Journal of General MicorbioIogy (1980), 120, 5 17-521. Printed in Great Britain
517
SHORT C O M M U N I C A T I O N
Anaerobic Growth and Denitrification by
Rhizobium japonicum and Other Rhizobia
ByR. M. D A N I E L , l * I . M. S M I T H , l J . A. D. P H I L L I P , ,
H . D. R A T C L I F F E , 3 J . W. D R O Z D 3 A N D A. T . B U L L 4
School of Science, University of Waikato, Hamilton, New Zealand
Department of Cell Biology, University of Glasgow, Glasgow G12 8QQ
3h-ShellBiosciencesLaboratory, Sittingbourne Research Centre, Sittingbourne, Kent ME9 8AG
Department of Applied Biology, University of Wales Institute of Science and
Technology, Cardifl CF 1 3NU
(Received 6 May 1980)
The product of nitrate respiration in Rhizobium japonicum and a number of other rhizobia
capable of anaerobic growth utilizing nitrate was N,O. No N, or ammonia was1formed.
Nitrate reduction was linked to ATP formation during anaerobic but not during aerobic
growth. Free-living rhizobia may remove fixed nitrogen from the soil by denitrification.
INTRODUCTION
Rhizobia are described in Bergey’s Manual of Determinative Bacteriology (Buchanan &
Gibbons, 1974) as aerobic bacteria utilizing oxygen as the terminal electron acceptor.
Nevertheless, it has been known for some time that the oxygen tension inside the root
nodule is low (see, for example, Appleby, 1962; Dilworth & Appleby, 1979) and that this is
an essential requirement for nitrogen fixation by both symbiotic and laboratory-grown freeliving rhizobia (Bergersen & Turner, 1968; Tjepkema & Evans, 1975; Keister, 1975). The
ability of free-living Rhizobium japonicum strain 505 (Wisconsin) to grow anaerobically
utilizing nitrate as a terminal electron acceptor (Daniel & Appleby, 1972) provided a useful
tool for studying anaerobically induced aspects of the rhizobium-legume symbiosis (Keister,
1975; Daniel & Appleby, 1972; Phillips et al., 1973a, b ; Bishop et al., 1975) and the possible
role of nitrate reductase in the control of nitrogenase (Evans & Russel, 1971; Kondorosi
et al., 1973). However, anaerobic growth is poor and relatively uncharacterized, its field
significance is unknown, and attempts to grow other rhizobia anaerobically have been
generally unsuccessful (see, for example, Dilworth & Appleby, 1979). We describe here
some of the features and bioenergetics of anaerobic growth, indicate how the low final
biomass concentrations may be increased, and show that some other rhizobia are able to
grow anaerobically. We also present evidence that under laboratory conditions free-living
R.japonicum 505 exhibits substantial rates of denitrification and is therefore potentially
capable of removing fixed nitrogen from soils.
METHODS
Bacterial strains and growth. RhizobiLm japonicum strain 505 (Wisconsin) was obtained as CC 705 from
the Division of Plant Industry, CSIRO, Canberra, Australia; R. japonicum ATCC 10324 (received as PDD
2864), R. japonicum WB 61 (received as PDD 2882), R . lupini PDD 4771, R. leguminosarumTA 101 (received
Downloaded
from www.microbiologyresearch.org by
0022-1287/80/oooO-9327 $02.00
0 1980 SGM
IP: 88.99.165.207
On: Sun, 18 Jun 2017 10:30:35
518
Short communication
as PDD 3661), R. meliloti SU 47 (received as PDD 2752) and R. phaseoli CC 511 (received as PDD 3305)
were all obtained from the Plant Diseases Division, DSIR, Mount Albert, Auckland, New Zealand.
Rhizobium japonicum 505 was grown aerobically and anaerobically as described by Daniel & Appleby
(1972). Glass bottles filled with a yeast extract/mannitol medium (Daniel & Appleby, 1972) and layered
with 1 cm of liquid paraffinwere used for anaerobiccultures of all rhizobia. Cultures were flushed with argon
,
otherwise stated.
and sealed with Suba-seals:the medium contained 6 ~ M - K N Ounless
Growth experimentsto detect any N, formation were carried out in 250 ml Quicklit flasks, fitted with taps,
which were flushed with argon before sampling.
Test tubes flushed with argon and sealed with Suba-seals were used for anaerobic dialysis cultures. Most
(80 to 95 %) of the growth medium was enclosed in a sealed dialysis sac, and remained sterile during the
experiment. Test tube and dialysis sac diameters were chosen so that the remainder of the medium was
spread in a thin layer between the dialysis sac and the test tube wall; this was inoculated, and bacterial growth
only occurred in this portion of the medium.
Anaerobic plates were incubated in GasPak (BBL) anaerobic jars in an atmosphere of 95 % H,/5 %
COz(v/v). All cultures were incubated at 25 "C.
Growth was assessed turbidimetrically at 680 nm in a cuvette with a 1 cm light path, and by measurement
of dry weight and cell protein (Herbert ot a/., 1971). Viable numbers were determined as described by
Postgate (1969). Plates for viable counts of both aerobically and anaerobically grown rhizobia were incubated aerobically.
Nodulution. Tests of the nodulation ability of R. japonicum 505 were made by conventional methods (see,
for example, Vincent, 1970; Rigaud et a/., 1973) using soybean (Glycine max Merr., cv. Lincoln) grown in
sterile vermiculite.
Gas analysis. NzO, CO, and NH3 were measured using a Varian 3700 gas chromatograph fitted with a
thermal conductivitydetector using a Poropak Q column with He as the carrier gas. N2was assayed similarly
but using a molecular sieve 5A (Linde) column.
Respiration-drivenproton translocation. These experimentswere made according to the method of Haddock
& Jones (1977). The electron donor in all cases was the endogenous substrate remaining after three washings
of the bacteria in 0.1 M-N~,HPO,/KH,PO~
buffer, pH 7.0.Data are cited as means k standard deviation,
with the number of observations in parentheses.
Other methods. Nitrate and nitrite were determined as described by Daniel & Appleby (1972).
RESULTS A N D DISCUSSION
Anaerobic growth of R. japonicum 505 appeared to be completely dependent upon the
presence of nitrate. Attempts to replace nitrate with alternative electron acceptors, fermentable substrates or other nitrogen sources were all unsuccessful.
Anaerobic final biomass concentrations were poor (Fig. 1). Bacterial numbers and biomass, assessed by viable counts, AaW,dry weight (not shown) and cell protein (not shown),
were less than 15 % of those attained with aerobically grown cultures. During anaerobic
growth the concentration of nitrate decreased and disappeared by about day 10; nitrite
appeared transiently and was about 1 m~ between days 5 and 10 (Daniel & Appleby, 1972).
The final biomass concentration was not improved by the addition of further nitrate at any
point during growth including after the disappearance of nitrite. We were unable to determine why anaerobic biomass concentrations were so poor. They were unaffected by the
, by 3 m-N,O (the end-product of nitrate respiration : see below) ;
presence of 2 ~ M - K N Oor
nor were final biomass concentrations improved in cultures flushed continuously with high
purity nitrogen or argon. However, high anaerobic bacterial densities, comparable with those
commonly found after aerobic growth, could be obtained by dialysis culture (for details, see
Methods). The dry weight of R. japonicum organisms produced per unit of total culture
volume (inside and outside the dialysis sac) was approximately the same as for undialysed
cultures. Growth rates were difficult to assess since bacteria tended to adhere to the dialysis
tubing. Growth was not improved in controls with open dialysis tubing or in the absence of
nitrate. The two most likely explanations for these results are either that an essential
dialysable medium constituent was present in limiting amounts, or that a dialysable growth
inhibitor was produced during growth. Increasing the concentration of individual medium
constituents did not improve the final anaerobic biomass concentrations and we believe the
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sun, 18 Jun 2017 10:30:35
Short communiration
519
0.20
0.16
2 0.12
\o
T:
0.08
0.04
2
4
6
8
10
Time (d)
12
14
16
Fig. 1. Viable counts (0,
0)
and Adso(0,a) during anaerobic growth of R. jupnicurn 505 in the
presence (0,
0)and absence (0,
a) of 6 ~ M - K N O ,and
, N 2 0production ( x ) in the presence of
6 mM-KNo3.No N20was produced in the absence of KNOs.
latter explanation to be the more likely. Demolon (1952) reported the accumulation of a
toxic factor in spent rhizobium growth medium which prevented fresh growth of the same
organism. It is conceivable that this effect, and the much more marked effect reported here,
are related to the cessation of bacterial growth which occurs early in root nodule development. However, the growth inhibition reported here was completely reversible. We found
that 10 d cultures of anaerobically grown R. japonicum 505 would successfully inoculate
aerobic and anaerobic cultures and nodulate soybean.
The final product of nitrate respiration was N20. Nitrate-nitrogen was quantitatively
recovered as N 2 0 (Fig. 1) and we could detect no N2 or NH3 in the culture fluid or gas
headspace after anaerobic growth. These results suggest that nitrous oxide reductase is
absent, as it is from many other denitrifying bacteria. Cowpea bacteroids also produced
N20but not N2during denitrification (Zablotowicz & Focht, 1979). These authors suggested
that this was due to the inhibition of nitrous oxide reductase by the acetylene in their assays
rather than to the absence of this enzyme.
The concentration of K N 0 3 used in our experiments was comparable with soil nitrate
concentrations and the numbers of rhizobia were within an order of magnitude of those
found in soils in which legumes are growing or have recently grown : an important implication is that in poorly aerated soils, free-living rhizobia are likely to have a substantial
capacity for denitrification.
Respiration-driven proton translocation experiments showed that anaerobic R. japonicum
was capable of using either oxygen [+H+/acceptor ratio (g equivalents of H+ translocated
outwards per g atom acceptor consumed) = 5.7 & 0.48(6)] or nitrate [+H+/acceptor ratio =
2.7 & 0.53 (6)] as electron acceptor for proton translocation and presumably ATP generation.
Aerobic bacteria were capable of using oxygen [+H+/acceptor ratio = 5.8 & 1.26 (6)] but
not nitrate [-+H+/acceptor ratio=O]. For oxygen the ratios correspond to a P/O ratio of
about 3 for NAD-linked substrates if an +H+/ATP ratio of 2 is assumed (Mitchell, 1966).
The +H+/acceptor ratio with nitrate is in good agreement with the relative efficiency of
anaerobic nitrogen fixation supported by nitrate respiration in bacteroids (Rigaud et al.,
1973) and corresponds to a P/acceptor ratio of 1 or 2. Oxygen uptake and nitrate reduction
rates are similar in aerobic, anaerobic and symbiotic R. japonicum (Daniel, 1979), with the
exception of nitrate reductase in aerobic bacteria which was only about 20 % of the other
rates. Aerobically grown R. japonicum was unable to couple this nitrate reductase to proton
translocation, suggesting that unlike the nitrate reductase in anaerobically grown bacteria,
it has an assimilatory rather than a respiratory function. This functional difference between
the nitrate reductase found in aerobic R . japonicum and that in anaerobic bacteria and
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sun, 18 Jun 2017 10:30:35
520
Short communication
bacteroids supports an earlier report (Daniel & Gray, 1976) of differences in molecular
weight and inhibitor sensitivities. Nitrite reductase is found only in anaerobically grown
R. japonicum (Daniel & Appleby, 1972) and the -+H+/acceptor ratio with nitrite was zero
for both aerobic cultures and bacteria grown anaerobically with nitrate as the terminal
electron acceptor.
We have tested a number of other rhizobia for the ability to grow anaerobically with
nitrate, using the criteria of increase in absorbance at 680 nm, C 0 2 production and N20
production, compared with controls lacking nitrate. Like R. japonicum 505, R. japonicum
ATCC 10324 (the type strain) and R . leguminosarum TA 101 were positive by all three
criteria. Rhizobium leguminosarum TA 101 and R . japonicum 505 were the only organisms
consistently to exhibit visible growth on anaerobic agar plates containing 6 mM-KNO,.
Rhizobium japonicum WB 61 and R . lupini PDD 4771 grew anaerobically (the latter only
weakly) and produced much more C 0 2than anaerobic controls lacking nitrate, but did not
produce N20. We did not test whether these organisms metabolized nitrate to N, rather
than N,O. Rhizubiumphaseuli CC 511 did not grow anaerobically or produce N,O nor did it
produce more COz than controls. We were unable to obtain consistent results with R.
meliloti SU 47.
Our results show that some rhizobia are capable of utilizing nitrate respiration to support
anaerobic growth and we suggest that free-living rhizobia may remove fixed nitrogen from
the soil by denitrification. Zablotowicz & Focht (1979) suggest that the presence of a full
nitrogen oxide reductase system may be an important criterion in the selection of rhizobial
strains for some agricultural conditions. We agree in principle, although whether or not the
nitrous oxide reductase component is present seems irrelevant. However, although more
evidence is necessary, we believe that under most conditions denitrificationis likely to prove
an undesirable characteristic of rhizobial strains rather than a desirable one.
We thank A. W. Limmer for valuable assistance. Part of the work was supported by an
ARC grant to one of us (R.M.D.) and H.D.R. thanks the SRC for a CASE Research
Studentship.
REFERENCES
APPLEBY,
C. A. (1962). The oxygen equilibrium of
leghaemoglobin. Biochimica et biophysica acta 60,
226-235.
BERGERSEN,
F. J. &TURNER,
G. L. (1968). Comparative studies of nitrogen fixation by soybean root
nodules, bacteroid suspensions and cell-free
extracts. Journal of General Microbiology 53,
205-220.
BISHOP,P. E., EVANS,H. J., DANIEL,R. M. &
HAMPTON,
R. 0. (1975). Immunological evidence
for the capability of free-living Rhizobium
japonicum to synthesise a portion of a nitrogenase
component. Biochimica et biophysica acta 381,
248-256.
BUCHANAN,
R. E. & GIBBONS,
N. E. (editors) (1974).
Bergey’s Manual of Determinative Bacteriology,
8th edn. Baltimore: Williams & Wilkins.
DANIEL,
R. M. (1979). The occurrence and role of
ubiquinone in electron transport to oxygen and
nitrate in aerobically, anaerobically and symbiotically grown Rhizobium japonicum. Journal of
General Microbiology 110,333-337.
DANIEL,R. M. & APPLEBY,
C. A. (1972). Anaerobicnitrate, symbiotic and aerobic growth of Rhizobium japonicum: effects on cytochrome P-450,
other haemoproteins, nitrate and nitrite redue
tases. Biochimica et biophysica acta 275,347-354.
DANIEL,
R. M. & GRAY,J. (1976). Nitrate reductase
from anaerobically grown Rhizobium japonicum.
Journal of General Microbiology 96,247-25 1.
DEMOLON,
A. (1952). Contribution a I’dtude de la
symbiose bacterienne chez les legumineuses.
Revue g&n&ralsde botanique 58,489-519.
DILWORTH,
M. J. & APPLEBY,
C. A. (1979). Leghaemoglobin and Rhizobium haemoproteins. In
A Treatise on Dinitrogen Fixation, Section II,
pp. 691-764. Edited by R. W. F. Hardy & R. C.
Burns. New York: John Wiley.
EVANS,H. J. & RUSSEL,
S. A. (1971). Physiological
chemistry of symbiotic nitrogen fixation by
legumes. In The Chemistry and Biochemistry of
Nitrogen Fixation, pp. 191-244. Edited by J. R.
Postgate. London: Plenum Press.
HADDOCK,
B. A. & JONES, C. W. (1977). Bacterial
respiration. Bacteriological Reviews 41, 47-99.
HERBERT,
D., PHIPPS, P. J. & STRANGE,
R. E. (1971).
Chemical analysis of microbial cells. Methods in
Microbiology 5B, 209- 344.
KEISTER,D. L. (1975). Acetylene reduction by pure
cultures of rhizobia. Journal of Bacteriology 123,
1265-1268.
KONDOROSI,
A., BARABAS,
I., SVAB,Z., OROSZ,L. &
SIK, T. (1973). Evidence for common genetic
determinants of nitrogenase and nitrate reductase
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sun, 18 Jun 2017 10:30:35
Short communication
in Rhizobium meliloti. Nature New Biology 246,
153-154.
MITCHELL,
P. (1966). Chemiosmotic coupling in
oxidative and photosynthetic phosphorylation.
Biological Reviews 41,445-502.
PHILLIPS,
D. A., DANIEL,
R. M., APPLEBY,
C. A. &
EVANS,
H. J. (1973 a). Isolation from Rhizobium of
factors which transfer electrons to soybean
nitrogenase. Plant Physiology 51,136-138.
PHILLIPS,D. A., HOWARD,
R. L. & EVANS,H. J.
(19736). Studies on the genetic control of a
nitrogenase component in leguminous root
nodules. Physiologia plantarum 28, 248-253.
POSTGATE,
J. R. (1969). Viable counts and viability.
Methods in Microbiology 1,611-628.
RIGAUD,J., BERGERSEN,
F. J., TURNER,G. L. &
521
DANIEL,
R. M. (1973). Nitrate-dependent anaerobic acetylene-reduction and nitrogen-fixation by
soybean bacteroids. Journal of General Microbiology 77, 137-144.
TJEPKEMA,
J. & EVANS,H. J. (1975). Nitrogen
fixation by free-living rhizobium in a defined
liquid medium. Biochemical and Biophysical
Research Communications65,625-628.
VINCENT,
J. M. (1970). A Manual for the Practical
Study of the Root Nodule Bacteria. Oxford &
Edinburgh : Blackwell Scientific Publications.
ZABLOTOWICZ,
R. M. & FOCHT,D. D. (1979).
Denitrification and anaerobic, nitrate-dependant
acetylene reduction in cowpea rhizobium. Journal
of General Microbiology 111,445448.
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sun, 18 Jun 2017 10:30:35