Polyhydroxybutyrate (PHB) synthases catalyze the conversion of beta-hydroxybutyryl coenzyme A (HBCoA) to PHB. These enzymes require an active site cysteine nucleophile for covalent catalysis. A protein BLASTp search using the Class III Chromatium vinosum synthase sequence reveals high homology to prokaryotic lipases whose crystal structures are known. The homology is very convincing in the alpha-beta-elbow (with the active site nucleophile)-alpha-beta structure, residues 131-175 of the synthase. A conserved histidine of the Class III PHB synthases aligns with the active site histidine of the lipases using the ClustalW algorithm. This is intriguing as this histidine is approximately 200 amino acids removed in sequence space from the catalytic nucleophile. Different threading algorithms suggest that the Class III synthases belong to the alpha/beta hydrolase superfamily which includes prokaryotic lipases. Mutagenesis studies were carried out on C. vinosum synthase C149, H331, H303, D302, and C130 residues. These studies reveal that H331 is the general base catalyst that activates the nucleophile, C149, for covalent catalysis. The model indicates that C130 is not involved in catalysis as previously proposed [Müh, U., Sinskey, A. J., Kirby, D. P., Lane, W. S., and Stubbe, J. (1999) Biochemistry 38, 826-837]. Studies with D302 mutants suggest D302 functions as a general base catalyst in activation of the 3-hydroxyl of HBCoA (or a hydroxybutyrate acyl enzyme) for nucleophilic attack on the covalently linked thiol ester intermediate. The relationship of the lipase model to previous models based on fatty acid synthases is discussed.

A total of four pathways are known for the catabolism by microorganisms of gentisate (2,5-dihydroxybenzoate) and homogentisate (2,5-dihydroxyphenylacetate). Both of these dihydric phenols can be degraded by either a glutathionedependent or a glutathione-independent reaction sequence. We found that it is not always possible to unequivocally assign glutathione dependence or independence to a particular catabolic sequence by using the well-established spectrophotometric assays at 330 nm (gentisate pathway) or 320 nm (homogentisate pathway). This paper reports a modification of the classical spectrophotometric assays that allowed an unequivocal differentiation between glutathione-dependent and glutathione-independent pathways, even when crude cell extracts contained significant quantities of cell-derived, reduced glutathione. This was accomplished by performing assays in the presence of an approximately 10-3 M solution of the sulfhydryl-binding agent N-ethylmaleimide. A number of para-dihydric phenols are known to be substrates for ring-fission dioxygenases (7). The two most frequently encountered examples have been gentisate (2,5-dihydroxybenzoate) and homogentisate (2,5-dihydroxyphenylacetate). Figure 1 illustrates the catabolic sequences described for the dissimilation of these two compounds, both of which can be degraded by glutathione-dependent or glutathione-independent routes. These pathways are usually differentiated spectrophotometrically, since the ring-fission products (compounds II and VIII, Fig. 1) have characteristic ultraviolet-visible absorption spectra (12, 13). Figure 2 illustrates the spectral changes that are characteristic of each of the pathways of Fig. 1. Crude cell extracts usually do not contain sufficient reduced glutathione (GSH) to catalyze GSH-dependent reactions such as those shown in steps 2 and 6 of Fig. 1. To observe further degradation of maleylpyruvate or maleylacetoacetate via sequences A and C (Fig. 1), GSH must be added to supplement reaction mixtures. This requirement for added GSH usually allows an easy differentiation between GSH-dependent and GSH-independent pathways (1, 3-6, 8-10, 13). In our recent investigations of gentisate and homogentisate catabolism by bacteria, we found that spectral changes at 330 or 320 nm sometimes are not as simple as the classic patterns of Fig. 2. We have frequently encountered spectral changes such as those illustrated in

A strain of Bacillus brevis isolated from a polluted section of the Mississippi River was shown to utilize 5-chloro-2-hydroxybenzoate (5-chlorosalicylate) as a sole source of carbon and energy. Enzymic analyses of cell-free extracts prepared from 5-chlorosalicylate-grown cells demonstrated that the initial step in the pathway involved cleavage of the aromatic ring between Cl and C2 by a specific 5-chlorosalicylate 1,2-dioxygenase. Loss of chloride from the growth substrate occurred after ring fission and was probably enzyme mediated. An intermediate chlorolactone apparently lost chloride by enzymatic hydrolysis with formation of maleylpyruvate. Maleylpyruvate was further degraded by both glutathione-dependent and glutathione-independent mechanisms, with these reactions being identical to the terminal reactions of the gentisate pathway. It was suggested that this novel 5-chlorosalicylate pathway may have evolved by recruitment of enzymes from an ancestral gentisate pathway.

No abstract