MBE Advance Access originally published online on February 14, 2008
Molecular Biology and Evolution 2008 25(5):972-979; doi:10.1093/molbev/msn046
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Research Articles |
Sexual Selection and Maintenance of Sex: Evidence from Comparisons of Rates of Genomic Accumulation of Mutations and Divergence of Sex-Related Genes in Sexual and Hermaphroditic Species of Caenorhabditis
Department of Biology, McMaster University, Hamilton, Ontario, Canada
E-mail: singh{at}mcmaster.ca.
 |
Abstract |
|---|
 TOP Abstract IntroductionMaterials and MethodsResultsDiscussionSupplementary MaterialAcknowledgementsReferences |
|---|
Several hypotheses have been proposed to explain the persistence of dioecy despite the reproductive advantages conferred to hermaphrodites, including greater efficiency at purging deleterious mutations in the former. Dioecy can benefit from both mutation purging and accelerated evolution by bringing together beneficial mutations in the same individual via recombination and shuffling of genotypes. In addition, mathematical treatment has shown that sexual selection is also capable of mitigating the cost of maintaining separate sexes by increasing the overall fitness of sexual populations, and genomic comparisons have shown that sexual selection can lead to accelerated evolution. Here, we examine the advantages of dioecy versus hermaphroditism by comparing the rate of evolution in sex-related genes and the rate of accumulation of deleterious mutations using a large number of orthologs (11,493) in the dioecious Caenorhabditis remanei and the hermaphroditic Caenorhabditis briggsae. We have used this data set to estimate the deleterious mutation rate per generation, U, in both species and find that although it is significantly higher in hermaphrodites, both species are at least 2 orders of magnitude lower than the value required to explain the persistence of sex by efficiency at purging deleterious mutations alone. We also find that genes expressed in sperm are evolving rapidly in both species; however, they show a greater increase in their rate of evolution relative to genes expressed in other tissues in C. remanei, suggesting stronger sexual selection pressure acting on these genes in dioecious species. Interestingly, the persistence of a signal of rapid evolution of sperm genes in C. briggsae suggests a recent evolutionary origin of hermaphrodism in this lineage. Our results provide empirical evidence of increased sexual selection pressure in dioecious animals, supporting the possibility that sexual selection may play an important role in the maintenance of sexual reproduction.
Key Words: sexual selection • deleterious mutation rate • sperm • Caenorhabditis
|
Introduction |
|---|
|
TOPAbstractIntroductionMaterials and MethodsResultsDiscussionSupplementary MaterialAcknowledgementsReferences |
|---|
Males and females of dioecious species (species with independent sexes) often display highly dimorphic secondary sexual traits (Eberhardt 1985
). Darwin (1871) proposed the theory of sexual selection in order to explain why traits that increase the reproductive success of individuals (irrespective of their survival) may be favored and maintained. Mathematical treatments of sexual selection in dioecious species have shown that the strength of selection is dependent on the amount of effort that is invested in searching out and acquiring mates (usually by males) versus taking part in other parental activities such as maturing eggs or rearing offspring (Sutherland 1985, 1987). Under such theory, self-fertile hermaphroditic species are believed to be subject to weaker pressures of sexual selection, due to reduced competition for mates (Greeff and Michiels 1999). The same will also be true of androdioecious species (i.e., species with self-fertile hermaphrodites in addition to separate males) albeit with expectations of a less extreme effect. Recent empirical observations support such theoretical predictions. For example, Chasnov et al. (2007) found that hermaphrodites of the species Caenorhabditis elegans and Caenorhabditis briggsae have lost the ability to produce male-attracting pheromones, presumably because reproduction in these species occurs primarily by hermaphroditic self-fertilization and there is little selection pressure to retain the ability to attract males.
The mainstream hypothesis explaining the maintenance of dioecy is the mutational deterministic (MD) hypothesis (Kondrashov 1988), which postulates that outcrossing will offer a net fitness advantage over selfing when the deleterious mutation rate per generation, U, is sufficiently large (
1.4) and only under conditions of synergistic epistasis (i.e., when the negative fitness impact of multiple deleterious mutations is greater than strictly additive) (Charlesworth 1990; Cutter and Payseur 2003). The strength of sexual selection has also been used to explain the maintenance of obligate outcrossing in the face of the reproductive advantage conferred by an asexual reproductive system (the so-called 2-fold cost of sex). Treatment of the subject has indicated that, all else being equal, if sexual selection causes deleterious mutations to have greater fitness impact on males, such mutations will be maintained in sexual populations at a lower equilibrium frequency than in asexual populations (Agrawal 2001; Siller 2001). Such a model may be extended into a comparison between dioecious and androdioecious species in that the reproductive advantage of the latter may be mitigated by the advantage of increased strength of sexual selection in the former. Although most work in this field has been performed in the model research genus of Drosophila, such a system does not allow us the opportunity to directly compare the effects of sexual selection and the deleterious mutation rate between different mating types (i.e., dioecious vs. androdioecious species).
The effect of mating type on genome evolution has been studied in some detail in plants where it is common to find closely related species with different mating types (e.g., Arabidopsis thaliana and Arabidopsis lyrata). Such studies have found evidence that hermaphroditic species display reduced amounts of neutral polymorphism, higher levels of linkage disequilibrium, and a generally reduced efficiency of selection relative to dioecious species due in large part to increased Hill–Robertson interference in heavily inbred selfers (Hill and Robertson 1966; Charlesworth and Wright 2001; Glémin et al. 2006). However, estimates of U between A. thaliana and A. lyrata have produced values that are well below the threshold value of 1.4 required to explain the maintenance of dioecy by the MD hypothesis alone (0.22 to 0.58; Wright et al. 2002). It should be noted that these estimates have been calculated from a limited subset of genes as the whole-genome sequence of A. lyrata has not yet been completed.
The nematode genus Caenorhabditis comprises both dioecious species (Caenorhabditis remanei, Caenorhabditis brenneri, Caenorhabditis japonica, etc.) as well as 2 known androdioecious species (C. elegans and C. briggsae). As the full-genome sequences of C. elegans, C. briggsae, and C. remanei are available, comparison of the patterns of evolution of genes among these species provides an excellent opportunity to compare how different mating systems affect the broad patterns of genome evolution in animals. Numerous studies have shown that sex- and reproduction-related (SRR) traits, particularly those involved in male reproductive function, evolve rapidly at both the morphological and genetic levels in a wide variety of taxa and that many of these genes evolve adaptively, indicating that their functions are beneficial to reproduction (Civetta and Singh 1998; Swanson and Vacquier 2002). Such observations are consistent with sexual selection in terms of female choice and competition for mates (especially sperm competition) acting as the primary driving force accelerating male evolution (Jagadeeshan and Singh 2005). Whole-genome comparisons between these nematode species would allow us to examine the strength of sexual selection among different mating types by comparing the rates of evolution of SRR proteins. In addition, these data allow for more reliable estimates of U in C. briggsae and C. remanei than were previously possible through the use of comparative genomic data (Cutter and Payseur 2003).
To this end, we have generated a large data set of 3-way orthologous genes among these species in order to compare the rates of evolution of a number of gene categories involved in SRR and non-SRR function. We find that genes expressed in sperm are evolving rapidly in both C. remanei and C. briggsae; however, sperm genes in C. remanei are showing a greater increase in their rate of evolution relative to other gene categories, providing evidence of stronger sexual selection in dioecy. We also use our data set in order to obtain improved estimates of U in these species using previously established methods (Eyre-Walker and Keightley 1999; Cutter and Payseur 2003). In contrast, and unlike previous estimates of U, we find that hermaphrodites experience a significantly greater genomic rate of deleterious mutation as well as an overall greater rate of nonsynonymous substitution. These results shed light on the role of sexual selection and deleterious mutation in the maintenance of sexual reproduction in nematodes.
|
Materials and Methods |
|---|
|
TOPAbstractIntroductionMaterials and MethodsResultsDiscussionSupplementary MaterialAcknowledgementsReferences |
|---|
Genomic Data Sets
We obtained all unique, predicted peptides for C. elegans (release WS173), C. briggsae (release WS173), and C. remanei (release 11/29/2005) from the Wormbase FTP site (ftp://ftp.wormbase.org/pub/wormbase/genomes/). As the C. remanei predicted peptide data set is known to contain redundant copies of genes due to heterozygosity in the sequenced genome (Schwartz E, personal communication), we used Cluster Database at High Identity with Tolerance (CD-HIT) (Li and Godzik 2006
) in order to cluster and remove all additional transcripts that had greater than or equal to 98% sequence similarity at the protein level. The original data set of 25,948 transcripts was truncated down to 24,267 nonredundant transcripts that were used in further analysis.
INPARANOID (Remm et al. 2002) was run with default parameters, using Blastall version 2.2.14 with VT emulation, on all 3 pairwise species comparisons. One-to-one best-hit reciprocal orthologs were collected and clustered into 3-way best-hit reciprocal orthologous trios. This method generated 11,594 orthologous gene trios of which 11,493 annotated transcripts were free of in-frame stop codons in all 3 species and thus could be used in further analysis.
Evolutionary Rate Estimates
The amino acid sequences of each orthologous trio were aligned with Dialign 2.2 using default parameters (Morgenstern 1999). The nucleotide sequences were then aligned with RevTrans 1.4 according to their corresponding protein alignments (Wernersson and Pedersen 2003). Nonsynonymous (dN) and synonymous (dS) rates of divergence were computed for C. briggsae and C. remanei (using C. elegans as the outgroup species) using CODEML from phylogenetic analysis by maximum likelihood (PAML) 3.15 (Yang and Nielsen 2002) under Model 0, in which a single rate was calculated for the entire phylogeny, as well as Model 1 in which a separate rate is estimated for each branch of the phylogeny.
Classification of Gene Functions
Genes were classified into functional categories based on the tissue/sex in which they display their highest level of expression according to Reinke et al. (2004). Genes were pooled into the following categories (with the number of genes in each category shown in brackets next to the name of the category itself): sperm (361), oocyte (622), hermaphrodite (nonsperm, nonoocyte; 684), male (nonsperm, nonoocyte; 289), and nonsex (nonsex biased, nonsperm, and nonoocyte; 9 537).
Calculation of Genomic Deleterious Mutation Rate
Rates of total mutation (M) and deleterious mutation (U) per generations were computed according to the procedure established by Eyre-Walker and Keightley (1999) and modified by Cutter and Payseur (2003): M = Z(
L x dS)/L, U = M – Z(L x dN)/L, where L = gene length in nucleotides, Z = 2 (genomes) x number of genes x average gene length x (1/generation per year) x (1/divergence time). We used a genome size of 19,500 genes as per estimates of Stein et al. (2003), an average gene length of 1403.37 and 1414.62 for C. briggsae and C. remanei, respectively, computed from 10 822 genes with dS < 3, 90 generations per year (Denver et al. 2000) and a divergence time of 44.5 Myr (Cutter and Payseur 2003). U was also computed according to the neutral mutation rates proposed by Drake et al. (1998) (supplementary table 3, Supplementary Material online). In order to take into account potential selection on synonymous sites, which would reduce the reliability of using dS as a measure of the neutral mutation rate, we used a corrected measure of dS using the frequency of optimal codons (Fop, CodonW, http://codonw.sourceforge.net/) taking into account the codon usage bias of C. elegans (Stenico et al. 1994) for both C. briggsae (dS = –1.6 x Fop + 1.527, R2 = 0.133, P < 2.2 x 10–16) and C. remanei (dS = –1.302 x Fop + 1.257, R2 = 0.094, P < 2.2 x 10–16). We computed 95% confidence intervals (CIs) from U using 1 000 bootstraps (Manly 1991).
Statistics
All statistical analyses were performed using the R statistical package (R Development Core Team 2004). Permuted Kruskal–Wallis rank sum tests were performed with 10 000 permutations of the data using the "coin" package.
|
Results |
|---|
|
TOPAbstractIntroductionMaterials and MethodsResultsDiscussionSupplementary MaterialAcknowledgementsReferences |
|---|
Faster Evolution of Sperm Proteins
We identified 11 493 gene trios free of in-frame stop codons that were orthologous between C. elegans, C. briggsae, and C. remanei and were thus suitable for this analysis (supplementary table 1, Supplementary Material online). Using large-scale expression study of Reinke et al. (2004)
in C. elegans, we classified each of the 3-way orthologs into 1 of 5 functional categories (the number of genes in each category is shown in parentheses): sperm (361), oocyte (622), hermaphrodite (684), male (289), and nonsex (9 537). Rates of synonymous substitutions per synonymous site (dS) and nonsynonymous substitutions per nonsynonymous site (dN) were estimated for all orthologs using PAML (Yang and Nielsen 2002) under Model 0, allowing for a single rate among all branches (supplementary fig. 1, Supplementary Material online). Given the long estimates of divergence times between the 3 species, dS was saturated (i.e., >3 substitutions per site, supplementary table 1 [Supplementary Material online]) and thus only dN was used in further analysis of the entire data set. We found that genes classified as having their highest level of expression in sperm according to Reinke et al. (2004) had a significantly higher dN than other gene categories (Tukey Honestly Significant Difference [HSD] test, P < 0.001 in all comparisons; supplementary fig. 1 [Supplementary Material online]). In contrast, nonsex genes are evolving significantly slower than all other gene categories (Tukey HSD test, P < 0.001 in all comparisons). In order to control for a possible bias in our results due to saturation of dS, we performed the analysis again, this time using only genes with dS < 3. Our conclusions remain the same as sperm genes are showing a greater dN than other gene categories (Tukey HSD test, P < 0.001 in all comparisons; supplementary table 2 [Supplementary Material online]). Furthermore, using the truncated data set, we were able to compare the dN/dS ratio between gene categories, revealing a significantly higher dN/dS for both male and sperm genes as compared with other gene categories (Tukey HSD test, P < 0.001 in all comparison). These 2 classes were not significantly different from one another (P = 0.986). We repeated the analysis of rates of evolutionary divergence, this time calculating a separate dN for each branch of the phylogeny (Model 1 in PAML). There was a significantly higher median dN in the branch leading to C. briggsae than in the branch leading to C. remanei (median dN of 0.0459 and 0.0346 for C. briggsae and C. remanei, respectively; Kruskal–Wallis rank sum test, 10 000 replicates P < 2.2·10–16). Sperm genes evolve significantly more rapidly than all other gene categories in C. remanei (Tukey HSD test, P < 0.001 in all comparisons), whereas in C. briggsae sperm genes are showing a significantly higher dN than nonsex genes and male genes (P < 0.001 in both comparisons) but do not differ from hermaphrodite genes (P = 0.293) nor oocyte genes (P = 0.4). However, there was no significant difference in the evolutionary rates of sperm genes (P = 0.998) nor male genes (P = 0.247) between mating types (i.e., dioecy vs. androdioecy; fig. 1). When limiting the analysis to genes showing dS < 3 in both branches, our conclusions remain the same (supplementary table 2, Supplementary Material online).
|
Accelerated Sperm Evolution in Dioecy
Previous mutational models predict that hermaphroditic species should have a higher rate of fixation of slightly deleterious mutations due to inbreeding, leading to reduced effective population size (Charlesworth 1990
), the results of which could lead to a higher rate of substitutions in hermaphrodites and thus the higher overall dN observed in C. briggsae. With this in mind, we sought to determine whether the degree of accelerated evolution of sperm proteins relative to other gene classes was similar for both mating types using an average centering approach and, in doing so, control for the effect of a higher rate of nonsynonymous substitution in hermaphrodites. We compared only the residuals of the dN estimates for each gene category (i.e., the value of the dN for each gene from which is subtracted the mean dN among all genes within that species), allowing us to determine the degree to which the dN for each category diverges from the species average. Pooling the residuals of the dN into sperm and nonsperm categories and performing an analysis of variance on these 2 groups while taking into account the effect of mating type, we observed a significant interaction between mating type and the variation of dN between the 2 gene categories (F1,22982 = 4.7937, P = 0.02857), indicating a faster rate of evolution for sperm genes relative to nonsperm genes in the dioecious species, C. remanei, when compared with the androdioecious C. briggsae (fig. 1 inset). In order to rule out any potential bias in our data due to the long divergence times between species, we also repeated the analysis of variance using only genes with dS < 3. Our conclusions remain the same (F1,21640 = 5.1462, P = 0.02334).
Higher Genomic Rate of Deleterious Mutations in Hermaphrodites
In order for the MD hypothesis alone to explain the maintenance of dioecy against invasion from reproductively advantageous hermaphrodites, the deleterious mutation rate per genome per generation, U, would have to be on the order 1.4 (Cutter and Payseur 2003). Furthermore, as U itself appears to be an adaptive trait (Bjedov et al. 2003; Baer et al. 2005), we would predict that the deleterious mutation rate per genome per generation would be higher in dioecious species as compared with related androdioecious species. A previous study attempted to compare the genomic deleterious mutation rate, U, between Caenorhabditis species using a small sample of genes and found no significant differences between mating types (Cutter and Payseur 2003). Using our whole-genome ortholog data set, we attempted to obtain a more accurate estimate of U for both C. briggsae and C. remanei using a modified version of method of Cutter and Payseur (2003) (see Materials and Methods). We used the following parameters when estimating U: 19 500 genes in both species (Stein et al. 2003), an average gene length of 1403.37 and 1414.62 nt for C. briggsae and C. remanei, respectively, a divergence time of 44.5 Myr between the 2 species (Cutter and Payseur 2003) and 90 generations per year. Under these conditions, we find that C. briggsae has a significantly higher U than C. remanei (0.01104 with 95% CI: 0.0108–0.0112 and 0.00962 with 95% CI: 0.0094–0.0097; alternate values of U calculated from a variety of biologically realistic parameters are presented in supplementary table 3 [Supplementary Material online]).
Effect of Chromosome Position on Gene Evolution
Previous studies have shown that chromosomal position can have a significant effect on the rates of evolution of genes, especially those involved in SRR function (Torgerson and Singh 2003, 2006; Stevison et al. 2004). Several studies have found that sex chromosomes and autosomes can display markedly different rates of evolution due to the unique, sex-dependent patterns of transmission of the former (Charlesworth et al. 1987; Torgerson and Singh 2003; Counterman et al. 2004; Musters et al. 2006; Mank et al. 2007). However, both C. elegans and C. briggsae reproduce overwhelmingly by self-fertilization and we should expect the sex chromosome, X, to behave like an autosome as it spends the majority of its time in the homogametic state. Because the C. remanei genome has not yet had contigs assigned to chromosomes, we determined the chromosomal positions for all the 3-way orthologous genes for which data were available in C. briggsae and C. elegans (Bieri et al. 2007). Gene synteny information was available for a total 10,304 genes from our data set, of which 403 genes do not present conserved synteny between the 2 species and 1,189 genes are not assigned to chromosomes in C. briggsae. We first compared the rate of evolution of genes between different chromosomes for the genes whose synteny was conserved between the 2 species. As expected, we were unable to detect a significant difference between the X chromosome as compared with the autosomes in the average rate of nonsynonymous substitution in genes (Tukey HSD test, P > 0.05 in each comparison), with the exception of chromosome IV, which is showing a significantly lower dN than genes on chromosomes X, I, and V (P = 0.0354, 1.6·10–4, and 7.63·10–3, respectively). Among the genes whose synteny was not conserved between C. elegans and C. briggsae, we determined whether any of the functional categories were overrepresented. We found that sperm genes are significantly overrepresented among genes with nonconserved synteny (
2 test = 425, degree of freedom [df] = 1, P = 0; table 2), whereas nonsex genes are significantly underrepresented among these (2 test = 23.37, df = 1, P = 1.3 x 10–6; table 2). The differences between other categories were nonsignificant. Genes lacking conserved synteny are also presenting a significantly higher dN (under Model 0) than genes with conserved synteny (Kruskal–Wallis rank sum test, P < 2.2 x 10–16). Removal of sperm genes from this analysis does not change the results (P < 2.2 x 10–16), indicating that this phenomenon is not simply due to an enrichment of rapidly evolving sperm genes.
|
|
Discussion |
|---|
|
TOPAbstractIntroductionMaterials and MethodsResultsDiscussionSupplementary MaterialAcknowledgementsReferences |
|---|
Evolution of Sperm Proteins
Accelerated evolution of spermatogenesis proteins is believed to result from sexual selection in the form of sperm competition and gametic interactions (Jagadeeshan and Singh 2005
; Singh and Kulathinal 2005). Theoretical modeling has suggested that the opportunity for sexual selection, and thus the strength of selection itself, is reduced in hermaphroditic species as compared with dioecious species, due primarily to reduced competition for mates among self-fertilizing hermaphrodites (Greeff and Michiels 1999). Despite these considerations, we find evidence for a persistent signal of accelerated evolution among genes involved in spermatogenesis within the lineage leading to C. briggsae (fig. 1). A previous study also found that sperm genes evolve more rapidly than other gene categories when performing a pairwise analysis between C. briggsae and C. elegans (Cutter and Ward 2005). Such observations could result from 2 possible evolutionary scenarios, neither of which is mutually exclusive.
It is possible that in natural populations of C. briggsae, sufficient sperm competition occurs to drive the rapid evolution of spermatogenic proteins. It is unlikely that gametic competition within individual selfers could occur as hermaphrodites are presenting near 100% efficiency of sperm usage (Ward and Carrel 1979). However, sexual selection will favor efficient male sperm in order to outcompete those of the hermaphrodite upon successful mating (Chasnov and Chow 2002). Although the frequency of outcrossing should be low in natural populations of C. briggsae (Cutter et al. 2006), there may still remain enough opportunity to produce a detectable signal of selection among sperm proteins. Alternatively, the rapid evolution of spermatogenic proteins within hermaphrodites could also be explained if androdioecy has evolved recently within the lineage leading to C. briggsae. In this case, our results would reflect the accelerated evolution of sex genes within the dioecious ancestor to C. briggsae (Cutter and Payseur 2003), the signal of which remains detectable in the present hermaphroditic state. Such a possibility is further supported by evolutionary studies performed in plants which suggest that despite the reproductive advantages acquired by evolving hermaphrodism, this mating type may ultimately prove to be an evolutionary "dead end" (Takebayashi and Morrell 2001). If hermaphrodite genomes have reduced efficiency of natural selection and thus reduced efficiency in adaptation, as theory predicts, this mating type could be difficult to sustain over long periods of evolutionary time. Thus, although hermaphrodism may evolve frequently in some lineages, it may not be evolutionarily stable over long periods. The extremely low frequency of males in natural populations of C. briggsae, coupled to the reduced opportunity for sexual selection in androdioecious species, would suggest that accelerated evolution of sperm proteins is most likely a remnant of a recent ancestral, male–female state of the C. briggsae lineage.
As shown in table 1, C. briggsae has a higher rate of nonsynonymous substitution than does C. remanei in most gene categories, with the exception of male genes and those involved in spermatogenesis. The increased rate of nonsynonymous substitution observed in C. briggsae may be the result of 2 nonexclusive processes: First, the rate of fixation of slightly deleterious amino acid substitutions is predicted to be higher in nonoutcrossing species relative to obligate outcrossers due to interference of selection between closely linked loci, the "Hill–Robertson effect" (Hill and Robertson 1966; Ohta 1973). As most amino acid substitutions are presumed to be slightly deleterious (Kimura 1968), a higher dN is to be expected in hermaphrodites as compared with dioecious species. Second, it is possible that C. briggsae has evolved an increased genomic mutation rate relative to that of C. remanei. Experimental evidence suggests that the mutation rate of C. briggsae may be higher than that of C. elegans (Ostrow et al. 2007); however, there is no experimental evidence indicating that C. briggsae experiences a higher genomic mutation rate than C. remanei (Cutter and Payseur 2003). Therefore, a comparison of the rates of evolution of gene categories between species of different mating types would require an appropriate correction for the expected differences in the rate of nonsynonymous substitution. We find that C. remanei shows a greater acceleration in the rate of evolution of proteins expressed in sperm relative to other tissues (fig. 1) than does C. briggsae. Such an observation is consistent with previous, theoretical predictions that species with separate males and females should be exposed to stronger sexual selection pressure than simultaneous hermaphrodites (Greeff and Michiels 1999).
|
Effect of Chromosome Position and Synteny on Gene Evolution
We observed a significant overrepresentation of genes involved in spermatogenesis among those that are nonsyntenic between C. elegans and C. briggsae (table 2). These results support previous studies in other phyla that found a significant enrichment of testis-expressed genes among those that are translocated as well as a high frequency of retrotransposition of genes off of the X chromosome (Betrán et al. 2002
; Emerson et al. 2004). Despite their current hermaphroditic status, it would appear that these species were subject to similar selective pressures as other dioecious taxa, which act to translocate sex-biased genes into more favorable positions within the genome (Miller et al. 2004; Kulathinal and Singh 2005). It is unlikely that such pressure would continue to act in the context of hermaphrodism, which implies that these observations reflect pressures that existed in an ancestral dioecious state. Coupled to the observation of a persistent signal of rapid evolution of sperm proteins, this suggests a relatively recent origin of hermaphrodism in these species. The discovery of a new dioecious species, JU727, more closely related to C. briggsae than C. remanei (Kiontke and Sudhaus 2006) further supports this hypothesis.
Comparison of Deleterious Mutation Rate between Mating Types
Given the reproductive advantages possessed by hermaphrodites over dioecious populations (Charlesworth 1990; Cutter 2005), the persistence of separate males and females calls for an explanation. Several hypotheses have been forwarded in order to explain the evolution and maintenance of different mating types. The MD hypothesis argues that, within the context of synergistic epistasis, dioecy is maintained due to its efficiency at purging deleterious mutations (Kondrashov 1988). In order for the MD hypothesis to explain the maintenance of separate sexes, it would require that dioecious species experience a greater U than hermaphrodites and be above a threshold of 1.4 (Lloyd 1980; Kondrashov 1988; Cutter and Payseurs (2003) small-scale analysis of ten orthologous loci between C. elegans, C. briggasae, and C. remanei found no significant differences in U, nor the rates of evolution of genes between the three species. Using our whole-genomic data set, we found a significant difference in U between the 2 mating types; however, it was in the opposite direction from theoretical expectations. Our estimates of U in both species (0.01 to 0.5; see supplementary table 3, Supplementary Material online) are between 1 and 2 orders of magnitude lower than those required for the MD hypothesis alone to explain the maintenance of obligate outcrossing in C. remanei. Our observation of a greater U in hermaphrodites also contradicts previous theoretical expectations suggesting that the change to hermaphroditism from dioecy would lead to the evolution of reduced deleterious mutation rates in order to slow the accumulation of deleterious substitutions (Birky 1999; West et al. 1999). However, under the assumption that hermaphrodism is of recent origin in C. briggsae, such a reduction in mutation rate could potentially occur in the future.
It should also be noted the C. remanei genome appears to be 15–30% larger than that of C. briggsae (Stein et al. 2003) (http://genome.wustl.edu/pub/organism/Invertebrates/Caenorhabditis_remanei/assembly/Caenorhabditis_remanei-15.0.1/ASSEMBLY), which could imply that it also may have more genes than the latter. Although this remains to be determined, a larger amount of genes in one species would affect estimates of U and could lead to opposite results (i.e., C. remanei having a statistically significantly higher U than C. briggsae; supplementary table 3 [Supplementary Material online]). In terms of magnitude, our estimates of U agree with those of study of Cutter and Payseur (2003) as well as that of Baer et al. (2006); however, they are up to 50 times lower than those computed by Denver et al. (2000) in C. elegans from direct sequencing of mutation accumulation lines. A variety of possibilities could explain such a discrepancy, including variation due to parameter estimates (i.e., divergence time and number of generations per year) as well as error associated with using dS as a measure of the neutral mutation rate (Halligan et al. 2004). It should also be noted that, as previously indicated, estimates of dS between C. remanei and C. briggsae were saturated and therefore could also lead to inaccurate estimates of U. However, reducing the data set such that we calculated U only for genes with dS lower than 3 (or lower than 1) produced very similar results, suggesting that the saturation of dS at some loci is not significantly biasing our estimates of U (supplementary table 3, Supplementary Material online). Regardless, after taking into account these considerations U remains under the threshold of 1.4 under a variety of biologically realistic parameters.
Sexual reproduction varies widely in form and expression from single-celled eukaryotes to sexually dimorphic organisms with exaggerated differences in sexual traits between closely related species. In the past, the leading population genetics theories as to the benefits of sex have emphasized recombination and mutation purging and thus could apply to all sexual organisms—regardless of mating type or system (Bell 1982; Kondrashov 1988). Varying degrees of sexual dimorphism apply to almost all multicellular animals. Its intricate association with the process of sexual reproduction itself would suggest a causal association with the maintenance of sex; one that has recently been made explicit in theoretical form (Agrawal 2001; Siller 2001). Sexual selection has the potential to drive rapid evolution, not only via increased opportunity for mutation purging due to increased variance in male mating success but also by direct effect of positive selection resulting selection on traits that are beneficial to the reproductive success of individual sexes. The aggregated effect of all these beneficial properties derived from the opportunity for sexual selection under the common situation of sexual dimorphism may contribute to paying the "2-fold" cost of sex, which theories focusing on recombination and mutational purging seem to be unable to do alone (Crow and Kimura 1970; Cutter and Payseur 2003).
In conclusion, we find evidence for stronger sexual selection pressure in the dioecious C. remanei as compared with the androdioecious C. briggsae as evidenced by its increased rate of spermatogenesis-related gene evolution relative to other gene categories. We have also used our whole-genome data set in order to compare estimates of the deleterious mutation rate per generation among these 2 mating types and find that our estimates are on the order of 2 orders of magnitude below what would be required for the MD hypothesis to explain the maintenance of dioecy alone. Despite these considerations, environmental constraints, parasitism, and sexual selection have also been suggested to explain the maintenance of dioecy against invasion of hermaphrodites and/or asexuals (Bell 1982; Agrawal 2001; Siller 2001). Our study provides evidence that efficient sexual selection and dioecy are tightly linked within nematodes. Comparisons of different mating systems using a greater number of organisms as well as representative taxa will be required to determine whether this is a general phenomenon among sexually reproducing metazoans. If sexual selection is indeed important in the maintenance of obligate outcrossing in Caenorhabditis, we are left with an intriguing paradox: does loss of efficient sexual selection precede the evolution of hermaphrodism or does evolution of hermaphrodism precede relaxation of sexual selection? Further experimental data and theoretical modeling will be required to answer such questions.
|
Supplementary Material |
|---|
|
TOPAbstractIntroductionMaterials and MethodsResultsDiscussionSupplementary MaterialAcknowledgementsReferences |
|---|
Supplementary figure 1 and tables 1–3 are available at Molecular Biology and Evolution online (http://www.mbe.oxfordjournals.org/).
|
Acknowledgements |
|---|
|
TOPAbstractIntroductionMaterials and MethodsResultsDiscussionSupplementary MaterialAcknowledgementsReferences |
|---|
We are grateful to Aneil Agrawal, Asher Cutter, Ben Evans, and 2 anonymous referees for the comments they raised on the early version of the manuscript. This work was funded by a National Sciences and Engineering Research Council of Canada (NSERC) Post-Graduate Doctoral Scholarship to C.G.A. and NSERC grants to R.S.S. and B.P.G.
|
Footnotes |
|---|
* Both of these labs contributed to this publication.
|
References |
|---|
|
TOPAbstractIntroductionMaterials and MethodsResultsDiscussionSupplementary MaterialAcknowledgementsReferences |
|---|
Agrawal AF. Sexual selection and the maintenance of sexual reproduction. Nature (2001) 411:692–695.[CrossRef][Medline]
Baer CF, Phillips N, Ostrow D, Avalos A, Blanton D, Boggs A, Keller T, Levy L, Mezerhane E. Cumulative effects of spontaneous mutations for fitness in Caenorhabditis: role of genotype, environment and stress. Genetics (2006) 174:1387–1395.
Baer CF, Shaw F, Steding C, Baumgartner M, Hawkins A, Houpper A, Mason N, Reed M, Simonelic K, Woodard W, Lynch M. Comparative evolutionary genetics of spontaneous mutations affecting fitness in rhabditid nematodes. Proc Natl Acad Sci USA (2005) 102:5785–5790.
Bell G. The masterpiece of nature: the evolution and genetics of sexuality (1982) Berkeley (CA): University of California Press. 635.
Betrán E, Thornton K, Long M. Retroposed new genes out of the X in Drosophila. Genome Res (2002) 12:1854–1859.
Bieri T, Blasiar D, Ozersky P, et al, (31 co-authors). WormBase: new content and better access. Nucleic Acids Res (2007) 35:D506–D510.
Birky W. An even broader perspective on sex and recombination. J Evol Biol (1999) 12:1013–1016.[CrossRef][Web of Science]
Bjedov I, Tenaillon O, Gérard B, Souza V, Denamur E, Radman M, Taddei F, Matic I. Stress-induced mutagenesis in bacteria. Science (2003) 300:1404–1409.
Charlesworth B. Mutation-selection balance and the evolutionary advantage of sex and recombination. Genet Res (1990) 55:199–221.[Web of Science][Medline]
Charlesworth B, Coyne JA, Barton N. The relative rates of evolution of sex chromosomes and autosomes. Am Nat (1987) 130:113–146.[CrossRef][Web of Science]
Charlesworth D, Wright SI. Breeding systems and genome evolution. Curr Opin Genet Dev (2001) 11:685–690.[CrossRef][Web of Science][Medline]
Chasnov JR, Chow KL. The species, sex and stage specificity of a Caenorhabditis sex pheromone. Genetics (2002) 160:983–994.
Chasnov JR, So WK, Chan CM, Chow KL. The species, sex, and stage specificity of a Caenorhabditis sex pheromone. Proc Natl Acad Sci USA (2007) 104:6730–6735.
Civetta A, Singh RS. Sex-related genes, directional sexual selection, and speciation. Mol Biol Evol (1998) 15:901–909.[Abstract]
Counterman BA, Ortíz-Barrientos D, Noor MA. Using comparative genomic data to test for fast-X evolution. Evol Int J Org Evol (2004) 58:656–660.
Crow JF, Kimura M. An introduction to population genetics theory (1970) New York: Harper and Row. 591.
Cutter AD. Mutation and the experimental evolution of outcrossing in Caenorhabditis elegans. J Evol Biol (2005) 18:27–34.[CrossRef][Web of Science][Medline]
Cutter AD, Félix MA, Barrière A, Charlesworth D. Patterns of nucleotide polymorphism distinguish temperate and tropical wild isolates of Caenorhabditis briggsae. Genetics (2006) 173:2021–2031.
Cutter AD, Payseur BA. Rates of deleterious mutation and the evolution of sex in Caenorhabditis. J Evol Biol (2003) 16:812–822.[CrossRef][Web of Science][Medline]
Cutter AD, Ward S. Sexual and temporal dynamics of molecular evolution in C. elegans development. Mol Biol Evol (2005) 22:178–188.
Darwin C. The descent of man and selection in relation to sex (1871) London: John Murray. 672.
Denver DR, Morris K, Lynch M, Vassilieva LL, Thomas WK. High direct estimate of the mutation rate in the mitochondrial genome of Caenorhabditis elegans. Science (2000) 289:2342–2344.
Drake JW, Charlesworth B, Charlesworth D, Crow JF. Rates of spontaneous mutation. Genetics (1998) 148:1667–1686.
Eberhardt WG. Sexual selection and animal genitalia (1985) Harvard (MA): Harvard University Press. 244.
Emerson JJ, Kaessmann H, Betrán E, Long M. Extensive gene traffic on the mammalian X chromosome. Science (2004) 303:537–540.
Eyre-Walker A, Keightley PD. High genomic deleterious mutation rates in hominids. Nature (1999) 397:344–347.[CrossRef][Medline]
Glémin S, Bazin E, Charlesworth D. Impact of mating systems on patterns of sequence polymorphism in flowering plants. Proc Biol Sci (2006) 273:3011–3019.
Greeff JM, Michiels NK. Low potential for sexual selection in simultaneously hermaphroditic animals. Proc R Soc Lond B Biol Sci (1999) 266:1671–1676.[CrossRef]
Halligan DL, Eyre-Walker A, Andolfatto P, Keightley PD. Patterns of evolutionary constraints in intronic and intergenic DNA of Drosophila. Genome Res (2004) 14:273–279.
Hill WG, Robertson A. The effect of linkage on limits to artificial selection. Genet Res (1966) 8:269–294.[Web of Science][Medline]
Jagadeeshan S, Singh RS. Rapidly evolving genes of Drosophila: differing levels of selective pressure in testis, ovary and head tissues between sibling species. Mol Biol Evol (2005) 22:1793–1801.
Kimura M. Evolutionary rate at the molecular level. Nature (1968) 217:624–626.[CrossRef][Medline]
Kiontke K, Sudhaus W. Ecology of Caenor habditis species [Internet]. WormBook, ed. In: The C. elegans Research Community WormBook (2006) [cited 2006 January 09]. Available from: http://www.wormbook.org.
Kondrashov AS. Deleterious mutations and the evolution of sexual reproduction. Nature (1988) 336:435–440.[CrossRef][Medline]
Kulathinal RJ, Singh RS. Male sex-drive and masculinization of the genome. Genetica (2005) 120:245–252.[Web of Science]
Li W, Godzik A. Cd-hit: a fast program for clustering and comparing large sets of protein or nucleotide sequences. Bioinformatics (2006) 22:1658–1659.
Lloyd DG. Benefits and handicaps of sexual reproduction. Evol Biol (1980) 13:69–111.
Mank JE, Axelsson E, Ellegren H. Fast-X on the Z: rapid evolution of sex-linked genes in birds. Genome Res (2007) 17:618–624.
Manly BFJ. Randomization, bootstrap and Monte Carlo methods in biology (1991) New York: Chapman & Hall. 281.
Miller MA, Cutter AD, Yamamoto I, Ward S, Greenstein D. Clustered organization of reproductive genes in the C. elegans genome. Curr Biol (2004) 14:1284–1290.[CrossRef][Web of Science][Medline]
Morgenstern B. DIALIGN 2: improvement of the segment-to-segment approach to multiple sequence alignment. Bioinformatics (1999) 15:211–218.
Musters H, Huntley MA, Singh RS. A genomic comparison of faster-sex, faster-X, and faster-male evolution between Drosophila melanogaster and Drosophila pseudoobscura. J Mol Evol (2006) 62:693–700.[CrossRef][Web of Science][Medline]
Ohta T. Slightly deleterious mutant substitutions in evolution. Nature (1973) 246:96–98.[CrossRef][Medline]
Ostrow D, Phillips N, Avalos A, Blanton D, Boggs A, Keller T, Levy L, Rosenbloom J, Bacr CF. Mutational bias for body size in rhabditid nematodes. Genetics (2007) 176:1653–1661.
R Development Core Team. R: a language and environment for statistical computing. (2004) Vienna (Austria): R Foundation for Statistical Computing.
Reinke V, Gil IS, Ward S, Kazmer K. Genome-wide germline-enriched and sex-biased expression profiles in Caenorhabditis elegans. Development (2004) 131:311–323.
Remm M, Storm CE, Sonnhammer EL. Automatic clustering of orthologs and in-paralogs from pairwise species comparisons. J Mol Biol (2002) 314:1041–1052.[CrossRef][Web of Science]
Siller S. Sexual selection and the maintenance of sex. Nature (2001) 411:689–692.[CrossRef][Medline]
Singh RS, Kulathinal RJ. Male sex-drive and the masculination of the genome. Bioessays (2005) 27:518–525.[CrossRef][Web of Science][Medline]
Stein LD, Bao Z, Blasiar D, et al, (33 co-authors). The genome sequence of Caenorhabditis briggsae: a platform for comparative genomics. PLoS Biol (2003) 1:E45.[Medline]
Stenico M, Lloyd AT, Sharp PM. Codon usage in Caenorhabditis elegans: delineation of translational selection and mutational biases. Nucleic Acids Res (1994) 22:2437–2446.
Stevison LS, Counterman BA, Noor MA. Molecular evolution of X-linked accessory gland proteins in Drosophila pseudoobscura. J Hered (2004) 95:114–118.
Sutherland WJ. Measures of sexual selection. Oxf Surv Evol Biol (1985) 2:90–101.
Sutherland WJ. Random and deterministic components of variance in mating success. In: Sexual selection: testing the alternatives?—Bradbury JW, Andersson MB, eds. (1987) New York: Wiley. 90–101.
Swanson WJ, Vacquier WD. Rapid evolution of reproductive proteins. Nat Rev Genet (2002) 3:137–144.[Web of Science][Medline]
Takebayashi N, Morrell PL. Is self-fertilization an evolutionary dead end? Revisiting an old hypothesis with genetic theories and a macroevolutionary approach. Am J Bot (2001) 88:1143–1150.
Torgerson DG, Singh RS. Sex-linked mammalian sperm proteins evolve faster than autosomal ones. Mol Biol Evol (2003) 20:1705–1709.
Torgerson DG, Singh RS. Enhanced adaptive evolution of sperm-expressed genes on the mammalian X chromosome. Heredity (2006) 96:39–44.[Web of Science][Medline]
Ward S, Carrel JS. Fertilization and sperm competition in the nematode C. elegans. Dev Biol (1979) 73:304–321.[CrossRef][Web of Science][Medline]
Wernersson R, Pedersen AG. RevTrans: multiple alignment of coding DNA from aligned amino acid sequences. Nucleic Acids Res (2003) 31:3537–3539.
West SA, Lively CM, Read AF. A pluralist approach to sex and recombination. J Evol Biol (1999) 12:1003–1012.[CrossRef][Web of Science]
Wright SI, Lauga B, Charlesworth D. Rates and patterns of molecular evolution in inbred and outbred Arabidopsis. Mol Biol Evol (2002) 19:1407–1420.
Yang Z, Nielsen R. Codon-substitution models for detecting molecular adaptation at individual sites along specific lineages. Mol Biol Evol (2002) 19:908–917.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  A. D. Cutter, A. Dey, and R. L. Murray Evolution of the Caenorhabditis elegans Genome Mol. Biol. Evol., June 1, 2009; 26(6): 1199 - 1234. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||