Limits...
Compensatory relationship between splice sites and exonic splicing signals depending on the length of vertebrate introns.

Dewey CN, Rogozin IB, Koonin EV - BMC Genomics (2006)

Bottom Line: In contrast, for longer introns, this effect was not detectable, and instead, an increase in the strength of the donor and acceptor splice sites was observed.Several weak but statistically significant correlations were observed between vertebrate intron length, splice site strength, and potential exonic splicing signals.Taken together, these findings attest to a compensatory relationship between splice sites and exonic splicing signals, depending on intron length.

View Article: PubMed Central - HTML - PubMed

Affiliation: National Center for Biotechnology Information NLM, National Institutes of Health, Bethesda, MD 20894, USA. cdewey@biostat.wisc.edu <cdewey@biostat.wisc.edu>

ABSTRACT

Background: The signals that determine the specificity and efficiency of splicing are multiple and complex, and are not fully understood. Among other factors, the relative contributions of different mechanisms appear to depend on intron size inasmuch as long introns might hinder the activity of the spliceosome through interference with the proper positioning of the intron-exon junctions. Indeed, it has been shown that the information content of splice sites positively correlates with intron length in the nematode, Drosophila, and fungi. We explored the connections between the length of vertebrate introns, the strength of splice sites, exonic splicing signals, and evolution of flanking exons.

Results: A compensatory relationship is shown to exist between different types of signals, namely, the splice sites and the exonic splicing enhancers (ESEs). In the range of relatively short introns (approximately, < 1.5 kilobases in length), the enhancement of the splicing signals for longer introns was manifest in the increased concentration of ESEs. In contrast, for longer introns, this effect was not detectable, and instead, an increase in the strength of the donor and acceptor splice sites was observed. Conceivably, accumulation of A-rich ESE motifs beyond a certain limit is incompatible with functional constraints operating at the level of protein sequence evolution, which leads to compensation in the form of evolution of the splice sites themselves toward greater strength. In addition, however, a correlation between sequence conservation in the exon ends and intron length, particularly, in synonymous positions, was observed throughout the entire length range of introns. Thus, splicing signals other than the currently defined ESEs, i.e., potential new classes of ESEs, might exist in exon sequences, particularly, those that flank long introns.

Conclusion: Several weak but statistically significant correlations were observed between vertebrate intron length, splice site strength, and potential exonic splicing signals. Taken together, these findings attest to a compensatory relationship between splice sites and exonic splicing signals, depending on intron length.

Show MeSH
Nucleotide composition of exon ends flanking introns in human varies with intron length. Median standard error bars are plotted for each point. Values for constitutive and alternative introns are shown with solid and dashed lines, respectively.
© Copyright Policy - open-access
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC1713244&req=5

Figure 3: Nucleotide composition of exon ends flanking introns in human varies with intron length. Median standard error bars are plotted for each point. Values for constitutive and alternative introns are shown with solid and dashed lines, respectively.

Mentions: Whereas for splice sites we observed correlations between signal strength and intron length for long introns (Figure 1), we found little, if any, correlation with ESE density for the same intron length range (Figure 2, Figure S4; see Additional File 1). However, for relatively short introns (<1.5 kb), we observed a positive correlation between intron length and the hexamer ESE density for both constitutive (R = 0.159, P ≈ 0) and alternative (R = 0.115, P ≈ 0) introns (Figure 2, Figure S4; see Additional File 1). As with splice site strength, ESE density was significantly higher in exons flanking constitutive introns than in those flanking alternative introns over all intron lengths (Mann-Whitney test). Since the putative ESE sequences have distinctive base compositions and variation in exon end base composition with intron length was observed (Figure 3, Figure S5; see Additional File 1), it seemed possible that the trends in ESE density were due entirely to the biases in base composition in exon ends. To test for this possibility, for each exon end, we calculated the difference (Eenrich) between the observed ESE density (Eobs) and the expected ESE density (Eexpect), given the base composition of that exon end. Note that Eenrich is in terms of density, and is therefore normalized by exon end length. The resulting Eenrich values had weaker but still significant correlations with intron length, for short constitutive (R = 0.086, P ≈ 0) and alternative (R = 0.060, P ≈ 0) introns. A slightly different test, in which the relative enrichment (Eenrich/Eexpect) was calculated, yielded weaker correlation values for short constitutive (R = 0.012, P = 4.37e-3) and alternative (R = 0.011, P = 0.102) introns. The results of these tests suggest a specific increase in hexamer ESE density with increasing intron length among relatively short introns, as opposed to a more general trend in base composition.


Compensatory relationship between splice sites and exonic splicing signals depending on the length of vertebrate introns.

Dewey CN, Rogozin IB, Koonin EV - BMC Genomics (2006)

Nucleotide composition of exon ends flanking introns in human varies with intron length. Median standard error bars are plotted for each point. Values for constitutive and alternative introns are shown with solid and dashed lines, respectively.
© Copyright Policy - open-access
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC1713244&req=5

Figure 3: Nucleotide composition of exon ends flanking introns in human varies with intron length. Median standard error bars are plotted for each point. Values for constitutive and alternative introns are shown with solid and dashed lines, respectively.
Mentions: Whereas for splice sites we observed correlations between signal strength and intron length for long introns (Figure 1), we found little, if any, correlation with ESE density for the same intron length range (Figure 2, Figure S4; see Additional File 1). However, for relatively short introns (<1.5 kb), we observed a positive correlation between intron length and the hexamer ESE density for both constitutive (R = 0.159, P ≈ 0) and alternative (R = 0.115, P ≈ 0) introns (Figure 2, Figure S4; see Additional File 1). As with splice site strength, ESE density was significantly higher in exons flanking constitutive introns than in those flanking alternative introns over all intron lengths (Mann-Whitney test). Since the putative ESE sequences have distinctive base compositions and variation in exon end base composition with intron length was observed (Figure 3, Figure S5; see Additional File 1), it seemed possible that the trends in ESE density were due entirely to the biases in base composition in exon ends. To test for this possibility, for each exon end, we calculated the difference (Eenrich) between the observed ESE density (Eobs) and the expected ESE density (Eexpect), given the base composition of that exon end. Note that Eenrich is in terms of density, and is therefore normalized by exon end length. The resulting Eenrich values had weaker but still significant correlations with intron length, for short constitutive (R = 0.086, P ≈ 0) and alternative (R = 0.060, P ≈ 0) introns. A slightly different test, in which the relative enrichment (Eenrich/Eexpect) was calculated, yielded weaker correlation values for short constitutive (R = 0.012, P = 4.37e-3) and alternative (R = 0.011, P = 0.102) introns. The results of these tests suggest a specific increase in hexamer ESE density with increasing intron length among relatively short introns, as opposed to a more general trend in base composition.

Bottom Line: In contrast, for longer introns, this effect was not detectable, and instead, an increase in the strength of the donor and acceptor splice sites was observed.Several weak but statistically significant correlations were observed between vertebrate intron length, splice site strength, and potential exonic splicing signals.Taken together, these findings attest to a compensatory relationship between splice sites and exonic splicing signals, depending on intron length.

View Article: PubMed Central - HTML - PubMed

Affiliation: National Center for Biotechnology Information NLM, National Institutes of Health, Bethesda, MD 20894, USA. cdewey@biostat.wisc.edu <cdewey@biostat.wisc.edu>

ABSTRACT

Background: The signals that determine the specificity and efficiency of splicing are multiple and complex, and are not fully understood. Among other factors, the relative contributions of different mechanisms appear to depend on intron size inasmuch as long introns might hinder the activity of the spliceosome through interference with the proper positioning of the intron-exon junctions. Indeed, it has been shown that the information content of splice sites positively correlates with intron length in the nematode, Drosophila, and fungi. We explored the connections between the length of vertebrate introns, the strength of splice sites, exonic splicing signals, and evolution of flanking exons.

Results: A compensatory relationship is shown to exist between different types of signals, namely, the splice sites and the exonic splicing enhancers (ESEs). In the range of relatively short introns (approximately, < 1.5 kilobases in length), the enhancement of the splicing signals for longer introns was manifest in the increased concentration of ESEs. In contrast, for longer introns, this effect was not detectable, and instead, an increase in the strength of the donor and acceptor splice sites was observed. Conceivably, accumulation of A-rich ESE motifs beyond a certain limit is incompatible with functional constraints operating at the level of protein sequence evolution, which leads to compensation in the form of evolution of the splice sites themselves toward greater strength. In addition, however, a correlation between sequence conservation in the exon ends and intron length, particularly, in synonymous positions, was observed throughout the entire length range of introns. Thus, splicing signals other than the currently defined ESEs, i.e., potential new classes of ESEs, might exist in exon sequences, particularly, those that flank long introns.

Conclusion: Several weak but statistically significant correlations were observed between vertebrate intron length, splice site strength, and potential exonic splicing signals. Taken together, these findings attest to a compensatory relationship between splice sites and exonic splicing signals, depending on intron length.

Show MeSH