Post on 20-May-2020
Long term survival of high quality sperm: Insights 1
into the sperm proteome of the honeybee Apis 2
mellifera 3 4
5 6
7 8
Reza Zareie1,2, Holger Eubel 1,3, A. Harvey Millar1,2 & 9
Boris Baer1,2,4* 10
11 12 13
1 Centre for Integrative Bee Research (CIBER), ARC CoE in Plant Energy Biology, MCS 14 Building M316, The University of Western Australia, 6009 Crawley, Australia 15 16 2 Centre for Comparative Analysis of Biomolecular Networks, MCS Building M316, The 17 University of Western Australia, 6009 Crawley, Australia 18 19 3 Institute for Plant Genetics, Department of Plant Proteomcis, Leibniz University 20 Hannover, 30419 Hannover, Germany 21 22 4 Centre for Evolutionary Biology, School of Animal Biology (MO92), The University of 23 Western Australia, 6009 Crawley, Australia 24 25 26
E-mail contacts: 27
Reza Zareie: reza.zareie@uwa.edu.au 28
Holger Eubel: heubel@genetik.uni-hannover.de 29
Harvey Millar: harvey.millar@uwa.edu.au 30
*Boris Baer: boris.baer@uwa.edu.au (Corresponding author) 31
32
33
34
Proteomics of honeybee sperm 2
Abstract 35
In the social bees, ants and wasps, females (queens) only mate during a brief period early 36
in their lives and afterwards store a lifetime supply of sperm in a specialized organ, the 37
spermatheca. In some species, stored sperm can remain viable for several decades and is 38
used by queens to fertilize millions of eggs. The physiological adaptations that allow this 39
prolonged survival are unknown. To unravel them, we conducted proteomic analyses on 40
the sperm of the honeybee Apis mellifera to define proteins that are bee-specific or 41
highly-divergent from sequences in the sperm proteomes of flies or mammals, and which 42
might therefore be associated with long-term sperm survival. We identified a honeybee 43
sperm proteome of 343 members and define the subset of proteins or protein networks 44
that cannot be discerned in the sperm proteomes of fruit flies and humans. This subset 45
contained a significant number of proteins that are predicted to act in enzyme regulation, 46
or in nucleic acid binding and processing. From our analysis we conclude that long-term 47
survival of sperm in social insects could be underpinned by substantial changes in only a 48
specific subset of sperm proteins that allow physiological adaptation to storage. The 49
unexpected preponderance of proteins predicted to be involved in transcriptional 50
processes and enzyme regulation suggest these are the primary targets of this adaptation. 51
52
Keywords: Sperm viability, Sperm senescence, Comparative Proteomics, Sperm Proteins 53
54
Proteomics of honeybee sperm 3
Introduction 55
Sperm undergoes dramatic modifications during and after spermatogenesis making it a 56
highly specialized cell type 1, which evolved in response to natural and sexual selection to 57
maximize paternity success. The most obvious characteristic of sperm physiology is their 58
motility, but recent research has revealed that these cells are more than simple DNA 59
delivering vehicles 2. For example, sperm can react to environmental stimuli such as pH 60
and temperature and have evolved traits to compete against rival sperm as part of 61
postcopulatory sexual selection 1. The selective forces of natural and sexual selection 62
generated a spectacular variation in sperm form and functioning 3, 4 making sperm the 63
most diverse cell type known to date. We have ample knowledge about sperm 64
morphology and ultrastructure. However, sperm physiology and its molecular interplay 65
with male and female derived glandular secretions underlying the above-mentioned traits, 66
remain substantially understudied. 67
Eusocial hymenopteran insects, being the social bees, ants and wasps, provide unique 68
model systems to study the biology of sperm, since their social lifestyle resulted in the 69
evolution of a number of highly specialized reproductive traits 5, 6. For example, pair 70
formation and copulations only occur during a very short period and prior to colony 71
initiation 7, 8. As females (queens) never re-mate later in life they store a lifetime supply 72
of sperm in a specialized organ, the spermatheca. As a consequence social insect sperm 73
has to survive for prolonged periods of time, which can be up to several decades in some 74
ant species 9, 10 and queens have to economize sperm use so very few sperm are required 75
to fertilize an individual egg 10, 11. These characteristics of the mating biology of social 76
insects have resulted in the evolution of large ejaculate sizes and sperm of high quality 12, 77
Proteomics of honeybee sperm 4
13. Leaf cutting ants are a good example to illustrate the extreme levels of sperm survival, 78
quality and economy. In these species queens initially store several hundred million 79
sperm 14, 15, which bears significant costs for the queens 16. The sperm survives alongside 80
the queen for 20 or more years and is used to fertilize tens of millions of eggs 10. 81
The key molecular mechanisms of how social insect sperm are able to achieve such long-82
term storage and delay their senescence remain largely unknown. It seems obvious that 83
sperm must have evolved specific adaptations that allow them to achieve such 84
exceptional life history traits. Furthermore, such sperm traits have coevolved with the 85
physiological environments the sperm operate in such as the male’s seminal fluid 17 and 86
the secretions provided to sperm by queens during storage 18. The recent availability of 87
sperm proteomes of other species such as the fruit flies or humans now allows us to 88
compare sperm components from different species to identify the specific proteins or 89
protein networks that distinguish social insect sperm and underpin their long term 90
survival. 91
Here we used the honeybee Apis mellifera as a model where queens mate with up to 90 92
males during one of few nuptial flights. Ejaculates compete against each other for storage 93
inside the female’s sexual tract and proteins within the seminal fluid target sperm of rival 94
males and kill them, a process known as sperm incapacitation 19. Only about 3-5% or 95
around 6 million of the sperm initially acquired during copulations will ultimately 96
become stored in the spermatheca 5. This allows queens to consequently fertilize more 97
than 1.5 million eggs over a time frame of up to seven years. Whilst in storage, queens 98
support sperm through glandular secretions from the spermathecal gland 18, and sperm 99
appear to undergo further developmental processes to accommodate storage 20. We have 100
Proteomics of honeybee sperm 5
previously identified a number of honeybee sperm proteins as part of other experiments 101
17, 20, but these earlier datasets were too small to gain a general understanding of the full 102
componentry of bee sperm or to identify honeybee-specific sets of proteins that might be 103
present. Here we performed an extended proteomic analysis of honeybee sperm using gel 104
and non-gel based proteomic approaches with the aim of identifying a large number of 105
honeybee sperm proteins. We then used comparative approaches to identify a subset of 106
proteins, and predicted their associated biochemical functions, which are found in 107
honeybees and have not been found in sperm proteomes of fruit fly or human. 108
109
Material & Methods 110
Male breeding and sperm sampling 111
All males used for experimental work originated from colonies of Apis mellifera that we 112
kept in an animal yard at the University of Western Australia. We collected sperm from 113
males at an age of 2-3 weeks after enclosure to ensure that males had reached sexual 114
maturity. Sperm was collected by using a technique previously developed to artificially 115
inseminate honeybees 20, 21. In brief, males were anaesthetized with chloroform to initiate 116
male ejaculation, which was proceeded by gently squeezing the male abdomen between 117
two fingers. As soon as the ejaculate appeared at the tip of the male’s endophallus it was 118
collected with a glass capillary connected to a syringe. To separate sperm from its 119
surrounding seminal fluid, we applied a previously developed method 20. In short, semen 120
was diluted in Hayes solution (9.0 g/l NaCl, 0.2 g/l CaCl2, 0.2 g/l KCl, 0.1 g/l, NaHCO3, 121
pH 8.7), briefly mixed and centrifuged for 25 minutes at 3000 rpm (850 xg) and 4°C. 122
This procedure was repeated three times. Final sperm pellets were re-suspended in 50µl 123
Proteomics of honeybee sperm 6
of Hayes solution and frozen at -80°C prior to further analyses. To identify proteins 124
present in honeybee sperm we used both gel based and non-gel based approaches as 125
follows: 126
127
Gel based analysis of the honeybee sperm proteome 128
To separate sperm proteins on 2D-PAGE gels we used a previously developed protocol 17, 129
20. In brief 100 μl of sperm (equivalent to ~1 mg protein from ~250 drones) collected as 130
described above was acetone-precipitated, and dissolved in IEF solubilisation buffer. The 131
sample was loaded onto a 24-cm non-linear pH 3-11 IPG strip (GE) and run using the 132
following settings: 12 h at 30 V (rehydration step), 1 h at 500 V, 1 h gradient from 500 to 133
1000 V, 1 h gradient from 1000 to 3000 V, 2 h gradient from 3000 to 8000 V and 5 h at 134
8000 V. The strip was reduced and alkylated following the manufacturer instruction and 135
resolved on a 12% SDS polyacrylamide gel. Protein spots were visualized with 136
Coomassie Blue (G 250) colloidal staining. To identify proteins using tandem mass 137
spectrometry, individual spots were cut out of the gel, digested with Sequencing Grade 138
Modified Trypsin (Promega, V5111) and peptides identified on an Agilent LC/MSD Trap 139
XCT Ultra 6330 mass spectrometer coupled to an Agilent 1100 Series capillary LC 140
system. Peptides were resolved on a 0.5 x 50 mm Microsorb (Varian) C18 column eluted 141
over 15 minutes with a 5 to 60% acetonitrile gradient and 0.1% formic acid at 10 μl/min. 142
Eluents were sprayed into the mass spectrometer under positive ion mode via an ESI low-143
flow nebulizer and analyzed with the MS scan over 300 to 1400 m/z at the speed of 5500 144
m/z per second and ion charge control of 150000. Ions were selected for MS/MS after 145
reaching an intensity of 20000 cps. 146
Proteomics of honeybee sperm 7
147
Non-gel based analysis of the honeybee sperm proteome 148
Protein samples were also analyzed using peptide mixture LC-MS/MS analyses. To do 149
this we used 100 µl of sperm sample that we first centrifuged at 800 xg for 1 min. The 150
pellet was re-suspended in 500 µl of digestion buffer (10 mM NH4HCO3, 100 µg 151
Trypsin/ml) and incubated overnight at 37°C. This step was repeated in order to increase 152
the peptide yield. For the gel-free analysis disulfide-forming cysteine residues were 153
intentionally left unmodified to exclude highly abundant cysteine-rich protamine-like 154
proteins from subsequent experiments 23,25. After the second digestion, insoluble 155
components were removed by centrifugation at 20’000 xg for 5 min. The supernatant was 156
dried by vacuum centrifugation and stored at 4°C before peptide fractionation. For this, 157
the peptide pellet was suspended in SCX buffer A (10 mM KH2PO4 in 25% acetonitrile, 158
pH 3.0) and bound to the column according to the manufacturer’s instructions. Separation 159
of peptides into ten fractions was achieved by step-elution with 10%-100% of buffer B 160
(1M KCl in 10 mM KH2PO4 and 25% acetonitrile, pH 3.0) in buffer A on a SCX column 161
(4.6mm x 50mm, Optimize Technologies, Oregon, USA). The eluent fractions were 162
desalted using C18 cartridges (Nest Group, MA, USA) and dried by vacuum 163
centrifugation. We found measurable amounts of peptides in the fractions containing 164
10%, 20%, 30% and 40% of buffer B based on the Bradford assays and pooled all 165
remaining fractions into a single fraction. All fractions were analyzed using an Agilent 166
6510 Q-TOF mass spectrometer (Agilent Technologies) with an HPLC Chip Cube 167
source. The Chip consisted of a 40 nl enrichment column (Zorbax 300SB-C18 5 u) and a 168
150 mm separation column (Zorbax 300SB-C18 5 u) driven by an Agilent Technologies 169
Proteomics of honeybee sperm 8
1100 series nano/capillary liquid chromatography system. Both systems were controlled 170
by a MassHunter Workstation Data Acquisition for Q-TOF (ver B.01.02, Build 65.4, 171
Patches 1,2,3,4, Agilent Technologies). Peptides were eluted from the enrichment column 172
and run through the separation column during a 1h-gradient (15% (v/v) acetonitrile – 173
60% (v/v) acetonitrile containing 0.1% formic acid) directly into the mass spectrometer 174
running in positive ion mode and scanning over 275 to 1500 m/z at 4 spectra x sec-1. 175
Precursor ions were selected for auto MS/MS at an absolute threshold of 500 and a 176
relative threshold of 0.01, with maximum 3 precursors per cycle, and active exclusion set 177
at 2 spectra and released after 1 minute. Precursor charge-state selection and preference 178
was set to 2+ and then 3+ and precursors selected by charge then abundance. Resulting 179
MS/MS spectra were analyzed further using MassHunter Workstation Qualitative 180
Analysis software (ver B.01.02, Build 1.2.122.1, Patches 3 Agilent Technologies) and 181
MS/MS compounds detected by “Find Auto MS/MS” using default settings. The 182
resulting compounds were then exported as mzData.xml files for subsequent analyses as 183
outlined below. 184
Following an initial run in peptide mixture analysis, the resulting mzdata.xml files were 185
then searched against the honeybee proteome using Mascot 2.2. The resulting peptides 186
matching with ion scores ≥ 37 were exported along with their respective peptide charge 187
into a .csv file. This file was then used to construct an exclusion list, based on peptide 188
(m/z) and charge (z). Isolation width was set to ‘medium~4 m/z’ precursor type set to 189
‘Exclude’, retention time set to “0” and Δm/z set to ‘100 ppm’. This table was then 190
loaded into the MassHunter Workstation Data Acquisition for Q-TOF (ver B.01.02, Build 191
65.4, Patches 1, 2, 3, 4, Agilent Technologies) software and a next sample of peptides 192
Proteomics of honeybee sperm 9
was run on the mass spectrometer. After each MS-MS run the new list of excluded 193
peptides was added to the previous list and the new list loaded for the consequent runs. 194
195
Bioinformatics analyses 196
Spectra from the gel based experiment were analyzed using ProteinScape™ version 2.1.0 197
(Bruker Daltonics, Bremen, Germany) that in turn triggered Mascot 2.2 core algorithm 198
(Matrix Science) to match the data against the honeybee sequences (an in-house database 199
build from 10618 Apis mellifera protein sequences from NCBI RefSeq release 48 plus 200
common contaminants) with the following options: scoring: standard, enzyme: semi-201
trypsin with 1 missed cleavage, fixed modification: cysteine carbamidomethylation, 202
variable modifications: Met oxidation and Gln and Asn deamidation, precursor-ion 203
tolerance: 100 ppm, fragment-ion tolerance: 0.5 Da. To further improve the identification 204
confidence of proteins with significant Mascot scores below 70 and above 50, proteins 205
were only accepted if their observed molecular masses were within 30% of the calculated 206
values. Identifications with Mascot scores above 70 were accepted regardless of their 207
positions on the gel 57. 208
Spectra from the non-gel based analyses were converted using MSConvrt (ProteoWizard 209
release-2_1_2705 51) with the default options except for the vendor peak picking filtering. 210
Spectra were searched against honeybee sequences, as previously explained except that 211
decoy (reversed) sequences were also included, using Mascot 2.2, X! Tandem (GPM) and 212
Omssa (NCBI) with the following options: enzyme: semi-trypsin with 1 missed cleavage, 213
fixed modification: none, variable modifications: Met oxidation and Gln and Asn 214
deamidation, precursor-ion tolerance: 100 ppm for Mascot and 1.1 Da for X! Tandem and 215
Proteomics of honeybee sperm 10
Omssa, fragment-ion tolerance: 0.5 Da. Results were subsequently combined and 216
validated by the Trans Proteomic Pipeline (TPP) v4.4 with default conditions 54, 55. The 217
use of multiple search engines was found to increase the number of confidently identified 218
proteins compared to when only one search engine was used, similar to what has been 219
reported in the literature 55, 56. Protein identities with P values of at least 95% (or a false 220
discovery rate of ~0.1%) were accepted. 221
RefSeq sequence databases (release 48) were obtained from the NCBI Web site 222
(http://www.ncbi.nlm.nih.gov). Databases of the honeybee tissue proteome survey 22 and 223
a predicted honeybee proteome database were obtained from the Peptide Atlas site 224
(http://www.peptideatlas.org/). Fruit fly and human sperm protein lists were obtained 225
from the literature 23-25 while their sequences were acquired from the Internet sites of the 226
International Protein Index (ftp://ftp.ebi.ac.uk/pub/databases/IPI/) and FlyBase 227
(http://flybase.org/static_pages/downloads/ID.html). 228
Homology search and comparisons were conducted using BLAST 2.2.24+ release, 229
obtained from the NCBI FTP site (ftp://ftp.ncbi.nlm.nih.gov/blast/executables/blast+). To 230
determine homologous proteins in databases with 10’000 – 100’000 sequences we used a 231
cut-off value of E ≤ 1e-11 (for functional homology) or E ≤ 1e-23 (for the more stringent 232
bioprocess homology). When using databases with 1’000 to 10’000 sequences, we used a 233
cut-off value of E ≤ 1e-12 for functional homology and E ≤ 1e-24 for bioprocess 234
homology49 (and the statistics of sequence similarity scores in BLAST, see 235
http://www.ncbi.nlm.nih.gov/BLAST/tutorial/Altschul-1.html for further details). When 236
comparing two databases against each other, the smaller database was used as the query 237
and we only used the best match reported for each query. 238
Proteomics of honeybee sperm 11
Functional prediction analyses were performed using PANTHER, Pfam and Prosite. 239
PANTHER version 1.03 was downloaded from the PANTHER FTP site 240
(ftp://ftp.pantherdb.org/). Pfam version 1.3 was downloaded from the Sanger Institute 241
FTP site (ftp://ftp.sanger.ac.uk/pub/databases/Pfam). Prosite Revision: 1.75 was 242
downloaded from the Expasy FTP site (ftp://ftp.expasy.org/databases/prosite/). 243
The PANTHER database already contained GO terms but both Pfam and Prosite required 244
conversion tables to convert their functional predictions into GO terms. These conversion 245
tables and the GO standard definition file (Revision: 1.2329 dated 16:10:2011) were 246
downloaded from the GO consortium FTP site (ftp://ftp.geneontology.org/pub/go/). 247
When a GO term was assigned to a protein all its parent entries were also included in the 248
assignment. Redundant assignments resulting from the use of multiple predictive 249
algorithms were then removed and the resulting list was used for all subsequent analyses. 250
Pathways were mapped based on the Reactome Pathway Analysis tool 251
(http://www.reactome.org/ReactomeGWT/entrypoint.html) and Kegg Pathway Database 252
(http://www.kegg.jp/kegg/pathway.html). Reactome pathways are searchable by gene 253
names whereas Kegg pathways may be mapped using protein GI numbers. When needed, 254
translation between protein and gene identifiers were carried out using BridgeDB 255
(http://www.bridgedb.org/, 26). Alternatively PICR (http://www.ebi.ac.uk/Tools/picr/, 27) 256
was used. Given the small number of honeybee proteins present in Kegg and in 257
Reactome, Drosophila-homologues for honeybee proteins were used to conduct this 258
analysis. To do this, honeybee sperm sequences were searched against Drosophila total 259
proteins (RefSeq, 22316 sequences). Resulting unique sequences with E ≤ 1e-23 were 260
filtered by removing sequences that had less than 30% relative homology length (as 261
Proteomics of honeybee sperm 12
defined by BLAST) of either query or hit protein lengths or their size ratios were >200% 262
or <50% (amino acid number). A similar approach was also used to obtain Drosophila-263
homologues of all honeybee proteins (RefSeq) that were then used to predict the 264
maximum number of honeybee proteins in all pathways. 265
266
267
Results 268
1. The honeybee sperm proteome (AmSp) 269
We identified a total of 121 honeybee sperm proteins within the gel spots cut from SDS-270
PAGE gels and 277 proteins from the gel free LC – MS/MS analysis (Supplemental 271
Table S1 and Figure S1). Combining the two data sets resulted in a final list of 343 272
individual sperm proteins that we refer to as the Apis mellifera sperm proteome (AmSp) 273
below. The protein sequences for 92 of the AmSp are annotated as hypothetical proteins 274
with no known function (Supplemental Table S1). The remaining 251 AmSp proteins 275
include 53 mitochondrial proteins and 30 cytoskeletal proteins and contain a number of 276
sperm specific proteins such as testis-specific tubulin alpha chain (RefSeq: XP|396338.2), 277
outer dense fiber protein 3 and 3B (RefSeq: XP|001123232.2 and XP_001121292.2), 278
sperm-associated antigen 6 (RefSeq: XP|394968.3), dnaJ homolog subfamily B member 279
13 (RefSeq: XP|001123348.1), and organic solute carrier partner 1 (RefSeq: 280
XP|001121796.2). We found that only 47% of the AmSp had previously been reported 281
through proteomics analysis in other honeybee tissues and only 31% had been found in 282
proteomic analysis of honeybee testes 22. Consequently our analyses of purified honeybee 283
sperm samples provides a substantially enlarged set of proteins linked to the reproductive 284
Proteomics of honeybee sperm 13
biology of honeybees. 285
286
2. Interspecific comparisons of the AmSp 287
When we compared the AmSp with the sperm proteome of Drosophila melanogaster 288
(DmSp) and Homo sapiens (HsSp) we found functional homologues (as defined in 289
Materials and Methods) for only 49% (168/343) of these proteins within DmSp and 45% 290
(155/343) of these proteins within HsSp (Figure 1.A). The number of AmSp without 291
homologs in the other two species is substantially larger than expected by random from 292
the total protein complement of each species. When we compared these total protein sets 293
of the three species, D. melanogaster homologs were found for 78% (8267/10618) of all 294
A. mellifera predicted proteins, and a similar number of homologs were even found for H. 295
sapiens (72% homologs (7644/10618)) (Figure 1.B). We then divided the AmSp into two 296
groups. The AmSp group for which we found functional homologues in DmSp and HsSp 297
we termed the common protein set and the 136 AmSp without homologs in either DmSp 298
or HsSp we termed the specific protein set (Supplemental Table S2). Parallel with this the 299
common proteins are frequently found in other honeybee tissues (averaged 19.1 300
tissues/protein) whereas the specific ones are detectable in markedly fewer other tissues 301
(averaged 2.4 tissues/protein) (Table S2). 302
When we conducted a biochemical pathway analysis on the entire set of honeybee sperm 303
proteins (based on the Kegg and Reactome databases as outlined in Materials and 304
Methods) we identified at least 13 pathways that were highly represented as shown in the 305
Supplemental Table S3.A (based on a confidence level ≥ 99% and at least 4 identified 306
proteins per pathway, see Supplemental Table S3.B for complete listing of all 3 307
Proteomics of honeybee sperm 14
proteomes). We found that honeybee sperm contain proteins involved in energy and 308
amino acid metabolism, maintaining the cytoskeleton, protein folding and defenses 309
against oxidative stress. We also identified the number of human and fruit fly sperm 310
proteins that could also be mapped to these pathways (see Supplemental Table S3.A). 311
Compared to these well-studied proteomes, fewer honeybee proteins mapped to these 312
pathways, likely because of the overall smaller number of proteins we identified in 313
honeybee sperm. The only exception to this was the glycine, serine and threonine 314
metabolism pathway that appears to be over-represented in honeybee sperm, relative to 315
these other species. 316
Only 10 out of 57 honeybee sperm proteins mapped to these pathways belonged to the 317
specific protein set. In order to gain more information about the nature and possible 318
function of the latter proteins we conducted a further set of analyses based on gene 319
ontology that did not rely on pre-processed mapped sequences that were required for 320
attribution of proteins to known pathways. 321
322
3. Comparative gene ontology 323
Gene ontology (GO) analysis of three sperm proteomes revealed that 76% (259/343) of 324
AmSp, 81% of DmSp and 88% of HsSp had at least one GO annotation. Overall, the 325
major molecular functions and biological activities were similar between the proteomes, 326
indicating a substantial similarity in the machinery of the three sperm and thus their likely 327
physiological processes (Supplemental Tables S4.A and S5.A, see also Figures 2.A and 328
2.B). When we analyzed the molecular function terms of common and specific sets 329
separately we found that proteins with nucleotide (especially GTP) binding, nucleoside-330
Proteomics of honeybee sperm 15
triphosphatase, and peptidase activities were significantly more frequent in the common 331
AmSp while those with nucleic acid binding and enzyme regulator functions were 332
significantly over-represented in the specific AmSp set (Supplemental Table S4.B, see 333
Supplemental Table S6 for statistical analyses). Similarly, analyses of the biological 334
process terms for these protein sets indicated that two classes of processes related to 335
energy usage (glycolysis, alcohol, glucose and nucleotide metabolism), and ultra-336
structure (microtubule-based process, organelle morphogenesis and organization) are 337
found more often within the common set of honeybee sperm proteins. The specific sperm 338
protein set was enriched for proteins with biological process terms for cellular 339
macromolecule biosynthetic process, and nucleic acid (especially RNA) related processes 340
(Supplemental Table S5.B, also see Supplemental Tables S7.A and S7.B). Further 341
examination of the individual proteins in the latter group showed that these proteins can 342
be classified into two major sub-groups based on the predicted functions of their 343
domains. The first sub-group contained 14 proteins involved in nucleic acid protection 344
and expression, and a second sub-group of 6 proteins were involved in enzyme regulation 345
(Tables 1 and S7.B). 346
347
Discussion 348
Our identification of more than three hundred honeybee sperm proteins has allowed us to 349
gain new insight into the molecular machinery present in sperm and to provide a set of 350
analyses that can direct studies to define the mechanism of successful physiological 351
adaptation to long-term storage. Our comparative analysis found that a remarkably large 352
set of the proteins are highly specific to honeybee sperm. In most cases, these proteins 353
Proteomics of honeybee sperm 16
have not been identified previously through in-depth proteomic analysis in other tissues 354
of honeybees, including their reproductive organs, and could not readily be found to be 355
homologs of proteins in sperm of other species. We therefore provide the first empirical 356
evidence of the components in honeybee sperm that are likely to underpin its specific 357
adaptations in response to the demand to store large numbers of high quality sperm over 358
prolonged periods of time. The experimental data on the AmSp and its comparative 359
dissection provides the primary information required for future opportunities to 360
specifically study individual proteins of interest, and to unravel their effects on sperm 361
physiology and male reproductive success. 362
We found that the specific AmSp set contained honeybee sperm proteins that are 363
involved in nucleic acid interactions and enzyme regulation. Although these processes are 364
also important during mating or the early development of the zygote in all species, the 365
absence of clear homologs of the honeybee proteins in sperm of fruit flies and humans 366
suggests that these proteins are linked to the specific life history of honeybee sperm. 367
In general, sperm cells are characterised by a highly condensed nucleus and an extremely 368
reduced cytoplasm, which has led to the widespread idea that sperm is transcriptionally 369
and translationally silent. More recent studies, however, showed that mammalian 370
spermatozoa contain significant amounts of nuclear and mitochondrial mRNA 28-31 as 371
well as microRNA 32. Both mitochondrial and nuclear genes of mature sperm cells are 372
capable of generating transcripts 33, 34 suggesting that mRNA in sperm is not a leftover of 373
cytoplasmic condensation during spermatogenesis. Furthermore, such transcripts can be 374
translated into proteins, rather unexpectedly, by mitochondrial-type ribosomes 35, 36. 375
Finally, Liu et al. reported that sperm also contain the machinery that allows microRNA 376
Proteomics of honeybee sperm 17
to maturate and then form the RNA-induced silencing complex, RISC, which is important 377
for gene regulation and as an antiviral response 32. Consequently, while all sperm cells 378
seem to have some transcriptional or translational activity, its extent, biological relevance 379
and factors required for its maintenance during storage are unknown. Our identification 380
of a specific set of proteins in honeybee predicted to be involved in transcription and 381
RNA processing provide evidence of an adapted machinery for these processes that could 382
operate during storage. Further experimental work will be needed to confirm the activities 383
of these proteins and their role in honeybee sperm prior to, during and immediately after 384
storage. 385
In the specific AmSP group, a number of the proteins appear to be directly involved in 386
gene expression with four of them classified as transcription factors (TFs). These proteins 387
are expected to activate genes downstream and thereby act as initial triggers for more 388
complex physiological processes. Interestingly the presence of these transcription factors 389
and their involvement in gene regulation could explain why honeybee sperm is able to 390
survive in vitro with minimal intervention for up to nine months 37. In this context the 391
role of the sperm transcriptome may be also relevant. A survey of mammalian sperm 392
mRNA shows that a significant portion of the mRNAs code for nuclear and membrane 393
proteins involved in signal transduction as well as apoptosis and survival processes 29. 394
Although it was proposed that many of these mRNA are directly translated in the sperm 395
cell, it is impossible to discount their role early after fertilization. An example of such an 396
indirect role of sperm proteins was recently shown for the processing enzymes of a 397
microRNA that silences specific ovule genes post-fertilization 32. Alternatively, sperm 398
may contain a reservoir of (pioneer) transcription factors 48 required shortly after 399
Proteomics of honeybee sperm 18
fertilization to unlock either the sperm or ovule genetic archives. Long-lived transcription 400
factors, that maintain their function during storage, may thus be important for future 401
fertilization success. 402
The group of specific sperm proteins also contained a number of enzyme regulators that 403
have domains for signalling and signal transduction. Specifically, the enzyme regulator 404
GO grouping was dominated by GTPase regulators (RefSeq: Xp|00325039.1 405
XP|392541.4 and XP|397025.3) and kinase activity regulators (RefSeq: XP|001120113.1, 406
XP|396081.2 and XP|624139.3). All of these regulator proteins could be involved in 407
signalling pathways implicated in processes such as G-protein reception and vesicle 408
transport in sperm cells. Specifically, proteins with calcium binding and protein kinase A 409
(PKA) regulator activities may play roles in specific signalling steps in capacitation 410
and/or motility 38-40. While acrosome reaction and vesicular transport may be triggered by 411
G-protein coupled receptors and regulated by small GTPases 41-43. Alternatively, a 412
physiological role for these proteins during storage, or as long-lived agents for action 413
immediately after storage, is possible. Interestingly two of these proteins (RefSeq: 414
XP|003250394.1 and XP|001120113.1) are found to contain domains that would allow 415
the proteins to interact with each other, supporting the idea that this subset of proteins are 416
involved in related physiological processes. 417
A requirement in reducing sperm senescence in long term storage is to protect the 418
spermatozoa against oxidative damage. Both antioxidative enzyme activity and reduced 419
production of reactive oxygen species (ROS) associated with reduced metabolic rates are 420
implicated in achieving this 52,53. We found many antioxidative enzymes and one related 421
pathway in the honeybee sperm (Tables S1 and S3) but failed to identify any significant 422
Proteomics of honeybee sperm 19
differences between the antioxidative components of the 3 sperm proteomes. However, it 423
is conceivable that both translational/ transcriptional control and enzyme regulation 424
mechanisms are upstream elements for the control of the metabolic rate, ROS production, 425
antioxidative enzyme content, and also the DNA protective proteins reported here. A 426
protective system comprising of these elements should be able to (1.) maintain cell 427
viability while reducing its metabolic rate and thus limit the production of ROS during 428
the storage period; (2.) further protect the cellular components and its genetic load by 429
maintaining the levels of antioxidant and DNA protective elements in storage; and (3.) 430
when required (by appropriate signals) rapidly lift the cellular metabolic rate for 431
activation of the cellular functions, including the locomotion apparatus, while adjusting 432
the content of the cellular antioxidant and DNA protective proteins to a levels consistent 433
with the new ROS production. It could be also predicted that sperm with no need for long 434
term storage in their lifetime should possess a less sophisticated DNA-protecting 435
mechanism and less ability to significantly and safely alter their metabolic rates. 436
We consider that the use of multiple search engines and downstream statistical treatment 437
of the MS data to limit false-positive rate is a strong indicator of the presence of the 438
functional groups and pathways reported here. Future research on individual proteins of 439
interest could consequently benefit from the targeted proteomics approach used here, 440
such as multiple reaction monitoring (MRM) after the presence of individual proteins are 441
confirmed by appropriate techniques. 442
443
Conclusions: 444
Proteomics of honeybee sperm 20
We find that honeybee sperm consist of two distinct groups of proteins. Firstly there is 445
one group of well-known sperm proteins, which are clear homologs of proteins from 446
other species, and which are mainly involved in physiological processes that are 447
important for sperm metabolism and locomotion. However, we also find that more than 448
half of the sperm proteins we identified do not have homologs reported in sperm of other 449
species. A number of these honeybee specific sperm proteins are involved in transcription 450
and translation, enzyme regulation, and DNA protection. Consequently, a key adaptation 451
of social insect sperm to facilitate long term protection against cellular senescence could 452
be that these sperm possess a regulatory/signalling mechanism linked to a 453
translation/transcription control system with the ability to safely and quickly switch the 454
cellular metabolic rate from active to much less active with special attention to 455
adjustment of anti-senescence mechanisms, including antioxidative and DNA-protecting 456
components. The sperm of social insects, like other spermatozoa, is likely to have 457
maintained some level of de novo protein synthesis, during storage and/or immediately 458
afterwards but prior to fertilization but this is likely limited to absolutely crucial proteins 459
needed during storage. This represents a consolidating hypothesis of the relevant 460
observations to date and defines a path and a molecular component list for future research 461
aimed at better understanding the physiology of sperm storage in insects. 462
463
464
Additional data files 465
Further supplementary information is provided in an Excel file “Supplementary 466
Data.xls”. The file includes 11 Tables with the following content: 467
Proteomics of honeybee sperm 21
• Table S1: A complete list of all sperm proteins identified in honeybee sperm. 468
• Table S2: AmSp homologues in DmSp and HsSp; and AmSp proteins detected in 469
other honeybee tissues and as reported in Peptide Atlas). 470
• Table S3.A: Proteins mapped to over-represented pathways of honeybee sperm. 471
This table also include numbers of proteins mapped to corresponding pathways in 472
DmSp and HsSp. 473
• Table S3.B: Pathways predicted in AmSp, DmSp and HsSp. 474
• Table S4.A: Molecular functions predicted in AmSp, DmSp and HsSp. 475
• Table S4.B: Major molecular functions predicted in honeybee sperm proteins with 476
ratios of common and specific proteins. 477
• Table S5.A: Biological processes predicted in AmSp, DmSp and HsSp 478
• Table S5.B: Major biological processes predicted in honeybee sperm proteins 479
with ratios of common and specific proteins. 480
• Table S6: Fisher's exact tests for distribution of AmSp proteins mapped to 481
molecular functions and biological processes. 482
• Table S7.A: Proteins mapped to GO terms overrepresented in the common Sp 483
group. 484
• Table S7.B: Proteins mapped to GO terms overrepresented in the specific AmSp 485
group. 486
• Figure S1: The 2D PAGE used to identify honeybee sperm proteins. 487
The mass spectrometry proteomics data have been deposited to the ProteomeXchange 488
Consortium (http://proteomecentral.proteomexchange.org) via the PRIDE partner 489
repository with the dataset identifier PXD000169. 490
Proteomics of honeybee sperm 22
491
Acknowledgements 492
We were supported by the Australian Research Council (ARC) through a Queen 493
Elizabeth II and Future Fellowship to BB and an Australian Professorial and Future 494
Fellowship to AHM, an ARC Linkage Project to BB and AHM and the ARC Centre of 495
Excellence in Plant Energy Biology. We thank the honeybee keepers of Western 496
Australia, especially BetterBees of Western Australia for providing the necessary 497
honeybee material for this study. 498
499
Proteomics of honeybee sperm 23
References 500
1. Birkhead, T. R.; Hosken, D. J.; Pitnick, S., Sperm Biology: An evolutionary 501
perspective. Elsevier: Amsterdam, 2009; p 642. 502
2. Ainsworth, C., The secret life of sperm. Nature 2005, 436, 770-771. 503
3. Jamieson, B. G. M., The Ultrastructure and Phylogeny of Insect Spermatozoa. 504
Cambridge University Press: New York, Cambridge, 1987; p 320. 505
4. Jamieson, B. G. M.; Dallai, R.; Afzelius, B. A., Insects: Their spermatozoa and 506
phylogeny. Scientific Publishers: Enfield, NH, USA, 1999; p 564. 507
5. Baer, B., Sexual selection in Apis bees. Apidologie 2005, 36, 187-200. 508
6. Baer, B., The copulation biology of ants (Hymenoptera: Formicidae). Myrmecol. 509
News 2011, 14, 55-68. 510
7. Hölldobler, B.; Bartz, S. H., Sociobiology of reproduction in ants. In Experimental 511
Behavioural Ecology and Sociobiology, Hölldobler, B.; Lindauer, M., Eds. Gustav 512
Fischer Verlag: Stuttgart, New York, 1985; pp 237 - 257. 513
8. Boomsma, J. J.; Baer, B.; Heinze, J., The evolution of male traits in social insects. 514
Annu. Rev. Entomol. 2005, 50, 395-420. 515
9. Pamilo, P., Life span of queens in the ant Formica exsecta. Insect. Soc. 1991, 38, (2), 516
111-120. 517
10. den Boer, S. P. A.; Baer, B.; Dreier, S.; Aron, S.; Nash, D. R.; Boomsma, J. J., 518
Prudent sperm use by leaf-cutter ant queens. Proc. R. Soc. B-Biol. Sci. 2009, 276, 519
(1675), 3945-3953. 520
Proteomics of honeybee sperm 24
11. Tschinkel, W. R.; Porter, S. D., Efficiency of sperm use in queens of the fire ant, 521
Solenopsis invicta (Hymenoptera: Formicidae). Ann. Entomol. Soc. Am. 1988, 81, 522
(5), 777-781. 523
12. Hunter, F. M.; Birkhead, T. R., Sperm viability and sperm competition in insects. 524
Curr. Biol. 2002, 12, (2), 121-123. 525
13. den Boer, S. P. A.; Boomsma, J. J.; Baer, B., Seminal fluid enhances sperm viability 526
in the leafcutter ant Atta colombica. Behav. Ecol. Sociobiol. 2008, 62, 1843-1849. 527
14. Fjerdingstad, E. J.; Boomsma, J. J., Variation in size and sperm content of sexuals in 528
the leafcutter ant Atta colombica. Insect. Soc. 1997, 44, (3), 209-218. 529
15. Weber, N. A., Gardening ants: The attines. The American Philosophical Society: 530
Philadelphia, 1972; p 146. 531
16. Baer, B.; Armitage, S. A. O.; Boomsma, J. J., Sperm storage induces an immunity 532
cost in ants. Nature 2006, 441, 872-875. 533
17. Baer, B.; Heazlewood, J. L.; Taylor, N. L.; Eubel, H.; Millar, A. H., The seminal 534
fluid proteome of the honeybee Apis mellifera. Proteomics 2009, 9, 2085-2097. 535
18. Baer, B.; Eubel, H.; Taylor, N. L.; O'Toole, N.; Millar, A. H., Insights into female 536
sperm storage from the spermathecal fluid proteome of the honeybee Apis mellifera. 537
Genome Biology 2009, 10, R 67. 538
19. den Boer, S. P. A.; Baer, B.; Boomsma, J. J., Seminal fluid mediates ejaculate 539
competition in social insects. Science 2010, 327, 1506-1509. 540
20. Poland, V.; Eubel, H.; King, M.; Solheim, C.; Millar, A. H.; Baer, B., Stored sperm 541
differs from ejaculated sperm by proteome alterations associated with energy 542
metabolism in the honeybee Apis mellifera. Mol. Ecol. 2011, 20, 2643-2654. 543
Proteomics of honeybee sperm 25
21. Ruttner, F., The instrumental insemination of the queen bee. 2 ed.; Apimondia: 544
Bucharest, 1976; p 123. 545
22. Chan, Q. W. T.; Parker, R.; Sun, Z.; Deutsch, E. W.; Foster, L. J., A honey bee (Apis 546
mellifera L.) PeptideAtlas crossing castes and tissues. Bmc Genomics 2011, 12. 547
23. Dorus, S.; Busby, S. A.; Gerike, U.; Shabanowitz, J.; Hunt, D. F.; Karr, T. L., 548
Genomic and functional evolution of the Drosophila melanogaster sperm proteome. 549
Nat. Genet. 2006, 38, (12), 1440-1445. 550
24. Wasbrough, E. R.; Dorus, S.; Hester, S.; Howard-Murkin, J.; Lilley, K.; Wilkin, E.; 551
Polpitiya, A.; Petritis, K.; Karr, T. L., The Drosophila melanogaster sperm 552
proteome-II (DmSP-II). J. Proteomics 2010, 73, (11), 2171-2185. 553
25. Baker, M. A.; Reeves, G.; Hetherington, L.; Muller, J.; Baur, I.; Aitken, R. J., 554
Identification of gene products present in Triton X-100 soluble and insoluble 555
fractions of human spermatozoa lysates using LC-MS/MS analysis. Proteom. Clin. 556
Appl. 2007, 1, (5), 524-532. 557
26. van Iersel, M. P.; Pico, A. R.; Kelder, T.; Gao, J. J.; Ho, I.; Hanspers, K.; Conklin, B. 558
R.; Evelo, C. T., The BridgeDb framework: standardized access to gene, protein and 559
metabolite identifier mapping services. BMC Bioinformatics 2010, 11. 560
27. Cote, R. G.; Jones, P.; Martens, L.; Kerrien, S.; Reisinger, F.; Lin, Q.; Leinonen, R.; 561
Apweiler, R.; Hermjakob, H., The Protein Identifier Cross-Referencing (PICR) 562
service: reconciling protein identifiers across multiple source databases. BMC 563
Bioinformatics 2007, 8. 564
28. Alcivar, A. A.; Hake, L. E.; Millette, C. F.; Trasler, J. M.; Hecht, N. B., Mitochodrial 565
gene expression in male germ cells of the mouse. Dev. Biol. 1989, 135, (2), 263-271. 566
Proteomics of honeybee sperm 26
29. Ostermeier, G. C.; Dix, D. J.; Miller, D.; Khatri, P.; Krawetz, S. A., Spermatozoal 567
RNA profiles of normal fertile men. The Lancet 2002, 360, (9335), 772-777. 568
30. Yang, C. C.; Lin, Y. S.; Hsu, C. C.; Wu, S. C.; Lin, E. C.; Cheng, W. T. K., 569
Identification and sequencing of remnant messenger RNAs found in domestic swine 570
(Sus scrofa) fresh ejaculated spermatozoa. Anim. Reprod. Sci. 2009, 113, (1-4), 143-571
155. 572
31. Bonache, S.; Mata, A.; Ramos, M. D.; Bassas, L.; Larriba, S., Sperm gene expression 573
profile is related to pregnancy rate after insemination and is predictive of low 574
fecundity in normozoospermic men. Hum. Reprod. 2012, 27, (6), 1556-1567. 575
32. Liu, W. M.; Pang, R. T. K.; Chiu, P. C. N.; Wong, B. P. C.; Lao, K. Q.; Lee, K. F.; 576
Yeung, W. S. B., Sperm-borne microRNA-34c is required for the first cleavage 577
division in mouse. Proc. Natl. Acad. Sci. U. S. A. 2012, 109, (2), 490-494. 578
33. Hecht, N. B.; Williams, J. L., Synthesis of RNA by separated heads and tails from 579
bovine spermatozoa. Biol. Reprod. 1978, 19, (3), 573-579. 580
34. Wang, L. B.; Fan, J. S.; Yu, M. J.; Zheng, S. Y.; Zhao, Y. J., Association of goat 581
(Capra hircus) CD4 gene exon 6 polymorphisms with ability of sperm internalizing 582
exogenous DNA. Mol. Biol. Rep. 2011, 38, (3), 1621-1628. 583
35. Gur, Y.; Breitbart, H., Mammalian sperm translate nuclear-encoded proteins by 584
mitochondrial-type ribosomes. Genes Dev. 2006, 20, 411-416. 585
36. Gur, Y.; Breitbart, H., Protein synthesis in sperm: Dialog between mitochondria and 586
cytoplasm. Mol. Cell. Endocrinol. 2008, 282, (1-2), 45-55. 587
Proteomics of honeybee sperm 27
37. Poole, H. K.; Taber, S., In-vitro preservation of honeybee (Hymenoptera Apidae) 588
semen enhanced by storage at 13-15 degrees C. Ann. Entomol. Soc. Am. 1970, 63, 589
(6), 1673-&. 590
38. Ficarro, S.; Chertihin, O.; Westbrook, V. A.; White, F.; Jayes, F.; Kalab, P.; Marto, J. 591
A.; Shabanowitz, J.; Herr, J. C.; Hunt, D. F.; Visconti, P. E., Phosphoproteome 592
analysis of capacitated human sperm - Evidence of tyrosine phosphorylation of a 593
kinase-anchoring protein 3 and valosin-containing protein/p97 during capacitation. J. 594
Biol. Chem. 2003, 278, (13), 11579-11589. 595
39. Krapf, D.; Arcelay, E.; Wertheimer, E. V.; Sanjay, A.; Pilder, S. H.; Salicioni, A. M.; 596
Visconti, P. E., Inhibition of Ser/Thr phosphatases induces capacitation-associated 597
signaling in the presence of Src kinase inhibitors. J. Biol. Chem. 2010, 285, (11), 598
7977-7985. 599
40. Salicioni, A. M.; Platt, M. D.; Wertheimer, E. V.; Arcelay, E.; Allaire, A.; Sosnik, J.; 600
Visconti, P. E., Signalling pathways involved in sperm capacitation. Soc Reprod 601
Fertil Suppl. 2007, 65, 245-259. 602
41. Breitbart, H.; Spungin, B., The biochemistry of the acrosome reaction. Molecular 603
Human Reproduction 1997, 3, (3), 195-202. 604
42. De Blas, G. A.; Roggero, C. M.; Tomes, C. N.; Mayorga, L. S., Dynamics of 605
SNARE assembly and disassembly during sperm acrosomal exocytosis. PLoS. Biol. 606
2005, 3, (10), 1801-1812. 607
43. Branham, M. T.; Bustos, M. A.; De Blas, G. A.; Rehmann, H.; Zarelli, V. E. P.; 608
Trevino, C. L.; Darszon, A.; Mayorga, L. S.; Tomes, C. N., Epac activates the small 609
Proteomics of honeybee sperm 28
G proteins Rap1 and Rab3A to achieve exocytosis. J. Biol. Chem. 2009, 284, (37), 610
24825-24839. 611
44. Eklund, H.; Uhlin, U.; Färnegårdh, M.; Logan, D.T.; Nordlund, P., Structure and 612
function of the radical enzyme ribonucleotide reductase. Prog Biophys Mol Biol 613
2001, 77,(3), 177-268. 614
45. Tanaka, T.; Hosokawa, M.; Vagin, V. V. et al., Tudor domain containing 7 615
(Tdrd7) is essential for dynamic ribonucleoprotein (RNP) remodeling of 616
chromatoid bodies during spermatogenesis. PNAS 2011, 108, (26), 10579-10584. 617
46. Low, M. G., Glycosyl-phosphatidylinositol: a versatile anchor for cell surface 618
proteins. FASEB 1989, 3, (5), 1600-1605. 619
47. Toha, Y. K.; Kamariaha, N.; Maurer-Stroh, S. et al., Structural insight into the 620
glycosylphosphatidylinositol transamidase subunits PIG-K and PIG-S from yeast. 621
J Structural Biol 2011, 173, (2), 271–281. 622
48. Smale, S. T., Pioneer factors in embryonic stem cells and differentiation. Cur 623
Opin Gen Dev 2010, 20, 519–526. 624
49. Thomas, P. D; Kejariwal, A.; Guo, N.; Mi, H.; Campbell, M. J.; Muruganujan, A.; 625
Lazareva-Ulitsky, B., Applications for protein sequence–function evolution data: 626
mRNA/protein expression analysis and coding SNP scoring tools. Nuc Acid Res 627
2006, 34, W645–W650. 628
50. Lachke, S. A.; Alkuraya, F. S.; Kneeland, S. C.; Ohn, T. et al, Mutations in the 629
RNA granule component TDRD7 cause cataract and glaucoma. Science 2011, 630
331, (6024), 1571-157 631
Proteomics of honeybee sperm 29
51. Kessner, D.; Chambers, M.; Burke, R.; Agus, D.; Mallick, P., ProteoWizard: 632
Open Source Software for Rapid Proteomics Tools Development. Bioinformatics 633
2008, 24, (21), 2534-2536. 634
52. Collins, A.M.; Williams, V.; Evans, J.D., Sperm storage and antioxidative enzyme 635
expression in the honey bee, Apis mellifera. Insect Mol. Biol. 2004, 13, 141–146. 636
53. Ribou, A.C.; Reinhardt, K., Reduced metabolic rate and oxygen radicals 637
production in stored insect sperm. Proceedings of the Royal Society B 2011, 1098: 638
2422-2426. 639
54. Deutsch, E.W.; Mendoza, L.; Shteynberg, D.; Farrah, T.; Lam, H.; Tasman, N.; 640
Nilsson, E,; Pratt, B.; Prazen, B.; Eng, J.K.; Martin, D.B.; Nesvizhskii, A.I.; 641
Aebersold, R., A guided tour of the Trans‐Proteomic Pipeline. Proteomics 2010, 642
10(6): 1150-1159. 643
55. Shteynberg, D.; Deutsch, E.W.; Lam, H.; Eng, J.K.; Sun, Z.; Tasman, N.; 644
Mendoza, L.; Moritz, R.L.; Aebersold, R,; Nesvizhskii, A. I., iProphet: multi-level 645
integrative analysis of shotgun proteomic data improves peptide and protein 646
identification rates and error estimates. Molecular and Cellular Proteomics 2011, 647
10(12): M111.007690. 648
56. Kapp, E.A.; Schütz, F.; Connolly, L.M.; Chakel, J.A.; Meza, J.E.; Miller, C.A.; 649
Fenyo, D.; Eng, J.K; Adkins, J.N.; Omenn, G.S.; Simpson, R.J., An evaluation, 650
comparison, and accurate benchmarking of several publicly available MS/MS 651
search algorithms: sensitivity and specificity analysis. Proteomics 2005, 5(13): 652
3475-3490. 653
Proteomics of honeybee sperm 30
57. Baer, B.; Zareie, R.; Paynter, E.; Poland, V.; Millar, A.H., Seminal fluid proteins 654
differ in abundance between genetic lineages of honeybees. Journal of proteomics 655
2012, 75(18): 5646-5653. 656
657
Proteomics of honeybee sperm 31
658
Figure legends 659
660
661
Figure 1: Number of homologue proteins in sperm (A) and total (B) proteomes of the 662
honeybee, fruit fly and human 663
664
665
666
Figure 2: Major sub-groups of molecular functions (A) and biological processes (B) 667
identified in honeybee sperm proteome 668
669
Table 1: Major classes identified in the specific proteins of honeybee sperm proteome. Classifications derived on protein’s predicted functions are shown in the two left columns. GO terms abundantly found in the AmSp specific group are in the next five columns (see Supplemental Table S7b for further details). Protein accessions are provided in the next column. Predicted protein functions and references used to predict these functions are shown in the last two columns.
Classification GO
:000
3676
GO
:003
0234
GO
:003
4645
GO
:009
0304
GO
:001
6070
RefSeq accession Predicted function Reference
Nucleic acid
binding and/or
metabolism
DNA protection
X X X XP|003249954.1 DNA repair CDD
X X XP|001121495.2 DNA replication and repair 22, 44, 45
X XP|003251300.1 Binds DNA for protection or repair Prosite, Pfam
Transcription and splicing
X X X X XP|392622.4 Transcription regulation Prosite, Pfam, PANTHER
X X X X XP|001120219.2 Transcription regulation Prosite
X X X X XP|003250668.1 Transcription regulation Prosite, PANTHER
X X X X XP|392215.4 RNA polymerase PANTHER,
X X X XP|394637.2 Histone acetylation (transcription regulation) Prosite, Pfam, PANTHER
X X X XP|397019.4 mRNA processing (perhaps splicing) Prosite, Pfam, PANTHER
X X XP|003251829.1 mRNA splicing PANTHER
Translation and post-
translational modification
X X X X XP|001123169.2 Seryl-tRNA synthetase (translation) Prosite, Pfam, PANTHER
X X X XP|394641.3 Cysteinyl-tRNA synthetase (translation) Pfam, PANTHER
X XP|003249197.1 mRNA binding and regulation Pfam, PANTHER, 50
X XP|624738.2 PTM and expression regulation Pfam, PANTHER, 46, 47
Enzyme regulation
Small GTPase
regulation
X XP|003250394.1 Small GTPase regulator (EF-hand calcium-binding) Prosite, PANTHER
X XP|397025.3 Activation of small GTPases Prosite, Pfam, PANTHER
X XP|392541.4 Small GTPase regulation, vesicle transport Prosite, Pfam, PANTHER
Kinase activity
regulation
X XP|624129.3 E3 ligase, promoting Cdk activity Prosite, Pfam, PANTHER
X XP|001120113.1 Regulation of PKA (binds EF-hand proteins) Prosite, Pfam, PANTHER
X XP|396081.2 kinase activator Pfam, PANTHER