Subfamily relationships and clustering of rabbit C repeats

Abstract

C repeats constitute the predominant family of short interspersed repeats (SINEs) in the rabbit genome. Determination of the nucleotide sequence 5′ to rabbit ζ-globin genes reveals clusters of C repeats, and analysis of these and other sequenced regions of rabbit chromosomes shows that the C repeats have a strong tendency to insert within or in close proximity to other C repeats. An alignment of 44 members of the C repeat family shows that they are composites of different sequences, including a tRNA-like sequence, a conserved central core, a stretch of repeating CT dinucleotides, and an A-rich tract. Cladograms generated by both parsimony and cluster analysis subdivide the C repeats into at least three distinct subfamilies. Nucleotides at sites diagnostic for subfamilies appear to have changed in a punctuated and progressive manner during evolution, indicating that a limited number of progenitors have given rise to new repeats in waves of dispersion. C repeats that insert into preexisting C repeats belong to subfamilies that are proposed to have been propagated more recently; hence, these data support the model of dispersion in successive waves. The divergence among the oldest group of C repeats is greater than that observed for the analogous Alu repeats in humans, indicating that rabbit C repeats have been propagating longer than human Alu repeats. The improved consensus sequence for these repeats is similar to that of the predominant artiodactyl SINE in both the tRNA-like region and a central region. Because members of different subfamilies cross-hybridize very poorly, hybridization data with representatives of each subfamily provide a new minimal estimate, 234,000, for the copy number of C repeats in the rabbit haploid genome, although it is likely that the actual value is closer to 1 million.