Gross gene rearrangements - Human Gene Evolution - page 397

unusual cause of human genome pathology (reviewed by Cooper and Krawczak
1993). Such fusions are however thought to have had an important role in evolu-
tion through the creation of novel genes encoding novel combinations of func-
tional protein domains; some examples of this phenomenon are described below.
As we have seen in Chapter 3, section 3.6, exon shuffling has been an important
process in evolution, serving to bring together coding sequence blocks from dif-
ferent sources to generate new proteins capable of novel interactions and there-
fore, potentially, with novel functions. In a sense, therefore, all proteins have
evolved by a series of gene fusion events. In this section, however, we concentrate
solely on the generation of novel genes through the fusion of once distinct and
independently functional gene sequences by recombination.
9.3.1 Gene fusion during evolution
Carboxyesterase E1, an enzyme responsible for the detoxification of ingested
xenobiotics, exhibits sequence homology both with acetylcholinesterase and the
carboxy terminal end of thyroglobulin, a precursor of thyroid hormone (Takagi et
al ., 1991). Phylogenetic analysis has suggested that both acetylcholinesterase
(human ACHE gene on 7q22) and thyroglobulin (human TG gene on 8q24.2-
q24.3) evolved from a common ancestral gene that encoded a carboxyesterase and
that the emergence of thyroglobulin must have preceded the divergence of the
vertebrates and invertebrates (Takagi et al ., 1991). The evolutionary origin of the
amino terminal portion of extant thyroglobulin is however unknown.
The human multidrug resistance ( PGY1 ; 7q21.1) gene encodes a membrane-
associated pump protein called P-glycoprotein. Sequence homology between the
amino and carboxy terminal halves of the protein was initially held to be consis-
tent with the view that the protein evolved by duplication of a primordial gene.
However, once the structure of the 29 exon PGY1 gene had been determined, it
became clear that only two intron pairs (both within nucleotide binding domains)
were located in conserved positions in the two halves of the protein. Thus, rather
than a primordial duplication, Chen et al . (1990) proposed that primordial pro-
teins corresponding to the left and right halves of P-glycoprotein were formed
independently by the fusion of closely related genes encoding the nucleotide-
binding domain with genes for different transmembrane domains. Subsequent
fusion of these two independently derived genes then resulted in the formation of
the PGY1 gene. Pauly et al . (1995) have speculated that a highly conserved
poly(CA).poly(TG) sequence in intron 15 of the PGY1 gene could have, with Alu
sequences in introns 14 and 17, mediated such a fusion event by recombination.
The human glutaminyl-tRNA synthetase ( EPRS ) gene serves to encode two
distinct aminoacyl-tRNA synthetase activities joined as part of a multi-enzyme
synthetase complex. The EPRS gene, located at chromosome 1q41-q42, com-
prises 29 exons spanning some 90 kb (Kaiser et al ., 1994). Exons 4 to 10 encode a
glutaminyl-tRNA synthetase (Kaiser et al ., 1992) whilst exons 19 to 29 encode a
prolyl-tRNA synthetase (Kaiser et al ., 1994). The function of the intervening
region (exons 11 to 18) is unclear (actin binding?; Kaiser et al ., 1994). Since the
glutaminyl- and prolyl-tRNA synthetases belong to different enzyme classes
which are believed to have evolved by separate pathways (Nagel and Doolittle,