Adaptive evolution of SCML1in primates, a gene involved in male reproduction
- 4.3k Downloads
Genes involved in male reproduction are often the targets of natural and/or sexual selection. SCML1 is a recently identified X-linked gene with preferential expression in testis. To test whether SCML1 is the target of selection in primates, we sequenced and compared the coding region of SCML1 in major primate lineages, and we observed the signature of positive selection in primates.
We analyzed the molecular evolutionary pattern of SCML1 in diverse primate species, and we observed a strong signature of adaptive evolution which is caused by Darwinian positive selection. When compared with the paralogous genes (SCML2 and SCMH1) of the same family, SCML1 evolved rapidly in primates, which is consistent with the proposed adaptive evolution, suggesting functional modification after gene duplication. Gene expression analysis in rhesus macaques shows that during male sexual maturation, there is a significant expression change in testis, implying that SCML1 likely plays a role in testis development and spermatogenesis. The immunohistochemical data indicates that SCML1 is preferentially expressed in germ stem cells of testis, therefore likely involved in spermatogenesis.
The adaptive evolution of SCML1 in primates provides a new case in understanding the evolutionary process of genes involved in primate male reproduction.
KeywordsSexual Selection Rhesus Macaque Primate Species Adaptive Evolution Amino Acid Site
human –Homo sapiens
chimpanzee –Pan troglodytes
gorilla –Gorilla gorilla
orangutan –Pongo pygmaeus
white-browed gibbon –Bunopithecus hoolock
white-cheeked gibbon –Nomascus leucogenys
rhesus monkeys –Macaca mulatta
Yunnan snub-nosed monkey –Rhinopithecus bieti
common marmoset –Callithrix jacchus.
Proteins involved in sexual reproduction often evolve rapidly due to positive selection [1, 2, 3, 4]. Although the selective forces are unclear, a variety of hypotheses have been proposed including mate choice, intra-sexual competition and sexual conflict, which are different forms of sexual selection. The rapid evolution of these proteins may contribute to several important biological aspects such as reproduction and speciation. It has long been recognized that gene duplication is a major source of genomic novelties. Therefore, the newly duplicated genes involved in reproduction are likely the targets of natural and/or sexual selection.
Using exon trapping, van de Vosse et al  identified a novel gene in human located on Xp22, named as SCM-like-1 (SCML1), which is similar with the Scm gene in Drosophila. In the human genome, SCML1 spans 18 kb and contains 8 exons. Northern blot analysis detected a major SCML1 transcript of approximately 3-kb in all human adult and fetal tissues tested .
SCML1 gene is a polycomb group (PcG) gene. Most of the PcG genes are expressed throughout embryonic, larval and pupal development, and are required continuously to maintain restricted homeotic expression in Drosophila. [6, 7, 8, 9, 10]. Most mammalian PcG genes have Drosophila homologs [11, 12]. Compared to Drosophila, the mammalian PcG genes have acquired novel functions during evolution because PcG knockout mice exhibit numerous phenotypes including hematopoietic defects, neural crest defects, cardiac anomalies, and sex reversal [12, 13]. SCML1 is likely a recently duplicated gene during mammalian evolution due to the absence of orthologs in Drosophila, zebrafish and chicken.
In the SCM family, there are other two genes, SCML2 and SCMH1, which have orthologs in all vertebrate species and are located on chromosome Xp22  and chromosome 1p34  respectively. SCMH1 is a core component of polycomb repressive complex 1 (PRC1) [16, 17, 18] which is involved in the maintenance of repression and can block chromatin remodeling, and it plays an important role in regulation of homeotic genes in embryogenesis. SCML2 is also involved in PRC1's regulation. A recent study showed that SCML2 is over-expressed in acute myeloid leukaemia, suggesting its role in differentiation and cell cycle regulation. As SCML2 and SCMH1 are the ancient copies in the SCM family, they would serve as the ideal reference genes when dissecting the molecular evolution of SCML1 in primates.
Through a genome-wide comparison, we have identified 34 candidate genes including SCML1 that showed rapid nonsynonymous sequence divergence between human and chimpanzee , therefore an implication of adaptive evolution of these genes during primate evolution. To test whether SCML1 is the target of selection in primates, we sequenced and compared the coding region of SCML1 in major primate lineages, and we observed the signature of positive selection.
The major lineages of primates were sampled, including three great ape species (chimpanzee-Pan troglodytes, gorilla-Gorilla gorilla and orangutan-Pongo pygmaeus), two lesser ape species (white-browed gibbon-Bunopithecus hoolock and white-cheeked gibbon-Nomascus leucogenys), two Old World monkey species (rhesus macaque-Macaca mulatta and Yunnan snub-nosed monkey-Rhinopithecus bieti) and one New World monkey species (common marmoset-Callithrix jacchus). The common ancestor of the tested primate species can be traced back to about 45 million years ago . All the DNA samples were from collection in Kunming Cell Bank of CAS and Kunming Blood Center in China.
PCR and sequencing
All the samples were sequenced for the full length coding region of SCML1. Primers for all the primates were designed by aligning the published sequences of human (Esembl ID: ENSG00000047634) and macaque (Ensembl ID: ENSMMUG00000012899, Ensemble genome browser ). The primer sequences are listed [see Additional file 1].
PCRs were performed with rTaq under conditions recommended by the manufacturer (TAKARA Company). Sequencing was performed in both directions with forward and reverse primers using the BigDye terminator sequencing kit on an ABI 3130 automated sequencer. There are 8 exons in SCML1 gene, and the first exon is non-translational, therefore, not sequenced in this study. Overlapping chromatogram files retrieved from the sequencer were analyzed and edited using the SeqMan program in the Lasergene software package (DNASTAR Inc).
The DNA sequences were aligned with the CLUSTALW program implanted in Mega [25, 26] and checked manually. There are several in-dels (do not change the reading frame) in the coding region of common marmoset, and those sites were removed in the sequence analysis. The known phylogeny of primate species was used[23, 27]. The ancestral sequences were inferred by PAML 3.15 . The synonymous (ds) and nonsynonymous (d N ) substitution rates of each branch were calculated with the use of the maximum likelihood method under the free-ratio model .
Test of selection
Positive selection can be inferred from a higher proportion of nonsynonymous than synonymous substitutions per site (d N /d S > 1). To detect specific amino acid sites under positive selection, we applied the site models in the codeml program of the PAML package. Using this set of models, we obtained the log likelihood estimates (L) of a tree topology under models that impose alternative assumptions in terms of rate variation (ω = dN/dS) over different codon sites [29, 30]. The model M0 was used to evaluate the general sequence substitution pattern of SCML1 in primates assuming a constant ω ratio across codon sites. M0 estimates the overall ω for the data. The M1a model estimates single parameter, p0, with ω0 = 0, and the remaining sites with frequency p1 (p1 = 1-p0) assuming ω1 = 1. We first compared model M0 with M1a to determine which model is more realistic for the data and M1a tuned out to be the better one. Then we compared model M2a (selection) and M1a (nearly neutral) to test if invoking of positive selection in model M2a would better explain the data [31, 32]. It was suggested that under certain scenarios, a beta distribution of ω is more realistic, therefore, we also conducted the selection test by comparing model M7 and M8, in which a beta distribution of ωwas assumed. We also conducted a more stringent test by comparing M8 with M8a. The LRTs between nested models were conducted by comparing twice the difference of the log-likelihood values (2ΔL) between two models . If the log likelihood test suggests the presence of sites under positive selection, we then identified these sites by using a Bayesian method to estimate posterior probabilities (P) .
Comparative evolutionary analysis among SCML1, SCML2 and SCMH1
Sequences of SCML2 and SCMH1 genes were obtained using BLAST (GenBank and Ensembl) for five primate species including human, chimpanzee, orangutan, rhesus macaque and common marmoset. The sequence IDs are: SCML2, Homo sapiens (ENST00000398048), Macaca mulatta (ENSMMUG00000005084), Pan troglodytess (ENSPTRG00000021710); SCMH1, Homo sapiens (ENST00000326197), Pan troglodytess (ENSPTRG00000000601), Macaca mulatta (ENSMMUG00000017104). With the use of human SCML2 and SCMH1, we searched the genomes of orangutan and common marmoset with Blastn and obtained the coding sequences of these two genes.
Protein sequences were aligned with the CLUSTALW program implanted in Mega4  and the ω calculation was conducted using the codeml program of PAML3.15 . The ratios of d N and d S were estimated by using PAML3.15, and the Z test(data not show) was used to evaluate the ratio difference between each branches . Similar neutrality tests described above were used in comparing the evolutionary patterns among the three genes.
RT-PCR analysis RNAs were extracted using the Tri-Reagent kit based on the manufacturer's specifications (Invitrogen Inc.). For gene expression analysis of rhesus macaques during development, a total of 20 testis samples were analyzed including ten 1–2 year old monkeys (sexually immature) and ten 7–8 years old monkeys (sexually matured). The T test was used for statistical evaluation of expression difference.
For tissue expression analysis in rhesus macaques, a total of 12 tissue types (1–2 year old male macaques) were analyzed including brain, cartilage, heart, large intestine, small intestine, liver, lung, muscle, pancreas, spleen, stomach and testis. All the rhesus macaque tissue samples were collected from the Kunming Primate Research Center, Chinese Academy of Sciences.
For real-time quantitative RT-PCR analysis, cDNAs were synthesized with SuperScript™ III (Invitrogen) from 5 μg of total RNA in a total volume of 20 μl with oligo(dT) primer in accordance with the manufacturer's instructions. SYRB Green I-based real-time PCR was carried out using the DNA Engine Opticon® 2 Continuous Fluorescence Detection System (MJ, BioRad). After an initial denature step for 5 min at 94°C, conditions for cycling are 40 cycles of 20 sec at 94°C, 20 sec at 58°C, 20 sec at 72°C. At the end of the PCR cycles, a melting curve was generated to identify specificity of the PCR product. For each run, serial dilutions of rhesus macaque GAPDH (glyceraldehyde-3-phosphate dehydrogenase) plasmids were used as standards for quantitative measurement of the amount of amplified DNA. In addition, for normalization of each sample, mGAPDH primers were used to measure the amount of GAPDH cDNA. All samples were run in triplicates and the data were presented as a ratio of SCML1/GAPDH. The ΔCt values were calculated and then converted into the linear-scale expression levels. Oligonucleotides were obtained from Invitrogen. Negative controls were performed with water as template. The primer sequences are:
GAPDH F primer 5'ACTTCAACAGCGACACCCACTC3'
GAPDH R primer 5'CCCTGTTGCTGTAGCCAAATTC3'
SCML1 F primer 5'CTCCTACCCTGAAAGTTATAGCC3'
SCML1 R primer 5'TCTGAGGGATGCACTGGAC3'
The liquid nitrogen stored tissue was sectioned (10 μm) using a HM550 tissue processor (Microm). The frozen section slides were stored at -80°C in a sealed slide box. Sections were stained using the standard immunohistochemical method. The mouse monoclonal antibodies generated using human SCML1 protein (dilution 1:100, Abnova) and the goat anti-mouse IgG antibody (dilution 1:200, Bethyl) were used following the manufacturer's instruction. The negative control used is the buffer-only samples with no mouse antibodies. Immuno-reactivity was visualized by using 0.025% 3.3'-diaminobenzidine tetrachloride/0.001% H2O2. These slides were washed with phosphate-buffered saline (pH 7.4). The sections were counterstained with hematoxylin for a few seconds.
Sequence substitution pattern of SCML1 in primates
Test of selection on SCML1 in primates
Neutrality tests of SCML1 in primates using maximum likelihood estimates (site-model)
Estimates of parameters
Positively selected sites
ω = 1.169
p0 = 0.278 p1 = 0.722
p0 = 0.217 p1 = 0.660
p2 = 0.123 ω2 = 5.25
23N 153L 201T (95 =< P < 99%) 92H 242G (P > 99%)
p = 1.096 q = 0.005
p0 = 0.892 p = 0.020
q = 0.005(p1 = 0.108)
ω = 5.78
3N 95S 153L 201T 270L (P > 95%) 92H 242G (P > 99%)
p0 = 0.278 p = 0.005
q = 1.728 (p1 = 0.722)
ω = 1.0
We next compared M1a and M2a (selection model), and M2a fits the data significantly better than M1a (2ΔL = 26.52 P < 0.0001), a strong signature of positive selection on SCML1 in primates. M2a suggests that 12.3% of the sites are under positive selection with ω2 = 5.25. In addition, to avoid the potential bias caused by the assumed substitution pattern in M1a and M2a, we also conducted the selection test by comparing M8 (selection model) and M7 (neutral model), in which a beta distribution for ω over sites was assumed. M8 provides significantly better fit to the data than M7(2ΔL = 33.9, P < 0.0001), again suggesting positive selection on SCML1 in primates. M8 suggests that 10.8% of the sites were under positive selection with ω = 5.78. Interestingly, M8 demonstrates a U-shaped distribution of beta values, suggesting that most sites are either highly conserved with dN/dS close to 0 or nearly neutral with dN/dS = 1, and only a small percentage of the sites were under positive selection. A more stringent test comparing M8 and M8a also supports the proposed positive selection (2ΔL = 27.3, P < 0.0001). The positively selected sites are shown in Table 1 and Figure 1[33, 38]. Collectively, all the tests on selection can be better explained by the evolutionary model that invokes positive selection in primates.
Evolutionary pattern comparison between SCML1, SCML2 and SCMH1
Expression analysis of SCML1
We demonstrate that SCML1 evolves rapidly in primates, which was caused by Darwinian positive selection. Genes expressed exclusively or preferentially in testis are likely involved in male reproduction and have been shown to evolve rapidly under positive selection in previous studies ([1, 2, 3, 49, 50, 51, 52, 53, 54, 55]. Our observation of rapid evolution in SCML1 provides another example of male reproductive gene under Darwinian positive selection in primates.
Darwinian positive selection may lead to functional changes of the target genes during evolution. The SAM domain located in the C terminal (Figure 1) is the only known functional domain of SCML1. Among primates, the amino acid sequences of the SAM domains are relatively conserved across species and there is one site under positive selection (Figure 1). The SAM domain is known to exhibit diverse protein-protein interaction modes, and is involved in developmental regulation . Through functional domain prediction [56, 57, 58, 59, 60, 61, 62, 63], besides of the SAM domain, we identified a total of six fragments[64, 65] (amino acid position 1–9, 13–23, 72–79, 88–114, 125–157 and 224–239) in SCML1 containing potential functional domains. For example, the fragment 1–9 is a signal peptide. The positively selected sites using 95% cutoff are listed in Table 1, and most of them are also located in the potential functional domains other than the structural domains (4/1 and 5/2 for model M2a and model M8 respectively). This distribution bias of the positively selected sites indicates that Darwinian positive selection on SCML1 targets the putative functional domains, which is consistent with the proposed functional modification of SCML1 during primate evolution.
Sexual selection is the favored explanation for the observed adaptive evolution of male reproductive genes [35, 66]. The immunohistochemical and RT-PCR data suggests that SCML1 is important for the development and normal function of testis in primates. Therefore, it is reasonable to propose that the adaptive evolution of SCML1 in primates is likely due to sexual selection. Sperm competition, one of the major mechanisms for sexual selection has been used to define the driving force of selection in promiscuous species, e.g. chimpanzee and human, which seems to explain the observed adaptive evolution of SCML1 since both chimpanzee and human are among the rapidly evolving lineages (ω > 1, Figure 2A). However, gibbon is a monogamous species with a high ω value, and the highly promiscuous rhesus monkey does not show accelerated evolution (ω = 0.09). Therefore, the branch-specific rapid evolution of SCML1 in primates does not provide consistent support for the sexual selection hypothesis. Other evolutionary mechanisms need to be tested, e.g. speciation [67, 68, 69, 70].
The origin of SCML1 probably occurred at the early stage of mammalian radiation about 100 million years ago because we do not identify SCML1 in non-mammalian species, neither in mouse and rat, but in dog, cow and primates. The comparison of evolutionary and expression patterns among the three genes of the same SCML family suggests that the rapid evolution of SCML1 likely led to function modification of testis development and spermatogenesis in primates [71, 72, 73].
The adaptive evolution of SCML1 in primates provides a new case in understanding the evolutionary process of genes involved in primate male reproduction.
We thank Drs. Qi XB, Wang YQ and Han L for their help in data analysis. We are also thankful to the technical help of Zhang H and Yu YC. This study was supported by grants from the National 973 project of China (2007CB947701and 2007CB815705), the Chinese Academy of Sciences (KSCX1-YW-R-34), the National Natural Science Foundation of China (30525028, 30630013 and 30623007), and the Natural Science Foundation of Yunnan Province of China.
- 5.Vosse van de E, Walpole SM, Nicolaou A, Bent van der P, Cahn A, Vaudin M, Ross MT, Durham J, Pavitt R, Wilkinson J, et al: Characterization of SCML1, a new gene in Xp22, with homology to developmental polycomb genes. Genomics. 1998, 49 (1): 96-102. 10.1006/geno.1998.5224.CrossRefPubMedGoogle Scholar
- 13.Takihara Y, Tomotsune D, Shirai M, Katoh-Fukui Y, Nishii K, Motaleb MA, Nomura M, Tsuchiya R, Fujita Y, Shibata Y, et al: Targeted disruption of the mouse homologue of the Drosophila polyhomeotic gene leads to altered anteroposterior patterning and neural crest defects. Development. 1997, 124 (19): 3673-3682.PubMedGoogle Scholar
- 14.Montini E, Buchner G, Spalluto C, Andolfi G, Caruso A, den Dunnen JT, Trump D, Rocchi M, Ballabio A, Franco B: Identification of SCML2, a second human gene homologous to the Drosophila sex comb on midleg (Scm): A new gene cluster on Xp22. Genomics. 1999, 58 (1): 65-72. 10.1006/geno.1999.5755.CrossRefPubMedGoogle Scholar
- 19.Tomotsune D, Takihara Y, Berger J, Duhl D, Joo S, Kyba M, Shirai M, Ohta H, Matsuda Y, Honda BM, et al: A novel member of murine Polycomb-group proteins, Sex comb on midleg homolog protein, is highly conserved, and interacts with RAE28/mph1 in vitro. Differentiation. 1999, 65 (4): 229-239. 10.1046/j.1432-0436.1999.6540229.x.CrossRefPubMedGoogle Scholar
- 21.Grubach L, Juhl-Christensen C, Rethmeier A, Olesen LH, Aggerholm A, Hokland P, Ostergaard M: Gene expression profiling of Polycomb, Hox and Meis genes in patients with acute myeloid leukaemia. Eur J Haematol. 2008Google Scholar
- 24.The ensembl website. [http://www.ensembl.org]
- 30.Yang Z: Complexity of the simplest phylogenetic estimation problem. Proceedings. 2000, 267 (1439): 109-116.Google Scholar
- 34.The Primates' sequencing database. [http://genome.wustl.edu/genome_group.cgi?GROUP=1]
- 39.GNF SymAtlas database. [http://symatlas.gnf.org/SymAtlas/]
- 56.Schultz J, Milpetz F, Bork P, Ponting CP: SMART, a simple modular architecture research tool: identification of signaling domains. Proceedings of the National Academy of Sciences of the United States of America. 1998, 95 (11): 5857-5864. 10.1073/pnas.95.11.5857.PubMedCentralCrossRefPubMedGoogle Scholar
- 60.Tsutsui K, Asai Y, Fujimoto A, Yamamoto M, Kubo M, Hatta N: A novel p63 sterile alpha motif (SAM) domain mutation in a Japanese patient with ankyloblepharon, ectodermal defects and cleft lip and palate (AEC) syndrome without ankyloblepharon. Br J Dermatol. 2003, 149 (2): 395-399. 10.1046/j.1365-2133.2003.05423.x.CrossRefPubMedGoogle Scholar
- 64.The DisEMBL on line. [http://biodatabase.org/index.php/DisEMBL]
- 65.The Smart site on line. [http://smart.embl-heidelberg.de/]
- 67.Gavrilets S: Waiting time to parapatric speciation. Proceedings. 2000, 267 (1461): 2483-2492.Google Scholar
- 74.The GEO database. [http://www.ncbi.nlm.nih.gov/geo]
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.