CD48-deficient T-lymphocytes from DMBA-treated rats have de novo mutations in the endogenous Pig-a gene

A major question concerning the scientific and regulatory acceptance of the rodent red blood cell-based Pig-a gene mutation assay is the extent to which mutants identified by their phenotype in the assay are caused by mutations in the Pig-a gene. In this study, we identified T-lymphocytes deficient for the glycosylphosphatidylinositol-anchored surface marker, CD48, in control and 7,12-dimethylbenz[a]anthracene (DMBA)-treated rats using a flow cytometric assay and determined the spectra of mutations in the endogenous Pig-a gene in these cells. CD48-deficient T-cells were seeded by sorting at one cell per well into 96-well plates, expanded into clones, and exons of their genomic Pig-a were sequenced. The majority (78%) of CD48-deficient T-cell clones from DMBA-treated rats had mutations in the Pig-a gene. The spectrum of DMBA-induced Pig-a mutations was dominated by mutations at A:T, with the mutated A being on the nontranscribed strand and A → T transversion being the most frequent change. The spectrum of Pig-a mutations in DMBA-treated rats was different from the spectrum of Pig-a mutations in N-ethyl-N-nitrosourea (ENU)-treated rats, but similar to the spectrum of DMBA mutations for another endogenous X-linked gene, Hprt. Only 15% of CD48-deficient mutants from control animals contained Pig-a mutations; T-cell biology may be responsible for a relatively large fraction of false Pig-a mutant lymphocytes in control animals. Among the verified mutants from control rats, the most common were frameshifts and deletions. The differences in the spectra of spontaneous, DMBA-, and ENU-induced Pig-a mutations suggest that the flow cytometric Pig-a assay detects de novo mutation in the endogenous Pig-a gene.