Molybdate Transport by Bradyrhizobium japonicum Bacteroids

JOURNAL

OF

BACTERIOLOGY, Dec. 1988, p. 5613-5619

Vol. 170, No. 12

0021-9193/88/125613-07$02.00/0 Copyright © 1988, American Society for Microbiology

Molybdate Transport by Bradyrhizobium japonicum Bacteroids ROBERT J. MAIER* AND LENNOX GRAHAM Department of Biology, The Johns Hopkins University, Baltimore, Maryland 21218 Received 27 July 1988/Accepted 7 September 1988

Bacteroid suspensions of Bradyrhizobiumjaponicum USDA 136 isolated from soybeans grown in Mo-deficient conditions were able to transport molybdate at a nearly constant rate for up to 1 min. The apparent Km for molybdate was 0.1 ,uM, and the Vmax was about 5 pmol/min per mg (dry weight) of bacteroid. Supplementation of bacteroid suspensions with oxidizable carbon sources did not markedly increase molybdate uptake rates. Anaerobically isolated bacteroids accumulated twice as much Mo in 1 h as aerobically isolated cells did, but the first 5 min of molybdate uptake was not dependent on the isolation condition with respect to 02. Respiratory inhibitors such as cyanide, azide, and hydroxylamine did not appreciably affect molybdate uptake, even at concentrations that inhibited 02 uptake. The uncouplers carbonyl cyanide m-chlorophenylhydrazone (CCCP) and carbonyl cyanide p-trifluoromethoxyphenylhydrazone (FCCP) and the ionophores nigericin and monensin significantly inhibited molybdate uptake. The electrogenic ionophores valinomycin and gramicidin stimulated molybdate uptake. Rapid pH shift experiments indicated that molybdate transport depends on a transmembrane proton gradient (ApH), and it is probably transported electroneutrally as H2MoO4. Most of the "MoO42- taken up was not exchangeable with a 100-fold excess of unlabeled MoO42-. Tungstate was a competitive inhibitor of molybdate uptake, with a Ki of 0.034 ,uM, and vanadate inhibited molybdate uptake slightly.

The conventional nitrogenase from N2-fixing bacteria, including the bacterial symbiont of legumes, contains molybdenum. Optimal nitrogenase specific activities are observed in preparations that contain two Mo atoms per molecule (7). It is thought that the lack of Mo in natural environments can sometimes limit biological N2 fixation, since Mo supplementation can increase the rate of symbiotic N2 fixation (5, 16, 20). The average abundance of Mo in the Earth's crust is only about 15 pmol/kg (30); thus, it is unlikely that organisms will commonly encounter high levels of Mo. Also, much of the Mo in natural environments is insoluble; therefore, specific high-affinity systems for Mo metabolism would be helpful. Thus, it would seem likely that N2-fixing bacteria express efficient mechanisms for scavenging trace levels of Mo. Indeed, high-affinity molybdatesequestering processes have been described for N2-fixing Klebsiella pneumoniae (17, 18, 36), Clostridium pasteurianum (8, 9), and free-living (25) and bacteroid (12) forms of Bradyrhizobium japonicum. The studies have been done mostly on whole cells, although extracellular factors implicated in Mo binding from Azotobacter vinelandii (28) and B. japonicum (25) have been described. It is likely that Mo metabolism in the N2-fixing bacteria involves many steps, including binding of Mo to the cell surface, sequestering and internalization from the free molybdate pool, and synthesis of intermediates such as the iron-molybdenum cofactor. The identification of intracellular Mo-binding intermediates prior to Mo bound in nitrogenase in several N2-fixing bacteria (15, 29, 36), as well as the assignment of at least five genes to intracellular Mo metabolism in K. pneumoniae (17, 18, 32, 36, 38), justifies this conclusion. In addition, Mo may play regulatory roles, such as the Mo-mediated transcriptional regulation of nitrogenase synthesis in A. vinelandii (19). The study of mutants with mutations in Mo metabolism (17, 18, 38), including some in the early steps of molybdate binding (18, 25), should help in *

separating and defining the various metabolic reactions of Mo. We previously screened 20 strains of B. japonicum bacteroids for molybdate uptake ability (12). The variability among the strains in their ability to transport molybdate was attributed at least in part to the differences in affinity of their whole cells for molybdate. Experiments in which the nodulated soybean plants were deprived of Mo indicated that the molybdate transport ability of the B. japonicum strain could ultimately influence the symbiotic N2 fixation ability (12). In the current study, the molybdate uptake process by the B. japonicum bacteroid strain best able to bind molybdate was characterized. The results indicate that molybdate is transported by a high-affinity system and that it is dependent on a transmembrane proton gradient. MATERIALS AND METHODS

Chemicals and media. Strain USDA 136 was obtained from Peter VanBerkum, U.S. Department of Agriculture, Beltsville, Md. Ultrapure chemicals were Puratronic grade 1 obtained from Johnson Matthey Chemicals, Ltd., Hertfordshire, England. These included all trace elements and MgSO4. Ultrex NaCl was from J. T. Baker Chemical Co., Phillipsburg, N.J. For Mo-free medium, distilled water was further treated to remove traces of Mo by filtration through 8-hydroxyquinoline immobilized on glass (10). Controlledpore glass 8-hydroxyquinoline was from Pierce Chemical Co., Rockford, Ill. The efficacy of this procedure for removal of Mo was demonstrated previously (12, 25). Stock cultures were inoculated by using sterile plastic loops, and medium was kept in polycarbonate flasks that were never contaminated with Mo-containing medium. Carrier-free Na299MoO4, at a concentration of approximately 100 mCi/ ml, was obtained from Cintichem, Tuxedo, N.Y. One millicurie of carrier-free Na299MoO4 is equivalent to 0.021 nmol of Mo. Growth of soybeans. Maryland certified soybean seeds (cultivar Essex) were surface sterilized and germinated in

Corresponding author. 5613

5614

MAIER AND GRAHAM

the dark for 48 h as described previously (12). The soybeans were planted (five seeds per pot) in sterile plastic 9-in (22-cm) pots containing vermiculite as described previously

(12). The vermiculite had been washed with Mo-free water, and 0.1 ml of B. japonicum strain USDA 136 was inoculated onto each seed (12). The plants were grown in a greenhouse with supplemental lighting as described previously (26). Each pot received nutrient solution (39) without Mo under the regimen described previously (12). All of the procedures for growing the inoculant strain and the soybean plants under Mo-deficient conditions were previously described in detail (12). Anaerobic versus aerobic harvesting of bacteroids. For comparison of the aerobic versus the anaerobic isolation of bacteroids (see Fig. 3), the bacteroids were harvested from 2 g of nodules by the micromethod described by Maier and Brill (24), with the slight modifications previously described (12). The 10 ml of 0.05 M potassium phosphate buffer used to macerate the nodules for the anaerobically isolated cells contained 5 mg of sodium dithionite and 0.2 M sodium ascorbate, and the cells were harvested (by centrifugation or suspension) in sealed test tubes in a 100% N2 atmosphere. The buffer for the aerobically isolated cells was identical, except that it lacked ascorbate and dithionite. Both types of cells were finally suspended in 0.025 M 2-(N-morpholino)ethanesulfonic acid (MES) (pH 6.5). These suspensions (6 ml of about 3 mg [dry weight] per ml) received 0.5 FLM 99Mo and were incubated in an argon-flushed 70-ml tightly stoppered vial. The vial was placed on a rotary shaker (28°C, shaking at 50 cycles per min), and 02 was injected to make an atmosphere of 1% 02 (balance argon). Triplicate samples of 0.2 ml were removed through the serum stopper by use of a syringe, and 99Mo accumulation was determined. At 20 and 40 min, a 0.5-ml gas sample was removed from each vial to assay for 02 concentration (26), and then 02 was injected to maintain levels at about 1.0% partial pressure. Molybdenum transport assays. Bacteroids were harvested under aerobic conditions, unless otherwise specified, from soybean nodules from 5- to 6-week-old soybean plants. Nodules were picked from the plants, and the bacteroids were harvested the same day. One gram of nodules was crushed (12) in 5 ml of 0.05 M MES buffer (pH 6.5). Bacteroids were suspended in Mo-free 0.05 M MES buffer (pH 6.5) to a concentration of 2.4 to 7.5 mg (dry weight) of bacteroids per ml. Samples of this suspension were transferred to a plastic tube, and Na299MoO4 was added to initiate the assay. Unless otherwise specified, the Mo concentration for transport assays was 0.5 puM and 1% of the Mo (5 nM) was radioactive. Five replicate 0.2- or 0.3-ml samples from each treatment were rapidly filtered (12) for each time point or inhibitor test. After the samples had been rapidly filtered, the cells on the filter were washed three times with 5 ml of ice-cold 0.05 M potassium phosphate buffer (pH 7.0) containing 0.5 mM sodium molybdate. The filters were Durapore no. HVLP 02500 (pore size, 0.45 ,um; Millipore Corp.) and were saturated, prior to use, with the buffer containing 1.0 mM Na2MoO4 (12). The filtering unit was a 12-sample filtering manifold (Millipore Corp.). Background counts correcting for radioactive Mo retention by filters without cells were subtracted from all the sample values, and these values were less than 5% of the total sample counts. For inhibitor studies, values from control samples with the inhibitor and 99Mo but without cells were subtracted from the data obtained with the cells included. After the filters had been dried overnight in the 6-ml scintillation vial, 5 ml of Aquasol 2 (Du Pont, NEN Research Products) was added, and each vial

J. BACTERIOL.

was capped and vortexed vigorously. After 30 min at room temperature, vials were counted in an LKB model 1218 scintillation spectrometer on the 32p channel. Data were adjusted to take into account the short (66-h) half-life of 99molybdate. Mo uptake was calculated for each sample, and bacteroid dry weight was determined by drying 0.5-ml samples of the suspensions at 65°C. The y axes in the figures show the amount of Mo associated with the 0.2- or 0.3-ml

samples of filtered cells. Each datum point is the average for the five filtered samples. The pH optimum for 99MoO42- uptake was determined by suspending bacteroids in buffer composed of 20 mM each MES, MOPS [3-(N-morpholine)propanesulfonic acid], and glycylglycine at pH 4 to 9 (in increments of 0.5 units) for 20 min. After the addition of Na299MoO4 to a concentration of 0.25 ,uM, five 0.3-ml samples were filtered for each datum point. For the kinetic experiments in which the substrate concentration (molybdate) was varied (see Fig. 2), the bacteroids were suspended in MES buffer (pH 6.5), and sodium molybdate (radioactive plus carrier) was added to initiate the assay. The eight lower Mo concentrations from 0.005 to 0.06 ,uM contained 1% of radioactive molybdate, whereas the higher four concentrations (0.10, 0.125, 0.15, and 0.20 ,uM) contained 0.1% of radioactive molybdate. Samples were removed 1 min after the addition of molybdate and treated as described above. Other experimental details are given in the tables and figure legends. RESULTS Previous results have shown that strain USDA 136 bacteroids from soybeans grown under Mo-deficient conditions took up molybdate at a rate of about 3 pmol/min per mg (dry weight) of bacteroid (12). In that study molybdate uptake in strain USDA 136 was assayed only to provide an initial comparison with other B. japonicum strains, and the major characteristics of Mo uptake were not reported. In the present study, Mo uptake is further characterized for bacteroids of strain USDA 136 that were obtained from soybean plants grown under Mo-deficient conditions. Strain USDA 136 had the greatest rate of molybdate uptake among 20 strains (12). Measurement of molybdate uptake with time by a bacteroid suspension of USDA 136 is shown in Fig. 1. Mo uptake by the bacteroids was nearly constant for about 1 min and then began to level off as Mo was (presumably) internalized. These types of results are typical of many reported bacterial transport systems for various solutes, including metal ions (13, 33). The concentration dependence of molybdate uptake by bacteroids was studied by varying the substrate concentration and measuring the velocity of molybdate uptake in 60 s (Fig. 2). Molybdate uptake was a saturable process; it showed Michaelis-Menten kinetics, and from these data an apparent Km for Mo was calculated to be about 0.1 pzM. The Vmax was about 5 pmol/min per mg (dry weight) of bacteroid. Previously (12), we reported a Km for molybdate for strain USDA 136 bacteroids of 0.045 ,uM, and this was the lowest Km value we obtained among the B. japonicum strains screened. From five recent experiments we have obtained apparent Km values ranging from 0.05 to 0.12 ,uM for strain USDA 136. We do not know whether the apparent Km value would be higher if the soybeans had been supplemented with Mo. Effects of carbon substrates on Mo uptake. We attempted to define the conditions for maximal molybdate uptake by bacteroids by incubating the cells in various oxidizable

MOLYBDATE TRANSPORT BY B. JAPONICUM BACTEROIDS

VOL. 170, 1988

5615

TABLE 1. Effect of carbon substrates on bacteroid respiratory activity and molybdate transport Mo uptake (%M 02 uptake (%)b Carbon source" 1 l4

None Succinate Malate Fumarate Lactate a-Ketoglutarate Gluconate Citrate Arabinose Acetate

E 12

10_ 0

8

0

E

42

1

2

3

4

5

Time (min) FIG. 1. Mo accumulation by bacteroids. A 10-ml bacteroid suspension of 7.2 mg (dry weight) of bacteroids per ml in 0.05 M MES (pH 6.5) received 0.5 ,uM Na2MoO4 containing 2 mCi (0.042 nmol) of radioactive '9M\o. Samples (0.3 ml) were filtered at the times indicated, and the amount of Mo accumulated was determined as described in the text.

carbon substrates. Supplementation of bacteroid suspensions, with organic acids in particular, has been shown previously to stimulate respiration (35) as well as nitrogenase activity (1, 31). The addition of various carbon sources, including organic acids, significantly increased respiratory activity (Table 1) but did not greatly affect molybdate uptake. It is well known that bacteroids contain adequate internal energy reserves to respire even without the addition of a carbon source to the suspension. All subsequent experiments were performed without the addition of carbon sources.

Effects of aerobic versus anaerobic bacteroid isolation. The Mo-containing nitrogenase enzyme in N2-fixing root nodule bacteria is 02 labile (11), and this enzyme is presumably a major sink for the transported Mo. It was possible that the ability to transport and incorporate Mo by bacteroids would depend on the exposure of the cells to 02 during the harvest of the bacteroids from nodules. We thus prepared bacteroids

100 317 271 271 243 202 152 179 147 135

100 150 133 152 163 105 105 165 150 133

± 28

± 34 ± 19

± 41 ± 44 ± 37 ± 25 ± 39 ± 13 ± 14

a Bacteroids were suspended with the indicated carbon substrate (stock suspensions of acids were first adjusted to pH 6.5 with NaOH) for 10 min prior to the start of the 02 uptake or Mo uptake assays. b The endogenous 02 uptake rate (no carbon source) was 0.42 ,umol of 02 per h per mg (dry weight) of bacteroid. The 02 data are the average of two independent determinations of a bacteroid suspension (0.33 mg [dry weight] of bacteroid per ml) in 25 mM MES (pH 6.5). 02 uptake was determined polarographically, as described previously (26), in the presence of 165 to 200

ILM 02-

' For Mo uptake assays, 0.8 mg (dry weight) of bacteroids was filtered, and the control (without carbon substrates) gave 2.2 ± 0.6 pmol of Mo uptake per min. The results are the mean ± standard deviation for five replicate filtered samples.

anaerobically and aerobically and then compared the ability of the two preparations to transport Mo (Fig. 3). The isolation conditions with respect to 02 did not affect the molybdate uptake ability in the first 5 min of molybdate transport. However, anaerobically isolated bacteroids steadily accumulated more Mo (Fig. 3), so that at 1 h they contained twice as much Mo as the aerobically isolated cells. It is possible that the anaerobically isolated cells contain intact but Mo-free or Mo-deficient (apo)nitrogenase as a major sink for Mo, whereas the aerobically isolated bacteroids contain 02-inactivated nitrogenase. Imperial et al. (17) were able to reactivate Mo-free nitrogenase from whole cells of K. pneumoniae by adding Mo to previously Mo-starved cells. Presumably, this reactivation ability was due to a pool of Mo-free nitrogenase that had accumulated under Mo starvation conditions. However, our previous attempts to reactivate nitrogenase by adding molybdate to bacteroids of

6

5 0

,

, 4

Q

8

0

0

< 6.

M 3

0

,, 4-

E 2

Q

0

E

0

0.2~

0.05

0.10

0.15

0.20

S (MM Mo)

FIG. 2. Velocity of Mo uptake versus Mo concentration. Bacteroid suspensions (5.5 mg [dry weight]/ml) in 25 mM MES (pH 6.5) received various 9Mo concentrations (see text), and the Mo uptake in 1 min was determined.

Time (min) FIG. 3. Comparison of aerobic (A) versus anaerobic (0) bacteroid isolation on Mo uptake activity in 60 min. Bacteroids were harvested as described in the text. Error bars represent standard deviations.

5616

MAIER AND GRAHAM

J. BACTERIOL.

TABLE 2. Effect of respiratory inhibitors and ionophores on 0, and molybdate uptake Inhibitor None KCN

Azide Hydroxylamine

Dinitrophenol CCCP

FCCP Valinomycin

Concn

02 uptake"

Mo uptake"

0.35 0.25 0.17 0.26 0.18 0.32 0.16 0.03 0.35 0.18 0.36 0.21 0.21 0.95

2.7 ± 0.6 2.9 ± 0.5 2.7 ± 0.3 2.6 ± 0.6 2.4 ± 0.6 2.8 ± 0.8 2.5 ± 0.7 2.2 ± 0.6 2.9 ± 0.5 2.2 ± 0.6 2.0 ± 0.4 1.2 ± 0.3 1.0 ± 0.2 8.8 ± 1.4

2 ,uM (+ 50 mM

0.32

1.8 ± 0.3

KCI) 20 ,uM (+ 50 mM KCI)

0.80

1.0 ± 0.2

5 ,uM (+ 10 mM

0.36

1.9 ± 0.5

0.55

1.9 ± 0.5

assay.

0.56

1.1 ± 0.3

0.53 0.33 0.28

4.8 ± 0.9 2.9 ± 0.5 2.7 ± 0.4

of the membrane potential, but it does not dissipate a ApH (14, 33). The electrogenic uniporter ionophores valinomycin and gramicidin stimulated 02 uptake (Table 2). This is an expected result for effective membrane potential dissipation. Gramicidin and valinomycin both stimulated molybdate uptake, with valinomycin stimulating it almost threefold. This result suggested that ApH, and not membrane potential, may be important in driving molybdate uptake. The stimulation of molybdate uptake by valinomycin plus K+ could be due to an increase in pH gradient across the membrane; valinomycin-mediated dissipation of the membrane potential can result in a simultaneous increase in the pH gradient (13, 14), and stimulation of anion uptake (particularly phosphate, but also others) by valinomycin plus K+ has been described for mitochondrial suspensions (14). pH shift experiments. The inhibitor studies suggested that Mo042- uptake depended on a transmembrane proton gradient (ApH). More direct evidence would be to show that Mo042- uptake depended on the pH of the medium. In fact, in determining the pH optimum for molybdate uptake, we found a general response of increased molybdate uptake with a decrease in pH, even though within this pH-versusactivity slope a distinct peak was evident at pH 6.5 (12). Therefore, rapid pH shift experiments were performed and molybdate uptake was monitored at 10, 20, and 30 s (Fig. 4). The uptake was strongly dependent on ApH, with a ApH of 3.5 causing a 10-fold increase in Mo042- uptake in 10 s compared with no pH change. Therefore, it appears that Mo042- uptake depends on a transmembrane proton gradient (ApH, inside alkaline) and is therefore probably transported electroneutrally as H2MoO4. Exchangeability of the molybdate. We studied the ability of externally added molybdate to exchange with the radioactive molybdate that had been taken up in the first 1 and 5 min. Most of the molybdate taken up in the first 1 or 5 min was not exchangeable with a 100-fold excess of unlabeled Moo42(Fig. 5); in both cases (at 1 and 5 min) the amount of radioactivity associated with the cells decreased only about 20 to 25% in the first 1 min after exposure to the excess nonradioactive molybdate. Most of the Mo that was exchangeable was released from the cells in the first 1 min (Fig.

0.05 mM 0.5 mM 1 mM 10 mM 0.1 mM 1 mM 5 mM 1 mM 5 mM 2 ,uM 20 p.M 20 ,uM 20 ,uM (+ 50 mM

KCI) Nigericin

Monensin

Gramicidin D Indoleacetic acid

NaCl) 20 ,uM (+ 10 mM NaCI) 50 ,uM (+ 10 mM NaCI) 20 ,uM 1 mM

10 mM a 02 uptake is expressed as micromoles of 0, consumed per hour per milligram (dry weight) of bacteroid, and Mo uptake is expressed as picomoles of Mo per minute per milligram (dry weight) of bacteroid. Assay conditions were as described in the footnote to Table 1 and the text. 02 and Mo uptake was not significantly affected by 50 mM KCI or 10 mM NaCI alone.

strain USDA 136 were not successful (12). To prevent oxidation and inactivation of bacteroid enzymes by phenolic compounds, soybean root nodule bacteroids are usually harvested from nodules in the presence of polyvinylpolypyrrolidone plus ascorbate (11). However, the rate of molybdate uptake (at least during the first 5 min) by bacteroids was not different whether the cells were isolated with or without polyvinylpolypyrrolidone and ascorbate in the suspension for crushing nodules (data not shown). Subsequent experiments used bacteroids harvested aerobically, and no polyvinylpolypyrrolidone or reductant was used. Inhibitor experiments. Respiratory inhibitors such as cyanide, azide, and hydroxylamine did not appreciably affect molybdate uptake, even at concentrations that inhibited 02 uptake (Table 2). Carbonyl cyanide m-chlorophenylhydrazone (CCCP) was previously shown to greatly inhibit the formation of a pH gradient by B. japonicum bacteroid suspensions (2). The uncouplers CCCP and carbonyl cyanide p-trifluoromethoxyphenylhydrazone (FCCP) inhibited molybdate uptake, suggesting that at least a portion of the molybdate taken up (in 1 min) may be coupled to proton movement. Other ionophores were therefore used to examine the possible involvement of a transmembrane proton gradient (ApH) in molybdate uptake. The K+-H+ antiporter nigericin and the Na+-H+ antiporter monensin can be used to dissipate a pH gradient in an electroneutral manner (14). Both of these ionophores stimulated 02 uptake and inhibited molybdate uptake (Table 2). When valinomycin (with KCI) is added to respiring cells (that have established a membrane potential), it allows K+ uptake with concomitant dissipation

10

20

30

Time (sec) FIG. 4. Mo accumulation after rapid pH shifts. Bacteroid suspensions (ca. 3 mg [dry weight]/ml) in 50 mM HEPES (pH 7.0) received 0.10 pM molybdate containing 5 nM radioactive 9Mo and then sufficient microliter amounts of concentrated HCI to cause the indicated pH shift from pH 7.0 (ApH). After addition of acid, the suspension was rapidly mixed by shaking, and samples (0.3 ml) were rapidly removed and filtered at 10, 20, and 30 s. The ApH = 0 received CCCP (50 p.M) 30 min prior to 9Mo addition to initiate the

MOLYBDATE TRANSPORT BY B. JAPONICUM BACTEROIDS

VOL. 170, 1988

(Fig. 6) were analyzed showed that W042- competitively inhibited MoO42- uptake, with an apparent Ki of 0.034 FM. This value is near the K,JM for Mo by the cells. The results indicate that W042- binds to the same component(s) that is responsible for MoO42- binding or transport. Vanadate also inhibited Mo uptake but not as strongly as W042-; incubation of bacteroids simultaneously with both 1 FM MoO42and 20 ,M V03- resulted in a 40% decrease of Mo uptake compared with controls without V03- (data not shown).

10

> 35 _ 0

El

c0

5617

0.~~~~~

/

DISCUSSION /

,

1

2

3

5

4

6

,

.

. .,

7

8

9

lo

Time (min) FIG. 5. Exchangeability of transported Mo. The amount of Mo taken up by a bacteroid suspension (2.9 mg [dry weight]/ml) in 1.0 puM 'Mo was determined at the time points indicated (0). At 1 and at 5 min (arrows), nonradioactive Mo was added to a concentration of 100 ,uM to separate replicate vials, and the amount of 'Mo still associated with the cells was determined (0) by rapidly filtering samples (see text).

5), and even after a 5- or 10-min exposure to excess external molybdate, little more Mo was exchanged than after the first 30 s of exposure to external molybdate. The results indicate that most of the Mo taken up in the first few minutes is rapidly and tightly associated with cellular components, and therefore it probably is internalized rather than loosely associated with the outside of the cell. Effect of tungstate on molybdate uptake. It is well established that tungsten can replace Mo in some enzyme systems (21), resulting in W-containing inactive enzymes. We therefore tested the effect of W042- on MoO42- transport by B. japonicum bacteroids. Bacteroids incubated with both 1 puM MoO42- and 20 FLM W042- accumulated almost no Mo in 1 min. Experiments in which the concentrations of both Mo and W were varied and then the subsequent Dixon plots *Ss0.0 15^M

-I.0

-0.9

-0.8

SO0.025p&M

0.7

0.6 I--

0.5 o0 0.3

Su0.05

FM

A

0-.l

IK; 1 -0.03 -0.01 0 -0.04 -0.02

-0.05

0.01

0.03 0.02

0.05 0.04

Tungsten ( ELM) FIG. 6. Analysis of tungstate as an inhibitor of molybdate transport. Individual bacteroid suspensions (2.4 mg [dry weight]/ml) received 99Mo at the S concentration indicated (1% of the [S] was radioactive) along with W (sodium tungstate) at 0.01, 0.02, 0.03, 0.04, or 0.05 ,uM, and the Mo uptake rate in 1 min was determined. S = 0.05 FM (A), 0.025 puM (0), or 0.015 p.M (0).

Component I or dinitrogenase of the conventional nitrogenase system of N2-fixing bacteria contains molybdenum. For the proper expression and functioning of this nitrogenase, root nodule bacteroids must therefore be able to sequester Mo, and this ability presumably must be especially efficient when the legume is grown under Mo-deficient conditions. Mo-deficient soybean growth occurs in nature (16), and the ability of soybean bacteroids to sequester Mo can vary greatly on the basis of the bacterial strain used as inoculant (12). Bacteroids from Mo-limited soybeans nodulated by strain USDA 136 express a high-affinity molybdate uptake system component for Mo binding, with apparent K,,, for MoO42- of about 0.1 p.M. The affinity for required ions by bacteria can be extraordinarily high, as is best exemplified by the iron-binding chelators produced in response to iron deprivation (27). In regard to anion-sequestering systems, bacterial phosphate and sulfate transport systems generally have affinities for the anion in the 0.1 to 5 FM range (3, 23, 34). Some nitrogen-fixing bacteria have high-affinity Mobinding ability. The affinity for MoO42- by K. pneumoniae grown on Mo-deficient medium is very high, with an apparent Km of less than 0.03 ,uM (calculation by us from data in reference 17), whereas the apparent Km reported for Molimited N2-fixing C. pasteurianum was 48 ,uM (8). A. vinelandii also produced a high-affinity Mo-uptake system which can scavenge even trace amounts of Mo from the medium (29, 36). It is of interest that N2-fixing A. vinelandii produces siderochromes that have high affinity for MoO42- as well as for iron (28, 37), and the properties of molybdate uptake by K. pneumoniae are consistent with the existence of chelators on the cell surface (36). Free-living B. japonicum wild-type cells produce an extracellular factor that restores Mo uptake and nitrogenase activity in mutant strains lacking a highaffinity molybdate-binding component (25). Whether an extracellular factor(s) is involved in the symbiosis is not known. As previously reported (12), bacteroids of strain USDA 136 took up molybdate at a rate of about 3 pmol/min per mg of protein. In the current study, the Vmax was calculated to be 5 pmol/min/per mg (dry weight) of bacteroid. This value is considerably lower than the Vmax for molybdate uptake by C. pasteurianum (8), K. pneumoniae (17, 36), and, probably, A. vinelandii (36). However, Mo accumulation in those organisms was determined under conditions which derepress nitrogenase during the molybdate uptake assay; those conditions are probably physiologically very different from those used when studying isolated (differentiated) bacteroids. The isolation of B. japonicum bacteroids under anaerobic conditions to prevent the denaturation of a major Mo sink (nitrogenase) did improve Mo accumulation ability but not short-term molybdate transport ability. Interestingly, Mo accumulation in K. pneumoniae and C. pasteurianum is coregulated with the expression of nif genes, and when

5618

MAIER AND GRAHAM

nitrogenase synthesis is repressed by ammonium or by 02 Mo accumulation is prevented (19, 29, 36). The inhibitor studies suggested that MoO42- is transported by an active transport system that depended on a H+ gradient across the cytoplasmic membrane. Other evidence for an H+-MoO42- cotransport mechanism was the demonstration that molybdate uptake was driven by an artificially imposed (inwardly directed) pH gradient. Phosphate transport by Escherichia coli (34) and Paracoccus denitrificans (4), sulfate (3) and nitrate (22, 23) transport in P. denitrificans, and sulfate uptake by Desulfovibrio desulfuricans (6) are also postulated to involve H+ symport mediated systems. Uptake of sulfate, sulfite, and thiosulfate by D. desulfuricans was recently shown to be dependent on ApH, and an electroneutral proton/anion symport mechanism was proposed for all three anions (6). These systems, apparently like MoO42- uptake in B. japonicum, are therefore postulated to take advantage of the directed proton gradient produced by an electron transport chain to facilitate electroneutral uptake of the anion (14). Molybdate transport in B. japonicum bacteroids has some similarities to as well as differences from the systems described in the free-living nitrogen-fixing bacteria C. pasteurianum (8, 9), K. pneumoniae (17, 18, 36), and A. vinelandii (29, 36). Tungstate is a competitive inhibitor of molybdate uptake in C. pasteurianum as well as in A. vinelandii and K. pneumoniae, and so it was not surprising to find that tungstate competitively inhibited molybdate uptake by B. japonicum bacteroids. Many studies have shown that Mo is more efficiently used than W for the effective functioning of nitrogenase (36). However, most of the studies, including ours, on B. japonicum bacteroids demonstrate that the transport process itself does not discriminate between the two elements. Vanadate also inhibited molybdate uptake in B. japonicum but to a much lesser extent than tungstate. In contrast to our results with B. japonicum bacteroids, molybdate uptake by C. pasteurianum (8) was dependent on the availability of a metabolizable carbon source in the medium during the molybdate transport assay. Little of the 99Mo transported by B. japonicum bacteroids was exchangeable with a 100-fold excess of unlabeled Mo. These results are similar to those observed for K. pneumoniae, A. vinelandii, and C. pasteurianum and indicate that Mo in all these organisms is rapidly assimilated into intracellular intermediates like the iron-molybdenum cofactor (36, 38). This rapid association of Mo into intermediates must serve to deplete internal pools of initially bound Mo and thus help to drive molybdate from outside to inside the cell. The current study addresses molybdenum uptake as molybdate, and this is justified, since it is the most abundant form in solution. It is likely that molybdate is the primary form of Mo supplied to bacteroids in the root nodule, but it should be remembered that other forms of Mo could be used; thus, those other forms of Mo may also require specific transport systems. These other forms of Mo could even include relatively insoluble forms, since MoS2, for example, can serve as a Mo source for nitrogenase in K. pneumoniae (36). Thus, many more aspects of Mo uptake and metabolism by B. japonicum must be studied. ACKNOWLEDGMENT

This work was supported by a grant from the Allied Chemical

Corporation.

J. BACTERIOL. LITERATURE CITED 1. Bergersen, F. J., and G. L. Turner. 1967. Nitrogen fixation by the bacteroid fraction of bries of soybean root nodules. Biochim. Biophys. Acta 387:461-474. 2. Bhandari, B., and D. J. D. Nicholas. 1985. Proton motive force in washed cells of Rhizobium japonicum and bacteroids from Glycine max. J. Bacteriol. 164:1383-1385. 3. Burnell, J. N., P. John, and F. R. Whatley. 1975. The reversibility of active sulfate transport in membrane vesicles of Paracoccus denitrificans. Biochem. J. 150:527-536. 4. Burnell, J. N., P. John, and F. R. Whatley. 1975. Phosphate transport in membrane vesicles of Paracoccus denitrificans. FEBS Lett. 58:215-218. 5. Coventry, D. R., J. R. Hirth, T. G. Reeves, and V. F. Burnett. 1985. Growth and nitrogen fixation by subterranean clover in response to inoculation, molybdenum application, and soil amendment with lime. Soil Biol. Biochem. 17:791-796. 6. Cypionka, H. 1987. Uptake of sulfate, sulfite and thiosulfate by proton-anion symport in Desulfovibrio desulfuricans. Arch. Microbiol. 148:144-149. 7. Eady, R. R. 1980. Isolation and characterization of various nitrogenases. Methods Enzymol. 69:753-778. 8. Elliot, B. B., and L. E. Mortenson. 1975. Transport of molybdate by Clostridium pasteurianum. J. Bacteriol. 124:1295-1301. 9. Elliot, B. B., and L. E. Mortenson. 1976. Regulation of molybdate transport by Clostridium pasteurianum. J. Bacteriol. 127: 770-779. 10. Eskew, D. L., R. M. Welch, and E. E. Cary. 1984. A simple plant nutrient solution purification method for effective removal of trace elements using controlled pore glass-8-hydroxyquinoline chelation column chromatography. Plant Physiol. 76:103-105. 11. Evans, H. J., B. Koch, and R. Klucas. 1972. Preparation of nitrogenase from nodules and separation into components. Methods Enzymol. 24:470-476. 12. Graham, L., and R. J. Maier. 1987. Variability in molybdenum uptake activity in Bradyrhizobium japonicum strains. J. Bacteriol. 169:2555-2560. 13. Harold, F. M. 1972. Conservation and transformation of energy by bacterial membranes. Bacteriol. Rev. 36:172-230. 14. Harold, F. M. 1986. The vital force: a study of bioenergetics. W. H. Freeman and Co., New York. 15. Hinton, S. M., and L. E. Mortenson. 1985. Regulation and order of involvement of molybdoproteins during synthesis of molybdoenzymes in Clostridium pasteurianum. J. Bacteriol. 162:485493. 16. Huang, C. Y. 1979. The responses of soybean plants to molybdenum treatment. Taiwania 24:38-46. 17. Imperial, J., R. A. Ugalde, V. K. Shah, and W. J. Brill. 1984. Role of the nifQ gene product in the incorporation of molybdenum into nitrogenase in Klebsiella pneumoniae. J. Bacteriol. 158:187-194. 18. Imperial, J., R. A. Ugalde, V. K. Shah, and W. J. Brill. 1985. Mol- mutants of Klebsiella pneumoniae requiring high levels of molybdate for nitrogenase activity. J. Bacteriol. 163:1285-1287. 19. Jacobsen, M. R., R. Premakumar, and P. E. Bishop. 1986. Transcriptional regulation of nitrogen fixation by molybdenum in Azotobacter vinelandii. J. Bacteriol. 167:480-486. 20. Janssen, K. A., and M. L. Vitosh. 1974. Effect of lime, sulfur and Mo on nitrogen fixation and yield of dark red kidney beans. Agron. J. 66:736-739. 21. Keeler, R. F., and J. E. Varner. 1957. Tungstate as an antagonist of molybdate in Azotobacter vinelandii. Arch. Biochem. Biophys. 70:585-590. 22. Kristjansson, J. K., B. Walter, and T. C. Hollocher. 1978. Respiration-dependent proton translocation and the transport of nitrate and nitrite in Paracoccus denitrificans and other denitrifying bacteria. Biochemistry 17:5014-5019. 23. Lanyi, J. K. 1987. Chloride, nitrate, and sulfate transport in bacteria, p. 249-267. In B. P. Rosen and S. Silver (ed.), Ion transport in prokaryotes. Academic Press, Inc., New York. 24. Maier, R. J., and W. J. Brill. 1976. Ineffective and nonnodulating mutants of Rhizobium japonicum. J. Bacteriol. 127: 763-769.

VOL. 170, 1988

MOLYBDATE TRANSPORT BY B. JAPONICUM BACTEROIDS

25. Maier, R. J., L. Graham, R. G. Keefe, T. Pihi, and E. Smith. 1987. Bradyrhizobium japonicum mutants defective in nitrogen fixation and molybdenum metabolism. J. Bacteriol. 169:25482554. 26. Maier, R. J., and D. M. Merberg. 1982. Rhizobium japonicum mutants that are hypersensitive to repression by H2 uptake by oxygen. J. Bacteriol. 150:161-167. 27. Nielands, J. B. 1981. Iron absorption and transport in microorganism. Annu. Rev. Nutr. 1:27-48. 28. Page, W. J., and M. von Tigerstrom. 1982. Iron- and molybdenum-repressible outer membrane proteins in competent Azotobacter vinelandii. J. Bacteriol. 151:237-242. 29. Pienkos, P. T., and W. J. Brill. 1981. Molybdenum accumulation and storage in Klebsiella pneumoniae and Azotobacter vinelandii. J. Bacteriol. 145:743-751. 30. Pope, M. T., E. R. Steel, and R. J. P. Williams. 1980. The comparison between the chemistry and biochemistry of molybdenum and related elements, p. 3-40. In M. Coughlan (ed.), Molybdenum and molybdenum-containing enzymes. Pergamon Press, Oxford. 31. Rigaud, J., F. J. Bergersen, G. L. Turner, and R. M. Daniel. 1973. Nitrate dependent anaerobic acetylene reduction and nitrogen fixation by soybean bacteroids. J. Gen. Microbiol. 77: 137-144. 32. Roberts, G. P., T. MacNeil, D. MacNeil, and W. J. Brill. 1978. Regulation and characterization of protein products coded by

33. 34.

35. 36. 37.

38.

39.

5619

njf (nitrogen fixation) genes of Klebsiella pneumoniae. J. Bacteriol. 136:267-279. Rosen, B. P., and S. Silver (ed.). 1987. Ion transport in prokaryotes. Academic Press, Inc., N.Y. Rosenberg, H. 1987. Phosphate transport in prokaryotes, p. 205248. In B. P. Rosen and S. Silver (ed.), Ion transport in prokaryotes. Academic Press, Inc., New York. Ruiz-Argueso, T., D. W. Emerich, and H. J. Evans. 1979. Characteristics of the H2 oxidizing system in soybean nodule bacteroids. Arch. Microbiol. 121:199-206. Shah, V. K., R. A. Ugalde, J. Imperial, and W. J. Brill. 1984. Molybdenum in nitrogenase. Annu. Rev. Biochem. 53:231-237. Stiefel, E. I., B. K. Burgess, S. Wherland, W. E. Newton, J. L. Corbin, and G. D. Watt. 1980. Azotobacter vinelandii biochemistry: H2 (D2)/N2 relationships of nitrogenase and some aspects of iron metabolism, p. 211-222. In W. E. Newton and W. H. Orme-Johnson (ed.), Nitrogen fixation, vol. 1. University Park Press, Baltimore. Ugalde, R. A., J. Imperial, V. K. Shah, and W. J. Brill. 1985. Biosynthesis of the iron-molybdenum cofactor and the molybdenum cofacter in Klebsiella pneumoniae: effect of sulfur source. J. Bacteriol. 164:1081-1087. Vincent, J. M. 1970. A manual for the practical study of root nodule bacteria. International Biological Programme. Blackwell Scientific Publications, Ltd., Oxford.