Purification of DNA ligase II from calf thymus and preparation of rabbit antibody against calf thymus DNA ligase II

DNA ligase II has been purified about 4,000-fold to apparent homogeneity from a calf thymus extract. The ligase consists of a single polypeptide with a molecular weight of 68,000 as estimated by sodium dodecyl sulfate-polyacrylamide gel electrophoresis. On fluorography after electrophoresis, a DNA ligase-[3H]AMP complex gave a single band corresponding to a molecular weight of 68,000. The Km values of the ligase for ATP and nicked DNA (5'-phosphoryl ends) were obtained to be 40 and 0.04 microM, respectively. Antibody against calf thymus DNA ligase II was prepared by injecting the purified enzyme into a rabbit. The antibody cross-reacted with DNA ligase II but not with calf thymus DNA ligase I. DNA ligase II was not affected by antibody against calf thymus DNA ligase I with a molecular weight of 130,000 (Teraoka, H. and Tsukada, K. (1982) J. Biol. Chem. 257, 4758-4763). These results indicate that DNA ligase II (Mr = 68,000) is immunologically distinct from DNA ligase I (Mr = 130,000).     

3-Hydroxybenzoate:coenzyme A ligase and 4-coumarate: coenzyme A ligase from cultured cells of Centaurium erythraea

3-Hydroxybenzoate:coenzyme A ligase, an enzyme involved in xanthone biosynthesis, was detected in cell-free extracts from cultured cells of Centaurium erythraea Rafn. The enzyme was separated from 4-coumarate:coenzyme A ligase by fractionated ammonium sulphate precipitation and hydrophobic interaction chromatography. The CoA ligases exhibited different substrate specificities. 3-Hydroxybenzoate:coenzyme A ligase activated 3-hydroxybenzoic acid most efficiently and lacked affinity for cinnamic acids. In contrast, 4-coumarate:CoA ligase mainly catalyzed the activation of 4-coumaric acid but did not act on benzoic acids. The two enzymes were similar with respect to their relative molecular weight, their pH and temperature optima, their specific activity and the changes in their activity during cell culture growth. Received: 23 September 1996 / Accepted: 28 November 1996     

DNA ligase from mouse Ehrlich ascites tumor cells

The molecular (Mr = 120,000; s20, w = 5S) and catalytic properties (Km (ATP) = 3 microM; Km (nicked DNA) = 0.2 microM; Km (Mg2+) = 3 mM) of DNA ligase from mouse Ehrlich ascites tumor cells are similar to those of the enzymes from calf thymus and rodent liver. The activity level of DNA ligase from the tumor cells is about 10-fold higher than that from mouse liver. Immunochemical titration of DNA ligase with antibodies against the calf thymus enzyme showed that the higher level of DNA ligase activity in the tumor cells is due to an increase in enzyme quantity and not to elevation of the catalytic efficiency of the enzyme molecule. These results suggest that there is little apparent difference between the qualities of DNA ligases from the tumor cells and normal tissues of rodents and calf.     

Transport of coenzyme M (2-mercaptoethanesulfonic acid) and methylcoenzyme M [(2-methylthio)ethanesulfonic acid] in Methanococcus voltae: identification of specific and general uptake systems.

A transport system for coenzyme M (2-mercaptoethanesulfonic acid [HS-CoM]) and methylcoenzyme M [(2-(methylthio)ethanesulfonic acid (CH3-S-CoM)] in Methanococcus voltae required energy, showed saturation kinetics, and concentrated both forms of coenzyme M against a concentration gradient. Transport required hydrogen and carbon dioxide for maximal uptake. CH3-S-CoM uptake was inhibited by N-ethylmaleimide and monensin. Both HS-CoM and CH3-S-CoM uptake showed sodium dependence. In wild-type M. voltae, HS-CoM uptake was concentration dependent, with a Vmax of 960 pmol/min per mg of protein and an apparent Km of 61 microM. Uptake of CH3-S-CoM showed a Vmax of 88 pmol/min per mg of protein and a Km of 53 microM. A mutant of M. voltae resistant to the coenzyme M analog 2-bromoethanesulfonic acid (BES) showed no uptake of CH3-S-CoM but accumulated HS-CoM at the wild-type rate. While the higher-affinity uptake system was specific for HS-CoM, the lower-affinity system mediated uptake of HS-CoM, CH3-S-CoM, and BES. Analysis of the intracellular coenzyme M pools in metabolizing cells showed an intracellular HS-CoM concentration of 14.8 mM and CH3-S-CoM concentration of 0.21 mM.     

The purification and characterisation of 4-chlorobenzoate:CoA ligase and 4-chlorobenzoyl CoA dehalogenase from Arthrobacter sp. strain TM-1

4-Chlorobenzoate:CoA ligase, the first enzyme in the pathway for 4-chlorobenzoate dissimilation, has been partially purified from Arthrobacter sp. strain TM-1, by sequential ammonium sulphate precipitation and chromatography on DEAE-Sepharose and Sephacryl S-200. The enzyme, a homodimer of subunit molecular mass approximately 56 kD, is dependent on Mg2+-ATP and coenzyme A, and produces 4-chlorobenzoyl CoA and AMP. Besides Mg2+, Mn2+, Co2+, Fe2+ and Zn2+ are also stimulatory, but not Ca2+. Maximal activity is exhibited at pH 7.0 and 25 degrees C. The ligase demonstrates broad specificity towards other halobenzoates, with 4-chlorobenzoate as best substrate. The apparent Michaelis constants (Km) of the enzyme for 4-chlorobenzoate, CoA and ATP were determined as 3.5, 30 and 238 microM respectively. 4-Chlorobenzoyl CoA dehalogenase, the second enzyme, has been purified to homogeneity by sequential column chromatography on hydroxyapatite, DEAE-Sepharose and Sephacryl S-200. It is a homotetramer of 33 kD subunits with an isoelectric point of 6.4. At pH 7.5 and 30 degrees C, Km and kcat for 4-CBCoA are 9 microM and 1 s(-1) respectively. The optimum pH is 7.5, and maximal enzymic activity occurs at 45 degrees C. The properties of this enzyme are compared with those of the 4-chlorobenzoyl CoA dehalogenases from Arthrobacter sp. strain 4-CB1 and Pseudomonas sp. strain CBS-3, which differ variously in their N-terminal amino acid sequences, optimal pH values, pI values and/or temperatures of maximal activity.     

Purification and characterization of benzoate-coenzyme A ligase and 2-aminobenzoate-coenzyme A ligases from a denitrifying Pseudomonas sp.

The enzymes catalyzing the formation of coenzyme A (CoA) thioesters of benzoate and 2-aminobenzoate were studied in a denitrifying Pseudomonas sp. anaerobically grown with these aromatic acids and nitrate as sole carbon and energy sources. Three different rather specific aromatic acyl-CoA ligases, E1, E2, and E3, were found which catalyze the formation of CoA thioesters of benzoate, fluorobenzoates, and 2-aminobenzoate. ATP is cleaved into AMP and pyrophosphate. The enzymes were purified, their N-terminal amino acid sequences were determined, and their catalytic and molecular properties were studied. Cells anaerobically grown on benzoate and nitrate contain one CoA ligase (AMP forming) for benzoic acid (E1). It is a homodimer of Mr 120,000 which prefers benzoate as a substrate but shows some activity also with 2-aminobenzoate and fluorobenzoates, although with lower Km. Cells anaerobically grown on 2-aminobenzoate and nitrate contain three different CoA ligases for aromatic acids. The first one is identical with benzoate-CoA ligase (E1). The second enzyme is a 2-aminobenzoate-CoA ligase (E2). It is a monomer of Mr 60,000 which prefers 2-aminobenzoate but also activates benzoate, fluorobenzoates and, less effectively, 2-methylbenzoate, with lower affinities to the latter substrates. The enzymes E1 and E2 have similar activity levels; a third minor CoA ligase activity is due to a different 2-aminobenzoate-CoA ligase. The enzyme (E3) is a monomer of Mr, 65,000 which 2-aminobenzoate pathway (U. Altenschmidt, C. Eckerskorn, and G. Fuchs, Eur. J. Biochem. 194:647-653, 1990); apparently, it is not completely repressed under anaerobic conditions and therefore also is induced to a small extent by 2-aminobenzoate under anaerobic growth conditions.     

Benzoate-coenzyme A ligase, encoded by badA, is one of three ligases able to catalyze benzoyl-coenzyme A formation during anaerobic growth of Rhodopseudomonas palustris on benzoate.

The first step of anaerobic benzoate degradation is the formation of benzoyl-coenzyme A by benzoate-coenzyme A ligase. This enzyme, purified from Rhodopseudomonas palustris, is maximally active with 5 microM benzoate. To study the molecular basis for this reaction, the benzoate-coenzyme A ligase gene (badA) was cloned and sequenced. The deduced amino acid sequence of badA showed substantial similarity to other coenzyme A ligases, with the highest degree of similarity being that to 4-hydroxybenzoate-coenzyme A ligase (50% amino acid identity) from R. palustris. A badA mutant that was constructed had barely detectable levels of ligase activity when cell extracts were assayed at 10 microM benzoate. Despite this, the mutant grew at wild-type rates on benzoate under laboratory culture conditions (3 mM benzoate), and mutant cell extracts had high levels of ligase activity when assayed at a high concentration of benzoate (1 mM). This suggested that R. palustris expresses, in addition to BadA, a benzoate-activating enzyme(s) with a relatively low affinity for benzoate. A possible role of 4-hydroxybenzoate-coenzyme A ligase (encoded by hbaA) in this capacity was investigated by constructing a badA hbaA double mutant. Although the double mutant grew more slowly on benzoate than badA cells, growth rates were still significant, suggesting the involvement of a third enzyme in benzoate activation. Competition experiments involving the addition of a small amount of cyclohexanecarboxylate to ligase assay mixtures implicated cyclohexanecarboxylate-coenzyme A ligase as being this third enzyme. These results show that wild-type R. palustris cells synthesize at least three enzymes that can catalyze the initial step in anaerobic benzoate degradation during growth on benzoate. This observation supports previous suggestions that benzoyl-coenzyme A formation plays a central role in anaerobic aromatic compound biodegradation.     

Isolation and characterization of the three polypeptide components of 4-chlorobenzoate dehalogenase from Pseudomonas sp. strain CBS-3.

The three genes encoding the 4-chlorobenzene dehalogenase polypeptides were excised from a Pseudomonas sp. CBS-3 DNA fragment and separately cloned and expressed in Escherichia coli. The three enzymes were purified from the respective subclones by using an ammonium sulfate precipitation step followed by one or two column chromatographic steps. The 4-chlorobenzoate:coenzyme A ligase was found to be a homodimer (57-kDa subunit size), to require Mg2+ (Co2+ and Mn2+ are also activators) for activity, and to turn over MgATP (Km = 100 microM), coenzyme A (Km = 80 microM), and 4-chlorobenzoate (Km = 9 microM) at a rate of 30 s-1 at pH 7.5 and 25 degrees C. Benzoate, 4-bromobenzoate, 4-iodobenzoate, and 4-methylbenzoate were shown to be alternate substrates while 4-hydroxybenzoate, 4-aminobenzoate, 2-aminobenzoate, 2,3-dihydroxybenzoate, 4-coumarate, palmate, laurate, caproate, butyrate, and phenylacetate were not substrate active. The 4-chlorobenzoate-coenzyme A dehalogenase was found to be a homotetramer (30 kDa subunit size) to have a Km = 15 microM and kcat = 0.3 s-1 at pH 7.5 and 25 degrees C and to be catalytically inactive toward hydration of crotonyl-CoA, alpha-methylcrotonyl-CoA, and beta-methylcrotonyl-CoA. The 4-hydroxybenzoate-coenzyme A thioesterase was shown to be a homotetramer (16 kDa subunit size), to have a Km = 5 microM and kcat = 7 s-1 at pH 7.5 and 25 degrees C, and to also catalyze the hydrolyses of benzoyl-coenzyme A and 4-chlorobenzoate-coenzyme A. Acetyl-coenzyme A, hexanoyl-coenzyme A, and palmitoyl-coenzyme A were not hydrolyzed by the thioesterase.     

Involvement of coenzyme A thioesters in anaerobic metabolism of 4-hydroxybenzoate by Rhodopseudomonas palustris.

The initial steps of anaerobic 4-hydroxybenzoate degradation were studied in whole cells and cell extracts of the photosynthetic bacterium Rhodopseudomonas palustris. Illuminated suspensions of cells that had been grown anaerobically on 4-hydroxybenzoate and were assayed under anaerobic conditions took up [U-14C]4-hydroxybenzoate at a rate of 0.6 nmol min-1 mg of protein-1. Uptake occurred with high affinity (apparent Km = 0.3 microM), was energy dependent, and was insensitive to external pH in the range of 6.5 to 8.2 Very little free 4-hydroxybenzoate was found associated with cells, but a range of intracellular products was formed after 20-s incubations of whole cells with labeled substrate. When anaerobic pulse-chase experiments were carried out with cells incubated on ice or in darkness, 4-hydroxybenzoyl coenzyme A (4-hydroxybenzoyl-CoA) was formed early and disappeared immediately after addition of excess unlabeled substrate, as would be expected of an early intermediate in 4-hydroxybenzoate metabolism. A 4-hydroxybenzoate-CoA ligase activity with an average specific activity of 0.7 nmol min-1 mg of protein-1 was measured in the soluble protein fraction of cells grown anaerobically on 4-hydroxybenzoate. 4-Hydroxybenzoyl-CoA was the sole product formed from labeled 4-hydroxybenzoate in the ligase reaction mixture. 4-Hydroxybenzoate uptake and ligase activities were present in cells grown anaerobically with benzoate, 4-hydroxybenzoate, and 4-aminobenzoate and were not detected in succinate-grown cells. These results indicate that the high-affinity uptake of 4-hydroxybenzoate by R. palustris is due to rapid conversion of the free acid to its CoA derivative by a CoA ligase and that this is also the initial step of anaerobic 4-hydroxybenzoate degradation.     

Inhibition of acetoacetyl-CoA synthetase from rat liver by fatty acyl-CoAs

The activity of acetoacetyl-CoA synthetase from rat liver was found to be negatively regulated by coenzyme A, fatty acyl-CoAs and acetoacetyl-CoA in vitro. With increasing concentrations of coenzyme A (substrate inhibition occurring at concentrations higher than 50 microM) the pH optimum shifted toward the acidic side (7.5-8.5 with 5 microM coenzyme A and 6.5-7.0 with 500 microM coenzyme A), in parallel with progressively decreasing enzyme activity. Fatty acyl-CoAs of various chain lengths dose-dependently inhibited acetoacetyl-CoA synthetase from rat liver, but much less effectively a similar enzyme from a bacterium, Zoogloea ramigera I-16-M. Palmitoyl-CoA, the most potent inhibitor of the rat liver enzyme, with an apparent Ki value of 9.8 microM, apparently inhibited the enzyme below its critical micellar concentration, not due to its detergent action. Acetoacetyl-CoA showed product inhibition with a Ki value of 15 microM. These results suggest a possible physiological regulation mechanism for this enzyme with respect to fatty acid biosynthesis.     

Evidence for the regulation of fatty acid utilization in human sperm by docosahexaenoic acid

The effects of stearic (18:0), linolenic (18:3), and docosahexaenoic (22:6) acids on palmitoyl coenzyme A (CoA) formation by a long-chain fatty acid:CoASH ligase (adenosine monophosphate) (E. C. 6.2.1.3-enriched fraction from human spermatozoa were studied. Both 18:0 and 18:3 were competitive inhibitors for palmitic (16:0) acid activation with Kis of 17.7 and 5.7 microM, respectively. In contrast, 22:6 was a noncompetitive inhibitor demonstrating a Ki of 9.5 microM. These data coupled with previous studies support the conclusion that 16:0, 18:0, and 18:3 and other saturated and unsaturated fatty acids are activated by the same ligase enzyme in sperm. Although the kinetics and interactions of 22:6 are unique compared to the other fatty acids found in sperm phospholipids, we cannot discern from our data if it is activated by a separate enzyme. We propose that 22:6, or a metabolite of 22:6, may regulate free fatty acid utilization in human sperm and that this hypothesis may provide an enzymatic explanation for the changes observed in phospholipid-bound fatty acids during the epididymal maturation of sperm.     

DNA ligase from Drosophila melanogaster embryos. Substrate specificity and mechanism of action

DNA ligase has been purified to homogeneity from 6-12 h Drosophila melanogaster embryos (Rabin, B. A., Hawley, R. S., and Chase, J. W. (1986) J. Biol. Chem. 261, 10637-10645). This enzyme had an apparent Km for ATP of 1.6 microM. Of a variety of nucleotides tested, only adenosine 5'-O-(3-thio)triphosphate could substitute for ATP in the joining reaction. The enzyme was competitively inhibited by dATP, with an apparent Ki of 2.3 microM. The apparent Km for DNA using p(dT)20 annealed with poly(dA) as substrate was 1.0 microM. Studies utilizing synthetic homopolymers showed that in addition to joining DNA to DNA, this enzyme could join the 5'-phosphoryl termini of RNA to the 3'-hydroxyl termini of DNA or RNA, when they were annealed with DNA. In addition, p(dT)7U could be joined when annealed with poly(dA). No joining was detected when RNA served as the template. Drosophila DNA ligase also catalyzed the joining of oligonucleotides containing a single mismatched nucleotide at their 3'-hydroxyl termini, as well as DNA containing short, complementary 5'-protruding ends, and in the presence of polyethylene glycol 6000, blunt-ended duplex DNA. The overall reaction mechanism was shown to be identical to that of the homologous prokaryotic DNA ligases. The joining reactions catalyzed by the Drosophila and T4 DNA ligases were shown to be reversible. Incubation of superhelical closed circular DNA molecules with the purified enzymes and AMP resulted in the production of a population of DNA molecules which had lost most, if not all, of their superhelical density.     

Synthesis and properties of 2-azido-NAD+. A study of interaction with glutamate dehydrogenase

A photoactive coenzyme analog of NAD+ has been synthesized by chemically coupling [32P]2-azido-AMP and NMN to produce [32P]nicotinamide 2-azidoadenosine dinucleotide (2-azido-NAD+). The utility of 2-azido-NAD+ as an effective active-site-directed photoprobe was demonstrated using bovine liver glutamate dehydrogenase as a model enzyme. In the absence of ultraviolet light, 2-azido-NAD+ is a substrate for this enzyme. Photoincorporation of probe was saturable with two different apparent dissociation constants of 10 microM and 40 microM. Protection of photoinsertion was seen with the natural substrate NAD+ with apparent dissociation constants of less than 5 microM and 25 microM. This observation may be explained on the basis of negative cooperative interaction between the subunits. The photoinsertion of 2-azido-NAD+ was increased by GTP and decreased by ADP in accordance with their known effects on NAD+ binding. When the enzyme was covalently modified by photolysis in the presence of saturating amounts of photoprobe, an approximately 40% inhibition of the enzyme activity was observed. These results demonstrate that the photoaffinity coenzyme analog has potential application as a probe to characterize NAD(+)-binding proteins and to identify the active sites of these proteins.     

Purification of rabbit myocardial cytosolic acyl-CoA hydrolase, identity with lysophospholipase, and modulation of enzymic activity by endogenous cardiac amphiphiles

Rabbit myocardial cytosolic acyl coenzyme A (acyl-CoA) hydrolase activity was purified to near-homogeneity by ammonium sulfate precipitation and ion-exchange, gel filtration, chromatofocusing, and hydroxylapatite chromatographies. Kinetic analysis of the purified protein demonstrated a maximum velocity of 24 mumol/(mg . min) and an apparent Michaelis constant of 50 microM. Cytosolic acyl-CoA hydrolase and lysophospholipase activities cochromatographed in every fraction of every step. The purified protein was a single band (Mr 23 000) after sodium dodecyl sulfate-polyacrylamide gel electrophoresis and silver staining. These results suggest that cytosolic lysophospholipase and palmitoyl-CoA hydrolase activities are catalyzed by a single polypeptide with dual activities. Palmitoyl-CoA competitively inhibited lysophospholipase activity (Ki = 4 microM). Low concentrations (20 microM) of lysophosphatidylcholine or L-palmitoylcarnitine increased palmitoyl-CoA hydrolase activity at low palmitoyl-CoA concentrations but had little effect at high concentrations of palmitoyl-CoA. In contrast, high concentrations (100 microM) of lysophosphatidylcholine or L-palmitoylcarnitine inhibited palmitoyl-CoA hydrolase activity. The results suggest that interactions between endogenous cardiac amphiphiles and palmitoyl-CoA hydrolase contribute to the regulation of intracellular long-chain acyl-CoA concentrations and therefore potentially modulate fluxes of fatty acid through several biochemical pathways.     

Third form of calcium-activated neutral proteinase from calf brain: purification, partial characterization and comparison of properties with other forms

A third form (CANP3) of calcium-activated neutral proteinase (CANP) has been purified, 3900-fold, to near homogeneity from calf brain cortex. The purification procedure is based on the one recently developed for the purification of CANP1 and CANP2. The molecular weight of CANP3, as judged on SDS-polyacrylamide gel electrophoresis was Mr 78,000. A protein with an apparent Mr 17,000 co-purified with the proteinase. At neutral pH (7.2), it was maximally active at 260 microM CaCl2. In the presence of CaCl2, CANP1 and CANP3 were autolyzed very rapidly, whereas the autolysis of CANP2 was slow and gradual. The autolyzed CANP1 and CANP3 responded differently to CaCl2; CANP1 lost activity completely, whereas CANP3 was fully active at 0.5 microM CaCl2. Despite the opposite behavior of these proteinases in the presence of Ca2+, no significant differences in the peptide maps of the three proteinases were observed. Neurofilaments, neurotubules and myelin basic protein (MBP) were degraded by each of the proteinases. Monoclonal antibodies raised against CANP2 reacted almost equally with CANP1 and CANP3. As with CANP1 and CANP2, leupeptin and sulfhydryl-modifying compounds, NEM and iodoacetic acid, inhibited the activity of CANP3.     

Capacitative calcium influx in human epithelial breast cancer and non-tumorigenic cells occurs through Ca2+ entry pathways with different permeabilities to divalent cations

The operation of capacitative Ca(2+) entry (CCE) in human breast cancer (SKBR3) and non-tumorigenic (HBL100) cell lines was investigated as an alternative Ca(2+) entry route in these cells. Ca(2+) readdition after thapsigargin-induced store depletion showed activation of CCE in both cell lines. SKBR3 cells exhibited retarded store depletion and CCE decay kinetics compared to the non-tumorigenic HBL100 cells, suggesting alterations in Ca(2+) homeostasis. CCE was also highly permeable to Mn(2+) and to a lesser extent to Sr(2+), but not to Ba(2+). In HBL100 cells, CCE is contributed (30%) by a Ca(2+)/Mn(2+) permeable route insensitive to low (1 microM) Gd(3+) and a Ca(2+)/Sr(2+)/Mn(2+) permeable non-selective pathway (70%) sensitive to 1 microM Gd(3+). In SKBR3 cells, the relative contribution to CCE of both routes was opposite to that in non-tumorigenic cells.     

4-Hydroxybenzoate-coenzyme A ligase from Rhodopseudomonas palustris: purification, gene sequence, and role in anaerobic degradation.

Anaerobic metabolism of most aromatic acids is initiated by coenzyme A thioester formation. Rhodopseudomonas palustris grows well under anaerobic, phototrophic conditions with many aromatic acids, including benzoate and 4-hydroxybenzoate, as a carbon source. A coenzyme A ligase that reacts with 4-hydroxybenzoate was purified from 4-hydroxybenzoate-grown cells of R. palustris. This enzyme required MgATP, reduced coenzyme A, and 4-hydroxybenzoate, benzoate, or cyclohex-1,4-dienecarboxylate for optimal activity but also used phosphopantetheine, cyclohex-2,5-dienecarboxylate, and 4-fluorobenzoate at lower rates. The 4-hydroxybenzoate-coenzyme A ligase differed in molecular characteristics from a previously described benzoate-coenzyme A ligase from R. palustris, and the two ligases did not cross-react immunologically. The gene encoding the 4-hydroxybenzoate enzyme was cloned and sequenced. The deduced gene product showed about 20% amino acid identity with bacterial coenzyme A ligases involved in aerobic degradation of aromatic acids. An R. palustris mutant carrying a disrupted 4-hydroxybenzoate-coenzyme A ligase gene was unable to grow with 4-hydroxybenzoate under anaerobic conditions, indicating that the enzyme is essential for anaerobic degradation of this compound.     

Purification, properties, and N-terminal amino acid sequence of homogeneous Escherichia coli 2-amino-3-ketobutyrate CoA ligase, a pyridoxal phosphate-dependent enzyme

Starting with 100 g (wet weight) of a mutant of Escherichia coli K-12 forced to grow on L-threonine as sole carbon source, we developed a 6-step procedure that provides 30-40 mg of homogeneous 2-amino-3-ketobutyrate CoA ligase (also called aminoacetone synthetase or synthase). This ligase, which catalyzes the cleavage/condensation reaction between 2-amino-3-ketobutyrate (the presumed product of the L-threonine dehydrogenase-catalyzed reaction) and glycine + acetyl-CoA, has an apparent molecular weight approximately equal to 85,000 and consists of two identical (or nearly identical) subunits with Mr = 42,000. Computer analysis of amino acid composition data, which gives the best fit nearest integer ratio for each residue, indicates a total of 387 amino acids/subunit with a calculated Mr = 42,093. Stepwise Edman degradation provided the N-terminal sequence of the first 21 amino acids. It is a pyridoxal phosphate-dependent enzyme since (a) several carbonyl reagents caused greater than 90% loss of activity, (b) dialysis against buffer containing hydroxylamine resulted in 89% loss of activity coincident with an 86% decrease in absorptivity at 428 nm, (c) incubation of the apoenzyme with 20 microM pyridoxal phosphate showed a parallel recovery (greater than 90%) of activity and 428-nm absorptivity, and (d) reduction of the holoenzyme with NaBH4 resulted in complete inactivation, disappearance of a new absorption maximum at 333 nm. Strict specificity for glycine is shown but acetyl-CoA (100%), n-propionyl-CoA (127%), or n-butyryl-CoA (16%) is utilized in the condensation reaction. Apparent Km values for acetyl-CoA, n-propionyl-CoA, and glycine are 59 microM, 80 microM, and 12 mM, respectively; the pH optimum = 7.5. Added divalent metal ions or sulfhydryl compounds inhibited catalysis of the condensation reaction.     

Substrate specificity of the 3-methylcrotonyl coenzyme A (CoA) and geranyl-CoA carboxylases from Pseudomonas aeruginosa

Biotin-containing 3-methylcrotonyl coenzyme A (MC-CoA) carboxylase (MCCase) and geranyl-CoA (G-CoA) carboxylase (GCCase) from Pseudomonas aeruginosa were expressed as His-tagged recombinant proteins in Escherichia coli. Both native and recombinant MCCase and GCCase showed pH and temperature optima of 8.5 and 37 degrees C. The apparent K(0.5) (affinity constant for non-Michaelis-Menten kinetics behavior) values of MCCase for MC-CoA, ATP, and bicarbonate were 9.8 microM, 13 microM, and 0.8 microM, respectively. MCCase activity showed sigmoidal kinetics for all the substrates and did not carboxylate G-CoA. In contrast, GCCase catalyzed the carboxylation of both G-CoA and MC-CoA. GCCase also showed sigmoidal kinetic behavior for G-CoA and bicarbonate but showed Michaelis-Menten kinetics for MC-CoA and the cosubstrate ATP. The apparent K(0.5) values of GCCase were 8.8 microM and 1.2 microM for G-CoA and bicarbonate, respectively, and the apparent K(m) values of GCCase were 10 microM for ATP and 14 microM for MC-CoA. The catalytic efficiencies of GCCase for G-CoA and MC-CoA were 56 and 22, respectively, indicating that G-CoA is preferred over MC-CoA as a substrate. The enzymatic properties of GCCase suggest that it may substitute for MCCase in leucine catabolism and that both the MCCase and GCCase enzymes play important roles in the leucine and acyclic terpene catabolic pathways.     

4-hydroxybenzoyl coenzyme A reductase (dehydroxylating) is required for anaerobic degradation of 4-hydroxybenzoate by Rhodopseudomonas palustris and shares features with molybdenum-containing hydroxylases.

The anaerobic degradation of 4-hydroxybenzoate is initiated by the formation of 4-hydroxybenzoyl coenzyme A, with the next step proposed to be a dehydroxylation to benzoyl coenzyme A, the starting compound for a central pathway of aromatic compound ring reduction and cleavage. Three open reading frames, divergently transcribed from the 4-hydroxybenzoate coenzyme A ligase gene, hbaA, were identified and sequenced from the phototrophic bacterium Rhodopseudomonas palustris. These genes, named hbaBCD, specify polypeptides of 17.5, 82.6, and 34.5 kDa, respectively. The deduced amino acid sequences show considerable similarities to a group of hydroxylating enzymes involved in CO, xanthine, and nicotine metabolism that have conserved binding sites for [2Fe-2S] clusters and a molybdenum cofactor. Cassette disruption of the hbaB gene yielded a mutant that was unable to grow anaerobically on 4-hydroxybenzoate but grew normally on benzoate. The hbaB mutant cells did not accumulate [14C]benzoyl coenzyme A during short-term uptake of [14C]4-hydroxybenzoate, but benzoyl coenzyme A was the major radioactive metabolite formed by the wild type. In addition, crude extracts of the mutant failed to convert 4-hydroxybenzoyl coenzyme A to benzoyl coenzyme A. This evidence indicates that the hbaBCD genes encode the subunits of a 4-hydroxybenzoyl coenzyme A reductase (dehydroxylating). The sizes of the specified polypeptides are similar to those reported for 4-hydroxybenzoyl coenzyme A reductase isolated from the denitrifying bacterium Thauera aromatica. The amino acid consensus sequence for a molybdenum cofactor binding site is in HbaC. This cofactor appears to be an essential component because anaerobic growth of R. palustris on 4-hydroxybenzoate, but not on benzoate, was retarded unless 0.1 microM molybdate was added to the medium. Neither tungstate nor vanadate replaced molybdate, and tungstate competitively inhibited growth stimulation by molybdate.