Analysis of similarity within 142 pairs of orthologous intergenic regions of Caenorhabditis elegans and Caenorhabditis briggsae

Abstract

Patterns of similarity between genomes of related species reflect the distribution of selective constraint within DNA. We analyzed alignments of 142 orthologous intergenic regions of Caenorhabditis elegans and Caenorhabditis briggsae and found a mosaic pattern with regions of high similarity (phylogenetic footprints) interspersed with non-alignable sequences. Footprints cover approximately 20% of intergenic regions, often occur in clumps and are rare within 5' UTRs but common within 3' UTRs. The footprints have a higher ratio of transitions to transversions than expected at random and a higher GC content than the rest of the intergenic region. The number of footprints and the GC content of footprints within an intergenic region are higher when genes are oriented so that their 5' ends form the boundaries of the intergenic region. Overall, the patterns and characteristics identified here, along with other comparative and experimental studies, suggest that many footprints have a regulatory function, although other types of function are also possible. These conclusions may be quite general across eukaryotes, and the characteristics of conserved regulatory elements determined from genomic comparisons can be useful in prediction of regulation sites within individual DNA sequences.

Figure 1

Sample alignment of one intergenic region of C.elegans and C.briggsae. Bordering exons are indicated in capital letters, and footprints are indicated in bold. Caenorhabditis elegans sequence from accession number AF036702 (cosmid F33D4): 21750–22253, and C.briggsae sequence from accession number AC084481 (cosmid G02P14): 11868–12399.

Figure 2

(A) Regression of the length of the intergenic region against the class of intergenic region in C.elegans (y = 337.4x + 433.1). (B) Regression of the number of footprints within an intergenic region against the class of intergenic region in C.elegans (y = 1.0x + 1.6). The regressions for C.briggsae are similar (not reported).

Figure 3

Distribution of similarity for (A) footprints and (B) intergenic regions for C.elegans. The distributions for C.briggsae are similar (not reported).

Figure 4

Frequency profile of footprints covering each nucleotide in C.elegans from (A) the 3′ end of translation to 300 nt into the intergenic region (n = 52) and (B) from 300 nt into the intergenic region to the 5′ end of translation (n = 62) for intergenic regions >600 nt in length. The white bar indicates the length of the average 3′ UTR (A) and 5′ UTR (B). The profiles for C.briggsae are similar (not reported).

Figure 5

Example of clumping of footprints within an intergenic region. Lines indicate where footprints fall within the intergenic region, and 0 is the beginning of the intergenic region. Caenorhabditis elegans sequence from accession number Z74040 (cosmid K10D6): 8203–13293, and C.briggsae sequence from cosmid MM10A5: 17365–21824.

Figure 6

The distribution of interfootprint distances in C.elegans measured as the length from the end of one footprint to the beginning of the next. The bars represent the observed distribution of 225 interfootprint distances. The expected line is from an exponential distribution fitted with the same parameter as the observed data. The distribution for C.briggsae is similar (not reported).

Figure 7

Regression of %GC content within a footprint against the class of intergenic region in C.elegans (y = 0.03x + 0.34). The regression for C.briggsae is similar (not reported).