Pleomorphic Forms of Bradyrhizobium japonicum

Transcription

1 APPLIED AND ENVIRONMENTAL MICROBIOLOGY, Mar. 1989, p /89/ $02.00/0 Copyright C 1989, American Society for Microbiology Vol. 55, No. 3 Physiological Characterization of Dicarboxylate-Induced Pleomorphic Forms of Bradyrhizobium japonicum H. KEITH REDINGt AND JOE EUGENE LEPOt* Department of Biology, The University of Mississippi, University, Mississippi Received 12 August 1988/Accepted 12 December 1988 When Bradyrhizobiumjaponicum was transferred into medium containing 40 mm succinate or 40 mm fumarate, over 90% of the bacteria acquired a swollen, pleomorphic form similar to that of bacteroids. The induction of pleomorphism was dependent on the carbon substrate and concentration but was independent of the hydrogen ion and sodium ion concentration. Cell extracts of rod-shaped and pleomorphic cells contained enzymes required for sugar catabolism and gluconeogenesis. Variations in these enzyme profiles were correlated with the carbon source used and not with the conversion to the bacteroid-like morphology. Rod-shaped cells cultured on glucose or 10 mm succinate transported glucose and succinate; however, the pleomorphic cells behaved similarly to symbiotic bacteroids in that they lacked the ability to transport glucose and transported succinate at lower rates than did rod-shaped cells. Rhizobia form symbioses with a variety of leguminous plants. The plants produce photosynthate, some of which supplies the reductant and ATP required for nitrogen fixation. This substrate may limit nitrogen fixation rates (1, 14). Although sucrose and sugar alcohols are the most abundant carbon-containing compounds in the nodule cytosol (14), dicarboxylic acids are the most likely class of compounds to support symbiotic nitrogen fixation for the following reasons: (i) bacteroids do not transport (16) or oxidize (41) sugars, (ii) bacteroids do oxidize (9, 26, 41) and transport (6, 11, 29) tricarboxylic acid cycle intermediates, (iii) mutants defective in sugar metabolism form effective symbioses (10, 31), and (iv) mutants defective in dicarboxylate metabolism form ineffective symbioses (6, 7, 30). The recent introduction of dicarboxylate transport (dct) genes from Rhizobium meliloti into Bradyrhizobium japonicum produced a strain with enhanced succinate uptake and freeliving nitrogen-fixing activities (2). Early in the symbiotic association, the rhizobia develop structurally and physiologically into bacteroids. The bacteria change from a rod shape to a swollen, pleomorphic form (4) in which cell division ceases (13). Free-living Rhizobium trifolii acquire a similar morphology when cultured in media containing succinate (42), as do rhizobia cultured in the presence of alkaloids (43) or yeast extract (17, 18, 35, 40). The physiological differentiation, aside from the development of the nitrogen-fixing mechanism, consists of changes in carbon substrate metabolic capabilities. In general, sugarcatabolizing pathways, e.g., Entner-Doudoroff (ED) and Embden-Meyerhoff-Parnas, are shut down (28, 32, 34, 39). Moreover, free-living rhizobia actively transport glucose (33, 38) and dicarboxylates (5, 16, 24), whereas bacteroids transport dicarboxylates but not sugars (6, 12, 29, 33). This study further elucidates the role of dicarboxylates in the Rhizobium-legume symbiosis. We have characterized requirements for the induction of a bacteroid-like morphology in free-living cells as well as the effects of factors such as * Corresponding author. t Present address: Department of Microbiology, University of Georgia, Athens, GA t Present address: ECOGEN Inc., 2005 Cabot Boulevard West, Langhorne, PA ph and mono- and divalent cation concentrations. In addition, we report the activity profiles of carbohydrate-catabolizing enzymes and substrate uptake activities of rod-shaped and pleomorphic cells. Thus, we have compared the physiology of rod-shaped and pleomorphic cells with that of symbiotic bacteroids to determine whether the substrateinduced morphological transformation in free-living cells is accompanied by an altered physiology analogous to that of the bacteroids. We found that the pleomorphic cells had a carbon substrate catabolic enzyme profile similar to that of the rod-shaped cells; however, such pleomorphic cells had glucose and succinate transport capabilities characteristic of bacteroids. MATERIALS AND METHODS Organism and cultivation. B. japonicum strains USDA and USDA 136 were obtained from the U.S. Department of Agriculture culture collection at Beltsville, Md. Unless otherwise indicated, all studies were performed with strain Bacteria were maintained on agar slants of hydrogen uptake medium (HUM; 22) containing 20 mm sodium gluconate as the sole source of carbon. Experimental cultures were grown in a modified HUM broth supplemented with biotin (1 mg/liter) and NH4Cl (1 g/liter). Other carbon substrates replaced the gluconate as indicated. The medium was adjusted to ph 7.0 with 5 N NaOH or concentrated NH40H before autoclaving. Phosphates (10 mm NaPO4, ph 7.0), iron-edta, and carbon substrates were autoclaved separately and added to the sterile salts-vitamin solution. Solid medium contained 15 g of agar (Sigma Chemical Co., St. Louis, Mo.) per liter. Experimental cultures grown on solid media were incubated at 29 C. Broth cultures in 20-mm test tubes were shaken at 29 C in a shaker-incubator (model G25; New Brunswick Scientific Co., Inc., Edison, N.J.). Growth was monitored by noting the optical density at 540 nm, using a Spectronic 501 spectrophotometer (Bausch & Lomb, Inc., Rochester, N.Y.). To determine cell morphology, heat-fixed smears were prepared at the desired time, stained for 1 min with crystal violet, and viewed under bright-field oil immersion, using a Nikon Optiphot microscope.

2 VOL. 55, 1989 DICARBOXYLATE-INDUCED PLEOMORPHISM OF B. JAPONICUM 667 FIG. 1. B. japonicum grown on (A) 20 mm gluconate and (B) 40 mm succinate. Photomicrographs were taken with a phase-contrast Nikon Optiphot microscope (magnification, x4,000). Inoculation of cultures for enzyme and transport studies. Starter cultures grown on HUM-gluconate (20 mm) or HUM-L-arabinose (20 mm) were transferred to sterile centrifuge tubes and centrifuged at 7,000 x g and 4 C for 10 min. The cells were washed twice with 10 mm phosphate-buffered HUM salts, resuspended in the buffered salts, and used to inoculate experimental cultures for enzyme and transport studies. Protein determination. For the cell extract, protein was estimated by the dye-binding method of Bradford (3), with bovine serum albumin as a standard. For uptake assays, whole cells and bovine serum albumin standards were first digested by a modification of the method of Stickland (36) in which 1 ml of whole-cell suspension was heated to 100 C for 5 min in 3% NaOH. Preparation of cell extract. Two liters of early-stationaryphase cells were harvested by centrifugation, washed twice with 10 mm NaPO4-buffered HUM salts, suspended in 50 mm HEPES (N-2-hydroxyethylpiperazine-N'-2-ethanesulfonic acid; ph 7.5) with 0.2 mm dithiothreitol, and passed twice through a French press at 15,000 lb/in2. The crude extract was then centrifuged at 44,000 x g for 20 min, and the supernatant fluid was assayed for the indicated enzymes. Enzyme assays. Enzymes were assayed by published procedures as follows: glucokinase (21), glucose-6-phosphate dehydrogenase (21), fructokinase (10, 21), gluconokinase (20), hexose diphosphatase (38), and succinate dehydrogenase (15). Fructose-1,6-bisphosphate aldolase and the ED enzyme were assayed by monitoring the reduction of NAD+ at 340 nm, using a glyceraldehyde-3-phosphate dehydrogenase- 3-phosphoglycerate phosphokinase-coupled assay system. The final reaction mixture contained 2.5 mm NaPO4, 0.2 mm 3-NAD+, 1.66 mm ADP, and excess commercial glyceraldehyde-3-phosphate dehydrogenase-3-phosphoglycerate phosphokinase in a total volume of 1 ml. For each assay, additions were made to contain the following in the final reaction mixture: for fructose-1,6-bisphosphate aldolase, 40 mm Tricine {N-[2-hydroxy-1,1-bis(hydroxymethyl)ethyl]-glycine}, ph 8.0; Sigma Chemical Co., St. Louis, Mo.} 100 mm KCl, 0.5 mm CoCl2, and 25 mm fructose-1,6-bisphosphate; and for the ED enzyme, 40 mm Tricine (ph 8.0) and 50 mm 6-phosphogluconate. This method effectively blanks out background activity caused by endogenous NADH oxidase. All enzymes were assayed at 30 C in a quartz cuvette with a 1-cm light path. The change in absorbance was monitored by using a Bausch & Lomb Spectronic 501 spectrophotometer; in each assay, minus-substrate controls were used. All commercial enzymes were purchased from Sigma. Transport of glucose and succinate by whole cells. Cells from 500-ml liter cultures were collected by centrifugation, washed twice with the uptake medium (HUM salts, NH4Cl, vitamins, and 10 mm NaPO4, ph 7.0), and suspended in 5 ml of uptake medium. The whole-cell suspension was diluted with uptake medium to contain 1 mg of protein per ml. Each uptake assay required 2.5 ml of cell suspension. Glucose or succinate was added to the cell suspension at a final concentration of 2 mm. The assay was initiated by the addition of [2,3-14C]glucose (0.13 to 0.28,uCi per assay mixture) or [2,3-_4C]succinate (0.91 to 1.76,uCi per assay mixture). The assay mixtures were shaken at 30 C in a water bath; 0.5-ml portions were removed when desired, vacuum filtered through a 0.45-,um-pore-size nitrocellulose filters (TCM-450; Gelman Sciences, Inc., Ann Arbor, Mich.), and washed with 10 ml of uptake medium without the carbon substrate. The filters were air dried, placed in 7-ml glass miniscintillation vials, and completely covered with 6 ml of Safety-Solve scintillation cocktail (Research Products International Corp., Mount Prospect, Ill.). Radioactivity was determined by using an LS6800 liquid scintillation counter (Beckman Instruments, Inc., Fullerton, Calif.). The counts from time zero were subtracted from each time reading to correct for nonspecific binding of the substrate to the cells. Transport rates were calculated from the linear portion of the curve. RESULTS Induction of pleomorphism. Figure 1A shows the typical rod shape of B. japonicum The pleomorphic cells (Fig. 1B) were induced within 38 h by 40 mm succinate and were typical of the altered morphology produced by several carboxylic acid substrates. To determine the carbon substrate(s) and the concentration required to produce pleomorphic forms in free-living cells, we grew B. japonicum on various carbon substrates, including sugars, amino acids, and dicarboxylates (Table 1). The sugars supported growth at all concentrations tested but did not induce pleomorphism up to 60 mm. The four-carbon

3 668 REDING AND LEPO APPL. ENVIRON. MICROBIOL. Carbon substrate TABLE 1. Growth substrate and pleomorphism of B. japonicum Growth/pleomorphism" at substrate concn (mm) of: Glucose G/- G/- GI- GI- GI- GI- Fructose G/- G/- GI- GI- GI- GI- L-Arabinose G/- G/- G/- GI- GI- GI- Gluconate G/- G/- GI- GI+ GI+ GI+ + Pyruvate G/- G/- GI- GI+ GI+ + Gl++++ Glutamate G/- G/+ G/- GI- GI+ GI+ + Citrate G/- GI- G/+ GI++++ GI++++ G+++++ o-ketoglutarate G/- GI- GI+ Gl++++ Gl++++ Gl++++ Succinate G/+ GI+ + G/+ + + G/++++ N/- N/- Fumarate G/+ GI+ + GI+++ GI++++ N/- N/- Malate G/- GI+ GI+ + G/+++ N/- N/- Oxalacetate G/- GI+ N/- N/- N/- N/- Malonate G/- GI- GI++ G/++++ GI++++ a Symbols: G, growth on indicated substrate; N, no growth on indicated substrate; +, <10% pleomorphism; + +, 11 to 30% pleomorphism; + + +, 31 to 50% pleomorphism; , >50% pleomorphism. dicarboxylates, succinate and fumarate, were the only substrates that induced pleomorphic cells at 10 mm. Furthermore, at 40 mm, these compounds produced pleomorphism in approximately 90% of the total cells. Higher concentrations of succinate or fumarate completely inhibited growth. Growth on malate also generated pleomorphic cells but less efficiently than did growth on succinate or fumarate. Other dicarboxylates, ot-ketoglutarate and malonate, and a tricarboxylate, citrate, also generated pleomorphic cells but only at concentrations of 30 mm or higher. Gluconate, pyruvate, and glutamate were poor inducers of pleomorphic cells, requiring 50 mm before more than 10% of the cells became pleomorphic. In analogous experiments, similar results were obtained for B. japonicum strains C33, PJ18, USDA 136, and USDA L-110 (data not shown). The issue of whether the induction of pleomorphic cells is a physiological-genetic effect or a physical-chemical effect was addressed. Finan et al. (5) have suggested that succinate chelates the divalent cations required for normal cell development. Therefore, we grew cells in broth cultures of HUM-succinate (40 mm) containing different concentrations of Mg2" (MgSO4 7H2O) and Ca2+ (CaCl2-2H20). Higher concentrations of these divalent cations in the media allowed cells to divide for a longer period of time (Fig. 2). Nevertheless, at the stationary phase of growth, such cultures contained about 90% pleomorphic cells, although mid-log-phase cultures contained less than 10% pleomorphic cells (data not shown). Possible chelation effects were also tested by growing the bacteria in HUM-L-arabinose (20 mm) broth containing different concentrations of EDTA. The bacteria grew in the presence of 2 mm EDTA but were unable to grow in 4 mm EDTA (data not shown). However, no pleomorphic forms were found in cultures containing 2 or 4 mm EDTA. Furthermore, when we cultured B. japonicum on 20 mm L-arabinose in the presence of 40 mm itaconic acid (a succinate analog that does not support growth), the cells did not become pleomorphic. We also found that L-arabinose did not prevent the induction of pleomorphism if the bradyrhizobia were grown in media that also contain succinate (data not shown). Because we supplied succinate in the growth medium as disodium succinate, the possible effects of sodium concentration on cell morphology were tested. Succinate was supplied as free succinic acid, and the ph of the medium was adjusted by using NH40H, with 40 mm MOPS [3-(Nmorpholino)propanesulfonic acid] serving as the buffer. The sodium concentration was adjusted over a range of 10 to 90 mm (supplied as NaCl). Over 90% of the cells became pleomorphic when cultured in medium containing 40 mm succinate regardless of the sodium concentration. In cultures containing 40 mm sodium gluconate, which usually does not generate pleomorphic cells, 60% of the bacteria became pleomorphic when the sodium concentration was increased to 90 mm. However, cells cultured on 40 mm L-arabinose produced no pleomorphic forms even in the presence of 90 mm NaCl. Pleomorphism could be induced over a ph range of 6.0 to 9.0 (data not shown). Enzyme profiles. Because the Embden-Meyerhoff-Parnas pathway, the ED pathway, and the tricarboxylic acid cycle exist in most rhizobia and bradyrhizobia (39), we chose to examine the initial enzymes of the Embden-Meyerhoff- Parnas pathway, the enzymes of the ED pathway, succinate dehydrogenase and hexose diphosphatase. Gluconate or succinate was used as the growth substrate. We found that extracts of B. japonicum cultured on 20 mm gluconate contained glucokinase, fructokinase, gluconokinase, glucose-6-phosphate dehydrogenase, the ED enzyme, and succinate dehydrogenase but lacked fructose- 1,6-bisphosphate aldolase and hexose diphosphatase (Table 2). Comparable levels of sugar-catabolizing enzymes and the gluconeogenic enzymes, hexose diphosphatase and fructose- 1,6-bisphosphate aldolase, were found in bacteria cultured on succinate. We could not detect fructokinase in cells grown on succinate; however, fructokinase was found in very low amounts when cells were cultured on gluconate (Table 2) or fructose (data not shown). Under similar conditions, B. japonicum USDA 136 expressed sugar-catabolizing enzymes at levels comparable to those of strain (data not shown for strain USDA 136). Active transport of glucose and succinate. We cultured B. japonicum on defined media and examined the ability of the cells to transport glucose and succinate. Glucose transport was highest in cells grown on 20 mm glucose. Cells cultured on 20 mm glucose plus 10 mm succinate transported glucose at 0.7 nmol/min per mg of whole-cell protein, a rate half that seen with glucose cultures, which was 1.4 nmol/min per mg of whole-cell protein. In cultures grown on 40 mm succinate or 40 mm succinate plus 20 mm glucose (both are conditions that induce pleomorphism), no glucose uptake could be detected. These bacteria were also tested for ability to transport

4 VOL. 55, 1989 DICARBOXYLATE-INDUCED PLEOMORPHISM OF B. JAPONICUM 669 E co L B TIM1E (hours) FIG. 2. Effect of divalent cations on growth on succinate. Cultures were inoculated from a washed suspension of early-log-phase cells grown on 20 mm gluconate. (A) Cultures containing 40 mm succinate, 1 mm CaCl2, plus MgSO4 at concentrations of 1 (0), 10 (*), and 40 (U) mm. (B) Cultures containing 40 mm succinate, 4 mm CaCI2, and MgSO4 at concentrations of 1 (LO), 10 (+), and 40 (E) mm. succinate. Transport was highest in cells cultured on 10 mm succinate or 20 mm glucose. When the bradyrhizobia were grown in medium containing 40 mm succinate, which generated 90%opleomorphic cells, succinate transport (determined with 1-mmn assays) was 4 nmol/min per mg of protein. This is a very low rate compared with that of rod-shaped cells produced by growth on 10 mm succinate or 20 mm glucose, which was 51 or 64 nmollmin per mg of whole-cell protein, respectively. When we grew the bacteria in 40 mm succinate with 40 mm MgSO4 and 4 mm CaCi2, the increase in Mg2+ Ca24 and had no effect on glucose or succinate transport (data not shown). However, increasing the concentration of these divalent cations does enhance succinate transport in B. japonicum grown on 15 mm succinate (24). DISCUSSION The induction of pleomorphism appears to be a substratedependent physiological-genetic phenomenon, with succinate and fumarate being the best substrates for inducing the transformation. Interestingly, strains of R. meliloti that lack succinate dehydrogenase but have the ability to transport succinate fail to produce pleomorphic forms when grown in the plant or in media containing succinate (8). This morphological transformation could not be prevented by increasing the magnesium or calcium concentration in the medium, nor could it be produced solely by growing the bacteria in the presence of EDTA; therefore, the chelation of divalent cations alone is not capable of producing pleomorphism. Magnesium and calcium have similar effects in some fast-growing rhizobia. Urban and Dazzo (42) reported that in media containing 16.6 mm succinate, R. trifolii continue to divide if 40 mm MgSO4 and 46 mm CaCl2 are added, although pleomorphic cells are still formed. The sodium concentration or the ph of the medium does not seem relevant to this morphological transformation. Enzyme profiles. Gluconate is metabolized primarily by the ED pathway (20); therefore, enzymes of this pathway should be present in extracts of gluconate-grown cells. Succinate was used because it is capable of generating either rod-shaped or pleomorphic forms, depending on its concentration in the medium (Table 1). Furthermore, in Rhizobium sp. strain 32H1, succinate represses the synthesis of sugarcatabolizing enzymes (39). Succinate-grown cells of cowpea Rhizobium sp. strain NGR 234 resemble bacteroids in that they lack fructokinase and show very low levels of other sugar-catabolizing enzymes (34). Bacteroids of B. japonicum USDA 110 lack the ED enzyme (the coupled activity of 6-phosphogluconate dehydratase plus 2-keto-3-deoxy-6-phosphogluconate aldolase), NADP-6-phosphogluconate dehydrogenase, and fructokinase, although other sugar-catabolizing enzymes are present (28, 32). Our results are similar to those reported by Mulongoy and Elkan (25), who used yeast extract mannitol-hepes-morpholineethanesulfonic acid (MES) as the culture medium. We found that the gluconate-grown cells, used as a positive control, contained the enzymes of the ED pathway, as did the cells grown 10 or 40 mm succinate. Two different serogroups (USDA 110 and USDA 136) of B. japonicum showed similar enzyme profiles; thus, this pattern was not confined to a single strain and serogroup. However, a much broader study would be necessary to further assess the prevalence of the pattern among the rather heterogeneous bradyrhizobia. Possibly, when B. japonicum I-110 was grown on 40 mm succinate, growth was inhibited before the bacteria had adequate time to repress the synthesis of sugar-catabolizing TABLE 2. Specific enzyme activities of B. japonicum grown on different substrates Enzyme Sp acta (nmol of substrate oxidized/ min per mg of protein) 20 mm Succinate gluconate 10 mm 40 mm Glucokinase Fructokinase Fructose-1,6-bisphosphate aldolase Gluconokinase Glucose-6-phosphate dehydrogenase ED enzymeb Hexose diphosphatase Succinate dehydrogenase a Mean of at least two experiments. b Combined activity of 6-phosphogluconate deoxy-6-phosphogluconate aldolase. dehydratase and 2-keto-3-

5 670 REDING AND LEPO enzymes and dissipate them from the cell. To test this possibility, we grew the bacteria in HUM-succinate (40 mm) in which MgSO4 and CaCl2 were increased to 40 and 4 mm, respectively. Increasing the Mg2' and Ca2+ concentrations in medium containing 40 mm succinate extended the growth period (Fig. 2). However, the sugar-catabolizing enzymes were still expressed, and repression did not occur (data not shown). Therefore, in contrast to reports on other rhizobial strains (34, 39), succinate did not repress the synthesis of sugar-catabolizing enzymes in B. japonicum In regard to cell morphology, cells grown on 10 tnm succinate (rod shaped) express enzyme profiles similar to those of cells cultured on 40 mm succinate (pleomorphic). Thus, enzyme expression in B. japonicum is dependent on the carbon growth substrate and is independent of cell morphology. Our results indicate that pleomorphic cells of B. japonicum express enzyme profiles similar to those of symbiotic bacteroids except for the ED enzyme, which Reibach and Streeter (28) find absent in extracts from bacteroids of B. japonicum USDA 110 and USDA 138. Salminen and Streeter (32) also were unable to detect the ED enzyme in either bacteroids or cultured bacteria. The presence of the E-D enzyme in cultured rhizobia, including the strains used by Streeter and co-workers, has been fully established by many scientists (11, 19, 20, 22, 23, 25, 31, 34, 37-39). Therefore, it is conceivable that the assay used by Streeter and coworkers may not be valid for B. japonicum. They assayed the ED enzyme by monitoring the oxidation of NADH, using lactate dehydrogenase. We were likewise unable to detect any ED enzyme activity by this assay. Our assay protocol allows for detection of the ED enzyme. In light of this, the reported lack of the ED enzyme in bacteroids of B. japonicum is questionable. Transport of glucose and succinate. In regard to carbon substrate uptake, free-living rhizobia actively transport glucose (33, 38) and succinate (5, 16, 24) into the cell. However, bacteroids of B. japonicum USDA 110 lack the ability to actively transport glucose, but they are able to actively transport succinate and other dicarboxylates (6, 12, 29). The transport of glucose appears to be regulated by the growth substrate, since succinate-grown cells of cowpea Rhizobium sp. strain 32H1 are unable to transport glucose, and succinate- or malate-grown cells of B. japonicum show very low glucose uptake rates (33). When B. japonicum was grown on 10 mm succinate or 10 mm succinate plus 20 mm glucose, the rate of glucose transport was one-half that seen in glucose-grown cells, which indicates that succinate represses glucose uptake. The inability of our pleomorphic cells to transport glucose suggests that either (i) the higher concentration of succinate (40 mm) completely repressed active transport of glucose or (ii) since the pleomorphic cells also transported succinate at low rates, these cells lost transport system proteins during the transition from a rod-shaped form to a bacteroid-like morphology. In Bradyrhizobium sp. strain 32H1, growth on succinate also leads to similar swollen, bacteroid-like forms which show an absence of several high-molecular-weight outer membrane polypeptides (27). Our pleomorphic bacteria generated by growth on 40 mm succinate showed glucose and succinate transport capabilities similar to those of symbiotic bacteroids. Both the pleomorphic cells and the bacteroids (29) lack the ability to transport glucose; in addition, the pleomorphic cells transported succinate at 4 nmol/min per mg of protein, and the bacteroids transport it at 3.7 nmol/min per mg of protein (29). In summary, growth on succinate, as well as other dicarboxylates, transformed free-living bradyrhizobia into cells having a bacteroid-like morphology. In addition, these transformed cells possessed a bacteroid-like physiology with respect to ability to transport glucose and succinate. These pleomorphic cells also expressed an enzyme profile similar to that reported for symbiotic bacteroids except that the bacteroids lacked the ED enzyme, although this result is questionable (see Discussion). ACKNOWLEDGMENTS APPL. ENVIRON. MICROBIOL. This work was supported by National Science Foundation grant BSR to J.E.L., by National Science Foundation equipment grant PCM , and by funds provided by the Department of Biology and the Office of Research of The University of Mississippi. We thank Harold L. Drake for helpful discussion. LITERATURE CITED 1. Bach, M. K., W. E. Magee, and R. H. Burris Translation of photosynthetic products to soybean nodules and their role in nitrogen fixation. Plant Physiol. 33: Birkenhead, K., S. S. Manian, and F. O'Gara Dicarboxylic acid transport in Bradyrhizobium japonicum: use of Rhizobium meloliti dct gene(s) to enhance nitrogen fixation. J. Bacteriol. 170: Bradford, M. M A rapid and sensitive method for the quantification of microgram quantities of protein utilizing the principle of protein-dye binding. Anal. Biochem. 72: Dazzo, F. B., C. A. Napoli, and D. H. Hubbell Adsorption of bacteria to roots as related to host specificity in the Rhizobium-clover symbiosis. Appl. Environ. Microbiol. 32: Finan, T. M., J. M. Wood, and D. C. Jordan Succinate transport in Rhizobium leguminosarum. J. Bacteriol. 148: Finan, T. M., J. M. Wood, and D. C. Jordan Symbiotic properties of C4-dicarboxylic acid transport mutants of Rhizobium leguminosarum. J. Bacteriol. 154: Gardiol, A., A. Arias, C. Cervenansky, and G. Martinez-Drets Succinate dehydrogenase mutant of Rhizobium meliloti. J. Bacteriol. 151: Gardiol, A., G. L. Truchet, and F. B. Dazzo Requirement of succinate dehydrogenase activity for symbiotic bacteroid differentation of Rhizobium meliloti in alfalfa nodules. Appl. Environ. Microbiol. 53: Glenn, A., and M. J. Dilworth Oxidation of substrates by isolated bacteroids and free-living cells of Rhizobium leguminosarum. J. Gen. Microbiol. 126: Glenn, A. R., R. Arwas, I. A. McKay, and M. J. Dilworth Fructose metabolism in wild-type, fructokinase-negative and revertant strains of Rhizobium leguminosarum. J. Gen. Microbiol. 130: Glenn, A. R., I. A. MacKay, R. Arwas, and M. J. Dilworth Sugar metabolism and the symbiotic properties of carbohydrate mutants of Rhizobium leguminosarum. J. Gen. Microbiol. 130: Glenn, A. R., P. Poole, and J. Hudman Succinate uptake by free-living and bacteroid forms of Rhizobium leguminosarum. J. Gen. Microbiol. 119: Gresshoff, P. M., and B. G. Rolfe Viability of Rhizobium bacteroids isolated from soybean nodule protoplasts. Planta 142: Hardy, R. W. F., and U. D. Havelka Photosynthate as a major limitation to N2 fixation by field grown legumes with an emphasis on soybeans, p In P. S. Nutman (ed.), International biological programme, symbiotic nitrogen fixation in plants, vol. 14. Cambridge University Press, New York. 15. Hederstedt, L., E. Holmgren, and L. Rutberg Characterization of succinate dehydrogenase complex solubilized from the cytoplasmic membrane of Bacillus subtilis with the nonionic detergent Triton X-100. J. Bacteriol. 138: Hudman, J., and A. R. Glenn Glucose uptake by free-

6 VOL. 55, 1989 DICARBOXYLATE-INDUCED PLEOMORPHISM OF B. JAPONICUM 671 living and bacteroid forms of Rhizobium leguminosarum. Arch. Microbiol. 128: Jordan, D. C The bacteroids of the genus Rhizobium. Bacteriol. Rev. 26: Jordan, D. C., and W. H. Coulter On the cytology and synthetic capacities of natural and artificially produced bacteroids of Rhizobium leguminosarum. Can. J. Microbiol. 11: Keele, B. B., P. B. Hamilton, and G. H. Elkan Glucose catabolism in Rhizobium japonicum. J. Bacteriol. 97: Keele, B. B., P. B. Hamilton, and G. H. Elkan Gluconate catabolism in Rhizobium japonicum. J. Bacteriol. 101: Lynch, W. H., J. MacLeod, and M. Franklin Effect of temperature on the activity and synthesis of glucose-catabolizing enzymes in Pseudomonasfiuorescens. Can. J. Microbiol. 21: Maier, R. J., N. E. R. Campbell, F. J. Hanas, F. B. Simpson, S. A. Russell, and H. J. Evans Expression of hydrogenase activity in free-living Rhizobium japonicum. Proc. Natl. Acad. Sci. USA 75: Martinez-Drets, G., and A. Arias Enzymatic basis for differentation of rhizobium into fast- and slow-growing groups. J. Bacteriol. 109: McAllister, C. F., and J. E. Lepo Succinate transport by free-living forms of Rhizobium japonicum. J. Bacteriol. 153: Mulongoy, K., and G. H. Elkan Glucose catabolism in two derivatives of a Rhizobium japonicum strain differing in nitrogen-fixing efficiency. J. Bacteriol. 131: Peterson, J. B., and T. A. La Rue Utilization of aldehydes and alcohols by soybean bacteroids. Plant Physiol. 68: Ramaswamy, P., and A. K. Bal Media-induced changes in the asymbiotic nitrogen-fixing bacteroid of Bradyrhizobium sp. 32H1. Curr. Microbiol. 14: Reibach, P. H., and J. G. Streeter Metabolism of 14Clabeled photosynthate and distribution of enzymes of glucose metabolism in soybean nodules. Plant Physiol. 72: Reibach, P. H., and J. G. Streeter Evaluation of active versus passive uptake of metabolites by Rhizobium japonicum bacteroids. J. Bacteriol. 159: Ronson, C. W., P. Lyttleton, and J. G. Robertson C4- dicarboxylate transport mutants of Rhizobium trifolii form ineffective nodules on Trifolium repens. Proc. Natl. Acad. Sci. USA 78: Ronson, C. W., and S. B. Primose Carbohydrate metabolism in Rhizobium trifoli: identification and symbiotic properties of mutants. J. Gen. Microbiol. 112: Salminen, S. O., and J. G. Streeter Uptake and metabolism of carbohydrates by Bradyrhizobium japonicum bacteroids. Plant Physiol. 83: San Francisco, M. J. D., and G. R. Jacobson Glucose uptake and phosphorylating activities in two species of slowgrowing rhizobium. FEMS Microbiol. Lett. 35: Saroso, S., M. J. Dilworth, and A. R. Glenn The use of activities of carbon catabolic enzymes as a probe for the carbon nutrition of snakebean nodule bacteroids. J. Gen. Microbiol. 132: Staphoist, J. L., and B. W. Strijdom The effect of yeast extract concentration in media on strains of Rhizobium meliloti. Phytophylactica 4: Stickland, L. H The determination of small quantities of bacteria by means of the biuret reaction. J. Gen. Microbiol. 5: Stowers, M. D Carbon metabolism in rhizobium species. Annu. Rev. Microbiol. 39: Stowers, M. D., and G. H. Elkan Transport and metabolism of glucose in cowpea rhizobia. Can. J. Microbiol. 29: Stowers, M. D., and G. H. Elkan Regulation of hexose catabolism in Rhizobium sp. 32H1. FEMS Microbiol. Lett. 26: Strijdom, B. W., and 0. N. Allen Medium supplementation with L- and D-amino acids related to growth and efficiency of Rhizobium meliloti. Can. J. Microbiol. 12: Tuzimuro, K., and H. Megura Respiration substrate of Rhizobium in the nodules. J. Biochem. 47: Urban, J. E., and F. B. Dazzo Succinate-induced morphology of Rhizobium trifolii 0403 resembles that of bacteroids in clover nodules. Appl. Environ. Microbiol. 44: Vincent, J. M Rhizobium: general microbiology, p In R. W. F. Hardy and W. S. Silver (ed.), A treatise on dinitrogen fixation. John Wiley & Sons, Inc., New York.