INTRODUCTION
Seminal fluid proteins of male insects have a multitude of functions and many alter the female reproductive physiology to the male's benefit (Wolfner, Reference Wolfner2002; Gillott, Reference Gillott2003; Chapman and Davies, Reference Chapman and Davies2004; Poiani, Reference Poiani2006). However, in Drosophila, seminal fluids also accelerated female mortality directly (Chapman et al. Reference Chapman, Hutchings and Partridge1993; Rice, Reference Rice1996). Because of such direct or indirect interference with the female reproductive output, i.e. her evolutionary fitness, seminal proteins have become the target of researchers interested in evolutionary biology and bioinformatics. Genes coding for seminal proteins evolve with an unusually rapid rate (Civetta and Singh, Reference Civetta and Singh1995; Swanson et al. Reference Swanson, Clark, Waldrip-Dail, Wolfner and Aquadro2001; Andres et al. Reference Andrés, Maroja, Bogdanowicz, Swanson and Harrison2006), like many other reproductive traits. Rapid evolution of reproductive traits has been shown to arise when male adaptations suppress female fitness (Rice, Reference Rice1992, Reference Rice1996).
For many insect taxa the knowledge about seminal fluids is relatively scarce and ‘pre-proteomic’, i.e. not amenable to bioinformatic analyses (e.g. Chen, Reference Chen1984; Lange and Loughton, Reference Lange and Loughton1984, Reference Lange and Loughton1985; Cheeseman and Gillott, Reference Cheeseman and Gillott1989; Hartmann and Loher, Reference Hartmann and Loher1999). Genome sequencing has verified evolutionary predictions about the rapid evolution and high diversity at the genetic and RNA level in Drosophila species (Fiumera et al. Reference Fiumera, Dumont and Clark2005; Mueller et al. Reference Mueller, Ravi-Ram, McGraw, Qazi, Siggia, Clark, Aquadro and Wolfner2005; Haerty et al. Reference Haerty, Jagadeeshan, Kulathinal, Wong, Ram, Sirot, Levesque, Artieri, Wolfner, Civetta and Singh2007) and Anopheles (Dottorini et al. Reference Dottorini, Nicolaides, Ranson, Rogers, Crisanti and Catteruccia2007). The coupling of mass spectrometry and bioinformatics tools (e.g. Domon and Aebersold, Reference Domon and Aebersold2006) will allow such tests at the protein level and will open avenues into research of other important organisms.
Here we address the suitability of mass spectrometry and bioinformatics to characterize the seminal proteome of the common bed bug, Cimex lectularius. This was stimulated by three visions. First, the current increase in bed bug infestations throughout much of the western world (see Reinhardt and Siva-Jothy, Reference Reinhardt and Siva-Jothy2007 for a review of such studies) calls for pest control measures that specifically target the reproduction of bed bugs. Seminal proteins may play an important role in such specific targets. It is, therefore, desirable that target proteins are those not found in other taxa, i.e. producing no returns in Drosophila databases.
Second, bed bugs are a model system for sexual conflict (Chapman, Reference Chapman2006) because several studies have detailed the co-evolution between male and female physiological, morphological and behavioural traits (Stutt and Siva-Jothy, Reference Stutt and Siva-Jothy2001; Morrow and Arnqvist, Reference Morrow and Arnqvist2003; Reinhardt et al. Reference Reinhardt, Naylor and Siva-Jothy2003, Reference Reinhardt, Naylor and Siva-Jothy2005; Siva-Jothy, Reference Siva-Jothy2006). As sexual conflict generally leads to very rapid evolution of reproductive traits (Rice, Reference Rice1992, Reference Rice1996; Holland and Rice, Reference Holland and Rice1998) seminal proteins were expected to evolve rapidly in bed bugs.
Third, the sperm biology of the bed bug is highly unusual (reviewed by Davis, Reference Davis and Usinger1966; Reinhardt and Siva-Jothy, Reference Reinhardt and Siva-Jothy2007). During copulation the male deposits sperm and seminal fluid into a female organ, the spermalege which is unique to the Cimicoidea. From the spermalege, sperm travel through the haemolymph of the female, through the walls of the oviduct until they reach either the ovaries or a secondary storage organ. During this passage sperm cells aggregate and re-aggregate and are being attacked by female cells (haemocytes) (see references in Davis, Reference Davis and Usinger1966; Reinhardt and Siva-Jothy, Reference Reinhardt and Siva-Jothy2007). While the interest in the seminal proteome of the honey bee is largely governed by identifying proteins that assist in the extraordinarily long sperm storage (Collins et al. Reference Collins, Caperna, Williams, Garrett and Evans2006 and references therein), seminal fluid properties of the bed bug are more likely to have evolved to assist the sperm in withstanding the rather harsh female environments.
MATERIALS AND METHODS
Study species and rearing
Three populations/species of bed bugs were used, 2 populations of the common bed bug, Cimex lectularius, henceforth Cimex A and Cimex B and Afrocimex constrictus, a species parasitic on fruit bat in Africa (Reinhardt et al. Reference Reinhardt, Naylor and Siva-Jothy2007).
Cimex A originates from a donation by the Medical Entomology Centre, Cambridge, UK in 1998. The bugs where maintained in an incubator at 26±1°C, at 70% relative humidity with a light cycle of L:D 12 h:12 h. Bugs were fed weekly using established protocols (Reinhardt et al. Reference Reinhardt, Naylor and Siva-Jothy2003) until 2 weeks after adult eclosion. Females were separated from males in order to ensure virginity. Males were not fed for 3 weeks just prior to the sampling of the seminal fluid.
Cimex B originates from a donation by Insect Control & Research, Inc. Maryland Inc. (USA) in June 2003. Individuals were cultured under identical conditions as Cimex A.
Individuals of a related species, Afrocimex constrictus, were collected in a cave in Kenya and transported live to the laboratory in the UK (see Reinhardt et al. Reference Reinhardt, Naylor and Siva-Jothy2007 for details). Eight males that were kept in sexual isolation for 2 months were killed and their seminal fluid containers dissected out and used for further analysis.
Bed bug anatomy
The anatomy of male Cimex lectularius has been detailed by Usinger (Reference Usinger1966). Briefly, in the male 1 seminal vesicle descends from each of the 2 testes. Sperm is constantly produced by the testes and sperm accumulates and is stored in the seminal vesicles (Fig. 1). Another paired structure, the accessory glands (or mesadenal glands sensuUsinger, Reference Usinger1966) produce the seminal fluid which is stored until copulation, separately from sperm, in a pair of containers called the seminal reservoir. During a 60-sec mating, males transfer approximately 20% of their seminal fluids to the female (Reinhardt, unpublished data). Female anatomy has also been reviewed by Usinger (Reference Usinger1966) and involves a secondary genitalic organ into which the male deposits his sperm. The sperm then travel freely through the female body (see Introduction section).
The reproductive tract of a virgin male bed bug, Cimex lectularius (30 days old). Note the sperm and seminal fluid containers are clearly separated.
Dissection
All Cimex males were sexually rested for 2 weeks, Afrocimex for 2 months (see above). Sperm and seminal fluid containers are well separated and can be dissected out separately (Fig. 1). The containers were placed into an Eppendorf tube on ice until used further.
Measurement of protein content
Protein content was determined using the Bradford method. Fifteen accessory containers (removed from 17 males) in Cimex A, 8 containers removed from 8 males in Afrocimex constrictus and 9 containers from 9 males in Cimex B. All males had not fed for 4 weeks. Containers were placed in 100 μl of PBS with protease inhibitors (2 mm PMSF and 1 mm EDTA) in Eppendorf tubes. The tubes were briefly vortexed, spun down and the supernatant used for the Bradford reaction. Distilled water was added to the pellet and all samples were placed in a hotblock (100°C) in order to denature the proteins.
No attempt was made to measure the protein content of sperm but individual males of this age contain approximately 0·08 mg of sperm dry weight (K. Reinhardt, unpublished observations).
1-D gel electrophoresis
One third of each of the samples was run on a ready-made gel (NuPage, Invitrogen) at 200 V for 40 min. The gel was Coomassie Blue stained and fixed with acetic acid. The gel was scanned and the image saved as a .jpg file. Bands were scored independently by 2 observers (K.R., S.A.G.) after manually adjusting the contrasts and the brightness using Photoshop C52 (Adobe).
2-D electrophoresis
Twenty accessory containers of Cimex A and 20 containers from Cimex B were used.
In order to get information about how many proteins found in the seminal containers are also found in the female, we dissected the spermalege from 19 virgin females and from 20 females 30–45 min after mating (a time at which sperm and seminal fluid are still in the spermalege). The spermalege tissues of all virgin females were pooled as were the tissues of all mated females. During mating sperm is obviously passed to female alongside seminal fluid. We therefore used sperm from 3 males in strain A and used this in the electrophoresis. Note that the analysis of sperm proteins was not intended (as by Dorus et al. Reference Dorus, Busby, Gerike, Shabanowitz, Hunt and Karr2006) but merely used as a control to see whether any proteins detected in the mated female by our method could have been attached to or transferred by sperm.
The organs were briefly rinsed in PBS and placed in 100 μl of PBS containing protease inhibitors (2 mm PMSF and 1 mm EDTA) in Eppendorf tubes. The tubes were vortexed for a few sec and centrifuged for 30 sec at 20 800 g to separate cellular components. The supernatant was frozen at −80°C until further use.
Proteins were resolved in the first dimension by isoelectric focusing (IEF) using 18 cm, pH 3-10 IPG strips (Amersham Biosciences) for a total of 33 500 Vh, in an IPGphor Isolectric Focusing system (Amersham Biosciences). The IEF program started with 500 V for 500 Vh, followed by a step-and-hold increase to 1000 V for 1000 Vh, and finally to a step-and-hold increase to 8000 V for 32 000 Vh. After focusing, IPG strips were equilibrated to reduce protein disulfide bonds in 10 ml of equilibrating solution per strip (6 m urea, 30% (v/v) glycerol, 2% (w/v) SDS, 50 mm Tris-HCl, pH 8·8, and 0·25% (w/v) bromophenol blue and 1% (w/v) DTT) with gentle rocking for 15 min. The free cysteine residues of proteins were then alkylated to prevent reformation of disulfide bonds by rocking each strip for 15 min in 10 ml of solution containing 6 m urea, 30% (v/v) glycerol, 2% (w/v) SDS, 50 mm Tris-HCl, pH 8·8, and 0·25% (w/v) bromophenol blue and 2·5% (w/v) iodoacetamide. These strips were then affixed onto homogeneous 12·5% polyacrylamide SDS-PAGE slab gels (2550×2100×1 mm). The second dimension was performed in the EttanDalt vertical system (Amersham Biosciences) at 25°C. The resolved 2-D spots were visualized with SYPRO Ruby fluorescent dye (Bio-Rad) according to the manufacturer's instructions.
The absence and presence of protein spots were scored by eye (see Fig. 3) independently by 2 observers (K.R., C.H.W.) after manually adjusting the contrasts and the brightness using Photoshop C52 (Adobe).
Mass spectrometry and protein identification
Differentially expressed proteins were excised from the gel, and spots were incubated with acetonitrile (ACN) for 15 min at room temperature. ACN was subsequently removed and the spots were dried in a vacuum centrifuge. Dried spots were stored at 4°C until required. Proteins were digested with 20 ng/μl of sequencing-grade modified trypsin (Promega, Southampton, UK) in 50 mm ammonium bicarbonate at 37°C for 12 h. The supernatant from the trypsin digest was transferred to a siliconized microcentrifuge tube. Peptides were sequentially extracted 3 times by incubation with peptide extraction solution, consisting of 25 mm ammonium bicarbonate (10 min at room temperature), 5% formic acid (15 min at 37°C) and ACN (15 min at 37°C). Each extraction was followed by centrifugation and removal of supernatants. The original supernatant and the supernatants from the 3 sequential extractions were combined and dried in a vacuum centrifuge for 4–6 h. The dried peptides were dissolved in 7 μl of 0·1% (v/v) formic acid in 3% (v/v) ACN in water. Samples were centrifuged for 5 min at 12 000 g and the supernatants were subjected to liquid chromatography electrospray ionisation tandem mass spectrometry (LC-ESI-MS/MS).
Liquid chromatographic (LC) separations of the tryptic digests were performed using a reverse phase CapLCTM system (Waters, Manchester, UK). Peptides were desalted by a PepMap C18 microguard column (300 μm internal diameter×1 mm) (LC-Dionex, Leeds, UK) and were then transferred to the analytical column (PepMap C18; 75 μm internal diameter×15 cm column, LC-Dionex). The peptides were eluted in a 60 min gradient. The compositions of the hydrophilic and hydrophobic solvents were 5% ACN, 0·1% formic acid and 95% ACN, 0·1% formic acid. The column eluent was sprayed directly into the nanoESI source of a Q-TOF micro (Waters). An initial MS scan was performed and selection of ions for collision-induced dissociation (CID) was automated by Mass Lynx software (Waters). CID selection criteria were set for 2+ and 3+ ions within the range of 400–2000 m/z above 10 ion counts.
Spectra were searched against the National Centre for Biotechnology Information non-redundant protein databases NCBInr and MSDB in a sequence query search using MASCOT 2.0 software (www.matrixscience.com). The taxonomy was limited to filter for all Metazoa but only Arthropoda were considered, with special reference to Drosophila and to Acyrthosiphon pisum, the closest sequenced relative of the bedbug. Trypsin was used as the cleavage enzyme, with 2 missed cleavage sites allowed. The peptide tolerance was set to 1·0 Da and the MS/MS tolerance was set to 0·3 Da. Carbamidomethyl modification of cysteine and oxidized methionine were set as variable modifications. Peptide/protein matches were only considered significant for mowse scores P<0·05 (Pappin et al. Reference Pappin, Hojrup and Bleasby1993). Mowse scores and P values were generated by MASCOT software. Briefly, based on the total protein mass, Mowse scores use empirically determined factors to statistically weight each individual peptide match from a protein sequence. Statistical significance can be derived by calculating the probability that an observed match is a random event. MASCOT provides mowse scores as −10*LOG10(P), where P is the absolute probability, i.e. a probability of 10−4 corresponds to a mowse score of 40. In MSDB mowse scores above 45 are a significant hit.
RESULTS
Bradford determination revealed the following average amounts of seminal fluid proteins per male: Cimex A: 7·4 μg, Afrocimex: 5 μg and Cimex B: 5·4 μg.
Twelve bands were found on the 1-D gel for Cimex A. Of these bands, all but 1 (approximately 300 kDa) were also found in Cimex B (Fig. 2). One protein band at 97 kDa was found in Cimex B but not in Cimex A. Afrocimex shared 5 bands with Cimex A and an additional one with Cimex B, but had 6 that it did not share with Cimex.
1-D PAGE showing differences between two species of cimicid bugs, Cimex lectularius and Afrocimex constrictus.
Using 20 seminal fluid containers, and thus only about 54–74 μg of protein, we were able to obtain informative gels: 2-D PAGE revealed at least 69 protein spots in the seminal fluid of Cimex A, 58 of which were also found in Cimex B (Fig. 3). Seven proteins found in Cimex B were not detectable in Cimex A. Thus even the small amounts of protein used resulted in 79% shared protein spots by both samples (note the similar difference in the 1-D PAGE). The low protein amounts of 30 μg available for Afrocimex did not produce protein spots.
2D gel showing enumerated proteins. Numbers above 60 denote proteins found in Cimex A but not B. The position of the number denotes the position of the protein. Compare Table 1 for proteins that have been found in the mated females. Protein numbers above 80 denote proteins that have not been used in mass spectrometry. The mass spectrometry data files (.pkl) can be obtained from the authors.
Out of the 76 spots found in the seminal fluid, 41 were found in the mated female, 2 of which were only found in Cimex A seminal fluid. Of these 41 proteins, 32 were not found in virgin females. Of the 9 proteins that were found in virgin females only 1 was identified, the gag protein, which with a mowse score of 36 was not significant.
Two control runs with sperm did not return any protein spots with the method used here. Therefore, most proteins are from the seminal fluid and up to 32 seminal fluid proteins may be transferred from male to female during mating.
Out of the 69 proteins that were analysed by mass spectrometry, 18 (26·5%) returned entries in the protein data bases when tested against all arthropods, 17 (24·6%) against all insects, 10 (14·5%) against Drosophila and 3 (4·3%) against other Hemiptera species (Table 1). Many of them, however, were ubiquituous and cytoskeletal proteins and fewer had – in Drosophila – a distinct link to reproduction (see Discussion section). One protein (spot 31), predicted Fkbp13 CG9847-PA, isoform A (Accession number gi|17352457) had a mowse score of 38 and so was just below the 5% significance level.
Predicted proteins found in the seminal fluid of bed bugs, Cimex lectularius
(Numbers in column 1 refer to the protein number on the 2D gel shown in Fig. 3. Also provided are the theoretical molecular mass of the protein, the arthropod species whose protein provided the closest match in the database, as well as the respective mowse score and protein coverage. Some proteins were found in females, either in mated or virgin ones. The predicted protein function is given in the last column.)
1 Also found in Culex pipiens quinquefasciatus (gi|170042204), mowse score 64, protein coverage 3%.
2 Found with similar parameters in Anopheles gambiae str. PEST, Tribolium castaneum.
3 Found with similar parameters in many basal insects (e.g. Periplaneta americana, gi|4378573), mowse score 290, protein coverage 16%).
4 Also found in Acyrthosiphon pisum (gi|193704626), predicted mass 32294 kDa, mowse score 182, protein coverage 15%.
5 Also found in Anopheles gambiae str. PEST (Accession Q7QIV1_ANOGA), predicted mass 21802 kDa, mowse score 80, protein coverage 8%.
DISCUSSION
This study supports the notion that proteomics of seminal plasma is a fruitful technology to discover proteins that have not been detected by genomic analysis (see also Collins et al. Reference Collins, Caperna, Williams, Garrett and Evans2006; Walker et al. Reference Walker, Rylett, Keen, Audsley, Sajid, Shirras and Isaac2006; Braswell et al. Reference Braswell, Andrés, Maroja, Harrison, Howard and Swanson2006; Findlay et al. Reference Findlay, Yi, MacCross and Swanson2008). Such studies add to an increasing knowledge about the diversity of seminal plasma, about further development of methods and about the function of specific components of the semen. These 3 areas are discussed below.
Seminal protein diversity
The low overall amount of protein in our samples did not allow quantification in terms of higher or lower expression of a particular protein. Likewise, absent spots cannot be used as evidence for the absence of proteins. However, we can rely on proteins that have been detected. Such positive identification (independently of any undetected proteins) is in agreement with the notion that seminal fluids differ strongly between species, i.e. evolve rapidly for 3 reasons. First, there was a high departure between Cimex and Drosophila (only 14% shared proteins in databases) and between Cimex and all other arthropods (26% shared protein entries). Second, using a different method, the differences were substantially smaller, but still considerable, between Cimex and Afrocimex which are members of the same family (Cimicidae) – about 60% of the 1-D gel bands were shared. Thirdly, the 1-D and 2-D gel analysis suggests that even between two separated laboratory populations of the same species only between 90% and 80% of bands and protein spots, respectively, are shared.
Using genomic analysis these levels of similarity compare positively with Andrés et al. (Reference Andrés, Maroja, Bogdanowicz, Swanson and Harrison2006) who did not find shared transcripts of accessory proteins between Drosophila and field crickets, Gryllus spp., while Braswell et al. (Reference Braswell, Andrés, Maroja, Harrison, Howard and Swanson2006) found 7–40% similarity between Drosophila and either Gryllus or Allonemobius. The level of similarity is less comparable to that between D. melanogaster and the honey bee semen where a staggering 96% of the proteins could be matched to Drosophila proteins (Collins et al. Reference Collins, Caperna, Williams, Garrett and Evans2006, but see Findlay et al. Reference Findlay, Yi, MacCross and Swanson2008 for additional proteins). Again using genomic analyses, the within-genus comparison between Drosophila melanogaster and D. pseudoobscura, 58% shared putative seminal proteins genes were found (Mueller et al. Reference Mueller, Ravi-Ram, McGraw, Qazi, Siggia, Clark, Aquadro and Wolfner2005), being closer to the 1-D gel similarity found between Cimex and Afrocimex.
Methodology
We have also demonstrated that with just 20 seminal fluid containers, i.e. around 60 μg of protein can be obtained. Although we note that a third population of the same species might further improve the repeatability and confidence in the data we suggest that our method may be used to compare bed bug samples from the ‘field’, i.e. by using the seminal fluid containers of only about 20 males per infestation.
Furthermore, the abundance of many proteins in our 2-D gels strongly suggests that the proteins can be subjected to fragmentation. This may increase the return rate in databases and, given good bed bug breeding facilities, may also enable the purification of larger amounts of some proteins. Larger amounts of proteins will enable the assessment of more detailed physiological properties of bed bug seminal fluid, given that in other insects seminal fluids affect ovulation (Qazi et al. Reference Qazi, Heifetz and Wolfner2003), increase the oviposition rate (Wolfner, Reference Wolfner2002; Gillot, Reference Gillott2003; Chapman and Davies, Reference Chapman and Davies2004; Poiani, Reference Poiani2006), reduce female attractiveness (Tram and Wolfner, Reference Tram and Wolfner1998) and egg hatchability immediately after mating (Prout and Clark, Reference Prout and Clark2000), influence sperm storage (Neubaum and Wolfner, Reference Neubaum and Wolfner1999), affect paternity by differential spermicide (Fry and Wilkinson, Reference Fry and Wilkinson2004), influence sperm organization in the female (Viscuso et al. Reference Viscuso, Narcisi, Sottile and Brundo2001) and sperm motility (Ruknudin and Raghavan, Reference Ruknudin and Raghavan1988).
Our study has identified several predicted proteins in both the seminal fluid and in the mated (but not the unmated) female. These are candidate proteins for being actually transferred during mating. Given the complex sperm biology of the bed bug (Davis, Reference Davis and Usinger1966; Rao and Davis, Reference Rao and Davis1969; Reinhardt and Siva-Jothy, Reference Reinhardt and Siva-Jothy2007) and that the female copulatory organ of bed bugs is capable of phagocytosing sperm (see references in Reinhardt and Siva-Jothy, Reference Reinhardt and Siva-Jothy2007) one may speculate about the sperm protective functions of several of the proteins.
Specific functions of some identified predicted proteins
The presence of actin, calreticulin, tropomyosin or paramyosin indicates a cellular component of the seminal fluid analysed here. These proteins unlikely originate from sperm cells because seminal plasma and sperm are stored separately in the bed bug and because sperm material analysed in the same way as seminal fluids did not produce any protein spots. It is possible that these mostly cytoskeletal proteins escaped from the cells of the seminal fluid container during centrifugation. On the other hand, intracellular material can be found in gland discharges as shown in the careful study by Walker et al. (Reference Walker, Rylett, Keen, Audsley, Sajid, Shirras and Isaac2006) who found calreticulin in male accessory glands of Drosophila.
Even if we assume (as a worst case scenario) that all proteins that were identified represent cellular contaminants, only the unidentified protein spot may then represent seminal fluid proteins, i.e. a maximum of 80·3% out of 76 (61 proteins). This number would still resemble the number of proteins found in honey bee semen (Collins et al. Reference Collins, Caperna, Williams, Garrett and Evans2006), were higher than that reported by Walker et al. (Reference Walker, Rylett, Keen, Audsley, Sajid, Shirras and Isaac2006) but much lower than that reported by Findlay et al. (Reference Findlay, Yi, MacCross and Swanson2008).
Other predicted proteins have been directly associated with insect reproduction. The gene moj 29 that was found mostly expressed in the testis and/or the accessory glands of Drosophila mojavensis and D. arizonae (Wagstaff and Begun, Reference Wagstaff and Begun2005) but its function is unknown. A proteomic study of Drosophila seminal fluid (Walker et al. Reference Walker, Rylett, Keen, Audsley, Sajid, Shirras and Isaac2006) reported several proteins that were also found in the present study, including the Fkbp13 protein (a non-significant hit in our study), the protein disulfide isomerase, and a similar sized heat shock protein.
Ubiquitin is very abundant in seminal plasma of humans (Lippert et al. Reference Lippert, Seeger, Schieferstein and Voelter1993) and is here reported from the bed bug. Its extracellular role is currently debated. While the amount of ubiquitin per ejaculate can be positively correlated with the proportion of abnormal sperm (Sutovsky et al. Reference Sutovsky, Turner, Hameed and Sutovsky2003, Reference Sutovsky, Hauser and Sutovsky2004a) other studies do not support such a notion (Muratori et al. Reference Muratori, Marchiani, Forti and Baldi2005) and show that ubiquitin enables or facilitates fertilization (Sutovsky et al. Reference Sutovsky, Manandhar, McCauley, Caamano, Sutovsky, Thompson and Day2004b; Sawada et al. Reference Sawada, Sakai, Abe, Tanaka, Takahashi, Fujino, Kodama, Takizawa and Yokosawa2002). In mammalian males ubiquitinated sperm are phagocytosed (Sutovsky et al. Reference Sutovsky, Moreno, Ramalho-Santos, Dominko, Winston and Schatten2001) and may, therefore, serve as a control mechanism ensuring that fertilization-competent cells are being ejaculated.
A similar controversy revolves around the elongation factor EF-1alpha which we detected in the seminal fluid. This factor is involved in protein synthesis, regulation of apoptosis and interacts with actin and the ubiquitin-dependent proteolysis. Some, but not all, genotypes bearing additional EF-1 alpha gene copies showed an enhanced life span (Stearns and Kaiser, Reference Stearns and Kaiser1993; Shikama et al. Reference Shikama, Ackermann and Brack1994). While such lines do not necessarily express more EF-1 alpha (Shikama et al. Reference Shikama, Ackermann and Brack1994), other studies show that increased expression of EF-1 alpha under stress lead to life-span extension (Wang et al. Reference Wang, Kazemi-Esfarjani and Benzer2004; Talapatra et al. Reference Talapatra, Wagner and Thompson2002).
Finally, the occurrence of peroxiredoxin, an antioxidant (e.g. Kawazu et al. Reference Kawazu, Komaki-Yasuda, Oku and Kano2008), in the seminal plasma poses the question whether this protects sperm or benefits the female, which is an exciting future research topic requiring the separation of the effect of other beneficial ejaculate substances (see above).
While female life-span prolongation due to ejaculate substances has been found in a cricket species (Wagner et al. Reference Wagner, Kelley, Tucker and Haper2001) there is much less evidence for that in bed bugs. One study found that the contribution of copulatory wounding to female mortality was larger than the total negative effect of copulation on females (Morrow and Arnqvist, Reference Morrow and Arnqvist2003) raising the possibility of a beneficial effect of the male's ejaculate (Reinhardt and Siva-Jothy, Reference Reinhardt and Siva-Jothy2007 counter to Morrow's and Arnqvist's (Reference Morrow and Arnqvist2003) suggestion). This intriguing possibility is currently being tested directly. Given sufficient protein material the method presented here may, in the future, enable us to reveal whether such a proposed effect is due to EF-1 alpha or other proteins.
Conclusion
We have shown that even the small amount of seminal fluid used here can provide informative gels. The proteins positively identified in the bed bug (Cimex lectularius) can be used as a reference gel (mass spectrometry files in .pkl format are available from K.R. on request) and be further analysed by fragmentation and bioinformatic tools. Comparisons with other species pose a challenge for interfering with reproductive pathways in the bed bug because the predicted rapid evolution of seminal proteins will result in divergence between different populations. It is desirable to focus further analyses and direct physiological assays towards proteins that are found conserved within the family of the Cimicidae or even within the species Cimex lectularius.
Sveta Sedelnikova, University of Sheffield, carried out the Bradford analysis and ran the 1-D gel. The study was funded through a VIP award by the Wellcome Trust to K.R. We thank Richard Naylor, Matt Pitts-Tucker and Mike Siva-Jothy for help in collecting Afrocimex constrictus. We also thank Robin Todd, Insect Control & Research, Inc., Baltimore, Maryland and Woodbridge Foster, Ohio State University, for donation of bed bugs. Richard Naylor helped with some of the dissections. Christian Wegener provided helpful comments on an earlier version of the paper.
