PCR-based analysis of the T-cell receptor V beta multigene family: experimental parameters affecting its validity.

The validity of semiquantitative, PCR-based analysis of gene expression within a multigene family, the human T-cell receptor (TCR) beta chain variable region family, was investigated. Primer comparability was addressed by grouping hybridization temperatures and limiting the size range of amplified fragments. Primers selected satisfied criteria for comprehensiveness, match to targets and discrimination of nontargets. Specificity was enhanced by maximizing mismatches with nontargets and using an elevated hybridization temperature. Reaction conditions are described that ensure specificity while maintaining sensitivity. Several results confirmed primer specificity. Limits on precision were documented: probable error was 3%-7% of mean value for target prevalences (% of all TCR mRNA represented by a particular V beta) in the 5%-40% range. Accuracy was limited by the nonlinear relationship between target prevalence and signal obtained. Because of this relationship, the effect of the observed limits of precision varied. Valid distinctions were possible between sufficiently separated prevalences, i.e., 0%-1%, 3%-5%, 10%, 30%, greater than 50%. Additional concerns addressed include: standardization of signals, coamplifiation and effects of primer artifacts, and the nature of the mRNA pool. Only when theoretical and practical limits in precision and accuracy are acknowledged can semiquantitative, PCR-based analysis be used with confidence to assess gene usage within a large, multigene family.