GENES-IN-PIECES REVISITED
GENES-IN-PIECES REVISITED
复制标题
DOI：
10.1126/science.4001923
复制
发表时间：
1985-01-01
期刊：
影响因子：
56.9
通讯作者：
GILBERT, W
中科院分区：
文献类型：
--
作者：
GILBERT, W
Mammalian genesare discontinuous, broken up along the DNA into alternating regions: coding sequences or exons, which are interspaced with other sequences, and introns that will bespliced out of the RNA transcript. What is the meaning of this arrangement? In 1977, I conjectured that genes in eukaryotic cells arose as collections of exons brought together by recombination within intron sequences, and that the introns were the remnants of a process thatspeeded up evolution (1, 2). This hypothesis predicts that the exons code for useful portions of protein structure: functional regions, folding ele-ments, domains, or subdomains-any segment that can be sorted independent-ly during evolution (3). As relics of the recombination process that brought the exons together, the introns would be long, random sequences that would drift rapidly in sequence and size since the last act thatassembled the gene. Ford Doolittle (4) realized that there is no reason for this speeding of the evolu-tionary process to be restricted to eu-karyotes, and he argued that the earliest organisms shouldhave had split genes; the present day intron-less genomes of prokaryotes and of lower eukaryotes would then be the result of streamlining, a consequence of the evolutionary pressure forrapid replication. How well have these ideas fared?Today we can draw some general con-clusions about intron patterns. Introns are common invertebrate genes, essen-tially absent in prokaryotes, and rare in lower eukaryotes such as yeast, with some dramatic exceptions such as the bithorax complex genes of drosophila. The distribution of exon sizes is rather narrow and peaks at 40 to 50 amino acid residues, but the sizes of introns scatter randomly and range from 50 to 10,000 to 20,000 bases in length. Since the introns are, on the average, so much larger than the exons, vertebrate genes have turned out to be much larger than we expected a decade ago, about 10 to 30 times larger than the coding sequence. The largest gene yet found is 200 kilobases long
