E0023

Structural Genomics of Short Chain Oxido Reductase Enzymes. William L. Duax^1,2, Anthony Addlagatta⁴, R. Huether¹, L. Habegger¹, Vladimir Pletnev^1,3 and C. M. Weeks¹, ¹Structural Biology Dept., Hauptman-Woodward Inst., 73 High St., Buffalo, NY 14203, ²SUNY, Dept. of Structural Biology, Buffalo, NY, ³Shemyakin-Ovchinnikov Institute, Moscow, Russian Federation and ⁴U. Oregon, Eugene, OR 97403.

The short chain oxidoreductase (SCOR) family of enzymes extends from bacteria to humans and includes over 2000 members identified in sequenced genomes. 300 of these enzymes have been functionally characterized, and the crystal structures of three dozen are known. On the basis of amino acid sequence alignments we are characterizing the cofactor, function, aggregation state and substrate of 1700 hypothetical members of the family. A search for covariance of residues in the SCOR family detected ten residues that define the substrate binding pocket of the β-keto acyl carrier protein reductase (ACPR) subfamily of the SCOR enzymes. Nucleic acid sequence analysis reveals that 25% of the SCOR genes have an antisense open reading frame overlapping the entire sense gene. Furthermore 5% of them have a third frame shift ORF. Analysis of 325 SCOR genes having double open reading frames (DORFs) and 82 having triple open reading frames (TORFs) reveal a remarkable codon bias. Over 85% of the 250 amino acids in the proteins encoded by these genes are coded for by the GC rich codons. 14 codons whose definitions vary with species are rarely used in the TORFs. These and other data suggest that the SCOR families of enzymes diverged from a common ancestor that evolved before the AT rich half of the genetic code was defined. This work is supported by NIH Grant No. DK26546.