Homology between coding and noncoding sequences within the human class I HLA antigen gene.

Abstract

In our previous paper, we identified 32 recurring base oligomers (1 decamer, 4 octamers, 9 heptamers and 18 hexamers) within the coding sequence for the mouse class I major histocompatibility (MHC) antigen H-2Kb. The compilation of these recurring base oligomers led to the conclusion that the entire ancestral coding sequence for class 1 MHC antigens evolved from tandem repeats of the one 45 base-long primordial building block base sequence. As with most other mammalian genes, the gene for each class I MHC antigen is an admixture of coding and noncoding segments interspersed with each other. Thus, the status of noncoding segments should be clarified in relation to the concept of the primordial building block. The published 4,122 base-long sequence of human pHLA 12.4 germline gene afforded me an opportunity for clarification. Since the number of recurring base oligomers residing within the entire 4,000 + base-long sequence proved unmanageably numerous, the two portions 1,200 bases in the total length were singled out. Within these portions containing five noncoding and four coding segments, noncoding and coding segments shared 1 nonomer, 3 octamers, 6 heptamers and 5 hexamers; all the above-noted base oligomers were derived from different parts of the same 45 base-long primordial building block. It was thus concluded that not only the coding segments, but the entire ancestral gene for class I MHC antigens evolved from tandem repeats of the 45-base-long primordial building block. This new concept of primordial building block was discussed in relation to the mechanism of evolution by gene duplication.