The biosynthesis of thymidylate synthase (TS) has recently been found to be down-regulated by binding of the protein to its own mRNA, and this regulation is involved in the development by cancer cells of resistance to several antimetabolites targeted at this enzyme. Because both TS-mRNA binding to TS and TS-enzyme activity were observed to increase similarly with increasing mercaptoethanol (ME) concentration, we tested whether these affinity and activity increases were consistent with either the hypothesis of control by a single common redox site or the hypothesis of independent control of binding and activity by two separate sites. Thus, data sets describing both the binding of TS to TS-mRNA and the activity of TS (no mRNA present) with added ME were fit separately to models accounting for binding to two known mRNA sites, for thymidylate production, and for appropriate choice of putative redox control. Likelihood ratio testing for either redox model revealed that the redox parameter vector is very significantly different when estimated from the binding or the activity data, thus leading to the rejection of both hypotheses. Hence, redox regulation of TS synthesis appears to be complex, involving coupled interactions between multiple sites. In addition, some structural investigation of the TS-mRNA binding region was begun. Melting points of two possible 35-mer stem-loop structures, identified previously, were examined for UV absorbance and found to be consistent with predicted G-C-rich stems. Furthermore, because the stem-loop structures differ in having a single adenine either free or base-paired, 1D-NMR was performed to seek evidence for far down-field-shifted imidazole hydrogens of U-paired adenine. Shifting was less than 13 ppm, thus indicating the need for 2D-NMR.
