Biosynthetic incorporation of telluromethionine into dihydrofolate reductase and crystallographic analysis of the distribution of tellurium atoms in the protein molecule

Recent successes in crystallographic studies of proteins with methionine (Met) residues replaced with SeMet, pioneered by Hendrickson and coworkers, inspired us to replace Met with TeMet in Escherichia coli dihydrofolate reductase (DHFR). E. coli DHFR, which catalyzes the NADPH-dependent reduction of dihydrofolate to tetrahydrofolate, consists of 159 residues, 5 of which are Met. TeMet was incorporated into DHFR using the Met auxotroph, E. coli DL41, carrying the expression vector pWT8 with an IPTG inducible promoter and ampicillin resistance gene. The enzyme was purified by successive chromatography on Q-Sepharose and PHenyl Sepharose resins, yielding milligram quantities of homogeneous enzyme with a specific activity of 40 units/mg. TeMet DHFR exhibits kinetic properties similar to those of wt DHFR. Amino acid analysis indicated 3 authentic Met residues in TeMet DHFR, whereas atomic absorption spectroscopy detected 2 Te per protein molecule. Amino acid sequence analysis results suggested that only authentic Met was present in the first three Met positions (1,16,and 20). Crystals of Te-DHFR were grown in the presence of methotrexate from PEG 4000 and were isomorphous with wt-DHFR crystals grown from ethanol. Difference Fourier maps and restrained least-squares refinement show very little, if any, Te in the first three Met positions: Met1, Met16, and Met20, whereas the occupancy of Te in positions 42 and 92 is 0.64. Apparently, the process of folding, subsequent purification, and crystallization select DHFR molecules with Te in Met42 and Met92. Replacing Met with TeMet provides an internal probe that should facilitate structural and mechanistic studies of proteins