediates alternative splicingof protein kinase C 120575VIII (PKC120575VIII) isoform via splicingfactor SC35rdquo Journal of Biological Chemistry vol 285 no 34pp 25987ndash25995 2010

[10] D Y Jun C R Han M S Choi M A Bae M H Woo and YH Kim ldquoEffect of mollugin on apoptosis and adipogenesis of3T3-L1 preadipocytesrdquo Phytotherapy Research vol 25 no 5 pp724ndash731 2011

[11] S M Kim H S Park D Y Jun et al ldquoMollugin induces apop-tosis in human Jurkat T cells through endoplasmic reticulumstress-mediated activation of JNK and caspase-12 and subse-quent activation of mitochondria-dependent caspase cascaderegulated by Bcl-xLrdquo Toxicology and Applied Pharmacology vol241 no 2 pp 210ndash220 2009

[12] N A Patel S S Song and D R Cooper ldquoPKC120575 alternativelyspliced isoforms modulate cellular apoptosis in retinoic acid-induced differentiation of human NT2 cells and mouse embry-onic stem cellsrdquo Gene Expression vol 13 no 2 pp 73ndash84 2005

[13] J J Peluso A Pappalardo and G Fernandez ldquoBasic fibrob-last growth factor maintains calcium homeostasis and gran-ulosa cell viability by stimulating calcium efflux via a PKC120575-dependent pathwayrdquo Endocrinology vol 142 no 10 pp 4203ndash4211 2001

[14] L E Kilpatrick J Y Lee K M Haines D E Campbell KE Sullivan and H M Korchak ldquoA role for PKC-120575 and PI3-kinase in TNF-120572-mediated antiapoptotic signaling in thehuman neutrophilrdquo The American Journal of Physiology vol283 no 1 pp C48ndashC57 2002

[15] A Zrachia M Dobroslav M Blass et al ldquoInfection of gliomacells with Sindbis virus induces selective activation and tyrosinephosphorylation of protein kinase C 120575 implications for sindbisvirus-induced apoptosisrdquo Journal of Biological Chemistry vol277 no 26 pp 23693ndash23701 2002

[16] A Apostolatos S Song S Acosta et al ldquoInsulin promotesneuronal survival via the alternatively spliced protein kinaseCdeltaII isoformrdquo Journal of Biological Chemistry vol 287 pp9299ndash9310 2012

[17] K J Strissel Z Stancheva H Miyoshi et al ldquoAdipocytedeath adipose tissue remodeling and obesity complicationsrdquoDiabetes vol 56 no 12 pp 2910ndash2918 2007

[18] D Papineau A Gagnon and A Sorisky ldquoApoptosis of humanabdominal preadipocytes before and after differentiation intoadipocytes in culturerdquo Metabolism vol 52 no 8 pp 987ndash9922003

[19] C Weyer J E Foley C Bogardus P A Tataranni and RE Pratley ldquoEnlarged subcutaneous abdominal adipocyte sizebut not obesity itself predicts type II diabetes independent ofinsulin resistancerdquo Diabetologia vol 43 no 12 pp 1498ndash15062000

[20] L B Tanko A Siddiq C Lecoeur et al ldquoACDCadiponectinand PPAR-120574 gene polymorphisms implications for features ofobesityrdquo Obesity Research vol 13 no 12 pp 2113ndash2121 2005

[21] N Ouchi S Kihara T Funahashi Y Matsuzawa and KWalshldquoObesity adiponectin and vascular inflammatory diseaserdquo Cur-rent Opinion in Lipidology vol 14 no 6 pp 561ndash566 2003

8 ISRN Obesity

kinase C delta (PKC120575) is a crucial signaling kinase affectingdownstream apoptotic pathways [8 12] Our data indicatesincreased expression of PKC120575VIII along with an early switchbetween the splice variants associated with apoptosis duringadipogenesis process of preadipocytes derived from obesepatients

PKC120575 is a serinethreonine kinase which plays a centralrole in apoptosis PKC120575 has dual functions as a mediator ofapoptosis and as an antiapoptosis effector Its splice variantsfunction as a switch that determines cell survival and fateThis could explain the opposing effects of PKC120575 on cellularapoptosis cited in the literature [13ndash15] Genetic switchesbased on alternative splicing are important for many cellularand developmental processes PKC120575I promotes apoptosiswhile PKC120575VIII promotes survival We have shown thatPKC120575II (the mouse splice variant) and PKC120575VIII functionas prosurvival proteins [8 12] the functions of other PKC120575splice variants are not yet established Our previous studiesin neurons have demonstrated that the prosurvival variant ofPKC120575 increases the expression of Bcl2 and Bcl-xL which areknown to increase cellular survival [16]

Obesity is associated with a chronic low inflamma-tory state due to the changes in secretory functions ofthe adipocytes Enlarged adipocytes (hypertrophy) lead toincreased adipocyte-derived free fatty acids which stimulatemacrophages In severely obese patientsmacrophages invadethe adipose tissue and release chemokines leading to necrosisand autophagy of adipocytes [17] These complications addto obese-associated metabolic syndrome diabetes insulinresistance and hepatic steatosis These forms of cell deathare distinct from apoptosis (programmed cell death) inadipocytes Apoptosis is shown to be substantially greaterin omental depots compared to subcutaneous depots inhealthy individuals [18] Here we show that obese samplesare resistant to apoptosis and express the antiapoptosis genesfourfold greater than in lean samples This may explain theincreased difficulty to lose weight and maintain the weightloss in overweight and obese individuals Larger adipocytesare an indicator of type II diabetes independent of insulinresistance [19] Previous studies showed that PPAR120574 andadiponectin expression was lower in obesity and type IIdiabetes [20 21] Our data indicated that obese patients havedysregulated adipogenesis program This may be modulatedby the signaling kinase PKC120575VIII which is overexpressedin obese individuals Our data also indicated dysregulatedalternative splicing of apoptosis genes in obese differentiatingcells This aberrant molecular profile may be a causativeeffect leading to obesity or may be a consequence of obesityand its comorbidities We are evaluating this further in ourlaboratory Here we have defined the differences in geneexpression between lean and obese adipocytes

In conclusion it may be possible to modulate alternativesplicing during adipogenesis for therapeutic treatments ofobesity Our results suggest a novel role for a key signalingkinase PKC120575VIII in the process of adipogenesis Expressionlevels of PKC120575VIII splice variant are increased in obesity andmay be linked to the cause or consequence of dysregulatedadipogenesis in obese patients Since new adipocytes arerequired for the maintenance and the endocrine function of

adipocytes balancing the gene expression of apoptotic genesmay be a better approach rather than complete inhibition ofadipogenesis

Conflict of Interests

The authors have no conflict of interests to declare

Authorsrsquo Contribution

GCarter andAApostolatos contributed equally to the paper

Acknowledgments

This work was supported by the Department of VeteransAffairs Medical Research Grant (NAP)The contents do notrepresent the views of the Department of Veterans Affairs ortheUnited States GovernmentThe authors thankDr AndreaMoor for help with microscopic images and adipocyte sizedetermination They thank Sean Yoder from Moffitt arraycore for array analysis

References

[1] S Rodrıguez-Acebes N Palacios J I Botella-Carretero et alldquoGene expression profiling of subcutaneous adipose tissue inmorbid obesity using a focused microarray distinct expressionof cell-cycle- and differentiation-related genesrdquo BMC MedicalGenomics vol 3 article 61 2010

[2] M Guzey D Jukic J Arlotti M Acquafondata R Dhir andR H Getzenberg ldquoIncreased apoptosis of periprostatic adiposetissue in VDR null micerdquo Journal of Cellular Biochemistry vol93 no 1 pp 133ndash141 2004

[3] P Fischer-Posovszky H Tornqvist K M Debatin and MWabitsch ldquoInhibition of death-receptor mediated apoptosis inhuman adipocytes by the insulin-like growth factor I (IGF-I)IGF-I receptor autocrine circuitrdquo Endocrinology vol 145 no4 pp 1849ndash1859 2004

[4] O Ishiko T Sumi H Yoshida Y Hyun and S Ogita ldquoCom-parison of expression of apoptosis regulatory proteins in theadipose tissue of tumor-bearing and diet-restricted rabbitsrdquoInternational Journal of Molecular Medicine vol 8 no 5 pp543ndash547 2001

[5] M A Della-Fera Y H Choi D L Hartzell J Duan MHamrick and C A Baile ldquoSensitivity of obob mice to Leptin-induced adipose tissue apoptosisrdquo Obesity Research vol 13 no9 pp 1540ndash1547 2005

[6] D H Kim S C Woods and R J Seeley ldquoPeptide designedto elicit apoptosis in adipose tissue endothelium reduces foodintake and body weightrdquo Diabetes vol 59 no 4 pp 907ndash9152010

[7] W C Earnshaw L M Martins and S H Kaufmann ldquoMam-malian caspases structure activation substrates and functionsduring apoptosisrdquo Annual Review of Biochemistry vol 68 pp383ndash424 1999

[8] K Jiang A H Apostolatos T Ghansah et al ldquoIdentificationof a novel antiapoptotic human protein kinase C 120575 isoformPKC120575VIII in NT2 cellsrdquo Biochemistry vol 47 no 2 pp 787ndash797 2008

ISRN Obesity 9

[9] H Apostolatos A Apostolatos T Vickers et al ldquoVitamin Ametabolite all-trans-retinoic acid mediates alternative splicingof protein kinase C 120575VIII (PKC120575VIII) isoform via splicingfactor SC35rdquo Journal of Biological Chemistry vol 285 no 34pp 25987ndash25995 2010

[10] D Y Jun C R Han M S Choi M A Bae M H Woo and YH Kim ldquoEffect of mollugin on apoptosis and adipogenesis of3T3-L1 preadipocytesrdquo Phytotherapy Research vol 25 no 5 pp724ndash731 2011

[11] S M Kim H S Park D Y Jun et al ldquoMollugin induces apop-tosis in human Jurkat T cells through endoplasmic reticulumstress-mediated activation of JNK and caspase-12 and subse-quent activation of mitochondria-dependent caspase cascaderegulated by Bcl-xLrdquo Toxicology and Applied Pharmacology vol241 no 2 pp 210ndash220 2009

[12] N A Patel S S Song and D R Cooper ldquoPKC120575 alternativelyspliced isoforms modulate cellular apoptosis in retinoic acid-induced differentiation of human NT2 cells and mouse embry-onic stem cellsrdquo Gene Expression vol 13 no 2 pp 73ndash84 2005

[13] J J Peluso A Pappalardo and G Fernandez ldquoBasic fibrob-last growth factor maintains calcium homeostasis and gran-ulosa cell viability by stimulating calcium efflux via a PKC120575-dependent pathwayrdquo Endocrinology vol 142 no 10 pp 4203ndash4211 2001

[14] L E Kilpatrick J Y Lee K M Haines D E Campbell KE Sullivan and H M Korchak ldquoA role for PKC-120575 and PI3-kinase in TNF-120572-mediated antiapoptotic signaling in thehuman neutrophilrdquo The American Journal of Physiology vol283 no 1 pp C48ndashC57 2002

[15] A Zrachia M Dobroslav M Blass et al ldquoInfection of gliomacells with Sindbis virus induces selective activation and tyrosinephosphorylation of protein kinase C 120575 implications for sindbisvirus-induced apoptosisrdquo Journal of Biological Chemistry vol277 no 26 pp 23693ndash23701 2002

[16] A Apostolatos S Song S Acosta et al ldquoInsulin promotesneuronal survival via the alternatively spliced protein kinaseCdeltaII isoformrdquo Journal of Biological Chemistry vol 287 pp9299ndash9310 2012

[17] K J Strissel Z Stancheva H Miyoshi et al ldquoAdipocytedeath adipose tissue remodeling and obesity complicationsrdquoDiabetes vol 56 no 12 pp 2910ndash2918 2007

[18] D Papineau A Gagnon and A Sorisky ldquoApoptosis of humanabdominal preadipocytes before and after differentiation intoadipocytes in culturerdquo Metabolism vol 52 no 8 pp 987ndash9922003

[19] C Weyer J E Foley C Bogardus P A Tataranni and RE Pratley ldquoEnlarged subcutaneous abdominal adipocyte sizebut not obesity itself predicts type II diabetes independent ofinsulin resistancerdquo Diabetologia vol 43 no 12 pp 1498ndash15062000

[20] L B Tanko A Siddiq C Lecoeur et al ldquoACDCadiponectinand PPAR-120574 gene polymorphisms implications for features ofobesityrdquo Obesity Research vol 13 no 12 pp 2113ndash2121 2005

[21] N Ouchi S Kihara T Funahashi Y Matsuzawa and KWalshldquoObesity adiponectin and vascular inflammatory diseaserdquo Cur-rent Opinion in Lipidology vol 14 no 6 pp 561ndash566 2003

ISRN Obesity 9

[9] H Apostolatos A Apostolatos T Vickers et al ldquoVitamin Ametabolite all-trans-retinoic acid mediates alternative splicingof protein kinase C 120575VIII (PKC120575VIII) isoform via splicingfactor SC35rdquo Journal of Biological Chemistry vol 285 no 34pp 25987ndash25995 2010

[10] D Y Jun C R Han M S Choi M A Bae M H Woo and YH Kim ldquoEffect of mollugin on apoptosis and adipogenesis of3T3-L1 preadipocytesrdquo Phytotherapy Research vol 25 no 5 pp724ndash731 2011

[11] S M Kim H S Park D Y Jun et al ldquoMollugin induces apop-tosis in human Jurkat T cells through endoplasmic reticulumstress-mediated activation of JNK and caspase-12 and subse-quent activation of mitochondria-dependent caspase cascaderegulated by Bcl-xLrdquo Toxicology and Applied Pharmacology vol241 no 2 pp 210ndash220 2009

[12] N A Patel S S Song and D R Cooper ldquoPKC120575 alternativelyspliced isoforms modulate cellular apoptosis in retinoic acid-induced differentiation of human NT2 cells and mouse embry-onic stem cellsrdquo Gene Expression vol 13 no 2 pp 73ndash84 2005

[13] J J Peluso A Pappalardo and G Fernandez ldquoBasic fibrob-last growth factor maintains calcium homeostasis and gran-ulosa cell viability by stimulating calcium efflux via a PKC120575-dependent pathwayrdquo Endocrinology vol 142 no 10 pp 4203ndash4211 2001

[14] L E Kilpatrick J Y Lee K M Haines D E Campbell KE Sullivan and H M Korchak ldquoA role for PKC-120575 and PI3-kinase in TNF-120572-mediated antiapoptotic signaling in thehuman neutrophilrdquo The American Journal of Physiology vol283 no 1 pp C48ndashC57 2002

[15] A Zrachia M Dobroslav M Blass et al ldquoInfection of gliomacells with Sindbis virus induces selective activation and tyrosinephosphorylation of protein kinase C 120575 implications for sindbisvirus-induced apoptosisrdquo Journal of Biological Chemistry vol277 no 26 pp 23693ndash23701 2002

[16] A Apostolatos S Song S Acosta et al ldquoInsulin promotesneuronal survival via the alternatively spliced protein kinaseCdeltaII isoformrdquo Journal of Biological Chemistry vol 287 pp9299ndash9310 2012

[17] K J Strissel Z Stancheva H Miyoshi et al ldquoAdipocytedeath adipose tissue remodeling and obesity complicationsrdquoDiabetes vol 56 no 12 pp 2910ndash2918 2007

[18] D Papineau A Gagnon and A Sorisky ldquoApoptosis of humanabdominal preadipocytes before and after differentiation intoadipocytes in culturerdquo Metabolism vol 52 no 8 pp 987ndash9922003

[19] C Weyer J E Foley C Bogardus P A Tataranni and RE Pratley ldquoEnlarged subcutaneous abdominal adipocyte sizebut not obesity itself predicts type II diabetes independent ofinsulin resistancerdquo Diabetologia vol 43 no 12 pp 1498ndash15062000

[20] L B Tanko A Siddiq C Lecoeur et al ldquoACDCadiponectinand PPAR-120574 gene polymorphisms implications for features ofobesityrdquo Obesity Research vol 13 no 12 pp 2113ndash2121 2005

[21] N Ouchi S Kihara T Funahashi Y Matsuzawa and KWalshldquoObesity adiponectin and vascular inflammatory diseaserdquo Cur-rent Opinion in Lipidology vol 14 no 6 pp 561ndash566 2003