Azorhizobium caulinodans pyruvate dehydrogenase activity is dispensable for aerobic but required for microaerobic growthb

David C. Pauling1, Jerome P. Lapointe1, Carolyn M. Paris1 and Robert A. Ludwig1

Department of MCD Biology, Sinsheimer Laboratories, University of California, Santa Cruz, CA 95064, USA1

Author for correspondence: Robert A. Ludwig. Tel: +1 831 459 4084. e-mail: ludwig{at}darwin.ucsc.edu


   ABSTRACT
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Azorhizobium caulinodans mutant 62004 carries a null allele of pdhB, encoding the E1ß subunit of pyruvate dehydrogenase, which converts pyruvate to acetyl-CoA. This pdhB mutant completely lacks pyruvate oxidation activities yet grows aerobically on C4 dicarboxylates (succinate, L-malate) as sole energy source, albeit slowly, and displays pleiotropic growth defects consistent with physiological acetyl-CoA limitation. Temperature-sensitive (ts), conditional-lethal derivatives of the pdhB mutant lack (methyl)malonate semialdehyde dehydrogenase activity, which thus also allows L-malate conversion to acetyl-CoA. The pdhB mutant remains able to fix N2 in aerobic culture, but is unable to fix N2 in symbiosis with host Sesbania rostrata plants and cannot grow microaerobically. In culture, A. caulinodans wild-type can use acetate, ß-D-hydroxybutyrate and nicotinate – all direct precursors of acetyl-CoA – as sole C and energy source for aerobic, but not microaerobic growth. Paradoxically, acetyl-CoA is thus a required intermediate for microaerobic oxidative energy transduction while not itself oxidized. Accordingly, A. caulinodans energy transduction under aerobic and microaerobic conditions is qualitatively different.

Keywords: (methyl)malonate semialdehyde dehydrogenase, microaerobiosis, Rhizobiaceae, microaerophilic bacteria

Abbreviations: DOT, dissolved O2 tension; MSDH, (methyl)malonate semialdehyde dehydrogenase (acylating); PDH, pyruvate dehydrogenase; PHB, poly-ß-hydroxybutyrate

b The GenBank accession number for the sequence determined in this work is AF299324.


   INTRODUCTION
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Microaerophilic bacteria, an evolutionarily diverse lot, share the remarkable ability to grow oxidatively in the presence of nanomolar O2 concentrations. Such microaerophiles include the family Rhizobiaceae, commonly referred to as rhizobia, which form symbiotic root associations (nodules) with host legume plants. During nodule development, rhizobia invade and fill, by endocytosis, proliferating host cells derived from root cortical tissue; these symbiotic nodules fix atmospheric N2 at high rates. To do so, nodules contain excess leghaemoglobin, which buffers O2, allowing free O2 levels to be maintained at nanomolar, steady-state levels. In this microaerobic environment, endosymbiotic rhizobia employ ultrahigh-affinity cytochrome oxidases that maintain high oxidative phosphorylation rates in spite of nanomolar O2 (Bergersen et al., 1986 ). As nodule free O2 is vanishingly low, the O2-labile dinitrogenase complex within endosymbiotic rhizobia is presumably physiologically protected.

Uniquely among rhizobia, Azorhizobium caulinodans fixes N2 both in aerobic culture and in microaerobic symbiosis with the host legume Sesbania rostrata (Dreyfus & Dommergues, 1981 ). Optimal free O2 levels for these processes span some four orders of magnitude. In fully aerobic culture, where dissolved O2 tension (DOT) approximates physiological saturation (<=200 µM), A. caulinodans grows vigorously. At 10 µM DOT, A. caulinodans optimally fixes N2. Under true microaerobic conditions (<1 µM DOT), A. caulinodans still grows avidly. At 10 nM DOT in S. rostrata nodules, endosymbiotic A. caulinodans rapidly fixes N2.

For these, various, physiological processes, A. caulinodans, like all microaerophilic rhizobia, prefers as C and energy source the common C4 dicarboxylates succinate and L-malate. Under aerobic conditions, succinate and L-malate are conventionally metabolized via citric acid cycle activity. However, in response to nanomolar DOT, symbiotic nodules require a division of metabolic labour. Infected host plant tissues strictly ferment primary photosynthate (sugars), yielding L-malate; endosymbiotic rhizobia take up and oxidize this L-malate as energy source. Indeed, Rhizobium spp. mutants which lack C4 dicarboxylate active transport capabilities cannot sustain symbiotic N2 fixation (Ronson et al., 1981 ; Finan et al., 1981 , 1983 ).

In microaerophilic bacteria, how does metabolic energy transduction adapt when DOT levels change by several orders of magnitude? To help address this question, we sought, identified and studied A. caulinodans mutants defective for microaerobic energy transduction. One such mutant carries a null allele for the E1ß subunit of the pyruvate dehydrogenase (PDH) complex, long considered indispensable to obligate aerobic organisms that lack distinct pyruvate decarboxylase activity. Aerobic growth on sugars and organic acids as C and energy source yields pyruvate as oxidative intermediate. PDH complex activity oxidizes pyruvate using NAD+ as oxidant, yielding acetyl-CoA. Accordingly, PDH complex activity is often referred to as the ‘priming step’ for citric acid cycle activity, which employs acetyl-CoA as substrate. Surprisingly, in A. caulinodans, PDH activity is indispensable for microaerobic growth but is dispensable for aerobic growth. As corroborated by additional physiological studies, the aerobic and microaerobic energy transduction of A. caulinodans are indeed qualitatively different.


   METHODS
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Bacterial strains and culture methods.
A library of Azorhizobium caulinodans vector-insertion (Vi) strains (Donald et al., 1985 , 1986 ) was screened for mutants with phenotypes of interest. Escherichia coli strain BNN45 was used for bacteriophage {lambda} propagation; E. coli strain DH5{alpha} was used for transformations and plasmid preparations (Table 1). A. caulinodans rich (GYPC) medium comprises: 10 mM potassium phosphate pH 6·3, 0·2% salt-free casein acid hydrolysate, 0·2% yeast extract, 0·4% D-glucose. Defined minimal medium (ORS-MM) comprises: 10 mM potassium phosphate pH 6·3, 15 mM (NH4)2SO4, 1 mM MgSO4, 0·5 mM CaCl2 16 µg nicotinate ml-1, 1·0 µg pantothenate ml-1, 0·2 µg ml-1 biotin. N2 fixation (NIF) salts comprises: 50 mM potassium phosphate pH 6·4, 1·25 mM MgSO4, 300 µM potassium nicotinate, 0·5 mM CaCl2, 1 µg NaMoO4 ml-1, 1 µg FeCl3 ml-1. E. coli strains were cultured on Luria–Bertani (LB) broth: 1% tryptone, 0·5% yeast extract, 0·5% sodium chloride. For liquid batch culture experiments, test strains were first cultured in GYPC medium. Cells were harvested in mid-exponential phase, washed, and resuspended at 5x107 cells ml-1 in ORS-MM supplemented with the indicated C source and N source, and 100 µM nicotinate. Culture growth was monitored spectrophotometrically (OD600). To measure N2-dependent growth, A. caulinodans strains were first cultured in liquid ORS-MM supplemented with 10 mM ammonium and either 0·4% (w/v) succinate or 0·4% acetate under air sparge to late-exponential phase (OD600 0·7). Bacterial cells were recovered by centrifugation, washed twice, and diluted to 1x108 cells ml-1 (OD600 0·05) in NIF medium containing the indicated C source at 40 meq l-1. Cell samples were placed in stoppered serum vials and again cultured at 30 °C under sparge with 1% O2/1% CO2/98% N2. To establish microaerobiosis, A. caulinodans strains were grown in liquid ORS-MM supplemented with the indicated C and energy source, nicotinate (100 µM) as vitamin, and ammonium (15 mM) as N source in air to a cell density of 1x108 cells ml-1 and then shifted to sparge with 0·1% O2/1·0% CO2/balance argon, all at 30 °C. For colony growth tests, media were solidified with acid-washed (1 M HCl) agarose, and plates were incubated in sealed jars at 30 °C under continuous sparge with 0·1% O2/1·0% CO2/balance argon. Bacterial colonies were examined after 14 d incubation.


View this table:
[in this window]
[in a new window]
 
Table 1. Bacterial strains, plasmids and phages

 
Temperature-sensitive mutants were selected by repetitive enrichment. Cultures were grown at 30 °C in ORS-MM containing 0·4% L-valine as sole C and N source, and nicotinate (16 µM) as vitamin source, to early exponential phase (1x108 cells ml-1) and shifted to 42 °C. After 30 min, Timentin (0·1 mg ml-1), a proprietary mixture of ticarcillin and ß-clavulanate, was added and cultures were incubated with vigorous agitation for 6 h at 42 °C. In control experiments, viable cell counts decreased some 103-fold after this treatment. Surviving cells were recovered by centrifugation, washed twice with phosphate buffer, and subcultured at 30 °C in ORS-MM containing 0·4% L-valine. Enrichments were repeated a total of four cycles. Surviving cells were then replica-plated on solid ORS-MM containing 0·4% L-valine as sole C and N source and (100 µM) nicotinate; plates were incubated at 30 and 42 °C, and temperature-sensitive mutants were identified and further characterized.

Molecular cloning and DNA sequencing.
For DNA sequencing purposes, the wild-type A. caulinodans pdhABC locus on pSKC4 was subcloned and sequenced by standard techniques (Sambrook et al., 1989 ). The multiple short sequences were assembled into a 3134 bp continuous sequence (GenBank accession no. AF299324), which was scanned for ORFs and then translated into presumed protein sequences, which were used to probe the SWISS-PROT database using the BLASTP program at the US National Center for Biotechnology Information; possible homologues identified by these searches were multiply aligned (PIMA; Smith & Smith, 1991 ).

2-Oxoacid and pyridine nucleotide pool measurements.
For 2-oxoacid quantitation, 2,4-dinitrophenylhydrazone derivatives from cleared, cell-free supernatants were prepared and separated by liquid chromatography (Supelco Gel C-610H) using 0·1% phosphoric acid as mobile phase; the flow rate was 0·5 ml min-1 at 30 °C, the spectrophotometric absorbance wavelength was 360 nm. Pyridine nucleotides were measured by modified cycling assays (Bernovsky & Swan, 1973 ). Culture samples were obtained at the indicated times, vigorously mixed with 0·5 vol. of either 7% perchloric acid (for total NAD) or 1 M NaOH (for NADH), heated for 10 min at 60 °C, and neutralized. For measurments of NADH, reactions included 40 mM Bicine pH 7·8, 0·5 mM EDTA, 15 mM phenazine methosulfate, 4 mM MTT [3-(4,5-dimethythiazol-2-yl)-2,5-diphenyltetrazolium bromide], 0·5 M ethanol and neutralized, cell-free alkali extract; reactions were conducted in the dark at 30 °C. For measurements of NAD+ by cycling, reactions contained neutralized, cell-free acid extract and were supplemented and initiated with 0·5 mg (200 U) purified Saccharomyces cerevisiae alcohol dehydrogenase. All reactions were monitored by spectrophotometric absorbance at 570 nm.

Pyruvate dehydrogenase assay.
Mid-exponential-phase cultures in ORS-MM were isolated and cell-free extracts were prepared by ultrasonication. Extracts were then cleared by centrifugation (1 h at 100000 g); extract protein concentrations were measured by standard Bradford assays. PDH activity at 30 °C was inferred as pyruvate-dependent reduction of NAD+ measured as spectrophotometric absorbance at 339 nm. Reactions comprised 0·1 M Tris/HCl pH 7·5, 10 mM MgCl2, 0·12 mM coenzyme A, 6 mM dithiothreitol, 1·2 mM NAD+, supernatant fractions of cell-free extracts (500 µg total protein); they were initiated with 10 mM pyruvate.

Cytochrome-dependent pyruvate oxidase assay.
Pyruvate-dependent cytochrome reduction was assayed by ferricyanide reduction. Batch cultures were grown on ORS-MM supplemented with 0·2% potassium succinate, 0·1% monosodium glutamate and 100 µM potassium nicotinate; early-exponential-phase cultures were harvested and washed; cell pellets were resuspended and lysed by ultrasonication. Crude cell-free extracts were then used to measure pyruvate-dependent reduction of ferricyanide at 30 °C measured as spectrophotometric absorbance at 420 nm. Reactions contained 0·1 M Tris/HCl pH 7·5, 10 mM MgCl2, 0·07 mM potassium ferricyanide and added crude cell-free extract (100 µg total protein), and were initiated with 20 mM pyruvate.

(Methyl)malonate semialdehyde dehydrogenase (acylating) assay.
Strains were cultured at 30 °C in ORS-MM supplemented with 0·4% L-valine as sole C and N source; mid-exponential-phase cultures were isolated and cell-free extracts were prepared by ultrasonication. Extracts were then cleared by centrifugation (1 h at 100000 g); extract protein concentrations were measured by standard Bradford assays. MSDH activity at 30 and 42 °C was inferred as (methyl)malonate semialdehyde-dependent reduction of NAD+. Reactions contained: 0·1 M Tris/HCl pH 8·5, 10 mM MgCl2, 0·20 mM coenzyme A, 1 mM dithiothreitol, cell-free extract (500 µg total protein), and either 10 mM malonate semialdehyde or 10 mM methylmalonate semialdehyde. These substrates were freshly prepared by acid hydrolysis of 3,3-diethoxypropionate or 3,3-diethoxyisobutyrate, respectively, as described by Hayaishi et al. (1961) . Reactions were initiated with addition of 1·5 mM NAD+ and monitored by spectrophotometric absorbance at 340 nm. Both activities were absolutely dependent upon added coenzyme A.

Dinitrogenase (N2ase) assay.
A. caulinodans N2 fixation was measured in culture by acetylene reduction and N2-dependent growth. Cells were grown to late-exponential phase in ORS-MM supplemented with either 0·4% succinate or 0·4% acetate as C source, 0·1% L-glutamate as N source, and 100 µM nicotinate, harvested, washed twice, resuspended (4x108 cells ml-1) in NIF medium, placed in stoppered serum vials and sparged with 3% O2/1% CO2/96% argon for 8 h at 30 °C. Vials were then injected with 0·2 atm acetylene (substrate), freshly generated by hydration of calcium carbide, and assayed for ethylene production versus time using gas chromatography (Donald et al., 1985 ).

Sesbania rostrata nodulation tests.
S. rostrata seedlings were germinated aseptically and grown on sterile, defined medium under N limitation (Kwon & Beevers, 1992 ). Three-week-old plants, about 50 cm in height, were inoculated with the desired A. caulinodans strain between the first and second stem internodes, for which region stem-nodule development is synchronized (Donald et al., 1986 ). Starting 6 d post-inculation, maturing stem-nodules were excised and tested for N2 fixation activity by acetylene reduction. Thereafter, nodules were recovered, crushed and homogenized (Polytron), and cell-free supernatants were analysed for leghaemoglobin, measured as spectrophotometric absorbance at 540 nm. All developmental nodulation tests were done in triplicate.


   RESULTS
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Molecular cloning of the mutated locus in A. caulinodans strain 62004
A genetic screen of A. caulinodans vector-insertion (Vi) mutants (Donald et al., 1986 ) defective for microaerobic growth on defined, minimal (ORS-MM) medium with 0·4% succinate added as sole C source (and with 16 µM added nicotinate) was conducted (see Methods). One identified strain, 62004, was unable to grow microaerobically and also yielded microcolonies under the permissive growth condition (aerobic culture on the same medium). Strain 62004 showed further, pleiotropic growth defects when cultured on defined medium with various C and N sources. From genomic DNA hybridization experiments, strain 62004 carried a single Vi insertion (data not presented). The DNA constituting this Vi insertion was physically isolated and circularized by treatment with DNA ligase (Donald et al., 1986 ), yielding recombinant plasmid pVi62004 (Table 1). Next, pVi62004 was used to screen, also by DNA hybridization, a library of recombinant {lambda}EMBL3 coliphage carrying A. caulinodans wild-type DNA insertions (Table 1). Among recombinant phage identified by DNA homology was {lambda}C4. As determined by physical mapping, {lambda}C4 carried a 20 kb A. caulinodans DNA insert which spanned the mutated locus in strain 62004. {lambda}C4 insert DNA was then subcloned in plasmid pSUP202, which stably replicates in A. caulinodans, yielding pSUPC4x-series plasmids (Table 1).

For genetic complementation tests, pSUPC4x-series plasmids were transmitted to A. caulinodans 62004 by intergeneric conjugation using, as agent, E. coli DH5{alpha}/pRK2073 (Table 1). Resulting A. caulinodans 62004/pSUPC4x partial diploids were tested for normal aerobic growth (large, opaque colonies) on defined (ORS-MM) salts plus 0·4% succinate as sole C and energy source, 15 mM as N source, and 16 µM nicotinate. In such complementation tests, several 62004/pSUPC4x partial diploid strains grew as wild-type. From physical analysis, all complementing, recombinant plasmids carried the same 3·1 kbp SalI DNA fragment derived from {lambda}C4. Accordingly, one representative plasmid, pSUPC4, was chosen for further study.

Structure of the A. caulinodans pdhABC locus
The complete sequence of the 3·1 kbp insert DNA of pSUPC4 was determined (see Methods). Four long ORFs were identified. Three of the ORFs shared substantial homology with pyruvate dehydrogenase subunits E1{alpha} and E1ß, as well as dihydrolipoamide S-acetyltransferase E2, all components of the PDH complex; ORF4 remained unidentified. The inferred A. caulinodans genes, respectively labelled pdhA, pdhB and pdhC, were tightly organized, presumably as a single operon (Fig. 1). The DNA sequence of recombinant pVi62004 (Table 1) was similarly determined. Co-alignment of the wild-type genome and recombinant pVi62004 DNA sequences placed the insertion mutation in strain 62004 at nucleotide 925 of the pdhB gene (pdhB4 allele; Fig. 1).



View larger version (7K):
[in this window]
[in a new window]
 
Fig. 1. Structure of the A. caulinodans pdhABC locus. The 3·1 kbp SalI DNA fragment complements the pdhB4 mutant. The position of the insertion mutation in the pdhB4 mutant is indicated. ORFs are discussed in the text.

 
The inferred A. caulinodans pdhA product, PDH E1{alpha}, and orthologues/homologues from various eukaryotic and prokaryotic species were analysed by multiple alignment. For the (Gram-negative) A. caulinodans E1{alpha}, the Zymomonas mobilis E1{alpha} orthologue, which shares 55% identity and 69% similarity, was best conserved. The alignment is shown in a supplementary figure with the online version of the paper at http://mic.sgmjournals.org. Interestingly, the next-nearest homologues included various eukaryotic and Gram-positive bacterial PDH E1 subunits, all of which share heterotetrameric E1{alpha}/E1ß structural organization. These PDH components all possess diagnostic thiamin-pyrophosphate-binding domains; in A. caulinodans E1{alpha} this domain spans residues 160–197. By contrast, other Gram-negative bacteria, including both E. coli and Alcaligenes eutrophus, possess a homodimeric PDH, whose single subunit shows considerable, but significantly less, homology with both A. caulinodans E1{alpha} and E1ß (data not presented).

A. caulinodans E1ß was likewise analysed. As with E1{alpha}, A. caulinodans E1ß was most similar to the Z. mobilis E1ß orthologue, sharing 62% identity and 75% similarity. The alignment is shown in a supplementary figure with the online version of the paper at http://mic.sgmjournals.org. Interestingly, A. caulinodans E1ß residues 4–87 define a lipoyl-binding domain of the type usually found with the eukaryotic-type E2, dihydrolipoamide S-acetyltransferase. Likewise, Z. mobilis E1ß also shows this domain organization. A. caulinodans E1ß also has a highly conserved E2 binding interface, spanning residues 163–197, again conserved in Z. mobilis E1ß. Distal to the A. caulinodans pdhB gene is a third ORF, the pdhC gene, whose inferred product carries a similar lipoyl-binding domain and which very likely encodes E2, dihydrolipoamide S-acetyltransferase, as is the case for Z. mobilis. The pdhC lipoyl-binding domain exhibits a high degree of homology (84% identity, 95% similarity) to amino acids 4–87 of E1ß as well as to the N-terminus of Z. mobilis E2. As evidenced by a high degree of sequence conservation, the A. caulinodans pdhABC locus, like that of Sinorhizobium meliloti (Cabanes et al., 2000 ) very likely encodes the first two catalytic activities of the PDH complex.

Genetic complementation and polarity
Transposon insertions frequently yield polar mutations that confound genetic complementation tests. As inferred from DNA sequence analysis, pSUPC4x plasmids carry an incomplete pdh operon yet complement strain 62004. So, either the genomic pdh locus carries internal promoter(s) distal to the pdhB4 insertion, or else VP2021 (8·7 kb) itself acts as mobile promoter. [Note that genomic VP2021 insertions carry direct, terminal IS50R element repeats (Donald et al., 1986 ).] To help distinguish these possibilities, strain 62004 derivatives in which the complex VP2021 insertion was resolved as a simple IS50R insertion were identified (Loroch et al., 1995 ). Such strains presumably arise by homologous IS50R excisive recombination. One resulting strain, 62004R (Table 1), now carrying the pdhB4R allele, was similarly tested for genetic complementation. The aerobic growth phenotype of strain 62004R/pSUPC4x partial diploids was intermediate between the 62004 parent and wild-type; some aerobic growth enhancement was observed when cells were tested with succinate as sole C and energy source. Therefore, internal VP2021 sequences might contain mobile promoter(s) when inserted and specifically oriented toward distal genes.

The A. caulinodans pdhB4R mutant completely lacks pyruvate dehydrogenase activity
For obligate aerobes such as A. caulinodans, PDH activity is considered indispensable for growth on both sugars and organic acids not directly yielding acetyl-CoA. Yet, for the pdhB4R mutant, the IS50R insertion presumably yielded a null allele. Accordingly, PDH catalytic activities for both wild-type and the pdhB4R mutant were measured and compared (see Methods). For these enzyme assays, strains were cultured aerobically on minimal, defined medium with succinate as sole C and energy source (which presumably requires PDH activity for synthesis of acetyl-CoA as citric acid cycle substrate). In cell-free extracts, pyruvate-dependent initial velocities of 0·29 µmol NADH min-1 (mg protein)-1 for wild-type and -0·03 µmol NADH min-1 (mg protein)-1 for the pdhB4R mutant were obtained. In this, as in all other similar assays, the pdhB4R mutant lacked detectable PDH activity. In accord with both biochemical and genetic evidence, the pdhB4R allele is, therefore, null.

The A. caulinodans pdhB4R mutant shows no cytochrome-dependent pyruvate oxidation
To help reconcile both the viability of the pdhB4R mutant and its lack of PDH activity, A. caulinodans was tested for cytochrome-dependent pyruvate oxidase activity. The archetypal E. coli POXB pyruvate oxidase activity is membrane associated (but not integral), yields CO2 and free acetate, and transfers reducing equivalents to bacterial respiration (Koland et al., 1984 ). Accordingly, crude cell extracts of both A. caulinodans wild-type and the pdhB4R mutant were tested for pyruvate oxidase activity, measured as pyruvate-dependent reduction of ferricyanide (Methods). Wild-type extracts showed pyruvate dependent ferricyanide reduction at high initial rates [0·35 µmol ferrocyanide min-1 (mg protein)-1], whereas the pdhB4R mutant showed no detectable pyruvate dependent ferricyanide reduction [<0·01 µmol ferrocyanide min-1 (mg protein)-1]. Thus, the pdhB4R mutant completely lacks discrete pyruvate oxidase activity. In wild-type crude cell extracts, the observed rates for ‘pyruvate oxidase’ activity may be attributable to coupled PDH and NADH-CoQ oxidoreductase activities. In summary, the pdhB4R mutant completely lacks any catalytic activity mediating pyruvate oxidation. Therefore, A. caulinodans PDH activity is dispensable for aerobic growth.

The alternative process for conversion of succinate to acetyl-CoA requires malonate semialdehyde dehydrogenase (acylating) activity
The pdhB4R mutant remains able to grow aerobically with C4 dicarboxylates (succinate, L-malate) as sole C and energy source, although growth is poor. In the absence of PDH activity, how is succinate converted to acetyl-CoA as substrate for citric acid cycle activity? Accordingly, conditional-lethal derivatives of strain 62004R carrying spontaneous mutants were isolated by penicillin-type enrichment. Strain 62004R was repeatedly subcultured on defined medium with, as sole C and N source, L-valine, on which A. caulinodans wild-type grows reasonably well, and in the presence of Timentin (see Methods) as selective agent. In these enrichment cultures, 30 °C was used as permissive and 42 °C as non-permissive temperature (see Methods). From 10 parallel culture-series, 51 temperature-sensitive (ts) spontaneous mutants that completely failed to grow on L-valine at 42 °C, but grew normally at 30 °C, were identified.

Upon replica-plate testing of these 51 candidates, 12 ts strains also failed to grow on defined ORS-MM medium with succinate and ammonium at 42 °C. To each of these 12 conditional-lethal mutants, plasmid pSUPC4 was introduced by intergeneric conjugation, as previously described. Resulting pdh partial-diploids were retested for growth on defined medium supplemented with succinate and ammonium, or alternatively with L-valine as sole C and N source, both at 42 °C. Two ts candidates, strains 62042 and 62047 (Table 1), recovered some growth on succinate and ammonium when carrying pSUPC4 (recall that pSUPC4 only weakly complements strain 62004R), but remained completely unable to grow with L-valine as sole C and N source at 42 °C; the other ts mutants remained conditional-lethal for all growth at 42 °C when carrying pSUPC4. Together with wild-type, strains 62042 and 62047 were tested for MSDH activity at both 30 and 42 °C (see Methods). For both ts mutants, MSDH activity was present at wild-type levels at 30 °C but was completely lost at 42 °C (Table 2). In subsequent assays, malonate semialdehyde was substituted for methylmalonate semialdehyde as substrate. Similar activities were recorded; again, both ts mutants specifically lacked malonate semialdehyde-dependent activity at 42 °C (Table 2). With both malonate semialdehyde and methylmalonate semialdehyde as substrates, observed activities were completely dependent on added coenzyme A; hence oxidative MSDH activity is acylating, yielding acetyl-CoA. In A. caulinodans wild-type, the relatively low MSDH activities measured (in comparison to PDH activities) may be responsible for the poor growth of the pdhB4 mutant on C4 dicarboxylates as sole C and energy source. Thus, MSDH activity also facilitates conversion of succinate to acetyl-CoA. Specifically, L-malate, as produced from succinate, might be diverted according to the following pathway (see Fig. 2):


View this table:
[in this window]
[in a new window]
 
Table 2. MSDH activities of cell-free extracts from A. caulinodans strains cultured on L-valine as sole C and N source at 30 °C

 


View larger version (12K):
[in this window]
[in a new window]
 
Fig. 2. A. caulinodans alternative pathway for the conversion of succinate or L-malate to acetyl-CoA. Relevant activities include: malate dehydrogenase, aspartate aminotransferase, aspartate {alpha}-decarboxylase, ß-alanine aminotransferase and (methyl)malonate semialdehyde dehydrogenase (acylating). Note: if the inferred ß-alanine aminotransferase prefers pyruvate as oxoacid substrate, then L-alanine aminotransferase activity must also contribute to this proposed pathway.

 

Relevant activities include: NAD-malate dehydrogenase, aspartate aminotransferase, aspartate {alpha}-decarboxylase, ß-alanine aminotransferase (Hayaishi et al., 1961 ) and MSDH (acylating).

The pdhB4R mutant displays growth properties consistent with physiological acetyl-CoA limitation
General patterns of C source utilization for both A. caulinodans wild-type and the pdhB4R mutant were compared. To mitigate NAD+ limitation with pdhB4R mutant cultures (see below), nicotinate levels of liquid batch cultures in defined media were increased to 100 µM (see Methods). Because growth yields for both wild-type and the pdhB4R mutant cultured without additional N source(s) were very low, 100 µM nicotinate as N source was still insufficient to sustain normal culture growth. In subsequent experiments, 15 mM ammonium as N source was included, and various C sources were tested. Both wild-type and the pdhB4R mutant were first cultured to mid-exponential phase in rich (GYPC) medium; cells were harvested by centrifugation, washed twice with phosphate buffer, and resuspended in defined (ORS-MM) medium containing various C sources and at cell densities of 5x107 ml-1. Subcultures were then vigorously aerated at 30 °C. Growth was monitored both spectrophotometrically by light scattering and by viable cell counts (see Methods). When wild-type was shifted to succinate, a preferred C and energy source, exponential growth resumed quickly; whereas with acetate as C source, resumption of growth was delayed some 16 h. By contrast, when the pdhB4R mutant was shifted to succinate, growth was delayed some 24 h and then very slow, whereas when the pdhB4R mutant was shifted to acetate, growth was decidedly faster. Because it adapts much more quickly than wild-type when shifted from rich medium to defined medium with acetate as sole C source, the pdhB4R mutant seems acetyl-CoA limited on rich medium. In contrast to the pdhB4R mutant, the ability of A. caulinodans wild-type to use acetate and other secondary C sources which yield acetyl-CoA independent of pyruvate seems inducible.

When cell doubling times were measured in mid-exponential phase after shift from rich to defined, minimal medium, growth responses were cleanly divided between preferred energy substrates (metabolically) upstream of PDH activity (succinate, L-malate, L-lactate) and those downstream (acetate, nicotinate and ß-D-hydroxybutyrate). The former were rapidly used by wild-type but slowly utilized by the pdhB4 mutant (Table 3).


View this table:
[in this window]
[in a new window]
 
Table 3. Effect on doubling time of shift from rich to minimal medium

 
Similar results were obtained with, as C source, high levels of nicotinate, whose catabolism directly yields (two equivalents of) acetyl-CoA. All wild A. caulinodans isolates are pyridine nucleotide (NAD+) auxotrophs and require nicotinate as vitamin precursor for growth in defined media (Dreyfus & Dommergues, 1981 ). During physiological C or N limitation, A. caulinodans also catabolizes nicotinate at high rates as both C and N source, yielding acetyl-CoA and ammonium (Kitts et al., 1989 , 1992 ; Buckmiller et al., 1991 ). Then, A. caulinodans cultures may exhaust nicotinate, become pyridine nucleotide limited, and cease further growth (Ludwig, 1986 ). Accordingly, nicotinate utilization was compared in wild-type and the pdhB4R mutant. Nicotinate catabolism was highly active in wild-type cultures fed either succinate or L-malate – preferred C and energy sources – but was repressed/inhibited by addition of either acetate, a secondary C source, or L-glutamine, the preferred N source (Fig. 3a): Like wild-type, pdhB4R mutant cultures showed repression/inhibition of nicotinate catabolism when supplemented with acetate. However, the pdhB4R mutant catabolized nicotinate at high rates with excess L-glutamine as N source (Fig. 3b). By inference, in the pdhB4R mutant, metabolic control of use of nicotinate as C source (yielding acetyl-CoA) was epistatic to control of its use as N source. Therefore, when grown on primary energy sources such as succinate, malate or lactate, the pdhB4R mutant is normally acetyl-CoA limited.



View larger version (16K):
[in this window]
[in a new window]
 
Fig. 3. Nicotinate depletion by A. caulinodans wild-type (a) and the pdhB4R mutant (b) in batch culture. Batch cultures were shifted at time zero from rich (GYPC) to defined, minimal (ORS-MM) medium supplemented with: 40 mM succinate and 10 mM ammonium ({bullet}); 20 mM succinate, 20 mM acetate and 10 mM ammonium ({blacksquare}); 40 mM succinate and 10 mM glutamine ({blacktriangleup}).

 
Both A. caulinodans wild-type and the pdhB4R mutant fix N2 in culture at 10 µM DOT
Remarkably, A. caulinodans wild-type fixes N2 both in culture under 10 µM DOT and in symbiosis at 10 nM DOT. The importance of PDH activity to the energetics of N2 fixation in culture at 10 µM DOT was assessed through physiological shift experiments. Starter cultures were established in defined liquid media containing fixed N source(s) under air sparge and, early in exponential phase, were shifted to defined medium devoid of N source under 1% O2 sparge. Consequently, cultures became N depleted, DOT declined to approximately 10 µM, and N2ase activity was then induced (see Methods). With succinate as sole C and energy source, in wild-type cultures N2ase activities were 556 nmol ethylene h-1 per 5x108 cells, whereas in the pdhB4R mutant cultures N2ase activities were 251 nmol ethylene h-1 per 5x108 cells. Again under these conditions, the pdhB4R mutant poorly used succinate as C and energy source. By contrast, both strains showed similar N2ase activities with acetate as sole C and energy source. With acetate as sole C and energy source, in wild-type cultures N2ase activities were 232 nmol ethylene h-1 per 5x108 cells, whereas, in the pdhB4R mutant cultures N2ase activities were 180 nmol ethylene h-1 per 5x108 cells. As a measure of N2 fixation, N2-dependent growth was also assessed by colony-forming ability on defined medium with either succinate or acetate as sole C and energy source and lacking combined N maintained under 98% N2/1% O2/1% CO2 sparge (see Methods). Seven days post-inoculation, wild-type yielded large, opaque colonies on succinate and small, opaque colonies on acetate. The pdhB4R mutant yielded small, translucent colonies with few cells on succinate and small, opaque colonies on acetate.

A. caulinodans microaerobic growth requires PDH activity
To study C source utilization under true microaerobic conditions, A. caulinodans wild-type and the pdhB4R mutant were aerobically cultured to early exponential phase (1x108 ml-1) in defined medium with 15 mM ammonium as N source and shifted to continuous sparge with a prepared gas mixture of 0·1% O2/1% CO2/balance (~99%) argon. After such shifts, DOT values for all cultures dropped to submicromolar levels within 2 h (see Methods). Upon shift from aerobic to microaerobic conditions, the wild-type continued to grow normally with succinate as sole C and energy source. However, with acetate as sole C source, further growth ceased immediately after shift (Fig. 4a). Upon shift from aerobic to microaerobic conditions, the pdhB4R mutant failed to grow whether provided with either succinate or acetate as sole C and energy source (Fig. 4b). Microaerobic growth was also assessed as colony-forming ability on solid, defined medium (Methods). A. caulinodans cytochrome cbb3 oxidase mutant 64611, which cannot grow in microaerobic culture (Kaminski et al., 1996 ) was included as negative control in these experiments. The wild-type was able to grow with either succinate, L-malate, L-lactate, 2-oxoglutarate or L-valine, but not with acetate, ß-D-hydroxybutyrate or nicotinate as sole C and energy source (Table 4). By contrast, the pdhB4R mutant did not grow under microaerobic conditions with any combination of these, various C and energy sources (Table 4). As measured by both physiological shift experiments and colony-forming ability, the pdhB4R mutant failed to show microaerobic growth. Therefore, PDH activity is indispensable under microaerobic conditions.



View larger version (17K):
[in this window]
[in a new window]
 
Fig. 4. Physiological shift of A. caulinodans wild-type (a) and the pdhB4R mutant (b), cultured in liquid, defined (ORS-MM) medium, from aerobic to microaerobic conditions. {blacksquare}, {blacktriangleup}, Cultures provided with succinate as sole C and energy source and shifted from from aerobic ({blacksquare}) to microaerobic ({blacktriangleup}) conditions; {diamondsuit}, {bullet}, cultures provided with acetate as sole C and energy source shifted from from aerobic ({diamondsuit}) to microaerobic ({bullet}) conditions.

 

View this table:
[in this window]
[in a new window]
 
Table 4. Microaerobic growth of A. caulinodans wild-type 57100, pdhB4R mutant 62004R and fixN mutant 64611 on defined ORS-MM

 
When cultures are shifted to microaerobic conditions, intracellular 2-oxoacid pools change considerably and the NADH:NAD+ ratio increases markedly
To gain further insight into microaerobic growth, A. caulinodans wild-type cultures were analysed for changes in intracellular 2-oxoacid and pyridine nucleotide pools upon microaerobic shift. A. caulinodans wild-type culture samples (100 ml) were removed at the time of microaerobic shift, and 4 h later (Fig. 4), and processed as described in Methods. Intracellular 2-oxoacid levels changed notably. Whereas endogenous pyruvate levels increased, both oxaloacetate and 2-oxoglutarate levels decreased substantially after microaerobic shift (Table 5). Pyridine nucleotide pools also changed markedly: the NADH:NAD+ ratio increased from 0·08 in aerobic culture to 0·41 upon microaerobic shift (Table 5). In control experiments in which cultures were maintained in aerobic conditions, both 2-oxoacid pools and the endogenous NADH:NAD+ ratio varied minimally. Paradoxically, such responses to microaerobic conditions may reflect a new steady state in which acetyl-CoA is in physiological excess, yet citric acid cycle activity becomes rate-limiting for growth (see Discussion).


View this table:
[in this window]
[in a new window]
 
Table 5. Physiological changes in A. caulinodans wild-type cells upon microaerobic shift

 
When growing microaerobically, A. caulinodans produces poly-ß-hydroxybutyrate
In response to steady-state acetyl-CoA excess, microaerophilic bacteria, including rhizobia, polymerize acetyl-CoA equivalents, yielding polyhydroxyalkanoates, principally poly-ß-hydroxybutyrate (PHB) (Wong & Evans, 1971 ; Jackson & Dawes, 1976 ). Twenty-four hours after shift to microaerobic conditions, A. caulinodans cells were recovered and tested for PHB production using a qualitative, Sudan Black cytological staining assay (Steinbüchel & Schlegel, 1991 ). An absolute correlation with microaerobic growth was observed. Bacteria sampled from cultures which grew also stained black; those which grew little or not at all did not stain black. When portions of these cultures maintained under vigorously growing aerobic conditions were similarly tested, in no case did cells stain black. Such PHB production was further evidence of altered physiological responses to microaerobic growth conditions.

In S. rostrata nodules, symbiotic N2 fixation requires PDH activity
In culture under 10 µM DOT, the pdhB4R mutant retains the ability to fix N2. In nodule symbiosis, N2 fixation operates in the presence of excess leghaemoglobin and, hence, is a microaerobic process. As inferred from reconstruction experiments, DOT in active nodule tissues decreases to nanomolar levels in response to leghaemoglobin induction (Bergersen et al., 1986 ; Bergersen & Turner, 1990 ). To assess the role of PDH in symbiotic N2 fixation, a (temporal) developmental study was conducted with, as host plant, S. rostrata, which yields determinate (developmentally synchronous) nodules on both stems and roots. Three-week-old S. rostrata plants were inoculated at stem internodes with fresh A. caulinodans cultures (see Methods). In incited S. rostrata stem nodules, leghaemoglobin synthesis dramatically increased, beginning 11 d after inoculation, and became fully active 14 d after inoculation. Starting 6 d and continuing to 24 d post-inoculation, individual stem nodules were excised and tested for N2 fixation activity by acetylene reduction (Fig. 5). For wild-type stem nodules, some N2 fixation activity was evident prior to leghaemoglobin induction (presumably at >1 µM DOT). Nevertheless, N2 fixation (acetylene reduction) activity increased dramatically in parallel with leghaemoglobin (spectrophotometric absorbance). The pdhB4R mutant also showed incremental N2 fixation activity early in stem nodule development. However, N2 fixation responded adversely to leghaemoglobin induction, which was both delayed and limited. In mature stem nodules elicited by the pdhB4R mutant, N2 fixation not only failed to induce, it decayed (Fig. 5). Therefore, the A. caulinodans pdhB4R mutant is symbiotically incompetent. In summary, A. caulinodans N2 fixation in S. rostrata nodules requires its own PDH activity.



View larger version (14K):
[in this window]
[in a new window]
 
Fig. 5. Developmental study of S. rostrata nodules. N2 fixation activities for detached nodules of A. caulinodans wild-type ({diamondsuit}) and the pdhB4R mutant ({blacksquare}) are plotted versus time after inoculation. Strong leghaemoglobin induction occurred at day 11 following inoculation (not indicated).

 

   DISCUSSION
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
A. caulinodans aerobic and microaerobic energy transduction are thus distinct, as evident from two type of experiments reported here. Firstly, with preferred C4 dicarboxylates (succinate, L-malate) as sole energy source, PDH activity is dispensable for aerobic, but indispensable for microaerobic energy transduction. Secondly, whereas acetyl-CoA is readily oxidized under aerobic conditions, it is poorly oxidized under microaerobic conditions. The PDH complex has long been considered a core metabolic activity and one indispensable to obligate aerobic organisms for growth on sugars or organic acids not directly metabolized to acetyl-CoA. Accordingly, the A. caulinodans pdhB4 mutant might well accumulate various intermediates, notably pyruvate itself, when cultured on C4 dicarboxylates (succinate, L-malate). However, after mass-spectral analysis of conditioned media from batch cultures, no mass peaks in pdhB4 mutant conditioned media that were not also present in wild-type conditioned media were identifiable (Pauling, 1999 ). Therefore, the pdhB4 mutant circumvents the metabolic block resulting from lack of PDH activity. Under aerobic conditions, the A. caulinodans pdhB4 mutant remains able to use C4 dicarboxylates (succinate, L-malate), albeit slowly. Because acetyl-CoA, the product of PDH activity, is a required metabolite, the pdhB4 mutant must somehow make acetyl-CoA from C4 dicarboxylates. Hence, PDH activity is not the sole source of acetyl-CoA under aerobic growth conditions.

From genetic experiments, MSDH activity was identified as alternative source of acetyl-CoA in aerobic culture. Notably, in wild-type cell extracts, valine-induced MSDH activities were quite low in comparison to succinate-induced PDH activities, which may help explain why MSDH-dependent acetyl-CoA production is growth-limiting under aerobic conditions with C4 dicarboxylates as C and energy source. Under microaerobic conditions, the MSDH dependent pathway does not function. By inference, oxaloacetate, and thus NAD+-dependent malate dehydrogenase (MDH) activity, is then required for acetyl-CoA synthesis. Conceivably, MDH activity is sufficiently inhibited by the high microaerobic NADH:NAD+ ratio so as to limit malonate semialdehyde production. Inhibition of MSDH-dependent acetyl-CoA synthesis would then render PDH activity indispensable. While L-valine strongly induces MSDH activity in aerobic cultures and sustains microaerobic growth of wild-type, it does not allow microaerobic growth of the pdhB4R mutant. Accordingly, this growth defect is not attributable to lack of MSDH activity.

The ts alleles carried by strains 62042 and 62047 lack MSDH activity at the restrictive temperature. For both ts mutants, activity with both malonate semialdehyde and methylmalonate semialdehyde as substrate is then completely lost. Most likely, these ts mutants involve a single, catalytic protein. In that case, either a single subunit is used to make two, different enzymes or else a single enzyme activity uses both substrates. In mammals, MSDH accepts as substrate both methylmalonate semialdehyde, yielding propanoyl-CoA, and malonate semialdehyde, yielding acetyl-CoA (Yamada & Jakoby, 1960 ). Hypothetically, A. caulinodans might produce malonate semialdehyde from L-aspartate via ß-alanine. However, such N-containing compounds accumulate in rhizobia (Reid et al., 1996 ) and thus are poorly oxidizable energy substrates. Further complicating this analysis, both L-aspartate and ß-alanine, when catabolized at high rates, also liberate (excess) ammonium. Therefore, confirmation of this pathway is problematical and requires experiments other than simple physiological growth tests.

In culture, the A. caulinodans pdhB4R mutant remains able to fix N2 at 10 µM DOT, a partially aerobic condition. However, during symbiosis, the pdhB4R mutant displays an absolute defect in N2 fixation that correlates with leghaemoglobin induction. Conceivably, dihydrolipoamide dehydrogenase, the E3 flavoprotein of the PDH complex, might serve as metabolic electron source for chemical reduction of N2 under microaerobiosis. However, changes in energy metabolism under microaerobic conditions characteristic of symbiotic N2 fixation in planta almost certainly contribute to this phenotype. In symbiotic nodules, bacteroids accumulate large amounts (30–60% dried cell mass) of PHB (Bergersen & Turner, 1990 ), as do microaerophilic bacterial cultures generally (Jackson & Dawes, 1976 ; Senior et al., 1972 ; Steinbüchel & Schlegel, 1991 ). Indeed, A. caulinodans PHB synthase mutants are symbiotically defective (Mandon et al., 1998 ).

For obligate aerobic organisms, physiological O2 limitation frequently leads to steady-state excess of acetyl-CoA, the fuel of oxidative energy transduction (citric acid cycle; glyoxylate cycle). Formally, this condition obtains when acetyl-CoA production rates exceed oxidation rates; the metabolic steady-state is then constrained by available oxidant. To compensate, some acetyl-CoA may then itself serve as endogenous oxidant, in which case partially reduced compounds, among them ß-hydroxybutyrate and other ‘ketone bodies’, accumulate. In mammals, to help mitigate adverse effects of such ‘ketosis’, ß-hydroxybutyrate equivalents are then excreted. In microaerophilic bacteria, ß-hydroxybutyrate equivalents are instead polymerized in situ, forming poly-hydroxyalkanoate (such as PHB) granules. Both bacterial and mammalian processes presumably help reoxidize pyridine nucleotide (cofactor) pools and thus revitalize stalled acetyl-CoA oxidation via concerted glyoxylate and citric acid cycle activities.

Pushing this limit further, microaerophilic bacteria remain capable of oxidative energy transduction under nanomolar DOT, which represents extreme O2 limitation. In principle, this capability may be owed to either peripheral factors and/or fundamentally different core processes. As an example of the former, in response to extreme O2 limitation, A. caulinodans produces ultrahigh-affinity terminal oxidases, such as cytochrome cbb3 and cytochrome bd oxidases, which possess apparent Km(O2) values sufficient to allow growth at submicromolar DOT (Kaminski et al., 1996 ). However, even such ultrahigh-affinity terminal oxidases may not be fully O2-saturated at nanomolar DOT, so overall metabolic rates may then be limited by terminal oxidase rates.

However, rates are not the whole story. Consider the overall equation for respiration:


In microaerophilic bacteria such as A. caulinodans, oxidative phosphorylation continues unabated and at high efficiency during microaerobic growth (Senior et al., 1972 ). To maintain a similarly high ~P:O ratio at nanomolar DOT (Bergersen et al., 1986 ), the steady-state NADH:NAD+ ratio needs to increase significantly. Correspondingly, A. caulinodans cultures show a dramatic increase in NADH:NAD+ upon microaerobic shift. In principle, this high NADH:NAD+ ratio might lead to qualitative metabolic changes, such as increased diversion of L-malate to pyruvate in lieu of oxaloacetate. In contrast to the reaction catalysed by conservative NAD-malate dehydrogenase (yielding oxaloacetate), a highly endergonic process, that catalysed by decarboxylative malate dehydrogenase (also known as ‘malic enzyme’), yielding pyruvate is exergonic under standard conditions. Indeed, in E. coli, NAD-dependent ‘malic enzyme’ is the only such activity among this group of L-malate-dependent enzymes not allosterically inhibited by NADH (Sanwal, 1970 ). In Sinorhizobium meliloti, NAD-dependent ‘malic enzyme’ activity is indeed required for microaerobic energy transduction in Medicago sativa nodules (Driscoll & Finan, 1993 ).

For aerobic energy transduction via the citric acid cycle, oxaloacetate and acetyl-CoA undergo bimolecular condensation to yield citrate, regenerating coenzyme A. In Saccharomyces cerevisiae, NAD-malate dehydrogenase and citrate synthase activities are metabolically coupled (Vélot & Srere, 2000 ). In E. coli, both citrate synthase and NAD-malate dehydrogenase activities are powerfully allosterically inhibited by NADH (Sanwal, 1969 , 1970 ). Accordingly, upon shift to microaerobic conditions, an increased NADH:NAD+ ratio may render acetyl-CoA oxidation rates growth-limiting. Consequently, citric acid cycle and glyoxylate cycle activities might then be restricted. Indeed, both Bradyrhizobium japonicum aconitase mutants (Thöny-Meyer & Künzler, 1996 ) and 2-oxoglutarate dehydrogenase mutants (Green et al., 2000 ) remain competent for symbiotic N2 fixation with host soybean (Glycine max) plants and, by inference, remain capable of microaerobic energy transduction.

Thus, excess acetyl-CoA serves as endogenous oxidant, being converted to PHB as metabolic end product. Nevertheless, energy transduction with C4 dicarboxylate (succinate, L-malate) as energy substrate must remain oxidative and NADH as substrate for oxidative phosphorylation must be produced (Fig. 6). However, the (linear) conversion of L-malate to PHB is only partially (<50%) oxidative, yielding four CO2 per two substrate L-malate molecules converted to ß-D-hydroxybutyryl-CoA for polymerization. Beneficially, a less oxidative energy transduction helps microaerophilic bacteria maintain high substrate fluxes, and requisite high enzyme activities, whether microaerobic or aerobic. In turn, such oxidative ‘metabolic gearing’ facilitates rapid adaptation in response to dramatic changes in environmental DOT.



View larger version (25K):
[in this window]
[in a new window]
 
Fig. 6. A. caulinodans energy metabolism: aerobic conditions (dashed lines, normal text) versus microaerobic conditions (solid lines, bold-italicised text); ({vdash}), strictly aerobic reactions inhibited under microaerobic conditions.

 

   ACKNOWLEDGEMENTS
 
This work was supported by grants from the US National Science Foundation and the US National Institutes of Health to R.A.L.


   REFERENCES
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Bergersen, F. J. & Turner, G. L. (1990). Bacteroids from soybean root nodules: accumulation of poly-ß-hydroxybutyrate during supply of malate and succinate in relation to N2 fixation in flow-chamber reactions. Proc R Soc Lond B 240, 39-59.

Bergersen, F., Turner, G. L., Bogusz, D., Wu, Y.-Q. & Appleby, C. A. (1986). Effects of O2 concentrations on respiration and nitrogenase activity of bacteroids from stem and root nodules of Sesbania rostrata and of the same bacteria from continuous culture. J Gen Microbiol 132, 3325-3336.

Bernovsky, C. & Swan, M. (1973). An improved cycling assay for nicotinamide adenine dinucleotide. Anal Biochem 53, 452-458.[Medline]

Buckmiller, L. M., Lapointe, J. P. & Ludwig, R. A. (1991). Physical mapping of the Azorhizobium caulinodans nicotinate catabolism genes and characterization of their importance to N2 fixation. J Bacteriol 173, 2017-2025.[Medline]

Cabanes, D., Boistard, P. & Batut, J. (2000). Symbiotic induction of pyruvate dehydrogenase genes from Sinorhizobium meliloti. Mol Plant–Microbe Interact 13, 483-493.[Medline]

Ditta, G. (1986). Tn5 mapping of Rhizobium nitrogen fixation genes. Methods Enzymol 118, 519-528.

Donald, R. G. K., Raymond, C. K. & Ludwig, R. A. (1985). Vector insertion mutagenesis of Rhizobium sp. strain ORS571: direct cloning of mutagenised DNA sequences. J Bacteriol 162, 317-323.[Medline]

Donald, R. G. K., Nees, D., Raymond, C. K., Loroch, A. I. & Ludwig, R. A. (1986). Three genomic loci encode Rhizobium sp. ORS571 N2 fixation genes. J Bacteriol 165, 72-81.[Medline]

Dreyfus, B. L. & Dommergues, Y. R. (1981). Nitrogen fixing nodules induced by Rhizobium on stems of the tropical legume Sesbania rostrata. FEMS Microbiol Lett 10, 313-317.

Driscoll, B. T. & Finan, T. M. (1993). NAD-dependent malic enzyme of Rhizobium meliloti is required for symbiotic nitrogen fixation. Mol Microbiol 7, 865-873.[Medline]

Finan, T. M., Wood, J. M. & Jordan, D. C. (1981). Succinate transport in Rhizobium leguminosarum. J Bacteriol 148, 193-202.[Medline]

Finan, T. M., Wood, J. M. & Jordan, D. C. (1983). Symbiotic properties of C4-dicarboxylic acid transport mutants of Rhizobium leguminosarum. J Bacteriol 154, 1403-1413.[Medline]

Green, L. S., Li, Y., Emerich, D. W., Bergersen, F. J. & Day, D. A. (2000). Catabolism of {alpha}-ketoglutarate by a sucA mutant of Bradyrhizobium japonicum: evidence for an alternative tricarboxylic acid cycle. J Bacteriol 182, 2838-2844.[Abstract/Free Full Text]

Hayaishi, O., Nishizuka, Y., Tatibana, M., Takeshita, M. & Kuno, S. (1961). Enzymatic studies on the metabolism of ß-alanine. J Biol Chem 236, 781-790.[Medline]

Jackson, F. A. & Dawes, E. A. (1976). Regulation of the tricarboxylic acid cycle and poly-ß-hydroxybutyrate metabolism in Azotobacter beijerinckii grown under nitrogen or oxygen limitation. J Gen Microbiol 97, 303-312.[Medline]

Kaminski, P. A, Kitts, C. L., Zimmerman, Z. & Ludwig, R. A. (1996). Azorhizobium caulinodans uses both Cytbd (quinol) and Cytcbb3 (Cytc) terminal oxidases for symbiotic N2 fixation. J Bacteriol 178, 5989-5994.[Abstract]

Kitts, C. L., Schaechter, L. E., Rabin, R. R. & Ludwig, R. A. (1989). Identification of cyclic intermediates in Azorhizobium caulinodans nicotinate catabolism. J Bacteriol 171, 3406-3411.[Medline]

Kitts, C. L., Lapointe, J. P., Lam, V. T. & Ludwig, R. A. (1992). Elucidation of the complete Azorhizobium nicotinate catabolism pathway. J Bacteriol 174, 7791-7797.[Abstract]

Koland, J. G., Miller, M. J. & Gennis, R. B. (1984). Reconstitution of the membrane-bound, ubiquinone-dependent pyruvate oxidase respiratory chain of Escherichia coli with the cytochrome d terminal oxidase. Biochemistry 23, 445-453.[Medline]

Kwon, D. K. & Beevers, H. (1992). Growth of Sesbania rostrata (Brem) with stem nodules under controlled conditions. Plant Cell Environ 15, 939-945.

Loroch, A. I., Nguyen, B. & Ludwig, R. A. (1995). FixLJK and NtrBC signals interactively regulate Azorhizobium nifA transcription via overlapping promoters. J Bacteriol 177, 7210-7221.[Abstract]

Ludwig, R. A. (1986). Rhizobium sp. strain ORS571 grows synergistically on N2 and nicotinate as N sources. J Bacteriol 165, 304-307.[Medline]

Mandon, K., Michel-Reydellet, N., Encarnación, S., Kaminski, P. A., Cevallos, M. A., Elmerich, C. & Mora, J. (1998). Poly-ß-hydroxybutyrate turnover in Azorhizobium caulinodans is required for growth and affects nifA expression. J Bacteriol 180, 5070-5076.[Abstract/Free Full Text]

Pauling, D. C. (1999). Identification and characterization of the pyruvate dehydrogenase locus in Azorhizobium caulinodans. PhD thesis, University of California, Santa Cruz.

Reid, C. J., Walshaw, D. L. & Poole, P. S. (1996). Aspartate transport by the Dct system in Rhizobium leguminosarum negatively affects nitrogen-regulated operons. Microbiology 142, 2603-2612.[Abstract]

Ronson, C. W., Lyttleton, P. & Robertson, J. G. (1981). C4-dicarboxylate transport mutants of Rhizobium trifolii form ineffective nodules on Trifolium repens. Proc Natl Acad Sci USA 78, 4284-4288.[Abstract]

Sambrook, J., Fritsch, E. F. & Maniatis, T. (1989). Molecular Cloning: a Laboratory Manual. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory.

Sanwal, B. D. (1969). Control characteristics of malate dehydrogenase. J Biol Chem 244, 1831-1837.[Abstract/Free Full Text]

Sanwal, B. D. (1970). Allosteric controls of amphibolic pathways in bacteria. Bacteriol Rev 34, 20-39.[Medline]

Senior, P. J., Beech, G. A., Ritchie, F. & Dawes, E. A. (1972). The role of oxygen limitation in the formation of poly-ß-hydroxybutyrate during batch and continuous culture of Azotobacter beijerinckii. Biochem J 128, 1193-1201.[Medline]

Simon, R., Priefer, U. & Pühler, A. (1983). A broad host range mobilization system for in vivo genetic engineering: transposon mutagenesis in Gram-negative bacteria. Biotechnology 1, 784-791.

Smith, R. F. & Smith, T. (1991). Automatic generation of primary sequence patterns from sets of related protein sequences. Proc Natl Acad Sci USA 87, 118-122.[Abstract]

Steinbüchel, A. & Schlegel, H. G. (1991). Physiology and molecular genetics of poly-ß-hydroxyalkanoic acid synthesis in Alcaligenes eutrophus. Mol Microbiol 5, 535-542.[Medline]

Thöny-Meyer, L. & Künzler, P. (1996). The Bradyrhizobium japonicum aconitase gene is important for free-living growth but not for an effective root nodule symbiosis. J Bacteriol 178, 6166-6172.[Abstract]

Vélot, C. & Srere, P. A. (2000). Reversible transdominant inhibition of a metabolic pathway. J Biol Chem 275, 12926-12933.[Abstract/Free Full Text]

Wong, P. P. & Evans, H. J. (1971). Poly-ß-hydroxybutyrate utilization by soybean (Glycine max Merr.) nodules and assessment of its role in maintenance of nitrogenase activity. Plant Physiol 47, 750-755.

Yamada, E. W. & Jakoby, W. B. (1960). Direct conversion of malonic semialdehyde to acetyl-coenzyme A. J Biol Chem 235, 589-594.[Medline]

Received 24 January 2001; revised 28 March 2001; accepted 2 April 2001.