Escherichia coli protease I is assayed as an esterase active with certain synthetic model chymotrypsin substrates. However, the gene encoding protease I has the same DNA sequence and genomic location as tesA, a gene that encodes E. coli thioesterase I. We report that both hydrolase activities utilize the same active site and demonstrate that the protein functions as a thioesterase in vivo.Escherichia coli protease I was first reported by Pacaud and Uriel (13). The partially purified enzyme was reported to convert the native form of polynucleotide phosphorylase into a smaller, active form of the enzyme. Later, Pacaud et al. (12) further purified protease I by using N-acetyl-DL-phenylalanine-2-naphthyl ester (NAPNE) as a model substrate and reported cleavage of oxidized bovine insulin by the purified enzyme. Kowit et al. (7) reported the isolation of an E. coli mutant deficient in NAPNE hydrolysis and also reported that purified protease I failed to hydrolyze casein. Salmonella typhimurium mutants unable to cleave NAPNE were isolated by Miller et al. (9), who mapped the gene and designated it apeA. Recently, Ichihara et al. (5) screened the Clarke-Carbon clone bank for strains that overproduced an esterase activity that hydrolyzed another synthetic protease I substrate, N-benzyloxycarbonyl-Lphenylalanine ,B-naphthyl ester. One class of these clones was shown to encode protease I, and the gene was subcloned, sequenced, and designated apeA by analogy with the S. typhimurium mutant gene. They also purified the enzyme to homogeneity and showed that its properties were those reported by Pacaud and coworkers (12, 13).In 1967, Kass and coworkers reported the presence of thioesterase activity in extracts of E. coli (6). Two distinct enzymes that catalyze the hydrolysis of fatty acyl coenzyme A (acyl-CoA) substrates were detected after fractionation of cell extracts (1). Barnes and Wakil partially purified and characterized the lower-molecular-weight enzyme, thioesterase I (2). Recently, we reported the cloning and DNA sequencing of the tesA gene encoding E. coli thioesterase I as well as studies of the physiological role of the enzyme (4).A comparison of our data (4) and those of Ichihara et al. (5) showed that the apeA and tesA genes have identical nucleotide sequences and the same location on the E. coli chromosome. Moreover, the proteins purified by following either enzyme activity are both of periplasmic origin and have identical molecular weights and amino-terminal sequences (4, 5). Thus, protease I and thioesterase I are the same protein, which raises the question of which activity (if not both or neither) is physiologically relevant. We report that the same active site is used for both fatty acid and amino acid substrates and show that overproduction of this protein results in accumulation of free fatty acids, indicating that the protein has thioesterase activity in vivo.First, we performed substrate competition assays with the * Corresponding author. Phone: (217) 333-0425. Fax: (217) 244-6697. E-mail: John_Crona...