Abstract
The specificity of sperm–egg recognition is crucial to species independence, and two proteins (Izumo1 and JUNO) are essential for gamete adhesion/fusion in mammals. However, hybridization, which is very common in turtles, also requires specific recognition of sperm–egg binding proteins. In this study, we discovered that natural selection plays an important role in the codon usage bias of Tu-Izumo1 and Tu-JUNO. Positively selected sites and co-evolutionary analyses between Tu-Izumo1 and Tu-JUNO have been previously reported, and we confirm these results in a larger analysis containing 25 turtle species. We also showed that Tu-JUNO is expressed on the oocyte surface and that Tu-Izumo1 and Tu-JUNO interact with each other directly in different species hybridization combinations. Co-immunization assays revealed that this interaction is evolutionarily conserved in turtles. The results of avidity-based extracellular interaction screening between Tu-Izumo1 and Tu-JUNO for sperm–oocyte binding pairs (both within and across species) likely suggest that the interaction force between Izumo1 and JUNO has a certain correlation in whether the turtles can hybridize. Our results lay a theoretical foundation for the subsequent development of techniques to detect whether different turtle species can interbreed, which would provide the molecular basis for breeding management and species protection of turtles.
Introduction
In sexual reproduction, fertilization is a fundamental biological process that produces a new organism. Successful recognition, adhesion, and fusion between sperm and egg are crucial parts of this process, and several receptor proteins play important roles in these events (Ikawa et al. 2010). For example, the proteins Izumo1 on sperm (Inoue et al. 2005) and JUNO on the egg (Grayson & Civetta 2012, Satouh et al. 2012, Bianchi et al. 2014, Nishimura et al. 2016) interact during fertilization, and this interaction is conserved within mammals. Other sperm proteins such as SOF1, TMEM95, FIMP, and SPACA6 have been shown to be required for fertilization (Lorenzetti et al. 2014, Lamas-Toranzo et al. 2020, Noda et al. 2020), whereas the disruption of the sperm protein FIMP and the egg protein CD9 causes severe subfertility (Zhou et al. 2009, Fujihara et al. 2020). Additionally, in mammals, this particular interaction event is involved in maintaining species-specificity at the level of sperm–egg fusion (Bianchi et al. 2014, Jean et al. 2019).
In contrast to species-specificity, hybridization is very common in nature, especially in turtles (Pan et al. 2009, Xia et al. 2011, Suzuki et al. 2013). Hybridization means that sperm and eggs from different species can combine and fuse with each other. For example, hamster oocytes lacking the zona pellucida were able to recognize and fuse with sperm from different mammalian species, including the mouse, pig, and human; these results show that sperm–egg recognition plays an important role in cross-species hybridization (Hanada & Chang 1972, Creighton & Houghton 1987). In another study, Bianchi et al. showed that hamster JUNO can directly interact with mouse, pig, and human Izumo1, which provided a molecular explanation for the ability of gametes from different species to recognize and fuse with each other in mammals (Bianchi & Wright 2015).
We previously reported that Izumo1 and JUNO are also present in turtles (Tu-Izumo1 and Tu-JUNO), and we inferred that they interact and suggested that these proteins are involved in the hybridization of a turtle based on positively selected sites and co-evolutionary analyses (Dong et al. 2019). In this study, the positively selected sites and co-evolutionary analyses for Tu-Izumo1 and Tu-JUNO were performed in 25 turtles. Moreover, the codon usage bias (CUB) of Tu-Izumo1 and Tu-JUNO was also calculated, which was affected by mutation pressure and natural selection (Sheng et al. 2007, Zhong et al. 2007). Importantly, we first used hematoxylin–eosin staining and immunofluorescence to verify that Tu-JUNO is expressed on the surface of turtle oocytes. We then used co-immunoprecipitation and avidity-based extracellular interaction screening (AVEXIS) to measure the interaction force between Tu-Izumo1 and Tu-JUNO proteins when eggs from one turtle species and sperm from another were mixed. Our results provide a molecular explanation for the role of the interaction between Izumo1 and JUNO in cross-species gamete recognition and fusion. These findings can be used to devise good management practices for the breeding and species protection of turtles.
Materials and methods
Animals and ethics
Eight species used to amplify Tu-Izumo1 and Tu-JUNO genes were all from Wuhu, Anhui Province, China. For each species, one male and one female gonad were used for RNA extraction (Supplementary Table 1, see section on supplementary materials given at the end of this article). Procedures involved in animals and their care were approved by the Animal Care and Use Committee of Anhui Normal University.
Sequence annotation and analysis of Tu-Izumo1 and Tu-JUNO
The coding region sequences of Tu-Izumo1 and Tu-JUNO in Dermatemys mawii, Sternotherus odoratus, Kinosternon carinatum, and Macroclemys leucostomum were obtained by amplification and sequencing using the reported primer (Dong et al. 2019). The coding region sequences of the remaining species were retrieved from NCBI (National Center for Biotechnology Information), see Supplementary Table 2.
Relative synonymous codon usage (RSCU) was calculated using the Sequence Manipulation Suite (Stothard 2000), and a heat map was drawn with R x64 4.0.2. The effective number of codons (ENC) is a parameter to measure the usage bias of synonymous codons, which was determined with DnaSP 5.0 (Librado & Rozas 2009), and the neutrality plot was drawn to compare the relative neutrality between GC12 (the ratio of GC vs AT in the first and second codon positions) and GC3 (the ratio of GC vs AT in the three codon positions) of Tu-Izumo1 and Tu-JUNO to estimate the most important factors that influence codon usage bias. If the corresponding data points should be on the abscissa then less influence of directional mutation pressure, while if the corresponding data points should be distributed along the bisector and a statistically significant correlation between G3 and G12, it indicates that codon usage bias is totally formed by directional mutation pressure (Sueoka 1988).
Construction of phylogenetic trees of Tu-Izumo1 and Tu-JUNO
Phylogenetic trees based on the amino acid sequences of Tu-Izumo1 and Tu-JUNO were conducted, respectively, with maximum likelihood (ML) and Bayesian inference (BI) methods. The sequence alignment was first carried out using MEGA 7 (Kumar et al. 2016). Then the ML analysis was conducted with RAxML v.8.2.12 using the PROTGAMMAAUTO model with rapid bootstrap for 1000 replicates. Subsequently, Bayesian analysis was conducted with MrBayes 3.2.6 using parameter 'aamodelpr = mixed' with two simultaneous runs and four independent Markov chains for 10,000,000 generations (Ronquist & Huelsenbeck 2003, Guindon et al. 2010). In addition, two species, including A. mississippiensis and A. sinensis, were used as the outgroups.
Positive selection test of Tu-Izumo1 and Tu-JUNO
To evaluate the influence of natural selection on the evolution of Tu-Izumo1 and Tu-JUNO, the positively selected sites of these proteins of 25 species were tested by using the Codeml tool implemented in PAML package v4.9 to run the site model and branch-site model with the null model M7, the alternative model M8 (Yang 2007), and a third test compared the likelihood of the model M8 to the likelihood of a null model M8a in which ω was fixed to 1 in avoidance of detecting false signatures of positively selected as a result of functional relaxation (Wong et al. 2004). Positively selected sites were detected when ω > 1 and the LRT was significant (P < 0.05), and the Bayes empirical Bayes (BEB) was used to determine which codon positions have experienced positive selection
Analyses of coevolution based on the amino acid sequences of Tu-Izumo1 and Tu-JUNO
The linear regression analyses among root-to-tip ω ratios of Tu-Izumo1 and Tu-JUNO in the phylogeny were performed by using a free-ratio model implemented in Codeml (PAML package v4.9). The tree topology was determined on the basis of the mitochondrial genome phylogenies for turtles. Linear regressions were conducted with R x64 4.0.2.
Recombinant protein production and purification
The regions encoding the entire extracellular domains of Tu-Izumo1 and Tu-JUNO of Mauremys reevesii, Mauremys mutica, Cistoclemmys flavomarginata, and Chrysemys picta bellii were cloned and sequenced using RT-PCR methods (Dong et al. 2019). Primers with restriction enzyme sites were synthesized by Shanghai Sangon Biotech (Shanghai, China) (Table 1).
Nucleotide sequences of primers used for plasmid construction in the experiment.
| Primer sequences (5’–3’) | AT (°C) | ||
|---|---|---|---|
| Forward | Reverse | ||
| Tu-Izumo1-Flag | GGAATTCGCCGCCACCATGGGATGGGCACTG | ATAGTTTAGCGGCCGCATGACGGTCCAGTCT | 67 |
| Tu-JUNO-Myc | GGAATTCCGGCCGCCACCATGGCAGCACGAT | CCGCTCGAGCTAGCAGAGACAGGAGA | 67 |
| Tu-Izumo1-COMP | AATGCGGCCGCTCCTAGCCGCCACCATGGCACCATGG | TTTAGGCGCGCCCACGGTCCAGTCTCCA | 68 |
| Tu-JUNO-bio | AATGCGGCCGCTCCTAGCCGCCACCATGGCACCATGG | TTTAGGCGCGCCGAGCAGGGACAGGAG | 63 |
AT, annealing temperature.
For co-immunoprecipitation, the ectodomains of Tu-Izumo1 and Tu-JUNO were flanked by EcoRI, NotI, and XhoI sites (New England BioLabs, Ipswich, MA, USA). Fragments then were transferred to protein expression vectors pCMV-Myc-C and pECMV-3×FLAG-C, respectively (Miaoling Bio, Wuhan, China). For AVEXIS, all ectodomains were flanked by unique NotI and AscI sites (New England BioLabs) and subcloned into the CD200Cd4d3+4-COMP-blac-3xFLAG-6his, CD200Cd4d3+4-bio, and secreted BirA-8his plasmids (Addgene, Watertown, MA, USA) (Bushell et al. 2008). All proteins were expressed by transient transfection using PolyJet™ in Vitro DNA Transfection Reagent (SignaGen Laboratories, Rockville, MD, USA) in HEK293T cells grown in suspension culture and collected from the cell culture supernatant 5 days post-transfection. His-tagged proteins were purified from the culture supernatant by affinity chromatography on HisTrap HP columns (GE Healthcare) using an AKTAxpress system (GE Healthcare) according to the manufacturer’s instructions.
Microscopic analysis of oocyte and sperm and immunofluorescence of Tu-JUNO
The epididymis of sexually mature M. reevesii was cut up with scissors in 0.65% normal saline and filtered. The sperm suspension was smeared and fixed with methanol for 5 min and stained with 2% eosin for 1 h, and then the sections were observed under the light microscope. Oocytes sampled from the ovary of sexually mature M. reevesii were fixed in 4% paraformaldehyde, embedded in paraffin, and serially sectioned (6 μM). Some of the sections were stained with hematoxylin and eosin for observation using a light microscope (Bian et al. 2013). The other sections were treated with 3% H2O2 to disrupt endogenous peroxidase. After 3 × 5 min washes in ddH2O, the slides were incubated in PBS in boiling water to repair antigen activity. The slides then were incubated in 5% BSA for 30 min at room temperature to block non-specific binding. The sections were incubated with rabbit polyclonal antibody to turtle Tu-JUNO (1:1000) (produced privately through Zoonbio Biotechnology, Nanjing, China) at 4°C overnight. In parallel, anti-rabbit IgG LCS (Abbkine) was prepared and used as the negative control. After 3 × 5 min washes with PBS, the sections were incubated with goat anti-rabbit IgG(H+L) Alexa Fluor 488 antibody (1:1000) (Beyotime, Shanghai, China) for 2 h at room temperature. After 3 × 5 min washes with PBS, the sections were incubated with antifade solution (Boster, Pleasanton, CA, USA).
Co-immunoprecipitation of Tu-Izumo1 and Tu-JUNO
Co-immunoprecipitation procedures were performed at 4°C unless otherwise indicated. The total protein was extracted from the transfected cells, then the mouse control IgG together with a resuspended volume of protein A/G PLUS-Agarose (Santa Cruz Biotechnology) was added to the protein supernatant to avoid non-specific binding. The protein A/G PLUS-Agarose slurry was washed twice with PBS buffer before use. One microgram anti-Myc Tag Mouse Monoclonal Antibody (MAB) (2D5) (Abbkine, Wuhan, China) and anti-DDDDK Tag Mouse MAB (1B10) (Abbkine) were added to the above culture supernatant, respectively, and incubated for 1 h. In parallel, anti-mouse IgG LCS (Abbkine) was prepared and used as the negative control. A resuspended volume of protein A/G PLUS-Agarose was added to these mixtures, which then were incubated on a rocker platform overnight. Immunoprecipitates were collected and washed four times in lysis buffer, and the eluted proteins were analyzed using standard Western blot procedures.
Western blot analysis
Western blotting was performed as described previously (Bartholdson et al. 2012). Briefly, samples were electrophoresed in 10% Bis-Tris gels at 100 V using sodium dodecyl sulfate running buffer. Proteins were electrophoretically transferred to a PVDF membrane at 300 mA for 1 h. The membrane was blocked with 5% BSA in PBS at room temperature for 1 h. Blots were incubated with the appropriate primary antibody (1:5000) overnight at 4°C. After three 10 min washes with PBS with Tween, membranes were incubated with peroxidase-conjugated secondary antibody (1:10000) for 1 h at room temperature. Mouse Control IgG was obtained from ABclonal Technology (Woburn, MA, USA) and His-Tag Antibody (1:1000) was obtained from Boster. After washing three times with TBS with Tween, blots were incubated with ECL detection reagent and imaged with a Gel Logic 440 Imaging system (Tanon, Shanghai, China).
Extracellular protein interaction screening by AVEXIS
We used the AVEXIS method to measure the interaction forces between Tu-Izumo1 and Tu-JUNO from egg and sperm of different related turtles. This technique can detect direct and transient interactions. We first expressed the entire ectodomains of both Tu-Izumo1 and Tu-JUNO as soluble recombinant proteins in HEK293T cells either as an enzymatically monobiotinylated ‘bait’ or a pentamerized enzyme-tagged ‘prey’. Bait and prey proteins were normalized to activities suitable for the AVEXIS assay as described previously (Kerr & Wright 2012). For both orientations of the bait-prey interaction, the interactions of Tu-Izumo1–Tu-JUNO within four species were tested. We first measured the intraspecific interactions of Tu-Izumo1–Tu-JUNO within each species. Subsequently, the interspecific interactions were measured between Tu-Izumo1 of M. reevesii and Tu-JUNO from the remaining species. Similarly, we measured the interactions between Tu- JUNO of M. reevesii and Tu-Izumo1 from the rest of species. The experiments were repeated three times by AVEXIS using independent protein preparations.
Results
Analysis of codon usage bias in Tu-Izumo1 and Tu-JUNO
To understand the pattern of random usage of synonymous codons for Tu-Izumo1 and Tu-JUNO in turtles, observed codon pair frequency heatmaps of codon pairs excluding stop codons (Supplementary Fig. 1) and relative synonymous codon usage (RSCU) of individual codons (Table 2) were calculated and compared. Among 60 codons, 12 codons in Tu-Izumo1 and 16 codons in Tu-JUNO are over-represented. However, 20 codons are under-represented in Tu-Izumo1 and Tu-JUNO, respectively (Table 2). The RSCU analysis showed that turtles exhibit comparatively higher CUB toward G or C at the third position and less toward A or T in Tu-Izumo1 and Tu-JUNO.
The over-represented codons and the under-represented codons of Tu-Izumo1 and Tu-JUNO.
| RSCU > 1.6 | RSCU < 0.6 | |
|---|---|---|
| Tu-Izumo1 | ACC, CAG, AGC, CGC, CGG, CTG, GAC, GAG, GCC, GGG, GTG, TTC | AGT, ATA, ATT, CAA, CAT, CGT, CTA, CAA, CTT, CAT, GCA, GCT, GGA, GTA, TCA, TGG, TTA, TTG, TTG, TTT |
| Tu-JUNO | AAC, ACC, ATC, CAC, CAG, CGC, CGG, CTG, GAC, GAG, GCC, GGG, GTG, TAC, TCC, TTC | AAT, ACA, ACT, AGT, ATA, CAA, CAT, CCT, CTA, CTT, GAA, GAT, GGA, GGT, GTA, GTT, TAT, TGG, TTA, TTT |
To further investigate the codon usage bias, ENC and the G+C were analyzed for Tu-Izumo1 and Tu-JUNO in 25 turtles. The actual ENC values for all turtles of Tu-Izumo1 and Tu-JUNO were just below the ENC curve (Supplementary Fig. 2), an indication that not only mutational bias but also natural selection plays an important role in the codon bias in turtles.
Further, a neutrality plot (GC3 Vs GC12) is commonly employed to analyze relationships between GC3 and GC12. Based on the neutrality analysis, we found that the genes have a wide range of GC3 value distribution, and there is no significant positive correlation between GC3 and GC12 (r = 0.774, P > 0.01 in Tu-Izumo1; r = 0.741, P > 0.01 in Tu-JUNO), and lower values of the slope regression line (0.395 and 0.383, respectively, for Tu-Izumo1 and Tu-JUNO) (Supplementary Fig. 3), suggesting that natural selection, rather than mutational pressure, is the main determinant of CUB in turtles.
Phylogenetic analyses of Tu-Izumo1 and Tu-JUNO
Discrepancies in topology were observed between the results of the maximum likelihood and the Bayesian approach for Tu-Izumo1 and Tu-JUNO phylogenies (Supplementary Figs 4 and 5). In addition, phylogenetic relationships among 25 turtles were inferred from mitochondrial genome sequences (Supplementary Fig. 6). None of the gene trees showed a topology that was in agreement with the phylogeny trees of the species. This discordance suggests that both Tu-Izumo1andTu-JUNO are undergoing independent evolutionary trajectories.
Positive selection test of Tu-Izumo1 and Tu-JUNO
Tu-JUNO in 25 turtles appears to be evolving under positive selection, and the LRTs (M8 vs M7 and M8 vs M8a) were in favor of these lection models. There are 5 BEB sites to be reported in Tu-JUNO (ω(M8) ratio = 1.573). However, there are no positively selected sites existed in Tu-Izumo1 (Table 3).
Tests of positive selection for Tu-Izumo1 and Tu-JUNO. BEB sites were only produced for clades evolving under positive selection.
| Gene | n | LRT | ω | BEB sites | |
|---|---|---|---|---|---|
| M7vsM8 | M8vsM8a | M8 | |||
| Tu-Izumo1 | 25 | 0.0825 | 0.3125 | 1.000 | Null |
| Tu-JUNO | 25 | 0.0000 | 0.0000 | 1.573 | 5; 59N*; 224R*; 239P*; 240A*; 243L* |
*P > 95%.
LRTs, likelihood-ratio tests; M7 allows ω to vary between 0 and 1; M8a (M7 with ω = 1) is the null hypothesis for M8; n, number of compared lineages.
Coevolution among Tu-Izumo1 and Tu-JUNO
We used an approach based on averaging root-to-tip ω ratio to test the correlation between Tu-Izumo1 and Tu-JUNO (Vicens & Roldan 2014). We first evaluated the influence of using a reconstructed species phylogeny based on the mitochondrial genome. A significant positive correlation between Tu-Izumo1 and Tu-JUNO ω estimates was obtained (r = 0.606, P < 0.001) (Supplementary Fig. 7A). We also analyzed topologies of trees derived from an alignment that included Tu-Izumo1 and Tu-JUNO coding sequences. These two genes yielded different phylogenies. Nevertheless, regression between Tu-Izumo1 and Tu-JUNO ω estimates were statistically significant (Tu-Izumo1: r = 0.708, P < 0.001; Tu-JUNO: r = 0.582, P < 0.001) (Supplementary Fig. 7B and C). There is a significant correlation between the omega values for Tu-Izumo1 and Tu-JUNO regardless of the phylogeny used, but the r values vary greatly with the phylogeny that is used, indicating that the phylogeny does have an impact on the co-evolutionary relationship.
Microstructure of oocytes and sperm and immunofluorescent localization of Tu-JUNO in oocytes
The microstructures of the oocytes and sperm of M. reevesii are shown in Supplementary Fig. 8. Rabbit anti-Tu-JUNO antibody treatment showed that Tu-JUNO protein was present at the turtle oocyte surface, and the distribution of the fluorescent signal appeared homogeneous (Fig. 1).
Expression of Tu-JUNO protein at the turtle oocyte surface, which was detected using inverted fluorescence microscope (Leica). The use of anti-turtle Tu-JUNO mAb, revealed by a goat anti-rabbit Alexa Fluor 488 antibody, showed the presence of Tu-JUNO protein on the membrane of turtle oocytes. Anti-rabbit IgG was used as negative control.
Citation: Reproduction 162, 4; 10.1530/REP-21-0124
Expression of Tu-JUNO protein at the turtle oocyte surface, which was detected using inverted fluorescence microscope (Leica). The use of anti-turtle Tu-JUNO mAb, revealed by a goat anti-rabbit Alexa Fluor 488 antibody, showed the presence of Tu-JUNO protein on the membrane of turtle oocytes. Anti-rabbit IgG was used as negative control.
Citation: Reproduction 162, 4; 10.1530/REP-21-0124
Expression of Tu-JUNO protein at the turtle oocyte surface, which was detected using inverted fluorescence microscope (Leica). The use of anti-turtle Tu-JUNO mAb, revealed by a goat anti-rabbit Alexa Fluor 488 antibody, showed the presence of Tu-JUNO protein on the membrane of turtle oocytes. Anti-rabbit IgG was used as negative control.
Citation: Reproduction 162, 4; 10.1530/REP-21-0124
Co-immunoprecipitation of Tu-Izumo1 with Tu-JUNO in different species
To detect the interaction between Tu-Izumo1 and Tu-JUNO, we expressed the entire ectodomain of Tu-JUNO and detected its binding to Tu-Izumo1 using co-immunoprecipitation. The results of Western blotting of the co-immunoprecipitation products revealed direct interactions between Tu-Izumo1 and Tu-JUNO of M. reevesii, M. mutica, C. flavomarginata, and C. picta bellii (Fig. 2). To test whether Tu-Izumo1 and Tu-JUNO of different turtle species can interact, we also conducted co-immunoprecipitation experiments across turtle species. The Tu-Izumo1 of M. reevesii interacted with the Tu-JUNO of closely related species (M. mutica and C. flavomarginata) and with that of C. picta bellii (Fig. 3A1, B1 and C1). Similarly, the Tu-Izumo1 of M. mutica, C. flavomarginata, and C. picta bellii interacted with the Tu-JUNO of M. reevesii (Fig. 3A2, B2 and C2). Clear binding between Tu-Izumo1 and Tu-JUNO was observed, which illustrates that the interaction between these proteins is conserved within turtles.
Western blot analysis of co-immunoprecipitation of Tu-Izumo1 and Tu-JUNO in Mauremys reevesii (A), Mauremys mutica (B), Cistoclemmys flavomarginata (C), Chrysemys picta bellii (D).
Citation: Reproduction 162, 4; 10.1530/REP-21-0124
Western blot analysis of co-immunoprecipitation of Tu-Izumo1 and Tu-JUNO in Mauremys reevesii (A), Mauremys mutica (B), Cistoclemmys flavomarginata (C), Chrysemys picta bellii (D).
Citation: Reproduction 162, 4; 10.1530/REP-21-0124
Western blot analysis of co-immunoprecipitation of Tu-Izumo1 and Tu-JUNO in Mauremys reevesii (A), Mauremys mutica (B), Cistoclemmys flavomarginata (C), Chrysemys picta bellii (D).
Citation: Reproduction 162, 4; 10.1530/REP-21-0124
Western blot analysis of co-immunoprecipitation of cFlag-Izumo1 and cMyc-JUNO across different turtles. (A1) The interaction between cFlag-Izumo1 of M. reevesii and cMyc-JUNO of M. mutica; (B1) the interaction between cFlag-Izumo1 of M reevesii and cMyc-JUNO of C. flavomarginata; (C1) the interaction between cFlag-Izumo1 of M. reevesii and cMyc-JUNO of C. picta bellii (A2) the interaction between cFlag-Izumo1 of M. mutica and cMyc-JUNO of M. reevesii; (B2) the interaction between cFlag-Izumo1 of C. flavomarginata and cMyc-JUNO of M. reevesii; (C2) the interaction between cFlag-Izumo1 of C. picta bellii and cMyc-JUNO of M. reevesii.
Citation: Reproduction 162, 4; 10.1530/REP-21-0124
Western blot analysis of co-immunoprecipitation of cFlag-Izumo1 and cMyc-JUNO across different turtles. (A1) The interaction between cFlag-Izumo1 of M. reevesii and cMyc-JUNO of M. mutica; (B1) the interaction between cFlag-Izumo1 of M reevesii and cMyc-JUNO of C. flavomarginata; (C1) the interaction between cFlag-Izumo1 of M. reevesii and cMyc-JUNO of C. picta bellii (A2) the interaction between cFlag-Izumo1 of M. mutica and cMyc-JUNO of M. reevesii; (B2) the interaction between cFlag-Izumo1 of C. flavomarginata and cMyc-JUNO of M. reevesii; (C2) the interaction between cFlag-Izumo1 of C. picta bellii and cMyc-JUNO of M. reevesii.
Citation: Reproduction 162, 4; 10.1530/REP-21-0124
Western blot analysis of co-immunoprecipitation of cFlag-Izumo1 and cMyc-JUNO across different turtles. (A1) The interaction between cFlag-Izumo1 of M. reevesii and cMyc-JUNO of M. mutica; (B1) the interaction between cFlag-Izumo1 of M reevesii and cMyc-JUNO of C. flavomarginata; (C1) the interaction between cFlag-Izumo1 of M. reevesii and cMyc-JUNO of C. picta bellii (A2) the interaction between cFlag-Izumo1 of M. mutica and cMyc-JUNO of M. reevesii; (B2) the interaction between cFlag-Izumo1 of C. flavomarginata and cMyc-JUNO of M. reevesii; (C2) the interaction between cFlag-Izumo1 of C. picta bellii and cMyc-JUNO of M. reevesii.
Citation: Reproduction 162, 4; 10.1530/REP-21-0124
Interaction forces between Tu-Izumo1 and Tu-JUNO across different species
Two proteins were recognized in the Western blot analysis of HEK-JUNO (HEK cells expressing turtle JUNO in CD200Cd4d3+4-bio) and HEK-Izumo1 (HEK cells expressing turtle Izumo1 in CD200Cd4d3+4-COMP-blac-3xFLAG-6his) extracts, each with an apparent MW of 100 kDa (Fig. 4), which is consistent with the predicted molecular weight (HEK-JUNO: 89 kDa, HEK-Izumo1: 96 kDa). Bait and prey proteins were normalized using the ability of biotin to bind to streptomycin and β-lactamase enzyme activity, respectively (Supplementary Figs 9 and 10). Tu-JUNO of M. reevesii was expressed as bait and captured in individual wells of a streptavidin-coated microtiter plate before probing it for interactions with Tu-Izumo1 of M. mutica, C. flavomarginata, and C. picta bellii. Tu-JUNO of M. reevesii was able to interact directly with Tu-Izumo1 of all three other species. Similarly, Tu-Izumo1 of M. reevesii was able to interact with Tu-JUNO of M. mutica, C. flavomarginata, and C. picta bellii (Supplementary Table 3).
Western blot of Tu-Izumo1 and Tu-JUNO protein with anti-His antibodies. (A) M. reevesii; (B) M. mutica; (C) C. flavomarginata; (D) C. picta bellii.
Citation: Reproduction 162, 4; 10.1530/REP-21-0124
Western blot of Tu-Izumo1 and Tu-JUNO protein with anti-His antibodies. (A) M. reevesii; (B) M. mutica; (C) C. flavomarginata; (D) C. picta bellii.
Citation: Reproduction 162, 4; 10.1530/REP-21-0124
Western blot of Tu-Izumo1 and Tu-JUNO protein with anti-His antibodies. (A) M. reevesii; (B) M. mutica; (C) C. flavomarginata; (D) C. picta bellii.
Citation: Reproduction 162, 4; 10.1530/REP-21-0124
Both the interspecific interaction forces, between Tu-JUNO of M. mutica and Tu-Izumo1 of M. reevesii as well as between Tu-JUNO of M. reevesii and Tu-Izumo1 of M. mutica, were not significantly different from the intraspecific interaction force within M. reevesii (see the blue line in Fig. 5) (Supplementary Table 3). Notably, the interaction force between Tu-Izumo1 of M. reevesii and Tu-JUNO of C. flavomarginata did not differ significantly from intraspecific interaction force within C. flavomarginata, whereas the force between Tu-JUNO of M. reevesii and Tu-Izumo1 of C. flavomarginata was significantly weaker than the intraspecific interaction force (P = 0.027, t ratio = 3.394) (see the red line in Fig. 5) (Supplementary Table 3). Moreover, both the interspecific interaction forces, between Tu-JUNO of C. picta bellii and Tu-Izumo1 of M. reevesii as well as Tu-JUNO of M. reevesii and Tu-Izumo1 of C. picta bellii, were significantly weaker than the intraspecific interaction force within C. picta bellii (Tu-JUNO of M. reevesii and Tu-Izumo1 of C. picta bellii: P = 0.007, t ratio = 5.075; Tu-JUNO of C. picta bellii and Tu-Izumo1 of M. reevesii: P = 0.001, t ratio = 8.512) (see the black line in Fig. 5) (Supplementary Table 3).
The Tu-Izumo1 and Tu-JUNO interaction is direct and more conservative across turtles. Quantification of the absorbance values of the same screen was performed in triplicate. AVEXIS was used for binding analysis using recombinant Tu-Izumo1 proteins as prey and Tu-JUNO proteins as bait. Negative control 1 is a Cd4d3+4 bait probed with the prey protein; negative control 2 is a COMP prey probed with the bait protein. Bar graphs represent mean ± s.d., n = 3. t-Tests showed that the interaction between Tu-Izumo1 and Tu-JUNO is significantly different among different species (P < 0.05).
Citation: Reproduction 162, 4; 10.1530/REP-21-0124
The Tu-Izumo1 and Tu-JUNO interaction is direct and more conservative across turtles. Quantification of the absorbance values of the same screen was performed in triplicate. AVEXIS was used for binding analysis using recombinant Tu-Izumo1 proteins as prey and Tu-JUNO proteins as bait. Negative control 1 is a Cd4d3+4 bait probed with the prey protein; negative control 2 is a COMP prey probed with the bait protein. Bar graphs represent mean ± s.d., n = 3. t-Tests showed that the interaction between Tu-Izumo1 and Tu-JUNO is significantly different among different species (P < 0.05).
Citation: Reproduction 162, 4; 10.1530/REP-21-0124
The Tu-Izumo1 and Tu-JUNO interaction is direct and more conservative across turtles. Quantification of the absorbance values of the same screen was performed in triplicate. AVEXIS was used for binding analysis using recombinant Tu-Izumo1 proteins as prey and Tu-JUNO proteins as bait. Negative control 1 is a Cd4d3+4 bait probed with the prey protein; negative control 2 is a COMP prey probed with the bait protein. Bar graphs represent mean ± s.d., n = 3. t-Tests showed that the interaction between Tu-Izumo1 and Tu-JUNO is significantly different among different species (P < 0.05).
Citation: Reproduction 162, 4; 10.1530/REP-21-0124
Discussion
Characteristics of Tu-Izumo1 and Tu-JUNO in turtles
The study of CUB was reported in various organisms (Chaney & Clark 2015, Galtier et al. 2018), and previous studies have shown that synonymous mutation plays a major role in affecting the protein function, as well as molecular evolution (Wang et al. 2001, Carpen et al. 2006, Matsuo et al. 2007). Two major factors (mutation pressure and natural selection) are known to influence the CUB. However, no such study till date was reported in turtles Tu-Izumo1 and Tu-JUNO that is necessary for sperm–egg fusion. In this study, we used different bioinformatics tools (ENC plot and neutrality plot) to analyze the evolutionary pressure that shapes the CUB of Tu-Izumo1 and Tu-JUNO in turtles. Analysis indicates that turtles exhibit comparatively higher CUB toward G or C at the third position and less toward A or T in Tu-Izumo1 and Tu-JUNO, and natural selection, rather than mutational pressure, is the main determinant of CUB in turtles. Moreover, Tu-JUNO has positive selection sites and there is a correlation between Tu-Izumo1 and Tu-JUNO based on 25 turtles, which is highly consistent with previous results (Dong et al. 2019).
Tu-JUNO is present on the surface of turtle oocytes
In turtles, the oocyte structure from outside to inside consists of theca follicular epithelium, follicular cells, zona pellucid, yolk space, yolk granules, and vacuoles (Supplementary Fig. 7A). Follicular cells secrete steroid hormones and contain abundant microfilaments, which play an important role in promoting the synthesis of vitellogenin and the maturation and expulsion of eggs. The sperm of turtles consists of a cone-shaped head, a cricoid midpiece, and a slender tail (Supplementary Fig. 7B). Previous research has shown that both male Izumo1-deficient and female JUNO-deficient mice are infertile, suggesting that the interaction between Izumo1 and JUNO is essential for normal fertilization (Bianchi & Wright 2015). Many researchers have studied sperm–egg binding proteins and their interaction mechanism in mammals (Hanada & Chang 1972, Bianchi & Wright 2015). However, this is the first study of the expression of turtle sperm–egg binding proteins. We report for the first time that Tu-JUNO is expressed on the surface of turtle oocytes and that there is a direct interaction between Tu-Izumo1 and Tu-JUNO. Three key amino acids are conserved between mice and humans, which explain the interaction between mouse Izumo1 and human JUNO (Aydin et al. 2016, Ohto et al. 2016), and a similar situation exists in turtles (Dong et al. 2019).
The role of Tu-Izumo1 and Tu-JUNO in turtle hybridization
There are many reported cases of cross-breeding among M. reevesii, M. mutica, and C. flavomarginata (McCord 1997, Pan et al. 2009, Xia et al. 2011), which have overlapping geographical distributions. However, hybridization between these three species and C. picta bellii has not been reported. In our previous study, we speculated that low selection pressure in turtles on the binding area of Tu-Izumo1 and Tu-JUNO may be one of the reasons why closely related turtle species can hybridize whereas more distantly related ones cannot (Dong et al. 2019). In the current study, clear binding between Tu-Izumo1 and Tu-JUNO in different turtle species was observed by co-immunoprecipitation, which demonstrates that the interaction is conserved within turtles.
As reported by Noda et al. protein–protein interactions might be transiently established after Izumo1 translocation upon acrosome reaction, therefore, it would be difficult to detect the interaction between Tu-Izumo1 and all other known fusion-related proteins in testis or spermatozoa by co-IP. However, co-IP analyses using HEK293T cells demonstrated such interactions in mice (Noda et al. 2020). Our results demonstrate the feasibility of this method for detecting the interactions between Tu-Izumo1 and Tu-JUNO in turtles.
In addition, we directly measured the interaction force between Tu-Izumo1 and Tu-JUNO of different turtles. For the species that can hybridize with each other (M. reevesii, M. mutica, and C. flavomarginata), no significant difference in the interaction forces between sperm and egg binding proteins was found, both for crosses within the same species and for different species pairs. However, for these species that cannot hybridize with each other (M. reevesii and C. picta bellii), weaker interaction forces were found. Therefore, we hypothesize that the fusibility of the oocyte with spermatozoa in turtles could be related to a threshold level of Tu-Izumo1 and Tu-JUNO binding force.
Possible molecular mechanism regulating the interaction between Tu-Izumo1 and Tu-JUNO during hybridization of turtles
Gene editing studies have uncovered the essential roles of the sperm proteins SPACA6, TMEM95, and SOF1 in mice (Lorenzetti et al. 2014, Lamas-Toranzo et al. 2020, Noda et al. 2020), whereas the ablation of the sperm protein of FIMP1 causes severe subfertility in mice (Fujihara et al. 2020). The possible roles of these proteins on fertilization in other vertebrates such as turtles remain to be explored. In this study, results of the AVEXIS analysis showed that the interaction force between Tu-Izumo1 and Tu-JUNO gradually decreased as the genetic distance between hybrid turtles increased. CD9 tetraspanin is another protein that plays key role in sperm–egg fusion and is essential for fertilization in many animals (Li et al. 2004, Ziyyat et al. 2006, Zhou et al. 2009). The binding of sperm protein Izumo1 and its egg receptor JUNO can drive CD9 accumulation in the intercellular contact area prior to fusion during mammalian fertilization (Charrin et al. 2009, Jégou et al. 2011, Chalbi et al. 2014). Therefore, we speculated that the interaction force between Tu-Izumo1 and Tu-JUNO in turtles may affect the occurrence of hybridization by participating in the CD9 recruitment in turtles. Further experiments are needed to explore the mechanism of hybridization.
Conclusions
In conclusion, our results suggested that the codon usage bias is high in Tu-Izumo1 and Tu-JUNO, and most probably due to the role of compositional constraint in the presence of natural selection. It was also found that Tu-Izumo1 and Tu-JUNO interact with each other and this interaction is conserved in turtles. We proposed that the interaction between Izumo1 and JUNO may be involved in hybridization. Our results provided a theoretical basis for turtle breeding and protection of species independence. Besides, through our endeavor, a feasible method could be developed to assess the possibility of turtle hybridization by identifying the molecules involved in turtle hybridization in future work.
Supplementary materials
This is linked to the online version of the paper at https://doi.org/10.1530/REP-21-0124.
Declaration of interest
The authors declare that there is no conflict of interest that could be perceived as prejudicing the impartiality of the research reported.
Funding
This work was supported by the National Natural Science Foundation of China (NSFC, No. 31970499).
Author contribution statement
Conceptualization, L N and J D; Methodology, J D, Z M, and S L; Investigation, S L, K Z, M Y and J L; Writing – Original Draft, J D and Z M; Writing – Review and Editing, L N and J D; Supervision, L N. All authors read and approved the final manuscript.
References
Aydin H, Sultana A, Li S, Thavalingam A & Lee JE 2016 Molecular architecture of the human sperm IZUMO1 and egg Juno fertilization complex. Nature 534 562–565. (https://doi.org/10.1038/nature18595)
Bartholdson SJ, Bustamante LY, Crosnier C, Johnson S, Lea S, Rayner JC & Wright GJ 2012 Semaphorin-7A is an erythrocyte receptor for P. falciparum merozoite-specific TRAP homolog, MTRAP. PLoS Pathogens 8 e1003031. (https://doi.org/10.1371/journal.ppat.1003031)
Bian X, Gandahi JA, Liu Y, Yang P, Liu Y, Zhang L, Zhang Q & Chen Q 2013 The ultrastructural characteristics of the spermatozoa stored in the cauda epididymidis in Chinese soft-shelled turtle Pelodiscus sinensis during the breeding season. Micron 44 202–209. (https://doi.org/10.1016/j.micron.2012.06.010)
Bianchi E & Wright GJ 2015 Cross-species fertilization: the hamster egg receptor, Juno, binds the human sperm ligand, Izumo1. Philosophical Transactions of the Royal Society of London: Series B, Biological Sciences 370 20140101. (https://doi.org/10.1098/rstb.2014.0101)
Bianchi E, Doe B, Goulding D & Wright GJ 2014 Juno is the egg Izumo receptor and is essential for mammalian fertilization. Nature 508 483–487. (https://doi.org/10.1038/nature13203)
Bushell KM, Söllner C, Schuster-Boeckler B, Bateman A & Wright GJ 2008 Large-scale screening for novel low-affinity extracellular protein interactions. Genome Research 18 622–630. (https://doi.org/10.1101/gr.7187808)
Carpen JD, Schantz MV, Smits M, Skene DJ & Archer SN 2006 A silent polymorphism in the PER1 gene associates with extreme diurnal preference in humans. Journal of Human Genetics 51 1122–1125. (https://doi.org/10.1007/s10038-006-0060-y)
Chalbi M, Barraud-Lange V, Ravaux B, Howan K, Rodriguez N, Soule P, Ndzoudi A, Boucheix C, Rubinstein E & Wolf JP et al.2014 Binding of sperm protein Izumo1 and its egg receptor Juno drives CD9 accumulation in the intercellular contact area prior to fusion during mammalian fertilization. Development 141 3732–3739. (https://doi.org/10.1242/dev.111534)
Chaney JL & Clark PL 2015 Roles for synonymous codon usage in protein biogenesis. Annual Review of Biophysics 44 143–166. (https://doi.org/10.1146/annurev-biophys-060414-034333)
Charrin S, Le Naour F, Silvie O, Milhiet PE, Boucheix C & Rubinstein E 2009 Lateral organization of membrane proteins: tetraspanins spin their web. Biochemical Journal 420 133–154. (https://doi.org/10.1042/BJ20082422)
Creighton P & Houghton JA 1987 Visualization of pig sperm chromosomes by in-vitro penetration of zona-free hamster ova. Journal of Reproduction and Fertility 80 619–622. (https://doi.org/10.1530/jrf.0.0800619)
Dong J, Jiang H, Xiong L, Zan J, Liu J, Yang M, Zheng K, Wang Z & Nie L 2019 Detecting coevolution of positively selected in turtles sperm-egg fusion proteins. Mechanisms of Development 156 1–7. (https://doi.org/10.1016/j.mod.2019.02.001)
Fujihara Y, Lu Y, Noda T, Oji A, Larasati T, Kojimakita K, Yu Z, Matzuk RM, Matzuk MM & Ikawa M 2020 Spermatozoa lacking fertilization influencing membrane protein (FIMP) fail to fuse with oocytes in mice. PNAS 117 9393–9400. (https://doi.org/10.1073/pnas.1917060117)
Galtier N, Roux C, Rousselle M, Romiguier J, Figuet E, Glémin S, Bierne N & Duret L 2018 Codon usage bias in animals: disentangling the effects of natural selection, effective population size, and GC-biased gene conversion. Molecular Biology and Evolution 35 1092–1103. (https://doi.org/10.1093/molbev/msy015)
Grayson P & Civetta A 2012 Positive selection and the evolution of izumo genes in mammals. International Journal of Evolutionary Biology 2012 958164. (https://doi.org/10.1155/2012/958164)
Guindon S, Dufayard JF, Lefort V, Anisimova M, Hordijk W & Gascuel O 2010 New algorithms and methods to estimate maximum-likelihood phylogenies: assessing the performance of PhyML 3.0. Systematic Biology 59 307–321. (https://doi.org/10.1093/sysbio/syq010)
Hanada A & Chang MC 1972 Penetration of zona-free eggs by spermatozoa of different species. Biology of Reproduction 6 300–309. (https://doi.org/10.1093/biolreprod/6.2.300)
Ikawa M, Inoue N, Benham AM & Okabe M 2010 Fertilization: a sperm’s journey to and interaction with the oocyte. Journal of Clinical Investigation 120 984–994. (https://doi.org/10.1172/JCI41585)
Inoue N, Ikawa M, Isotani A & Okabe M 2005 The immunoglobulin superfamily protein Izumo is required for sperm to fuse with eggs. Nature 434 234–238. (https://doi.org/10.1038/nature03362)
Jean C, Haghighirad F, Zhu Y, Chalbi M, Ziyyat A, Rubinstein E, Gourier C, Yip P, Wolf JP & Lee JE et al.2019 Juno, the receptor of sperm IZUMO1, is expressed by the human oocyte and is essential for human fertilisation. Human Reproduction 34 118–126. (https://doi.org/10.1093/humrep/dey340)
Jégou A, Ziyyat A, Barraud-Lange V, Perez E, Wolf JP, Pincet F & Gourier C 2011 CD9 tetraspanin generates fusion competent sites on the egg membrane for mammalian fertilization. PNAS 108 10946–10951. (https://doi.org/10.1073/pnas.1017400108)
Kerr JS & Wright GJ 2012 Avidity-based extracellular interaction screening (AVEXIS) for the scalable detection of low-affinity extracellular receptor-ligand interactions. Journal of Visualized Experiments 61 e3881. (https://doi.org/10.3791/3881)
Kumar S, Stecher G & Tamura K 2016 MEGA7: molecular evolutionary genetics analysis version 7.0 for bigger datasets. Molecular Biology and Evolution 33 1870–1874. (https://doi.org/10.1093/molbev/msw054)
Lamas-Toranzo I, Hamze JG, Bianchi E, Fernández-Fuertes B, Pérez-Cerezales S, Laguna-Barraza R, Fernández-González R, Lonergan P, Gutiérrez-Adán A & Wright GJ et al.2020 TMEM95 is a sperm membrane protein essential for mammalian fertilization. eLife 9 e53913. (https://doi.org/10.7554/eLife.53913)
Li YH, Hou Y, Ma W, Yuan JX, Zhang D, Sun QY & Wang WH 2004 Localization of CD9 in pig oocytes and its effects on sperm-egg interaction. Reproduction 127 151–157. (https://doi.org/10.1530/rep.1.00006)
Librado P & Rozas J 2009 DnaSP v5: a software for comprehensive analysis of DNA polymorphism data. Bioinformatics 25 1451–1452. (https://doi.org/10.1093/bioinformatics/btp187)
Lorenzetti D, Poirier C, Zhao M, Overbeek PA, Harrison W & Bishop CE 2014 A transgenic insertion on mouse chromosome 17 inactivates a novel immunoglobulin superfamily gene potentially involved in sperm–egg fusion. Mammalian Genome 25 141–148. (https://doi.org/10.1007/s00335-013-9491-x)
Matsuo M, Shiino Y, Yamada N, Ozeki Y & Okawa M 2007 A novel SNP in hPer2 associates with diurnal preference in a healthy population. Sleep and Biological Rhythms 5 141–145. (https://doi.org/10.1111/j.1479-8425.2007.00264.x)
McCord WP 1997 Mauremys pritchardi, a new batagrid turtles from Myanmar and Yunnan, China. Chelonian Conservation and Biology 2 555–562.
Nishimura K, Han L, Bianchi E, Wright GJ, De Sanctis D & Jovine L 2016 The structure of sperm Izumo1 reveals unexpected similarities with Plasmodium invasion proteins. Current Biology 26 R661–R662. (https://doi.org/10.1016/j.cub.2016.06.028)
Noda T, Lu Y, Fujihara Y, Oura S, Koyano T, Kobayashi S, Matzuk MM & Ikawa M 2020 Sperm proteins SOF1, TMEM95, and SPACA6 are required for sperm-oocyte fusion in mice. PNAS 117 11493–11502. (https://doi.org/10.1073/pnas.1922650117)
Ohto U, Ishida H, Krayukhina E, Uchiyama S, Inoue N & Shimizu T 2016 Structure of IZUMO1–JUNO reveals sperm–oocyte recognition during mammalian fertilization. Nature 534 566–569. (https://doi.org/10.1038/nature18596)
Pan D, Chen K, Zhu X, Zheng G, Liu Y, Chen Y & Li KB 2009 The morphologic characters of hybrid (Mauremys mutica ♀ × Cuora trifasciata ♂) and comparison with their paren. Acta Hydrobiologica Sinica 33 620–626. (https://doi.org/10.3724/SP.J.1035.2009.40620)
Ronquist F & Huelsenbeck JP 2003 MrBayes 3: bayesian phylogenetic inference under mixed models. Bioinformatics 19 1572–1574. (https://doi.org/10.1093/bioinformatics/btg180)
Satouh Y, Inoue N, Ikawa M & Okabe M 2012 Visualization of the moment of mouse sperm-egg fusion and dynamic localization of IZUMO1. Journal of Cell Science 125 4985–4990. (https://doi.org/10.1242/jcs.100867)
Sheng Z, Qin Z, Chen Z, Zhao Y & Zhong J 2007 The factors shaping synonymous codon usage in the genome of Burkholderia mallei. Journal of Genetics and Genomics 34 362–372. (https://doi.org/10.1016/S1673-8527(0760039-3)
Stothard P 2000 The sequence manipulation suite: JavaScript programs for analyzing and formatting protein and DNA sequences. BioTechniques 28 1102, 1104. (https://doi.org/10.2144/00286ir01)
Sueoka N 1988 Directional mutation pressure and neutral molecular evolution. PNAS 85 2653–2657. (https://doi.org/10.1073/pnas.85.8.2653)
Suzuki D, Yabe T & Hikida T 2013 Hybridization between Mauremys japonica and Mauremys reevesii inferred by nuclear and mitochondrial DNA analyses. Journal of Herpetology 48 445–454. (https://doi.org/10.1670/11-320)
Vicens A & Roldan ERS 2014 Coevolution of positively selected IZUMO1 and CD9 in rodents: evidence of interaction Between gamete fusion proteins? Biology of Reproduction 90 113. (https://doi.org/10.1095/biolreprod.113.116871)
Wang HC, Badger J, Kearney P & Li M 2001 Analysis of codon usage patterns of bacterial genomes using the self-organizing map. Molecular Biology and Evolution 18 792–800. (https://doi.org/10.1093/oxfordjournals.molbev.a003861)
Wong WS, Yang Z, Goldman N & Nielsen R 2004 Accuracy and power of statistical methods for detecting adaptive evolution in protein coding sequences and for identifying positively selected sites. Genetics 168 1041–1051. (https://doi.org/10.1534/genetics.104.031153)
Xia X, Wang L, Nie L, Huang Z, Jiang Y, Jing W & Liu L 2011 Interspecific hybridization between Mauremys reevesii and Mauremys sinensis: evidence from morphology and DNA sequence data. African Journal of Biotechnology 10 6716–6724. (https://doi.org/10.5897/AJB11.063)
Yang Z 2007 PAML 4: phylogenetic analysis by maximum likelihood. Molecular Biology and Evolution 24 1586–1591. (https://doi.org/10.1093/molbev/msm088)
Zhong J, Li Y, Zhao S, Liu S & Zhang Z 2007 Mutation pressure shapes codon usage in the GC-Rich genome of foot-and-mouth disease virus. Virus Genes 35 767–776. (https://doi.org/10.1007/s11262-007-0159-z)
Zhou GB, Liu GS, Meng QG, Liu Y, Hou YP, Wang XX, Li N & Zhu SE 2009 Tetraspanin CD9 in bovine oocytes and its role in fertilization. Journal of Reproduction and Development 55 305–308. (https://doi.org/10.1262/jrd.20099)
Ziyyat A, Rubinstein E, Monier-Gavelle F, Barraud V, Kulski O, Prenant M, Boucheix C, Bomsel M & Wolf JP 2006 CD9 controls the formation of clusters that contain tetraspanins and the integrin α6β1, which are involved in human and mouse gamete fusion. Journal of Cell Science 119 416–424. (https://doi.org/10.1242/jcs.02730)
