The distribution of interspersed repeats is nonuniform and conserved in the mouse and human genomes

Abstract

We investigated the genomic distribution of mouse and human repeated sequences by assessing their relative amounts in the four major components into which these genomes can be resolved by density gradient centrifugation techniques. These components are families of fragments that account for most or all of main-band DNAs, range in dG + dC content from 37% to 49%, and are derived by preparative breakage from long DNA segments (greater than 300 kb) of fairly homogeneous composition, the isochores. The results indicate that the short repeats of the B1 family of mouse and of the Alu I family of man are most frequent in the heavy components, whereas the long repeats of the BamHI family of mouse and of the Kpn I family of man are mainly present in the two light components. These results show that the genomic distribution of repeated sequences is nonuniform and conserved in two mammalian species. In addition, we observed that the base composition of two classes of repeats (60% dG + dC for short repeats; 39% dG + dC for long repeats) is correlated with the composition of the major components in which they are embedded. Finally, we obtained evidence that not only the short repeats but also the long repeats are transcribed, these transcripts having been found in mouse poly(A)+ mRNA.