Skip Navigation


Human Molecular Genetics Advance Access originally published online on June 22, 2007
Human Molecular Genetics 2007 16(17):2053-2060; doi:10.1093/hmg/ddm153
This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow Supplementary Data
Right arrow All Versions of this Article:
16/17/2053    most recent
ddm153v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrowRequest Permissions
Google Scholar
Right arrow Articles by Wang, X.
Right arrow Articles by Zhang, J.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Wang, X.
Right arrow Articles by Zhang, J.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?

© The Author 2007. Published by Oxford University Press. All rights reserved. For Permissions, please email: journals.permissions@oxfordjournals.org

Rapid evolution of primate ESX1, an X-linked placenta- and testis-expressed homeobox gene

Xiaoxia Wang and Jianzhi Zhang*

Department of Ecology and Evolutionary Biology, University of Michigan, Ann Arbor, MI, USA

* To whom correspondence should be addressed at: Department of Ecology and Evolutionary Biology, University of Michigan, 1075 Natural Science Building, 830 North University Avenue, Ann Arbor, MI 48109, USA. Tel: +1 7347630527; Fax: +1 7347630544; Email: jianzhi{at}umich.edu

Received April 13, 2007; Revised June 1, 2007; Accepted June 14, 2007


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 RESULTS
 DISCUSSION
 MATERIALS AND METHODS
 SUPPLEMENTARY MATERIAL
 REFERENCES
 
Homeobox genes encode transcription factors that play important roles in various developmental processes and are usually evolutionarily conserved. Here we report a case of rapid evolution of a homeobox gene in humans and non-human primates. ESX1 is an X-linked homeobox gene primarily expressed in the placenta and testis, with physiological functions in placenta/fetus development and spermatogenesis. ESX1 is paternally imprinted in mice, but is not imprinted in humans. We provide evidence for a significantly higher non-synonymous substitution rate than synonymous rate in ESX1 between humans and chimps as well as among a total of 15 primate species. Population genetic data also show signals of recent selective sweeps within humans. Positive selection appears to be concentrated in the C-terminal non-homeodomain region, which has been implicated in regulating human male germ cell division by prohibiting the degradation of cyclins. In contrast, mouse Esx1 has a substantively different C-terminal region subject to strong purifying selection. These and other results suggest that even the fundamental process of spermatogenesis has been targeted by positive selection in primate and human evolution and that mouse may not be a suitable model for studying human reproduction.


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
RESULTS
 DISCUSSION
 MATERIALS AND METHODS
 SUPPLEMENTARY MATERIAL
 REFERENCES
 
Homeobox genes are characterized by the presence of a sequence motif known as the homeobox, which encodes the ~60-amino-acid homeodomain, a helix-turn-helix DNA binding domain (1). In humans, there are about 230 homeobox genes (2), encoding a large family of transcription factors that play key roles in various developmental processes such as body-plan specification, pattern formation and cell-fate determination (1). Due to their functional importance, most homeodomain proteins are evolutionarily highly conserved in sequence (1,3,4). Hence, the identification of non-conserved homeobox genes would be particularly interesting, because such homeobox genes may regulate important developmental processes that vary among relatively closely related species. Three such rapidly evolving homeobox genes are known, from fruit flies (OdsH), rodents (Rhox5) and primates (TGIFLX), respectively. OdsH is an X-linked gene involved in spermatogenesis and it is partly responsible for the hybrid male sterility between Drosophila simulans and D. mauritiana (5). Mouse Rhox5 (also known as Pem) is expressed in both male and female reproductive tissues (6). Targeted disruption of Rhox5 increases male germ cell apoptosis and reduces sperm production, sperm motility and fertility (7). In fact, Rhox5 is just one member of a recently expanded homeobox gene cluster known as the Rhox cluster on the mouse X chromosome (7–10). Several other members of the cluster are also expressed in reproductive tissues (7) and evolve rapidly (9,11). TGIFLX is a retroduplicate formed in the common ancestor of primates and rodents by retroposition of the autosomal gene TGIF2 to the X chromosome, and is specifically expressed in the germ cells of adult testis (12). Interestingly, each of the three cases involves a homeobox gene that is X-linked and testis-expressed. Here we report yet another case of rapid evolution of an X-linked testis-expressed homeobox gene, ESX1.

Human ESX1, also known as ESX1L and ESXR1, is a paired-like homeobox gene located on Xq22.1 (13). ESX1 protein contains two functional domains, the homeodomain and the proline-rich domain (Fig. 1A) (13). Esx1, the mouse ortholog, has an extra domain known as the PN/PF motif, located at the C-terminus (Fig. 1B) (14). In humans, ESX1 is specifically expressed in placenta from 5 weeks of gestation until term (15) and in adult testis (13). A recent study shows decreased ESX1 expression in human pre-term idiopathic fetal growth restriction, a clinically significant pregnancy disorder in which the fetus fails to achieve its full growth potential in utero (16). In mice, Esx1 is also expressed in placenta and testis (17,18). More specifically, during embryogenesis, it is expressed in the extraembryonic tissues, including the endoderm of the visceral yolk sac, the ectoderm of the chorion and subsequently the labyrinthine trophoblast of the chorioallantoic placenta (17). In adults, Esx1 is expressed in male germ cells only, particularly the spermatogonia/preleptotene spermatocytes and round spermatids of spermatogenic stages IV–VII (17,18). These restricted temporal and spatial expression patterns suggest that ESX1/Esx1 is involved in placental development and spermatogenesis. Mouse Esx1 is paternally imprinted in the placenta, with only the maternally derived allele expressed (19). Heterozygous female mice inheriting a null Esx1 allele from their mother are born 20% smaller than normal, suggesting that Esx1 is required for placental development and fetal growth in mice (19). In contrast, biparental expression of ESX1 is found in human placenta (20).


Figure 1
View larger version (8K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Figure 1. Structures of the orthologous (A) human ESX1 and (B) mouse Esx1 genes, adapted from Fohn and Behringer (13) and Li et al. (17). Exons are boxed, with coding regions shown in grey and homeobox shown by hatches. The approximate length of each intron is given in parentheses. Pro-rich and PN/PF motifs are indicated underneath the gene structure.

 
Our preliminary comparison between human ESX1 and mouse Esx1 proteins showed an unexpectedly high level of sequence divergence (34%), suggesting that the gene might be evolving rapidly in primates and/or rodents as a result of positive Darwinian selection (12). Below we first describe the evolutionary pattern of ESX1 in primates and then compare it to the evolutionary pattern in rodents. We show that positive selection has acted on ESX1 within humans, between humans and chimpanzees, and among a large array of primate species, whereas purifying selection has dominated Esx1 evolution in rodents. We discuss these evolutionary patterns in light of the structure and function of the gene.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
DISCUSSION
 MATERIALS AND METHODS
 SUPPLEMENTARY MATERIAL
 REFERENCES
 
Comparison of ESX1 sequences between humans and chimps and within humans
We obtained the ESX1 gene sequence from the chimpanzee genome sequence (http://genome.wustl.edu/) and compared it with the human ESX1 sequence available in GenBank (AY114148). The alignment shows a high level of sequence divergence. Of the aligned 406 amino acid sites, there are 25 amino acid replacements, in addition to two gaps totaling 12 amino acids (Fig. 2). A comparison of synonymous (dS) and non-synonymous (dN) nucleotide distances between gene sequences can inform us about the nature and strength of selection acting on a gene. A higher dN than dS indicates positive selection, whereas a lower dN than dS indicates negative or purifying selection. The vast majority of genes in the human genome are under negative selection, with a genomic average dN/dS ratio of 0.26 (21). In ESX1, however, dN (0.031) is significantly greater than dS (0.009) [P = 0.028; Fisher's exact test (22)]. Because the dS value of ESX1 is not significantly different from the genomic average dS of 0.012 (23), the above observation strongly suggests that positive selection has promoted non-synonymous substitutions in ESX1 during the divergence between humans and chimps.


Figure 2
View larger version (14K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Figure 2. Alignment of human and chimpanzee ESX1 protein sequences. The human sequence is from GenBank (accession number AY114148). The homeodomain is boxed. SNPs detected in humans are shaded. For each non-synonymous SNP, the alternative amino acid is shown above the human sequence. For each synonymous SNP, no alternative amino acid is shown. Triangles indicate indel polymorphisms observed in humans, with deletions shown by triangles pointing upwards and insertions shown by triangles pointing downwards. The width of the triangle shows the size of the indel.‘.’ indicates identity to the human sequence and ‘-’ indicates a gap.

 
To identify the regions where positive selection has been operating, we divided the ESX1 protein sequence into three segments, the N-terminus, homeodomain and C-terminus (Fig. 2). The homeodomain is completely identical in amino acid sequence between human and chimp and thus has not been targeted by positive selection (dN = 0, dS = 0.019, P = 0.29; Fisher's exact test). In the N-terminus, dN (0.024) and dS (0) are not significantly different (P = 0.1) and hence neutrality cannot be rejected. In the C-terminus, however, dN (0.047) is significantly greater than dS (0.012) (P = 0.035). Thus, positive selection has been concentrated in the C-terminus. As aforementioned, the C-terminus is mainly composed of proline-rich repeats. The two alignment gaps between human and chimp also occur in the C-terminus (Fig. 2). Compared to the human sequence, the chimp sequence lost a complete nine-amino-acid repeat and part of another repeat.

To further examine whether the positive selection might have happened in the recent history of human evolution, we sequenced exon 4 of ESX1 in 32 unrelated male humans of diverse geographic origins (4 Pygmy Africans, 6 African Americans, 12 Caucasians, 3 Southeast Asians, 2 Chinese, 2 Pacific Islanders, and 3 Andes Indians). Exon 4 encodes 14 amino acids of the homeodomain, corresponding roughly to the third helix of the homeodomain, and the complete C-terminus of ESX1, the likely target of positive selection (Figs 1 and 2). From the 32 alleles, we observed four insertion/deletion (indel) polymorphisms and nine single nucleotide polymorphisms (SNPs). All of these polymorphisms occur in the C-terminus (non-homeodomain) region and none of them disrupt the open reading frame (Fig. 2). Supplementary Material, Table S1 lists the polymorphisms and their associated allele frequencies. The polymorphic data allow us to compute the level of DNA polymorphism in exon 4. Nucleotide diversity per sequence ({pi}) is 0.897 and Watterson's polymorphism per sequence ({theta}) is 1.27. A comparison between expected and observed distributions of allelic frequencies can tell us whether a genomic region is likely to have been subject to recent selective sweeps, which render {pi} lower than {theta} and high-frequency alleles enriched, generating negative values of Tajima's D (24) and Fay and Wu's H (25). Combining the information from D and H, Zeng and colleagues recently invented a new test known as the DH test of positive selection (26). This test is superior to the individual D and H tests because it is more powerful and is insensitive to common confounding factors such as background selection, population growth and population subdivision (26). We found that the DH test rejects the neutral hypothesis for the exon 4 sequences of 32 humans (P < 0.039). For samples with African, Caucasian and Asian origins, the tail probability of the DH test is 0.067, 0.31 and 0.26, respectively. Thus, selective sweeps might have occurred among Africans. Consistent with this result, the H test also yields a significant result for the African samples (P = 0.044), and this P-value is lower than 128 of the 132 genes that were recently surveyed in Africans (27). In other words, ESX1 is among the bottom 3% of human genes for H value in Africans. These results support the hypothesis of recent selective sweeps at human ESX1 or linked genomic regions. We also sequenced exons 1, 2 and 3 of ESX1 in eight male humans with diverse geographic origins (one Pygmy African, three African Americans, one Caucasian, two Chinese and one Pacific Islander), but observed no polymorphisms.

Positive selection at the C-terminus of ESX1 in many primates
To examine whether ESX1 has also been under positive selection in other primates, we obtained the rhesus monkey ESX1 gene sequence by searching its recently completed draft genome sequence (http://www.ncbi.nlm.nih.gov/). We then sequenced exon 4 of ESX1 in 12 additional primate species, including three hominoids, four Old World monkeys and five New World monkeys (see Materials and Methods). Together with the three known sequences from human, chimp and rhesus, a total of 15 primate sequences of exon 4 were conceptually translated and aligned by Clustal W with manual adjustment. DNA sequences were subsequently aligned by following the protein alignment (Fig. 3A). A gene tree of the 15 sequences was reconstructed using the neighbor-joining method (28). The tree topology is consistent with the known species tree, suggesting that the sequences analyzed are orthologous to each other. We found that the length of exon 4 is highly variable among species. The shortest proline-rich repeat region is found in marmoset and tamarin, whereas the longest is observed in orangutan. The nine-amino-acid repeat unit has variable sequences among the primate species, although proline is always the most frequent amino acid.


Figure 3
View larger version (39K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Figure 3. Protein sequence alignment for exon 4 of ESX1 (Esx1) in (A) 12 primates and (B) 4 Mus species.‘.’ indicates identity to the first sequence in each alignment. ‘-’ indicates an alignment gap and ‘*’ indicates a stop codon. The partial homeodomain region is indicated. ‘OW’, Old World; ‘NW’, New World.

 
To examine the potential action of positive selection in exon 4 of primate ESX1, we computed pairwise dN and dS among the 15 sequences. Excluding alignment gaps, we analyzed a total of 312 nucleotide sites. Higher dN than dS is observed in 79 (75.2%) of 105 pairwise comparisons (Fig. 4A). There is no apparent difference in this pattern among hominoids, Old World monkeys and New World monkeys. When only the (non-homeodomain) C-terminus is analyzed, 86 (81.9%) comparisons showed dN > dS (Fig. 4B). In our dataset, the average dS is 0.12 between hominoids and Old World monkeys, 0.18 between hominoids and New World monkeys and 0.16 between Old World monkeys and New World monkeys. All three numbers are greater than the corresponding values (0.08, 0.12 and 0.15, respectively) previously estimated from multiple different intron and non-coding sequences of the same species pairs (29). Thus, the synonymous substitution rate of ESX1 is not reduced and the overall higher dN over dS suggests positive selection.


Figure 4
View larger version (6K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Figure 4. Pairwise synonymous (dS) and non-synonymous (dN) nucleotide distances for (A) the entire exon 4 of ESX1 among 15 primates, (B) the C-terminal non-homeodomain region of ESX1 among 15 primates and (C) the exon 4 of Esx1 among 4 Mus species.

 
Due to the occurrence of many indels in the proline-rich region of primate ESX1, the sequence alignment may not be reliable. Because closely related species are more likely to share the same repeat sequence, which facilitates alignment, we made separate alignments for hominoids, Old World monkeys and New World monkeys, respectively (Supplementary Material, Figure S1). The dN and dS values were then computed for species pairs within each of the three groups. It happened that each of the three groups has 5 species. Of the 30 pairwise comparisons, 23 (76.7%) show dN > dS. This finding is similar to the above result when all 15 sequences are aligned and analyzed together.

To test positive selection in primate ESX1 more rigorously, we conducted a phylogeny-based analysis (22). We inferred the ancestral sequences for the C-terminus at all interior nodes of the primate tree (Fig. 5) using PAML (30) and then counted the numbers of non-synonymous and synonymous substitutions on each tree branch. We found that the ratio of the total number of non-synonymous substitutions and that of synonymous substitutions over all branches of the tree equals n/s = 139/38 = 3.66, significantly greater than the expected value of N/S = 177/93 = 1.90 (P = 0.002, Fisher's exact test). Here N and S are the numbers of non-synonymous and synonymous sites, respectively, in the C-terminus.


Figure 5
View larger version (10K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Figure 5. Numbers of synonymous (s) and non-synonymous (n) substitutions in the evolution of primate ESX1. Shown on each branch are the n and s values for the C-terminal non-homeodomain region. The numbers of non-synonymous (N) and synonymous (S) sites for the same region are also given.

 
We also conducted a likelihood-based analysis to detect positive selection on individual codons within the C-terminus using PAML. We compared the null model M8a with the alternative model M8. M8a, introduced by Swanson et al. (31), assumes that the dN/dS ratio of individual codons follows a beta distribution between 0 and 1, with an extra class of codons with fixed dN/dS of 1. M8 is identical to M8a except for the presence of an additional class of codons with any dN/dS. M8 is found to fit the data significantly better than M8a ({chi}2 = 49.63, df = 2, P < 10–10). M8 suggests that ~66% of codons in the C-terminus have been subject to positive selection with dN/dS = 4.04. Analysis using another pair of models, M1a and M2a, also supports a large proportion of codons under positive selection ({chi}2 = 49.77, df = 2, P < 10–10). Taken together, various analyses provide strong evidence that positive selection has acted in the C-terminus of primate ESX1 to promote amino acid substitutions. We note that although sequence alignment is not easy for ESX1, alignment errors cannot render dN significantly greater than dS, because even when the alignment is completely random, dN is expected to be equal to dS.

Purifying selection on rodent Esx1
To test whether positive selection on ESX1 extends to non-primate mammals, we turn to rodents. We first obtained the Esx1 gene sequence of Mus musculus from GenBank and determined the sequence of exons 2 and 4 of Esx1 from M. spretus. Coding for only two and 15 amino acids, respectively, exons 1 and 3 are not studied here. Esx1 possesses a unique PF/PN motif at its C-terminus, consisting of proline-phenylalanine (PF) tandem repeats followed by proline-asparagine (PN) tandem repeats (Fig. 1B). An earlier study found that the PF/PN motif can inhibit both nuclear localization and DNA binding activity of the Esx1 protein (14). A comparison of exons 2 and 4 sequences of M. musculus and M. spretus shows strong purifying selection on Esx1, as dN (0.006) is significantly lower than dS (0.032) (P = 0.01, Fisher's exact test). Exon 4 was also sequenced in M. cervicolor and M. cookii (Fig. 3B). Significantly lower dN than dS is observed in all pairwise comparisons among the four Mus species with the exception of the comparison between M. cervicolor and M. cookii, probably owing to the small number of substitutions involved (Fig. 4C). Sequence length variation is observed in the PN/PF motif but not in the proline-rich region. Overall, our results suggest that Esx1 has been subject to purifying selection in the Mus genus of rodents.


   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
 DISCUSSION
MATERIALS AND METHODS
 SUPPLEMENTARY MATERIAL
 REFERENCES
 
In this work, we provide evidence for positive selection acting in the C-terminus region of the homeodomain-containing protein ESX1 during primate evolution as well as in human populations. In adult humans, ESX1 is primarily expressed in testis. A previous study showed that ESX1 is proteolytically processed into a 45-kDa N-terminal fragment (including the homeodomain) and a 20-kDa C-terminal fragment. The C-terminal fragment is found in cytoplasm and can inhibit the degradation of cyclin A and B1, causing cell-cycle arrest in human cells (32). Cyclins are a family of proteins controlling transitions through different phases of the cell cycle. Thus, it has been proposed that the C-terminal fragment of ESX1 plays a role in spermatogenesis, functioning as a checkpoint in male germ cell division (32). In contrast, the N-terminal fragment, including the homeodomain, functions as a transcriptional repressor in nucleus (32,33). Our observations of conserved sequences in the N-terminal fragment but rapid sequence changes in the C-terminal fragment are explainable by the distinct functions of the two regions. The finding of positive selection in the C-terminus of primate ESX1 suggests that even in the recent past of human and primate evolution, spermatogenesis has been subject to adaptive modifications (34). Because different species reach sexual maturity at different age, the optimal time of germ cell division may also vary among species. The observed positive selection on ESX1 may reflect such adaptations in individual species. In general, our finding is consistent with many reports of rapid evolution of proteins involved in animal male reproduction (35). Furthermore, mammalian sperm proteins on the X chromosome have been found to evolve faster than those on autosomes (36). Thus, the rapid evolution of ESX1 is likely related to its role in spermatogenesis as well as its location in the X chromosome.

Interestingly, male mice with null Esx1 are fertile, indicating that Esx1 is not essential for spermatogenesis in mice (19). The observation of purifying selection acting on the C-terminal region of Esx1 in mice may be explained by the fact that the gene function has changed between primates and rodents. It is likely that Esx1 is more important for placenta development rather than spermatogenesis in mice (19). Biochemical studies also showed that the nuclear localization of mouse Esx1 is regulated by the presence of the PF/PN motif (14), which is lacking in primate ESX1, further suggesting functional differences between primate ESX1 and rodent Esx1. To examine the C-terminal sequence of ESX1 in other mammals, we TBLASTN-searched the GenBank with human ESX1 and mouse Esx1 as queries. We found putative orthologous Esx1 genes in rat, dog and cow. In horse, only a partial sequence was identified by WISE2 (http://www.ebi.ac.uk/Wise2/). We did not find Esx1 orthologs in opossum and chicken genome sequences. The estimated dN/dS ratio between mouse and rat in the C-terminal region of Esx1 is significantly lower than 1 (P < 0.01), consistent with our findings in the Mus genus. Rat Esx1 has a similar domain structure as mouse Esx1, with the exception that all of the PF repeats are replaced by PN repeats in the PF/PN motif. In contrast, the C-terminus of cow and horse Esx1 proteins is similar to that of primates, with the proline-rich region but not the PF/PN motif. The putative Esx1 in dog has neither the proline-rich region nor the PF/PN motif at its C-terminus. It seems likely that the PF/PN motif was acquired by Esx1 in rodent evolution.

The different evolutionary patterns of primate ESX1 and rodent Esx1 suggest that the utility of the mouse model for studying human reproduction may be limited. Previous studies also reported several other reproduction-related genes that show substantive human–mouse differences. For example, SED1, a protein involved in sperm-egg binding in mice, has lost an important protein–protein binding domain in ancestral primates, which was accompanied by rapid sequence changes in another domain by positive selection (37). In another example, three human X-linked homeobox genes, PEPP1, PEPP2 and PEPP3, correspond to a cluster to 30 Rhox genes in mouse, due to dramatic expansions of the gene cluster in rodent evolution (9). The mouse Rhox genes are expressed in male and female reproductive tissues and at least one of them (Rhox5) is involved in male reproduction, evident from reduced fertilities of Rhox5-knockout mice (7).

Esx1 is paternally imprinted in mouse placenta and is functionally important to placenta morphogenesis and fetal growth (17,19). In contrast, ESX1 is not imprinted in human placenta (20). Imprinting is an important regulatory pathway involved in the development and function of the placenta in eutherian mammals. The imprinting of Esx1 is consistent with the general phenomenon in mice that the paternally derived X chromosome is preferentially inactivated in placental tissues of female embryos (38,39). Recently, Monk et al. (40) reported that several human orthologs of mouse placenta-imprinted genes are un-imprinted. In addition, an earlier investigation revealed a widespread reduction in the maintenance of imprinting in humans (41). If imprinted genes tend to be involved in intra-genomic conflict and hence evolve rapidly by arms race (42), our observation of rapid evolution of the un-imprinted primate ESX1 but slow evolution of imprinted rodent Esx1 is unexpected. While a change in spermatogenesis function might explain the unexpected evolutionary pattern for ESX1/Esx1, in the future, it would be interesting to test the genomic conflict hypothesis by comparing the evolutionary rates of all mouse imprinted genes with those of their un-imprinted human orthologs.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
 DISCUSSION
 MATERIALS AND METHODS
SUPPLEMENTARY MATERIAL
 REFERENCES
 
DNA samples
One individual from each of 12 primate species, 32 male humans, 1 Mus spretus, 1 Mus cookie and 1 Mus cervicolor were surveyed. The 12 primate species include three hominoids (gorilla Gorilla gorilla, orangutan Pongo pygmaeus and gibbon Hylobates lar), four Old World monkeys (green monkey Cercopithecus aethiops, langur Pygathrix nemaeus, talapoin Miopithecus talapoin and baboon Papio hamadryas) and five New World monkeys (marmoset Callithrix jacchus, tamarin Saguinus oedipus, owl monkey Aotus trivirgatus, squirrel monkey Saimiri sciureus and woolly monkey Lagothrix lagotricha). The animal DNA samples were from Wang and Zhang (12) and Podlaha et al. (43), whereas the human DNA samples were purchased from Coriell (http://ccr.coriell.org/).

Gene amplification and DNA sequencing
The amplified ESX1 regions in different species and the primers used for amplification are described in Supplementary Material, Table S2. Primers were designed according to the published human (NT_011651 [GenBank] ) and mouse (NM_007957 [GenBank] ) sequences. Polymerase chain reactions (PCRs) were performed with MasterTaq or TripleMasterTaq under conditions recommended by the manufacturer (Eppendorf, Hamburg, Germany). Dimethyl sulfoxide (DMSO) was used in PCR amplification and DNA sequencing of exon 4. Amplified exon 4 sequences from 12 primates were cloned into PCR4TOPO vector (Invitrogen) and then sequenced from both directions. Other PCR products were purified and directly sequenced from both directions. The dideoxy chain termination method was used in DNA sequencing by an automated sequencer. Sequencher (GeneCodes) was used to assemble the sequences and identify DNA polymorphisms in humans.

Human population genetic analysis
The DH test was conducted by program DH.jar (26). The population recombination rate used in the test was estimated to be R = 3Nr = 3 x 10 000 x(0.18 x 10–6 x 723) = 4 per sequence per generation. Here N = 10 000 is the effective population size of humans, 0.18 x 10–6 is the pedigree-based recombination rate per generation per nucleotide at the ESX1 locus (44) and 723 is the number of nucleotides of the human ESX1 exon 4 sequence. For samples of African, Caucasian and Asian origins, we used N = 10 000, 4000 and 4000, respectively, as their effective population sizes (45). The chimpanzee ESX1 sequence was used as the outgroup in computing DH except for one site where the gorilla sequence was used as the outgroup because the chimpanzee sequence is different from both human alleles. P-values in the DH test and H test were estimated using 100 000 replications of coalescent simulation.

Evolutionary analysis
The coding sequences of human ESX1 and mouse Esx1 were obtained from GenBank with accession numbers AY114148 and NM_007957, respectively. Clustal W (46) was used to conduct sequence alignment for the primates and the Mus species, respectively. MEGA3 (47) was used for the phylogenetic analysis. Pairwise synonymous (dS) and non-synonymous (dN) distances were calculated using the modified Nei-Gojobori method (48), with estimated transition/transversion ratios. Based on the phylogeny of 15 primates, we inferred ancestral ESX1 sequences at all interior nodes of the tree by using the likelihood method under the M8 model in PAML3.15 (30). The number of synonymous (s) and non-synonymous (n) substitutions on each branch of the tree was then counted. The number of synonymous (S) and non-synonymous (N) sites was also estimated by PAML.


   SUPPLEMENTARY MATERIAL
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
 DISCUSSION
 MATERIALS AND METHODS
 SUPPLEMENTARY MATERIAL
REFERENCES
 
Supplementary Material is available at HMG Online.


   ACKNOWLEDGEMENTS
 
We thank Kai Zeng for assistance with the DH test and Meg Bakewell, Soochin Cho and Ondrej Podlaha for valuable comments. This work was supported by research grants from the University of Michigan and National Institutes of Health to J.Z.

Conflict of Interest statement. None declared.


   FOOTNOTES
 
DNA sequences reported in this paper have been submitted to GenBank (accession nos EF650070–EF650084 and EF695414–EF695445)


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
 DISCUSSION
 MATERIALS AND METHODS
 SUPPLEMENTARY MATERIAL
 REFERENCES
 

  1. Gehring W.J., Affolter M., Burglin T. Homeodomain proteins. Annu. Rev. Biochem. (1994) 63:487–526.[CrossRef][Web of Science][Medline]

  2. Nam J., Nei M. Evolutionary change of the numbers of homeobox genes in bilateral animals. Mol. Biol. Evol. (2005) 22:2386–2394.[Abstract/Free Full Text]

  3. McGinnis W., Hart C.P., Gehring W.J., Ruddle F.H. Molecular cloning and chromosome mapping of a mouse DNA sequence homologous to homeotic genes of Drosophila. Cell (1984) 38:675–680.[CrossRef][Web of Science][Medline]

  4. Zhang J., Nei M. Evolution of Antennapedia-class homeobox genes. Genetics (1996) 142:295–303.[Abstract]

  5. Ting C.T., Tsaur S.C., Wu M.L., Wu C.I. A rapidly evolving homeobox at the site of a hybrid sterility gene. Science (1998) 282:1501–1504.[Abstract/Free Full Text]

  6. Sutton K.A., Wilkinson M.F. Rapid evolution of a homeodomain: evidence for positive selection. J. Mol. Evol. (1997) 45:579–588.[CrossRef][Web of Science][Medline]

  7. Maclean J.A. II, Chen M.A., Wayne C.M., Bruce S.R., Rao M., Meistrich M.L., Macleod C., Wilkinson M.F. Rhox: a new homeobox gene cluster. Cell (2005) 120:369–382.[CrossRef][Web of Science][Medline]

  8. MacLean J.A. II, Lorenzetti D., Hu Z., Salerno W.J., Miller J., Wilkinson M.F. Rhox homeobox gene cluster: recent duplication of three family members. Genesis (2006) 44:122–129.[CrossRef][Web of Science][Medline]

  9. Wang X., Zhang J. Remarkable expansions of an X-linked reproductive homeobox gene cluster in rodent evolution. Genomics (2006) 88:34–43.[CrossRef][Web of Science][Medline]

  10. Morris L., Gordon J., Blackburn C.C. Identification of a tandem duplicated array in the Rhox alpha locus on mouse chromosome X. Mamm. Genome (2006) 17:178–187.[CrossRef][Web of Science][Medline]

  11. Jackson M., Watt A.J., Gautier P., Gilchrist D., Driehaus J., Graham G.J., Keebler J., Prugnolle F., Awadalla P., Forrester L.M. A murine specific expansion of the Rhox cluster involved in embryonic stem cell biology is under natural selection. BMC Genomics (2006) 7:212.[CrossRef][Medline]

  12. Wang X., Zhang J. Rapid evolution of mammalian X-linked testis-expressed homeobox genes. Genetics (2004) 167:879–888.[Abstract/Free Full Text]

  13. Fohn L.E., Behringer R.R. ESX1L, a novel X chromosome-linked human homeobox gene expressed in the placenta and testis. Genomics (2001) 74:105–108.[CrossRef][Web of Science][Medline]

  14. Yan Y.T., Stein S.M., Ding J., Shen M.M., Abate-Shen C. A novel PF/PN motif inhibits nuclear localization and DNA binding activity of the ESX1 homeoprotein. Mol. Cell. Biol. (2000) 20:661–671.[Abstract/Free Full Text]

  15. Figueiredo A.L., Salles M.G., Albano R.M., Porto L.C. Molecular and morphologic analyses of expression of ESX1L in different stages of human placental development. J. Cell. Mol. Med. (2004) 8:545–550.[Web of Science][Medline]

  16. Murthi P., Doherty V.L., Said J.M., Donath S., Brennecke S.P., Kalionis B. Homeobox gene ESX1L expression is decreased in human pre-term idiopathic fetal growth restriction. Mol. Hum. Reprod. (2006) 12:335–340.[Abstract/Free Full Text]

  17. Li Y., Lemaire P., Behringer R.R. Esx1, a novel X chromosome-linked homeobox gene expressed in mouse extraembryonic tissues and male germ cells. Dev. Biol. (1997) 188:85–95.[CrossRef][Web of Science][Medline]

  18. Branford W.W., Zhao G.Q., Valerius M.T., Weinstein M., Birkenmeier E.H., Rowe L.B., Potter S.S. Spx1, a novel X-linked homeobox gene expressed during spermatogenesis. Mech. Dev. (1997) 65:87–98.[CrossRef][Web of Science][Medline]

  19. Li Y., Behringer R.R. Esx1 is an X-chromosome-imprinted regulator of placental development and fetal growth. Nat. Genet. (1998) 20:309–311.[CrossRef][Web of Science][Medline]

  20. Grati F.R., Sirchia S.M., Gentilin B., Rossella F., Ramoscelli L., Antonazzo P., Cavallari U., Bulfamante G., Cetin I., Simoni G., et al. Biparental expression of ESX1L gene in placentas from normal and intrauterine growth-restricted pregnancies. Eur. J. Hum. Genet. (2004) 12:272–278.[CrossRef][Web of Science][Medline]

  21. Bakewell M.A., Shi P., Zhang J. More genes underwent positive selection in chimpanzee evolution than in human evolution. Proc. Natl Acad. Sci. USA (2007) 104:7489–7494.[Abstract/Free Full Text]

  22. Zhang J., Kumar S., Nei M. Small-sample tests of episodic adaptive evolution: a case study of primate lysozymes. Mol. Biol. Evol. (1997) 14:1335–1338.[Web of Science][Medline]

  23. Consortium C.S.a.A. Initial sequence of the chimpanzee genome and comparison with the human genome. Nature (2005) 437:69–87.[CrossRef][Medline]

  24. Tajima F. Statistical method for testing the neutral mutation hypothesis by DNA polymorphism. Genetics (1989) 123:585–595.[Abstract/Free Full Text]

  25. Fay J.C., Wu C.I. Hitchhiking under positive Darwinian selection. Genetics (2000) 155:1405–1413.[Abstract/Free Full Text]

  26. Zeng K., Fu Y.X., Shi S., Wu C.I. Statistical tests for detecting positive selection by utilizing high-frequency variants. Genetics (2006) 174:1431–1439.[Abstract/Free Full Text]

  27. Akey J.M., Eberle M.A., Rieder M.J., Carlson C.S., Shriver M.D., Nickerson D.A., Kruglyak L. Population history and natural selection shape patterns of genetic variation in 132 genes. PLoS Biol. (2004) 2:e286.[CrossRef][Medline]

  28. Saitou N., Nei M. The neighbor-joining method: a new method for reconstructing phylogenetic trees. Mol. Biol. Evol. (1987) 4:406–425.[Abstract]

  29. Li W. Molecular Evolution (1997) Sunderland, Mass: Sinauer.

  30. Yang Z. PAML: a program package for phylogenetic analysis by maximum likelihood. Comput. Appl. Biosci. (1997) 13:555–556.[Free Full Text]

  31. Swanson W.J., Nielsen R., Yang Q. Pervasive adaptive evolution in mammalian fertilization proteins. Mol. Biol. Evol. (2003) 20:18–20.[Abstract/Free Full Text]

  32. Ozawa H., Ashizawa S., Naito M., Yanagihara M., Ohnishi N., Maeda T., Matsuda Y., Jo Y., Higashi H., Kakita A., et al. Paired-like homeodomain protein ESXR1 possesses a cleavable C-terminal region that inhibits cyclin degradation. Oncogene (2004) 23:6590–6602.[CrossRef][Web of Science][Medline]

  33. Yanagihara M., Ishikawa S., Naito M., Nakajima J., Aburatani H., Hatakeyama M. Paired-like homeoprotein ESXR1 acts as a sequence-specific transcriptional repressor of the human K-ras gene. Oncogene (2005) 24:5878–5887.[CrossRef][Web of Science][Medline]

  34. Wyckoff G.J., Wang W., Wu C.I. Rapid evolution of male reproductive genes in the descent of man. Nature (2000) 403:304–309.[CrossRef][Medline]

  35. Swanson W.J., Vacquier V.D. The rapid evolution of reproductive proteins. Nat. Rev. Genet. (2002) 3:137–144.[Web of Science][Medline]

  36. Torgerson D.G., Singh R.S. Sex-linked mammalian sperm proteins evolve faster than autosomal ones. Mol. Biol. Evol. (2003) 20:1705–1709.[Abstract/Free Full Text]

  37. Podlaha O., Webb D.M., Zhang J. Accelerated evolution and loss of a domain of the sperm-egg-binding protein SED1 in ancestral primates. Mol. Biol. Evol. (2006) 23:1828–1831.[Abstract/Free Full Text]

  38. West J.D., Frels W.I., Chapman V.M., Papaioannou V.E. Preferential expression of the maternally derived X chromosome in the mouse yolk sac. Cell (1977) 12:873–882.[CrossRef][Web of Science][Medline]

  39. Wagschal A., Feil R. Genomic imprinting in the placenta. Cytogenet. Genome Res. (2006) 113:90–98.[CrossRef][Web of Science][Medline]

  40. Monk D., Arnaud P., Apostolidou S., Hills F.A., Kelsey G., Stanier P., Feil R., Moore G.E. Limited evolutionary conservation of imprinting in the human placenta. Proc. Natl Acad. Sci. USA (2006) 103:6623–6628.[Abstract/Free Full Text]

  41. Morison I.M., Ramsay J.P., Spencer H.G. A census of mammalian imprinting. Trends Genet. (2005) 21:457–465.[CrossRef][Web of Science][Medline]

  42. Haig D. Genetic conflicts in human pregnancy. Q. Rev. Biol. (1993) 68:495–532.[CrossRef][Medline]

  43. Podlaha O., Webb D.M., Tucker P.K., Zhang J. Positive Selection for indel substitutions in the rodent sperm protein Catsper1. Mol. Biol. Evol. (2005) 22:1845–1852.[Abstract/Free Full Text]

  44. Kong A., Gudbjartsson D.F., Sainz J., Jonsdottir G.M., Gudjonsson S.A., Richardsson B., Sigurdardottir S., Barnard J., Hallbeck B., Masson G., et al. A high-resolution recombination map of the human genome. Nat. Genet. (2002) 31:241–247.[CrossRef][Web of Science][Medline]

  45. Tenesa A., Navarro P., Hayes B.J., Duffy D.L., Clarke G.M., Goddard M.E., Visscher P.M. Recent human effective population size estimated from linkage disequilibrium. Genome Res. (2007) 17:520–526.[Abstract/Free Full Text]

  46. Thompson J.D., Higgins D.G., Gibson T.J. CLUSTAL W: improving the sensitivity of progressive multiple sequence alignment through sequence weighting, position-specific gap penalties and weight matrix choice. Nucleic Acids Res. (1994) 22:4673–4680.[Abstract/Free Full Text]

  47. Kumar S., Tamura K., Nei M. MEGA3: Integrated software for Molecular Evolutionary Genetics Analysis and sequence alignment. Brief Bioinform. (2004) 5:150–163.[Abstract/Free Full Text]

  48. Zhang J., Rosenberg H.F., Nei M. Positive Darwinian selection after gene duplication in primate ribonuclease genes. Proc. Natl Acad. Sci. USA (1998) 95:3708–3713.[Abstract/Free Full Text]


Add to CiteULike CiteULike   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us    What's this?


This article has been cited by other articles:


Home page
 Mol Biol EvolHome page
Z. Wang, L. Yuan, S. J. Rossiter, X. Zuo, B. Ru, H. Zhong, N. Han, G. Jones, P. D. Jepson, and S. Zhang
Adaptive Evolution of 5'HoxD Genes in the Origin and Diversification of the Cetacean Flipper
Mol. Biol. Evol., March 1, 2009; 26(3): 613 - 622.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow Supplementary Data
Right arrow All Versions of this Article:
16/17/2053    most recent
ddm153v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrowRequest Permissions
Google Scholar
Right arrow Articles by Wang, X.
Right arrow Articles by Zhang, J.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Wang, X.
Right arrow Articles by Zhang, J.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?