Abstract

Thymidylate synthase (TS) provides the sole de novo source of the DNA precursor thymidylate (dTMP) in almost all organisms, and is one of the most conserved enzymes known. One salient feature of the eukaryotic version of the enzyme is the occurrence of two peptide inserts, absent in the prokaryotes, in surface loops peripheral to the core structure. The biological function of these inserts is unknown, but they have been shown in yeast TS to contribute to the structural integrity of the enzyme (Munro et al ., 1999). More precisely, the removal of one of the inserts (EUK1) leads to reduced affinity for both substrates as well as decreased enzyme activity, while removal of the other (EUK2) completely abolishes enzyme activity. In this study we further analyzed effects of deletion and point mutations in the loop (Loop 2) that harbours EUK2. Gel filtration chromatography indicated that inactive deletion mutants that mimic prokaryotic versions of this loop fail to form stable dimers. Activity could not be restored when a phenylalanine residue, presumed to be buried by the conformation of the normal eukaryotic loop, was substituted by a polar residue which even more closely mimicked the prokaryotic TS. Point mutations in EUK2 that substituted one or the other of two conserved tyrosines with phenylalanine partially affected enzyme activity; complementation of TS-deficient E. coli appeared to be gene-dosage dependent and tritium release activity was drastically reduced both in E. coli and yeast. Surprisingly however, yeasts that were auxotrophic for dTMP did not suffer any discernible deleterious effects when complemented by the single tyrosine/phenylalanine mutants, even when the mutant genes were expressed from single copy plasmids. A mutant with both tyrosines replaced by phenylalanine did not complement in E. coli , complemented in a yeast TS-knockout strain only when overexpressed, but showed heteroallelic complementation in a yeast strain that has a mutation in the active site of TS. All mutations introduced in Loop 2 so far dramatically reduce enzyme activity as determined by tritium release assays. These results indicate that Loop 2, despite its peripheral location, is highly sensitive to modification and contributes to the structural integrity of the protein. In contrast, prokaryotic TS (from E. coli ) was functionally expressed in a yeast TS-knockout strain, suggesting that although EUK1 and EUK2 are essential for structural integrity they do not provide an additional essential biological function. Moreover, our results suggest that the enzymatic impairment of Loop 2 is offset by a factor present in S. cerevisiae cells but not in E. coli