Although intron loss and gain have been widely observed, their mechanisms are still to be determined. In four Aspergillus genomes, we found 204 cases of intron loss and 84 cases of intron gain. Using this data, we tested common hypotheses of intron loss or gain. Statistical analysis showed that adjacent introns tend to be lost simultaneously and small introns were preferentially lost, supporting the model of mRNA-mediated intron loss. The lost introns reside in internal regions of genes, which is inconsistent with the traditional version of the model (partial length cDNAs are reverse transcribed from 3' ends of mRNAs), but consistent with an alternate version (partial length cDNAs are produced by self-primed reverse transcription). The latter version was not supported by examination of the abundance of T-rich segments in mRNAs. Preferential loss of internal introns might be explained by highly efficient recombination at internal regions of genes. Among the 84 cases of intron gain, we found a significantly higher frequency of short direct repeats near exon-intron boundary than in conserved introns, supporting the double-strand break repair model. We also found possible source sequences for two cases of intron gain, one by gene conversion and one by insertion of a mitochondrial sequence during double-strand break repair. Source sequences for most gained introns could not be identified and the possible reasons were discussed. In the four Aspergillus genomes studied, we did not find evidence of frequent parallel intron gains.

No abstract

Introns were found to enhance almost every steps of gene expression except increasing mRNA stability. By analyzing the genome-wide data of mRNA stability published by someone previously, we found that human intron-containing genes have more stable mRNAs than intronless genes, and the Arabidopsis thaliana genes with the most unstable mRNAs have fewer introns than other genes in the genome. After controlling for mRNA length, we found mRNA stability is still positively correlated with intron number in human intron-containing genes. But in yeast Saccharomyces cerevisiae, two different datasets on mRNA half-life gave conflicting results. The components of messenger ribonucleoprotein particles recruited during intron splicing may be retained in cytoplasmic mRNPs and act as signals of mRNA stability or simply insulators to avoid mRNA degradation.

No abstract

BackgroundThe costs and benefits of spliceosomal introns in eukaryotes have not been established. One recognized effect of intron splicing is its known enhancement of gene expression. However, the mechanism regulating such splicing-mediated expression enhancement has not been defined. Previous studies have shown that intron splicing is a time-consuming process, indicating that splicing may not reduce the time required for transcription and processing of spliced pre-mRNA molecules; rather, it might facilitate the later rounds of transcription. Because the densities of active RNA polymerase II on most genes are less than one molecule per gene, direct interactions between the splicing apparatus and transcriptional complexes (from the later rounds of transcription) are infrequent, and thus unlikely to account for splicing-mediated gene expression enhancement.Presentation of the hypothesisThe serine/arginine-rich protein SF2/ASF can inhibit the DNA topoisomerase I activity that removes negative supercoiling of DNA generated by transcription. Consequently, splicing could make genes more receptive to RNA polymerase II during the later rounds of transcription, and thus affect the frequency of gene transcription. Compared with the transcriptional enhancement mediated by strong promoters, intron-containing genes experience a lower frequency of cut-and-paste processes. The cleavage and religation activity of DNA strands by DNA topoisomerase I was recently shown to account for transcription-associated mutagenesis. Therefore, intron-mediated enhancement of gene expression could reduce transcription-associated genome instability.Testing the hypothesisExperimentally test whether transcription-associated mutagenesis is lower in intron-containing genes than in intronless genes. Use bioinformatic analysis to check whether exons flanking lost introns have higher frequencies of short deletions.Implications of the hypothesisThe mechanism of intron-mediated enhancement proposed here may also explain the positive correlation observed between intron size and gene expression levels in unicellular organisms, and the greater number of intron containing genes in higher organisms.ReviewersThis article was reviewed by Dr Arcady Mushegian, Dr Igor B Rogozin (nominated by Dr I King Jordan) and Dr Alexey S Kondrashov. For the full reviews, please go to the Reviewer's Reports section.