3-DIMENSIONAL STRUCTURE OF ADENOSYLCOBINAMIDE KINASE ADENOSYLEOBINAMIDE PHOSPHATE GUANYLYLTRANSFERASE FROM SALMONELLA-TYPHIMURIUM DETERMINED TO 2.3 ANGSTROM RESOLUTION

The X-ray structure of adenosylcobinamide kinase/adenosylcobinamide ph osphate guanylyltransferase (CobU) from Salmonella typhimurium has bee n determined to 2.3 Angstrom,resolution. This enzyme of subunit molecu lar weight 19770 plays a central role in the assembly of the nucleotid e loop for adenosylcobalamin where it catalyzes both the phosphorylati on of the 1-amino-2-propanol side chain of the corrin ring and the sub sequent attachment of GMP to form the product adenosylcobinamide-GDP. The kinase activity is believed to be associated with a P-loop motif, whereas the transferase activity proceeds at a different site on the e nzyme via a guanylyl intermediate. The enzyme was crystallized in the space group C222(1) with unit cell dimensions of a = 96.4 Angstrom, b = 114.4 Angstrom, and c = 106.7 Angstrom, with three subunits per asym metric unit. The structure reveals that the enzyme is a molecular trim er and appears somewhat Like a propeller with overall molecular dimens ions of approximately 64 Angstrom x 77 Angstrom x 131 Angstrom. Each s ubunit consists of a single domain that is dominated by a seven-strand ed mixed beta-sheet flanked on either side by a total of five alpha-he lices and one helical turn. Six of the seven beta-strands run parallel . The C-terminal strand Lies at the edge of the sheet and runs antipar allel to the others. Interestingly, CobU displays a remarkable structu ral and topological similarity to the central domain of the RecA prote in, although the reason for this observation is unclear. The structure contains a P-loop motif located at the base of a prominent cleft form ed by the association of two subunits and is most likely the kinase ac tive site. Each subunit,of CobU contains a cis peptide bond between Gl u(80) and Cys(81) where Glu(80) faces the P-loop and might serve to co ordinate the magnesium ion of the triphosphate substrate. Interestingl y, His(46), which is the putative site for guanylylation, lies similar to 21 Angstrom from the P-loop and is solvent-exposed. This suggests that the enzyme undergoes a conformational change when the substrates bind to bring these two active sites into closer proximity.