Maximum-likelihood approach for gene family evolution under functional divergence

According to the observed alignment pattern (i.e., amino acid configuration), we studied two basic types of functional divergence of a protein family. Type I functional divergence after gene duplication results in altered functional constraints (i.e., different evolutionary rate) between duplicate genes, whereas type II results in no altered functional constraints but radical change in amino acid property between them (e.g., charge, hydrophobicity, etc.). Two statistical approaches, i.e., the subtree likelihood and the whole-tree likelihood, were developed for estimating the coefficients of (type I or type II) functional divergence. Numerical algorithms for obtaining maximum-likelihood estimates are also provided. Moreover, a posterior-based site-specific profile is implemented to predict critical amino acid residues that are responsible for type I and/or type II functional divergence after gene duplication. We compared the current likelihood with a fast method developed previously by examples; both show similar results. For handling altered functional constraints (type I functional divergence) in the large gene family with many member genes (clusters), which appears to be a normal case in postgenomics, the subtree likelihood provides a solution that is computationally feasible and robust against the uncertainty of the phylogeny. The cost of this feasibility is the approximation when frequencies of amino acids are very skewed. The potential bias and correction are discussed.     

Functional divergence in the caspase gene family and altered functional constraints: statistical analysis and prediction.

In this article, we explore the pattern of type I functional divergence (i.e., altered functional constraints or site-specific rate difference) in the caspase gene family that is important for apoptosis (programmed cell death) and cytokine maturation. By taking advantage of substantial experimental data from caspases, the functional/structural basis of our posterior predictions from sequence analysis was extensively studied. Our results are as follows: (1) Phylogenetic analysis shows that the evolution of major caspase-mediated pathways has been facilitated by gene duplications, (2) type I functional divergence (altered functional constraints) is statistically significant between two major subfamilies, CED-3 and ICE, (3) 4 of 21 predicted amino acid residues (for site-specific rate difference between CED-3 and ICE) have been verified by experimental evidence, and (4) we found that some CED-3 caspases may inherit more ancestral functions, whereas other members may employ some recently derived functions. Our approach can be cost effective in functional genomics to make statistically sound predictions from amino acid sequences.     

Nucleotide polymorphism in colicin E2 gene clusters: evidence for nonneutral evolution

To explore the molecular mechanisms behind the diversification of colicin gene clusters, we examined DNA sequence polymorphism for the colicin gene clusters of 14 colicin E2 (ColE2) plasmids obtained from natural isolates of Escherichia coli. Two types of ColE2 plasmids are revealed, with type II gene clusters generated by recombination between type I ColE2 and ColE7 gene clusters. The levels and patterns of DNA polymorphism are different between the two types. Type I polymorphism is distributed evenly along the gene cluster, while type II accumulates polymorphism at an elevated rate in the 5' end of the colicin gene. These differences may be explained by recombinational origins of type II gene clusters. The pattern of divergence between the ColE2 gene cluster and its close relative ColE9 is not correlated with the pattern of polymorphism within ColE2, suggesting that this gene cluster is not evolving in a neutral fashion. A statistical test confirms significant departures from the predictions of neutrality. These data lend further support to the hypothesis that colicin gene clusters may evolve under the influence of nonneutral forces.      

Mathematical modeling for functional divergence after gene duplication.

In this paper, I present a statistical framework for modeling the functional divergence after gene duplication. A rate-component model to describe the rate covariation among homologous genes of a gene family is implemented when a phylogenetic tree is known. The Markov chain model is rigorous but may require a huge amount of computational time when the number of sequences is large. On the other hand, the Poisson-based model is mathematically analytical so that computation is very fast even for a large dataset. Moreover, under the posterior framework, we have developed a site-specific profile for predicting important amino acid residues responsible for these functional differences between member genes of a gene family. Our study may have great potential for functional genomics because it is cost-effective, and these predictions can be further tested by biological experimentation.     

Extensive Functional Diversification of the Populus Glutathione S-Transferase Supergene Family

Identifying how genes and their functions evolve after duplication is central to understanding gene family radiation. In this study, we systematically examined the functional diversification of the glutathione S-transferase (GST) gene family in Populus trichocarpa by integrating phylogeny, expression, substrate specificity, and enzyme kinetic data. GSTs are ubiquitous proteins in plants that play important roles in stress tolerance and detoxification metabolism. Genome annotation identified 81 GST genes in Populus that were divided into eight classes with distinct divergence in their evolutionary rate, gene structure, expression responses to abiotic stressors, and enzymatic properties of encoded proteins. In addition, when all the functional parameters were examined, clear divergence was observed within tandem clusters and between paralogous gene pairs, suggesting that subfunctionalization has taken place among duplicate genes. The two domains of GST proteins appear to have evolved under differential selective pressures. The C-terminal domain seems to have been subject to more relaxed functional constraints or divergent directional selection, which may have allowed rapid changes in substrate specificity, affinity, and activity, while maintaining the primary function of the enzyme. Our findings shed light on mechanisms that facilitate the retention of duplicate genes, which can result in a large gene family with a broad substrate spectrum and a wide range of reactivity toward different substrates.     

A semi-quantitative, synteny-based method to improve functional predictions for hypothetical and poorly annotated bacterial and archaeal genes

During microbial evolution, genome rearrangement increases with increasing sequence divergence. If the relationship between synteny and sequence divergence can be modeled, gene clusters in genomes of distantly related organisms exhibiting anomalous synteny can be identified and used to infer functional conservation. We applied the phylogenetic pairwise comparison method to establish and model a strong correlation between synteny and sequence divergence in all 634 available Archaeal and Bacterial genomes from the NCBI database and four newly assembled genomes of uncultivated Archaea from an acid mine drainage (AMD) community. In parallel, we established and modeled the trend between synteny and functional relatedness in the 118 genomes available in the STRING database. By combining these models, we developed a gene functional annotation method that weights evolutionary distance to estimate the probability of functional associations of syntenous proteins between genome pairs. The method was applied to the hypothetical proteins and poorly annotated genes in newly assembled acid mine drainage Archaeal genomes to add or improve gene annotations. This is the first method to assign possible functions to poorly annotated genes through quantification of the probability of gene functional relationships based on synteny at a significant evolutionary distance, and has the potential for broad application.     

Sequence and structural aspects of the functional diversification of plant alcohol dehydrogenases

The glycolytic proteins in plants are coded by small multigene families, which provide an interesting contrast to the high copy number of gene families studied to date. The alcohol dehydrogenase (Adh) genes encode glycolytic enzymes that have been characterized in some plant families. Although the amino acid sequences of zinc-containing long-chain ADHs are highly conserved, the metabolic function of this enzyme is variable. They also have different patterns of expression and are submitted to differences in nonsynonymous substitution rates between gene copies. It is possible that the Adh copies have been retained as a consequence of adaptative amino acid replacements which have conferred subtle changes in function. Phylogenetic analysis indicates that there have been a number of separate duplication events within angiosperms, and that genes labeled Adh1, Adh2 and Adh3 in different groups may not be homologous. Nonsynonymous/synonymous ratios yielded no signs of positive selection. However, the coefficients of functional divergence (theta) estimated between the Adh1 and Adh2 gene groups indicate statistically significant site-specific shift of evolutionary rates between them, as well as between those of different botanical families, suggesting that altered functional constraints may have taken place at some amino acid residues after their diversification. The theoretical three-dimensional structure of the alcohol dehydrogenase from Arabis blepharophylla was constructed and verified to be stereochemically valid.     

cDNA clones encoding rabbit immunoglobulin lambda chains. Evidence for length variation of the third hypervariable region and for a novel constant region.

Five cDNA clones designated pDH2, pDH8, pDH9, pDH31 and pDH101 encoding rabbit immunoglobulin lambda light chain sequences have been characterized. Comparison of the V lambda sequences suggests that, in addition to an increased divergence in all of the complementarity-determining regions (CDRs), variable-region diversity is amplified by the length heterogeneity of the CDR3, at the V lambda-J lambda junction. An insertion of four codons at positions 48a-d has been noted in three cDNA sequences. This insert, not found in lambda nor kappa light chains of other species, has a variable sequence, suggesting its possible implication in expanding variability of the CDR2. One of the cDNA clones was shown to encode a novel C lambda region which differs by four amino acid substitutions from the C lambda region common to all the other clones. Thus, the rabbit can use two different C lambda genes, which might correlate with the expression of the two known allotypes of lambda chains, C7 and C21. Southern blotting experiments indicate a small number of germ-line V lambda genes and the cDNA nucleotide sequence data reported here suggest that several of these genes can be expressed. The possibility of at least two V-J-C gene clusters is discussed.     

Phylogeography of speciation: allopatric divergence and secondary contact between outcrossing and selfing Clarkia

The origins of hybrid zones between parapatric taxa have been of particular interest for understanding the evolution of reproductive isolation and the geographic context of species divergence. One challenge has been to distinguish between allopatric divergence (followed by secondary contact) versus primary intergradation (parapatric speciation) as alternative divergence histories. Here, we use complementary phylogeographic and population genetic analyses to investigate the recent divergence of two subspecies of Clarkia xantiana and the formation of a hybrid zone within the narrow region of sympatry. We tested alternative phylogeographic models of divergence using approximate Bayesian computation (ABC) and found strong support for a secondary contact model and little support for a model allowing for gene flow throughout the divergence process (i.e. primary intergradation). Two independent methods for inferring the ancestral geography of each subspecies, one based on probabilistic character state reconstructions and the other on palaeo-distribution modelling, also support a model of divergence in allopatry and range expansion leading to secondary contact. The membership of individuals to genetic clusters suggests geographic substructure within each taxon where allopatric and sympatric samples are primarily found in separate clusters. We also observed coincidence and concordance of genetic clines across three types of molecular markers, which suggests that there is a strong barrier to gene flow. Taken together, our results provide evidence for allopatric divergence followed by range expansion leading to secondary contact. The location of refugial populations and the directionality of range expansion are consistent with expectations based on climate change since the last glacial maximum. Our approach also illustrates the utility of combining phylogeographic hypothesis testing with species distribution modelling and fine-scale population genetic analyses for inferring the geography of the divergence process.     

Variation in V lambda genes in the genus Mus

The complement of Ig V lambda genes in nine species of feral mice representing the four extant subgenera of the genus Mus was examined and compared with that of BALB/c inbred mice. Although all inbred strains examined have two V lambda genes, there is variation in the number of copies of V lambda genes in the wild mice. All feral representatives of M. musculus domesticus, from which inbred strains are derived, have at least three V lambda genes, indicating that a V lambda gene may have been lost during the inbreeding process. At least three V lambda genes are also found in representatives of three other M. musculus subspecies, including the stock of M. musculus musculus "Czech II" shown to have at least 12 C lambda genes. In comparing the complement of V lambda and C lambda genes in these animals, evidence is found that supports a mechanism of lambda gene reiteration involving duplication of a unit containing a V lambda and two C lambda genes. However, the possibility that C lambda gene amplification occurred independent of V lambda gene evolution cannot be ruled out. M. spicelegus and M. spretus, species that are semifertile with M. musculus, have one to three V lambda genes. Species more distantly related to M. musculus, such as M. cookii and M. platythrix, appear to have more (four to six) V lambda genes. Greater V lambda gene heterogeneity is also found in these animals. We propose that the ancestors of the subgenus Mus had more V lambda genes than are seen in modern species and that the paucity of V lambda genes in M. musculus, M. spicelegus, and M. spretus may be the result of V lambda gene deletion events that occurred since the divergence of the ancestor of these three species and those of the distantly related species.     

Genomic background drives the divergence of duplicated amylase genes at synonymous sites in Drosophila

In some Drosophila species, there are two types of greatly diverged amylase (Amy) genes (Amy clusters 1 and 2), each encoding active amylase isozymes. Cluster 1 is located at the middle of its chromosomal arm, and the region has a normal local recombination rate. However, cluster 2 is near the centromere, and this region is known to have a reduced recombination rate. Although nonsynonymous substitutions follow a molecular clock, synonymous substitutions were accelerated in cluster 2 after gene duplications. This resulted in a higher GC content at the third codon position (GC3) and codon usage bias in cluster 1, and lower GC3 content and codon usage bias in the cluster 2. However, no systematic difference in GC content was observed in the first and second codon positions or the 3'-flanking regions. Therefore, differences in local recombination rate rather than mutation bias might explain the divergence at synonymous sites between the two Amy clusters within species (Hill-Robertson effect). Alternatively, the different patterns and levels of expression between the two clusters may imply that the reduced expression level in cluster 2 caused by chromatin potentiation decreased the codon bias. Both of these hypotheses imply the importance of the genomic background as a driving force of divergence between non-tandemly duplicated genes.     

Functional divergence after gene duplication and sequence-structure relationship: a case study of G-protein alpha subunits

In this article, we use animal G-protein alpha subunit family as an example to illustrate a comprehensive analytical pipeline for detecting different types of functional divergence of protein families, which is phylogeny-dependent, combined with ancestral sequence inference and available protein structure information. In particular, we focus on (i) Type-I functional divergence, or site-specific rate shift, as typically exemplified by amino acid residue highly conserved in a subset of homologous genes but highly variable in a different subset of homologous genes, and (ii) Type-II functional divergence, or the shift of cluster-specific amino acid property, as exemplified by a radical shift of amino acid property between duplicate genes, which is otherwise evolutionally conserved. We utilized the software DIVERGE2 to carry out these analyses. In the case of G-protein alpha subunit gene family, we have predicted amino acid residues that are related to either Type-I or Type-II functional divergence. The inferred ancestral sequences for these sites are helpful to explore the trends of functional divergence. Finally, these predicted residues are mapped to the protein structures to test whether these residues may have 3D structure or solvent accessibility preference.     

Structure of rat calmodulin processed genes with implications for a mRNA-mediated process of insertion

Two distinct processed calmodulin genes of rat (lambda SC8 and lambda SC9) were identified, cloned and their DNA sequences determined. The existence of direct repeats of 19 base-pairs for lambda SC8 or 9 base-pairs for lambda SC9 at both ends of the coding plus non-coding regions suggested a possible involvement of a mRNA-mediated process of insertion. Total genomic Southern hybridization suggested the existence of at least three different calmodulin-related genes in the rat genome. The other gene was the bona fide calmodulin gene (lambda SC4) which was split into at least five exons. lambda SC9 contained insertions of one nucleotide and two 17 base-pair direct repeats in the coding region. These insertions cause frameshift mutations probably preventing it from encoding a functional calmodulin. It also carried an insertion of a rat middle repetitive sequence, identifier sequence (IDS: Sutcliffe et al., 1982) in the 3'-non-coding region. Otherwise, it consisted of an almost identical DNA sequence to that of the bona fide calmodulin gene (lambda SC4), including the 3'-non-coding region down to the poly(A) recognition signal, A-A-T-A-A-A. On the other hand, lambda SC8 did not possess frameshift mutations in the coding region, and hence was capable of encoding a functional protein. In fact, a probe specific to the lambda SC8 sequence identified a band in Northern blotting whose size was 300 nucleotides smaller than that of authentic calmodulin mRNA. Comparison of the nucleotide sequences showed that only the coding regions of these two processed genes were homologous, indicating that the divergence of these two processed genes from the common ancestor calmodulin was an ancient event.     

Sequence-Dependent Gene Conversion: Can Duplicated Genes Diverge Fast Enough to Escape Conversion?

Conversion between duplicated genes limits their independent evolution. Models in which conversion frequencies decrease as genes diverge are examined to determine conditions under which genes can "escape" further conversion and hence escape from a gene family. A review of results from various recombination systems suggests two classes of sequence-dependence models: (1) the "k-hit" model in which conversion is completely inactivated by a few (k) mutational events, such as the insertion of a mobile element, and (2) more general models where conversion frequency gradually declines as genes diverge through the accumulation of point mutants. Exact analysis of the k-hit model is given and an approximate analysis of a more general sequence-dependent model is developed and verified by computer simulation. If mu is the per nucleotide mutation rate, then neutral duplicated genes diverging through point mutants are likely to escape conversion provided 2 mu/lambda much greater than 0.1, where lambda is the conversion rate between identical genes. If 2 mu/lambda much less than 0.1, the expected number of conversions before escape increases exponentially so that, for biological purposes, the genes never escape conversion. For single mutational events sufficient to block further conversions, occurring at rate nu per copy per generation, many conversions are expected if 2 nu/lambda much less than 1, while the genes essentially evolve independently if 2 nu/lambda much greater than 1. Implications of these results for both models of concerted evolution and the evolution of new gene functions via gene duplication are discussed.     

Genomic background predicts the fate of duplicated genes: evidence from the yeast genome

Gene duplication with subsequent divergence plays a central role in the acquisition of genes with novel function and complexity during the course of evolution. With reduced functional constraints or through positive selection, these duplicated genes may experience accelerated evolution. Under the model of subfunctionalization, loss of subfunctions leads to complementary acceleration at sites with two copies, and the difference in average rate between the sequences may not be obvious. On the other hand, the classical model of neofunctionalization predicts that the evolutionary rate in one of the two duplicates is accelerated. However, the classical model does not tell which of the duplicates experiences the acceleration in evolutionary rate. Here, we present evidence from the Saccharomyces cerevisiae genome that a duplicate located in a genomic region with a low-recombination rate is likely to evolve faster than a duplicate in an area of high recombination. This observation is consistent with population genetics theory that predicts that purifying selection is less effective in genomic regions of low recombination (Hill-Robertson effect). Together with previous studies, our results suggest the genomic background (e.g., local recombination rate) as a potential force to drive the divergence between nontandemly duplicated genes. This implies the importance of structure and complexity of genomes in the diversification of organisms via gene duplications.     

Molecular evidence of a peripatric origin for two sympatric species of field crickets (Gryllus rubens and G. texensis) revealed from coalescent simulations and population genetic tests

Species pairs that differ primarily in characters involved in mating interactions and are largely sympatric raise intriguing questions about the mode of speciation. When species divergence is relatively recent, the footprint of the demographic history during speciation might be preserved and used to reconstruct the biogeography of species divergence. In this study, patterns of genetic variation were examined throughout the geographical range of two cryptic sister taxa of field crickets, Gryllus texensis and G. rubens; mitochondrial cytochrome oxidase I (COI) was sequenced in 365 individuals sampled from 48 localities. Despite significant molecular divergence between the species, they were not reciprocally monophyletic. We devised several analyses to statistically explore what historical processes might have given rise to this genealogical structure. The analyses indicated that the biogeographical pattern of genetic variation does not support a model of recent gene flow between species. Instead, coalescent simulations suggested that the genealogical structure within G. texensis, namely a deep split between two geographically overlapping clades, reflects historical substructure within G. texensis. Additional tests that consider the concentration of G. rubens haplotypes in one of the two G. texensis genetic clusters suggest a model of speciation in which G. rubens was derived from one lineage of a geographically subdivided ancestor. These results indicate that, despite the contemporary sympatry of G. texensis and G. rubens, the data are indicative of an peripatric origin in which G. rubens was derived from one of the two historical partitions in the species currently recognized as G. texensis. This proposed model of species divergence suggests how the interplay of geography and selection may give rise to new species, although this requires testing with multilocus data. Specifically, the model highlights how that geographical partitioning of ancestral variation in the past may augment the selectively driven divergence of characters involved in the reproductive isolation of the species today.     

Molecular evolution and functional divergence of the cytochrome P450 3 (CYP3) Family in Actinopterygii (ray-finned fish)

Background

The cytochrome P450 (CYP) superfamily is a multifunctional hemethiolate enzyme that is widely distributed from Bacteria to Eukarya. The CYP3 family contains mainly the four subfamilies CYP3A, CYP3B, CYP3C and CYP3D in vertebrates; however, only the Actinopterygii (ray-finned fish) have all four subfamilies and detailed understanding of the evolutionary relationship of Actinopterygii CYP3 family members would be valuable.

Methods and Findings

Phylogenetic relationships were constructed to trace the evolutionary history of the Actinopterygii CYP3 family genes. Selection analysis, relative rate tests and functional divergence analysis were combined to interpret the relationship of the site-specific evolution and functional divergence in the Actinopterygii CYP3 family. The results showed that the four CYP3 subfamilies in Actinopterygii might be formed by gene duplication. The first gene duplication event was responsible for divergence of the CYP3B/C clusters from ancient CYP3 before the origin of the Actinopterygii, which corresponded to the fish-specific whole genome duplication (WGD). Tandem repeat duplication in each of the homologue clusters produced stable CYP3B, CYP3C, CYP3A and CYP3D subfamilies. Acceleration of asymmetric evolutionary rates and purifying selection together were the main force for the production of new subfamilies and functional divergence in the new subset after gene duplication, whereas positive selection was detected only in the retained CYP3A subfamily. Furthermore, nearly half of the functional divergence sites appear to be related to substrate recognition, which suggests that site-specific evolution is closely related with functional divergence in the Actinopterygii CYP3 family.

Conclusions

The split of fish-specific CYP3 subfamilies was related to the fish-specific WGD, and site-specific acceleration of asymmetric evolutionary rates and purifying selection was the main force for the origin of the new subfamilies and functional divergence in the new subset after gene duplication. Site-specific evolution in substrate recognition was related to functional divergence in the Actinopterygii CYP3 family.     

The Ghost of Selection Past: Rates of Evolution and Functional Divergence of Anciently Duplicated Genes

The duplication of genes and even complete genomes may be a prerequisite for major evolutionary transitions and the origin of evolutionary novelties. However, the evolutionary mechanisms of gene evolution and the origin of novel gene functions after gene duplication have been a subject of many debates. Recently, we compiled 26 groups of orthologous genes, which included one gene from human, mouse, and chicken, one or two genes from the tetraploid Xenopus and two genes from zebrafish. Comparative analysis and mapping data showed that these pairs of zebrafish genes were probably produced during a fish-specific genome duplication that occurred between 300 and 450 Mya, before the teleost radiation (Taylor et al. 2001). As discussed here, many of these retained duplicated genes code for DNA binding proteins. Different models have been developed to explain the retention of duplicated genes and in particular the subfunctionalization model of Force et al. (1999) could explain why so many developmental control genes have been retained. Other models are harder to reconcile with this particular set of duplicated genes. Most genes seem to have been subjected to strong purifying selection, keeping properties such as charge and polarity the same in both duplicates, although some evidence was found for positive Darwinian selection, in particular for Hox genes. However, since only the cumulative pattern of nucleotide substitutions can be studied, clear indications of positive Darwinian selection or neutrality may be hard to find for such anciently duplicated genes. Nevertheless, an increase in evolutionary rate in about half of the duplicated genes seems to suggest that either positive Darwinian selection has occurred or that functional constraints have been relaxed at one point in time during functional divergence. Received: 4 January 2001 / Accepted: 29 March 2001     

Relationships of the Escherichia coli O157, O111, and O55 O-antigen gene clusters with those of Salmonella enterica and Citrobacter freundii, which express identical O antigens

Escherichia coli O157, Salmonella enterica O30, and Citrobacter freundii F90 have identical O-antigen structures, as do E. coli O55 and S. enterica O50. The O-antigen gene cluster sequences for E. coli O157 and E. coli O55 have been published, and the genes necessary for O-antigen biosynthesis have been identified, although transferase genes for glycosidic linkages are only generic and have not been allocated to specific linkages. We determined sequences for S. enterica O30 and C. freundii F90 O-antigen gene clusters and compared them to the sequence of the previously described E. coli O157 cluster. We also determined the sequence of the S. enterica O50 O-antigen gene cluster and compared it to the sequence of the previously described E. coli O55 cluster. For both the S. enterica O30-C. freundii F90-E. coli O157 group and the S. enterica O50-E. coli O55 group of O antigens, the gene clusters have identical or nearly identical organizations. The two sets of gene clusters had comparable overall levels of similarity in their genes, which were lower than the levels determined for housekeeping genes for these species, which were 55 to 65% for the genes encoding glycosyltransferases and O-antigen processing proteins and 75 to 93% for the nucleotide-sugar pathway genes. Nonetheless, the similarity of the levels of divergence in the five gene clusters required us to consider the possibility that the parent gene cluster for each structure was in the common ancestor of the species and that divergence is faster than expected for the common ancestor hypothesis. We propose that the identical O-antigen gene clusters originated from a common ancestor, and we discuss some possible explanations for the increased rate of divergence that is seen in these genes.