THE CLASSICAL HUMAN PHOSPHOGLUCOMUTASE (PGM1) ISOZYME POLYMORPHISM ISGENERATED BY INTRAGENIC RECOMBINATION

The molecular basis of the classical human phosphoglucomutase 1 (PGM1) isozyme polymorphism has been established. In 1964, when this genetic polymorphism was first described, two common allelozymes PGM1 1 and P GM1 2 were identified by starch gel electrophoresis. The PGM1 2 isozym e showed a greater anodal electrophoretic mobility than PGM1 1. Subseq uently, it was found that each of these allelozymes could be split, by isoelectric focusing, into two subtypes; the acidic isozymes were giv en the suffix + and the basic isozymes were given the suffix -. Hence, four genetically distinct isozymes 1+, 1-, 2+, and 2- were identified . We have now analyzed the whole of the coding region of the human PGM 1 gene by DNA sequencing in individuals of known PGM1 protein phenotyp e. Only two mutations have been found, both C to T transitions, at nt 723 and 1320. The mutation at position 723, which changes the amino ac id sequence from Arg to Cys at residue 220, showed complete associatio n with the PGM1 2/1 protein polymorphism: DNA from individuals showing the PGM1 1 isozyme carried the Arg codon CGT, whereas individuals sho wing the PGM1 2 isozyme carried the Cys codon TGT. Similarly, the muta tion at position 1320, which leads to a Tyr to His substitution at res idue 419, showed complete association with the PGM1+/- protein polymor phism: individuals with the + isozyme carried the Tyr codon TAT, where as individuals with the - isozyme carried the His codon CAT. The charg e changes predicted by these amino acid substitutions are entirely con sistent with the charge intervals calculated from the isoelectric prof iles of these four PGM1 isozymes. We therefore conclude that the mutat ions are solely responsible for the classical PGM1 protein polymorphis m. Thus, our findings strongly support the view that only two point mu tations are involved in the generation of the four common alleles and that one allele must have arisen by homologous intragenic recombinatio n between these mutation sites.