Degenerate oligonucleotides based on the published Escherichia coli glutamate decarboxylase (GAD) protein sequence were used in a polymerase chain reaction to generate a DNA probe for the E. coli GAD structural gene. Southern blots showed that there were two cross-hybridizing GAD genes, and both of these were cloned and sequenced. The two GAD structural genes, designated gadA and gadB, were found to be 98% similar at the nucleotide level. Each gene encoded a 466-residue polypeptide, named, respectively, GAD alpha and GAD beta, and these differed by only five amino acids. Both GAD alpha and GAD beta contain amino acid residues which are highly conserved among pyridoxal-dependent decarboxylases, but otherwise the protein sequences were not homologous to any other known proteins. By restriction mapping and hybridization to the Kohara miniset library, the two GAD genes were located on the E. coli chromosome. gadA maps at 4046 kb and gadB at 1588 kb. Neither of these positions is in agreement with the current map position for gadS as determined by genetic means. Analysis of Southern blots indicated that two GAD genes were present in all E. coli strains examined, including representatives from the ECOR collection. However, no significant cross-hybridizing gene was found in Salmonella species. Information about the DNA sequences and map positions of gadA and gadB should facilitate a genetic approach to elucidate the role of GAD in E. coli metabolism.