Journal of General Microbwlogy (1984), 130, 1809-1 8 14. Printed in Great Britain
1809
Carbon Utilization by Free-living and Bacteroid Forms of Cowpea
Rhizobium Strain NGR234
B y S . SAROSO,* A. R. G L E N N A N D M. J . D I L W O R T H
Nitrogen Fixation Research Group, School of Environmental and Life Sciences,
Murdoch University, Murdoch, Western Australia 6150
(Received 14 June 1983; revised 12 January 1984)
Free-living cells of the fast-growing cowpea Rhizobium NGR234 were able to grow on a variety
of carbon substrates at growth rates varying from 2.5 h on glucose or fumarate to 15.6 h on p hydroxybenzoate. Free-living cells constitutively oxidized glucose, glutamate and aspartate but
were inducible for all the other systems investigated. Bacteroids from root nodules of snake
bean, however, were only capable of oxidizing C,-dicarboxylic acids and failed to oxidize any
other carbon sources. Free-living cells of NGR234 possess inducible fructose and succinate
uptake systems. These substrates are accumulated by active processes since accumulation is
in hi bi ted by azide, 2,4-dinitrop hen01 and carbonyl cyanide m-chlorophenylhydrazone.
Bacteroids failed to take up fructose although they actively accumulated succinate, suggesting
that the latter substrate is significant in the development of an effective symbiosis.
INTRODUCTION
An understanding of the survival of rhizobia in soil and of the complex interactions between
Rhizobium and legume required for the establishment of an effective symbiosis is likely only
when the physiology of these organisms has been carefully studied. Although recent advances in
the genetics and molecular biology of Rhizobium have been rapid, advances in understanding its
physiology have been much slower.
Previous work from this laboratory has examined various aspects of the carbon metabolism of
Rhizobium leguminosarum (e.g. Glenn et al., 1980; Dilworth & Glenn, 1981; Glenn & Dilworth,
1981a ; Dilworth et al., 1983). In general, inducible oxidation systems for carbon compounds
were found to be relatively uncommon for R . leguminosarum; most oxidation systems were
produced at significant levels in the absence of the particular substrate. In a survey of the uptake
and hydrolysis of disaccharides by rhizobia (Glenn & Dilworth, 1981b), it was found that while
R . leguminosarum synthesized significant levels of the uptake systems and hydrolytic enzymes
for a range of disaccharides in the absence of the substrate, such systems in R . meliloti and
cowpea Rhizobium NGR234 were inducible.
Cowpea Rhizobium NGR234 is an unusual strain in that it is fast-growing and forms effective
nodules on legumes typically nodulated by slow-growing bradyrhizobia. In addition, it has a
single megaplasmid responsible for its symbiotic properties (Morrison et al., 1983). Because of
these unusual properties, the general interest in the strain by workers in plasmid biology and
some clear differences in physiology in comparison with R . leguminosarurn MNF3841 (Glenn &
Dilworth, 1981a), the carbon metabolism of NGR234 has now been investigated in greater
detail.
METHODS
Organism. Cowpea Rhizobium NGR234 was obtained from Dr M. J . Trinick, CSIRO, Perth, Western Australia.
This strain is a typical fast-grower (Broughton & Dilworth, 1971) and nodulates both cowpea (Vigna unguiculata
(L.) Walp. ssp. unguiculata) and snake bean ( V . unguiculata ssp. sesquipedalis (L.) Verdc.).
0022-1287/84/0001-1318 $02.00
0 1984 SGM
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Mon, 31 Jul 2017 22:58:45
1810
S . SAROSO, A . R . G L E N N A N D M . J . D I L W O R T H
Media. Bacteria were grown at 28 "C in batch culture in the minimal salts medium of Brown & Dilworth (1975)
with NH4Cl (10mM) as the nitrogen source and phosphate at 0.3mM. Carbon sources were provided at 10 mM
except for isoleucine and p-hydroxybenzoate which were used at 3 m~ to avoid growth inhibition by higher
concentrations. HEPES buffer (40mM, pH 7.0)was added to maintain the pH.
Nodulation and preparation of bacteroids. Snake bean seeds (dwarf variety) obtained from Arthur Yates & Co.,
Perth, W. Australia, were surface sterilized (Vincent, 1977) and inoculated with strain NGR234. After 60 d,
bacteroids were prepared as described by Glenn et al. (1980).
Determination of oxygen consumption.Substrate-dependent O2consumption was measured polarographically at
25 "C using Hansatech oxygen electrodes (Hansatech, Norfolk, UK) as described by Glenn & Dilworth (1981a).
Results are expressed as nmol O2 min-' (mg protein)-'.
Uptake experiments. Cells were prepared for uptake experiments as described by Hudman & Glenn (1980),
except that the cells were centrifuged and washed rather than Millipore filtered. Uptake experiments were carried
out as described by Hudman & Glenn (1980).[2,3-14C]Succinicacid (59mCi mmol- ;2.18 GBq mmol-l) and D[U-14C]fructose(283mCi mmol- ;7.5 GBq mmol- l ) were obtained from Amersham. Competitors or inhibitors
were added 60 s before the addition of 0.5 m~-[~~C]succinate
or [ 14C]fructosein order to allow time for them to act
and produce linear uptake and incorporation kinetics. Cell-associated radioactivity was measured using the
scintillant described by Hudman & Glenn (1980)with the addition of Triton X-100 (300ml 1- l ) .
Analytical methods. Protein was determined by the Lowry method using bovine serum albumin (Fraction V,
Sigma) as the standard. Histidine was measured with a modified Paully reagent (Ray, 1967).
L-Amino-acid oxidase (EC 1.4.3.2)was measured in whole cells grown on glucose (A600approx. 0.8), washed
twice with minimal salts (Brown & Dilworth, 1975)and resuspended to the same density in a similar salts solution.
Cells were incubated for 15 min at 28 "C with histidine, isoleucine, glutamate or aspartate, each at 1.5 mM final
concentration. After 15 min incubation with 1 ml2,4-dinitrophenylhydrazine(0.1%, w/v, in 1 M-HCl),3 m12 MNaOH was added; after a further 15 min, the solutions were centrifuged for 10 min at 5000 r.p.m. and the AS40
was measured. Histidase (EC 4.3.1.3)was assaayed using the method described by Dilworth et al. (1983).
RESULTS A N D D I S C U S S I O N
Oxidation of substrates by cultured cells
Strain NGR234 grew on a wide range of amino acids, sugars, aliphatic and aromatic acids
and polyols as sole source of carbon and energy with the generation times shown in Table 1. A
similar metabolic versatility has been reported previously for R. leguminosarum 3841 (Glenn &
Dilworth, 1981a).
Of the 16 growth substrates examined, NGR234 possessed constitutive oxidation systems for
only glucose, glutamate and aspartate; the remainder were inducible (Table 1). Glucose
oxidation was low in cells grown on organic or amino acids, but relatively high on all other
Table 1. Substrate-dependent O2consumption by cowpea Rhizobium NGR234
Grown on
Mean
generation
time (h)
Glucose
Fructose
Sucrose
Arabinose
Fumarate
Succinate
Malate
Pyruvate
Malonate
p-Hydroxybenzoate
Glycerol
Mannitol
Glutamate
Aspartate
Isoleucine
Histidine
2.5
4.3
2.7
2.9
2.5
2.5
2.5
3.7
5.1
15.6
4.2
3.1
2.5
5.9
6.5
6.7
Oxidation [nmol min- (mg protein)-'] of:
A
I
Homologous
substrate
Succinate
88
80
109
154
48
69
46
69
56
161
90
49
109
73
67
137
5
1
16
5
25
69
27
4
9
20
7
8
14
10
5
8
Glucose
Glutamate
Aspartate
88
51
38
61
20
30
15
31
12
40
37
32
22
28
28
31
36
73
51
75
74
95
84
101
67
146
81
63
109
197
128
174
38
34
42
42
42
58
36
46
38
85
47
41
66
73
73
71
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Mon, 31 Jul 2017 22:58:45
Bacteroid
oxidation
rate
0.6
0.8
0.7
0.9
23.9
35.5
19.3
1.0
1.2
3.0
0.3
0.0
0.4
5.3
1.1
2.2
Carbon utilization in cowpea Rhizobium NGR234
1811
substrates, a pattern similar to that reported by Hudman & Glenn (1980) for R. leguminosarum
WU235. The glucose catabolic system appears to be somewhat different to those for other
sugars, since the latter are clearly inducible in this strain. Cells grown on mannitol had a high
level of fructose-dependent O2 consumption [114 nmol O2 rnin-l (mg protein)-'], while
sucrose-grown cells also oxidized fructose [56 nmol O2 min- (mg protein)- '1. These data
suggest that the fructose permease is induced in such cells, probably as a result of the
intracellular conversion of mannitol or sucrose to fructose.
Cell grown on fumarate or malate or succinate were capable of oxidizing the other two C4dicarboxylic acids, This suggests that there is a common uptake system in NGR234 similar to
that described in R . leguminosarum (Glenn et al., 1980; Finan et al., 1981) and R. trifolii (Ronson
et al., 1981). The other organic acids are also metabolized via inducible systems. Like other
rhizobia (Parker et al., 1977; Glenn & Dilworth, 1981a ; Chen et al., 1984), NGR234 is capable
of growth on the aromatic substrate p-hydroxybenzoate but it was unable to grow on benzoate,
catechol or p-toluate. It will be of interest to investigate whether this capacity is genetically
determined by the large plasmid reported in this strain (Morrison et al., 1983).
Aspartate-, glutamate-, histidine- and isoleucine-dependent oxygen consumption were
observed in cells grown on all carbon sources tested (Table 1). The usual catabolic route for
histidine degradation in Rhizobium is via urocanic acid, a reaction mediated by histidase
(Dilworth et al., 1983). When strain NGR234 was grown on L-histidine as the sole source of
carbon, histidase activity was high [159 nmol min-l (mg protein)-'], whereas cells grown on a
range of other carbon sources (glucose or fructose or arabinose or succinate) had low histidase
activities [0.2-1*2 nmol min- (mg protein)- '1. It seems unlikely that the L-histidine-dependent
O2 consumption by cells grown on other carbon sources is due to the presence of these basal
levels of histidase. An alternative explanation is that the cells contain an L-amino acid oxidase
which consumes oxygen and produces 2-0x0 acids. When glucose-growncells of NGR234 were
incubated in glutamate, aspartate or histidine there was no evidence for 0x0 acid formation. One
speculative hypothesis for the amino acid-induced O2 consumption is that endogenous
metabolism may be stimulated by organic nitrogen sources.
Oxidation of substrates by bacteroids
Isolated bacteroids of strain NGR234 from snake bean nodules oxidized the C4-dicarboxylic
acids fumarate, malate and succinate, the last named supporting the highest O2 consumption.
The uptake and catabolic enzyme systems for C,-dicarboxylic acid metabolism must therefore
be present in the bacteroid (Table 1). The rate of succinate oxidation by isolated bacteroids was
less than 50% that observed for free-living cells grown on succinate. This lower rate of succinate
oxidation by isolated bacteroids of NGR234 is similar to that described for isolated bacteroids
of R. feguminosarum MNF3841. Bacteroids of strain NGR234 were unable to oxidize glucose,
glutamate or aspartate even though these systems appear to be constitutive in the free-living
form. A similar loss of metabolic capabilities constitutive in free-living cells has been reported in
bacteroids of I?.leguminosarum (Glenn & Dilworth, 1981a). One possible explanation advanced
for this apparent loss of constitutive uptake systems from the bacteroid is that they depend on
periplasmic binding proteins which may be lost during bacteroid isolation (Dilworth & Glenn,
1981). However, it has been shown that isolated bacteroids of strain NGR234 retain most of
their periplasmic proteins (J. Smart, personal communication). It is likely, therefore, that the
loss of the constitutive uptake systems from the bacteroid is due to other causes.
The induction of the C4-dicarboxylic acid permease system in nodule bacteroids provides
clear evidence that they are exposed to, and capable of utilizing, these organic acids. The
production of the C,-dicarboxylic acid uptake system is thus likely to be an event of some
significance in the development of an effective symbiosis.
Nature of the fructose and succinate uptake systems
Cells of cowpea Rhizobium NGR234 grown on fructose or mannitol had a rapid rate of
[ 14C]fructoseuptake and incorporation [30-50 nmol min- (mg protein)- l]. Cells grown on
glucose or succinate were unable to transport [14C]fructose[ < 2.0 nmol min- (mg protein)- '1.
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Mon, 31 Jul 2017 22:58:45
1812
S. SAROSO, A . R . G L E N N A N D M . J . D I L W O R T H
- 200
200 r
i
i%
1
(b)
41
Q 100
&
m
9
Y
z
'5
2
50
L
2
ii
3
Time (min)
1
2
3
Time (min)
4
5
Fig. 1. (a) Uptake and incorporation of label from [14C]succinateby free-living cells of cowpea
Rhizobium NGR234 grown on: succinate (0);glucose, fructose, mannitol or glutamate (results
identical; 0).(b)Uptake and incorporation of label from [14C]succinate(0)and [14C]fructose( 0 )by
isolated bacteroids of cowpea Rhizobium NGR234.
This shows that the fructose uptake system in this cowpea strain is inducible during growth on
fructose or mannitol and contrasts with the situation in R . leguminosarum MNF3841 (Glenn &
Dilworth, 1981b ; Glenn et al., 1984b), where the fructose transport system is constitutive.
Cells of strain NGR234 grown on succinate showed an immediate capacity to take up
[14C]succinate [34 nmol min-l (mg protein)-'] (Fig. 1a), whereas cells grown on glucose,
fructose, mannitol or glutamate were unable to transport succinate [ < 1 nmol min- (mg
protein)- '1. The C,-dicarboxylic acids are thus transported by NGR234 via an inducible system
similar to that reported in R . trifolii by Ronson et al. (198 1) and in R . leguminosarum by Finan et
al. (1981), but different from that found in R . leguminosarum MNF3841 by Glenn et al. (1980).
Bacteroids of NGR234 isolated from snake bean nodules were able to take up [ 14C]succinate
[16 nmol min-l (mg protein)-'] but did not take up [14C]fructose, showing that the uptake
system for succinate was induced in the bacteroid in the nodule (Fig. lb).
EfSect of metabolic inhibitors on fructose and succinate uptake
[ 14C]Fructoseand [ 14C]succinateuptake, measured in cells grown to mid-exponential phase
(A&r approx. 0.6) on fructose and succinate, respectively, involved active processes since they
were inhibited by azide (2 mM), 2,4-dinitrophenol (1 mM) and carbonyl cyanide rnchlorophenylhydrazone (0.02 mM) by more than 96%. This is similar to the effects of these
inhibitors on the fructose and succinate uptake systems in R . leguminosarum MNF3841 (Glenn
et al., 1980, 1984b).
Effect of competitors on fructose and succinate uptake
The specificity of the two uptake systems in cowpea Rhizobium NGR234 was examined by
pre-incubating cells with potential competitors and then measuring the rate of uptake in
comparison with untreated controls.
Glucose (10 mM) and galactose (10 mM) had little effect on [14C]fructose uptake [26
nmol min- (mg protein)-' compared with 30.0 in untreated controls]. Sorbose (10 mM) and
mannose (10 mM) both markedly lowered fructose uptake to 1.8 nmol min- (mg protein)- l,
while mannitol was without effect. These results are similar to those of Glenn et al. (1984b),
where sorbose strongly inhibited [14C]fructoseuptake in R. leguminosarum MNF3841.
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Mon, 31 Jul 2017 22:58:45
Carbon utilization in cowpea Rhizobium NGR234
1813
[14C]Succinateuptake fell 98% in the presence of malate, fumarate or oxaloacetate (10 mM),
suggesting that these organic acids are transported on a common system. The succinate
analogues 2,2-dimethylsuccinate and itaconic acid (10 mM) lowered the uptake of [ 14C]succinate
by 84% and 87%, respectively.
These data suggest a picture of a versatile organism capable of using a wide range of carbon
compounds for growth. In strain NGR234 most of the systems for the catabolism of these carbon
substrates are inducible, in marked contrast to the pattern which has emerged with R .
leguminosarum MNF3841 (Glenn & Dilworth, 1981a). This difference highlights the danger of
generalizing about rhizobial physiology from data obtained with one or two strains.
Isolated bacteroids of NGR234 under the environmental conditions used in these
experiments appear only able to utilize C,-dicarboxylic acids. The succinate permease is
induced to a significant extent in the bacteroids, providing clear evidence for the utilization of
C4-dicarboxylic acids by bacteroids in the nodule. The fructose permease, which in laboratory
cultures is induced by fructose or sucrose appears not to be induced in bacteroids of NGR234. A
tentative conclusion from this observation is that the bacteroids are never exposed to sufficient
fructose or sucrose for the fructose permease to be induced.
Work with mutants of R . trifolii (Ronson et al., 1981)and R . leguminosarum (Glenn & Brewin,
1981 ;Finan et al., 1983) which cannot utilize C4-dicarboxylic acids has shown that such mutants
are able to nodulate but not to fix N2 (Nod+ Fix-). However, mutants of R . trifolii (Ronson &
Primrose, 1979) and R . leguminosarum (Glenn et al., 1984a) which cannot utilize sugars are able
to nodulate and fix N2.The data in this paper provide additional information on the importance
of dicarboxylic acids for bacteroid function and suggest that snake bean bacteroids may not
have access to significant concentrations of sugars.
S . S. gratefully acknowledges receipt of an ADAB studentship. Part of this work was supported by the
Australian Research Grants Scheme.
REFERENCES
BROUGHTON,
W.J. & DILWORTH,
M. J. (1971). Control
of leghaemoglobin synthesis in snake beans. Biochemical Journal 125, 1075-1 080.
BROWN,C. M. & DILWORTH,
M. J. (1975). Ammonia
assimilation by Rhizobium cultures and bacteroids.
Journal of General Microbiology 86, 39-48.
CHEN,Y. P., GLENN,A. R. & DILWORTH,
M. J. (1984).
Uptake and oxidation of aromatic asubstrates by
Rhizobium leguminosanun MNF3841 and Rhizobium
trifolii T A1. FEMS Microbiology Letters 21,20 1-205.
DILWORTH,
M. J. & GLENN,A. R. (1981). Control of
carbon substrate utilization by rhimbia. In Current
Perspectives in Nitrogen Fixation, pp. 244-256. Edited
by A. H. Gibson & W. E. Newton. Canberra,
Australia: Australian Academy of Science.
DILWORTH,M. J., MCKAY, I., FRANKLIN,
M. &
GLENN,A. R. (1983). Catabolite effects on enzyme
induction and substrate utilization in Rhizobium
leguminosarum.Journal of General Microbiology 129,
359-366.
FINAN,T. M., WOOD,J. M. & JORDAN,D. C. (1981).
Succinate transport in Rhizobium leguminosarum.
Journal of Bacteriology 146, 193-202.
FINAN,T. M., WOOD,J. M. & JORDAN,D. C. (1983).
Symbiotic properties of C,-dicarboxylic acid transport mutants of Rhizobium leguminosarum.Journal of
Bacteriology 154, 1403-1413.
GLENN,A. R. & BREWIN,N. J. (1981). Succinateresistant mutants of Rhizobium leguminosarum.Journal of General Microbiology 126, 237-241.
GLENN,A. R. & DILWORTH,
M. J. (1981 a). Oxidation
of substrates by isolated bacteroids and free-living
cells of Rhizobium leguminosarum 3841. Journal of
General Microbiology 126, 243-247.
GLENN,
A. R. & DILWORTH,
M. J. (1981 6). The uptake
and hydrolysis of disaccharides by fast- and slowgrowing species of Rhizobium. Archives of Microbiology 129, 233-239.
GLENN,A. R., POOLE,P. S. & HUDMAN,
J. F. (1980).
Succinate uptake by free-living and bacteroid forms
of Rhizobium leguminosarum. Journal of General
Microbiology 119, 267-271.
GLENN,A. R., MCKAY,I. A., ARWAS,R. & DILWORTH, M. J. (1984a). Sugar metabolism and the
symbiotic properties of carbohydrate mutants of
Rhizobium leguminosarum.Journal of General Microbiology 130, 239-245.
GLENN,A. R., ARWAS,R., MCKAY,I. A. & DILWORTH,
M. J. (1984b). Fructose metabolism in wildtype, fructokinase-negative and revertant strains of
Rhizobium leguminosarum.Journal of General Microbiology 130, 231-237.
HUDMAN,
J. F. &GLENN,A. R. (1980). Glucose uptake
by free-living and bacteroid forms of Rhizobium
leguminosarum. Archives of Microbiology 128, 7277.
MORRISON,
N. A., HAU,C. Y.,TRINICK,M. J., SHINE,
J. & ROLPE,B. G. (1983). Heat curing of a sym
plasmid in a fast-growing Rhizobium sp. that is able
to nodulate legumes and the nonlegume Parasponia
sp. Journal of Bacteriology 153, 527-531.
PARKER,
C. A., TRINICK,
M. J. & CHATEL,
D. L. (1977).
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Mon, 31 Jul 2017 22:58:45
1814
S . SAROSO, A . R . G L E N N A N D M . J . D I L W O R T H
Rhizobia as soil and rhizosphere inhabitants. In A
Treatise on Dinitrogen Fixation, Section IV : Agronomy and Ecology, pp. 31 1-352. New York: Wiley.
RAY,W. J. (1967). Photochemical oxidations. Methods
in Enzymology 11, 490-497.
RONSON,C. M., LYTTLETON,P. & ROBERTSON,
J. G.
(1981). C,-Dicarboxylate transport mutants of Rhizobium trifolii form ineffective nodules on Trifolium
repens. Proceedings of the National Academy of
Sciences of the United States of America 78, 42844288.
RONSON,C. W. & PRIMROSE,
S. B. (1979). Carbohydrate metabolism in Rhizobium trifolii: identification
and symbiotic properties of mutants. Journal of
General Microbiology 112, 77-88.
VINCENT,
J. M. (1977). A Manualfor the Practical Study
of the Root Nodule Bacteria. Oxford: Blackwell
Scientific Publications.
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Mon, 31 Jul 2017 22:58:45