Planta
Planta (1988) 174:51-58
9 Springer-Verlag 1988
Utilization of nitrate by bacteroids of Bradyrhizobiumjaponicum in the soybean root nodule C. Giannakis, D.J.D. Nicholas and W. Wallace* Department of Agricultural Biochemistry, Waite Agricultural Research Institute, University of Adelaide, Glen Osmond, S.A. 5064, Australia
Abstract. Bacteroids of Bradyrhizobium japonicum strain CB1809, unlike CC705, do not have a high level of constitutive nitrate reductase (NR; EC 1.7.99.4) in the soybean (Glycine max. Merr.) nodule. Ex planta both strains have a high activity of N R when cultured on 5 m M nitrate at 2% Oa (v/v). Nitrite reductase (NiR) was active in cultured cells of bradyrhizobia, but activity with succinate as electron donor was not detected in freshly-isolated bacteroids. A low activity was measured with reduced methyl viologen. When bacteroids of CC705 were incubated with nitrate there was a rapid production of nitrite which resulted in repression of N R . Subsequently when N i R was induced, nitrite was utilized and N R activity recovered. Nitrate reductase was induced in bacteroids of strain CB1809 when they were incubated in-vitro with nitrate or nitrite. Increase in N R activity was prevented by rifampicin (10 gg. m l - 1) or chloramphenicol (50 gg. m l - 1). Nitritereductase activity in bacteroids of strain CB1809 was induced in parallel with NR. When nitrate was supplied to soybeans nodulated with strain CC705, nitrite was detected in nodule extracts prepared in aqueous media and it accumulated during storage (1 ~ C) and on further incubation at 25 ~ C. Nitrite was not detected in nodule extracts prepared in ethanol. Thus nitrite accumulation in nodule tissue appears to occur only after maceration and although bacteroids of some strains of B. japonieum have a high level of a constitutive N R , they do not appear to reduce nitrate in the nodule because this anion does not gain access to the bacteroid zone. Soybeans nodulated with strains CC705 and CB1809 were equally sensitive to nitrate inhibition of N2 fixation. * To whom correspondence should be addressed Abbrev&tions: N R = nitrate reductase; NiR = nitrite reductase; Tris = 2-amino-2-(hydroxymethyl)-1,3-propanediol
Key words: Bacteroid - Bradyrhizobium - Glycine (N2 fixation) - Nitrate reductase - Nitrite reductase - Nitrogen fixation.
Introduction Under anaerobiosis, growth of rhizobia is dependent on their capacity to denitrify, i.e. utilize nitrate as a terminal electron acceptor and reduce nitrite produced to nitrogenous gases (Daniel et al. 1982). Both nitrate reductase (NR) and nitrite reductase (NiR) have been detected in the rhizobial cell, especially when cultured on nitrate at low O2 tension (Daniel and Appleby 1972). In the bacteroid zone of the legume root nodule where the 02 concentration is very low (10 nM; Wittenberg 1980) high levels of N R have been demonstrated. A constitutive N R was first identified in the soybean nodule by Evans (1954) and subsequently studied by Evans and associates (e.g. Lowe and Evans 1964) and in other laboratories (Kennedy et al. 1975; Stephens and Neyra 1983; Alikulov et al. 1980). Evidence for the occurrence of N i R and utilization of the nitrite by bacteroids is contradictory. O ' H a r a et al. (1983) showed that bacteroids of Bradyrhizobium japonicum, strain CC705 denitrifled nitrate to nitrous oxide. However, van Berkum and Keyser (1985) found that metabolism of nitrate by bacteroids of several strains of B. japonicure resulted only in nitrite accumulation, irrespective of the gaseous product obtained from freeliving cells cultured under the same conditions. Bhandari and Nicholas (1984) demonstrated a rapid utilization of nitrate by bacteroids of B. japonicum strain CC705, but further reduction of nitrite to nitrogenous gases only occurred after incubation of the bacteroids for at least 10 h under anaer-
52
C. Giannakis et al. : Nitrate metabolism of bacteroids of soybean nodules
obiosis. Similarly, nitrous oxide was only evolved from isolated bacteroids of cowpea rhizobia after a relatively long period of incubation (Zablotowicz and Focht 1979). Smith and Smith (1986) observed very low rates of denitrification to nitrous oxide in bacteroids of two strains of B. japonicum, about 50- and 340-fold less than in anaerobic cultures of the same strains. There are conflicting reports on the occurrence of N i R in bacteroids, either its absence (Daniel and Appleby 1972; Stephens and Neyra 1983) or its activity measured (Chen and Sung 1983; Streeter 1985). The aim of this study was to examine the utilization of nitrate by cultured cells and bacteroids of B. japonicurn. Nitrate reductase is expressed in bacteroids of some strains of B. japonicum but the occurrence of N i R in bacteroids is uncertain and it has not been established whether nitrate utilization (assimilation or dissimilation) occurs in the infected zone of the nodule.
then macerated in 0.1 M K-phosphate, pH 7.5, containing 0.3 M sucrose, with a mortar and pestle. The extract was centrifuged at 200-g for 10 rain and the resultant supernatant centrifuged at t0000-g for 10 rain. The bacteroid pellet obtained was resuspended in 0.1 M K-phosphate, pH 7.5, and recentrifuged at 10000-g. All media and apparatus were autoclaved at 135 kPa and 120 ~ C for 20 min before use. Bacteroids were disrupted in 100mM K-phosphate, pH 7.5, containing 5 mM NaMoO4, by treatment with an ultrasonic probe (Branson Sonic Power Co., Danbury, Conn., USA) for six 30-s periods at 70 W and 4 ~ C. All buffers were chilled and sparged with argon for 20 min before use and during tissue maceration. Centrifuge tubes were flushed with argon, sealed, and centrifuged at 2 ~ C. The g values indicated are gav. In studies on the nodule cytosol, nodules were macerated in 0.2 M 2-amino-2-(hydroxymethyl)-l,3-propanediol (Tris), pH 8.5 (2 ~ C) containing 0.05 M K-phosphate, pH 8.5, 1% (w/ v) casein (BDH Chemicals, Poole, U K ; Hammarsten grade), 2.5% (w/v) insoluble polyvinylpyrrolidone, 10 gM flavin adenine dinucleotide, 5 mM ethylenediaminetetraacetic acid (EDTA), 5 mM cysteine and 0.3 M sucrose (1 g tissue F W + 3 ml extraction medium). The extract was filtered through two layers of Miracloth (Chicopee Mills, Milltown, N.J., USA) and centrifuged at 40000.g for 10 min. The supernatant fraction was used as the nodule cytosol.
Material and methods
Nitrate reductase (EC 1- 7.99.4) and nitrite reductase (EC 1.7-2.1): (J) In-vivo study with cells' and bacteroids. The
Bradyrhizobium strains and culture procedures. Strains CC705
incubation mixture (1 ml) contained 100raM K-phosphate (pH 7.5), 5 mM KNO3, 20 mM Na-succinate, 10 mM glucose and 0.2 mM diethyldithiocarbamate (DIECA) to inhibit NiR. An aliquot of washed cells or bacteroids (1 mg protein) was added and after incubation for 30 rain at 2 5 ~ under argon, nitrite production was measured. The NiR was determined as for NR, except that KNO2 (1 mM) was substituted for nitrate, D I E C A omitted and nitrite utilization was determined. (2) In-vitro assay for bacteroids. Both N R and NiR were assayed as above but with succinate and glucose replaced by 0.8 m M methyl viologen and 8 m M Na2S204 in 1% NaHICO~
and CB1809 were supplied by Dr. F.J. Bergersen, CSIRO Division of Plant Industry, Canberra, A.C.T., Australia. The culture medium contained 5 m M KNO3 and 14 mM glucose in 10 mM K-phosphate pH 7.5 (1 1 medium in 2-1 Erlenmeyer flask) with the other mineral elements and vitamins described by Brown and Dilworth (1975). Cultures were grown on a reciprocating shaker (70 strokes'min -1) at 30~ and sparged continuously with a mixture of 2% 02 and 98% N2 (v/v). Cells were harvested at 5 d (late exponential stage of growth) by centrifugation at 10000.g (10 min) and washed twice in 100 m M K-phosphate pH 7.5.
Soybean plants. Seeds of Glycine max L. (Merr.) cv. Clark were obtained from the Agricultural Research Station, Leeton, N.S.W., Australia. They were surface-sterilised by a brief treatment in 25% (v/v) ethanol (10 s) and 0.5% (w/v) I-IgClz (2 min), rinsed, and left overnight in sterile water. They were then planted in a mixture of sand and loam (3 : 1, v/v), each of which had been heat-sterilized. Bradyrhizobium inoculant (strain indicated in text) was supplied first at seed sowing and then after 7 d. Nutrient soh/tion was supplied twice weekly and water at other times as required. The nutrient solution was as follows: 0.63 mM K2SO4, 1.25 mM CaSO4-2H20, 0.5 m M MgSO47H20, 0.25 m M KH2PO4, 66 gM organic iron complex (Ruffin, Dodge City, Kan., USA supplied by Chemical Recovery Co., Welland, S.A., Australia), 60 ~tM FeSO4, 0.15 gM NaMoO4"2HzO, 0.1 gM CoC12"6HzO, 46 gM H3BO3, 9.2 IxM MnC12 "4H20, 0.9 laM ZnSO4"H20 and 0.3 gM CuCI2-2H20. The pH was adjusted to 6.0 with NaOH. Plants were grown in a growth room equipped with multivapour lamps (MVR 400 VBU; General Electric Co., Cleveland, Oh., USA) supplying approx. 1000 gmol quanta- m - 2 s - 1 photosynthetically active radiation at leaf level. The 16-h light period was at 30 ~ C, with 20 ~ C in the dark period.
Nodule fractionation and bacteroid isolation. Nodules were surface-sterilized by rinsing several times with sterile water before and after a brief rinse in 50% (v/v) ethanol (30 s). They were
(w/v). (3) In-vitro assay of plant enzymes. Extracts were prepared as described for nodule tissue (sucrose omitted) and N R assayed as described by Aryan et al. (/983) except that the reaction was terminated and excess N A D H oxidized by the procedure of Scholl etal. (1974). Nitrite-reductase activity (reduced methyl viologen as electron donor) was determined by the method of Hucklesby et al. (1972).
Nitrogenase activity. Nitrogenase activity was estimated by the C2He-reduction technique (Hardy et al. 1968). Plants were carefully removed from the root medium and placed in 1-1 jars. These were sealed and 100 ml acetylene injected into each jar through a rubber septum. The amount of CzH2 reduced was measured at 30 and 60 min by injecting a 50-gl sample into a gas chromatograph (Model 9A; Shimadzu Corp., Kyoto, Japan) ; fitted with a column, 1 m long, 3 mm diameter, of Porapak N (80-100 mesh; Waters Associates, Milford, Mass., USA). The rate of C2H2 reduction, checked at 3-min intervals, was linear up to at least 60 min.
Sucrose-gradient fractionation of bacteroids. A 3-ml sample of washed bacteroids (approx. 40 mg protein from 2.5 g nodule FW) in 0.05 M K-phosphate, pH 7.5, was layered on a discontinuous sucrose gradient prepared in the above buffer and comprising the following sucrose steps (w/w): 9 ml, 45%; 9 ml, 50%; 9 ml, 52% and 7 ml, 57%. Centrifugation was for 4 h at 100000-g (1 ~ C) in a SW-28 rotor (Beckman Instruments,
C. Giannakis et al. : Nitrate metabolism of bacteroids of soybean nodules Palo Alto, Ca, USA). Fractions (1.5 ml) were collected by upward displacement using a density-gradient fractionator (ISCO, Lincoln, Neb., USA).
Nitrate, nitrite and protein measurements. Nitrate was measured by the Escherichia coli N R method (Wallace 1986) and nitrite by the method of Nicholas and Nason (1957). Protein was determined using the micro-biuret method (Itzhaki and Gill 1964), except in the analysis of the sucrose gradients where the Folin phenol reagent of Lowry et al. (1951) was employed after a preliminary precipitation of the protein with 10% (w/v) trichloroacetic acid. Bovine serum albumin was used as a reference protein. Serological studies. Antiserum against strain CB1809 was kindly provided by the Australian Inoculants Research and Control Service, Department of Agriculture, Gosford, N.S.W. Bacteroid samples were incubated at 37 ~ C for 1-2 h with an appropriate level of antiserum to promote agglutination. After centrifugation at 250 .g for 3 min the supernatant was tested for bacteroid N R activity. Controls were run with bacteroids of strain CC705 and with 0.85% NaC1 (w/v) instead of antiserum. Analys& of data. Data presented represent experiments that have been repeated at least twice. Where replicate plant samples were tested, a statistical analysis of the data is included.
Results
Activity of N R and NiR in free-living cells and bacteroids. Cells of strains CB1809 and CC705, grown under 2% Oz (v/v), had high activities of N R and NiR (Table 1). The N R activity was in excess of that of NiR, especially in strain CC705. Bacteroids isolated from soybean plants inoculated with strain CC705 had a similar N R activity to that of freeliving cells (Table 1). In contrast, bacteroids of strain CB1809 had very low N R activity, only 1% of that of the free-living cells. The bacteroids of neither strain had detectable NiR activity when assayed with succinate as the electron donor. A low level of enzymic nitrite utilization by bacteroids was observed with reduced methyl viologen as electron donor (approx. 0.9 nmol NO~ utilized. r a i n - t - ( r a g protein)-i). However, with this assay a high level of non-enzymic utilization of nitrite occurred with boiled bacteroid samples (approx. 1.6 nmol NO2 utilized, m i n - 1. (rag protein- 1)). Comparison of nitrate utilization by bacteroids of strains CC705 and CB1809. Washed bacteroids of strain CC705 reduced nitrate to nitrite within a short time (3 h) followed by a rapid decline in N R activity (Fig. 1 A). After a 16 h incubation there was an increase in NiR activity, resulting in the utilization of the nitrite and a recovery of N R activity. In-vitro activities of N R and NiR (Fig. 1 B) correlated well with the rates of nitrate and nitrite reduction in bacteroids (Fig. 1 A). Inclusion of ri-
53
Table 1. Comparison of N R and NiR activities in cultured cells and bacteroids of Bradyrhizobium japonicum. Cells were grown under 2% (v/v) 02 as described in Materials and methods. Bacteroids were isolated from 37-d-old soybean plants supplied with a nutrient medium lacking combined nitrogen. The N R and NiR activities were measured in washed cells and bacteroids as described in Mater&& and methods (in-vivo study) and expressed as follows: NR, nmol nitrite produced.min-1. (nag protein)-l; NiR, nmol nitrite utilized-rain-1-(rag protein)- 1 Strain
CB1809 CC705
Free-living cells
Bacteroids
NR
NiR
NR
NiR
113 53
61 7
1 42
0 0
fampicin in the incubation medium with nitrate prevented the development of NiR activity in bacteroids (Fig. 1 C) so that the nitrite produced was not re-utilized. A similar result was obtained with 50 gg. ml-1 chloramphenicol (data not shown). When CB1809 bacteroids were incubated with nitrate (Fig. 2A), N R and NiR activities increased after 14 h, a slower increase in NiR resulting in a temporary accumulation of nitrite. The same pattern including a lag in the development of N R and NiR activities occurred in bacteroids of CB1809 incubated with nitrite (Fig. 2B). Inclusion of rifampicin (10 gg.m1-1) prevented the increase in both enzyme activities. Bacteroids of CB1809, incubated without nitrate or nitrite showed only a small increase in N R activity (to about 7 nmol NO 2 produced, m i n - 1. (mg protein)- i). Since the bacteroid fi'action could contain nontransformed bradyrhizobia or other bacteria, it was necessary to establish that the induction of N R and NiR activity was occurring in the bacteroid component. Ching et al. (/977) demonstrated that bacteroids could be separated from bacteria by centrifugation on a discontinuous sucrose gradient. Using this procedure the bacteroid sample of strain CB1809 was resolved into two fractions separated in the 45% and 50% (w/w) sucrose steps (Fig. 3). No bacterial cells were detected. Cultured cells of strain CB1809 separated in the 52% and 57% sucrose (w/w) fractions, mainly in the latter (data not shown). Induction of N R activity was demonstrated in the two bacteroid fractions isolated on the sucrose gradient (Fig. 3, inset). To confirm that enzyme induction was not occurring in contaminating bacteria in the bacteroid preparation, the effect of antiserum specific for strain CB1809 was tested. When bacteroids of CB1809 (incubated for 24 h as described in Fig. 1)
54
C. Giannakis et al. : Nitrate metabolism of bacteroids of soybean nodules
A
50
50
5
40
4
5
~ 1 4 9 ~176 ~
40
I
4
~
30
/-
T 2
20
m o
&
E
Z
20
2O
30
2
I0
T
.....~"
.E N
\
......~......o. .......
o
I0
3
s
v
I0
o~ o
0
40
I
I0
20
3o
B
._~
E
50 ............=:..........~"---~-
5
40
4
30
3
20
2
.+2_ .>_
t_
~, 50 r._
o o o.
"..
z
Z
~ 4o
tY Z
s 30 o
10 9
.~_
I
t0
0
20
INCUBATION ~
20
I0
0
30
C .IP .............
9 ...........
9 .................
9 .................
9
40
5 4
'"
B
30
5 i
o
2 &
20
'E- 4
A
ql}
I0
I
I0
20 INCUBATION
~
Initial
1.0
1.0
After
22.3
19.5
z o
a
NR activity (nmol nitrite produced 9min-1. (mg protein)- 1) A B
incubation
r i
(h)
Fig. 2. A, B. Utilization of nitrate and nitrite by isolated bacteroids of B. japonieum strain CB1809. Washed bacteroids were incubated (A) as described in Fig. 1A or with 5 m M KNO2 in place of KNO3 (B). A - - A , in-vivo N R ; ~ - - z x , in-vivo NiR; 9 ...... e, nitrite
40
n,Z
50
PERIOD
30
2
iA
30 PERIOD
40 (h)
Fig. l A-C. Utilization of nitrate by isolated bacteroids of B. japonicum strain CC705. Washed bacteroids (1 mg p r o t e i n - m l ~) were incubated at 2 5 ~ under anaerobic conditions. The incubation medium (10 ml) contained 100 m M K-phosphate (pH 7.5), 20 m M Na-succinate, 10 m M glucose, 5 m M KNOa (A, B) plus 10 pg-ml -~ rifampicin (C). In-vivo N R and NiR activities (A, C) and in-vitro N R and N i R activities (B) were measured at the times indicated, as described in Materials and methods. A - - A , N R ; zx--zx, NiR. Nitrite content of the incubation medium was also determined, e - - - e
1 0
~ - I ~ 0
!~1-~o50~__~1 n4_, 52% _.1~~1.S7%.~, 5
10 15 Fraction No
20
Fig. 3. A bacteroid fraction from nodules of 32-d-old soybean plants inoculated with B.japonicum strain CB1809 was fractionated on a discontinuous sucrose gradient as described in the Materials and methods and with the sucrose steps (w/w) indicated. Nitrate reductase activities of A and B are shown in the inset, before and after incubation at 25 ~ C for 24 h as described in Fig. 1
C. Giannakis et al. : Nitrate metabolism of bacteroids of soybean nodules
55
Table 2. Effects of nitrate on growth and nitrogenase activity of soybean. Plants inoculated with B. japonicum strains CC705 or CB1809 were harvested at 36 d after treatment with nitrate for 4 d as indicated. The data represent the mean of at least four replicates with SE Nitrate
FW/plant (g)
(mM)
Shoot
Root
Nodule
Nitrogenase (gmol C2H2 reduced h - 1. (g nodule F W ) - 1)
CC705
0 10
10.9• 14.6•
8.8• 11.2•
2.0• 1.8•
8.8• 3.3!0.5
CB1809
0 10
13.3• 17.5•
11.3• 14.4•
1.4• 1.2•
9.9• 3.8•
Strain
were treated with the antiserum, agglutination occurred and 80% of the N R activity was sedimented by centrifugation at 250.g for 3 min. Nitrate-reductase activity in bacteroids of CC705 was not inhibited by this antiserum.
Comparison of N2 fixation in soybeans inoculated with CB1809 and CC705 and their sensitivity to inhibition by nitrate. Plants inoculated with strain CC705 exhibited a much more pronounced "nitrogen-hunger phase" between 15 and 25 d after planting and thus in experiments undertaken at approx. 30 d (Table 2) the plants inoculated with CC705 were consistently smaller than those inoculated with CB1809. Nodules of the CC705 plants evolved a 70-fold higher amount of H 2 than that of CB 1809 (data not shown). Nitrate treatment (10 mM) for 4 d (Table 2) resulted in a 63% inhibition of nitrogenase activity in plants nodulated with either strain. Occurrence of NR and NiR and accumulation of nitrate and nitrite in soybean nodules and roots. In plants treated with nitrate for 4 d (Table 3) there was a relatively small change in bacteroid N R activity (succinate-dependent) or nodule cytosol N R activity (NADH-dependent), whereas activity of the root enzyme was increased. Bacteroid N R activity of strain CC705 was considerably in excess of that in either the nodule cytosol or root. The nodule cytosol had a low N i R activity not influenced by nitrate supply to the roots, in contrast to the activity of the root enzyme which was increased about 50-fold. A greater accumulation of nitrate was apparent in the root than nodule cytosol (Table 3), and only a trace was detected in the bacteroid fraction. Nodules extracted in Tris buffer (pH 8.5) contained nitrite in the cytosol fraction (Table 4). Further production of nitrite occurred when the extracts were incubated at 2 5 ~ and even at 1~ C. When the nodules were extracted with ethanol no
3. Influence of nitrate on the N R and NiR activities and nitrate content of soybean root and nodule tissues in plants inoculated with B. japonicum strains CB1809 or CC705. Bacteroid and nodule cytosol data are expressed on the basis of nodule FW. Details of the plant material are given in Table 2
Table
CB1809 Nitrate 0
CC705 Nitrate 10 m M
0
10 m M
30.8 1.2 0
37.9 0.6 0.5
N R - gmol NO 2 produced, h - 1 (g F W ) - 1 Bacteroid Nodule cytosol Root NiR
-
0.6 0.7 0
1.8 0.7 0.9
Ixmol N O 2 utilised, h - 1. (g FW) - 1
Bacteroid Nodule cytosol Root
0 1.1 0.5
0 1.4 27.3
0 1.9 0.5
0 1.9 25.6
0.02 0.3 1.6
0.05 13.8 40.2
0.04 0.1 L1
0.06 12.1 38.7
NO 3 - gmol. (g F W ) - 1 Bacteroid Nodule cytosol Root
TaMe 4. Nitrite contents of nodule extracts prepared in various media. Nodules from plants inoculated with B. japonicum strain CC705 were extracted in a mortar and pestle using the medium indicated (3 vol: 1 g nodule FW). Aliquots of the extract were centrifuged immediately (1 rain) and after 30 min at 25 ~ C. Nitrite in the supernatant was determined as described in Materials and methods Extraction medium
0.2 M Tris/0.05 M K-phosphate, pH 8.5 Ethanol, 100% (v/v) I M zinc acetate 0.025 M K-phosphate, pH 10.0 (80 ~ C)
Nitrite (gmol-(g nodule F W ) - 1) 0 min
30 min
1.41 a
8.64
0 0.I0 0.36
0 0.08 8.17
After 60 min at 1~ C this value increased to 3.0 gmol.(g nodule FW) - 1
56
C. Giannakis et al. : Nitrate metabolism of bacteroids of soybean nodules
nitrite was detected or produced. Nitrite added to the nodule sample (0.25 gmol) was fully recovered in the ethanol extract. Only a trace o f nitrite was detected when I M zinc acetate was the extractant. Extraction with a buffer at p H 10, preheated to 80 ~ C, did not prevent nitrite production in vitro. Nitrite accumulation also occurred in extracts of CB1809 nodules, but at a much slower rate, corresponding to their low level of bacteroid N R (data not shown). Discussion
There are two main types of N R in bacteria associated with nitrate assimilation and dissimilation, respectively (see review by Beevers and Hageman 1983). The N R in anaerobically cultured bradyrhizobia has the same relative molecular mass (Mr) as that in bacteroids (69000; Daniel and Gray 1976) and is likely to be a dissimilatory enzyme. Bacteroid N R purified from tupin nodules had an Mr of 67000 and utilized reduced ferredoxin as an electron donor (Alikulov et al. 1981). In aerobically-cultured cells Daniel and Gray (1976) identified an N R species with a larger Mr (180000). However, Kennedy et al. (1975) found that N R in aerobically grown cells had an Mr of 70 000. In this study DIECA was used to inhibit N i R of cultured cells without affecting N R . Diethyldithiocarbamate is an irreversible inhibitor of copper-dependent enzymes (Mann 1955) and one of the two types of N i R identified in denitrifying bacteria is a copper-containing flavoprotein (Knowles 1982). Relatively little is known about the N i R enzymes in rhizobia, but it is likely that there are separate enzymes for the assimilation of nitrite to ammonium and its dissimilation to nitrogenous gases. In the nodule cytosol there is an N A D H - d e p e n dent N R (Streeter 1982; Stephens and Neyra 1973) and an N i R assayed with reduced methyl viologen (Hunter 1984). These have similar properties to the assimilatory enzymes of root tissue. Bacteroids of Bradyrhizobiumjaponicum strain CB1809 have a low activity of N R compared to strain CC705, even though both have high enzyme activity ex planta. Nitrate-reductase activity in cultured cells was increased at reduced 02 concentrations (Daniel and Appleby 1972) and thus the low 02 environment o f the infected region of the nodule would be suitable for the expression of the enzyme in bacteroids. Vairinhos et al. (1986) showed that strain U S D A 7 6 was similar to CC705 in expressing a high level of N R in the bacteroids, while RGS527 and CB1003, like CB1809, had relatively
low bacteroid N R activities. Bergersen (1974) also noted that bacteroids o f B. japonicum strains CC705 and CB1809 differed widely in their nitrate reductase activities. Bacteroids of Rhizobium trifolii, R. phaseoli and R. leguminosarum do not have N R activity (Manhart and Wong 1979) but the enzyme was detected in bacteroids of R. meliloti and was induced by nitrate application to the alfalfa host (Becana et al. 1985 a). When bacteroids of strain CB1809 were isolated from the nodules and incubated anaerobically with nitrate or nitrite, N R was induced to the same level as that in bacteroids of CC705. In both cases, complete inhibition by chloramphenicol and rifampicin indicated a de-novo synthesis of the two enzymes. Rigaud (1976) has also demonstrated an induction of N R and N i R in anaerobic preparations of Phaseolus vulgaris bacteroids incubated for at least 10 h. After fractionation of the bacteroids of strain CB1809 on a sucrose gradient it was established that N R was induced in two mature bacteroid fractions (Fig. 3). With reference to the earlier work of Ching et al. (1977), the two bacteroid fractions identified in the current study represent a mature bacteroid fi'action (Fraction B Fig. 3) and a larger bacteroid component, possibly a group of bacteroids within the peribacteroid membrane (Bergersen and Appleby 1981). The latter component was not detected by Ching et al. but could have been masked in the large protein peak associated with the nodule cytosol, which they observed in this region of the sucrose gradient. In our study, the bacteroids were separated from the nodule cytosol by a preliminary centrifugation. Unlike Ching et al. we did not detect either transforming bacteria or bacteria in our bacteroid preparations. Our data thus demonstrate that mature bacteroids retain the ability to undertake de-novo synthesis of N R when supplied with nitrate. The observation that N R activity in bacteroids of CB1809 could be induced after isolation and incubation with nitrate and the lack of a similar response in situ, with nitrate supplied to the root medium, indicates that nitrate was not reaching the infected region of the nodule. From studies with the in-vivo assay of N R in bacteroids (with and without nitrate) Hunter (1983) also concluded that nitrate had limited access to the bacteroid zone of the nodule. Indeed, Pate and Atkins (1981) had earlier proposed that neither nitrate nor its reduction products reach the bacteroids. However, other workers have suggested that utilization of nitrate by bacteroids makes a contribution to its assimilation in the plant (Randall et al. 1978) or else results in nitrite accumulation (Stephens and
C. Giannakis et al. : Nitrate metabolism of bacteroids of soybean nodules
Neyra 1983; Becana et al. 1985b; Streeter 1985) which could inhibit nitrogenase (Kennedy et al. 1975; Rigaud and Puppo 1977). Sprent etal. (1987) have recently demonstrated that in both determinate and indeterminate nodules, nitrate has restricted access to the inner infected zone. It appears that nitrite accumulation in nodule tissue only occurs after maceration of the nodules, thus allowing the bacteroids access to the nitrate which is normally restricted to the outer cortex region. No nitrite was detected in ethanol extracts, while I M zinc acetate (as used by Manhart and Wong 1980) was also fairly effective in preventing nitrite accumulation. In our studies, extraction of nodules with a buffer at pH 10 preheated to 80 ~ C (the procedure of Stephens and Neyra 1983) did not prevent nitrite accumulation in vitro. We conclude that nitrite does not accumulate in the nodule and correlations between N R activity of bacteroids and nitrite accumulation in the nodule are likely to be an artefact of their isolation. Stephens and Neyra (1983) observed that nitrate inhibition of nitrogenase in isolated bacteroids was relieved in a NR-minus mutant strain of B. japonicum, but in detached nodules the effect of the mutant was less conclusive, Since nitrate available to the nodule appears not to gain access to the bacteroids (Sprent et al. 1987) the level of bacteroid N R will not affect the sensitivity of nodule nitrogenase to nitrate. Soybean plants nodulated with CC705 (high bacteroid NR) and CB1809 (low bacteroid NR) were equally sensitive to nitrate inhibition of nitrogenase (Table 2). Thus, as already concluded from studies with NR-deficient mutant strains of rhizobia (Gibson and Pagan 1977; Manhart and Wong 1980; Streeter 1982) there is no correlation between bacteroid N R activity and sensitivity of the N zfixation process to inhibition by nitrate. Bacteroids of CC705 and CB1809 when incubated with succinate exhibited no detectable NiR activity, in agreement with earlier studies on soybean by Daniel and Appleby (1972) and Stephens and Neyra (1983). Using reduced methyl viologen as an electron donor we detected a low level of NiR activity in bacteroids as observed by Chen and Sung (1983) and Streeter (1985). The activity measured (0.9 nmol NO~ utilized-min- 1 .(rag protein)- 1) was considerably lower than that induced in isolated bacteroids incubated with nitrate or nitrite and assayed with succinate as an electron donor (approx. 25 nmol NO~- utilized, m i n - 1. (mg protein)-1; Figs. 1, 2). When reduced methyl viologen was used to assay NiR activity in intact bacteroids, we observed that there was a relatively
57
high level of non-enzymic utilization of nitrite. Thus, although a boiled bacteroid sample was used as control, the estimate of enzymic activity may be inaccurate. The nodule cytosol had NiR activity, which was much lower than bacteroid N R and root NiR activities (Table 3, data for strain CC705). Unlike the root enzyme, nodule cytosol NiR did not increase when the plants were supplied with nitrate. Conclusion
Utilization of nitrate by bacteroids would result in either nitrite accumulation (not detected) or its further reduction by NiR (no induction of NiR activity observed in situ). Nitrite accmnulation would also repress the synthesis of bacteroid NR. We conclude that nitrate metabolism does not normally occur in the bacteroid zone of the soybean nodule, even when a dissimilatory N R is expressed, because of a restricted access of nitrate. Thus the inhibitory effect of nitrate on nitrogen fixation cannot be explained by nitrite accumulation in the nodule and its inhibition of nitrogenase. We thank R.G. Batt for technical assistance and C.G. acknowledges the receipt of a postgraduate scholarship from the University of Adelaide.
References Alikulov, Z.A., Burikhanov, Sh.S., Sergeev, N.S., L'vov, N.P., Kretovich, V.L. (1980) Nitrate reductase of lupine root nodule bacteroid. Prikl. Biokhim. Mikrobiol. (Transln.) I6, 372-376 Aryan, A.P., Batt, R.G., Wallace, W. (1983) Reversible inactivation of nitrate reductase by NADH and the occurrence of partially inactive enzyme in the wheat leaf. Plant Physiol. 71,582-587 Becana, M., Aparicio-Tejo, P.M. Sanchez-Diaz, M. (1985 a) Nitrate and nitrite reduction by alfalfa root nodules: Accumulation of nitrite in Rhizobium meliloti bacteroids and senescence of nodules. Physiol. Plant. 64, 353-358 Becana, M., Aparicio-Tejo, P.M., Sanchez-Diaz, M. (1985b) Nitrate and nitrite reduction in the plant fraction of alfalfa root nodules. Physiol. Plant. 65, 185-188 Beevers, L., Hageman, R.H. (1983) Uptake and reduction of nitrate: Bacteria and higher plants. In: Encyclopedia of plant physiology, N.S. vol. 15A: Inorganic plant nutrition, pp. 351-375, Lfiuchli, A., Bieleski, R.L., eds. Springer-Verlag, Berlin etc. Bergersen, F.J. (1974) Formation and function of bacteroids. In: The biology of nirogen fixation, pp. 473-497, Quispel, A., ed. North Holland. Publishing Co., Amsterdam Bergerson, F.J., Appleby, C.A. (1981) Leghaemoglobin within bacteroid-enclosingmembrane envelopes from soybean root nodules. Planta 152, 534~543 Bhandari, B., Nicholas, D.J.D. (1984) Denitrification of nitrate to nitrogen gas by washed cells of Rhizobium japonicum and by bacteroids from Glycine max. Planta 161, 8/-85
58
C. Giannakis et al. : Nitrate metabolism of bacteroids of soybean nodules
Brown, C.M., Dilworth, M.J. (1975) Ammonia assimilation by Rhizobium cultures and bacteroids. J. Gen. Microbiol. 86, 39-48 Chen, C-L., Sung, J-M. (1983) Effect of water stress on the reduction of nitrate and nitrite by soybean nodules. Plant Physiol. 73, 1065 1066 Ching, T.M., Hedtke, S., Newcomb, W. (1977) Isolation of bacteria, transforming bacteria and bacteroids from soybean nodules. Plant Physiol. 60, 771-774 Daniel, R.M., Appleby, C.A. (1972) Anaerobic-nitrate, symbiotic and aerobic growth of Rhizobium japonicum : effects on cytochrome P4so, other haemoproteins, nitrate and nitrite reductase. Biochim. Biophys. Acta 275, 347-354 Daniel, R.M., Gray, J. (1976) Nitrate reductase from anaerobically grown Rhizobiurn japonicum. J. Gen. Microbiol. 96, 247-251 Daniel, R.M., Limmer, A.W., Steele, K.W., Smith, I.M. (1982) Anaerobic growth, nitrate reduction and denitrification in 46 Rhizobium strains. J. Gen. Microbiol. 128, 1811-1815 Evans, H.J. (1954) Diphosphopyridine nucleotide-nitratereductase from soybean nodules. Plant Physiol. 29, 298-301 Gibson, A.H., Pagan, J.D. (1977) Nitrate effects on the nodulation of legumes inoculated with nitrate reductase-deficient mutants of Rhizobium. Planta 134, 17 22 Hardy, R.W.F., Holsten, R.D., Jackson, E.K., Burns, R.C. (1968) The acetylene-ethylene assay for N2-fixation : laboratory and field evaluation. Plant Physiol. 43, 1185-1207 Hucklesby, D.P., Dalling, M.J., Hageman, R.H. (1972) Some properties of two forms of nitrite reductase from corn (Zea mays L) scutellum. Planta 104, 220233 Hunter, W.J. (1983) Soybean root and nodule nitrate reductase. Physiol. Plant. 59, 471-475 Hunter, W.J. (1984) Purification and characterisation of soybean nodule nitrite reductase. Physiol. Plant. 60, 467-472 Itzhaki, R.F., Gill, D.M. (1964) A microbiuret method for estimating proteins. Anal. Biochem. 9, 401.410 Kennedy, I.R., Rigaud, J., Trinchant, J.C. (1975) Nitrate reductase from bacteroids of Rhizobium japonicum: enzyme characteristics and possible interaction with nitrogen fixation. Biochim. Biophys. Acta 397, 24-35 Knowles, R. (1982) Denitrification. Microbiol. Rev. 46, 43-70 Lowe, R.H., Evans, H.J. (1964) Preparation and some properties of a soluble nitrate reductase from Rhizobiumjaponicum. Biochim. Biophys. Acta 85, 377-389 Lowry, O.H., Rosebrough, N.J., Farr, A.L., Randall R.J. (1951) Protein measurement with the Folin phenol reagent. J. Biol. Chem. 193, 265-275 Manhart, J.R., Wong, P.R. (1979) Nitrate reductase activities of rhizobia and the correlation between nitrate reduction and nitrogen fixation. Can. J. Microbiol. 25, 1169-1174 Manhart, J.R., Wong, P.P. (1980) Nitrate effect on nitrogen fixation (acetylene reduction). Activities oflegmne root nodules induced by rhizobia with varied nitrate reductase activities. Plant Physiol. 65, 502-505 Mann, P.J.G. (1955) Purification and properties of the amine oxidase of pea seedlings. Biochem. J. 59, 609-620 Nicholas, D.J.D., Nason, A. (1957) Determination of nitrate and nitrite. Methods Enzymol. 3, 981-984
O'Hara, G.W., Daniel, R.M., Steele, K.W. (1983) Effect of oxygen on the synthesis, activity and breakdown of the Rhizobium denitrification system. J. Gem Microbiol. 129, 2405-2412 Pate, J.S., Atkins, C.A. (1981) Nitrogen uptake, transport and utilization. In: Ecology of nitrogen fixation, vol. 3, pp. 245298, Broughton, W.J., ed. Oxford University Press, Oxford, UK Randall, D.D., Russell, W.J., Johnson, D.R. (1978) Nodule nitrate reductase as a source of reduced nitrogen in soybean, Glycine max. Physiol. Plant. 44, 325-328 Rigaud, J. (1976) Effet des nitrates sur la fixation d'azote par les nodules de Haricot (Phaseolus vulgaris L.). Physiol. V6g. 14, 297-308 Rigaud, J., Puppo, A. (1977) Effect of nitrite upon leghemoglobin and interaction with nitrogen fixation. Biochim..Biophys. Acta 497, 702-706 Scholl, R.L., Harper, J.E., Hageman, R.H. (1974) Improvements of the nitrite color development in assays of nitrate reductase by phenazine methosulfate and zinc acetate. Plant Physiol. 53, 825-828 Smith, G.B., Smith, M.B. (1986) Symbiotic and free-living denitrification by Bradyrhizobium japonieum. J. Soil Sci. Soc. Am. 50, 349-353 Sprent, J.I., Giannakis, C., Wallace, W. (1987) Transport of nitrate and calcium into legume root nodules. J. Exp. Bot. 38, 1121-1128 Stephens, B.D., Neyra, C.A. (1983) Nitrate and nitrite reduction in relation to nitrogenase activity in soybean nodules and Rhizobiurn japonicum bacteroids. Plant Physiol. 71, 731 735 Streeter, J.G. (1982) Synthesis and accumulation of nitrite on soybean nodules supplied with nitrate. Plant Physiol. 69, 1429-1434 Streeter, J.G. (1985) Nitrate inhibition of legume nodule growth and activity. I. Long term studies with a continuous supply of nitrate. Plant Physiol. 77, 321-324 Vairinhos, F., Wallace, W., Nicholas, D.J.D. (1986) Denitrification versus assimilation of nitrate by Bradyrhizobiumjaponicurn. In: Proc. 8th Aust. Nitrogen Fixation Conf., pp. 137138, Wallace, W., Smith, S.E., eds. Publ. No. 25, Australian Institute of Agricultural Science, Melbourne Van Berkum, P., Keyser, H.H. (1985) Anaerobic growth and denitrification among different serogroups of soybean rhizobia. Appl. Environ. Microbiol. 49, 772-777 Wallace, W. (1986) Distribution of nitrate assimilation between the root and shoot of legumes and a comparison with wheat. Physiol. Plant. 66, 630-636 Wilttenberg, J.B. (1980) Utilization of leghemoglobin-bound oxygen by Rhizobium bacteroids. In: Nitrogen fixation, vol. II, pp. 53-67, Newton, W.E., Orme-Johnson, W.H., eds. University Park Press, Baltimore, Ma., USA Zablotowicz, R.M., Focht, D.D. (1979) Denitrification and anaerobic, nitrate-dependent acetylene reduction in cowpea Rhizobium. J. Gen. Microbiol. 111,445-448 Received 6 July; accepted 28 September 1987
Planta (1988) 174:51-58
9 Springer-Verlag 1988
Utilization of nitrate by bacteroids of Bradyrhizobiumjaponicum in the soybean root nodule C. Giannakis, D.J.D. Nicholas and W. Wallace* Department of Agricultural Biochemistry, Waite Agricultural Research Institute, University of Adelaide, Glen Osmond, S.A. 5064, Australia
Abstract. Bacteroids of Bradyrhizobium japonicum strain CB1809, unlike CC705, do not have a high level of constitutive nitrate reductase (NR; EC 1.7.99.4) in the soybean (Glycine max. Merr.) nodule. Ex planta both strains have a high activity of N R when cultured on 5 m M nitrate at 2% Oa (v/v). Nitrite reductase (NiR) was active in cultured cells of bradyrhizobia, but activity with succinate as electron donor was not detected in freshly-isolated bacteroids. A low activity was measured with reduced methyl viologen. When bacteroids of CC705 were incubated with nitrate there was a rapid production of nitrite which resulted in repression of N R . Subsequently when N i R was induced, nitrite was utilized and N R activity recovered. Nitrate reductase was induced in bacteroids of strain CB1809 when they were incubated in-vitro with nitrate or nitrite. Increase in N R activity was prevented by rifampicin (10 gg. m l - 1) or chloramphenicol (50 gg. m l - 1). Nitritereductase activity in bacteroids of strain CB1809 was induced in parallel with NR. When nitrate was supplied to soybeans nodulated with strain CC705, nitrite was detected in nodule extracts prepared in aqueous media and it accumulated during storage (1 ~ C) and on further incubation at 25 ~ C. Nitrite was not detected in nodule extracts prepared in ethanol. Thus nitrite accumulation in nodule tissue appears to occur only after maceration and although bacteroids of some strains of B. japonieum have a high level of a constitutive N R , they do not appear to reduce nitrate in the nodule because this anion does not gain access to the bacteroid zone. Soybeans nodulated with strains CC705 and CB1809 were equally sensitive to nitrate inhibition of N2 fixation. * To whom correspondence should be addressed Abbrev&tions: N R = nitrate reductase; NiR = nitrite reductase; Tris = 2-amino-2-(hydroxymethyl)-1,3-propanediol
Key words: Bacteroid - Bradyrhizobium - Glycine (N2 fixation) - Nitrate reductase - Nitrite reductase - Nitrogen fixation.
Introduction Under anaerobiosis, growth of rhizobia is dependent on their capacity to denitrify, i.e. utilize nitrate as a terminal electron acceptor and reduce nitrite produced to nitrogenous gases (Daniel et al. 1982). Both nitrate reductase (NR) and nitrite reductase (NiR) have been detected in the rhizobial cell, especially when cultured on nitrate at low O2 tension (Daniel and Appleby 1972). In the bacteroid zone of the legume root nodule where the 02 concentration is very low (10 nM; Wittenberg 1980) high levels of N R have been demonstrated. A constitutive N R was first identified in the soybean nodule by Evans (1954) and subsequently studied by Evans and associates (e.g. Lowe and Evans 1964) and in other laboratories (Kennedy et al. 1975; Stephens and Neyra 1983; Alikulov et al. 1980). Evidence for the occurrence of N i R and utilization of the nitrite by bacteroids is contradictory. O ' H a r a et al. (1983) showed that bacteroids of Bradyrhizobium japonicum, strain CC705 denitrifled nitrate to nitrous oxide. However, van Berkum and Keyser (1985) found that metabolism of nitrate by bacteroids of several strains of B. japonicure resulted only in nitrite accumulation, irrespective of the gaseous product obtained from freeliving cells cultured under the same conditions. Bhandari and Nicholas (1984) demonstrated a rapid utilization of nitrate by bacteroids of B. japonicum strain CC705, but further reduction of nitrite to nitrogenous gases only occurred after incubation of the bacteroids for at least 10 h under anaer-
52
C. Giannakis et al. : Nitrate metabolism of bacteroids of soybean nodules
obiosis. Similarly, nitrous oxide was only evolved from isolated bacteroids of cowpea rhizobia after a relatively long period of incubation (Zablotowicz and Focht 1979). Smith and Smith (1986) observed very low rates of denitrification to nitrous oxide in bacteroids of two strains of B. japonicum, about 50- and 340-fold less than in anaerobic cultures of the same strains. There are conflicting reports on the occurrence of N i R in bacteroids, either its absence (Daniel and Appleby 1972; Stephens and Neyra 1983) or its activity measured (Chen and Sung 1983; Streeter 1985). The aim of this study was to examine the utilization of nitrate by cultured cells and bacteroids of B. japonicurn. Nitrate reductase is expressed in bacteroids of some strains of B. japonicum but the occurrence of N i R in bacteroids is uncertain and it has not been established whether nitrate utilization (assimilation or dissimilation) occurs in the infected zone of the nodule.
then macerated in 0.1 M K-phosphate, pH 7.5, containing 0.3 M sucrose, with a mortar and pestle. The extract was centrifuged at 200-g for 10 rain and the resultant supernatant centrifuged at t0000-g for 10 rain. The bacteroid pellet obtained was resuspended in 0.1 M K-phosphate, pH 7.5, and recentrifuged at 10000-g. All media and apparatus were autoclaved at 135 kPa and 120 ~ C for 20 min before use. Bacteroids were disrupted in 100mM K-phosphate, pH 7.5, containing 5 mM NaMoO4, by treatment with an ultrasonic probe (Branson Sonic Power Co., Danbury, Conn., USA) for six 30-s periods at 70 W and 4 ~ C. All buffers were chilled and sparged with argon for 20 min before use and during tissue maceration. Centrifuge tubes were flushed with argon, sealed, and centrifuged at 2 ~ C. The g values indicated are gav. In studies on the nodule cytosol, nodules were macerated in 0.2 M 2-amino-2-(hydroxymethyl)-l,3-propanediol (Tris), pH 8.5 (2 ~ C) containing 0.05 M K-phosphate, pH 8.5, 1% (w/ v) casein (BDH Chemicals, Poole, U K ; Hammarsten grade), 2.5% (w/v) insoluble polyvinylpyrrolidone, 10 gM flavin adenine dinucleotide, 5 mM ethylenediaminetetraacetic acid (EDTA), 5 mM cysteine and 0.3 M sucrose (1 g tissue F W + 3 ml extraction medium). The extract was filtered through two layers of Miracloth (Chicopee Mills, Milltown, N.J., USA) and centrifuged at 40000.g for 10 min. The supernatant fraction was used as the nodule cytosol.
Material and methods
Nitrate reductase (EC 1- 7.99.4) and nitrite reductase (EC 1.7-2.1): (J) In-vivo study with cells' and bacteroids. The
Bradyrhizobium strains and culture procedures. Strains CC705
incubation mixture (1 ml) contained 100raM K-phosphate (pH 7.5), 5 mM KNO3, 20 mM Na-succinate, 10 mM glucose and 0.2 mM diethyldithiocarbamate (DIECA) to inhibit NiR. An aliquot of washed cells or bacteroids (1 mg protein) was added and after incubation for 30 rain at 2 5 ~ under argon, nitrite production was measured. The NiR was determined as for NR, except that KNO2 (1 mM) was substituted for nitrate, D I E C A omitted and nitrite utilization was determined. (2) In-vitro assay for bacteroids. Both N R and NiR were assayed as above but with succinate and glucose replaced by 0.8 m M methyl viologen and 8 m M Na2S204 in 1% NaHICO~
and CB1809 were supplied by Dr. F.J. Bergersen, CSIRO Division of Plant Industry, Canberra, A.C.T., Australia. The culture medium contained 5 m M KNO3 and 14 mM glucose in 10 mM K-phosphate pH 7.5 (1 1 medium in 2-1 Erlenmeyer flask) with the other mineral elements and vitamins described by Brown and Dilworth (1975). Cultures were grown on a reciprocating shaker (70 strokes'min -1) at 30~ and sparged continuously with a mixture of 2% 02 and 98% N2 (v/v). Cells were harvested at 5 d (late exponential stage of growth) by centrifugation at 10000.g (10 min) and washed twice in 100 m M K-phosphate pH 7.5.
Soybean plants. Seeds of Glycine max L. (Merr.) cv. Clark were obtained from the Agricultural Research Station, Leeton, N.S.W., Australia. They were surface-sterilised by a brief treatment in 25% (v/v) ethanol (10 s) and 0.5% (w/v) I-IgClz (2 min), rinsed, and left overnight in sterile water. They were then planted in a mixture of sand and loam (3 : 1, v/v), each of which had been heat-sterilized. Bradyrhizobium inoculant (strain indicated in text) was supplied first at seed sowing and then after 7 d. Nutrient soh/tion was supplied twice weekly and water at other times as required. The nutrient solution was as follows: 0.63 mM K2SO4, 1.25 mM CaSO4-2H20, 0.5 m M MgSO47H20, 0.25 m M KH2PO4, 66 gM organic iron complex (Ruffin, Dodge City, Kan., USA supplied by Chemical Recovery Co., Welland, S.A., Australia), 60 ~tM FeSO4, 0.15 gM NaMoO4"2HzO, 0.1 gM CoC12"6HzO, 46 gM H3BO3, 9.2 IxM MnC12 "4H20, 0.9 laM ZnSO4"H20 and 0.3 gM CuCI2-2H20. The pH was adjusted to 6.0 with NaOH. Plants were grown in a growth room equipped with multivapour lamps (MVR 400 VBU; General Electric Co., Cleveland, Oh., USA) supplying approx. 1000 gmol quanta- m - 2 s - 1 photosynthetically active radiation at leaf level. The 16-h light period was at 30 ~ C, with 20 ~ C in the dark period.
Nodule fractionation and bacteroid isolation. Nodules were surface-sterilized by rinsing several times with sterile water before and after a brief rinse in 50% (v/v) ethanol (30 s). They were
(w/v). (3) In-vitro assay of plant enzymes. Extracts were prepared as described for nodule tissue (sucrose omitted) and N R assayed as described by Aryan et al. (/983) except that the reaction was terminated and excess N A D H oxidized by the procedure of Scholl etal. (1974). Nitrite-reductase activity (reduced methyl viologen as electron donor) was determined by the method of Hucklesby et al. (1972).
Nitrogenase activity. Nitrogenase activity was estimated by the C2He-reduction technique (Hardy et al. 1968). Plants were carefully removed from the root medium and placed in 1-1 jars. These were sealed and 100 ml acetylene injected into each jar through a rubber septum. The amount of CzH2 reduced was measured at 30 and 60 min by injecting a 50-gl sample into a gas chromatograph (Model 9A; Shimadzu Corp., Kyoto, Japan) ; fitted with a column, 1 m long, 3 mm diameter, of Porapak N (80-100 mesh; Waters Associates, Milford, Mass., USA). The rate of C2H2 reduction, checked at 3-min intervals, was linear up to at least 60 min.
Sucrose-gradient fractionation of bacteroids. A 3-ml sample of washed bacteroids (approx. 40 mg protein from 2.5 g nodule FW) in 0.05 M K-phosphate, pH 7.5, was layered on a discontinuous sucrose gradient prepared in the above buffer and comprising the following sucrose steps (w/w): 9 ml, 45%; 9 ml, 50%; 9 ml, 52% and 7 ml, 57%. Centrifugation was for 4 h at 100000-g (1 ~ C) in a SW-28 rotor (Beckman Instruments,
C. Giannakis et al. : Nitrate metabolism of bacteroids of soybean nodules Palo Alto, Ca, USA). Fractions (1.5 ml) were collected by upward displacement using a density-gradient fractionator (ISCO, Lincoln, Neb., USA).
Nitrate, nitrite and protein measurements. Nitrate was measured by the Escherichia coli N R method (Wallace 1986) and nitrite by the method of Nicholas and Nason (1957). Protein was determined using the micro-biuret method (Itzhaki and Gill 1964), except in the analysis of the sucrose gradients where the Folin phenol reagent of Lowry et al. (1951) was employed after a preliminary precipitation of the protein with 10% (w/v) trichloroacetic acid. Bovine serum albumin was used as a reference protein. Serological studies. Antiserum against strain CB1809 was kindly provided by the Australian Inoculants Research and Control Service, Department of Agriculture, Gosford, N.S.W. Bacteroid samples were incubated at 37 ~ C for 1-2 h with an appropriate level of antiserum to promote agglutination. After centrifugation at 250 .g for 3 min the supernatant was tested for bacteroid N R activity. Controls were run with bacteroids of strain CC705 and with 0.85% NaC1 (w/v) instead of antiserum. Analys& of data. Data presented represent experiments that have been repeated at least twice. Where replicate plant samples were tested, a statistical analysis of the data is included.
Results
Activity of N R and NiR in free-living cells and bacteroids. Cells of strains CB1809 and CC705, grown under 2% Oz (v/v), had high activities of N R and NiR (Table 1). The N R activity was in excess of that of NiR, especially in strain CC705. Bacteroids isolated from soybean plants inoculated with strain CC705 had a similar N R activity to that of freeliving cells (Table 1). In contrast, bacteroids of strain CB1809 had very low N R activity, only 1% of that of the free-living cells. The bacteroids of neither strain had detectable NiR activity when assayed with succinate as the electron donor. A low level of enzymic nitrite utilization by bacteroids was observed with reduced methyl viologen as electron donor (approx. 0.9 nmol NO~ utilized. r a i n - t - ( r a g protein)-i). However, with this assay a high level of non-enzymic utilization of nitrite occurred with boiled bacteroid samples (approx. 1.6 nmol NO2 utilized, m i n - 1. (rag protein- 1)). Comparison of nitrate utilization by bacteroids of strains CC705 and CB1809. Washed bacteroids of strain CC705 reduced nitrate to nitrite within a short time (3 h) followed by a rapid decline in N R activity (Fig. 1 A). After a 16 h incubation there was an increase in NiR activity, resulting in the utilization of the nitrite and a recovery of N R activity. In-vitro activities of N R and NiR (Fig. 1 B) correlated well with the rates of nitrate and nitrite reduction in bacteroids (Fig. 1 A). Inclusion of ri-
53
Table 1. Comparison of N R and NiR activities in cultured cells and bacteroids of Bradyrhizobium japonicum. Cells were grown under 2% (v/v) 02 as described in Materials and methods. Bacteroids were isolated from 37-d-old soybean plants supplied with a nutrient medium lacking combined nitrogen. The N R and NiR activities were measured in washed cells and bacteroids as described in Mater&& and methods (in-vivo study) and expressed as follows: NR, nmol nitrite produced.min-1. (nag protein)-l; NiR, nmol nitrite utilized-rain-1-(rag protein)- 1 Strain
CB1809 CC705
Free-living cells
Bacteroids
NR
NiR
NR
NiR
113 53
61 7
1 42
0 0
fampicin in the incubation medium with nitrate prevented the development of NiR activity in bacteroids (Fig. 1 C) so that the nitrite produced was not re-utilized. A similar result was obtained with 50 gg. ml-1 chloramphenicol (data not shown). When CB1809 bacteroids were incubated with nitrate (Fig. 2A), N R and NiR activities increased after 14 h, a slower increase in NiR resulting in a temporary accumulation of nitrite. The same pattern including a lag in the development of N R and NiR activities occurred in bacteroids of CB1809 incubated with nitrite (Fig. 2B). Inclusion of rifampicin (10 gg.m1-1) prevented the increase in both enzyme activities. Bacteroids of CB1809, incubated without nitrate or nitrite showed only a small increase in N R activity (to about 7 nmol NO 2 produced, m i n - 1. (mg protein)- i). Since the bacteroid fi'action could contain nontransformed bradyrhizobia or other bacteria, it was necessary to establish that the induction of N R and NiR activity was occurring in the bacteroid component. Ching et al. (/977) demonstrated that bacteroids could be separated from bacteria by centrifugation on a discontinuous sucrose gradient. Using this procedure the bacteroid sample of strain CB1809 was resolved into two fractions separated in the 45% and 50% (w/w) sucrose steps (Fig. 3). No bacterial cells were detected. Cultured cells of strain CB1809 separated in the 52% and 57% sucrose (w/w) fractions, mainly in the latter (data not shown). Induction of N R activity was demonstrated in the two bacteroid fractions isolated on the sucrose gradient (Fig. 3, inset). To confirm that enzyme induction was not occurring in contaminating bacteria in the bacteroid preparation, the effect of antiserum specific for strain CB1809 was tested. When bacteroids of CB1809 (incubated for 24 h as described in Fig. 1)
54
C. Giannakis et al. : Nitrate metabolism of bacteroids of soybean nodules
A
50
50
5
40
4
5
~ 1 4 9 ~176 ~
40
I
4
~
30
/-
T 2
20
m o
&
E
Z
20
2O
30
2
I0
T
.....~"
.E N
\
......~......o. .......
o
I0
3
s
v
I0
o~ o
0
40
I
I0
20
3o
B
._~
E
50 ............=:..........~"---~-
5
40
4
30
3
20
2
.+2_ .>_
t_
~, 50 r._
o o o.
"..
z
Z
~ 4o
tY Z
s 30 o
10 9
.~_
I
t0
0
20
INCUBATION ~
20
I0
0
30
C .IP .............
9 ...........
9 .................
9 .................
9
40
5 4
'"
B
30
5 i
o
2 &
20
'E- 4
A
ql}
I0
I
I0
20 INCUBATION
~
Initial
1.0
1.0
After
22.3
19.5
z o
a
NR activity (nmol nitrite produced 9min-1. (mg protein)- 1) A B
incubation
r i
(h)
Fig. 2. A, B. Utilization of nitrate and nitrite by isolated bacteroids of B. japonieum strain CB1809. Washed bacteroids were incubated (A) as described in Fig. 1A or with 5 m M KNO2 in place of KNO3 (B). A - - A , in-vivo N R ; ~ - - z x , in-vivo NiR; 9 ...... e, nitrite
40
n,Z
50
PERIOD
30
2
iA
30 PERIOD
40 (h)
Fig. l A-C. Utilization of nitrate by isolated bacteroids of B. japonicum strain CC705. Washed bacteroids (1 mg p r o t e i n - m l ~) were incubated at 2 5 ~ under anaerobic conditions. The incubation medium (10 ml) contained 100 m M K-phosphate (pH 7.5), 20 m M Na-succinate, 10 m M glucose, 5 m M KNOa (A, B) plus 10 pg-ml -~ rifampicin (C). In-vivo N R and NiR activities (A, C) and in-vitro N R and N i R activities (B) were measured at the times indicated, as described in Materials and methods. A - - A , N R ; zx--zx, NiR. Nitrite content of the incubation medium was also determined, e - - - e
1 0
~ - I ~ 0
!~1-~o50~__~1 n4_, 52% _.1~~1.S7%.~, 5
10 15 Fraction No
20
Fig. 3. A bacteroid fraction from nodules of 32-d-old soybean plants inoculated with B.japonicum strain CB1809 was fractionated on a discontinuous sucrose gradient as described in the Materials and methods and with the sucrose steps (w/w) indicated. Nitrate reductase activities of A and B are shown in the inset, before and after incubation at 25 ~ C for 24 h as described in Fig. 1
C. Giannakis et al. : Nitrate metabolism of bacteroids of soybean nodules
55
Table 2. Effects of nitrate on growth and nitrogenase activity of soybean. Plants inoculated with B. japonicum strains CC705 or CB1809 were harvested at 36 d after treatment with nitrate for 4 d as indicated. The data represent the mean of at least four replicates with SE Nitrate
FW/plant (g)
(mM)
Shoot
Root
Nodule
Nitrogenase (gmol C2H2 reduced h - 1. (g nodule F W ) - 1)
CC705
0 10
10.9• 14.6•
8.8• 11.2•
2.0• 1.8•
8.8• 3.3!0.5
CB1809
0 10
13.3• 17.5•
11.3• 14.4•
1.4• 1.2•
9.9• 3.8•
Strain
were treated with the antiserum, agglutination occurred and 80% of the N R activity was sedimented by centrifugation at 250.g for 3 min. Nitrate-reductase activity in bacteroids of CC705 was not inhibited by this antiserum.
Comparison of N2 fixation in soybeans inoculated with CB1809 and CC705 and their sensitivity to inhibition by nitrate. Plants inoculated with strain CC705 exhibited a much more pronounced "nitrogen-hunger phase" between 15 and 25 d after planting and thus in experiments undertaken at approx. 30 d (Table 2) the plants inoculated with CC705 were consistently smaller than those inoculated with CB1809. Nodules of the CC705 plants evolved a 70-fold higher amount of H 2 than that of CB 1809 (data not shown). Nitrate treatment (10 mM) for 4 d (Table 2) resulted in a 63% inhibition of nitrogenase activity in plants nodulated with either strain. Occurrence of NR and NiR and accumulation of nitrate and nitrite in soybean nodules and roots. In plants treated with nitrate for 4 d (Table 3) there was a relatively small change in bacteroid N R activity (succinate-dependent) or nodule cytosol N R activity (NADH-dependent), whereas activity of the root enzyme was increased. Bacteroid N R activity of strain CC705 was considerably in excess of that in either the nodule cytosol or root. The nodule cytosol had a low N i R activity not influenced by nitrate supply to the roots, in contrast to the activity of the root enzyme which was increased about 50-fold. A greater accumulation of nitrate was apparent in the root than nodule cytosol (Table 3), and only a trace was detected in the bacteroid fraction. Nodules extracted in Tris buffer (pH 8.5) contained nitrite in the cytosol fraction (Table 4). Further production of nitrite occurred when the extracts were incubated at 2 5 ~ and even at 1~ C. When the nodules were extracted with ethanol no
3. Influence of nitrate on the N R and NiR activities and nitrate content of soybean root and nodule tissues in plants inoculated with B. japonicum strains CB1809 or CC705. Bacteroid and nodule cytosol data are expressed on the basis of nodule FW. Details of the plant material are given in Table 2
Table
CB1809 Nitrate 0
CC705 Nitrate 10 m M
0
10 m M
30.8 1.2 0
37.9 0.6 0.5
N R - gmol NO 2 produced, h - 1 (g F W ) - 1 Bacteroid Nodule cytosol Root NiR
-
0.6 0.7 0
1.8 0.7 0.9
Ixmol N O 2 utilised, h - 1. (g FW) - 1
Bacteroid Nodule cytosol Root
0 1.1 0.5
0 1.4 27.3
0 1.9 0.5
0 1.9 25.6
0.02 0.3 1.6
0.05 13.8 40.2
0.04 0.1 L1
0.06 12.1 38.7
NO 3 - gmol. (g F W ) - 1 Bacteroid Nodule cytosol Root
TaMe 4. Nitrite contents of nodule extracts prepared in various media. Nodules from plants inoculated with B. japonicum strain CC705 were extracted in a mortar and pestle using the medium indicated (3 vol: 1 g nodule FW). Aliquots of the extract were centrifuged immediately (1 rain) and after 30 min at 25 ~ C. Nitrite in the supernatant was determined as described in Materials and methods Extraction medium
0.2 M Tris/0.05 M K-phosphate, pH 8.5 Ethanol, 100% (v/v) I M zinc acetate 0.025 M K-phosphate, pH 10.0 (80 ~ C)
Nitrite (gmol-(g nodule F W ) - 1) 0 min
30 min
1.41 a
8.64
0 0.I0 0.36
0 0.08 8.17
After 60 min at 1~ C this value increased to 3.0 gmol.(g nodule FW) - 1
56
C. Giannakis et al. : Nitrate metabolism of bacteroids of soybean nodules
nitrite was detected or produced. Nitrite added to the nodule sample (0.25 gmol) was fully recovered in the ethanol extract. Only a trace o f nitrite was detected when I M zinc acetate was the extractant. Extraction with a buffer at p H 10, preheated to 80 ~ C, did not prevent nitrite production in vitro. Nitrite accumulation also occurred in extracts of CB1809 nodules, but at a much slower rate, corresponding to their low level of bacteroid N R (data not shown). Discussion
There are two main types of N R in bacteria associated with nitrate assimilation and dissimilation, respectively (see review by Beevers and Hageman 1983). The N R in anaerobically cultured bradyrhizobia has the same relative molecular mass (Mr) as that in bacteroids (69000; Daniel and Gray 1976) and is likely to be a dissimilatory enzyme. Bacteroid N R purified from tupin nodules had an Mr of 67000 and utilized reduced ferredoxin as an electron donor (Alikulov et al. 1981). In aerobically-cultured cells Daniel and Gray (1976) identified an N R species with a larger Mr (180000). However, Kennedy et al. (1975) found that N R in aerobically grown cells had an Mr of 70 000. In this study DIECA was used to inhibit N i R of cultured cells without affecting N R . Diethyldithiocarbamate is an irreversible inhibitor of copper-dependent enzymes (Mann 1955) and one of the two types of N i R identified in denitrifying bacteria is a copper-containing flavoprotein (Knowles 1982). Relatively little is known about the N i R enzymes in rhizobia, but it is likely that there are separate enzymes for the assimilation of nitrite to ammonium and its dissimilation to nitrogenous gases. In the nodule cytosol there is an N A D H - d e p e n dent N R (Streeter 1982; Stephens and Neyra 1973) and an N i R assayed with reduced methyl viologen (Hunter 1984). These have similar properties to the assimilatory enzymes of root tissue. Bacteroids of Bradyrhizobiumjaponicum strain CB1809 have a low activity of N R compared to strain CC705, even though both have high enzyme activity ex planta. Nitrate-reductase activity in cultured cells was increased at reduced 02 concentrations (Daniel and Appleby 1972) and thus the low 02 environment o f the infected region of the nodule would be suitable for the expression of the enzyme in bacteroids. Vairinhos et al. (1986) showed that strain U S D A 7 6 was similar to CC705 in expressing a high level of N R in the bacteroids, while RGS527 and CB1003, like CB1809, had relatively
low bacteroid N R activities. Bergersen (1974) also noted that bacteroids o f B. japonicum strains CC705 and CB1809 differed widely in their nitrate reductase activities. Bacteroids of Rhizobium trifolii, R. phaseoli and R. leguminosarum do not have N R activity (Manhart and Wong 1979) but the enzyme was detected in bacteroids of R. meliloti and was induced by nitrate application to the alfalfa host (Becana et al. 1985 a). When bacteroids of strain CB1809 were isolated from the nodules and incubated anaerobically with nitrate or nitrite, N R was induced to the same level as that in bacteroids of CC705. In both cases, complete inhibition by chloramphenicol and rifampicin indicated a de-novo synthesis of the two enzymes. Rigaud (1976) has also demonstrated an induction of N R and N i R in anaerobic preparations of Phaseolus vulgaris bacteroids incubated for at least 10 h. After fractionation of the bacteroids of strain CB1809 on a sucrose gradient it was established that N R was induced in two mature bacteroid fractions (Fig. 3). With reference to the earlier work of Ching et al. (1977), the two bacteroid fractions identified in the current study represent a mature bacteroid fi'action (Fraction B Fig. 3) and a larger bacteroid component, possibly a group of bacteroids within the peribacteroid membrane (Bergersen and Appleby 1981). The latter component was not detected by Ching et al. but could have been masked in the large protein peak associated with the nodule cytosol, which they observed in this region of the sucrose gradient. In our study, the bacteroids were separated from the nodule cytosol by a preliminary centrifugation. Unlike Ching et al. we did not detect either transforming bacteria or bacteria in our bacteroid preparations. Our data thus demonstrate that mature bacteroids retain the ability to undertake de-novo synthesis of N R when supplied with nitrate. The observation that N R activity in bacteroids of CB1809 could be induced after isolation and incubation with nitrate and the lack of a similar response in situ, with nitrate supplied to the root medium, indicates that nitrate was not reaching the infected region of the nodule. From studies with the in-vivo assay of N R in bacteroids (with and without nitrate) Hunter (1983) also concluded that nitrate had limited access to the bacteroid zone of the nodule. Indeed, Pate and Atkins (1981) had earlier proposed that neither nitrate nor its reduction products reach the bacteroids. However, other workers have suggested that utilization of nitrate by bacteroids makes a contribution to its assimilation in the plant (Randall et al. 1978) or else results in nitrite accumulation (Stephens and
C. Giannakis et al. : Nitrate metabolism of bacteroids of soybean nodules
Neyra 1983; Becana et al. 1985b; Streeter 1985) which could inhibit nitrogenase (Kennedy et al. 1975; Rigaud and Puppo 1977). Sprent etal. (1987) have recently demonstrated that in both determinate and indeterminate nodules, nitrate has restricted access to the inner infected zone. It appears that nitrite accumulation in nodule tissue only occurs after maceration of the nodules, thus allowing the bacteroids access to the nitrate which is normally restricted to the outer cortex region. No nitrite was detected in ethanol extracts, while I M zinc acetate (as used by Manhart and Wong 1980) was also fairly effective in preventing nitrite accumulation. In our studies, extraction of nodules with a buffer at pH 10 preheated to 80 ~ C (the procedure of Stephens and Neyra 1983) did not prevent nitrite accumulation in vitro. We conclude that nitrite does not accumulate in the nodule and correlations between N R activity of bacteroids and nitrite accumulation in the nodule are likely to be an artefact of their isolation. Stephens and Neyra (1983) observed that nitrate inhibition of nitrogenase in isolated bacteroids was relieved in a NR-minus mutant strain of B. japonicum, but in detached nodules the effect of the mutant was less conclusive, Since nitrate available to the nodule appears not to gain access to the bacteroids (Sprent et al. 1987) the level of bacteroid N R will not affect the sensitivity of nodule nitrogenase to nitrate. Soybean plants nodulated with CC705 (high bacteroid NR) and CB1809 (low bacteroid NR) were equally sensitive to nitrate inhibition of nitrogenase (Table 2). Thus, as already concluded from studies with NR-deficient mutant strains of rhizobia (Gibson and Pagan 1977; Manhart and Wong 1980; Streeter 1982) there is no correlation between bacteroid N R activity and sensitivity of the N zfixation process to inhibition by nitrate. Bacteroids of CC705 and CB1809 when incubated with succinate exhibited no detectable NiR activity, in agreement with earlier studies on soybean by Daniel and Appleby (1972) and Stephens and Neyra (1983). Using reduced methyl viologen as an electron donor we detected a low level of NiR activity in bacteroids as observed by Chen and Sung (1983) and Streeter (1985). The activity measured (0.9 nmol NO~ utilized-min- 1 .(rag protein)- 1) was considerably lower than that induced in isolated bacteroids incubated with nitrate or nitrite and assayed with succinate as an electron donor (approx. 25 nmol NO~- utilized, m i n - 1. (mg protein)-1; Figs. 1, 2). When reduced methyl viologen was used to assay NiR activity in intact bacteroids, we observed that there was a relatively
57
high level of non-enzymic utilization of nitrite. Thus, although a boiled bacteroid sample was used as control, the estimate of enzymic activity may be inaccurate. The nodule cytosol had NiR activity, which was much lower than bacteroid N R and root NiR activities (Table 3, data for strain CC705). Unlike the root enzyme, nodule cytosol NiR did not increase when the plants were supplied with nitrate. Conclusion
Utilization of nitrate by bacteroids would result in either nitrite accumulation (not detected) or its further reduction by NiR (no induction of NiR activity observed in situ). Nitrite accmnulation would also repress the synthesis of bacteroid NR. We conclude that nitrate metabolism does not normally occur in the bacteroid zone of the soybean nodule, even when a dissimilatory N R is expressed, because of a restricted access of nitrate. Thus the inhibitory effect of nitrate on nitrogen fixation cannot be explained by nitrite accumulation in the nodule and its inhibition of nitrogenase. We thank R.G. Batt for technical assistance and C.G. acknowledges the receipt of a postgraduate scholarship from the University of Adelaide.
References Alikulov, Z.A., Burikhanov, Sh.S., Sergeev, N.S., L'vov, N.P., Kretovich, V.L. (1980) Nitrate reductase of lupine root nodule bacteroid. Prikl. Biokhim. Mikrobiol. (Transln.) I6, 372-376 Aryan, A.P., Batt, R.G., Wallace, W. (1983) Reversible inactivation of nitrate reductase by NADH and the occurrence of partially inactive enzyme in the wheat leaf. Plant Physiol. 71,582-587 Becana, M., Aparicio-Tejo, P.M. Sanchez-Diaz, M. (1985 a) Nitrate and nitrite reduction by alfalfa root nodules: Accumulation of nitrite in Rhizobium meliloti bacteroids and senescence of nodules. Physiol. Plant. 64, 353-358 Becana, M., Aparicio-Tejo, P.M., Sanchez-Diaz, M. (1985b) Nitrate and nitrite reduction in the plant fraction of alfalfa root nodules. Physiol. Plant. 65, 185-188 Beevers, L., Hageman, R.H. (1983) Uptake and reduction of nitrate: Bacteria and higher plants. In: Encyclopedia of plant physiology, N.S. vol. 15A: Inorganic plant nutrition, pp. 351-375, Lfiuchli, A., Bieleski, R.L., eds. Springer-Verlag, Berlin etc. Bergersen, F.J. (1974) Formation and function of bacteroids. In: The biology of nirogen fixation, pp. 473-497, Quispel, A., ed. North Holland. Publishing Co., Amsterdam Bergerson, F.J., Appleby, C.A. (1981) Leghaemoglobin within bacteroid-enclosingmembrane envelopes from soybean root nodules. Planta 152, 534~543 Bhandari, B., Nicholas, D.J.D. (1984) Denitrification of nitrate to nitrogen gas by washed cells of Rhizobium japonicum and by bacteroids from Glycine max. Planta 161, 8/-85
58
C. Giannakis et al. : Nitrate metabolism of bacteroids of soybean nodules
Brown, C.M., Dilworth, M.J. (1975) Ammonia assimilation by Rhizobium cultures and bacteroids. J. Gen. Microbiol. 86, 39-48 Chen, C-L., Sung, J-M. (1983) Effect of water stress on the reduction of nitrate and nitrite by soybean nodules. Plant Physiol. 73, 1065 1066 Ching, T.M., Hedtke, S., Newcomb, W. (1977) Isolation of bacteria, transforming bacteria and bacteroids from soybean nodules. Plant Physiol. 60, 771-774 Daniel, R.M., Appleby, C.A. (1972) Anaerobic-nitrate, symbiotic and aerobic growth of Rhizobium japonicum : effects on cytochrome P4so, other haemoproteins, nitrate and nitrite reductase. Biochim. Biophys. Acta 275, 347-354 Daniel, R.M., Gray, J. (1976) Nitrate reductase from anaerobically grown Rhizobiurn japonicum. J. Gen. Microbiol. 96, 247-251 Daniel, R.M., Limmer, A.W., Steele, K.W., Smith, I.M. (1982) Anaerobic growth, nitrate reduction and denitrification in 46 Rhizobium strains. J. Gen. Microbiol. 128, 1811-1815 Evans, H.J. (1954) Diphosphopyridine nucleotide-nitratereductase from soybean nodules. Plant Physiol. 29, 298-301 Gibson, A.H., Pagan, J.D. (1977) Nitrate effects on the nodulation of legumes inoculated with nitrate reductase-deficient mutants of Rhizobium. Planta 134, 17 22 Hardy, R.W.F., Holsten, R.D., Jackson, E.K., Burns, R.C. (1968) The acetylene-ethylene assay for N2-fixation : laboratory and field evaluation. Plant Physiol. 43, 1185-1207 Hucklesby, D.P., Dalling, M.J., Hageman, R.H. (1972) Some properties of two forms of nitrite reductase from corn (Zea mays L) scutellum. Planta 104, 220233 Hunter, W.J. (1983) Soybean root and nodule nitrate reductase. Physiol. Plant. 59, 471-475 Hunter, W.J. (1984) Purification and characterisation of soybean nodule nitrite reductase. Physiol. Plant. 60, 467-472 Itzhaki, R.F., Gill, D.M. (1964) A microbiuret method for estimating proteins. Anal. Biochem. 9, 401.410 Kennedy, I.R., Rigaud, J., Trinchant, J.C. (1975) Nitrate reductase from bacteroids of Rhizobium japonicum: enzyme characteristics and possible interaction with nitrogen fixation. Biochim. Biophys. Acta 397, 24-35 Knowles, R. (1982) Denitrification. Microbiol. Rev. 46, 43-70 Lowe, R.H., Evans, H.J. (1964) Preparation and some properties of a soluble nitrate reductase from Rhizobiumjaponicum. Biochim. Biophys. Acta 85, 377-389 Lowry, O.H., Rosebrough, N.J., Farr, A.L., Randall R.J. (1951) Protein measurement with the Folin phenol reagent. J. Biol. Chem. 193, 265-275 Manhart, J.R., Wong, P.R. (1979) Nitrate reductase activities of rhizobia and the correlation between nitrate reduction and nitrogen fixation. Can. J. Microbiol. 25, 1169-1174 Manhart, J.R., Wong, P.P. (1980) Nitrate effect on nitrogen fixation (acetylene reduction). Activities oflegmne root nodules induced by rhizobia with varied nitrate reductase activities. Plant Physiol. 65, 502-505 Mann, P.J.G. (1955) Purification and properties of the amine oxidase of pea seedlings. Biochem. J. 59, 609-620 Nicholas, D.J.D., Nason, A. (1957) Determination of nitrate and nitrite. Methods Enzymol. 3, 981-984
O'Hara, G.W., Daniel, R.M., Steele, K.W. (1983) Effect of oxygen on the synthesis, activity and breakdown of the Rhizobium denitrification system. J. Gem Microbiol. 129, 2405-2412 Pate, J.S., Atkins, C.A. (1981) Nitrogen uptake, transport and utilization. In: Ecology of nitrogen fixation, vol. 3, pp. 245298, Broughton, W.J., ed. Oxford University Press, Oxford, UK Randall, D.D., Russell, W.J., Johnson, D.R. (1978) Nodule nitrate reductase as a source of reduced nitrogen in soybean, Glycine max. Physiol. Plant. 44, 325-328 Rigaud, J. (1976) Effet des nitrates sur la fixation d'azote par les nodules de Haricot (Phaseolus vulgaris L.). Physiol. V6g. 14, 297-308 Rigaud, J., Puppo, A. (1977) Effect of nitrite upon leghemoglobin and interaction with nitrogen fixation. Biochim..Biophys. Acta 497, 702-706 Scholl, R.L., Harper, J.E., Hageman, R.H. (1974) Improvements of the nitrite color development in assays of nitrate reductase by phenazine methosulfate and zinc acetate. Plant Physiol. 53, 825-828 Smith, G.B., Smith, M.B. (1986) Symbiotic and free-living denitrification by Bradyrhizobium japonieum. J. Soil Sci. Soc. Am. 50, 349-353 Sprent, J.I., Giannakis, C., Wallace, W. (1987) Transport of nitrate and calcium into legume root nodules. J. Exp. Bot. 38, 1121-1128 Stephens, B.D., Neyra, C.A. (1983) Nitrate and nitrite reduction in relation to nitrogenase activity in soybean nodules and Rhizobiurn japonicum bacteroids. Plant Physiol. 71, 731 735 Streeter, J.G. (1982) Synthesis and accumulation of nitrite on soybean nodules supplied with nitrate. Plant Physiol. 69, 1429-1434 Streeter, J.G. (1985) Nitrate inhibition of legume nodule growth and activity. I. Long term studies with a continuous supply of nitrate. Plant Physiol. 77, 321-324 Vairinhos, F., Wallace, W., Nicholas, D.J.D. (1986) Denitrification versus assimilation of nitrate by Bradyrhizobiumjaponicurn. In: Proc. 8th Aust. Nitrogen Fixation Conf., pp. 137138, Wallace, W., Smith, S.E., eds. Publ. No. 25, Australian Institute of Agricultural Science, Melbourne Van Berkum, P., Keyser, H.H. (1985) Anaerobic growth and denitrification among different serogroups of soybean rhizobia. Appl. Environ. Microbiol. 49, 772-777 Wallace, W. (1986) Distribution of nitrate assimilation between the root and shoot of legumes and a comparison with wheat. Physiol. Plant. 66, 630-636 Wilttenberg, J.B. (1980) Utilization of leghemoglobin-bound oxygen by Rhizobium bacteroids. In: Nitrogen fixation, vol. II, pp. 53-67, Newton, W.E., Orme-Johnson, W.H., eds. University Park Press, Baltimore, Ma., USA Zablotowicz, R.M., Focht, D.D. (1979) Denitrification and anaerobic, nitrate-dependent acetylene reduction in cowpea Rhizobium. J. Gen. Microbiol. 111,445-448 Received 6 July; accepted 28 September 1987
