In vitro incorporation of molybdate into demolybdoproteins in Escherichia coli

NarJ is a specific chaperone required for molybdenum cofactor assembly in nitrate reductase A of Escherichia coli

The formation of active membrane-bound nitrate reductase A in Escherichia coli requires the presence of three subunits, NarG, NarH and NarI, as well as a fourth protein, NarJ, that is not part of the active nitrate reductase. In narJ strains, both NarG and NarH subunits are associated in an unstable and inactive NarGH complex. A significant activation of this complex was observed in vitro after adding purified NarJ-6His polypeptide to the cell supernatant of a narJ strain. Once the apo-enzyme NarGHI of a narJ mutant has become anchored to the membrane via the NarI subunit, it cannot be reactivated by NarJ in vitro. NarJ protein specifically recognizes the catalytic NarG subunit. Fluorescence, electron paramagnetic resonance (EPR) spectroscopy and molybdenum quantification based on inductively coupled plasma emission spectroscopy (ICPES) clearly indicate that, in the absence of NarJ, no molybdenum cofactor is present in the NarGH complex. We propose that NarJ is a specific chaperone that binds to NarG and may thus keep it in an appropriate competent-open conformation for the molybdenum cofactor insertion to occur, resulting in a catalytically active enzyme. Upon insertion of the molybdenum cofactor into the apo-nitrate reductase, NarJ is then dissociated from the activated enzyme.

Molybdenum cofactor requirement for in vitro activation of apo-molybdoenzymes of Escherichia coli

The apo-nitrate reductase precursor in an Escherichia coli chlB mutant preparation obtained following growth in the presence of tungstate is activated by incubation with protein FA and a heat-treated preparation from an E. coli crude extract. We show that the requirement for heat-treated E. coli crude extract can be fulfilled by material obtained from either of two heat-denatured purified E. coli molybdoenzymes, namely nitrate reductase or trimethylamine N-oxide reductase. Apo-trimethylamine N-oxide reductase precursor in the tungstate-grown chlB preparation can be activated in a similar manner with material from either heat-denatured molybdoenzyme. The active component in the denatured molybdoenzyme preparations is shown to be the molybdenum cofactor by Neurospora crassa nit1 molybdenum cofactor assay, size estimation and fluorimetric analysis. The direct demonstration of the requirement for molybdenum cofactor in the E. coli tungstate-grown chlB complementation system is an important step towards the molecular definition of the activation process and an understanding of the mechanism of cofactor acquisition during molybdoenzyme biosynthesis.

Involvement of a low-molecular-weight substance in in vitro activation of the molybdoenzyme respiratory nitrate reductase from a chlB mutant of Escherichia coli

The soluble subcellular fraction of a chlB mutant contains an inactive precursor form of the molybdoenzyme nitrate reductase, which can be activated by the addition to the soluble fraction of protein FA, which is thought to be the active product of the chlB locus. Dialysis or desalting of the chlB soluble fraction leads to the loss of nitrate reductase activation, indicating that some low-molecular-weight material is required for the activation. The protein FA-dependent activation of nitrate reductase can be restored to the desalted chlB soluble fraction by the addition of a clarified extract obtained after heating the chlB soluble fraction at 100 degrees C for 8 min. The heat-stable substance present in this preparation has a molecular weight of approximately 1,000. This substance is distinct from the active molybdenum cofactor since its activity is unimpaired in heat-treated extracts prepared from the organism grown in the presence of tungstate, which leads to loss of cofactor activity. Mutations at the chlA or chlE locus, which are required for molybdenum cofactor biosynthesis, similarly do not affect the activity of the heat-treated extract in the in vitro activation process. Moreover, the active material can be separated from the molybdenum cofactor activity by gel filtration. None of the other known pleiotropic chlorate resistance loci (chlD, chlG) are required for the expression of its activity. Magnesium ATP appears to have a role in the formation of the active substance. We conclude that a low-molecular-weight substance, distinct from the active molybdenum cofactor, is required to bestow activity on the molybdoenzyme nitrate reductase during its biosynthesis.

Activation in vitro of respiratory nitrate reductase of Escherichia coli K12 grown in the presence of tungstate. Involvement of molybdenum cofactor

The chlorate-resistant (chlR) mutants are pleiotropically defective in molybdoenzyme activity. The inactive derivative of the molybdoenzyme, respiratory nitrate reductase, present in the cell-free extract of a chlB mutant, can be activated by the addition of protein FA, the probable active product of the chlB locus. Protein FA addition, however, cannot bring about the activation if 10 mM sodium tungstate is included in the culture medium for the chlB strain. The inclusion of a heat-treated preparation of a wild-type or chlB strain prepared after growth in the absence of tungstate, restores the protein-FA-dependent activation of nitrate reductase. All attempts to activate nitrate reductase in extracts prepared from tungstate-grown wild-type Escherichia coli strains failed. It appears that during growth with tungstate, the possession of the active chlB gene product leads to the synthesis of a nitrate reductase derivative which is distinct from that present in the tungstate-grown chlB mutant. Heat-treated preparations from chlA and chlE mutants which do not possess molybdenum cofactor activity fail to restore the activation. Fractionation by gel filtration of the heat-treated preparation from a wild-type strain produced two active peaks in the eluate of approximate Mr 12000 and less than or equal to 1500. The active material in the heat-treated extract was resistant to exposure to proteinases, but after such treatment the active component, previously of approximate Mr 12000, eluted from the gel filtration column with the material of Mr less than or equal to 1500. The active material is therefore of low molecular mass and can exist either in a protein-bound form or in an apparently free state. Molybdenum cofactor activity, assayed by the complementation of the apoprotein of NADPH:nitrate oxidoreductase in an extract of the nit-1 mutant of Neurospora crassa, gave a profile following gel filtration similar to that of the ability to restore respiratory nitrate reductase activity to the tungstate-grown chlB mutant soluble fraction. This was the case even after proteinase treatment of the heat-stable fraction. Analysis of the chlC (narC) mutant, defective in the structural gene for nitrate reductase, revealed that heat treatment is not necessary for the expression of the active component. Furthermore both the active component and molybdenum cofactor activity are present in corresponding bound and free fractions in the non-heat-treated soluble subcellular fraction.(ABSTRACT TRUNCATED AT 400 WORDS)

Evidence for gene sharing in the nitrate reduction systems of Pseudomonas aeruginosa

Pseudomonas aeruginosa mutants unable to assimilate or dissimilate nitrate were isolated. Transduction and reversion analyses of these mutants revealed that single genetic lesions are responsible for the double phenotypes. The mutants were divided into two classes based on the ability to utilize hypoxanthine. It can be concluded from this study that at least two genes are shared between the two nitrate reduction systems.

Properties of dissimilatory nitrate reductase purified from the denitrifier Pseudomonas aeruginosa

Dissimilatory nitrate reductase was purified to homogeneity from anaerobic cultures of the denitrifying bacterium Pseudomonas aeruginosa. The following procedures were used in the rapid isolation of this unstable enzyme: induction by nitrate in semianaerobic cell suspension, heat-stimulated activation and solubilization from the membrane fraction, and purification by hydrophobic interaction chromatography. The molecular weight of the purified enzyme was estimated by nondenaturing polyacrylamide gel electrophoresis, sucrose density gradient sedimentation, and gel filtration chromatography. Subunit molecular weights were estimated by electrophoresis in sodium dodecyl sulfate-polyacrylamide gels. The active enzyme monomer, with a molecular weight of 176,000 to 260,000 (depending upon the method of determination), was composed of subunits with molecular weights of approximately 64,000 and 118,000. The monomer aggregated to form an inactive tetramer of about 800,000 molecular weight. Purified enzyme exhibited a broad pH optimum, between 6.5 and 7.5. Kinetic studies showed that the apparent Km was 0.30 mM for nitrate, and 2.2 to 2.9 microM for dithionite-reduced benzyl viologen. Azide was an effective inhibitor: the concentration required for half-maximal inhibition was 21 to 24 microM. Azide inhibition was competitive with nitrate (Ki = 2.0 microM) but uncompetitive with reduced benzyl viologen (Ki = 25 microM). Based upon spectral evidence, the purified molybdo-enzyme had no associated cytochromes but did contain nonhaem iron that responded to dithionite reduction and nitrate oxidation. The enzyme that was purified after being heat solubilized from membranes had properties essentially identical to those of the enzyme that was purified after deoxycholate solubilization.

Molybdenum cofactor requirement for biotin sulfoxide reduction in Escherichia coli

The bisC gene of Escherichia coli is tentatively identified as the structural gene for biotin sulfoxide reductase by the isolation of bisC(Ts) mutants that make thermolabile enzyme. The products of four other E. coli genes (chlA, chlB, chlE and chlG) are also needed for enzymatic activity. Mutations previously assigned to the bisA, bisB, and bisD genes belong to genes chlA, chlE, and chlG, respectively. The biotin sulfoxide reductase deficiency of a chlG, mutant is partially reversed by the addition of 10 mM molybdate to the growth medium. Mutational inactivation of the chlD gene reduces the specific activity of biotin sulfoxide reductase about twofold. This effect is reversed by the addition of 1 mM molybdate to the growth medium. The specific activity of biotin sulfoxide reductase is decreased about 30-fold by the presence of tungstate in the growth medium, an effect that has been observed previously with nitrate reductase and other molybdoenzymes. The specific activity of biotin sulfoxide reductase is not elevated in a lysate prepared by derepressing a lambda cI857 chlG prophage. Whereas biotin sulfoxide reductase prepared by sonic extraction of growing cells is almost completely dependent on the presence of a small heat-stable protein resembling thioredoxin, much of the enzyme obtained from lysates of thermoinduced lambda cI857 lysogens does not require this factor.

Kinetic Explanation for Accumulation of Nitrite, Nitric Oxide, and Nitrous Oxide During Bacterial Denitrification

The kinetics of denitrification and the causes of nitrite and nitrous oxide accumulation were examined in resting cell suspensions of three denitrifiers. An Alcaligenes species and a Pseudomonas fluorescens isolate characteristically accumulated nitrite when reducing nitrate; a Flavobacterium isolate did not. Nitrate did not inhibit nitrite reduction in cultures grown with tungstate to prevent formation of an active nitrate reductase; rather, accumulation of nitrite seemed to depend on the relative rates of nitrate and nitrite reduction. Each isolate rapidly reduced nitrous oxide even when nitrate or nitrite had been included in the incubation mixture. Nitrate also did not inhibit nitrous oxide reduction in Alcaligenes odorans , an organism incapable of nitrate reduction. Thus, added nitrate or nitrite does not always cause nitrous oxide accumulation, as has often been reported for denitrifying soils. All strains produced small amounts of nitric oxide during denitrification in a pattern suggesting that nitric oxide was also under kinetic control similar to that of nitrite and nitrous oxide. Apparent K m values for nitrate and nitrite reduction were 15 μM or less for each isolate. The K m value for nitrous oxide reduction by Flavobacterium sp. was 0.5 μM. Numerical solutions to a mathematical model of denitrification based on Michaelis-Menten kinetics showed that differences in reduction rates of the nitrogenous compounds were sufficient to account for the observed patterns of nitrite, nitric oxide, and nitrous oxide accumulation. Addition of oxygen inhibited gas production from 13 NO 3 − by Alcaligenes sp. and P. fluorescens , but it did not reduce gas production by Flavobacterium sp. However, all three isolates produced higher ratios of nitrous oxide to dinitrogen as the oxygen tension increased. Inclusion of oxygen in the model as a nonspecific inhibitor of each step in denitrification resulted in decreased gas production but increased ratios of nitrous oxide to dinitrogen, as observed experimentally. The simplicity of this kinetic model of denitrification and its ability to unify disparate observations should make the model a useful guide in research on the physiology of denitrifier response to environmental effectors.

Identification of the molybdenum cofactor in chlorate-resistant mutants of Escherichia coli

Experiments were performed to determine whether defects in molybdenum cofactor metabolism were responsible for the pleiotropic loss of the molybdoenzymes nitrate reductase and formate dehydrogenase in chl mutants of Escherichia coli. In wild-type E. coli, molybdenum cofactor activity was present in both the soluble and membrane-associated fractions when the cells were grown either aerobically or anaerobically, with and without nitrate. Molybdenum cofactor in the soluble fraction decreased when the membrane-bound nitrate reductase and formate dehydrogenase were induced. In the chl mutants, molybdenum cofactor activity was found in the soluble fraction of chlA, chlB, chlC, chlD, chlE, and chlG, but only chlB, chlC, chlD, and chlG expressed cofactor activity in the membrane fraction. The defect in the chlA mutants which prevented incorporation of the soluble cofactor into the membrane also caused the soluble cofactor to be defective in its ability to bind molybdenum. This cofactor was not active in the absence of molybdate, and it required at least threefold more molybdate than did the wild type in the Neurospora crassa nit-1 complementation assay. However, the cofactor from the chlA strain mediated the dimerization of the nit-1 subunits in the presence and absence of molybdate to yield the 7.9S dimer. Growth of chlA mutants in medium with increased molybdate did not repair the defect in the chlA cofactor nor restore the molybdoenzyme activities. Thus, molybdenum cofactor was synthesized in all the chl mutants, but additional processing steps may be missing in chlA and chlE mutants for proper insertion of cofactor in the membrane.

The influence of growth conditions on the synthesis of molybdenum cofactor in Proteins mirabilis

Cell-free extracts of Proteus mirabilis were able to reconstitute NADPH-dependent assimilatory nitrate reductase in crude extracts of the Neurospora crassa mutant strain nit-1, lacking molybdenum cofactor. Molybdenum cofactor was formed in the cytoplasm of the bacterium even in the presence of oxygen during growth though under these conditions no molybdo enzymes are formed. As a consequence no cofactor could be released by acid treatment from membranes of cells growth aerobically. The amount of cofactor released from membranes of cells grown anaerobically under various conditions was proportional to the amount of molybdo enzymes formed. During growth in the presence of tungstate a cofactor, which lacks molybdenum, was found in the cytoplasm. For detection of this so-called demolybdo cofactor the presence of molybdate during reconstitution was essential. Moreover, the cytoplasmic cofactor pool in cells grown in the presence of tungstate appeared to be two to three times higher than in cells grown under similar conditions without tungstate. After anaerobic growth in the presence of tungstate, the inactive demolybdo reductases were shown to contain partly no cofactor and partly a demolybdo cofactor. The P. mirabilis chlorate resistant mutant S 556 did not contain molybdenum cofactor. In two other chl-mutants the cofactor activity was the same as in the wild type.

Assimilatory nitrate uptake in Pseudomonas fluorescens studied using nitrogen-13

The mechanism of nitrate uptake for assimilation in procaryotes is not known. We used the radioactive isotope, 13N as NO3-, to study this process in a prevalent soil bacterium, Pseudomonas fluorescens. Cultures grown on ammonium sulfate or ammonium nitrate failed to take up labeled nitrate, indicating ammonium repressed synthesis of the assimilatory enzymes. Cultures grown on nitrite or under ammonium limitation had measurable nitrate reductase activity, indicating that the assimilatory enzymes need not be induced by nitrate. In cultures with an active nitrate reductase, the form of 13N internally was ammonium and amino acids; the amino acid labeling pattern indicated that 13NO3- was assimilated via glutamine synthetase and glutamate synthase. Cultures grown on tungstate to inactivate the reductase concentrated NO3- at least sixfold. Chlorate had no effect on nitrate transport or assimilation, nor on reduction in cell-free extracts. Ammonium inhibited nitrate uptake in cells with and without active nitrate reductases, but had no effect on cell-free nitrate reduction, indicating the site of inhibition was nitrate transport into the cytoplasm. Nitrate assimilation in cells grown on nitrate and nitrate uptake into cells grown with tungstate on nitrite both followed Michaelis-Menten kinetics with similar Km values, 7 muM. Both azide and cyanide inhibited nitrate assimilation. Our findings suggest that Pseudomonas fluorescens can take up nitrate via active transport and that nitrate assimilation is both inhibited and repressed by ammonium.

Nickel requirement for active hydrogenase formation in Alcaligenes eutrophus

The nickel-dependent chemolithoautotrophic growth of Alcaligenes eutrophus is apparently due to a requirement of nickel for active hydrogenase formation. Cells grown heterotrophically with fructose and glycerol revealed a specific activity of soluble and membrane-bound hydrogenase which was severalfold higher than the normal autotrophic level. The omission of nickel from the medium did not affect heterotrophic growth, but the soluble hydrogenase activity was reduced significantly. In the presence of ethylenediaminetetraacetic acid (EDTA), almost no hydrogenase activity was detected. The addition of nickel allowed active hydrogenase formation even when EDTA was present. When chloramphenicol was added simultaneously with nickel to an EDTA-containing medium, almost no hydrogenase activity was found. This indicates that nickel ions are involved in a process which requires protein synthesis and not the direct reactivation of a preformed inactive protein. The formation of the membrane-bound hydrogenase also appeared to be nickel dependent. Autotrophic CO2 assimilation did not specifically require nickel ions, since formate was utilized in the presence of EDTA and the activity of ribulosebisphosphate carboxylase was not affected under these conditions.

In vitro activation of inactive nitrogenase component I with molybdate

When Azotobacter vinelandii was derepressed for nitrogenase synthesis in the presence of WO42- rather than MoO42-, it synthesized active component II and inactive component I of nitrogenase. This inactive component I could be activated in vitro with the iron-molybdenum cofactor or with MoO42-. The latter reaction required adenosine 5'-triphosphate and was inhibited by adenosine 5'-diphosphate. FeMo cofactor and MoO42- produced different levels of activation, but there was no evidence that they acted upon different species of demolybdo component I. Rather, it may be that an additional factor necessary for MoO42-mediated activation but not for FeMo cofactor-mediated activation was limiting. Mo was inserted into component I during both FeMo cofactor- and MoO42- mediated activations.

Effects of molybdenum and tungsten on induction of nitrate reductase and formate dehydrogenase in wild type and mutant Paracoccus denitrificans

Molybdenum is required for induction of nitrate reductase and of NAD-linked formate dehydrogenase activities in suspensions of wild type Paracoccus denitrificans; tungsten prevents the development of these enzyme activities. The wild type forms a membrane protein Mr150,000 when incubated with tungsten and inducers of nitrate reductase and this is presumed to represent an inactive form of the enzyme. Suspensions of mutuant M-1 did not develop nitrate reductase or formate dehydrogenase activities but the membrane protein Mr150,000 was formed under all conditions tested, including without inducers and without molybdenum. Analysis of membranes, solubilized with deoxycholate, by polyacrylamide gel electrophoresis under nondenaturing conditions showed that the mutant protein had similar electrophoretic mobility to the active nitrate reductase formed by the wild type. Autoradiography of preparations from cells incubated with 55Fe showed that the mutant and wild type proteins contained iron. However, in similar experiments with 99Mo, incorporation of molybdenum into the mutant protein was not detectable. We conclude that mutant M-1 is defective in one or more steps required to process molybdenum for incorporation into molybdoenzymes. This failure affects the normal regulation of nitrate reductase protein with respect to the role of inducers.