Estimation of the functional consequences of single nucleotide polymorphisms (SNPs) is still in its infancy. Although some SNPs have clearly been shown to have a major impact on activity of the resultant drug metabolizing enzyme, these examples are likely to be the ''low hanging fruit'' of pharmacogenetics. There are a much larger number of functional SNPs that each have only a minor impact on gene function, but where individual subjects with specific combinations are likely to have major deviations from mean enzyme activity, and these permutations are likely to account for most of the variance in drug effect in a population. Thymidylate synthase (TS) is a good example of this issue. Specific SNPs in the upstream promoter have been shown to explain phenotypic variation only in specific situations, and attempts to extrapolate to more common clinical circumstances have not been as fruitful as expected. The need to look for more genetic variations and study them in concert has been identified. The recent discovery by us 1 and Gusella et al.2 of a new SNP in the TYMS promoter is an example of progress in this field.There is still much to learn about the functional consequences of the G > C promoter polymorphisms on the regulation of TS. In the interim, the need for a standardized nomenclature is apparent. Kawakami et al.3 designated a TYMS 2R allele with a C > G snp at the 12th nucleotide of the second 28-base pair repeat as 2G, with 2C being the wild type. Subsequently, Gusella et al.2 have called a 2R allele with a G > C polymorphism at the same position in the first repeat as either 2RC or 2RG. To avoid confusion, we feel the only solution is to indicate the actual nucleotide present at the 12th position of each repeat, prefixed by the number of repeats present in the allele; thus the latter most recently described 2R variant would be indicated by 2RCC. Presently, the functional consequences of these polymorphisms can only be postulated based on the number of potential upstream stimulatory factor (USF) binding sites (i.e., those containing a G at the polymorphic site) in each allele. The actual effect of these repeats on gene transcription has only been specifically addressed in three reports. The two most recent studies investigated TYMS reporter gene constructs in luciferase assay systems, and found no appreciable difference in TS mRNA levels from 2RGG 3 or 2RCC 4 (respectively) compared to wild type 2RGC in untreated cells. However, Mandola et al.4 took their experiments one step further and showed that the addition of exogenous USF-1 dramatically changed the results of the assay, and under these conditions both 2RCC and 3RCCC had significantly reduced transcription compared to their wild type counterparts. More general studies by Kaneda et al.
employed measurements of [3 H] deoxyuridine as an indication of TS-mediated incorporation of thymidylate into the genome of TS-negative murine cells after transient transfection with various TYMS plasmid constructs. Not only did this fail to show a difference between th...