MBE Advance Access originally published online on April 13, 2007
Molecular Biology and Evolution 2007 24(7):1443-1446; doi:10.1093/molbev/msm072
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Letters |
Opposite Evolutionary Effects between Different Alternative Splicing Patterns
* Division of Biostatistics and Bioinformatics, National Health Research Institutes, Miaoli County, Taiwan
Research Center for Biodiversity, Academia Sinica, Taipei, Taiwan
Genomics Research Center, Academia Sinica, Taipei, Taiwan
E-mail: trees{at}gate.sinica.edu.tw.
 |
Abstract |
|---|
 TOP Abstract MethodsAcknowledgementsReferences |
|---|
Alternative splicing (AS) has been recognized as a mechanism of relaxing selection pressure on protein subsequences. Here, we show that AS may also yield contrary evolutionary effects. We compare the evolutionary rates of 2 types of alternatively spliced exons (ASEs)—simple and complex. The former does not change the boundaries of its flanking exons, whereas the latter does. By analyzing over 26,000 human–mouse orthologous exons, we demonstrate that complex ASEs have lower Ka and Ka/Ks ratio and higher Ks than constitutively spliced exons (CSEs), whereas simple ASEs have evolutionary rates to the opposite of CSEs. Our results indicate that complex ASEs are subject to stronger selection pressure than CSEs at the protein level, but the trend is reversed at the RNA level. Therefore, the previous view that ASEs accelerate evolution of protein subsequences needs to be modified.
Key Words: simple alternatively spliced exons • complex alternatively spliced exons • constitutively spliced exons • selective constraint • lineage specificity • comparative genomics
Alternative splicing (AS) is a topic of extensive studies because of its importance in increasing proteome complexity and its role involved in a wide variety of biological processes (Brett et al. 2002
; Bracco and Kearsey 2003). The most common AS event is "cassette exon," in which an individual exon is either included or excluded in a transcript. For simplicity, cassette exons are sometimes referred to as alternatively spliced exons (ASEs) and suggested to have higher nonsynonymous (Ka) but lower synonymous (Ks) substitution rates than constitutively spliced exons (CSEs) (Iida and Akashi 2000; Hurst and Pal 2001; Modrek and Lee 2003; Filip and Mundy 2004; Xing and Lee 2005; Chen and Chuang 2006; Chen et al. 2006). However, whether conservation of AS patterns affects evolutionary rates has not been investigated.
The AS database (ASD) at the European Bioinformatics Institute (EBI-ASD) (Stamm et al. 2006) classifies cassette exons into simple and complex ones. Inclusion of a complex ASE can result in boundary changes of its 1 or 2 flanking exons, which is not observed for simple ASEs. Because the splicing processes of simple and complex ASEs may be quite different, we ask whether the 2 types of ASEs differ in evolutionary rates. In this study, we retrieved more than 26,000 human–mouse orthologous exon pairs from the EBI-ASD and analyzed their Ka and Ks values, which, respectively, give empirical measures of selection pressures against amino acid changes and synonymous nucleotide substitutions (Li 1997; Yang and Nielsen 2000). We demonstrate that simple and complex ASEs differ significantly in evolutionary features and propose that AS may either increase or decrease the evolutionary rates of protein subsequences.
As illustrated in table 1, the human–mouse orthologous exon pairs are divided into 4 groups for calculation of evolutionary rates. The ASE–ASE and CSE–CSE groups are referred to as "conserved ASEs" and "conserved CSEs," respectively, and the ASE–ASE, ASE–CSE, and CSE–ASE groups are collectively called "AS-affected groups." We find that conserved ASEs have the smallest median Ks value, the largest median Ka/Ks ratio, and the highest proportion of exons that fail the Ka/Ks ratio test (failing-test exons); followed by lineage-specific ASEs (i.e., ASE–CSE and CSE–ASE); and then conserved CSEs (all P < 0.01 by 2-tailed Wilcoxon rank sum test). Notice that conserved ASEs tend to evolve faster than lineage-specific ASEs, although the median Ka values of the 3 AS-affected groups are not significantly different. The similar Ka and Ks rates and Ka/Ks ratios between ASE–CSE and CSE–ASE groups (all P > 0.01) imply that the effects of AS conservation on exon evolution are not significantly affected by the lineage specificity of ASEs. This is surprising because mouse has a faster molecular clock than human (Li 1997; Nekrutenko et al. 2003) and both ASEs and CSEs are expected to evolve faster in mouse because it. Intuitively, we may suppose that when ASEs occur in mouse rather than in human (mouse ASEs vs. human CSEs), the Ka values should be higher than those observed in human (human ASEs vs. mouse CSEs). However, our results indicate that the Ka, Ks, and Ka/Ks ratios are similar between the 2 groups. This observation implies that the effects of AS dominates that of the molecular clock on Ka and Ks values.
|
To examine whether different AS types (i.e., simple/complex ASEs) affect the evolutionary rates of exons, we further divided the human–mouse orthologous exons into 3 groups for calculation of evolutionary rates: Group 1, ASEs versus all exons (including ASEs and CSEs); Group 2, ASEs versus CSEs; and Group 3, ASEs versus ASEs (table 2). For Group 1, simple ASEs have higher median Ka values, higher median Ka/Ks ratios, lower median Ks values, and a larger proportion of failing-test exons than complex ASEs regardless of lineage. The differences are all significant except for those of Ka values (by the Wilcoxon rank sum test, see table 2). Nevertheless, the accumulative distribution of Ka values exhibits significant difference (P < 0.01, by the Kolmogorov–Smirnov test). The overall trends also hold for Groups 2 and 3. Therefore, our results suggest that complex ASEs are under stronger selection pressure at the amino acid level than simple ones in both human and mouse.
|
We then compared the evolutionary rates of lineage-specific ASEs (Group 2) with conserved ASEs (Group 3). Note that simple ASEs in Group 3 have significantly lower Ks values, higher Ka/Ks ratios, and higher proportions of failing-test exons in both human and mouse than those in Group 2 (all P < 0.01). The changes in Ka/Ks ratios are mainly due to the changes in Ks values because Ka values remain approximately equal between the 2 groups. In contrast, for complex ASEs, none of the differences in evolutionary rates between the 2 groups are significant. Moreover, the evolutionary rates and the proportion of failing-test exons of conserved CSEs generally fall between those of simple and complex ASEs (tables 1 and 2). Therefore, complex ASEs appear to be under stronger selection pressure at the amino acid level, but less pressure at the RNA level, than CSEs.
Because the inclusion of CSEs can significantly affect the Ks values of human–mouse orthologous exons, we compare the Ks values between complex and simple ASEs of Group 3. It is found that the median Ks values of complex ASEs are 65–74% larger than those of simple ASEs (tables 1 and 2). Using CSEs as a reference, simple ASEs have higher Ka but lower Ks values, whereas the reverse is true for complex ASEs. Moreover, we performed a Gene Ontology (Gene Ontology Consortium 2001) analysis for both ASE types and found that they differ only in the "Transporter activity" category. Exclusion of transcripts in this category does not change the overall trends. Consequently, our results suggest that simple and complex ASEs have opposite evolutionary effects at both amino acid and RNA levels.
The differences in evolutionary rates between simple and complex ASEs may have resulted from the differences in nucleotide composition or codon usage bias. Table 3 shows that complex ASEs have significantly higher GC content than simple ones (P < 0.01 by Fisher's exact test), which is especially conspicuous at the 4-fold degenerate sites. Nonetheless, the 2 exon types do not significantly differ in their average CpG derivative contents. These data are consistent with the higher Ks values in complex ASEs, suggesting that complex ASEs have a stronger mutability both exon-wide and at the 4-fold degenerate sites, possibly because of higher GC content. Our linear regression analyses also demonstrate that Ks is strongly correlated with GC content (P < 0.01). We then analyzed the substitution rate at non–CpG-prone 4-fold degenerate sites, which represents the mutability at the 4-fold degenerate sites, excluding the CpG effect. For human, complex and simple ASEs have an almost identical non–CpG-prone substitution rates (0.135 vs. 0.137), but for mouse, simple ASEs have a higher non–CpG-prone substitution rate than complex ones (0.131 vs. 0.075). Note that the Ks value differences between Group 1 simple and complex ASEs are significant in both human and mouse (tables 2 and 3). These observations suggest that CpGs have very different effects on the 4-fold synonymous substitution rates between human and mouse. Moreover, complex ASEs are found to have a larger codon usage bias in both human and mouse. Our linear regression analyses indicate that the relationship between codon usage bias and Ks values is highly significant (P < 0.01) in both species, implying that codon usage bias has a strong effect on Ks values in mammalian ASEs.
|
In summary, we have shown that simple and complex ASEs are subject to opposite evolutionary effects. We therefore suggest that evolutionary analysis of AS should take the effects of different AS types into consideration and propose that AS may serve to relax or increase selection pressure at the amino acid level. The previous view that ASEs accelerate evolution of protein subsequences needs to be modified.
|
Methods |
|---|
|
TOPAbstractMethodsAcknowledgementsReferences |
|---|
We used 5,176 human–mouse orthologous gene pairs from the EBI database (http://www.ebi.ac.uk/) and extracted, using Blast, 26,106 reciprocal best-hit exon pairs. The human and mouse files used to annotate AS events (including the ASE types) were downloaded from the EBI-ASD (AltSplice Human Release 2 based on Ensembl 27.35a.1 and AltSplice Mouse Release 2 based on Ensembl 27.33c.1) at http://www.ebi.ac.uk/asd/altsplice/index.html. The Ka, Ks, and Ka/Ks values between orthologous exon pairs were computed using the yn00 program of the PAML package (Yang and Nielsen 2000
). The Ka/Ks ratio tests were performed as previously described (Nekrutenko et al. 2003). To study the effects of CG content on Ks, we extracted 4-fold degenerate sites from human ASEs and their mouse counterparts for calculation of Ks values. To exclude the effect of CpG dinucleotides, only sites that were neither preceded by a "C" nor followed by a "G" (non–CpG-prone sites) were considered. The effective number of codons and the codon bias index were evaluated using the DnaSP (Version 4.10.7) package (Rozas et al. 2003).
|
Acknowledgements |
|---|
|
TOPAbstractMethodsAcknowledgementsReferences |
|---|
This work was supported by the Genomics Research Center, Academia Sinica, and the National Health Research Institutes (NHRI), Taiwan, under the contract NHRI-EX96-9408PC to T.J.C., NHRI intramural funding to F.C.C., and in part by the Research Center for Biodiversity, Academia Sinica, to S.M.C. We thank the EBI-ASD Web interface for free downloaded data and Chuang-Jong Chen and Chia-Jung Chung for assistance in data collection. Special thanks are due to Dr Sudhir Kumar and the 3 anonymous reviewers who provided suggestive and helpful comments to the authors.
|
Footnotes |
|---|
Sudhir Kumar, Associate Editor
|
References |
|---|
|
TOPAbstractMethodsAcknowledgementsReferences |
|---|
Bracco L, Kearsey J. The relevance of alternative RNA splicing to pharmacogenomics. Trends Biotechnol (2003) 21:346–353.[CrossRef][Web of Science][Medline]
Brett D, Pospisil H, Valcarcel J, Reich J, Bork P. Alternative splicing and genome complexity. Nat Genet (2002) 30:29–30.[CrossRef][Web of Science][Medline]
Chen FC, Chuang TJ. The effects of multiple features of alternatively spliced exons on the Ka/Ks ratio test. BMC Bioinformatics (2006) 7:259.[CrossRef][Medline]
Chen FC, Wang SS, Chen CJ, Li WH, Chuang TJ. Alternatively and constitutively spliced exons are subject to different evolutionary forces. Mol Biol Evol (2006) 23:675–682.
Filip LC, Mundy NI. Rapid evolution by positive Darwinian selection in the extracellular domain of the abundant lymphocyte protein CD45 in primates. Mol Biol Evol (2004) 21:1504–1511.
Gene Ontology Consortium. Creating the gene ontology resource: design and implementation. Genome Res (2001) 11:1425–1433.
Hurst LD, Pal C. Evidence for purifying selection acting on silent sites in BRCA1. Trends Genet (2001) 17:62–65.[CrossRef][Web of Science][Medline]
Iida K, Akashi H. A test of translational selection at ‘silent’ sites in the human genome: base composition comparisons in alternatively spliced genes. Gene (2000) 261:93–105.[CrossRef][Web of Science][Medline]
Li W-H. Molecular evolution (1997) Sunderland (MA): Sinauer Associates.
Modrek B, Lee CJ. Alternative splicing in the human, mouse and rat genomes is associated with an increased frequency of exon creation and/or loss. Nat Genet (2003) 34:177–180.[CrossRef][Web of Science][Medline]
Nekrutenko A, Chung WY, Li WH. An evolutionary approach reveals a high protein-coding capacity of the human genome. Trends Genet (2003) 19:306–310.[CrossRef][Web of Science][Medline]
Rozas J, Sanchez-DelBarrio JC, Messeguer X, Rozas R. DnaSP, DNA polymorphism analyses by the coalescent and other methods. Bioinformatics (2003) 19:2496–2497.
Stamm S, Riethoven JJ, Le Texier V, Gopalakrishnan C, Kumanduri V, Tang Y, Barbosa-Morais NL, Thanaraj TA. ASD: a bioinformatics resource on alternative splicing. Nucleic Acids Res (2006) 34:D46–D55.
Xing Y, Lee C. Evidence of functional selection pressure for alternative splicing events that accelerate evolution of protein subsequences. Proc Natl Acad Sci USA (2005) 102:13526–13531.
Yang Z, Nielsen R. Estimating synonymous and nonsynonymous substitution rates under realistic evolutionary models. Mol Biol Evol (2000) 17:32–43.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  Y.-T. Huang, F.-C. Chen, C.-J. Chen, H.-L. Chen, and T.-J. Chuang Identification and analysis of ancestral hominoid transcriptome inferred from cross-species transcript and processed pseudogene comparisons Genome Res., July 1, 2008; 18(7): 1163 - 1170. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||