PhD student Boldina G.1, prof. Ivashchenko A.1, prof. Régnier M.2

1 – al-Farabi Kazakh National University, Almaty, Kazakhstan

2 – INRIA, Le Chesnay, France

IDENTIFICATION REGIONS, WHICH ARE ESSENTIAL FOR SPLICING, ON THE BASE OF HYDROPATHY PROFILES

 

Pre-mRNA splicing is a nuclear process conserved across eukaryotes. The spliceosome recognizes conserved sequences at the exon–intron boundaries. There are at two classes of pre-mRNA introns, based on the splicing machineries that catalyze the reaction: U2 and U12 snRNP-dependent introns. Most human introns, around 99.66% /1/, are likely to be U2- type introns. The U2-type introns have highly degenerate sequence motifs. It is still largely unknown how degenerate sequences at the U2 splice sites are recognized by spliceosome.

In order to find out regions with conservative properties, namely hydropathy, which may be recognized by spliceosome and to distinguish U2 and U12-types of introns, we defined hydropathy profiles.

Methods

In order to define a general hydropathy profile we built a set of 313 introns and a set of 385 exons from genes of 21st and 22nd chromosomes contained 1 to 3 introns from GenBank (http://www.ncbi.nlm.nih.gov). The flanking sequences (30 nt within the exon and 30 nt within the intron) were extracted at both exon–intron junctions, at 5'ss and 3'ss boundaries. We determined the background hydropathy value E that is -0.996 for exons and -1.01 for introns. The corresponding variances are VE = 0.0687 and VI = 0.0690. Regions whose hydropathy differs from the background value are expected to be essential for recognition by spliceosome. . At the hydropathy evaluating, the procedure is given below, the hydropathy coefficients provided by Guckian et al. in 2000 /2/ were associated to each base i. Given set of splice sites, one computes an average hydropathy value for each position as follows. For each base, its number of occurrences at a given position in the set is multiplied by its hydropathy coefficient. Summing over all the bases yields the average hydropathy value. We computed P-value with help of the large deviation formula /3/ for positions at the splice sites which deviate from approximately normal distribution. In order to define hydropathy profiles for splice sites of two subgroups of U2 and U12-type introns we built four sets of 100 introns with confirmed splice sites extracted from SpliceRack database (http://katahdin.cshl.edu:9331/SpliceRack/). Two sets are associated to human U2-type introns, with GT–AG and GC–AG termini, and two sets of U12-type introns, with GT–AG and AT–AC termini correspondingly. For each intron we extracted 8 nt within the exon and 30 nt within the intron..

Results

Our method attempts to point out regions which have conservative properties, namely hydropathy, from a variable background. Hydropathy profile of genes of 21st and 22nd chromosomes contained 1 to 3 introns is illustrated in Figure 1. For all pictures the numbers of nucleotides are marked on the x-axis and hydropathy values are indicated by the scale on y-axis. The termini of the introns are marked in red. Average values of background hydropathy are marked by red line. Limits of 99.9% confidence intervals are given by blue dotted lines.

 Figure 1. Distinguishing biochemically conservative regions from background values

The positions of nucleotides are marked on the x-axis and hydropathy values are indicated by the scale on y-axis.

 

At positions -30 to -3 within the introns and +8 to +30 within the exons at the 5’ss and at positions +2 to +30 within the exons at the 3’ss, deviations from the average are consistent with an approximately normal distribution of hydropathy values. Slow decay at positions -26 to -5 due to pyrimidine abundance correspond to polypyrimidine track.

Regions at the positions -2 to +6 at the 5’ss and -26 to +1 at the 3’ss deviate from the background hydropathy with significant P-values. General hydropathy profile of splice cites of genes with 1- 3 introns from 21st and 22nd chromosomes resembles to the U2-type introns hydropathy profile (Figure 2), because of the low proportion of U12-type introns that does not exceed 0,34% /1/.

In Figures 2 and 3, the hydropathy profiles of U2 and U12- introns with different termini are depicted.

U2-type introns

100 splice sites of U2-type introns with GT-AG as well as with GC-AG termini extracted from SpliceRack are considered separately in order to compare their hydropathy profiles.

Figure 2. The hydropathy profiles of the U2-type introns. The hydropathy profiles of two subtypes U2-type introns with GT-AG (A-B) and GC-AG (C-D) termini are shown.

The hydropathy profiles of GT–AG and GC–AG subtypes are quite similar (Figure 2), Indeed, nucleotide consensus at the 5’ss of U2-type introns mainly contain quite hydrophilic purines when termini are either GT-AG (Figure 2 A) or GC-AG (Figure 2 C). We were interested to reveal mechanisms of the competence of both U2-subtypes with inconsistent 3’ss for splicing machinery. We show that the 3’ss of both subtypes of U2-type introns have resembling hydropathy profiles (Figure 2 B, D). Intronic nucleotides of both subtypes of U2-type introns are enriched in pyrimidines and as a result hydrophobic. Therefore, our results indicate that the mechanism of functional regions definition is probably based on recognition of conservative features, but not only on nucleotide basepairing.

U12-type introns.

Figure 3. The hydropathy profiles of the U12-type introns. The hydropathy profiles of two subtypes U12-type introns with GT-AG (A-B) and AT-AC (C-D) termini are shown.

The 5'ss of U12-type introns (Figure 3 A, C), show a high degree of conservation at intronic positions +1 to +9 and correspondingly the large P-values. The BPS, which lies close to the 3’ end of the intron, shows quite high degree of conservation of hydropathy. Indeed, the BPS is rich of pirimidines for both U12-type GT-AG and AT-AC. In order to explain recognition mechanisms of the U2 and the U12 – dependent introns with the same terminal dinucleotides we compared the hydropathy profiles of the U2 and the U12 – dependent introns (Figure 2A, B and 3A, B). The 5'ss profile for U12-type GT-AG is different from the U2-type GT-AG. Indeed, exonic and intronic nucleotides of the 5’ss of U2 – dependent introns have hydrophilic purine rich consequence, while 5’ss of U12- dependent introns match hydrophobic pyrimidine rich canonical consensus (plateau in Figure 3 A and C) that succeeds terminal dinucleotides.

The 3'ss profile for U12-type GT-AG is different from the U2-type GT-AG (Figures 2B, D and 3B, D). U12-type introns lack obvious PPT at the 3’ss and the BPS lies close to the 3’ end of introns.

Conclusion

We showed similarity of hydropathy profiles inside intron types. On the one hand, GT–AG and GC–AG introns belonging to U2-type have resembling hydropathy profiles as well as AT–AC and GT–AG introns belonging to U12-type. On the other hand, hydropathy profiles of U2 and U12-types GT–AG introns are completely different. Our analysis should be a step forward for a general understanding of recognition of regions, which are essential for splicing, by spliceosome and for a distinction of U2 and U12-types of introns.

References:

1.                 Levine, A. and Durbin, R. (2001) A computational scan for u12-dependent introns in the human genome sequence Nucleic Acids Res,. 29, 4006–4013.

2.                 Guckian K.M., Schweitzer B.A., Ren R., X.-F., Sheils C.J., Tahmassebi D.C., Kool E.T. (2000) Factors conributing to aromatic stacking in water: evaluation in context of DNA  J. Am. Chem. Soc., 122, 2213-2222.

3.                 Régnier M., Vandenbogaert M. (2006) Comparison of statistical significance criteria. Journal of Bioinformatics and Computational Biology, 4, 537–551.