Search

Crc (catabolite repression control) protein of Pseudomonas aeruginosa has shown to be involved in carbon regulation of several pathways. In this study, the role of Crc in catabolite repression control has been studied in Pseudomonas putida. The bkd operons of P. putida and P. aeruginosa encode the inducible multienzyme complex branched-chain keto acid dehydrogenase, which is regulated in both species by catabolite repression. We report here that this effect is mediated in both species by Crc. A 13-kb cloned DNA fragment containing the P. putida crc gene region was sequenced. Crc regulates the expression of branched-chain keto acid dehydrogenase, glucose-6-phosphate dehydrogenase, and amidase in both species but not urocanase, although the carbon sources responsible for catabolite repression in the two species differ. Transposon mutants affected in their expression of BkdR, the transcriptional activator of the bkd operon, were isolated and identified as crc and vacB (rnr) mutants. These mutants suggested that catabolite repression in pseudomonads might, in part, involve control of BkdR levels.

The effect of growth in 2xYT medium on catabolite repression control in Pseudomonas putida has been investigated using the bkd operon, encoding branched-chain keto acid dehydrogenase. Crc (catabolite repression control protein) was shown to be responsible for repression of bkd operon transcription in 2xYT. BkdR levels were elevated in a P. putida crc mutant, but bkdR transcript levels were the same in both wild type and crc mutant. This suggests that the mechanism of catabolite repression control in rich media by Crc involves posttranscriptional regulation of the bkdR message.

BkdR can be isolated in nearly pure form as a tetramer by this procedure, which involves hyperexpressing bkdR from a plasmid, purification by chromatography on DEAE-Sepharose CL-6B, heparin-Sepharose CL-6B, and dialysis to precipitate BkdR. BkdR is relatively insoluble in aqueous buffers but can be kept in solution in buffer with 50% (v/v) glycerol and 0.2 M NaCl. Cultures of E. coli DH5 alpha (pJRS119) should be maintained at 30 degrees to promote plasmid stability. Because BkdR is prone to form intermolecular disulfide bonds, buffers for SDS-PAGE should contain fresh 0.5% (v/v) 2-mercaptoethanol.

The family of giant multienzyme complexes metabolizing pyruvate, 2-oxoglutarate, branched-chain 2-oxo acids or acetoin contains several of the largest and most sophisticated protein assemblies known, with molecular masses between 4 and 10 million Da. The principal enzyme components, E1, E2 and E3, are present in numerous copies and utilize multiple cofactors to catalyze a directed sequence of reactions via substrate channeling. The crystal structure of a heterotetrameric (alpha2beta2) E1, 2-oxoisovalerate dehydrogenase from Pseudomonas putida, reveals a tightly packed arrangement of the four subunits with the beta2-dimer held between the jaws of a 'vise' formed by the alpha2-dimer. A long hydrophobic channel, suitable to accommodate the E2 lipoyl-lysine arm, leads to the active site, which contains the cofactor thiamin diphosphate (ThDP) and an inhibitor-derived covalent modification of a histidine side chain. The E1 structure, together with previous structural information on E2 and E3, completes the picture of the shared architectural features of these enormous macromolecular assemblies.

BkdR is the transcriptional activator of the bkd operon, which encodes the four proteins of the branched-chain keto acid dehydrogenase multienzyme complex of Pseudomonas putida. In this study, hydroxyl radical footprinting revealed that BkdR bound to only one face of DNA over the same region identified in DNase I protection assays. Deletions of even a few bases in the 5' region of the BkdR-binding site greatly reduced transcription, confirming that the entire protected region is necessary for transcription. In vitro transcription of the bkd operon was obtained by using a vector containing the bkdR-bkdA1 intergenic region plus the putative rho-independent terminator of the bkd operon. Substrate DNA, BkdR, and any of the L-branched-chain amino acids or D-leucine was required for transcription. Branched-chain keto acids, D-valine, and D-isoleucine did not promote transcription. Therefore, the L-branched-chain amino acids and D-leucine are the inducers of the bkd operon. The concentration of L-valine required for half-maximal transcription was 2.8 mM, which is similar to that needed to cause half-maximal proteolysis due to a conformational change in BkdR. A model for transcriptional activation of the bkd operon by BkdR during enzyme induction which incorporates these results is presented.

Reinvestigation of the transcriptional start site of the bkd operon of Pseudomonas putida revealed that the transcriptional start site was located 86 nucleotides upstream of the translational start. There was a sigma 70 binding site 10 bp upstream of the transcriptional start site. The dissociation constants for BkdR, the transcriptional activator of the bkd operon, were 3.1 x 10(-7) M in the absence of L-valine and 8.9 x 10(-8) M in the presence of L-valine. Binding of BkdR to substrate DNA in the absence of L-valine imposed a bend angle of 92 degrees in the DNA. In the presence of L-valine, the angle was 76 degrees. BkdR did not bind to either of the two fragments of substrate DNA resulting from digestion with AgeI. Because AgeI attacks between three potential BkdR binding sites, this suggests that binding of BkdR is cooperative. P. putida JS110 and JS112, mutant strains which do not express any of the components of branched-chain keto acid dehydrogenase, were found to contain missense mutations in bkdR resulting in R40Q and T22I changes in the putative helix-turn-helix of BkdR. Addition of glucose to the medium repressed expression of lacZ from a chromosomal bkdR-lacZ fusion, suggesting that catabolite repression of the bkd operon was the result of reduced expression of bkdR. These data are used to present a model for the role of BkdR in transcriptional control of the bkd operon.

BkdR is the positive transcriptional activator of the inducible bkd operon of Pseudomonas putida. Evidence is accumulating that L-branched-chain amino acids are the inducers of the operon, and the data obtained in this study show that they induce a conformational change in BkdR. Addition of L-branched-chain amino acids increased the susceptibility of BkdR to trypsin with the cleavage between Arg-51 and Gln-52 on the C-terminal side of the DNA-binding domain. L-Valine also caused an increased fluorescence emission intensity and produced significant changes in the circular dichroism spectrum of BkdR. Analytical ultracentrifugation confirmed earlier data obtained from gel filtration that BkdR was a tetramer with a Stokes radius of 32 +/- 3 A and an axial ratio of 2:1.

BkdR is the transcriptional activator of the bkd operon of Pseudomonas putida, which encodes branched chain keto acid dehydrogenase. BkdR binds to a large region of DNA between its own structural gene and the first gene of the bkd operon. The object of the present studies was to determine the stoichiometry of binding as part of an effort to understand the action of BkdR in regulation of the bkd operon. [35S]BkdR was prepared and found to be essentially 100% active in the gel shift assay. Only one complex was formed under all the conditions used. The stoichiometry of BkdR binding to its specific substrate DNA was three tetramers per mold substrate DNA. L-valine did not affect the stoichiometry although this ligand was previously shown to affect the DNase I protection pattern. The addition of nonspecific DNA to the incubation mixture also did not affect this stoichiometry.

Active E1 component of Pseudomonas putida branched-chain-oxoacid dehydrogenase was purified from P. putida strains carrying pJRS84 which contains bkdR (encoding the transcriptional activator) and bkdA1 and bkdA2 (encoding the alpha and beta subunits). Expression was inducible, however, 45-, 39- and 37-kDa proteins were produced instead of the expected 45-kDa and 37-kDa proteins. The 45-kDa protein was identified as E1 alpha and the 37-kDa and 39-kDa proteins were identified as separate translational products of bkdA2 by their N-terminal sequences. The N-terminal amino acid of the 39-kDa protein was leucine instead of methionine. The 45-, 39- and 37-kDa proteins were also produced in wild-type P.putida. Translation of bkdA1 and bkdA2 from an Escherichia coli expression plasmid produced only 45-kDa and 39-kDa proteins, with N-terminal methionine on the 39-kDa protein. The insertion of guanine residues 5' to the first ATG of bkdA2 did not affect expression of E1 beta in P. putida including the N-terminal leucine which appears to eliminate the possibility of ribosome jumping. The Z-average molecular mass of the E1 component was determined by sedimentation equilibrium to be 172 +/- 9 kDa compared to a calculated value of 166 kDa for the heterotetramer and a Stokes radius of 5.1 nm. E1 alpha Ser313, which is homologous to the phosphorylated residue of rat liver E1 alpha, was converted to alanine resulting in about a twofold increase in Km, but no change in Kcat. S315A and S319A mutations had no effect on Km or Kcat indicating that these residues do not play a major part in catalysis of E1 alpha beta 2.

The bkd operon of Pseudomonas putida consists of the structural genes encoding the components of the inducible branched-chain ketoacid dehydrogenase. BkdR, a positive regulator of the bkd operon and a homolog of Lrp of Escherichia coli is encoded by a structural gene adjacent to, and divergently transcribed from, the bkd operon of P. putida. BkdR was purified from E. coli containing bkdR cloned into pCYTEXP1, an expression vector. The molecular weight of BkdR obtained by gel filtration indicates that BkdR is a tetramer, and the abundance of BkdR in P. putida was estimated to be about 25 to 40 copies of the tetramer per cell. BkdR bound specifically to the region between bkdR and bkdA1, the latter being the first gene of the bkd operon. One BkdR-DNA complex was observed in gel mobility shift patterns. Approximately 100 bp was protected from the action of DNase I by BkdR, and the addition of L-branched-chain amino acids enhanced the appearance of hypersensitive sites in the protected region. There are four potential BkdR-DNA binding sequences in this region based on similarity to Lrp-binding consensus sequences. Like many other transcriptional activators, BkdR regulates expression of its structural gene. DNAs from several gram-negative bacteria hybridized to a probe containing bkdR, indicating the presence of bkdR-like genes in these organisms.

The recA gene of Pseudomonas putida PpG2 was cloned by complementation of the recA mutations of Escherichia coli strains DH5 alpha and HB101. The nucleotide sequence of the DNA fragment was determined and shown to contain recA and a downstream partial open reading frame. Two mutants of P. putida PpG2, strains JS387 and JS388, were constructed by insertional inactivation of recA with a tetracycline-resistance gene in both orientations. Both mutants acquired sensitivity to methyl methanesulfonate (MMS) and both failed to undergo homologous recombination. While the recA mutation of P. putida JS388 was complemented in trans by recA of P. putida, the JS387 mutant was difficult to transform and transformants exhibited varying degrees of sensitivity to MMS. Therefore, P. putida JS388 can be used as a carrier of recombinant plasmids, but JS387 is not a suitable host for this purpose.

Branched-chain keto acid dehydrogenase is a multienzyme complex which is required for the metabolism of the branched-chain amino acids in Pseudomonas putida. The structural genes encoding all four proteins of the bkd operon have been cloned, and their nucleotide sequences have been determined (G. Burns, K. T. Madhusudhan, K. Hatter, and J. R. Sokatch, p. 177-184 in S. Silver, A. M. Chakrabarty, B. Iglewski, and S. Kaplan [ed.], Pseudomonas: Biotransformations, Pathogenesis, and Evolving Biotechnology, American Society for Microbiology, Washington D.C., 1990). An open reading frame which encoded a protein with 36.5% amino acid identity to the leucine-responsive regulatory protein (Lrp) of Escherichia coli was found immediately upstream of the bkd operon. Chromosomal mutations affecting this gene, named bkdR, resulted in a loss of ability to use branched-chain amino acids as carbon and energy sources and failure to produce branched-chain keto acid dehydrogenase. These mutations were complemented in trans by plasmids which contained intact bkdR. Mutations affecting bkdR did not have any effect on transport of branched-chain amino acids or transamination. Therefore, the bkdR gene product must affect expression of the bkd operon and regulation must be positive. Mutations affecting bkdR could also be complemented by plasmids containing lrp of E. coli. This is the first instance of a Lrp-like protein demonstrated to regulate expression of an operon outside of E. coli.

The three-dimensional structure of one of the three lipoamide dehydrogenases occurring in Pseudomonas putida, LipDH Val, has been determined at 2.45 A resolution. The orthorhombic crystals, grown in the presence of 20 mM NAD+, contain 458 residues per asymmetric unit. A crystallographic 2-fold axis generates the dimer which is observed in solution. The final crystallographic R-factor is 21.8% for 18,216 unique reflections and a model consisting of 3,452 protein atoms, 189 solvent molecules and 44 NAD+ atoms, while the overall B-factor is unusually high: 47 A2. The structure of LipDH Val reveals the conformation of the C-terminal residues which fold "back" into the putative lipoamide binding region. The C-terminus has been proven to be important for activity by site-directed mutagenesis. However, the distance of the C-terminus to the catalytically essential residues is surprisingly large, over 6 A, and the precise role of the C-terminus still needs to be elucidated. In this crystal form LipDH Val contains one NAD+ molecule per subunit. Its adenine-ribose moiety occupies an analogous position as in the structure of glutathione reductase. However, the nicotinamide-ribose moiety is far removed from its expected position near the isoalloxazine ring and points into solution. Comparison of LipDH Val with Azotobacter vinelandii lipoamide dehydrogenase yields an rms difference of 1.6 A for 440 well defined C alpha atoms per subunit. Comparing LipDH Val with glutathione reductase shows large differences in the tertiary and quaternary structure of the two enzymes. For instance, the two subunits in the dimer are shifted by 6 A with respect to each other. So, LipDH Val confirms the surprising differences in molecular architecture between glutathione reductase and lipoamide dehydrogenase, which were already observed in Azotobacter vinelandii LipDH. This is the more remarkable since the active sites are located at the subunit interface and are virtually identical in all three enzymes.

A 5417-base pair (bp) region of Pseudomonas aeruginosa PAO chromosomal DNA containing the mmsAB operon and an upstream regulatory gene (mmsR) has been cloned and characterized. The operon contains two structural genes involved in valine metabolism: mmsA, which encodes methylmalonate-semialdehyde dehydrogenase; and mmsB, which encodes 3-hydroxyisobutyrate dehydrogenase. mmsA and mmsB share the same orientation and are separated by a 16-bp noncoding region. The transcriptional start site for the operon has been pinpointed to a cytidine residue located 77 bp from the translational start site of the operon. mmsR is located on the opposite strand and begins 134 bp from the translational start site of mmsA. MmsR has been identified as a member of the XylS/AraC family of transcriptional regulators and appears to act as a positive regulator of the mmsAB operon. Sequence comparison of MmsA to other proteins in the data bases revealed that MmsA belongs to the aldehyde dehydrogenase (NAD+) superfamily. MmsB shares a 44% amino acid identity with 3-hydroxyisobutyrate dehydrogenase from rat liver. Mutants with insertionally inactivated mmsR, mmsA, and mmsB grow slowly on valine/isoleucine medium and exhibit reduced enzyme activity in cell-free extracts compared to P. aeruginosa PAO.

The third lipoamide dehydrogenase structural gene of Pseudomonas putida, lpd3, was isolated from a library of P. putida PpG2 DNA cloned in Escherichia coli TB1. The nucleotide sequence of lpd3 and its flanking regions indicate that lpd3 is not part of an operon, which is unique for a prokaryotic lipoamide dehydrogenase. An open reading frame was found 207 bases upstream from the start of transcription, but is encoded on the strand opposite lpd3. There is no evidence of an open reading frame immediately downstream from lpd3. The coding region of lpd3 consists of 1401 bp, providing for 466 amino acids plus a stop codon with a G/C content of 62.4%. The transcriptional start site was located 33-bp upstream from the start of translation. The third lipoamide dehydrogenase (LPD-3) shares amino acid identity with the other two lipoamide dehydrogenases of P. putida, 45% with that of the 2-oxoglutarate dehydrogenase and pyruvate multienzyme complexes, and 45.9% with the lipoamide dehydrogenase of the branched-chain oxoacid complex. LPD-3 is more closely related to eukaryotic lipoamide dehydrogenases since it has 53.6% amino acid sequence identity with pig and human lipoamide dehydrogenases and 51.1% identity with yeast lipoamide dehydrogenase. LPD-3 was not produced in wild-type P. putida PpG2 under a variety of growth conditions. However, LPD-3 was produced in P. putida PpG2 carrying pSP14, a pKT240-based clone with the entire lpd3 gene plus 104 bases of the leader. The only demonstrated role of LPD-3 in P. putida is as a substitute for lipoamide dehydrogenase of the 2-oxoglutarate dehydrogenase and pyruvate multienzyme complexes when the latter is inactive or missing.

Pseudomonas putida is able to produce three lipoamide dehydrogenases: (i) LPD-glc, which is the E3 component of the pyruvate and 2-ketoglutarate dehydrogenase complexes and the L-factor for the glycine oxidation system; (ii) LPD-val, which is the specific E3 component of the branched-chain keto acid dehydrogenase complex and is induced by growth on leucine, isoleucine, or valine; and (iii) LPD-3, which was discovered in a lpdG mutant and whose role is unknown. Southern hybridization with an oligonucleotide probe encoding the highly conserved redox-active site produced three bands corresponding to the genes encoding these three lipoamide dehydrogenases. The complete structural gene for LPD-glc, lpdG, was isolated, and its nucleotide sequence was determined. The latter consists of 476 codons plus a stop codon, TAA. The structural gene for LPD-glc is preceded by a partial open reading frame with strong similarity to the E2 component of 2-ketoglutarate dehydrogenase of Escherichia coli. This suggests that lpdG is part of the 2-ketoglutarate dehydrogenase operon. LPD-glc was expressed in Pseudomonas putida JS348 from pHP4 which contains a partial open reading frame corresponding to the E2 component, 94 bases of noncoding DNA, and the structural gene for lpdG. This result indicates that lpdG can be expressed separately from the other genes of the operon.

Branched-chain keto acid dehydrogenase is a multienzyme complex produced by Pseudomonas putida when it is grown in a minimal medium containing branched-chain amino acids. A 1.87-kilobase (kb) DNA fragment was cloned and sequenced which contained 0.24 kb of the E1 alpha structural gene and 1.6 kb of upstream DNA. There were 854 base pairs (bp) of noncoding DNA upstream of bkdA1, the first gene of the bkd operon, and 592 bp between the transcriptional and translational starts. The G + C content of the noncoding region was 56.7% compared with 65.2% for all the structural genes of the operon. A partial open reading frame was found on the strand opposite that of the bkd operon beginning at base 774. When the bkd promoter was cloned into the promoter probe vector pKT240, streptomycin resistance was obtained in P. putida but not Escherichia coli with the promoter in both orientations, which indicates either that the bkd promoter is bidirectional or that there are two promoters in this region. A series of ordered deletions on both sides of the proposed site of the start of transcription revealed that almost 700 bp upstream of the start of translation were required for expression. Streptomycin resistance was also obtained in an rpoN mutant of P. putida KT2440 containing constructs with the intact bkd promoter, indicating that the bkd operon does not require the rpoN sigma factor for expression. Another construct containing the bkd promoter, bkdA1, and bkdA2 in pKT240 was used to transform P. putida JS113, a mutant which was unable to produce the E1 subunits of the branched-chain keto acid dehydrogenase. In this case, very high inducible expression of the bkd operon was obtained.

Pseudomonas putida oxidized isoleucine to acetyl-coenzyme A (CoA) and propionyl-CoA by a pathway which involved deamination of d-isoleucine by oxidation and l-isoleucine by transamination, oxidative decarboxylation, and beta oxidation at the ethyl side chain. At least three separate inductive events were required to form all of the enzymes of the pathway: d-amino acid dehydrogenase was induced during growth in the presence of d-isoleucine; branched-chain keto dehydrogenase was induced during growth on 2-keto-3-methylvalerate and enzymes specific for isoleucine metabolism; tiglyl-CoA hydrase and 2-methyl-3-hydroxybutyryl-CoA dehydrogenase were induced by growth on isoleucine, 2-keto-3-methylvalerate, 2-methylbutyrate, or tiglate. Tiglyl-CoA hydrase and 2-methyl-3-hydroxybutyryl-CoA dehydrogenase were purified simultaneously by several enzyme concentration procedures, but were separated by isoelectric focusing. Isoelectric points, pH optima, substrate specificity, and requirements for enzyme action were determined for both enzymes. Evidence was obtained that the dehydrogenase catalyzed the oxidation of 2-methyl-3-hydroxybutyryl-CoA to 2-methylacetoacetyl-CoA. 2-Methyl-3-hydroxybutyryl-CoA dehydrogenase catalyzed the oxidation of 3-hydroxybutyryl-CoA, but l-3-hydroxyacyl-CoA dehydrogenase from pig heart did not catalyze the oxidation of 2-methyl-3-hydroxybutyryl-CoA; therefore, they appeared to be different dehydrogenases. Furthermore, growth on tiglate resulted in the induction of tiglyl-CoA hydrase and 2-methyl-3-hydroxybutyryl-CoA dehydrogenase, but these two enzymes were not induced during growth on crotonate or 3-hydroxybutyrate.

The incorporation of isotope from [6-14C]glucose into alginate by both Pseudomonas aeruginosa and Azotobacter vinelandii was 10-fold greater than that from either [1-14C]- or [2-14C]glucose, indicating preferential utilization of the bottom half of the glucose molecule for alginate synthesis. These data strongly suggest that the Entner - Doudoroff pathway plays a major role in alginate synthesis in both P. aeruginosa and A. vinelandii.

Pseudomonas putida produces two lipoamide dehydrogenases with molecular weights of 49,000 and 56,000 designated LPD-val and LPD-glc, respectively. LPD-val is required for oxidation of valine, since it is specifically utilized as the E3 component of branched-chain keto acid dehydrogenase. Since glycine oxidation by bacteria and mammals also requires lipoamide dehydrogenase, we desired to determine which lipoamide dehydrogenase would be used by the P. putida glycine oxidation system. When grown in a medium with glycine as the sole nitrogen source, P. putida produced a single lipoamide dehydrogenase with a molecular weight of 56,000 and which reacted with antiserum to LPD-glc. The partially purified glycine oxidation system from P. putida was stimulated by LPD-glc but not by LPD-val and was inhibited by anti-LPD-glc, but not by anti-LPD-val. It was not possible to detect LPD-val in extracts of cells grown in glucose-glycine medium by the use of anti-LPD-val. LPD-glc was five times as active as LPD-val in catalyzing the oxidation of purified protein H, the heat-stable, lipoic acid-containing protein of the glycine oxidation system. These results indicate that LPD-glc is specifically utilized for glycine oxidation in P. putida.

The purpose of the present study was to determine if the enoyl coenzyme A hydratase formed by Pseudomonas putida during growth on isoleucine was a unique enzyme specific for isoleucine metabolism. The highest levels of the hydratase were formed during growth on isoleucine intermediates and the lowest levels during growth on glutamate and glucose. Data from growth experiments revealed that 2-methyl-3-hydroxybutyryl coenzyme A hydratase, an enzyme unique to isoleucine metabolism and enoyl coenzyme A hydratase were coordinately induced, but that 3-hydroxyacyl coenzyme A dehydrogenase was under separate control. The hydratase was purified 180-fold from isoleucine cells, and its physical and catalytic properties reported. The highest activity was with crotonyl coenzyme A,Vmax = 1100 x 10(3) moles/min mole enzyme, next was tiglyl coenzyme A, Vmax = 61 x 10(3) moles/min mole enzyme, and last was 3-methyl-crotonyl coenzyme A, Vmax = 2.3 x 10(3) moles/min mole enzyme. Enzyme purified from butyrate cells had the same elution patterns during column chromatography and catalytic properties as the enzyme from isoleucine cells. These data support the conclusion that a single enzyme in P. putida is responsible for the hydration of both tiglyl coenzyme A and crotonyl coenzyme A.

The nucleotide sequence of bkdB, the structural gene for E2b, the transacylase component of branched-chain-oxoacid dehydrogenase of Pseudomonas putida has been determined and translated into its amino acid sequence. The start of bkdB was identified from the N-terminal sequence of E2b isolated from branched-chain-oxoacid dehydrogenase of the closely related species, P. aeruginosa. The reading frame was composed of 65.5% G + C with 82.3% of the codons ending in G or C. There was no intergenic space between bkdA2 and bkdB. No codons requiring minor tRNAs were utilized and the codon bias index indicated a preferential codon usage. The bkdB gene encoded 423 amino acids although the N-terminal methionine was absent from E2b prepared from P. aeruginosa. The relative molecular mass of the encoded protein was 45,134 (45,003 minus methionine) vs 47,000 obtained by SDS/polyacrylamide gel electrophoresis. There was a single lipoyl domain in E2b compared to three lipoyl domains in E2p, and one domain in E2o, the transacylases of pyruvate and 2-oxoglutarate dehydrogenases of Escherichia coli respectively. There was significant similarity between the lipoyl domain of E2b and of E2p and E2o as well as between the E1-E2 binding domains of E2b, E2p and E2o. There was no similarity between the E3 binding domain of E2b to E2p and E2o which may reflect the uniqueness of the E3 component of branched-chain-oxoacid dehydrogenase of P. putida. The conclusions drawn from these comparisons are that the transacylases of prokaryotic pyruvate, 2-oxoglutarate and branched-chain-oxoacid dehydrogenases descended from a common ancestral protein probably at about the same time.

The genes encoding proteins responsible for activity of the E1 component of branched-chain-oxoacid dehydrogenase of Pseudomonas putida have been subcloned and the nucleotide sequence of this region determined. Open reading frames encoding E1 alpha (bkdA1, 1233 bp) and E1 beta (bkdA2, 1020 bp) were identified with the aid of the N-terminal sequence of the purified subunits. The Mr of E1 alpha was 45,158 and of E1 beta was 37,007, both calculated without N-terminal methionine. The deduced amino acid sequences of E1 alpha and E1 beta had no similarity to the published sequences of the E1 subunits of pyruvate and 2-oxoglutarate dehydrogenases of Escherichia coli. However, there was substantial similarity between the E1 alpha subunits of Pseudomonas and rat liver branched-chain-oxoacid dehydrogenases. In particular, the region of the E1 alpha subunit of the mammalian branched-chain-oxoacid dehydrogenase which is phosphorylated, was found to be highly conserved in the Pseudomonas E1 alpha subunit. There was also considerable similarity between the E1 beta subunits of Pseudomonas branched-chain-oxoacid dehydrogenase and human pyruvate dehydrogenase.

We purified lipoamide dehydrogenase from cells of Pseudomonas putida PpG2 grown on glucose (LPD-glu) and lipoamide dehydrogenase from cells grown on valine (LPD-val), which contained branched-chain keto acid dehydrogenase. LPD-glu had a molecular weight of 56,000 as determined by sodium dodecyl sulfate-polyacrylamide gel electrophoresis, and LPD-val had a molecular weight of 49,000. The pH optimum for LPD-glu for reduced nicotinamide adenine dinucleotide oxidation was 7.4, compared with pH 6.5 for LPD-val. When oxidized nicotinamide adenine dinucleotide was included in the assay mixture, the pH optima were 7.1 and 5.7, respectively. There was also a difference in pH optima between the two enzymes for oxidized nicotinamide adenine dinucleotide reduction, but the Michaelis constants and maximum velocities were similar. A purified preparation of branched-chain keto acid dehydrogenase, which was deficient in lipoamide dehydrogenase, was stimulated 10-fold by LPD-val but not by LPD-glu, which suggested that the branched-chain keto acid dehydrogenase of P. putida has a specific requirement for LPD-val. In contrast, a partially purified preparation of 2-ketoglutarate dehydrogenase that was deficient in lipoamide dehydrogenase was stimulated by LPD-glu but not by LPD-val, indicating that this complex has a specific requirement of LPD-glu.

The generation time of Pseudomonas putida with l-leucine was 20 h in synthetic media but only 3 h with d-leucine. Slow growth in the presence of l-leucine was partially overcome by addition of 0.1 mM amounts of either d-valine, l-valine, or 2-ketoisovalerate. The activities of five enzymes which take part in the oxidation of leucine by P. putida were measured under various conditions of growth. Four enzymes were induced by growth with dl-leucine as sole source of carbon: d-amino acid dehydrogenase, branched-chain keto acid dehydrogenase, 3-methylcrotonyl-coenzyme A carboxylase, and 3-hydroxy-3-methylglutaryl-coenzyme A lyase. The segment of the pathway required for oxidation of 3-methylcrotonate was induced by growth on isovalerate or 3-methylcrotonate without formation of the preceding enzymes. The synthesis of carboxylase and lyase appeared to have been repressed by the addition of l-glutamate or glucose to cells growing on dl-leucine as the sole carbon source. Mutants unable to grow at the expense of isovalerate had reduced levels of carboxylase and lyase, whereas the levels of three enzymes common to the catabolism of all three branched-chain amino acids and those of two isoleucine catabolic enzymes were normal.

Branched-chain keto acid dehydrogenase, an enzyme in the common pathway of branched-chain amino acid catabolism of Pseudomonas putida, is a multienzyme complex which catalyzes the oxidative decarboxylation of branched-chain keto acids. The objective of the present study was to isolate strains with mutations of this and other keto acid dehydrogenases and to map the location of the mutations on the chromosome of P. putida. Several strains with mutations of branched-chain keto acid dehydrogenase, two pyruvate and two 2-ketoglutarate dehydrogenase, were isolated, and the defective subunits were identified by biochemical analysis. By using a recombinant XYL-K plasmid to mediate conjugation, these mutations were mapped in relation to a series of auxotrophic and other catabolic mutations. The last time of entry recorded was at approximately 35 min, and the data were consistent with a single point of entry. Branched-chain keto acid dehydrogenase mutations affecting E1, E1 plus E2, and E3 subunits mapped at approximately 35 min. One other strain affected in the common pathway was deficient in branched-chain amino acid transaminase, and the mutation was mapped at 16 min. The mutations in the two pyruvate dehydrogenase mutants, one deficient in E1 and the other deficient in E1 plus E2, mapped at 22 minutes. The 2-ketoglutarate dehydrogenase mutation affecting the E1 subunit mapped at 12 minutes. A 2-ketoglutarate dehydrogenase mutant deficient in E3 was isolated, but the mutation proved too leaky to map.

Pseudomonas putida grown on valine produces two lipoamide dehydrogenases, LPD-glu (Mr, 56,000 and LPD-val (Mr, 49,000). The 49,000-dalton protein is used by P. putida for branched-chain keto acid dehydrogenase, whereas the 56,000-dalton protein is presumably used for pyruvate and 2-ketoglutarate dehydrogenases. The objective of this study was to isolate and characterize mutants of P. putida with mutations affecting lipoamide dehydrogenases in order to study the relationship of these two proteins. Mutant JS287 lacked LPD-val, the lipoamide dehydrogenase which is induced by growth on valine and is specific for branched-chain keto acid dehydrogenase, and had normal amounts of LPD-glu, the lipoamide dehydrogenase which is formed during growth on glucose and which is probably used by both pyruvate and 2-ketoglutarate dehydrogenases. Mutant JS94 was a pleiotropic mutant with defects in 2-ketoglutarate, branched-chain, and lipoamide dehydrogenases. Proteolysis of LPD-glu and LPD-val produced completely different digestion products, suggesting that these two proteins are products of separate structural genes. Antisera prepared against LPD-glu reacted only with LPD-glu, whereas antisera prepared against LPD-val reacted with LPD-val and cross-reacted with LPD-glu. Although mutant JS94 did not produce active lipoamide dehydrogenase, cell-free extracts of this mutant contained a protein which cross-reacted with anti-LPD-val.

The production of two lipoamide dehydrogenases by Pseudomonas is so far unique. One, LPD-val, is the specific E3 component of the branched-chain-oxoacid dehydrogenase and the second, LPD-glc, is the E3 component of 2-oxoglutarate dehydrogenase and the L-factor of the glycine oxidation system. The objective of the present research was to determine the nucleotide sequence of the structural gene for LPD-val in order to compare its deduced amino acid structure with that of other redox-active disulfide flavoproteins. Northern blots using mRNA isolated from P. putida grown in media with branched-chain amino acids identified a transcript of 6.2 kb which is long enough to encode all the structural genes for the complex. The nucleotide sequence of the structural gene for LPD-val, lpdV, was determined and consists of 459 codons plus the stop codon. The open reading frame begins two bases after the stop codon for the E2 subunit and is composed of 66.3% G + C. Codon usage is characteristic of moderately strongly expressed genes. There is a ribosome-binding site preceding the ATG start codon and a strong candidate for a rho-independent terminator at the 3' end of the reading frame. The Mr of the protein encoded is 48,164 and when the Mr of FAD is added, the total Mr is 48,949, which is very close to the value of 49,000 obtained by SDS-polyacrylamide gel electrophoresis. Similarity comparisons of LPD-val with sequences of three other lipoamide dehydrogenases showed that LPD-val was somewhat more distantly related. It is probable that the lipoamide dehydrogenases and the glutathione and mercuric reductases evolved from a common ancestral flavoprotein.

Pseudomonas putida produces two lipoamide dehydrogenases, LPD-glc and LPD-val. LPD-val is specifically required as the lipoamide dehydrogenase of branched-chain keto acid dehydrogenase and LPD-glc fulfills all other requirements for lipoamide dehydrogenase. Both proteins are dimers with one FAD per subunit. LPD-glc has an absorption maximum at 455 nm, but LPD-val has a maximum at 460 nm. Comparison of amino acid compositions revealed that LPD-glc was more closely related to Escherichia coli and pig heart lipoamide dehydrogenase than to LPD-val. LPD-val did not appear to be closely related to any of the proteins compared with the possible exception of mercuric reductase.

We purified branched-chain keto acid dehydrogenase to a specific activity of 10 mumol/min per mg of protein from Pseudomonas putida grown on valine. The purified enzyme was active with 2-ketoisovalerate, 2-ketoisocaproate, and 2-keto-3-methylvalerate in a ratio of 1.0:0.8:0.7 but showed no activity with either pyruvate or 2-ketoglutarate. There were four polypeptides in the purified enzyme (molecular weights, 49,000, 46,000, 39,000, and 37,000). The purified enzyme was deficient in the specific lipoamide dehydrogenase produced during growth on valine (molecular weight, 49,000). Branched-chain keto acid dehydrogenase required L-valine, oxidized nicotinamide adenine dinucleotide, coenzyme A, thiamine pyrophosphate, and magnesium chloride. A partially purified preparation catalyzed the oxidation of 2-keto-[1-14C]isovalerate to [14C]carbon dioxide, isobutyryl-coenzyme A, and reduced nicotinamide adenine dinucleotide in equimolar amounts. Both the Km and the Vmax for 2-ketoisovalerate were affected by the addition of L-valine to the assay mixture. However, only the Vmax values for oxidized nicotinamide adenine dinucleotide and coenzyme A were affected when L-valine was present. This suggested that valine acted by affecting the binding of branched-chain keto acids to subunit E1 of the complex.

Pseudomonads are the only organisms so far known to produce two lipoamide dehydrogenases (LPDs), LPD-Val and LPD-Glc. LPD-Val is the specific E3 component of branched-chain oxoacid dehydrogenase, and LPD-Glc is the E3 component of 2-ketoglutarate and possibly pyruvate dehydrogenases and the L-factor of the glycine oxidation system. Three mutants of Pseudomonas putida, JS348, JS350, and JS351, affected in lpdG, the gene encoding LPD-Glc, have been isolated; all lacked 2-ketoglutarate dehydrogenase, but two, JS348 and JS351, had normal pyruvate dehydrogenase activity. The pyruvate and 2-ketoglutarate dehydrogenases of the wild-type strain of P. putida were both inhibited by anti-LPD-Glc, but the pyruvate dehydrogenase of the lpdG mutants was not inhibited, suggesting that the mutant pyruvate dehydrogenase E3 component was different from that of the wild type. The lipoamide dehydrogenase present in one of the lpdG mutants, JS348, was isolated and characterized. This lipoamide dehydrogenase, provisionally named LPD-3, differed in molecular weight, amino acid composition, and N-terminal amino acid sequence from LPD-Glc and LPD-Val. LPD-3 was clearly a lipoamide dehydrogenase as opposed to a mercuric reductase or glutathione reductase. LPD-3 was about 60% as effective as LPD-Glc in restoring 2-ketoglutarate dehydrogenase activity and completely restored pyruvate dehydrogenase activity in JS350. These results suggest that LPD-3 is a lipoamide dehydrogenase associated with an unknown multienzyme complex which can replace LPD-Glc as the E3 component of pyruvate and 2-ketoglutarate dehydrogenases in lpdG mutants.

Branched-chain oxo acid dehydrogenase was purified from Pseudomonas aeruginosa strain PAO with the objective of resolving the complex into its subunits. The purified complex consisted of four proteins, of Mr 36,000, 42,000, 49,000 and 50,000. The complex was resolved by heat treatment into the 49,000 and 50,000-Mr proteins, which were separated by chromatography on DEAE-Sepharose. The 49,000-Mr protein was identified as the E2 subunit by its ability to catalyse transacylation with a variety of substrates, with dihydrolipoamide as the acceptor. P. aeruginosa, like P. putida, produces two lipoamide dehydrogenases. One, the 50,000-Mr protein, was identified as the specific E3 subunit of branched-chain oxo acid dehydrogenase and had many properties in common with the lipoamide dehydrogenase LPD-val of P. putida. The second lipoamide dehydrogenase had Mr 54,000 and corresponded to the lipoamide dehydrogenase LPD-glc of P. putida. Fragments of C-terminal CNBr peptides of LPD-val from P. putida and P. aeruginosa corresponded closely, with only two amino acid differences over 31 amino acids. A corresponding fragment at the C-terminal end of lipoamide dehydrogenase from Escherichia coli also showed extensive homology. All three peptides had a common segment of eight amino acids, with the sequence TIHAHPTL. This homology was not evident in any other flavoproteins in the Dayhoff data base which suggests that this sequence might be characteristic of lipoamide dehydrogenase.

We cloned the structural genes for the individual subunits of the branched-chain keto acid dehydrogenase multienzyme complex on a 7.8-kilobase EcoRI-SstI restriction fragment of Pseudomonas putida chromosomal DNA by cloning into the broad-host-range vector pKT230. A direct selection system for growth on valine-isoleucine agar was achieved by complementation of P. putida branched-chain keto acid dehydrogenase mutants. The recombinant plasmid, pSS1-1, increased expression of branched-chain keto acid dehydrogenase up to five times in wild-type P. putida. The complex was expressed constitutively in P. putida(pSS1-1) but was inducible in Escherichia coli HB101(pSS1-1) by high valine. E. coli minicells transformed with pSS1-1 produced three polypeptides which did not match the four polypeptides of the purified complex. To resolve this problem, we inserted P. putida DNA from pSS1-1 into pUC18 and pUC19. The pUC-derived plasmids were used as DNA templates in an E. coli transcription-translation system. Four polypeptides were produced from the pUC18-derived plasmid which had the correct molecular weights, showing that the structural genes had been cloned. Since only weak bands were produced with the pUC19-derived plasmid, the direction of transcription was established. The locations and order of all the structural genes of branched-chain keto acid dehydrogenase were located by restriction enzyme mapping.

Sokatch, J. R. (University of Oklahoma School of Medicine, Oklahoma City). Alanine and aspartate formation during growth on valine-C(14) by Pseudomonas aeruginosa. J. Bacteriol. 92:72-75. 1966.-Pseudomonas aeruginosa grown with dl-valine-4,4'-C(14) synthesized alanine labeled mainly in carbons 1 and 3, indicating that the isopropyl carbons of valine were the precursors of pyruvate for alanine formation by a pathway which did not involve randomization of isotope. Alanine from cells grown on valine-1-C(14) contained isotope only in the carboxyl carbon, suggesting another route to pyruvate from valine by carbon dioxide fixation. Oxidation of valine to propionyl-coenzyme A (CoA), as it occurs in animal tissues, followed by the oxidation of propionyl-CoA to acrylyl-CoA, lactyl-CoA, and pyruvate, would account for the isotope data. Cells grown on valine oxidized valine, isobutyrate, and propionate immediately, whereas cells grown on acetate did not oxidize valine or isobutyrate and required an induction period before propionate was oxidized. P. aeruginosa grown with propionate-1-C(14) or propionate-2-C(14) formed alanine-1-C(14) and alanine-2-C(14), respectively, which agrees with the contention that at least part of the propionate is oxidized via the acrylate pathway. Aspartate formed from valine-1-C(14) was labeled only in the carboxyl carbons, whereas that formed from valine-4,4'-C(14) was labeled in all four carbons, but most heavily in carbons 1 and 3. These data suggest that the main route for the formation of the carbon skeleton of aspartate was by a C(3) plus C(1) condensation, with the C(3) unit derived from the isopropyl carbons of valine and the C(1) unit probably from carbon dioxide.

A particulate d-amino acid dehydrogenase has been partially purified from cell free extracts of Pseudomonas aeruginosa grown on dl-valine as the source of carbon and energy. A standard assay was developed which utilized 2,6-dichlorophenol-indophenol as the electron acceptor. The pH optimum for enzyme activity ranged from 6.0 to 8.0, depending on the amino acid assayed. The enzyme was most active with monoamino-monocarboxylic amino acids and histidine. The Michaelis constant for d-phenylalanine was found to be 1.3 x 10(-3)m d-phenylalanine. Constants could not be calculated for the other amino acids oxidized because anomalous plots of V as a function of V/S were obtained. Spectra of enzyme preparations reduced with d-valine or sodium hydrosulfite exhibited adsorption bands typical of the alpha, beta, and gamma bands of cytochromes as well as bleaching in the flavin region of the spectrum. When dl-valine was added to a medium with glycerol as the energy source, d-amino acid dehydrogenase was detected after the addition of valine and was produced at a rate directly proportional to the synthesis of total protein. The enzyme was formed when d-valine, l-valine, or dl-alanine was the source of carbon and energy, but not when glucose, glycerol, or succinate was the energy source.

The activities of six enzymes which take part in the oxidation of valine by Pseudomonas putida were measured under various conditions of growth. The formation of four of the six enzymes was induced by growth on d- or l-valine: d-amino acid dehydrogenase, branched-chain keto acid dehydrogenase, 3-hydroxyisobutyrate dehydrogenase, and methylmalonate semialdehyde dehydrogenase. Branched-chain amino acid transaminase and isobutyryl-CoA dehydrogenase were synthesized constitutively. d-Amino acid dehydrogenase and branched-chain keto acid dehydrogenase were induced during growth on valine, leucine, and isoleucine, and these enzymes were assumed to be common to the metabolism of all three branched-chain amino acids. The segment of the pathway required for oxidation of isobutyrate was induced by growth on isobutyrate or 3-hydroxyisobutyrate without formation of the preceding enzymes. d-Amino acid dehydrogenase was induced by growth on l-alanine without formation of other enzymes required for the catabolism of valine. d-Valine was a more effective inducer of d-amino acid dehydrogenase than was l-valine. Therefore, the valine catabolic pathway was induced in three separate segments: (i) d-amino acid dehydrogenase, (ii) branched-chain keto acid dehydrogenase, and (iii) 3-hydroxyisobutyrate dehydrogenase plus methylmalonate semialdehyde dehydrogenase. In a study of the kinetics of formation of the inducible enzymes, it was found that 3-hydroxyisobutyrate and methylmalonate semialdehyde dehydrogenases were coordinately induced. Induction of enzymes of the valine catabolic pathway was studied in a mutant that had lost the ability to grow on all three branched-chain amino acids. Strain PpM2106 had lowered levels of branched-chain amino acid transaminase and completely lacked branched-chain keto acid dehydrogenase when grown in medium which contained valine. Addition of 2-ketoisovalerate, 2-ketoisocaproate, or 2-keto-3-methylvalerate to the growth medium of strain PpM2106 resulted in induction of normal levels of branched-chain keto acid dehydrogenase; therefore, the branched-chain keto acids were the actual inducers of branched-chain keto acid dehydrogenase.

Two types of Pseudomonas putida PpG2 mutants which were unable to degrade branched-chain amino acids were isolated after mutagenesis and selection for ability to grow on succinate, but not valine, as a sole source of carbon. These isolates were characterized by growth on the three branched-chain amino acids (valine, isoleucine, and leucine), on the corresponding branched-chain keto acids (2-ketoisovalerate, 2-keto-3-methylvalerate, and 2-ketoisocaproate), and on other selected intermediates as carbon sources, and by their enzymatic composition. One group of mutants lost 2-ketoisovalerate-inducible branched-chain keto acid dehydrogenase that was active on all three keto acids. There was also a concomitant loss of ability to grow on all three branched-chain amino acids as well as on all three corresponding keto acids, but there was retention of ability to use subsequent intermediates in the catabolism of branched-chain amino acids. Another type of mutant showed a marked reduction in branched-chain amino acid transaminase activity and grew poorly at the expense of all three amino acids, but it utilized subsequent intermediates as carbon sources. Both the transaminase and branched-chain keto acid dehydrogenase mutants retained the ability to degrade camphor. These findings are consistent with the view that branched-chain amino acid transaminase and branched-chain keto acid dehydrogenase are common enzymes in the catabolism of valine, isoleucine, and leucine.