An archetypical extradiol-cleaving catecholic dioxygenase: the crystal structure of catechol 2,3-dioxygenase (metapyrocatechase) from Ppseudomonas putida mt-2

BACKGROUND

Catechol dioxygenases catalyze the ring cleavage of catechol and its derivatives in either an intradiol or extradiol manner. These enzymes have a key role in the degradation of aromatic molecules in the environment by soil bacteria. Catechol 2, 3-dioxygenase catalyzes the incorporation of dioxygen into catechol and the extradiol ring cleavage to form 2-hydroxymuconate semialdehyde. Catechol 2,3-dioxygenase (metapyrocatechase, MPC) from Pseudomonas putida mt-2 was the first extradiol dioxygenase to be obtained in a pure form and has been studied extensively. The lack of an MPC structure has hampered the understanding of the general mechanism of extradiol dioxygenases.

RESULTS

The three-dimensional structure of MPC has been determined at 2.8 A resolution by the multiple isomorphous replacement method. The enzyme is a homotetramer with each subunit folded into two similar domains. The structure of the MPC subunit resembles that of 2,3-dihydroxybiphenyl 1,2-dioxygenase, although there is low amino acid sequence identity between these enzymes. The active-site structure reveals a distorted tetrahedral Fe(II) site with three endogenous ligands (His153, His214 and Glu265), and an additional molecule that is most probably acetone.

CONCLUSIONS

The present structure of MPC, combined with those of two 2,3-dihydroxybiphenyl 1,2-dioxygenases, reveals a conserved core region of the active site comprising three Fe(II) ligands (His153, His214 and Glu265), one tyrosine (Tyr255) and two histidine (His199 and His246) residues. The results suggest that extradiol dioxygenases employ a common mechanism to recognize the catechol ring moiety of various substrates and to activate dioxygen. One of the conserved histidine residues (His199) seems to have important roles in the catalytic cycle.

Dopamine-degrading activity of monoamine oxidase is not detected by histochemistry in neurons of the substantia nigra pars compacta of the rat

Monoamine oxidase (MAO) activity was examined in neurons of the substantia nigra pars compacta (SNC) of the rat using a histochemical method, and compared to MAO activity in neurons of the locus coeruleus (LC) and dorsal raphe nucleus (DR). Using dopamine as a substrate, dopamine-degrading MAO activity was not detected in any SNC neurons, although LC and DR neurons were intensely stained for this activity. We further examined MAO activity in these neurons using other substrates, including serotonin (an MAO type A preferential substrate), beta-phenylethylamine (an MAO type B preferential substrate), and tyramine (a substrate common to both MAO types A and B). As for dopamine, no SNC neurons were stained for MAO activity using any of these other substrates. In contrast, LC neurons were intensely stained when either serotonin or tyramine was used, and DR neurons were darkly stained when either beta-phenylethylamine or tyramine was used. The lack of evidence of MAO activity in the SNC is surprising given that there are densely packed tyrosine hydroxylase (TH)-immunoreactive neurons in the SNC (i.e., dopaminergic neurons). By comparison, in the LC and DR the distribution patterns of the MAO-stained neurons were similar to those of TH-immunolabeled neurons (i.e., noradrenergic neurons) and serotonin-immunoreactive neurons, respectively. Our results suggest that dopamine-degrading MAO activity and MAO types A and B activities in SNC dopamine neurons are very low compared to MAO activity in LC noradrenaline neurons and in DR serotonin neurons.

D-serine content and D-[3H]serine binding in the brain regions of the senescence-accelerated mouse

An established senescence-accelerated model mouse strain, SAMP8, shows the deterioration of learning and memory compared with a normal control strain, SAMR1. D-Serine binds to strychnine-insensitive glycine binding sites of the N-methyl-D-aspartate (NMDA) receptor complex, and enhances glutamate binding to the receptor complex. To investigate the relationship of endogenous brain D-serine and the brain dysfunction caused by aging, the level of brain free D-serine and the D-[3H]serine binding to the brain samples were examined using the SAMP8 and SAMR1 mice. The free D-serine level was highest in the cerebral frontal and occipital cortices in both the SAMP8 and SAMR1; no difference in the D-serine level was shown between the two strains. A receptor autoradiographical analysis showed that the D-[3H]serine binding to the brain section was highest in the hippocampus, and the binding in the SAMP8 brains was lower than that of the SAMR1. The D-[3H]serine binding to the crude cerebral membranes indicated that the value of the total binding sites for the SAMP8 was lower than that for the SAMR1, whereas the value of the dissociation constant Kd for the SAMP8 was similar to that of the SAMR1. These results suggest that the number of D-[3H]serine binding sites was decreased in the SAMP8 compared to the SAMR1, but the affinity of D-[3H]serine to the binding sites was not altered. These results support the view that a decrease of NMDA receptor complex is involved in the age-related neural dysfunction of SAMP8 mice.

Altered activities of transcription factors and their related gene expression in cardiac tissues of diabetic rats

Gene regulation in the cardiovascular tissues of diabetic subjects has been reported to be altered. To examine abnormal activities in transcription factors as a possible cause of this altered gene regulation, we studied the activity of two redox-sensitive transcription factors--nuclear factor-kappaB (NF-kappaB) and activating protein-1 (AP-1)--and the change in the mRNA content of heme oxygenase-1, which is regulated by these transcription factors in the cardiac tissues of rats with streptozotocin-induced diabetes. Increased activity of NF-kappaB and AP-1 but not nuclear transcription-activating factor, as determined by an electrophoretic mobility shift assay, was found in the hearts of 4-week diabetic rats. Glycemic control by a subcutaneous injection of insulin prevented these diabetes-induced changes in transcription factor activity. In accordance with these changes, the mRNA content of heme oxygenase-1 was increased fourfold in 4-week diabetic rats and threefold in 24-week diabetic rats as compared with control rats (P < 0.01 and P < 0.05, respectively). Insulin treatment also consistently prevented changes in the mRNA content of heme oxygenase-1. The oral administration of an antioxidant, probucol, to these diabetic rats partially prevented the elevation of the activity of both NF-kappaB and AP-1, and normalized the mRNA content of heme oxygenase-1 without producing any change in the plasma glucose concentration. These results suggest that elevated oxidative stress is involved in the activation of the transcription factors NF-kappaB and AP-1 in the cardiac tissues of diabetic rats, and that these abnormal activities of transcription factors could be associated with the altered gene regulation observed in the cardiovascular tissues of diabetic rats.

Dopamine-degrading activity of monoamine oxidase in locus coeruleus and dorsal raphe nucleus neurons. A histochemical study in the rat

Dopamine-degrading activity of monoamine oxidase (MAO) was detected in the rat using a new histochemical method, with dopamine as the substrate. Our new method, designed to minimise the non-enzymatic oxidation of dopamine, was applied in combination with tyrosine hydroxylase (TH) and serotonin immunohistochemistry. We showed that the distribution pattern of MAO neurons was similar to that of TH-immunoreactive neurons (i.e. noradrenergic neurons) in the locus coeruleus (LC) and to that of serotonergic neurons in the dorsal raphe nucleus (DR). Since LC neurons form dopamine during noradrenaline biosynthesis, and DR neurons produce dopamine from exogenously administered L-dopa, our results indicate that dopamine produced in LC and DR neurons may be degraded, at least in part, by MAO.

Semi-micro-scale frontal gel chromatography of interacting systems of a protein and small molecules: binding of warfarin, tryptophan, or FMN to albumin, and of o-nitrophenol to catechol 2,3-dioxygenase

Frontal gel chromatography is a convenient and accurate method to obtain the free ligand concentration of a protein-ligand mixture. Because a large amount of sample (more than 6 ml) is required for the method, it has been rarely used for binding experiments. We have developed a system to carry out frontal gel chromatography on a semi-micro scale using short gel filtration columns (4.6 mm x 50-100 mm); frontal chromatograms could be obtained with small amounts of samples (1-2.5 ml) within 20 min. We used this technique to examine the binding of warfarin, L-tryptophan, or FMN to human serum albumin, the binding of warfarin to bovine serum albumin, and the interaction of catechol 2,3-dioxygenase with o-nitrophenol. The data fitted to a binding model in which a protein has one or several independent binding sites. Both human and bovine serum albumin showed the high-affinity bindings of two warfarin molecules. The binding number for L-tryptophan on human serum albumin was confirmed to be one, whereas maximal binding of FMN was 0.6 molecule per albumin molecule. o-Nitrophenol showed high-affinity binding only to holocatechol 2,3-dioxygenase. The absorption spectrum of the bound o-nitrophenol resembled that of anionic o-nitrophenol. These results demonstrated that frontal gel chromatography on a semi-micro scale is useful for the study of binding systems; the method is rapid and would be easy to automate fully.

Crystallization and preliminary X-ray diffraction studies of expressed Pseudomonas putida catechol 2,3-dioxygenase

Crystals of recombinant Pseudomonas putida catechol 2,3-dioxygenase, metapyrocate-chase, composed of four identical subunits, each with a molecular mass of 35 kDa and one nonheme ferrous iron, have been grown by the vapor diffusion method using sodium citrate as the precipitant. Repeated macroseeding and the addition of ethanol to protein solutions were together effective for obtaining crystals suitable for further crystallographic characterization. The crystals belong to the tetragonal space group P4(2)2(1)2 with unit-cell dimensions of a = b = 266 A, c = 60 A. They diffracted beyond 2.5 A resolution with synchrotron radiation. Assuming that one tetramer (alpha-Fe2+)4 is contained in an asymmetric unit, the crystal volume per unit molecular mass, Vm, is calculated to be 3.8 A3/Da, which corresponds to the solvent content of 67.6%.

A natural variant of bovine dopamine beta-monooxygenase with phenylalanine as residue 208: purification and characterization of the variant homo- and heterotetramers of (F208)4 and (F208)2(L208)2

Bovine dopamine beta-monooxygenase was purified from each of 18 individual adrenal glands by the method we have developed for the rapid purification of the enzyme from a single adrenal gland. Differential peptide mapping of the 18 enzyme preparations following fluorescence labeling of their cysteine residues revealed the presence of a novel variant with Phe as residue 208 in 14 adrenal glands; seven of them were homozygous for the variant allele and the remaining seven heterozygous. The variant enzyme was a tetramer and exhibited kinetic and structural properties similar to those of the wild-type tetramer (L208)4. These results indicate an allelic polymorphism and codominant expression of the two alleles of the enzyme gene.

Human group II phospholipase A2 in normal and diseased intervertebral discs

We measured calcium-dependent phospholipase A2 (PLA2) activity and immunoreactive group II PLA2 levels of 54 normal discs obtained from cadavers and 73 disc samples surgically obtained from patients with spinal disorders, including intervertebral disc herniations, spondylosis, and spondylolisthesis. Both cadaveric and surgical disc specimens contained about two-fold greater PLA2 activity than the ileal mucosa, one of the richest sources of group II PLA2. Discs of middle-aged cases had significantly higher activity than those of younger and elder cases. In cadaveric normal discs, calcium-dependent PLA2 activity was significantly higher in females than in males. Annulus fibrosus and nucleus pulposus contained the same PLA2 levels. In diseased disc, herniated fragments that had extruded or protruded out of the discs possessed lower activity than other parts of discs in the intervertebral space. Immunoreactive group II PLA2 levels of intervertebral discs closely correlated with PLA2 enzymatic activity. We purified a PLA2 from human intervertebral disc to homogeneity to further identify the isozymic nature of discal PLA2. Its NH2-terminal amino acid sequences and molecular weight were identical to those of human group II PLA2. Immunohistochemical analysis using a monoclonal anti-group II PLA2 antibody showed that in both annulus fibrosus and nucleus pulposus chondrocytes contained intense group II PLA2 immunoreactivity in their cytoplasm, and that the matrix contained no substantial immunoreactivity. These results suggest that group II PLA2 in chondrocytes has important physiological roles in discal ordinary metabolism, maintaining discal homeostasis.

Selective cleavage and modification of the intersubunit disulfide bonds of bovine dopamine beta-monooxygenase: conversion of tetramer to active dimer

Bovine dopamine beta-monooxygenase is a tetramer consisting of two disulfide-linked dimers. To examine the role(s) of the intersubunit disulfide bonds in the protein structure and activity, the enzyme was treated with DTT at pH 7.5 and 25 degrees C under nondenaturing conditions. A 15-min incubation with 0.5 mM DTT selectively cleaved half of the intersubunit disulfide bonds. The cleavage did not affect the activity or tetrameric structure of the enzyme. Upon chemical modification of the reduced cysteine residues with 0.1 M iodoacetamide (IAA) for 60 min, half of the tetramer was converted to a dimeric species. The resulting dimeric and tetrameric species exhibited similar kinetic properties, and the Vmax values were decreased by 30% compared to that of the native enzyme. Upon treatment with IAA alone, no dimer species was detected but the enzyme lost 30% of the original activity. Cys514 and Cys516 were selectively modified by the treatment with DTT and IAA. From these results, we concluded that: (i) chemical modification of the intermolecular disulfide bonds strongly destabilizes the intersubunit interaction; (ii) breakage of the intersubunit interaction does not affect the activity. The reduction mechanism of the intersubunit disulfide bonds and the roles of the intersubunit interactions are discussed.

Expression of intercellular adhesion molecules 1 (ICAM-1) via an osmotic effect in human umbilical vein endothelial cells exposed to high glucose medium

To clarify the etiology of accelerated atherosclerosis in patients with diabetes mellitus, we measured expression of intercellular adhesion molecule 1 (ICAM-1), vascular cellular adhesion molecule 1 (VCAM-1), and E-selection on the cell surface by enzyme-linked immunosorbent assay and ICAM-1 mRNA content in human umbilical vein endothelial cells exposed to 5.5 mM glucose (NG), 33 mM glucose (HG), or 27.5 mM mannitol plus 5.5 mM glucose (HM).1) Cell-surface ICAM-1 expression in HG and HM cells was maximally increased by 37% and 32% (P < 0.01), respectively. This effect was dependent on glucose concentration in the medium and was found as early as 24 h and maintained until 6 days after exposing cells of HG. However, neither VCAM-1 nor E-selection expression were affected by HG conditions. 2) Both HG and HM induced increased mRNA content between 6 and 12 h after the stimulation. 3) Adhesion of THP-1 cells to endothelial cells exposed to HG and HM was increased, when compared to NG conditions. These results indicate that osmotic effects can induce increased mRNA and cell-surface expression of ICAM-1 via an as yet unknown mechanism.

Overexpression of Pseudomonas putida catechol 2,3-dioxygenase with high specific activity by genetically engineered Escherichia coli

The cloned xylE gene encoding catechol 2,3-dioxygenase (metapyrocatechase) from TOL plasmid in Pseudomonas putida mt-2 has been expressed in Escherichia coli W3110 to a level of approximately 15% of the total soluble protein. Of the total iron in the crude extract, 45% was on the enzyme. The crystallized enzyme from E. coli had higher iron content (3.7 mol/mol enzyme) and specific activity (536 U/mg) than the enzyme from P. putida mt-2. However, no differences were observed in physicochemical, protein-chemical, and kinetic properties between the two enzymes. The enzyme was a homotetramer, and no changes were observed in the values of M(r) (136,000 +/- 5,000) and Stokes radius (4.26 nm) in the concentration range from 0.36 nM to 2.8 microM, indicating that the native enzyme neither dissociated into subunits nor polymerized in this range. The catalytic center activity and the Km values for catechol and dioxygen were 278 s-1, 1.87 and 7.45 microM, respectively, at pH 7.5 and 25 degrees C. The enzyme showed a broad substrate specificity. Among substrates, 4-methylcatechol and 4-chlorocatechol showed specificity constants (approximately 200 microM-1.s-1) higher than that for catechol. Acetone and phenol derivatives competitively inhibited the activity against catechol. The relationship between specific activity and iron content was not linear, suggesting some conformational changes in the partially iron-depleted enzyme.

D-amino-acid oxidase is confined to the lower brain stem and cerebellum in rat brain: regional differentiation of astrocytes

Based on enzymatic activity, the localization and the identification of D-amino-acid oxidase-containing cells in rat whole brain was systematically studied in serial fixed sections. The oxidase activity was absent or scarce in the forebrain, was confined to the brain stem (midbrain, pons and medulla oblongata) and cerebellum, and its localization was extended to the spinal cord. In the brain stem the oxidase was mainly localized in the tegmentum, particularly in the reticular formation. The intense oxidase reactions were present in the red nucleus, oculomotor nucleus, trochlear nucleus, ventral nucleus of the lateral lemniscus, dorsal and ventral cochlear nuclei, vestibular nuclei, nuclei of posterior funiculus, nucleus of the spinal tract of the trigeminal nerve, lateral reticular nucleus, inferior olivary nucleus, and hypoglossal nucleus. In the cerebellum the activity in the cortex was much more intense than that in the medulla. In all the fields described above, the oxidase-containing cells were exclusively astrocytes including Bergmann glial cells, and neither neuronal components, endothelial cells, oligodendrocytes nor ependymal cells showed oxidase activity. These results indicated that the astrocytes regionally differentiated into two distinct types, one of which expressed oxidase in the midbrain, rhombencephalon and spinal cord, and the other which did not in the forebrain. The localization of the oxidase was inversely correlated with the distribution of free D-serine in mammalian brains (Nagata, Y., Horiike, K. and Maeda, T., Brain Res., 634 (1994) 291-295). Based on the characteristic localization of the oxidase-containing astrocytes, we discussed the physiological role of the oxidase.

Distribution of free D-serine in vertebrate brains

Free D-serine distribution in vertebrate brains was investigated. In various brain regions of the lower vertebrate species, carp, frog and chick, free D-serine levels were low. On the contrary, in the mammals, mouse, rat and bull, the contents of free D-serine were high in the forebrain (around 400 nmol/g wet weight, and the ratio of D-serine to L-serine, was D/L = 0.4), and low in the hindbrain. In developing mice, D-serine levels in the cerebrum increased with age and attained the adult level (D/L = 0.40) 8 weeks after birth. In the cerebellum and brain stem, the free D-serine levels increased with age until 2 weeks, followed by a decrease to the adult levels: the D/L ratios remained constant until 2 weeks of age, then decreased to 0.03 in the cerebellum and 0.12 in the brain stem. Free D-serine was shown not to be of microbial origin using germ-free mice. In the rat forebrain, D-serine was evenly distributed in two cerebral regions, namely frontal and occipital lobes. The D/L ratios in other regions of forebrain, hippocampus and hypothalamus, were comparable to the cerebrum (D/L = 0.4), while that in the olfactory bulb was lower (D/L = 0.12). In the rat cerebrum, the D-serine content in the grey matter was significantly higher than that in the white matter. The contents of free D-serine in bovine cerebrum and cerebellum were similar to those in other mammalian brains, but the D/L ratio for bovine cerebral grey matter was lower than that for the cerebral white matter.(ABSTRACT TRUNCATED AT 250 WORDS)

Identification and characterization of a new plasmid carrying genes for degradation of 2,4-dichlorophenoxyacetate from Pseudomonas cepacia CSV90

Pseudomonas cepacia CSV90 is able to utilize 2,4-dichlorophenoxyacetate (2,4-D) and 2-methyl-4-chlorophenoxyacetate as sole sources of carbon and energy. Mutants of the strain CSV90 which had lost this ability appeared spontaneously on a nonselective medium. The wild-type strain harbored a 90-kb plasmid, pMAB1, whereas 2,4-D-negative mutants either lost the plasmid or had a 70-kb plasmid, pMAB2. The plasmid pMAB2 was found to have undergone a deletion of a 20-kb fragment of pMAB1. The plasmid-free mutants regained the ability to degrade 2,4-D after introduction of purified pMAB1 by electroporation. Cloning in Escherichia coli of a 10-kb BamHI fragment from pMAB1, the region absent in pMAB2, resulted in the expression of the gene tfdC encoding 3,5-dichlorocatechol 1,2-dioxygenase. After subcloning, the tfdC gene was located in a 1.6-kb HindIII fragment. The nucleotide sequence of the tfdC gene and the restriction map of its contiguous region are identical to those of the well-characterized 2,4-D-degradative plasmid pJP4 of Alcaligenes eutrophus, whereas the overall restriction maps of the two plasmids are different. The N-terminal 44-amino-acid sequence of the enzyme purified from the strain CSV90 confirmed the reading frame in the DNA sequence for tfdC and indicated that the initiation codon GUG is read as methionine instead of valine.

Purification of 3,5-dichlorocatechol 1,2-dioxygenase, a nonheme iron dioxygenase and a key enzyme in the biodegradation of a herbicide, 2,4-dichlorophenoxyacetic acid (2,4-D), from Pseudomonas cepacia CSV90

An enzyme which cleaves the benzene ring of 3,5-dichlorocatechol has been purified to homogeneity from Pseudomonas cepacia CSV90, grown with 2,4-dichlorophenoxyacetic acid (2,4-D) as the sole carbon source. The enzyme was a nonheme ferric dioxygenase and catalyzed the intradiol cleavage of all the examined catechol derivatives, 3,5-dichlorocatechol having the highest specificity constant of 7.3 microM-1s-1 in an air-saturated buffer. No extradiol-cleaving activity was observed. Thus, the enzyme was designated as 3,5-dichlorocatechol 1,2-dioxygenase. The molecular weight of the native enzyme was ascertained to be 56,000 by light scattering method, while the M(r) value of the enzyme denatured with 6 M guanidine-HCl or sodium dodecyl sulfate was 29,000 or 31,600, respectively, suggesting that the enzyme was a homodimer. The iron content was estimated to be 0.89 mol per mole of enzyme. The enzyme was deep red and exhibited a broad absorption spectrum with a maximum at around 425 nm, which was bleached by sodium dithionite, and shifted to 515 nm upon anaerobic 3,5-dichlorocatechol binding. The catalytic constant and the Km values for 3,5-dichlorocatechol and oxygen were 34.7 s-1 and 4.4 and 652 microM, respectively, at pH 8 and 25 degrees C. Some heavy metal ions, chelating agents and sulfhydryl reagents inhibited the activity. The NH2-terminal sequence was determined up to 44 amino acid residues and compared with those of the other catechol dioxygenases previously reported.

Triphasic vascular effects of thiol compounds and their oxidized forms on dog coronary arteries

The vascular effects of 2-mercaptoethanol, cysteamine, L-cysteine, glutathione (GSH), cystamine and oxidized GSH (GSSG) on the isometric tension of isolated dog coronary arterial strips were examined, and these effects were compared with the triphasic response induced by dithiothreitol (DTT); a rapid and weak contraction (phase A), an intervening slow relaxation (phase B) and a slowly-developing strong contraction (phase C) which we previously reported. The responses of the arteries induced by 2-mercaptoethanol, cysteamine and L-cysteine consisted of phases A, B and C. The order of contractile potency (ED50 of phase C) was DTT approximately L-cysteine > 2-mercaptoethanol approximately cysteamine, while the order of relaxant potency (ED50 of phase B) was DTT > cysteamine approximately 2-mercaptoethanol. GSSG and cystamine mainly produced relaxation, which corresponded to phase B. The phase C contraction was specific to the reduced forms of thiols, except for GSH, which produced only relaxation. The participation of endothelial cells was not essential for the contracting or relaxing effects of the thiol compounds. The phase C contraction was depressed by W-7, a calmodulin antagonist, while phase A was not. Therefore calmodulin-dependent protein kinases may participate in phase C, not in phase A.

Urate oxidase activity and copper content in the liver of macular mutant mouse, a model animal for human congenital copper deficiency, Menkes' kinky hair disease

The macular mouse is an X-linked recessive inherited mutant and is considered to be a model for human congenital copper deficiency, Menkes' kinky hair disease. The activity of urate oxidase, which has been believed to be a copper enzyme, and copper content in the liver of the mutant mouse were determined. The oxidase activity was maintained at normal level even though there was very low level of copper present in the liver through days 7 to 14. Copper administration increased the copper content in the liver to the normal level, but did not affect the oxidase activity.

Immunochemical relationship of D-amino acid oxidases in various tissues and animals

1. By means of an enzyme immunoassay, the contents of D-amino acid oxidase (DAO) were determined in kidney, liver, cerebellum and lung of hog, but the oxidase was not detectable in heart or cerebrum. 2. The oxidases in kidney, liver and cerebellum of hog were indistinguishable as regards immunoreactivity toward anti-hog kidney DAO antibody, specific activity and molecular weight. 3. The oxidases in rat and dog kidneys immunochemically cross-reacted with anti-hog DAO antibody. 4. The overall structure of the hog oxidase was more similar to that of the dog enzyme than that of the rat, while the structure around the catalytic site of the hog oxidase was more similar to that of the rat oxidase than that of the dog enzyme. 5. On immunoblot analysis, two forms of the oxidase were detected in extracts of hog, rat and dog kidneys.

Three isozymes of catechol 1,2-dioxygenase (pyrocatechase), alpha alpha, alpha beta, and beta beta, from Pseudomonas arvilla C-1

Three isozymes of catechol 1,2-dioxygenase (pyrocatechase) from Pseudomonas arvilla C-1 were separated using DEAE-Toyopearl chromatography. The specific activities of each isozyme were similar to one another. The molecular weights of isozymes 1, 2, and 3 were estimated to be approximately 67,000, 64,000, and 59,000, respectively, from gel filtration. On sodium dodecyl sulfate-polyacrylamide gel electrophoresis, isozymes 1 and 3 gave a single protein band, corresponding to Mr = 32,000 and 30,000, respectively, and isozyme 2 gave two bands corresponding to Mr = 32,000 and 30,000. These results indicated that isozymes 1 and 3 were homodimers, while isozyme 2 was a heterodimer. The NH2-terminal sequences up to 20 residues of these three isozymes confirmed that isozymes 1, 2, and 3 consisted of beta beta, alpha beta, and alpha alpha, respectively, based on our previous data (Nakai, C., Kagamiyama, H., Saeki, Y., and Nozaki, M. (1979) Arch. Biochem. Biophys. 195, 12-22). Properties of these isozymes such as absorption spectrum, iron content, substrate specificity, and kinetic constants were similar to one another. Subunit exchange between the different isozymes and dissociation of the isozymes into subunits was not observed under nondenaturing conditions. Available evidence indicates that these isozymes exist naturally in the bacterium and were not due to artifacts caused by purification.