Partial nucleotide sequence of the 300-nucleotide interspersed repeated human DNA sequences.

Abstract

In most eukaryotic genomes, including human, 300-nucleotide repeated DNA sequences are interspersed with longer (∼1,000 nucleotide) single copy sequences1–3. We have recently found that most 300-nucleotide interspersed repeats in human share a common site for cleavage by the restriction enzyme AluI and should be regarded as a single family of sequences4. We designate this as the Alu family of sequences. Similarly, most of the 300-nucleotide inverted repeated sequences, which are also interspersed with single copy DNA, share this same restriction site and belong to the Alu family4. There are approximately 300,000 members of this family of sequences, which together make up at least 3% of the human genome4. It is conceivable that individual members of the Alu family repeats share only very limited regions of homology, one of which happens to contain the restriction site for AluI and others which share the additional restriction sites reported here. In this case, members of the Alu family could be essentially different DNA sequences. DNA renaturation studies support the alternative view that members of the Alu family share extensive homology over the entire sequence length4. According to this view, individual members of the family could exhibit some divergence from the ancestral sequence but all members would be recognised as a closely related group of sequences5. We have distinguished between these alternatives by directly determining the base sequence of a part of the Alu family. This base sequence shows that individual members of the Alu family share a common ancestral nucleotide sequence.