.‘L ..‘ . .A .. . . _ . . A . . .+ . . 1 . . .A . . «C . . . . .. . A . .. J A u 4.....5n' .. . . m . .. . O . . .. . | I . ~ A. I n A .. .. A. . u . ..v v . l . A I I . . - v D .. A v A . .. . . . . . .. u A . .. . A . r N , A , A . D . . A A A . c .A . . D . . . , AA . . A . . . . A . . . A . . . A . . . . . . . A . 5 W. 1 A A . v A 0 . — . .4 . - A , A . . .. A v. . . . . .D, . . . . u . A . . ... w , _ . . . . . A . .. .. ..A‘ ..A. . . . A. ., . . . . _ . r .AA . I A . b . q . . . . A . . A . . . A . . A u , , . A .. . . . . . . . . . .. . . A u p A‘ 1 A o . . A .... . A AA. A A . , . n . . < o v . . . o I: II . . q . A . .. A: . . A A . u . . . .. . a . A... L A.A r A. A . A 1 . . A A . .. . . v A ul- 7 o a . . . ~ - p I..A . . t A A A . _. , . A . it]... . AA. ... .A ...? vy’s‘lvrhp . , . . u». .l .v..v..v. o l - .—,—> .——> O—-—-> { __________ 1 I :3 MIA 4» NM * NHL 4» (oxygen or low tixed nitrogen) nitHDK C > Figure 1: Nitrogen regulatory network (Ntr) in the enteric bacteria (18). Table 1. Components of nitrogen regulatory (Ntr) systems in enteric bacteria Gene (synonym) Gene product Function Regulatgry ntrA rpoN, glnF NtrA RNA polymerase binding of Ntr/Nif promoters ntrB glflL NtrB, NRII NtrC phosphorylation ntrg glnG NtrC, NRI Activator/Repressor of of Ntr promoters glnB GlnB, PII Stimulates GS adenylation and NtrC phosphorylation gan GlnD, UTase Uridylylates GlnB glnE GlnE, ATase Adenylylates GS nifA NifA Activator of Nif promoters nifL NifL Regulates NifA Regulon glnA Glutamine synthetase Nitrogen assimilation hg_ Histidine utilization Nitrogen assimilation enzymes glnHPQ Glutamine permease Glutamine transport nifHDK Nitrogenase Nitrogen fixation (ATase). Adenylylation is stimulated by the gene product encoded by glnB - a tetramer with small, 12,000 kd subunits (GlnB; 16). Deadenylylation is catalyzed by the adenylyltransferase (GlnE) and is stimulated by the uridylylated form of GlnB (GlnB-UMP 7, 25, 40). Thus the uridylylation of GlnB is a critical step in the activation/deactivation of GS. A uridylyltransferase (UTase), encoded by glnD, is responsible for this reaction (8, 25). GlnD may sense the nitrogen status of the cell via the glutamine:2-ketoglutarate ratio (42, 43). The activity of the positive transcriptional regulator, NtrC, is also regulated by GlnB, GlnD as well as NtrB (8, 16, 19, 25, 30). NtrB is a kinase/phosphatase that post-translationally modifies the NtrC protein (12, 54). This modification results in its activation (phosphorylation) or deactivation (dephosphorylation). As mentioned earlier GlnB is deuridylylated by GlnD and is then involved in adenylylating CS (8, 16). Coordinate with this reaction the deuridylylated GlnB also stimulates the dephosphorylation of NtrC by NtrB (16, 30). The function of NtrC in activating transcription has recently been clarified. It had been known that NtrC binds to specific sequence elements adjacent to Ntr-regulated promoters (39). Using a purified in vitro transcription system Popham et al. (36) found that after binding at these sites, NtrC catalyzes the isomerization of closed recognition complexes to open complexes in which DNA in the region of transcription start site is locally denatured. Whether NtrC makes physical contact with the NtrAzRNA polymerase complex in this reaction is not yet known (36). The specific interactions involved in enteric nitrogen global regulation are presented in model farm in Figure 1 (18). In summary: Under nitrogen limitation: 1. glnB or its gene product senses a low glutamine: 2-ketoglutarate ratio and GlnD uridylylates GlnB (GlnB-UMP), (8, 42). 2. Uridylylated GlnB is no longer active in stimulating the dephosphorylation of NtrC, (16). 3. NtrB phosphorylates NtrC (NtrC-P) which then becomes active (coordinate with NtrA, a sigma factor) in inducing transcription from the glnApZ promoter (30); GS activity increases. 4. Adenylylated GS already present in the cell is deadenylylated by ATase stimulated by GlnB-UMP, (40). 5. Newly transcribed/translated GS becomes available and is not adenylylated due to the presence of GlnB-UMP (16). Under nitrogen excess: 1. glnD or its gene product senses a high glutamine: Z-ketoglutarate ratio and deuridylylates GlnB (8, 42). 2. GlnB stimulates the phosphatase activity of NtrB and NtrC-P is dephosphorylated and thus made inactive, (30). 3. The decreased NtrC level causes transcription to revert to the weaker glnApl promoter thus lowering GS levels (38). 4. Concurrently, the deuridylylated GlnB stimulates the adenylyltransferase activity of GlnE and GS is inactivated, (16, 40). It is important to note that GS is maintained at a low level, unadenylylated, even under nitrogen excess in order to provide the cell with some nitrogen assimilatory capacity. In addition to GS, there are several other proteins under nitrogen control which can be considered part of the Ntr regulon. It has recently been reported in E. coli that the glutamine transport operon (gInHPQ) has two transcriptional start sites, one having homology to other Ntr promoters (33, 34). In S. typhimurium several amino acid transport functions are known to be under Ntr control including: the histidase (hisY), glutamine (glnH), and lysine-arginine-ornithine (argT) permeases (23). In the non-nitrogen- fixing strain of K. aerogenes studied by Magasanik and co-workers (25) it has been found that urease and the tryptophan permease are under nitrogen control, as well as enzymes involved in degrading histidine (Hut system) and proline (Put system). Nitrogen regulation in the enteric, nitrogen-fixers. Enteric free-living nitrogen-fixers such as K. pneumoniae share much of the regulatory circuitry modeled in the E. coli system (3, 9, 15, 18, 54). They however have additional genes in the Ntr regulon as result of their nitrogen-fixing capability. For example, it has been shown that NtrC is involved indirectly in the activation of nif structural genes via the activation of the nifLA operon (18, 45, 54). This operon encodes the negative regulatory factor NifL that under anaerobiosis or in the presence of sufficient nitrogen inhibits nifA (18, 28). NifA is the direct positive regulatory factor responsible for activating, along with NtrA, the transcription of the nifHOK operon and other nif genes which encode the components necessary for nitrogen-fixation (5, 15, 29, 35). Nitrogen regulation in nonenteric, free-living nitrogen-fixers. Nonenteric, free-living nitrogen-fixers with demonstrated Ntr systems include Azotobacter vinelandii (48) and Rhodopseudomonas capsulata (21). In R. capsulata regulatory genes were isolated that are unable to activate a nifH::lacZ fusion (21). 0f four such characterized genes one, nifRI, was found to be partially homologous to the E. coli ntrc gene based on Southern hybridizations. A ntrC::Tn5 mutant is Nif- but differs from the typical ntrC phenotype in several respects. The R. 10 capsulata ntrC strain is able to use proline, arginine, glutamine, or NH4 as sole nitrogen sources. In addition, in contrast to E. coli where ntrC transcript is 10-12 times more abundant in nitrogen-starved than nitrogen-excess cultures, R. capsulata ntrC transcript increased only two-fold in derepression conditions (21). The A. vinelandii ntrC gene was isolated by identifying cosmids that complemented E. coli ntrC strains (48). A mutant with a TnS insertion in the putative ntrC gene was unable to grow on N03“; nitrate reductase activity was 3-11% of wild type. However, this strain was able to fix nitrogen and was unaltered in glutamine synthetase activity. A. vinelandii expresses genes for an alternative nitrogenase when grown without molybdenum. This alternative nitrogenase was also found to be unaffected in the ntrC strain (48). Nitrogen regulation in symbiotic nitrogen-fixers The ntrC gene has been isolated and characterized in the symbiotic nitrogen-fixer Rhizobium meliloti (46). This species contains among others, two operons, nifHDK and fixABC, that require the nifA regulatory gene product for transcription under symbiotic conditions (15, 35). In addition, like most members of the Rhizobiaceae this organism contains two glutamine synthetases, I and II, encoded by different genes (13). R. meliloti ntrC::Tn5 insertion mutants were observed for their ability .to grow on various amino acids, NH4SO4, N03‘, for transcription of glnA, nifA, nifHD, nifB, and fixABC under free-living conditions and for symbiotic phenotype. Unlike K'pneumaniae the R. meliloti ntrC was not required for utilization of histidine or proline. The ntrC strain was 11 also able to grow on urea, glutamine, and NH4SO4. However, it was unable to grow on 0.5 mM N03“ indicating a similarity to A. vinelandii where nitrate reductase activity is decreased in an ntrC mutant (46, 48). Unlike the E. coli Ntr system (25), glnA transcript levels were found to be unaffected in the ntrC strain (46), thus providing an explanation for the ability to grow on NH4SO4. The glnII transcript level was not studied, but 6511 activity has been shown to be nitrogen- controlled in Bradyrhizobia (1, 2, 10, 11). Different results were obtained regarding nif expression depending on if free-living or symbiotic growth conditions were studied. Although nitrogen-deprived, free-living R. meliloti have not been shown to fix nitrogen, the ntrC gene was found to be required under these conditions for transcription of nifA, nifHDK, and fixABC, (21). The ntrC strains were able to nodulate alfalfa plants and showed wildtype levels of nitrogenase activity. Thus, in contrast to K. pneumoniae (18), the R. meliloti ntrC does not appear to activate nifA - at least not during symbiotic nitrogen fixation. Rather, it has been proposed that a symbiosis- specific signal is responsible for nifA activation in Rhizobia (46). Alternatively, nifA transcription may be constitutive or regulated in an ntrA-independent manner. Previous to the work presented here no ntrC mutants had been reported in any Bradyrhizobia. Nitrogen-regulated control of glnII expression under aerobic conditions had been demonstrated (10, 11), however, and 651 had been shown to exist in adenylylated and unadenylylated forms in response to nitrogen conditions (14, 24). These observations suggested that global nitrogen regulatory circuitry existed in the Bradyrhizobia. 12 In this dissertation I present experiments that utilized various B. japonicum mutants constructed by gene-directed mutagenesis. These experiments were designed to characterize the regulation of specific B. japonicum genes in response to shifts in nitrogen status and oxygen status. Chapter 1 reports the construction of glnA, glnII and glnA/glnII mutants and presents evidence indicating that neither glnA or glnII plays an essential role in the symbiotic association with soybean. Chapters 2 and 3 report the isolation of a regulatory gene, ntrC, and the construction of various mutants designed to elucidate the role of ntrC and another regulatory gene, nifA, in the control of gene expression in response to nitrogen and oxygen stress. The results of these studies suggest that separate regulatory networks exist in B. japonicum free-living and symbiotic cells. These networks are shown to be governed either by nitrogen or oxygen availability. Finally, Chapter 4 presents the isolation and regulation of the B. japonicum glnB gene. This gene was found to be regulated by the ntrC gene product at tandem promoters in response to nitrogen status, but not oxygen status. The work reported in chapters 1 and 2 has been presented elsewhere (11, 26). Chapters 3 and 4 will be submitted for publication to Molecular and General Genetics and Journal of Bacteriology. respectively. 13 LIST OF REFERENCES 1. Adams, T. H. (1986). Transcriptional regulation of nitrogen fixation and nitrogen assimilation genes in Bradyrhizobium japonicum. Ph.D. dissertation, Mich. State Univ., E. Lansing, MI. 2. Adams, T. H., and Chelm, B. K. (1988). Effects of oxygen levels on the transcription of nif and gln genes in Bradyrhizobium japonicum. J. Gen. Microbiol. 134:611-618. 3. Alvarez-Morales, A., Dixon, R. and Merrick, M. (1984). Positive and negative control of the glnA ntrBC regulon in Klebsiella pneunoniae. EMBO J. 3:501-507. 4. Ausubel, F. M. (1982). Molecular genetics of symbiotic nitrogen fixation. Cell 29:1-2. 5. Ausubel, F. M. (1984). Regulation of nitrogen fixation genes. Cell 37:5-6. 6. Backman, K., Chen, Y-M, and Magasanik, B. (1981). Physical and genetic characterization of the glnA-glnLG region of the Escherichia coli chromosome. P.N.A.S. USA 78:3743-3747. 7. Bancroft, S., Rhee, S. G. Neumann, C., and Kustu, S. (1978). Mutations that alter the covalent modification of glutamine synthetase in Salmonella typhimurium. 134:1046-1055. 00 Bloom, F. R., Levin, M. S., Foor, F., and Tyler, B. (1978). Regulation of glutamine synthetase formation in Escherichia coli: characterization of mutants lacking uridylyltransferase. J. Bacteriol. 134:569-577. \0 . Buck, H., Miller, S. Drummond, M., and Dixon, R. (1986). Upstream activator sequences are present in the promoters of nitrogen fixation genes. Nature 320:374-378. 10. Carlson, T. A. (1986). Characterization of the genes encoding glutamine synthetase I and glutamine synthetase II from Bradyrhizobium japonicum. Ph.d. dissertation, Mich. State Univ., E. Lansing, MI. 11. Carlson, T. A., Martin, G. B., and Chelm, B. K. (1987). Differential transcription of the two glutamine synthetase genes in Bradyrhizobium japonicum. J. Bacteriol. 169:5861-5866. 12. Chen, Y.-M., Backman, K., Magasanik, B. (1982). Characterization of a gene, glnL, the product of which is involved in the regulation of nitrogen utilization in Escherichia coli. J. Bacteriol. 150:214-220. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 14 Darrow, R. A., and Knotts, R. R. (1977). Two forms of glutamine synthetase in free-living root-nodule bacteria. Biochem. Biophys. Res. Comm. 78:554-559. Darrow, R. A., Crist, D., Evans, H. R., Jones, B. L., Keister, D. L., and Knotts, R. R. (1981). Biochemical and physiological studies on the two glutamine synthetases of Rhizobium, in Current perspectives in nitrogen fixation, A. H. Gibson and H. E. Newton, eds. (Australian Academy of Science: Canberra) pp. 182-185. Drummond, N., Clements, J., Merrick, M., and Dixon, R. (1983). Positive control and autogenous regulation of the nifLA promoter in Klebsiella pneumoniae. Nature 301:302-307. Foor, F., Reuveny, Z. and Magasanik, B. (1980). Regulation of the synthesis of glutamine synthetase by the P11 protein in Klebsiella aerogenes. P.N.A.S. USA 77:2636-2640. Gottesman, S. (1984). Bacterial regulation: Global regulatory networks. Ann. Rev. Genet. 18:415-441. Gussin, G. N., Ronson, C. H., and Ausubel, F. M. (1986). Regulation of nitrogen fixation genes. Ann. Rev. Genet. 20:567-591. Hunt, T. P., and Magasanik, B. (1985). Transcription of glnA by purified Escherichia cali components: Core RNA polymerase and the products of glnF, glnG, and glnL. Proc. Natl. Acad. Sci. USA 82:8453-8457. Jamieson, D. J. and Higgins, C. F. (1986). Two genetically distinct pathways for transcriptional regulation of anaerobic gene expression in Salmonella typhimurium. J. Bacteriol. 168:389-397. Keener, J., and S. Kustu. (1988). Protein kinase and phosphoprotein phosphatase activities of nitrogen regulatory proteins NtrB and NtrC of enteric bacteria: Roles of the conserved amino-terminal domain of NtrC. Proc. Natl. Acad. Sci. USA 85:4976-4980. Kranz, R. G. and Haselkarn, R. (1985). Characterization of nif regulatory genes in Rhodopseudomonas capsulata using lac gene fusions. Gene40:203-215. Kustu, S. G., McFarland, N. C., Hui, S. P., Esman, B., and Ames G. F-L. (1979). Nitrogen control in Salmonella typhimurium: Co- regulation of synthesis of glutamine synthetase and amino acid transport systems. J. Bacteriol. 138:218-234. Ludwig, R. A. (1980). Regulation of Rhizobium nitrogen fixation by the unadenylylated glutamine synthetase I system. P.N.A.S. USA 77:5817-5821. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. 15 Magasanik, B. (1982). Genetic control of nitrogen assimilation in bacteria. Ann. Rev. Genet. 16:135-168. Martin G. B., K. Chapman, and B. K. Chelm. (1988). Role of the Bradyrhizobium japanicum ntrC gene product in differential regulation of the glutamine synthetase II gene, glnII. J. Bacteriol. 170:5452-5459. McFarland, N., McCarter, L., Artz, S., and Kustu, S. (1981). Nitrogen regulatory locus 'glnR' of enteric bacteria is composed of cistrons ntrB and ntrC: Identification of their protein products. P.N.A.S. USA 78:2135-2139. Merrick, H., Hill, S., Hennecke, H., Hahn, M, Dixon, R., and Kennedy, C. (1982). Repressor properties of the nifL gene product in Klebsiella pneumoniae. Mol. Gen. Genet. 185:75-81. Morett, E., and M. Buck. (1988). NifA-dependent in vivo protection demonstrates that the upstream activator sequence of nif promoters is a protein binding site. Proc. Natl. Acad. Sci. USA 85:9401- 9405. Ninfa, A. J., and Magasanik, B. (1986). Covalent modification of the glnG product, NR1, by the glnL product, NR 1’ regulates transcription of the glnALG operon in Escheric 1a coli. Proc. Natl. Acad. Sci. USA 83: 5909-5913. Ninfa, A. J., E. Ninfa, A. Lupas, A. Stock, B. Magasanik, and J. Stock. (1988). Crosstalk between bacterial chemotaxis signal transduction proteins and regulators of transcription of the Ntr regulon: Evidence that nitrogen assimilation and chemotaxis are controlled by a common phosphotransfer mechanism. Proc. Natl. Acad. Sci. USA 85:5492-5496. Nixon, B. T., Ronson, C. H., and Ausubel, F. M. (1986). Two- component regulatory systems responsive to environmental stimuli share strongly conserved domains with the nitrogen assimilation regulatory genes ntrB and ntrC. Proc. Natl. Acad. Sci. USA 83:7850-7854. Nohno, T., and Saito, T. (1987). Two transcriptional start sites found in the promoter region of Escherichia coli glutamine permease operon, glnHPQ. Nuc. Acids Res. 15:2777. Nohno, T., Saito, T., and Hong, J-S. (1986). Cloning and complete nucleotide sequence of the Escherichia coli glutamine permease operon (glnHPQ). Mol. Gen, Genet. 205:260-269. 0w, 0. H., and Ausubel, F. M. (1983). Regulation of nitrogen metabolism genes by nifA gene product in Klebsiella pneumoniae. Nature 301, 307-313. 36. 37. 38. 39. 40. 41. 42. 43. 44. 45. 46. 47. 16 Popham D. L., D. Szeto, J. Keener, and S. Kustu. (1989). Function of a bacterial activator protein that binds to transcriptional enhancers. Science 243:629-635. Raibaud, O. and Schwartz, M. (1984). Positive control of transcription initiation in bacteria. Ann. Rev. Gen. 18:173-206. Reitzer, L. J., and Magasanik, B. (1985). Expression of glnA in Escherichia coli is regulated at tandem promoters. Proc. Natl. Acad. Sci. USA 82, 1979-1983. Reitzer, L. J., and Magasanik, B. (1986). Transcription of glnA in E. cali is stimulated by activator bound to sites far from the promoter. Cell 45, 785-792. Reuveny, 2., Foor, F. and Magasanik, B. (1981). Regulation of glutamine synthetase by regulatory protein P11 in Klebsiella aerogenes mutants lacking adenylyltransferase. J. Bacteriol. 146:740-745. Schlesinger, M. J., Ashburner, H., and Tissieres, A. (1982). Heat shock from Bacteria to Man. Cold Spring Harbor, N.Y. Segal, A., Brown, M. S., and Stadtman, E. R. (1974) Arch. Biochem. Biophys. 161:319-327. Stadtman, E. R., Mura, U., Chock, P. B., Rhee, S. G. (1980). The interconvertible enzyme cascade that regulates glutamine synthetase activity. In: Glutamine: Metabolism, Enzymology, and Regulation, Mora, J., and Palacios, R., eds. Acad. Press. pp. 41-59. Strauch, K. L., Lenk, J. B., Gamble, B. l., and Miller, C. G. (1985). Oxygen regulation in Salmonella typhimurium. J. Bacteriol. 161:673-680. Sundaresan, V., Ow, D. U., and Ausubel, F. M. (1983a). Activation of Klebsiella pneumoniae and Rhizobium melilati nitrogenase promoters by gln (ntr) regulatory proteins. Proc. Natl. Acad. Sci. USA 80:4030-4034. Szeto, H. H., Nixon, B. T. Ronson, C. H. and Ausubel, F. M. (1987). Identification and characterization of the Rhizobium meliloti ntrC gene: R. melilati has separate regulatory pathways for activation of nitrogen fixation genes in free-living and symbiotic cells. J. Bacteriol. 169:1423-1432. Tommassen, J., and Lugtenberg, B. (1982). Pho regulon of Escherichia coli K12: A minireview. Ann. Microbial. (Inst. Pasteur) 133A:243-249. 48. 49. 50. 51. 52. 53. 54. 17 Toukdarian, A., and Kennedy, C. (1986). Regulation of nitrogen metabolism in Azotobacter vinelandii: isolation of ntr and glnA genes and construction of ntr mutants. EMBO J. 5:399-407. Tyler, 8. (1978). Regulation of the assimilation of nitrogen compounds. Ann. Rev. Biochem. 47:1127-1162. Virts, E., S. Stanfield, D. Helinski, and G. Ditta. (1988). Common regulatory elements control symbiotic and microaerobic induction of nifA in Rhizobium meliloti. Proc. Natl. Acad. Sci. USA 85:3062- 3065. Hei, G. R. and Kustu, S. (1981). Glutamine auxotrophs with mutations in a regulatory gene ntrC, that is near glnA. Mol. Gen. Genet. 183:392-399. Hillsky, G. R., Bennet, R. L., Malamy, M. H. (1973). Inorganic phosphate transport in Escherichia coli: involvement of two genes which play a role in alkaline phosphatase regulation. J. Bacteriol. 113:529-539. Ninans, S., P. Ebert, S. Stachel, M. Gordon, and E. Nester. (1986). A gene essential for Agrobacterium virulence is homologous to a family of positive regulatory loci. Proc. Natl. Acad. Sci. USA. 83:8278-8282. Hang, P.-K., Popham, 0., Keener, J., and Kustu, S. (1987). In viva transcription of the nitrogen fixation regulatory operon nifLA of Klebsiella pneumoniae. J. Bacteriol. 169:2876-2880. Chapter 1 Physiological roles of the two isoforms of glutamine synthetase in ammonia assimilation and symbiotic nitrogen fixation in Bradyrhizobium japonicum INTRODUCTION During free-living growth Bradyrhizobium japanicum, the endosymbiont of soybean, assimilates ammonia primarily by the coordinate activity of glutamine synthetase (GS) and glutamate synthase (3, 23). However, in bacteroids, the symbiotic farm of these bacteria, GS activity decreases in concert with the derepression of nitrogenase activity (3). The physiological role of GS during free-living growth, incipient nodule formation and in mature bacteroids is therefore an integral part of the developmental process in B. japonicum. Bacteria of the Rhizobiaceae family, including B. japonicum, are unusual in that most members contain two GS enzymes GSI and GSII encoded by glnA and glnII, respectively (6, 12, 8, 9). The B. japonicum glnA gene has sequence similarity to the E. coli glnA gene and is not appreciably nitrogen-regulated (4). The activity of GSI is post- translationally regulated by reversible adenylylation (8). The CS encoded by glnII has sequence similarity with eucaryotic GS’s and it has ' been suggested that GSII was acquired from plants (5). GSII is not known to be post-translationally modified, but is instead transcriptionally regulated in response to nitrogen and carbon source (6, 9, 16) and oxygen concentration (20). 18 19 Because most members of the Rhizobiaceae can associate with plants it has been speculated that there may be a unique symbiotic role for one or both of the two glutamine synthetases found in rhizobial species (7, 9). Previously the lack of isolated genes has made it difficult to make well-defined mutations at these two loci and thereby rigorously test this hypothesis. I have used gene-directed mutagenesis to develop single glnA and glnII mutant strains and a double glnA/glnII mutant strain. I show here that glnA or glnII alone is sufficient to provide glutamine prototrophy while the double mutant glnA/glnII is a glutamine auxotroph. Both single mutants form effective symbioses with soybean. The double mutant, however, is unable to form nodules on soybean. MATERIALS AND METHODS Bacterial strains, plasmids, media and growth conditions. BJllOd is a small-colony derivative of B. japanicum 311b110 (13). All other bacterial strains and recombinant plasmids are listed in Table 1. To check growth on various nitrogen sources, 50 ml cultures of a mineral salts base medium adapted from O’Gara and Shanmugam (18) was used which had the following composition per liter:. KHZPO4, 0.3 g; NaZHPO4, 0.3 g; MgSO4.7H20, 0.12 9. Trace elements were added (18) and the pH was adjusted to 6.8. After autoclaving, 1 ml of a filter- sterilized vitamin solution (13), 0.05 g CaClz and 7.5 ml of filter- sterilized 20% (wt/vol) D-xylose were added per liter. The various nitrogen sources were added from filter-sterilized stock solutions at the concentrations indicated in Table 2. 20 DNA biochemistry. Southern hybridizations and isolation of plasmid DNA were carried out as described previously (1). Total genomic DNA from B. japonicum was purified by phenol extraction (17). The isolation of glnA and glnII has been described (4, 5). Gene-directed mutagenesis. Recombinant plasmids were manipulated in vitro and in E. coli to construct mutant alleles of the glnA and glnII loci (Table 1, Figure l). The mutant allele of the glnA gene was constructed by deletion of the EcoRI to BamHI region indicated in Figure 1 and replacement of this region with the HindIII Omega fragment from the plasmid pHC45 (19) to which EcoRI-HindIII and HindIII-BamHI adaptors (from pKC7; [21]) had been added. This fragment confers spectinomycin resistance upon B. japonicum. The mutant allele of the glnII gene was constructed by deletion of the XhoI fragment indicated in Figure 1 and replacement of this region with a SalI fragment containing the nptII gene from plasmid pRL161 (25). This fragment confers kanamycin resistance upon B. japonicum. The plasmids carrying the mutant alleles were conjugally transferred from E. coli to B. japonicum by methods previously described (13). Stable B. japonicum recombinants were selected by antibiotic resistance conferred by the constructed allele, and those in which the mutant allele had replaced the wild-type allele were screened by hybridization as described previously (13). All of the constructed genotypes were confirmed by Southern hybridization analyses of genomic DNA. In all of the steps of strain construction, the growth medium was supplemented with 10 mM glutamine. In the case of the double mutations, the confirmed glnII mutant strain, BJ2513, was 21 Table I. Bacteria strains and plasmids. Strain or Description Source or plasmid reference BJIIOd Hild type (13) BJ2841 glnA::spc; glnA mutant, Spcr This work BJ2513 glnII::nptII; glnII mutant, Kanr This work GM3-21 glnA::spc, glnII::nptII; This work glnA/glnII mutant, Spc', Kanr Hamid; pBJ196 Apr; glnII locus cloned into pBR322 (5) pBJSBA Apr; glnA locus cloned into pUC8 (4) pBJZSl Kan”; glnII deletion allele cloned This work into pBR322 pBJ284 Spc'; glnA deletion allele cloned This work into pKC7 pOS4lOl Apr; derivative of ColK (25) pHC45 Spc', source of Spc cassette (21) pRL161 Kan”, Source of nptII cassette P. Holk and J. Elhai (unpublished) pBR322 Apr, Tc': vector (2) pKC7 Kmr, Apr; vector (21) 22 processed through a second mating round using an E. coli strain carrying the plasmids pBJ284 and pDS4101 (see Table 1) to mutate the glnA gene. Plant inoculation and growth conditions. Soybean seeds (Glycine max [L.] Merr. cv. Amsoy 71) were sterilized as described previously (13) and then soaked for 1 hr in a late log phase culture of the appropriate strain of B. japonicum. Seeds were planted in modified Leonard jars as described by Vincent (24) using the nitrogen-free medium of Johnson et al. (15). The plants were grown with a 16 hr daily light period, day/night temperatures of 28/25°C, a relative humidity of 60%, and a photon fluence rate of 300 umol'm‘z °sec'1. All plants were harvested at 28 days and the nodules were removed, rinsed with H20, counted, weighed and then used in acetylene reduction assays (13, 14). RESULTS AND DISCUSSION Construction of B. japonicum glnA, glnII and glnA/glnII mutants. Since most members of the family Rhizobiaceae contain two forms of glutamine synthetase the function of these two enzymes has been a matter of speculation (e.g. 7, 8, 9, 16). In order to determine if both enzymes are capable of fulfilling the requirement for the biosynthesis of glutamine, we used the isolated glnA and glnII genes to construct mutant 8. japanicum strains that were missing either one, or both of the enzymes 651 and G511 (see Materials and Methods; Figure l). The resulting three derivatives of the wildtype strain BJllO were BJ2841, glnA; BJ2513, glnII; and GM3-21, glnA/glnII (Table 1). 23 .Amv omamuwc=F mew magma :aon Lee mcmaosoca ash .czmcu mm agape o» aka. soc; magma :uon com =o_aawcomcmeu ;»_z mvmmzzoccm acpcwmucco mcopmmg cmxon may an umumuwucw wee magma one Co meowmmg mcpeoo as» .Axv Hoax new .sz HHfiuemz .Amv ~moom .Amv -mw .Amv ~=smm "mch—oe ma cmamo_c:w mmu_m omemposccucm ;u_3 czogm wee m=o_mmc mcmm -=~u can <=~u as“ me mama cowuo_eummg ash .me_o_~m ucmuss use enauvp_z -=~u use <=~m as» no mama —mowma;a .~ wasmmm 24 .H mczmwd .anOOm. m w xux m 7 t _ .2 _ 2.3 R A A __ :53 x. 7.. \ I \ I I \ I \\ Ir . 23.3 _ 3 :3: m m mx 3 mm x m 3 mm m m r. _ __ _ P __ _ _ _ _ 0:3 Em A A C 2 <5... \. a \\ \\ ’I/A/ _vmwam fir _ _ 4 0am m I z m 535 £56.. 25 Growth of B. japanicum glnA, glnII and glnA/glnII mutant strains on various nitrogen sources. The growth properties of the glutamine synthetase deficient strains are summarized in Table 2. Both the glnA (BJ2841) and glnII (BJ2513) single mutant strains were glutamine protatrophs and were able to utilize nitrate, and ammonia as sole nitrogen sources. The glnA/glnII double mutant strain (GM3-21) was unable to grow an ammonium or nitrate as sole nitrogen sources and could be rescued by exogenous glutamine indicating glutamine auxotrophy (Table 2). The growth of GM3-21 in the presence of exogenous glutamine was not as good as for a glutamine prototroph indicating that there might be an additional defect, possibly in the transport of glutamine. This decreased growth was not investigated further. Symbiotic properties of B. japanicum glnA. glnII and glnA/glnII mutant strains. Both the glnA (BJ2841) and glnII (BJ2513) strains were able to effectively nodulate soybeans but surprisingly they induced greater numbers of nodules per plant than the wildtype (BJllOd) parental strain (Table 3). The reason for this hypernodulation is unknown. Higher levels of nitrogen fixation ability as assayed by acetylene reduction activity were also found for the plants inoculated with these two mutant strains. The glnA glnII double mutant (GM3-21) was unable to nodulate soybeans (Table 3). He found that the addition of glutamine to the potting medium resulted in the partial rescue of nodulation by the double mutant, yielding small white Fix' nodules. Unless there are polar effects of the constructed alleles on genes which are as yet uncharacterized, these results taken together indicate that GSI and GSII 26 Table 2. Growth properties of B. japanicum gln mutants.a Strain Nitrogen Sourceb Glutamine Ammonium Nitrate BJ110d (wt) ++ ++ ++ BJ2841 (glnA) ++ ++ ++ BJ2513 (glnII) ++ ++ ++ GM3-21 (glnA/glnII) + - - aGrowth was determined in minimal media (18), both in liquid and on plates, and is indicated as follows: ++, very good; +, good; -, none. bBoth glutamine and ammonium chloride were tested at 10 mM. Potassium nitrate was tested at a variety of concentrations within the range of 0.05 to 2 mM; the results were the same at all concentrations. 27 Table 3. Symbiotic phenotypes B. japonicum gln mutants.a Strain Number of nodulesa Acetylene reductionb (Total wet weight/ (umol of C2H4/hour plant, 9.) per plant root system) BJIIO (wt) 74.0 i 7.8 (1.61 i 0.28) 24.0 i 3.1 BJ2841 (glnA) 90.3 i 14.2 (1.51 i 0.24) 40.6 i 1.5 BJ2513 (glnII) 116.0 1 11.4 (1.82 i 0.3) 36.8 t 2.1 GM3-21 (glnA/glnII) 0 0 aThe number of nodules per plant and, in parentheses, the total wet weight of nodules per plant in grams, respectively, are indicated. The numbers represent the mean +/- SEM of 3 nodulation tests each. bNitrogen fixation activity was measured by the acetylene reduction assay as described previously (13, 14). Activity is reported as umoles of ethylene produced per hour per plant root system. The numbers represent the mean +/- SEM of the analyses of three plants each. 28 are the only two glutamine synthetases in B. japonicum which are active in glutamine biosynthesis under standard laboratory conditions and in symbiotic cells. Recently there have been reports of glutamine auxotrophs in other rhizobial species: Agrobacterium tumefaciens (22), Azorhizobium caulinodans (10), and Rhizobium meliloti (11). In each of these cases the isolated genes were used, as in this work, to make defined mutations in the gene(s) for glutamine synthetase. The A. tumefaciens and A. caulinodans glutamine auxotrophs were both found to be defective in association with their host plants (10, 22). A R. meliloti glnA/glnII mutant, however, was a glutamine auxotroph but still formed wild-type nodules on alfalfa plants (11). A possible explanation for these contrasting results is that the plant partners in the associations may differ in their ability to provide glutamine to their bacterial symbiont. 29 LIST OF REFERENCES 1. 10. Adams, T.H., and B.K. Chelm. 1984. The nifH and nifDK promoter regions from Rhizobium japanicum share structural homologies with each other and with nitrogen-regulated promoters from other organisms. J. Mol. Appl. Genet. 2:392-405. Bolivar F, Rodrigues RL, Greene PJ, Betlach H, Heynecker HL, Boyer HH, Crossa JH, Falkow S (1977) Construction and characterization of new cloning vehicles. A multiple cloning system. Gene 2:95- 100 Brown CM, Dilworth MJ (1975) Ammonia assimilation by Rhizobium cultures and bacteroids. J Gen Microbiol 86:39-48 Carlson, T.A., M. L. Guerinot, and B.K. Chelm. 1985. Characterization of the gene encoding glutamine synthetase I (glnA) from Bradyrhizobium japanicum. J. Bacteriol. 162:698-703. Carlson, T.A. and B.K. Chelm. 1986. Apparent eucaryotic origin of glutamine synthetase II from the bacterium Bradyrhizobium japanicum. Nature 322:568-570. Carlson TA, Martin GB, Chelm BK (1987) Differential transcription of the two glutamine synthetase genes of Bradyrhizobium japanicum. J Bacteriol 169:5861-5866 Darrow, R.A. 1980. Role of glutamine synthetase in nitrogen fixation, p. 139-166. In: J. Mara and R. Palacios (ed.), Glutamine synthetase: metabolism, enzymology. and regulation. Acad. Press, Inc. New York. Darrow, R.A. and R.R. Knotts. 1977. Two forms of glutamine synthetase in free-living root-nodule bacteria. Biochem. Bi0phys. Res. Comm. 78:554-559. Darrow, R.A., D. Crist, R.R. Evans, B.L. Jones, D.L. Keister and R.R. Knotts. 1981. Biochemical and physiological studies on the two glutamine synthetases of Rhizobium. In Current perspectives in nitrogen fixation, A.H. Gibson and H.E. Newton, eds. (Australian Academy of Science: Canberra) pp. 182-185. de Bruijn FJ, Pawlowski K, Ratet P, Hilgert U, Hang CH, Schneider M, Meyer H, Schell J (1988) Molecular genetics of nitrogen fixation by Azorhizobium caulinodans 0R5571, the diazotrophic stem-nodulating symbiont of Sesbania rostrata. In: Bathe H, de Bruijn FJ, Newton HE (eds) Nitrogen fixation: Hundred years after. Gustav Fischer, Stuttgart, pp 351-355 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 30 de BruiJn FJ, Rossbach S, Schneider M, Ratet P, Messmer S, Szeto NH, Ausubel FM, Schell J (1989) Rhizobium meliloti 1021 has three differentially regulated loci involved in glutamine biosynthesis, none of which is essential for symbiotic nitrogen fixation. J Bacteriol 171: (in press) Fuchs, R. L. and O. L. Keister. 1980. Comparative properties of glutamine synthetases I and II in Rhizobium and Agrobacterium spp. J. Bacteriol. 144:641-648. Guerinot, M.L. and B.K. Chelm. 1986. Bacterial -aminolevulinic acid synthase activity is not essential for leghemoglobin formation in the soybean / Bradyrhizobium japanicum symbiosis. Proc. Natl. Acad. Sci. USA 83:1837-1841. Hardy RHF, Holsten RD, Jackson EK, Burns RC (1968) The acetylene- ethylene assay for N fixation: laboratory and field evaluation. Plant Physiol 43:118 -1207 Johnson GV, Evans HJ, Ching TM (1966) Enzymes of the glyoxylate cycle in Rhizobia and nodules of legumes. Plant Physiol 41:1330- 1336 Ludwig, R.A. 1980. Physiological roles of glutamine synthetases I and II in ammonium assimilation in Rhizobium sp. 32H1. J. Bacteriol. 141:1209-1216. Maniatis T, Fritsch EF, Sambrook J (1982) Molecular cloning. Cold Spring Harbor Laboratory, Cold Spring Harbor, New York O’Gara, F. and K.T. Shanmugan. 1976. Regulation of nitrogen fixation by Rhizobia: export of fixed N2 as NH4. Biochim. Biophys. Acta 437:313-321. Prentki, P. and H.M. Krisch. 1984. In vitro insertional mutagenesis with a selectable DNA fragment. Gene 29:303-313. Rao, V.R., R.A. Darrow and D.L. Deister. 1978. Effect of oxygen tension on nitrogenase and on glutamine synthetases I and II in Rhizobium japanicum 61A76. Biochem. Biophys. Res. Comm. 81:224- 231. Rao RN, Rogers 56 (1979) Plasmid pKC7: a vector containing ten restriction endonuclease sites suitable for cloning DNA segments. Gene 7:79-82 Rossbach S, Schell J, de Bruin FJ (1987) The ntrC gene of Agrobacterium tumefaciens C58 controls glutamine synthetase (GSII) activity, growth on nitrate and chromosomal but not Ti-encoded arginine catabolism pathways. Mol Gen Genet 209:419-426 31 23. Vairinhos F, Bhandari B, Nicholas 000 (1983) Glutamine synthetase, glutamate synthase and glutamate dehydrogenase in Rhizobium japanicum strains grown in cultures and in bacteroids from root nodules of Glycine max. Planta 159:207-215 24. Vincent, J.M. 1978. Factors controlling the legume-Rhizobium symbiosis. In Nitrogen Fixation Volume II, H.E. Newton and H.H. Orme-Johnson, eds. (University Park Press: Baltimore, MD) pp. 103- 129. 25. Walk CP, Vonshak A, Kehoe P, Elhai J (1984) Construction of shuttle vectors capable of conjugative transfer from Escherichia cali to nitrogen-fixing filamentous cyanobacteria. Proc Natl Acad Sci USA 81:1561-1565 Chapter 2 Role of the Bradyrhizobium japanicum ntrC Gene Product in Differential Regulation of the Glutamine Synthetase II gene, glnII. SUMMARY The ntrC gene was isolated from Bradyrhizobium japanicum, the endosymbiont of soybean (Glycine max), and its role in regulating nitrogen assimilation was examined. Two independent ntrC mutants were constructed by gene replacement techniques. One mutant was unable to produce NtrC protein, while the other constitutively produced a stable, truncated NtrC protein that lacked a postulated DNA-binding region. Both ntrC mutants were unable to utilize potassium nitrate as a sole nitrogen source. In contrast to wild-type B. japanicum, the NtrC null mutant lacked glnII transcript in aerobic, nitrogen-starved cultures. However, the truncated-NtrC mutant expressed glnII under both nitrogen- starved and nitrogen-excess cultures. Both mutants expressed glnII under oxygen-limited culture conditions and in symbiotic cells. These results suggest that nitrogen assimilation in B. japanicum is regulated in response to both nitrogen limitation and oxygen limitation and that separate regulatory networks exist in free-living and symbiotic cells. INTRODUCTION Free-living Bradyrhizobium japanicum cells assimilate ammonia primarily by the coordinate activity of glutamine synthetase (GS) and glutamate synthase (6, 51). However, in bacteroids, the symbiotic form 32 33 of these bacteria, GS activity decreases in concert with the derepression of nitrogenase activity (6, 50). The regulation of nitrogen assimilation pathways is therefore an integral part of the bacterial developmental process and the symbiotic interaction. In the Enterobacteriaceae, genes involved in nitrogen metabolism are coordinately regulated by the Ntr system (22, 30, references below).- Genes regulated by the Ntr system have a unique promoter that bears no homology to the canonical Escherichia coli promoter (15, 37). These genes also have protein-binding sequences that display dyad symmetry (4). Transcriptional activation of Ntr promoters requires RNA polymerase core, an alternative sigma factor (54) which is the product of the ntrA gene (NtrA), and the regulatory protein NtrC encoded by the ntrC (glnG) gene (24, 25, 33). NtrC has sequence-specific DNA binding activity (4, 42, 49) that has been implicated in transcriptional activation (42), repression (23, 49) and antitermination (4). NtrC can be phosphorylated by the ntrB gene product which converts NtrC to its promoter activating form (34). In the Enterobacteriaceae, glnA, the only gene encoding glutamine synthetase, is expressed from two promoters (16, 41). Expression from the upstream promoter, glnApl (an E. coli consensus-type promoter) occurs under conditions of carbon deficiency and nitrogen excess and is repressed by NtrC (16, 41). Expression from the downstream promoter, glnApZ, requires NtrC and occurs under conditions of nitrogen limitation (16, 41). Bacteria of the Rhizobiaceae, including B. japanicum, are distinct from the Enterobactericeae in that most members contain two glutamine synthetase enzymes, GSI and GSII, encoded by glnA and glnII respectively (7, 8, 12, 34 13). Interestingly, the dual promoter control of glnA seen in the Enterobactericeae is replaced in B. japanicum with two genes coding for GS which are differentially regulated (9). The B. japanicum glnA gene is homologous to the E. coli glnA gene and is not appreciably nitrogen regulated (8, 9). In contrast, the GS encoded by glnII shares homology with eucaryotic glutamine synthetases, responds to nitrogen availability, has an Ntr-type promoter, and has sequence elements upstream that are homologous to the Enterobacteriaceae NtrC binding sequence (7, 9, T.A. Carlson, Ph.D. dissertation, Michigan State University, 1986). Several lines of evidence suggest the presence of an Ntr system in the Rhizobiaceae (35, 43, 45). The isolation of ntrC genes has been reported in Rhizobium meliloti (45), Bradyrhizobium paraspaniae (35), Agrobacterium tumefaciens (43), and Azorhizobium caulinodans (38). A. caulinodans ntrC mutants are affected in histidine and arginine utilization, in free-living nitrogen fixation and in their ability to form Fix+ nodules (38). A. tumefaciens ntrC mutants are unable to grow on nitrate, lack GSII activity, and are unable to use arginine unless the Ti-encoded arginine catabolism pathway is induced (43). In R. meliloti, ntrC is required for growth on nitrate and for the ex planta, but not symbiotic, transcription of several nif genes and the fixABC operon (45). However, neither R. meliloti or A. tumefaciens ntrC mutants are affected in their ability to interact normally with their plant hosts (43, 45). Because of the differential regulation of nif genes ex planta and symbiotically, Szeto et al. (45) postulated a dual regulatory system governing nitrogen utilization in the Rhizobiaceae. 35 This model posits that during aerobic nitrogen-limited growth, NtrC activates the transcription of genes required for nitrogen utilization but that during symbiotic growth another activator, the nifA gene product, regulates the expression of these genes. In this chapter I show that the B. japanicum ntrC gene product is required for the transcriptional regulation of glnII under conditions of nitrogen starved aerobic growth. Under microaerobic conditions and during symbiosis, however, glnII is expressed but does not require the ntrC gene product. Therefore glnII appears to be under the control of a dual regulatory system responding to nitrogen starvation and oxygen limitation. MATERIALS AND METHODS Bacterial strains, plasmids, media and growth conditions. BJ110d is a small-colony derivative of B. japanicum 311b110 isolated as described (21). All other bacterial strains and recombinant plasmids are listed in Table l. The complex media were YEM (0.2%.yeast extract, 1% mannitol, 3 mM KZHPO4, 0.8 mM M9504, 1.1 mM NaCl) and YEMAN (YEM with 10 mM NH4Cl, 0.5 mM KNO3). To check growth on various nitrogen sources, a mineral salts base medium adapted from O’Gara and Shanmugan (36) was used which had the following composition per liter: KHZPO4, 0.3 g; NazHP04, 0.3 g; MgS04.7H20, 0.12 9. Trace elements were added (36) and the pH was adjusted to 6.8. After autoclaving, 1 ml of a filter-sterilized vitamin soluion (21), 0.05 g CaClz and 7.5 ml of filter-sterilized 20% (wt/vol) O-xylose were added per liter. For growth observations, the various 36 Table 1. Bacterial strains and plasmids continued . . . ntrC fragment cloned into the BglII site of pIC19R (32) Strain or plasmid Description Source or . reference 3. iangnigum strains BJ110d Hild-type (21) BJ27147 ntrC::nptII, null mutant; Kmr This study BJ3028 ntrC::nptII, truncated NtrC; Kmr This study Plasmids (phages) M13mp18 (54) M13glnA 391 bp SalI fragment from (2) B. japanicum glnA promoter (8) region cloned into M13mp18 MIBglnII 2.1 kbp SalI fragment from (2) B. japanicum glnII promoter (7) region cloned into M13mp18 pBN386 Apr; 1.7 kbp EsaRI fragment 8. T. Nixon containing B. parasponiae ntrC in pSP64CS (18) pRJcos4-84 Tc"; cosmid clone (in pLAFRI [20]) This study containing 8. japanicum ntrC DNA pBJ262 Apr; 4.6 kbp 39711 a. japanicum This study 37 Table 1 (continued) pBJ312 Apr; 5.1 kbp XhoI B. japanicum This study ntrBC fragment cloned into the XhoI site of pIC19R (32) pBJ271 Apr, Kmr; 1.7 kbp SalI B. japanicum This study ntrC fragment cloned into the XhoI site and 1.3 kbp EcaRI B. japanicum ntrC downstream fragment cloned into the EcaRI site of pKC7 (39) pBJ302 Apr, Km”; 2.7 kbp B. japanicum ntrC This study fragment cloned into the EcoRI site of pBR322-S (see below). nptII gene contained in cassette CK.1 (19) inserted into the SalI site of ntrC pBR322-S pBR322 with the SalI site removed J. S.-Craig pDS4101 Ap"; derivative of 001K (53) pRK2013 Km”; carries conjugal transfer (14) genes from RK2 38 nitrogen sources were added at 0.5 mM from filter-sterilized stock solutions. For analyzing gene expression in aerobic cultures the mineral salts basal medium was used with either 10 mM glutamate added (XG) or 10 mM glutamate plus 10 mM NH4Cl (XGA). For microaerobic growth experiments ten-liter cultures of B. japanicum were grown in Microferm fermentors (New Brunswick Scientific) agitated at 200 r.p.m. and sparged with 0.2% 02 and 99.8% N2 at a rate of 500 ml min-l. Gas flow rate was controlled using thermal mass flowmeters (Brooks Instruments, model 5850 C). Each 10 liter culture was inoculated with 50 ml of a stationary-phase aerobic culture in YEH medium. All 8. japanicum cultures were grown at 30°C. DNA biochemistry. Southern hybridizations, colony hybridizations and isolation of plasmid DNA were carried out as described (3). Total genomic DNA from B. japanicum was purified by phenol extraction (31). A B. japanicum cosmid library constructed in pLAFRl (20) and described by Adams et al. (3) was screened using a 32P-labeled 1.7 kbp EcoRI fragment purified from the recombinant plasmid pBN386 (Table 1; pBN386 was generously provided by B. T. Nixon). This 1.7 kbp EcoRI fragment contains 1.4 kbp of the B. parasponiae ntrC gene (35). Gene-directed mutagenesis. Recombinant plasmids were manipulated in vitro and in E. cali to construct mutant alleles of ntrC (Table 1). The plasmid (pBJ302) used in the construction of BJ3028 was formed by first inserting the 2.7 kbp EcoRI fragment from pBJ262 into the EcoRI site of pBR322-S (Table 1). Into the SalI site of ntrC in this plasmid 39 was inserted a 1.35 kbp SalI fragment containing the nptII gene from Tn5. This nptII-containing SalI fragment was obtained by inserting the cassette C.K1 (19) with BamHI ends into the BamHI site of L.HEH1 (19) and then removing the cassette as a SalI fragment. The plasmid (pBJ271) used in the construction of BJ27147 was formed by inserting the 1.7 kbp SalI fragment from pBJ312 into the XhoI site of plasmid pKC7 (39). A 1.3 kbp EcoRI fragment from downstream of ntrC was inserted into the EcoRI site of the pKC7 recombinant plasmid but this fragment was not involved in the homologous recombination event resulting in BJ27147. Each mutant plasmid was transformed into H8101 carrying pDS4101 for use as the donor strain in a conjugal mating with B. japanicum. The mating procedure was as described previously (21) using B. japanicum 110d as the recipient, HBlOl(pRK2013), and either HBlOl(pBJ302, pDS4101) or HBlOl(pBJ271, pDS4101) as the donor. The cells were spread on YEM plates supplemented with 10 mM glutamine. After incubation for 4 days at 30°C, the cells were removed from the plates by suspension in 10 ml 0.01% Tween 80 and 0.2 ml was plated on YEM plates containing kanamycin (150 ug/ml) and chloramphenicol (30 ug/ml). Colonies from the selection plates were screened by colony hybridization. Southern hybridizations of restriction endonuclease digested genomic DNA were used to confirm the constructions. RNA biochemistry. B. japanicum cells (1 g) were suspended in 10 ml of 4.0% sarkosyl, 0.1 M Tris-HCl, pH 8.0 and passed two times through a French pressure cell (Aminco; Silver Spring, MD) at 12,000 psi. For 40 each milliliter of broken cell solution 1 g CsCl was added. RNA was isolated by discontinuous CsCl gradient centrifugation as described previously (3, 10). The RNA pellet was drained thoroughly, redissolved in H20 and then extracted twice with phenol and four times with diethyl ether. Following ethanol precipitation, the RNA was resuspended in sterile H20 and stored in aliquots at -70°C. Quantitative SI nuclease protection assays. Levels of transcript were quantitated by the SI nuclease protection method adapted from Berk and Sharp (5) and described previously (2) in which both glnA and glnII mRNAs could be quantified in a single reaction. Five micrograms of total cellular RNA was used per lane. Single-stranded 5’ end-labelled probes were synthesized by primer extension (2) using oligonucleotide primers described previously (2). The recombinant M13 phages used in synthesizing the single-stranded DNA hybridization probes are described in Table 1. The partially protected fragments from S1 analyses with each of the probes are 125 (glnA) and 170 (glnII) nucleotides long. Hybridization, SI digestion, and identification of partially protected products were as described previously (2). Immunoblotting. B. japanicum cells were harvested by centrifugation then resuspended in 10% sucrose, 20 mM Tris-HCl pH 7.5. Lysates were prepared by passing the cells twice through a French pressure cell at 16,000 psi. Cell debris was pelleted by centrifugation and discarded. Proteins in the cleared lysate were precipitated by adding 2 vol of saturated ammonium sulfate, chilling on ice 30 min and centrifuging for 41 10 min at 12,000 g. The proteins were separated by SOS-PAGE (27) and then transferred to cellulose nitrate paper (47). The immunoblot development procedure was a modification of the method of Tsang et al. (48) as described by Darr et al. (11) except for the following changes: 3% non-fat dry milk was used instead of BSA as the blocking agent, the concentration of Tween-20 in the wash buffer was 0.05% (vol/vol), and the bound immunoglobulin was detected using a l ug/ml solution of Staphylococcus aureus protein A-alkaline phosphatase conjugate (Sigma). The method of Leary et al. was used for the color development (28). Polyclonal antiserum prepared against the E. coli ntrC (glnG) gene product (40) was generously provided by L. Reitzer and B. Magasanik. Plant inoculation, growth conditions and nodule bacteria isolation. Soybean seeds (Glycine max [L.] Merr. cv. Amsoy 71) were sterilized as described previously (21) and then soaked for 1 hr in a late log phase culture of the appropriate strain of B. japanicum. Seeds were planted in modified Leonard jars as described by Vincent (52) using the nitrogen-free medium of Johnson et al. (26). The plants were grown with a 16 hr daily light period, day/night temperatures of 28/25°C, a relative humidity of 60%, and a photon fluence rate of 300 umol.m-2.sec- 1. All plants were harvested at 28 days and the nodules were removed, rinsed with H20, counted, weighed and either used in acetylene reduction assays or immediately frozen at -70°C for later RNA purification. B. japanicum cells were isolated from soybean nodules as described previously (1). The cells were used to isolate RNA as described above. 42 RESULTS Isolation of the B. japanicum ntrC gene. To determine whether B. japanicum contains an ntrC-like gene, total BJ110d DNA was examined by Southern hybridization using a 32P-labeled 1.7 kbp EcoRI fragment from pBN386 (this EcoRI fragment contains 1.4 kbp of the B. parasponiae ntrC gene; 35). A strong hybridization signal was detected at 5.1 kbp when BJ110d DNA was digested with XhoI (Figure 1, lane 1). The probe was therefore used to screen a B. japanicum cosmid library described previously (3). Five cosmids were identified that contained ntrC- homologous sequences (pRJcos4-84, pRJcosS-90, pRJcos9-9, pRJcole-62, and pRJcosll-89). These cosmids were found to share a common 5.1 kbp XhoI fragment that was subsequently determined to span the entire ntrBC operon. One cosmid, pRJcos4-84, was used to subclone a 4.6 kbp BglII fragment (pBJ262) and the 5.1 kbp XhoI (pBJ312) for use in restriction mapping and in the construction of mutant alleles. A partial restriction map of the ntrC-homologous region in BJ110d is shown in Figure 2. In E. coli, Klebsiella pneumoniae, R. meliloti, and B. parasponiae the nitrogen regulatory gene responsible for NtrC phosphorylation, ntrB, is situated directly upstream of the ntrC gene (29, 30, 35, 45). To determine whether an ntrB lies upstream of the B. japanicum ntrC gene a partial DNA sequence was determined for the ntrC-homologous region and a region one kilobase upstream of the presumed ntrC gene (data not shown). The 400 bp sequenced from the B. japanicum ntrC region was 86% homologous to B. parasponiae ntrC (35) and 61% homologous to R. meliloti ntrC (45) at the nucleic acid level. The 350 bp sequenced from the 43 Figure 1. Hybridization of B. japanicum genomic DNA to B. parasponiae ntrC DNA. Left: Ethidium bromide stained 0.7% agarose gel of XhoI digested total DNA. Right: Autoradiogram obtained after Southern transfer and hybridization with the 1.7 kbp EcoRI fragment from pBN386 corresponding to ntrC (35). Lanes: 1, BJ110d (wild-type); 2, BJ27147 (ntrC mutant); and 3, BJ3028 (ntrC mutant). 44 2 13 1 1 2 3 9kbp -—12. 5 -SJ 6. _ Figure 1. 45 region upstream of the B. japanicum ntrC was 87% homologous to B. parasponiae ntrB (35) and 75% homologous to the comparable region in R. meliloti ntrB (45). These data along with several conserved restriction sites with B. parasponiae confirmed the presence of an upstream ntrB and allowed us to localize the B. japanicum ntrBC operon within the isolated sequence (Figure 2). Generation of B. japanicum ntrC mutants. The recombinant plasmids pBJ271 and pBJ302 (see Materials and Methods) were conjugated into wild- type B. japanicum (BJ110d) and recombined into the BJ110d genome. Two independent mutants, 8027147 and BJ3028, were isolated and confirmed by Southern hybridization (Figure 1, lanes 2 and 3; Figure 2). The mutant BJ3028 was formed as the result of a double crossover event such that the ntrC genomic region in this mutant now carries the nptII gene inserted into a SalI site in the opposite orientation to the ntrC gene (Figure 2). i The mutant BJ27147 was formed as the result of a single crossover event in the SalI 1.7 kbp fragment that was used in the construction of the recombinant plasmid pBJ271. Since this SalI fragment is completely internal to the ntrBC operon the single crossover event resulted in the partial duplication of the operon (Figure 2). BJ27147 thus carries a copy of the operon that has a 3’-truncated ntrC gene and a copy that has a 5’-truncated ntrB gene. Since the B. japanicum ntrC is probably transcribed from a promoter located upstream of ntrB, I expected little, if any, NtrC protein to be produced in this mutant (see next section). 46 .Eeex .x m_~em .m ._ssz .z m_mee3 .3 m__~mm .3 “mee_m ammo—uacoucm co_uu_cammm .Amuogumz use mpmpemumz mmmv _-wma co cm_eemo we: “as“ Nuxg mo anx-_.¢ macmmmsawc ~¢_N~am me 335 may =_ :owmmc umguums-mmoeu esp .mmomam new ~e_-am mcwmcum Lou edema—uc. 3. mean -uq= egg mo co_umu=mpco use =o_»moop =o_aemm=_ ash .umop cu u;m_e scam mp copuawcomcmeu use cmamo_c:¢ mew «mean we»: use me»: egg mo m=o_umuop asp .mwomam use measuem «names: we“: as» can ecumam waaatuppz c. covmoe one»: szomcoq~w .m as» he nee pmoFmAEg .N egzmpm 47 "In“ .N mczmwu N ~ 0:: ~ 02: n ~92. 0:: m - . _ _ . _ .a. .:a: magnnl. . . _ a _ m m z x3. .3 u z m z x m naOOm :2: ~ 0:: e. ..l. ... wt, _ a _ w. _ _ . _ x .. m m .. m z x m e a t .11 L . _ . . . 1, .2e: . _ _ . x a mu m z w m z x m E... J 0:: . _ ...“ NVP hmfim owpfim mwomfim 48 Detection of the NtrC protein by immunoblotting. Since both BJ27147 and BJ3028 carried an insertion near the 3’ end of the ntrC gene it was possible that a truncated NtrC protein might still be produced. To investigate this possibility an antiserum raised against the E. coli NRI (NtrC) protein was used in immunoblot experiments. Extracts were prepared from cells grown in aerobic minimal media cultures that were either nitrogen-starved (XG) or nitrogen-surfeited (XGA). the proteins were separated by SOS-PAGE, and transferred to cellulose nitrate paper. In BJ110d (wildtype) a band that migrated with an apparent molecular weight of 59 kd reacted with the NtrC antiserum (Figure 3, lanes 5 and 6). (B. parasponiae NtrC is 53 kd based on DNA sequence, [35]). No strongly reacting protein to the NtrC antiserum was detected in the mutant BJ27147 (Figure 3, lanes 3 and 4). The mutant BJ3028 produced a protein that reacted with the E. coli NtrC antiserum but this protein had a smaller apparent molecular weight (57 kd) than that of wild-type B. japanicum (Figure 3, lanes 1 and 2). Although both 8027147 and BJ3028 carried insertions at the same point in the ntrC gene, the nptII cassettes were from different origins and thus the cassette sequence directly adjoining the ntrC sequence was different in each mutant (Figure 2). Apparently this difference is responsible for the formation of a stable truncated protein in one mutant (BJ3028) but not in the other (BJ27147). Finally, a band at approximately 56 kd was detected in all strains and thus did not appear to be associated with the B. japanicum ntrC locus reported here. Measurements of the intracellular concentration of NtrC protein in E. coli using immunological techniques have revealed 70 molecules of NtrC 49 Figure 3. Immunoblot analysis of the wild type and the two ntrC mutants. Samples of cell extracts from B. japanicum grown under nitrogen-excess (+) or nitrogen-starved (-) conditions were separated by SOS-PAGE, transferred to cellulose nitrate and probed with antiserum prepared against the E. coli ntrC gene product. Each lane contains 200 ug of total protein. Lanes: 1 and 2, BJ3028; 3 and 4, BJ27147; 5 and 6, BJ110d. 50 NH-CII- + — + — + 4 “I 2 3 4 5 6 59 kd— v -. ' w K'J' 57 kd— ' Figure 3. 51 dimers per cell in nitrogen-limited cultures and only 5 molecules per cell in nitrogen-excess medium (40). This ratio of 14:1 agrees with gene fusion data (44). My analysis of the BJ110d wild-type NtrC levels also detected a substantial increase in protein abundance under nitrogen-limiting conditions (Figure 3, lanes 5 and 6). However, this increase was not seen in the BJ3028 mutant. Rather, an NtrC abundance roughly equal to the wild-type nitrogen-limited cultures was detected in both nitrogen-excess and nitrogen-starved cultures of the BJ3028 mutant (Figure 3, lanes 1 and 2). In summary, the data presented in this section indicate that 8027147 is a null mutant lacking NtrC and that BJ3028 is a constitutive producer of a truncated NtrC protein. Growth of B. japanicum ntrC::nptII mutants on various nitrogen sources. It has been reported in several other species that NtrC is required for the utilization of nitrate. (43, 45, 46). Both BJ27147 and BJ3028 were unable to utilize potassium nitrate as the sole nitrogen source. The growth rates of BJ27147 and BJ3028 were the same as wild-type B. japanicum when grown in minimal medium with ammonium, glutamine, glutamate, proline, arginine, histidine or aspartate as the sole nitrogen source. These results indicate that the B. japanicum ntrC is similar to the ntrC genes reported for R. meliloti and A. tumefaciens and is unlike the A. caulinodans ntrC and the Enterobacteriaceae ntrC which are required for the utilization of several amino acids as sole nitrogen sources (30, 38). 52 Transcription of B. japanicum glnA and glnII genes in ntrC::nptII mutants grown in aerobic cultures. Previous work in our laboratory indicated that in aerobic cultures glnII, but not glnA, was transcriptionally regulated in response to nitrogen availability (9). To determine the role, if any, of the B. japanicum ntrC gene in this regulation, 81 nuclease protection anaylses were carried out using primer extended probes specific to glnA and glnII. RNA was purified from cells grown in aerobic minimal media cultures as those described for the immunoblot analyses. As expected from earlier work (9), the glnA transcript level did not vary significantly in response to nitrogen availability (Figure 4). In contrast, in wild-type B. japanicum, the glnII transcript abundance was barely detectable in nitrogen-excess cultures but was induced about 40-fold in nitrogen-limited cultures (Figure 4). In strain BJ27147, no glnII transcript was apparent in either nitrogen-limited or nitrogen-excess cultures. This result indicated that the ntrC gene product is involved in the transcriptional activation of glnII under aerobic nitrogen-limited conditions. In strain BJ3028, glnII transcript was present under both nitrogen-limited and nitrogen-excessculture conditions indicating that constitutive expresssion of NtrC affects glnII expression (Figure 4; see discussion). Transcription of B. japanicum glnA and glnII genes in ntrC::nptII mutants grown in nitrogen-excess microaerobic cultures. Earlier studies in our laboratory with B. japanicum demonstrated that in nitrogen-excess cultures (YEM plus 10 mM KNO3 or YEM plus 10 mMKNO3, 10 mM NH4Cl) glnII transcription could be induced by decreasing the oxygen concentration to 53 Figure 4. Expression of B. japanicum glnA and glnII genes in aerobic cultures. Samples of RNA purified from BJ110d, BJ27147 or BJ3028 grown under nitrogen-excess (+) or nitrogen starved (-) aerobic cultures were hybridized to 32P-labeled single- stranded DNA probes specific for the 5’ ends of glnA or glnII and analysed by the SI nuclease protection method (see Material and Methods). The migration positions of the protected fragments (P ln and P lnI ) are indicated. The other radioactive gangs reprgseni residual undigested probe DNA. Lanes: 1 and 2, BJ110d; 3 and 4, BJ3028; 5 and 6, BJ27147. 54 NHC|3—+—+v—+ 1 23456 A Figure 4. 55 5.0% or less (2). Adams and Chelm (2) suggested this oxygen-limited glnII induction may be mediated through an Ntr-like system. To test if the B. japonicum ntrC gene is involved in the oxygen-limited response of glnII, microaerobic (0.2% 02) cultures were grown that contained excess nitrogen (10 mM NH4Cl, 0.5 mM KNO3). RNA was purified from these cells and 51 protection analyses were conducted as described for the aerobic culture experiments. As observed previously (2), glnII expression was induced in the wild-type BJ110d strain under microaerobic conditions even though excess nitrogen is present (Figure 5). The ntrC null mutant, BJ27147, also expressed glnII under these conditions in contrast to the aerobic cultures of this strain (Figure 5). This establishes that the ntrC gene product is not required for the microaerobic induction of glnII. In BJ3028, glnII transcript abundance was similar to BJ27147. Symbiotic phenotype of B. japanicum ntrC::nptII mutants. Both BJ27147 land BJ3028 formed approximately the same number of nodules on soybean plants as the wild-type strain. In addition, the average nodule fresh weight per plant, average weight per nodule, and average plant dry weight from ntrC mutant-incited symbioses were not significantly different from a wild-type incited symbiosis at 4 weeks after inoculation. Moreover, nodules incited by ntrC mutants showed wild-type levels of acetylene reduction activity. These results indicate that, as in R. meliloti (45), the B. japanicum ntrC gene is not required for initiating effective nodulation or for activating genes essential to nitrogen fixation during symbiotic growth. 56 Figure 5. Expression of B. japanicum glnA and glnII genes in nitrogen-excess, microaerobic (0.2% 02) cultures. Samples of RNA purified from BJ110d, BJ27147, or BJ3028 grown in nitrogen- excess cultures that were maintained at 0.2% oxygen, 98.2% nitrogen were analysed by the SI nuclease protection method. Lanes: 1, BJIlOd; 2, BJ27147; and 3, BJ3028. 57 1.23 ‘d-n. a-f a a .— - Pglnll gn Figure 5. 58 Figure 6. Expression of B. japanicum glnA and glnII genes in bacteria isolated from soybean nodules. Samples of RNA purified from B. japanicum cells isolated from nodules incited by BJ110d, BJ27147 or BJ3028 were analysed by the S1 nuclease protection method. Lanes: 1, BJ110; 2, BJ27147; 3, BJ3028; 4, BJ2513 (glnII strain [9]); 5, BJ2841 (glnA strain [9]). 59 .12345 glnA Figure 6. 60 Transcription of B. japanicum glnA and glnII genes in soybean nodules. Transcripts of both glnA and glnII are present in wild-type B. japanicum incited soybean nodules (2). To determine if ntrC is involved in the regulation during symbiosis of either of these genes, RNA was purified from bacteria that were isolated from 4 week old soybean nodules and used in 51 protection analyses (Figure 6). In both BJ27147 and BJ3028 glnA and glnII transcripts were present in levels similar to microaerobically grown cells. This indicates that the B. japanicum ntrC gene is not required for the expression of either glutamine synthetase gene during symbiosis. DISCUSSION I have isolated the B. japanicum ntrC gene and examined its role in regulating nitrogen assimilation. Several lines of evidence indicate that this locus is ntrC. First, only one homologous sequence was observed in the B. japanicum genome when the B. parasponiae ntrC probe was used (Figure 1). This homology was confirmed by comparing a B. japanicum partial DNA sequence with the B. parasponiae ntrC sequence. Secondly, a region upstream of the B. japanicum ntrC was found to be highly homologous to the ntrB genes reported upstream of the ntrC genes from B. parasponiae and R. meliloti (35, 45). Thirdly, consistent with ntrC genes in several other diazotrophs (43, 45, 46), the B. japanicum ntrC is required for the utilization of potassium nitrate as a sole nitrogen source. Finally, an NtrC antiserum from E. coli recognized the protein produced by the B. japanicum ntrC (Figure 3). The expression of 61 the B. japanicum glnII gene was induced when aerobic growth was limited by nitrogen depletion as expected if regulated by an Ntr-like system (Figure 4). Indeed, the absence of glnII transcript in the NtrC null mutant (BJ27147) indicates that in B. japanicum the control of glnII expression under aerobic nitrogen-limited growth is dependent on the ntrC gene and that this control is at the transcriptional level. This observation is consistent with two aspects of the upstream region of glnII. First, the -8 to -26 region of the glnII promoter is homologous to promoter regions of several nif genes that are thought to be recognized by RNA polymerase associated with NtrA (9, 37). Secondly, the upstream region of glnII contains two sequence elements (at -103 to -119 [9] and at -282 to -299 [5’-GGCACCAGCTTGGTGAC-3’; T. A. Carlson, Ph.D. dissertation, Michigan State University, E. Lansing, 1986]) that match the 5’-TGCACCnnnnTGGTGCA- 3’ consensus NtrC binding site from the Enterobacteriaceae (4) in 10 of 13 conserved positions. In BJ3028, a truncated NtrC protein was produced under nitrogen-starved and nitrogen- excess conditions (Figure3). Also in BJ3028, in contrast to wild-type, glnII was transcribed under both nitrogen regimes. This indicates that the truncated NtrC protein remained functional as an activator of glnII expression, i.e. the portion of the NtrC protein that is missing was not required for glnII activation. The B. parasponiae NtrC has a region very near the carboxy-terminus that is characteristic of the helix-turn- helix motif thought to be involved in DNA binding (17). Assuming colinearity of the B. japanicum and B. parasponiae NtrC proteins over this region, it is possible that the truncated NtrC protein of BJ3028 lacks this putative DNA binding region. The observation that the 62 truncated NtrC retained function as an activator of glnII expression therefore raises the possibility that the helix-turn-helix region is not required for glnII activation. In E. coli, NtrC represses the transcription of the ntrBC operon under nitrogen-excess conditions by binding to a sequence upstream of the promoter (49). In B. japanicum, preliminary observations indicate that the constitutive production of the truncated NtrC in mutant BJ3028 is due to a loss of repressor activity at the promoter for the ntrBC operon. This suggests that the repressor activity is present in the carboxy-terminal region of the B. japanicum NtrC. Studies are underway to determine if B. japanicum ntrC- regulatable promoters have upstream regions that bind NtrC. The inability to utilize nitrate as a nitrogen source is a common ntrC phenotype having been reported in R. meliloti (45), A. tumefaciens (43), and Azotobacter vinelandii (46). Although BJ27147 and BJ3028 had different phenotypes for aerobic glnII regulation and NtrC production, neither was able to utilize potassium nitrate as a nitrogen source. Thus, the B. japanicum ntrC may also be involved in the induction of assimilatory nitrate reductase as has been demonstrated in A. vinelandii (46). The reason why the truncated-NtrC mutant is able to activate glnII but is unable to utilize nitrate is not clear. It is possible that the nitrate reductase promoter is "weaker” and requires NtrC binding activity present in the carboxy-terminus whereas the glnII promoter does not. Alternatively, the carboxy-terminal region of B. japanicum NtrC may be involved in repressing a repressor of nitrate reductase. 63 One of the most interesting aspects of the B. japanicum ntrC gene is its involvement in the differential regulation of glnII when cells are exposed to different oxygen conditions. The appearance of glnII transcript in the NtrC null mutant (BJ27147) when it was grown microaerobically (or symbiotically; Figure 5 and 6) indicates that another regulator of glnII may be active under conditions where oxygen becomes limiting. This oxygen-limited induction of glnII has been shown to require the B. japanicum nifA gene (T. H. Adams, Ph.D dissertation, Michigan State University, E. Lansing, 1986) although no Nif-specific sequence element is present upstream of glnII (9). Whether or not NifA protein interacts directly with an upstream glnII region requires further study. In nature, B. japanicum exists as a free-living organism relying upon soil nitrogen or as a bacteroid where it fixes atmospheric nitrogen for itself and its soybean host. As a bacteroid, B. japanicum survives in a microaerobic environment necessitated by the oxygen lability of nitrogenase. The data presented in this paper support a model for the involvement of the B. japanicum ntrC gene in a dual regulatory system for glnII that responds to nitrogen or oxygen availability depending upon the developmental stage of the organism. Such a model has been postulated for the regulation of the nifHD, nifB and fixABC genes in R. meliloti (45). In this case the model posits that during free-living, nitrogen-starved growth in soil, glnII is transcriptionally activated by the ntrC gene product in concert with the NtrA sigma factor. However, under the microaerobic conditions encountered during symbiotic growth, glnII is activated by another regulatory mechanism involving the nifA 64 gene product in concert with NtrA. In light of the apparent dispensability of glnII (9), the significance of such dual regulation for this gene remains to be elucidated. 65 LITERATURE CITED 1. 10. 11. 12. Adams, T.H., and B.K. Chelm. 1984. The niffl and nifDK promoter regions from Rhizobium japanicum share structural homologies with each other annd with nitrogen-regulated promoters from other organisms. J. Mol. Appl. Genet. 2:392-405. Adams, T. H. and B. K. Chelm. 1988. Effects of oxygen levels on the transcription of nif and gln genes in Bradyrhizobium japanicum. J. Gen. Microbiol. 134: 611- 618. Adams, T. H., C.R. McClung, and B.K. Chelm. 1984. Physical organization of the Bradyrhizobium japanicum nitrogenase gene region. J. Bacteriol. 159:857-862. Ames, G.F.-L., and K. Nikaido. 1985. Nitrogen regulation in Salmonella typhimurium. Identification of an ntrC protein-binding site and definition of a consensus binding sequence. EMBO J. 4:539-547. Berk, A.J., and P.A. Sharp. 1977. Sizing and mapping early adenovirus mRNAs by gel electrophoresis of SI endonuclease- digested hybrids. Cell 12:721-732. Brown, C. H., and M.J. Dilworth. 1975. Ammonia assimilation by Rhizobium cultures and bacteroids. J. Gen. Microbiol. 86:39-48. Carlson, T.A., and B.K. Chelm. 1986. Apparent eucaryotic origin of glutamine synthetase II from the bacterium Bradyrhizobium japanicum. Nature 322:568-570. Carlson, T.A., M.L. Guerinot, and B.K. Chelm. 1985. Characterization of the gene encoding glutamine synthetase I (glnA) from Bradyrhizobium japanicum. J. Bacteriol. 162:698-703. Carlson, T.A., G.B. Martin, and B.K. Chelm. 1987. Differential transcription of the two glutamine synthetase genes of Bradyrhizobium japanicum. J. Bacteriol. 169:5861-5866. Chelm, B.K., and R.B. Hallick. 1976. Changes in the expression of the chloroplast genome of Euglena gracilis during chloroplast development. Biochemistry 15:593-599. Darr, S.C., S.C. Somerville, and C.J. Arntzen. 1986. Monoclonal antibodies to the light-harvesting chlorophyll a/b protein complex of photosystem II. J. Cell Biol. 103:733-740. Darrow, R.A., and R.R. Knotts. 1977. Two forms of glutamine synthetase in free-living root-nodule bacteria. Biochem. Biophys. Res. Comm. 78:554-559. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 66 Darrow, R.A., D. Crist, R.R. Evans, B.L. Jones, D. Keister, and R.R. Knotts. 1981. Biochemical and physiological studies on the two glutamine synthetases of Rhizobium, pp. 182-185. In: Current perspectives in nitrogen fixation, A.H. Gibson and H.E. Newton, (ed). Australian Academy of Science, Canberra. Ditta, G., S. Stanfield, D. Corbin, and D.R. Helinski. 1980. Broad host range DNA cloning system for Gram-negative bacteria: Construction of a gene bank of Rhizobium meliloti. Proc. Natl. Acad. Sci. USA 77:7347-7351. Dixon, R.A. 1984. The genetic complexity of nitrogen fixation. J. Gen. Microbiol. 130:2745-2755. Dixon, R. 1984. Tandem promoters determine regulation of the Klebsiella pneumoniae glutamine synthetase (glnA) gene. Nucleic Acids Res. 12:7811-7830. Drummond, M., P. Nhitty, and J. Hooton. 1986. Sequence and domain relationships of ntrC and nifA from Klebsiella pneumoniae: homologies to other regulatory proteins. EMBO J. 5:441-447. Eckert, R. 1987. New vectors for rapid sequencing of DNA fragments by chemical degradation. Gene 51:245-252. Elhai, J., and C.P. Nolk. 1988. A versatile class of positive selection vectors based on the nonviability of palindrome- containing plasmids that allows cloning into long polylinkers. Gene 68:119-138. Friedman, A.H., S.R. Long,S.E. Brown, H.J. Buikema, and F.M. Ausubel. 1982. Construction of a broad host range cosmid cloning vector and its use in the genetic analysis of Rhizobium mutants. Gene 18:289-296. Guerinot, M.L., and B.K. Chelm. 1986. Bacterial aminolevulinic acid synthase activity is not essential for leghemoglobin formation in the soybean / Bradyrhizobium japanicum symbiosis. Proc. Natl. Acad. Sci. USA 83:1837-1841. Gussin, G.N., C.R. Ronson, and F.M. Ausubel. 1986. Regulation of nitrogen fixation genes. Ann. Rev. Genet. 20:567-592. Hawkes, T., M. Merrick, and R. Dixon. 1985. Interaction of purified NtrC protein with nitrogen regulated promoters from Klebsiella pneumoniae. Mol. Gen. Genet. 201:492-498. Hirschman, J., P.-K. Wong, K. Sei, J. Keener, and S. Kustu. 1985. Products of nitrogen regulatory genes ntrA and ntrC of enteric bacteria activate glnA transcription in vitro: Evidence that the ntrA product is a sigma factor. Proc. Natl. Acad. Sci. USA 82:7525-7529. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. 67 Hunt, T.P., and B. Magasanik. 1985. Transcription of glnA by purified Escherichia coli components: Core RNA polymerase and the products of glnF, glnG and glnL. Proc. Natl. Acad. Sci. USA 82:8453-8457. Johnson, G.V., H.J. Evans, and T.M. Ching. 1966. Enzymes of the glyoxylate cycle in Rhizobia and nodules of legumes. Plant Physiol. 41:1330-1336. Laemmli, U.K. 1970. Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227:680-685. Leary, J.J., D.J. Brigati, and D.C. Hard. 1983. Rapid and sensitive colorimetric method for visualizing biotin-labeled DNA probes hybridized to DNA or RNA immobilized in nitrocellulose: Bio-blots. Proc. Natl. Acad. Sci. USA 80:4045-4049. MacFarlane, S.A., and M. Merrick. 1985. The nucleotide sequence of the nitrogen regulation gene ntrB and the glnA-ntrBC intergenic region of Klebsiella pneumoniae. Nucleic. Acids Res. 13:7591- 7606. Magasanik, B. 1982. Genetic control of nitrogen assimilation in bacteria. Ann. Rev. Genet. 16:135-168. Marmur, J., and P. Doty. 1962. Determination of the base composition of deoxyribonucleic acid from its thermal denaturation temperature. J. Mol. Biol 5:109-118. Marsh, J.L., M. Erfle, and E.J. Hykes. 1984. The pIC plasmid and phage vectors with versatile cloning sites for recombinant selection by insertional inactivation. Gene 32:481-485. Merrick, M.J., and J.R. Gibbins. 1985. The nucleotide sequence of the nitrogen-regulation gene ntrA of Klebsiella pneumoniae and comparison with conserved features in bacterial RNA polymerase sigma factors. Nucleic Acids Res. 13:7607-7619. Ninfa, A.J., and B. Magasanik. 1986. Covalent modification of the glnG product, NRI, by the glnL product, NRII, regulates transcription of the glnALG operon in Escherichia coli. Proc. Natl. Acad. Sci. USA 83:5909-5913. Nixon, B. T., C.R. Ronson, and F.M. Ausubel. 1986. Two-component regulatory systems responsive to environmental stimuli share strongly conserved domains with the nitrogen assimilation regulatory genes ntrB and ntrC. Proc. Natl. Acad. Sci. USA 83:7850-7854. 36. 37. 38. 39. 40. 41. 42. 43. 44. 45. 46. .68 O’Gara, F., and K.T. Shanmugan. 1976. Regulation of nitrogen fixation by Rhizobia: export of fixed N2 as NH4. Biochim. Biophys. Acta 437:313-321. Ow, D.H., V. Sundaresan, D.M. Rothstein, S.E. Brown, and F.M. Ausubel. 1983. Promoters regulated by the glnG and nifA products share a heptameric consensus sequence in the -15 region. Proc. Natl. Acad. Sci. USA 80:2524-2528. Pawlowski, K., P. Ratet, J. Schell, and F.J. de Bruijn. 1987. Cloning and charcterization of nifA and ntrC genes of the stem nodulating bacterium ORSS71, the nitrogen fixing symbiont of Sesbania rostrata: Regulation of nitrogen fixation (nif) genes in the free living versus symbiotic state. Mol. Gen. Genet. 206:207-219. Rao, R. N., and S.C. Rogers. 1979. Plasmid pKC7: a vector containing ten restriction endonuclease sites suitable for cloning DNA segments. Gene 7:79-82. Reitzer, L.J., and B. Magasanik. 1983. Isolation of the nitrogen assimilation regulator NRI, the product of the glnG gene of Escherichia coli. Proc. Natl. Acad. Sci. USA 80:5554-5558. Reitzer, L.J., and B. Magasanik. 1985. Expression of glnA in Escherichia coli is regulated at tandem promoters. Proc. Natl. Acad. Sci. USA 82:1979-1983. Reitzer, L.J., and B. Magasanik. 1986. Transcription of glnA in Escherichia coli is stimulated by activator bound to sites far from the promoter. Cell 45:785-792. Rossbach, S., J. Schell, and F.J. de Bruijn. 1987. The ntrC gene of Agrobacterium tumefaciens C58 controls glutamine synthetase (GSII) activity, growth on nitrate and chromosomal but not Ti-encoded arginine catabolism pathways. Mol. Gen. Genet. 209:419-426. Rothstein, D.M., G. Pahel, 8. Tyler, and B. Magasanik. 1980. Regulation of expression from the glnA promoter of Escherichia coli in the absence of glutamine synthetase. Proc. Natl. Acad. Sci. USA 77:7372-7376. Szeto, H. H., B.T. Nixon, C.R. Ronson, and F.M. Ausubel. 1987. Identification and characterization of the Rhizobium meliloti ntrC gene: R. meliloti has separate regulatory pathways for activation of nitrogen fixation genes in free-living and symbiotic cells. J. Bacteriol. 169:1423-1432. Toukdarian, A., and C. Kennedy. 1986. Regulation of nitrogen metabolism in Azotobacter vinelandii: isolation of ntr and glnA genes and construction of ntr mutants. EMBO J. 5:399-407. 47. 48. 49. 50. 51. 52. 53. 54. 69 Towbin, H., T. Staehelin, and J. Gordon. 1979. Electrophoretic transfer of proteins from polyacrylamide gels to nitrocellulose sheets: Procedure and some applications. Proc. Natl. Acad. Sci. USA 76:4350-4354. Tsang, V.C.H., J.M. Peralta, and A.R. Simons. 1983. Enzyme-linked immunoelectrotransfer blot techniques (EITB) for studying the specificities of antigens and antibodies separated by gel electrophoresis. Methods Enzymol. 92:377-391. Ueno-Nishio, S., S. Mango, L.J. Reitzer, and B. Magasanik. 1984. Identification and regulation of the glnL operator-promoter of the complex glnALG operon of Escherichia coli. J. Bacteriol. 160:379- 384. Upchurch, R.G., and G.H. Elkan. 1978. Ammonia assimilation in Rhizobium japanicum colonial derivatives differing in nitrogen- fixing efficiency. J. Gen. Micro. 104:219-225. Vairinhos, F., B. Bhandari, and D.J.D. Nicholas. 1983. Glutamine synthetase, glutamate synthase and glutamate dehydrogenase in Rhizobium japanicum strains grown in cultures and in bacteroids from root nodules of Glycine max. Planta 159:207-215. Vincent, J.M. 1978. Factors controlling the legume-Rhizobium symbiosis, pp. 103-129. In: Nitrogen fixation, volume 11, H.E. Newton and H.H. Orme-Johnson, (ed.), University Park Press, Baltimore, MD. Holk, C.P., A. Vonshak, P. Kehoe, and J. Elhai. 1984. Construction of shuttle vectors capable of conjugative transfer from Escherichia coli to nitrogen-fixing filamentous cyanobacteria. Proc. Natl. Acad. Sci. USA 81:1561-1565. Yanisch-Perron, C., J. Vieira, and J. Messing. 1985. improved M13 phage cloning vectors and host strains: nucleotide sequences of the M13mp18 and pUC19 vectors. Gene 33:103-119. CHAPTER 3 Bradyrhizobium japanicum ntrBC/glnA and nifA/glnA mutants: Evidence that separate regulatory pathways govern glnII expression in free-living and symbiotic cells. SUMMARY The presence in Bradyrhizobium japanicum of two glutamine synthetase genes, glnA and glnII, makes it difficult to study the physiological effects of possible regulatory genes that differentially control these genes. Here I have investigated effects of the products of the B. japanicum ntrBC operon and the nifA gene on glnII expression during the formation of nodules on soybean roots (Glycine max). Two B. japanicum strains with dual mutations were constructed by gene-directed mutagenesis. Each mutant carried a deletion in glnA and in either the ntrBC operon or nifA. These double mutants were compared with strains carrying single mutations in glnA, ntrBC, and nifA, for the ability to grow on various nitrogen sources, nodulate soybeans, and fix atmospheric nitrogen. The ntrBC/glnA mutant, but not the nifA/glnA mutant, was unable to grow on nitrate or ammonia as the sole nitrogen source when grown in aerobic cultures. Both ntrBC/glnA and nifA/glnA mutant strains formed fewer nodules per plant than the wild-type strain or the single mutants and were altered in their ability to fix nitrogen. These observations are interpreted in the context of a model of separate regulatory networks controlling glnII expression in B. japanicum. The 70 71 networks involve the products of ntrBC and nifA, and respond to either nitrogen deprivation or a signal present in symbiotic cells. INTRODUCTION During free-living growth Bradyrhizobium japanicum, the endosymbiont of soybean, assimilates ammonia primarily by the coordinate activity of glutamine synthetase (GS) and glutamate synthase (4, 30). However, in bacteroids, the symbiotic form of these bacteria, GS activity decreases in concert with the derepression of nitrogenase activity (4). The regulation of GS during free-living growth, incipient nodule formation and in mature bacteroids is therefore an integral part of the developmental process in B. japanicum. Bacteria of the Rhizobiaceae family, including B. japanicum, are unusual in that most members contain two GS enzymes GSI and GSII, encoded by glnA and glnII, respectively (7, 8, 9). The B. japanicum glnA gene has sequence similarity to the E. cali glnA gene and is not appreciably nitrogen-regulated (6). In contrast, the GS encoded by glnII has sequence similarity with eucaryotic GS’s and is transcriptionally regulated in response to nitrogen availability (5, 7). In Klebsiella pneumoniae, the best studied nitrogen-fixing organism, genes involved in nitrogen metabolism are subject to two levels of positive regulation in response to ammonia and oxygen (2). The first level, which is specific to nitrogen-fixation genes (nif genes) is mediated by the transcriptional activator encoded by nifA (NifA; 2). The second level of regulation is mediated by the centralized Ntr system, which controls the expression of a variety of nitrogen assimilation genes in the enteric bacteria (18). In the Ntr system the 72 gene product of ntrC (NtrC) functions as a transcriptional activator of many genes including nifA (14). Both nifA and ntrC homologous genes are present in members of the Rhizobiaceae and these genes have been isolated and characterized in several species (12, 20, 23, 27, 28, 29). These studies have found that the rhizobial nifA gene is expressed independently of the centralized Ntr system and is not dependent on NtrC for expression. Azorhizobium caulinodans, where some NtrC control over nifA has been reported, appears to be the exception (23). I have shown previously that in B. japanicum NtrC is required for the transcriptional regulation of glnII under conditions of nitrogen-starved aerobic growth (20). Under microaerobic conditions and during symbiotic growth, however, glnII is expressed but does not require NtrC (20). Szeto et al. (28) have shown in Rhizobium meliloti that, under symbiotic conditions, NifA activates several genes that are regulated by NtrC during aerobic growth. I have suggested that the NtrC-independent expression of glnII during microaerobic and symbiotic growth also involves NifA control (20). Thus in both B. japanicum and R. meliloti, several genes involved in nitrogen metabolism appear to be regulated by separate regulatory networks in free-living and symbiotic cells. In order to examine the physiological effects of specific regulatory proteins on the expression of glnII, I have developed double mutants that carry a glnA deletion along with a deletion in either ntrBC or nifA. By eliminating the interfering GS activity encoded by glnA, this strategy allowed us to assess the role of NtrB, NtrC, and NifA in regulating glnII and the subsequent effects of this control on soybean nodulation. The results reported here support a model of separate 73 regulatory networks in B. japanicum responding to nitrogen deprivation or a symbiotic signal which probably involves oxygen limitation. MATERIALS AND METHODS . Bacterial strains, plasmids, media and growth conditions. BJ110d is a small-colony derivative of B. japanicum 311b110 (13). All other bacterial strains and recombinant plasmids are listed in Table 1. To check growth on various nitrogen sources, 50 ml cultures of a mineral salts base medium adapted from O’Gara and Shanmugam (22) was used which had the following composition per liter: KH2P04, 0.3 g; NazHPO4, 0.3 g; M9504.7H20, 0.12 9. Trace elements were added (22) and the pH was adjusted to 6.8. After autoclaving, 1 ml of a filter- sterilized vitamin solution (22), 0.05 g CaClZ and 7.5 ml of filter- sterilized 20% (wt/vol) D-xylose were added per liter. The various nitrogen sources were added at 1.0 mM from filter-sterilized stock solutions. DNA biochemistry. Southern hybridizations and isolation of plasmid DNA were carried out as described previously (1). Total genomic DNA from B. japanicum was purified by phenol extraction (19). The isolation of glnA and the ntrBC operon has been described (6, 20). The nifA locus was isolated from a B. japanicum cosmid library using a 950 base pair PstI- EcoRV fragment from the Klebsiella pneumoniae nifA gene as a probe (T. H. Adams, Ph.D. dissertation, Michigan State University, East Lansing, 1986). 74 Table 1. Bacterial strains and plasmids. Strain or Description Source or plasmid reference BJIlOd Nild type (14) BJ2841 glnA::spc; glnA mutant, Spcr (7) BJ3263 ntrBC::nptII; ntrBC mutant, Kmr This work 3.12101 nifA::nptII; nifA mutant, Km” T. H. Adams GMIO-l glnA::spc, ntrBC::nptII; This work glnA/ntrBC mutant, Spcr, Kmr GM4-33 glnAzzspc, nifA::nptII; This work glnA/nifA mutant, Spc”, Kmr Elasmids pBJ284 Spcr; glnA deletion allele cloned into pKC7 (7) pBJ326 Kmr, Apr; ntrBC deletion allele cloned This work into pBR322 pDS4101 Apr; derivative of ColK (33) pKC7 Km”, Apr; vector (27) pBR322 Apr, Tcr; vector (3) 75 Gene-directed mutagenesis. Recombinant plasmids were manipulated in vitro and in E. coli to construct mutant alleles of the glnA and ntrBC loci (Table 1, Fig. 1). In both cases, sizable deletions within the genes were created and, for the purposes of B. japanicum strain construction, the deleted regions were replaced with antibiotic resistance genes. The plasmids carrying the mutant alleles were conjugally transferred from E. coli to B. japanicum by methods previously described (13). Stable B. japanicum recombinants were selected by antibiotic resistance conferred by the constructed allele, and those in which the mutant allele had replaced the wild-type allele were screened by hybridization as described previously (13). All of the constructed genotypes were confirmed by Southern hybridization analyses of genomic DNA. In all of the steps of strain construction, the growth medium was supplemented with 10 mM glutamine. The nifA mutant, BJ2101, was constructed using identical procedures to those described for the glnA and ntrBC mutants (T. H. Adams, Ph. D. dissertation). In the case of the double mutations, the confirmed ntrBC and nifA deletion mutants (BJ3263 and BJ2101) were processed through a second mating round using an E. coli strain carrying the plasmids pBJ284 and pDS4101 (see Table 1) to mutate the glnA gene. Plant inoculation, growth conditions and acetylene reduction assays. Soybean seeds (Glycine max [L.] Merr. cv. Amsoy 71) were sterilized as described previously (13) and then soaked for 1 hr in a late log phase culture of the appropriate strain of B. japanicum. Seeds were planted in modified Leonard jars as described by Vincent (31) using the 76 nitrogen-free medium of Johnson et al. (17). The plants were grown with a 16 hr daily light period, day/night temperatures of 28/25°C, a relative humidity of 60%, and a photon fluence rate of 300 umol.m-2.sec- 1. All plants were harvested at 28 days and the nodules were removed, rinsed with H20, counted, weighed and then used in acetylene reduction assays (15). Strain verification of bacteria isolated from nodules. Nodules were chosen at random, immersed in 95% ethanol for 30 sec, then in 3% hydrogen peroxide for 4 min, and then rinsed six times in sterile water. These surface-sterilized nodules were crushed aseptically in 1 ml of sterile 0.01% Tween 80. Serial dilutions were plated onto nonselective medium and colonies were subsequently checked for genetic markers. Electron Microscopy. Nodules were harvested from 4 week-old root systems, rinsed in deionized water, and sliced with a razor to pieces of approximately 1 mm in the largest dimension. The slices were then immediately fixed in 4% glutaraldehyde in 0.15 M sodium cacadylate, pH 7.2, for 2 hr under vacuum and then at 4°C overnight. Following fixation, the tissue was rinsed three times over a period of 3 hr at room temperature with Buffer A (21) containing 0.2% (w/v) tannic acid and then for 15 min in Buffer A alone. Postfixation was in 1% (w/v) osmium tetroxide in Buffer A for 1 hr at room temperature. The tissue was then dehydrated through a graded ethanol series, cleaned in propylene oxide and embedded in VCD/HSXA ultra low viscosity medium (Ladd Research Industries, Burlington, VT). Thin sections were cut on 77 an LKB Ultrotome III with a Dupont diamond knife. Sections were subsequently stained with uranyl acetate and lead citrate and examined in a Phillips EM300 electron microscope. RESULTS AND DISCUSSION Construction of B. japanicum ntrBC/glnA and nifA/glnA mutants. We have reported the construction of a glnA mutant allele and its use in the generation of a B. japanicum glnA mutant (7). Here this allele was used in conjugal matings with ntrBC and nifA B. japanicum strains to create ntrBC/glnA and nifA/glnA mutants (Table 1, Fig. 1). The mutant glnA allele was constructed by deleting the glnA 300 bp EcoRI-BamHI region and replacing it with a HindIII fragment containing a spectinomycin resistance gene. The source of the fragment was a derivative of the plasmid pHC45 (25), to which EcoRI-HindIII and BamHI- HindIII adaptors obtained from plasmid pKC7 (26) had been added (Fig. l). The mutant ntrBC region was constructed by deleting the 1.8 kb SalI fragment from the center of the ntrBC operon and replacing it with a 1.35 kb kanamycin resistance cassette which has been described previously (20). This cassette is homologous to the kanamycin resistance gene present on pKC7. The mutant nifA allele carried a 300 bp deletion which was replaced with a pKC7-derived EcoRI-XhoI 1.6 kb kanamycin cassette (T. Adams, Ph.D. dissertation). The double mutants were constructed by conducting a second round of mutagenesis as described in the Materials and Methods section. The mutations in all the regions were confirmed by Southern hybridizations of XhoI-digested total genomic DNA with pKC7 (Fig. 2), and with plasmids containing the 78 Figure 1. Physical maps of the B. japanicum glnA, ntrBC and nifA wildtype and mutant alleles. The coding regions of the genes are indicated by the boxed regions containing arrowheads, and transcription is from left to right. The insertion locations are indicated for the spectinomycin resistance gene (spc) and the kanamycin resistance gene (nptII). See Materials and Methods for details of constructions. Restriction endonuclease sites: B, BamHI; Bg, BglII; E, EcoRI; H, HindIII; 5, Sell; N, NruI; X, XhoI. 79 LOCIJS 3.] Hill E14 )3 ) )7F) Ix IN spc glnA anl ntrBC lkb Figure 1. nptH nifA 80 Figure 2. Hybridization analysis of B. japanicum genomic DNA from wildtype and mutant strains to pKC7. XhaI digests of total cellular DNA were transferred to cellulose nitrate and hybridized to radiolabeled pKC7. Lanes: 1, GM4-33 (glnA/nifA mutant); 2, BJ2101 (nifA mutant); 3, BJ2841 (glnA mutant); 4, BJ3263 (ntrBC mutant); 5, GMIO-l (glnA/ntrBC mutant); and 6, BJ110d (wildtype). 1 81 23455 Figure 2. ——nirBC —glnA_ —nifA- 82 appropriate gene region (data not shown). The presence of pKC7- homologous sequences in all of the constructions allowed us to use this vector to verify the presence of the mutant alleles in all strains (Fig. 2). These results confirmed that the mutant alleles had been recombined into the expected regions of the B. japanicum chromosome. The glnA, ntrBC and nifA mutant alleles were carried on 4.4 kb, 4.7 kb and 2.3 kb XhoI fragments, respectively (Fig. 2). Free-living and symbiotic phenotypes of the B. japanicum glnA, ntrBC and ntrBC/glnA mutants. Previous work in our laboratory indicated that NtrC is required for the transcriptional activation of glnII under aerobic culture conditions (20). I therefore predicted that, lacking either glutamine synthetase, the ntrBC/ glnA mutant would be deficient in the utilization of some nitrogen sources under aerobic conditions. I also predicted, based on the observation that a glnA/glnII mutant is Nod‘ (7), that the ntrBC/glnA mutant would be altered in its ability to form an effectve symbiosis with soybean. A potentially critical difference, however, existed between the ntrBC/glnA and the glnA/glnII mutants. In symbiotic cells of the ntrBC/glnA mutant, glnII should still be expressed in an NtrC-independent manner (see Introduction). Thus, if ntrBC/glnA mutant cells were able to invade root cells then it was expected that they would produce GSII and, as a result, be capable of establishing a normal symbiosis. In order to detect nitrogen-utilization defects, 1 tested the ntrBC, glnA, and ntrBC/glnA mutants for growth on various nitrogen sources 83 (Table 2). It was reported for several rhizobial species, including 3. japanicum, that NtrC is required for aerobic growth on nitrate as a sole nitrogen source (20, 27, 28). I found, as expected, that the ntrBC region was required for growth on nitrate; it was not required for growth on ammonia, glutamate or glutamine (Table 2). The ntrBC/glnA mutant, GMIO-l, was also unable to utilize nitrate, but in addition, it could not utilize ammonia and grew poorly on glutamate and glutamine as sole nitrogen sources (Table 2). He reported previously that B. japanicum strains with single glnA and glnII mutations exhibit no nitrogen-utilization defects, but that a glnA/glnII double mutant is a glutamine auxotroph (7). The fact that the ntrBC/glnA mutant was unable to utilize ammonia as a sole nitrogen source indicated an absolute requirement of the ntrBC region for glnII expression during aerobic, free-living growth. The effects of the ntrBC, glnA, and ntrBC/glnA mutations on soybean nodulation and nitrogen-fixation were determined by growing cultures of minimal medium supplemented with 1 mM glutamine, washing the cultures in unsupplemented medium, and inoculating soybean seeds in modified Leonard jars. I found that the ntrBC mutant, BJ3263, formed similar numbers of nodules on soybean as the wild-type strain (BJ110d; Table 3). In addition, the average plant dry weight and acetylene reduction rates with this mutant were similar to wild-type (Table 3). However, the ntrBC/glnA mutant, GMlO-l, produced many fewer nodules than the wild- type or ntrBC strain and these nodules were of two distinct types (Table 3, Fig. 3C). The less common type were apparently normal (see below), leghemoglobin-containing nodules of a similar size as wild type nodules. 84 Table 2. Growth properties of B.,japonicum strainsa Nitrogen sourceb Strain Genotype Nitrate Ammonia Glutamate Glutamine BJ110 wild type ++ ++ ++ ++ BJ2841 glnA ++ ++ ++ ++ BJ3263 ntrBC - ++ ++ ++ BJ2101 nifA ++ ++ ++ ++ GMIO-l glnA/ntrBC - - + + GM4-33 glnA/nifiA ++ ++ ++ ++ a Growth was determined in liquid minimal medium (see Materials and Methods), and is indicated as fellows: ++ very good; + good; - none. ° Nitrogen sources were tested at 1.0 mM concentration. All nitrogen sources were added from freshly prepared, filter-sterilized stock solutions. 85 Table 3. Symbiotic properties of B. japanicum strainsa. Strain Genotype Number of Nodules Acetylene reduction Plant dry (umoles per plant) weight (gm) BJ110 wildtype 63.7 1 17.6 19.8 i 4.6 4.33 i .28 BJ2841 glnA 82.3 i 7.5 27.3 i 0.6 3.20 i .35 BJ3263 ntrBC 58.8 i 9.6 21.3 1 7.4 3.00 i .38 802101 nifA 151.5 1 15.2 0 .94 i .07 GMIO-l glnA/ntrBC 13.1 i 5.3 .2 i 0.1 .86 i .16 GM4-33 glnA/niffl 29.5 1 13.3 0 .93 i .15 a The numbers represent the means and standard error of the means from three nodulation tests per strain. Plants were inoculated, grown and harvested as described in Material and Methods. 86 Figure 3. Comparison of symbiotic phenotypes of B. japanicum BJ110d and mutant strains BJ2101, GMIO-l, GM4-33. Sterilized soybean seeds were inoculated with the appropriate B. japanicum strain and grown in modified Leonard jars (see Materials and methods). Plants were harvested four weeks after sowing. Panels: A, BJ110d (wildtype); B, BJ2101 (nifA mutant); C, GMlO-l (glnA/ntrBC mutant); D, GM4-33 (glnA/nifA mutant). 87 Figure 3. 88 The second, more common type, were smaller than wild-type nodules and were yellowish-white inside. Electron microscopy of the normal- appearing nodules revealed numerous mature bacteroids surrounded by well-formed peribacteroid membranes; in all respects the ultrastructure of these nodules was identical to wild type (BJ110d) nodule ultrastructure (Fig. 4A and 40). Electron microscopy of the greenish- white nodules revealed very few bacteroids (Fig. 4C). To ensure that these nodules had been formed by GMIO-I, bacteria were isolated from randomly chosen nodules of both types and these were checked for the presence of genetic markers. All of the 100 colonies tested were both kanamycin-resistant and spectinomycin-resistant indicating that all the bacteria present were mutant and that no reversion had occurred. These observations indicate that, in the absence of glnA, the activation of glnII by NtrC is necessary for prolific nodulation of soybean roots. The presence of some wild- type nodules, however, suggests that when ntrBC/glnA mutant cells do invade soybean root hairs that glnII is expressed and a normal symbiosis is possible. The low rate of acetylene reduction detected from the GMlO-l nodules is apparently attributable to these wild-type nodules (Table 3). The inability of GMlO-l to prolifically nodulate soybean is probably due simply to the nitrogen-utilization defects of this strain. A B. japanicum glnA/glnII mutant, also a glutamine auxotroph, is similarly defective in nodulation (7). An Azorhizobium caulinodans glutamine auxotroph also forms a defective symbiosis (10). It appears, however, that the inability of B. japanicum and A. caulinodans glutamine auxotrophs to form effective symbioses is not a general phenomenon in 89 Figure 4. Transmission electron micrographs of nodule tissue infected by B. japanicum BJ110d, and mutant strains GMlO-l and GM4-33. The tissue was harvested 4 weeks after seed inoculation and sowing. All micrographs were taken at the same magnification (SOOOX) with the bar in panel 0 representing 1 um. Panels: A, BJIlOd; B, GMIO-l (yellowish-green nodule); C, GM4- 33; D, GMlO-l (normal-type nodule). 90 ‘. .97. “‘-\ m ...-n. . a Figure 4. 91 rhizobial-plant associations. A Rhizobium meliloti glnA/glnII mutant is a glutamine auxotroph but is Nod+ Fix+ on Medicago sativa (11). A possible explanation for these contrasting results is that the plant partners in these associations may differ in their ability to provide glutamine to their bacterial symbiont. Free-living and symbiotic phenotype of the B. japanicum nifAzglnA mutant. No nitrogen-utilization deficiencies have been reported for free- living rhizobial nifA mutants (12, 29) and no nitrogen-utilization r“... defects were observed in a B. japanicum nifA mutant strain (BJ2101) in our laboratory (Table 2; T. Adams, Ph.D. dissertation). I did not expect that a nifA/glnA mutant would exhibit nitrogen-utilization defects in aerobic cultures, since glnII expression is activated by NtrC under these conditions (20). However, previous 51 protection experiments in our laboratory with RNA isolated from B. japanicum nifA bacteroids (BJ2101) revealed that NifA is required for the activation of glnII in symbiotic cells (T. H. Adams, Ph.D. dissertation). I expected, therefore, that nifA/glnA mutant cells inoculated onto soybean roots would rely on NtrC to activate glnII expression up until they encountered the symbiotic signal that normally transfers this control to NifA. In light of the pleiotropic nature of B. japanicum nifA mutants (12), I was particularly interested to determine at what developmental stage the loss of gin]! expression would be manifested during nodulation. 92 As expected, I observed no nitrogen-utilization deficiencies in the nifA/glnA mutant, GM4-33, in free-living cultures (Table 2). This indicated that a normal level of GSII activity was present in this strain and that nifA therefore does not play an important role in activating glnII expression during aerobic, free-living growth. B. japanicum nifA mutants fail to derepress many genes involved in an effective symbiosis including nifH, nifDK, fixA, and fixBCX (12, 16). E In addition, Fischer et al. (12) reported that B. japanicum nifA mutants exhibit an altered nodulation phenotype inducing numerous, small, widely distributed soybean nodules in which the bacteroids are subject to severe degradation. It was suggested that this hypernodulation is due I to a circumvention of a plant phenomenon termed "autoregulation" in which effective nodulation events inhibit further nodulation on other parts of the roots (12, 24). I used an independently constructed 8. japanicum nifA mutant (BJ2101) as a control in these experiments, and confirmed that more than double the number of nodules are observed on a nifA mutant-inoculated root system than are seen in a wild-type induced symbiosis (Table 3, Fig. 3B). In contrast, I found that the B. japanicum nifA/glnA mutant, GM4-33, formed significantly fewer nodules than the nifA mutant or the wild-type strain (Table 3, Fig. 3D). The nodules that were formed were larger than nifA mutant-induced nodules but, like those nodules, they were Fix' (Table 3, Fig. 30). The fact that some nodules were formed by strain GM4-33 suggested that glnII may be expressed, albeit at a low level. These results were consistent with the notion that NifA activates glnII expression and, in addition, indicated that the glutamine synthetase activity encoded by glnA is 93 required for the formation of the numerous, small nodules seen in the nifA mutant-incited symbiosis. Electron microscopy of the nifA/glnA- incited nodules revealed bacteroid degradation and the presence of numerous intercellular bacteria (Fig. 4C). It is not clear why the nifA/glnA mutant-induced nodules were larger than the nifA mutant- induced nodules. It is possible that the plant becomes limited in some way in the resources it allocates to an ineffective symbiosis involving a strain with a mutant nifA locus and that more plant resources are available to the few individual nodules formed with the nifA/glnA mutant than with the numerous nifA mutant-induced nodules. The glnA mutant, BJ2841, was also included in these experiments as a control and, as reported earlier (7), it formed slightly more nodules and had a higher acetylene reduction rate per plant than the wild-type incited symbiosis. A B. japanicum glnII mutant exhibits a similar phenotype (7). The significance of these observations is unclear. The results presented here are consistent with a model for glnII regulation that involves separate regulatory networks existing in free- living and symbiotic cells. The ntrBC/glnA mutant, according to this model, was defective in nodule formation because it lacked NtrC needed for the activation of glnII expression; it also lacked the ancillary GS activity encoded by glnA. An analogous glnA/glnII mutant is similarly ineffective in establishing a symbioisis with soybean (7). This defect probably results in an inability to maintain cell growth long enough to efficiently invade root hairs and thereby reach the microaerobic environment of the peribacteroid-membrane bound vesicle where NifA activation of glnII expression could occur. The nifA/glnA mutant- 94 induced symbiosis suffered from numerous defects caused by the absence of NifA and conclusions regarding the loss of glnII regulation by NifA are less clear. It can be concluded, however, that the inability of the nifA/glnA mutant to hypernodulate soybean indicates that the GS activity encoded by glnA allows the interaction induced by the nifA mutant, BJ2101, to proceed to the stage where numerous infections and subsequent hypernodulation occurs. In addition, the inability to hypernodulate indicates that sufficient ancillary GS activity is not supplied by glnII in this mutant. This suggests, but does not prove, that NifA is required for the activation of glnII in symbiotic cells. 95 REFERENCES 1. 10. 11. Adams TH, Chelm BK (1988) Effects of oxygen levels on the transcription of nif and gln genes in Bradyrhizobium japanicum. J Gen Microbiol 134:611-618. Ausubel FM (1984) Regulation of nitrogen fixation genes. Cell 37:5- 6. Bolivar F, Rodrigues RL, Greene PJ, Betlach H, Heynecker HL, Boyer HH, Crossa JH, Falkow S (1977) Construction and characterization of new cloning vehicles. A multiple cloning system. Gene 2:95-100 Brown CM, Dilworth MJ (1975) Ammonia assimilation by Rhizobium cultures and bacteroids. J Gen Microbiol 86:39-48 Carlson TA, Chelm BK (1986) Apparent eucaryotic origin of glutamine synthetase II from the bacterium Bradyrhizobium japanicum. Nature 322:568-570 Carlson TA, Guerinot ML, Chelm BK (1985) Characterization of the gene encoding glutamine synthetase I (glnA) from Bradyrhizobium japanicum. J Bacteriol 162:698-703 Carlson TA, Martin GB, Chelm BK (1987) Differential transcription of the two glutamine synthetase genes of Bradyrhizobium japanicum. J Bacteriol 169:5861-5866 Darrow RA, Knotts RR (1977) Two forms of glutamine synthetase in free-living root-nodule bacteria. Biochem Biophys Res Comm 78:554-559 Darrow RA, Crist D, Evans HR, Jones BL, Keister D, Knotts RR (1981) Biochemical and physiological studies on the two glutamine synthetases of Rhizobium. In: Gibson AH, Newton HE, (eds) Current perspectives in nitrogen fixation. Australian Academy of Science. Canberra, Australia, pp 182-185 de Bruijn FJ, Pawlowski K, Ratet P, Hilgert U, Hang CH, Schneider M, Meyer H, Schell J (1988) Molecular genetics of nitrogen fixation by Azorhizobium caulinodans 0RS571, the diazotrophic stem-nodulating symbiont of Sesbania rostrata. In: Bathe H, de Bruijn FJ, Newton HE (eds) Nitrogen fixation: Hundred years after. Gustav Fischer, Stuttgart, pp 351-355 de Bruijn FJ, Rossbach S, Schneider M, Ratet P, Messmer S, Szeto HH, Ausubel FM, Schell J (1989) Rhizobium meliloti 1021 has three differentially regulated loci involved in glutamine biosynthesis, none of which is essential for symbiotic nitrogen fixation. J Bacteriol 171: (in press) 12. 13. 14. 15. 16. 17. 18. 19. 21. 22. 96 Fischer H-F, Alvarez-Morales A, Hennecke H (1986) The pleiotropic nature of symbiotic regulatory mutants: Bradyrhizobium japanicum nifA gene is involved in control of nif gene expression and formation of determinate symbiosis. EMBO J 5:1165-1173 Guerinot ML, Chelm BK (1986) Bacterial aminolevulinic acid synthase activity is not essential for leghemoglobin formation in the soybean/Bradyrhizobium japanicum symbiosis. Proc Natl Acad Sci USA 83:1837-1841 Gussin GN, Ronson CH, Ausubel FM (1986) Regulation of nitrogen fixation genes. Ann Rev Genet 20:567-592 Hardy RHF, Holsten RD, Jackson EK, Burns RC (1968) The acetylene- ethylene assay for N fixation: laboratory and field evaluation. Plant Physiol 43:118 -1207 Hennecke H, Fischer H-M, Gubler M, Thony B, Anthamatten D, Kullik I, Ebeling S, Fritsche S, Zurcher T (1988) Regulation of nif and fix genes in Bradyrhizobium japanicum occurs by a cascade of two consecutive gene activation steps of which the second one is oxygen sensitive. In: Bathe H, de Bruijn FJ, Newton HE (eds) Nitrogen fixation: Hundred years after. Gustav Fischer, Stuttgart, pp 339-344. Johnson GV, Evans HJ, Ching TM (1966) Enzymes of the glyoxylate cycle in Rhizobia and nodules of legumes. Plant Physiol 41:1330- 1336 Magasanik, B (1982) Genetic control of nitrogen assimilation in bacteria. Ann Rev Genet 16:135-168 Maniatis T, Fritsch EF, Sambrook J (1982) Molecular cloning. Cold Spring Harbor Laboratory, Cold Spring Harbor, New York Martin GB, Chapman KA, Chelm BK (1988) Role of the Bradyrhizobium japanicum ntrC gene product in differential regulation of the glutamine synthetase II gene (glnII). J Bacteriol 170:5452-5459 Maupin P, Pollard TD (1983) Improved preservation and staining of HeLa cell actin filaments, clathrin-coated membranes, and other cytoplasmic structures by tannic acid-glutaraldehyde-saponin fixation. J Cell Biol 96:51-62 O’Gara F, Shanmugan KT (1976) Regulation of nitrogen fixation by Rhizobia: export of fixed N2 as NH4. Biochim Biophys Acta 437:313-321 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 97 Pawlowski K, Ratet P, Schell J, de Bruijn FJ (1987) Cloning and characterization of nifA and ntrC genes of the stem nodulating bacterium ORS571, the nitrogen fixing symbiont of Sesbania rostrata: regulation of the nitrogen fixation (nif) genes in the free living versus symbiotic state. Mol Gen Genet 206:207-219 Pierce M, Bauer HD (1983) A rapid regulatory response governing nodulation in soybean. Plant Physiol 73:286-290 Prentki P. Krisch HM (1984) In vitro insertional mutagenesis with a selectable DNA fragment. Gene 29:303-313 Rao RN, Rogers 56 (1979) Plasmid pKC7: a vector containing ten restriction endonuclease sites suitable for cloning DNA segments. | Gene 7:79-82 Rossbach S, Schell J, de Bruin FJ (1987) The ntrC gene of Agrobacterium tumefaciens C58 controls glutamine synthetase (GSII) 1 activity, growth on nitrate and chromosomal but not Ti-encoded ‘ arginine catabolism pathways. Mol Gen Genet 209:419-426 Szeto HH, Nixon BT, Ronson CH, Ausubel FM (1987) Identification and characterization of the Rhizobium meliloti ntrC gene: R. meliloti has separate regulatory pathways for activation of nitrogen fixation genes in free-living and symbiotic cells. J Bacteriol 169:1423-1432 Szeto HH, Zimmerman JL, Sundaresan V, Ausubel FM (1984) A Rhizobium meliloti symbiotic regulatory gene. Cell 36:535-543 Vairinhos F, Bhandari 8, Nicholas DJD (1983) Glutamine synthetase, glutamate synthase and glutamate dehydrogenase in Rhizobium japanicum strains grown in cultures and in bacteroids from root nodules of Glycine max. Planta 159:207-215 Vincent JM (1978) Factors controlling the legume-Rhizobium symbiosis. In: Newton HE, Orme-Johnson HH (eds) Nitrogen Fixation. Volume II. University Park Press. Baltimore, Maryland, pp 103-129 Holk CP, Vonshak A, Kehoe P, Elhai J (1984) Construction of shuttle vectors capable of conjugative transfer from Escherichia coli to nitrogen-fixing filamentous cyanobacteria. Proc Natl Acad Sci USA 81:1561-1565 CHAPTER 4 Bradyrhizobium japanicum glnB, a putative nitrogen regulatory gene, is regulated by NtrC at tandem promoters. SUMMARY The glnB gene from Bradyrhizobium japanicum, the endosymbiont of soybean (Glycine max), was isolated and sequenced and its expression under various culture conditions and in soybean nodules was examined. The B. japanicum glnB gene encodes a 12,237 dalton polypeptide that is highly homologous to the glnB gene products from Klebsiella pneumoniae and Esherichia coli. In contrast to the glnB genes from the enteric bacteria, glnB from B. japanicum is located directly upstream from glnA, the structural gene for glutamine synthetase. The glnB gene from B. japanicum is expressed from tandem promoters, which are differentially regulated in response to the nitrogen status of the medium. Expression from the downstream promoter involves the B. japanicum ntrC gene product (NtrC) in both free-living and symbiotic cells. Thus glnB, a putative nitrogen regulatory gene in B. japanicum, is itself Ntr-regulated, and NtrC is active in B. japanicum in the symbiotic state. INTRODUCTION Bradyrhizobium japanicum exists as a free-living organism, growing at the expense of soil nitrogen, or as a symbiont, reducing dinitrogen to ammonia for itself and its soybean host. Free-living B. japanicum cells assimilate ammonia primarily by the coordinate activities of glutamine 98 99 synthetase (GS) and glutamate synthase (12, 56). However, during the development of bacteroids, the symbiotic form of these bacteria, GS activity decreases in concert with the derepression of nitrogen-fixation activity (12, 56). The regulation of nitrogen assimilation pathways is therefore an integral part of the bacterial developmental process and the symbiotic interaction. ‘ In the family Enterobacteriaceae many genes involved in nitrogen assimilation, including glnA, the glutamine synthetase gene, are regulated by the Ntr system (36, 40, and references below). This regulation involves both transcriptional and post-translational control exerted by the products of the ntrA, ntrB, ntrC, glnB, glnD, and glnE genes (NtrA, NtrB, NtrC, GlnB, GlnD, and GlnE, respectively; [8, 10, 14, 23, 28, 31, 32, 35]). NtrB and NtrC are bifunctional regulatory proteins that act in concert to either repress or activate transcription (35, 43, 47). The functional state of NtrB is modulated by GlnB and GlnD in response to the ratio of intracellular glutamine to 2- ketoglutarate (14). Depending on its state, NtrB is able to switch NtrC between the activating phosphorylated form and the repressing dephosphorylated form (33, 41). NtrA is an alternative sigma factor that confers specificity on core RNA polymerase for NtrA-specific promoter sequences (28, 31). GlnB (P11 protein) plays a central role in coordinating the response of the Ntr system to combined nitrogen (13, 36, 37, 53). It is a tetramic protein with a subunit molecular weight of 12,387 daltons (29, 53) and exists in two interconvertible forms (13). Under nitrogen- deprived conditions, GlnB is uridylylated at a tyrosyl residue by GlnD, 72‘2“:- rql'h—Jfi’ if" '.- 'T‘Tflfifiw'. 31'1“???»- 100 a uridylyltransferase (UTase; [14, 53]). Under conditions of nitrogen excess GlnB is deuridylylated by GlnD (14). In the uridylylated form, GlnB stimulates the adenylyltransferase (ATase) activity of GlnE which mediates the deadenylylation (activation) of glutamine synthetase (CS; [4, 32]). In the deuridylylated form, GlnB acts in concert with GlnE to adenylylate (inactivate) GS (32). The deuridylylated GlnB also interacts with NtrB to dephosphorylate (inactivate) NtrC under conditions of nitrogen excess (23, 36). Thus, GlnB in the enteric bacteria is centrally involved in transmitting the nitrogen status of the cell to the GS modifying enzyme, GlnE, and the NtrC modifying enzyme, NtrB. Expression of nitrogen assimilation capacity in B. japanicum is coordinated by separate regulatory networks that respond to nitrogen or oxygen limitation (2, 17, 39). The network responsive to nitrogen status shares many features with Ntr control in the Enterobacteriaceae (26, 36) and genes that have similar functional roles to the enteric ntrC and ntrA genes have been characterized from several rhizobial species (39, 44, 50, 51, 54). The regulatory network responsive to oxygen limitation is activated in symbiotic cells and probably relies on the positive regulator, NifA, to activate the genes controlled by this system (39). Most members of the Rhizobiaceae contain two GS enzymes, GSI and GSII, encoded by glnA and glnII respectively (15, 16, 19, 24). In B. japanicum GSI is subject to adenylylation in response to ammonia whereas GSII is not known to be post-translationally modified (20, 24). Expression of glnII is controlled by both the nitrogen and oxygen .7 ‘ js- SWEET)!” u-n-n—I- “I“: . 101 regulatory networks whereas glnA is constitutively expressed (39). The adenylylation of GSI and the presence of NtrC suggest that a functional analog of the enteric glnB is present in B. japanicum. It was reported (18) that a region upstream from the Rhizobium leguminosarum glnA gene encodes a protein, part of which is highly homologous to a published ten-residue peptide sequence from the E. coli glnB gene product (49). Similar homologies were noted in published partial sequences upstream from the glnA genes from Azospirillum basilense and B. japanicum glnA (11, 16, 18). Although the role of glnB in nitrogen regulation in the enteric bacteria is well characterized, its expression in response to environmental stimuli has not been studied. The possibility that a glnB gene exists in B. japanicum raises questions as to its role in this symbiont and how its expression may be integrated into rhizobial oxygen and nitrogen regulation. In this paper I report the isolation of the complete B. japanicum glnB gene, the glnB sequence, and the finding that the B. japanicum glnB is regulated by nitrogen status at tandem promoters. In addition, I show that glnB expression is regulated by a mechanism involving NtrC in both free-living and symbiotic cells. MATERIALS AND METHODS Bacterial strains, plasmids, media, and growth conditions. BJ110d is small-colony derivative of B. japanicum 311b110 (25). The ntrC mutants, BJ27147 and BJ3028, are derivatives of BJ110d with chromosomal insertions and have been described previously (39). Recombinant plasmids and phages are listed in Table 1. The complex _" 6-"7' .‘ ..'.;“-.nh‘rI-' 'fl. ”1311- -lnii’ 102 Table 1. Bacterial strains and plasmids. Strain or plasmid Description Source or ‘ reference B. 132231233 strains BJ110d Hild-type (25) BJ27147 ntrC::nptII, null mutant; Kmr (37) BJ3028 ntrC::nptII, truncated NtrC; Kmr (37) Plasmids (phages) MIBglnA 391 bp SalI fragment from (2) B. japanicum glnA promoter (16) region cloned into M13mp18 (54) M13glnB 1000 bp EcoRI - SstI fragment This study from B. japanicum glnB region cloned into M13mp19 (54) 3616 Recombinant phage carrying entire (16) glnB and glnA B. japanicum genes in.xBF101 (36) pBJ47 Apr; 500 bp EcoRI - BamHI fragment This study containing entire B. japanicum glnB cloned into pBR322 (36) pBJ299 Kmr, Apr; 2.3 kbp XhoI fragment This study containing the B. japanicum glnB and partial glnA from 616 cloned into pKC7 (44) 103 medium YEMAN (YEM with 10 mM NH4Cl and 0.5 mM KN03) has been described before (39). For analyzing gene expression in aerobic cultures, a mineral salts basal medium (39) was used supplemented with either 10 mM glutamate (XG) or 10 mM glutamate plus 10 mM NH4Cl (XGA). For microaerobic growth experiments, ten-liter cultures of B. japanicum in YEMAN were grown in Microferm fermentors (New Brunswick Scientific) agitated at 200 r.p.m., and sparged with 0.2% 02 and 99.8% 2 N2 at a rate of 500 ml min‘l. The rate of gas flow was controlled using thermal mass flowmeters (Brooks Instruments, model 58500). Each 10 liter culture was inoculated with 50 ml of a stationary-phase aerobic culture in YEM medium. All 8. japanicum cultures were grown at 30°C. Isolation of bacteria from soybean nodules has been described previously (1). DNA biochemistry. Construction of the B. japanicum lambda library in ABFIOI and the isolation of recombinant phage.A6IG have been described (16). Phage particles were prepared by two rounds of cesium chloride block density gradient sedimentation (21) and phage DNA was extracted by formamide treatment. DNA restriction endonuclease fragments used in subcloning were isolated by separation on agarose gels and electroelution onto DE81 paper by the method of Dretzen (22). Standard procedures were used for restriction enzyme digestions, ligations and gel electrophoresis (38). DNA sequencing strategy. The B. japanicum glnB locus is contained on the large insert of the recombinant lambda clone.AGIG that we have r‘ 104 idescribed previously (Table 1, [14]). A 2.3 kbp XhoI fragment was subcloned from this lambda clone into pKC7 (46) to form p8J299. The plasmid pBJ299 and a subclone, pBJ47, (Table 1) were used to determine the partial restriction endonuclease map shown in Fig. 1. The fragments indicated in Fig. 1 were subcloned into M13mp18 or M13mp19 and sequenced by the dideoxy method (52) using [35$]dATP and either a primer hybridizing to the polylinker region (Bethesda Research Labs) or the primer, GM4, described below (51 protection analysis). The sequence “......" ...—3...".-. ..-.-fi. . .A_-"\o‘...' ...4‘1' presented was determined from the analysis of both DNA strands. .15-U a RNA biochemistry. B. japanicum cells (1 g) were suspended in 10 ml of 4.0% sarkosyl, 0.1 M Tris-HCl, pH 8.0 and passed two times through a French pressure cell (Aminco; Silver Spring, MD) at 12,000 psi. For each milliliter of broken cell solution 1 g CsCl was added. RNA was isolated by discontinuous CsCl gradient centrifugation as described previously (3). The RNA pellet was drained thoroughly, redissolved in H20, and then extracted twice with phenol and four times with diethyl ether. Following ethanol precipitation, the RNA was resuspended in sterile H20 and stored in aliquots at -70°C. SI protection analysis. Levels of transcript were quantitated by the $1 nuclease protection method adapted from Berk and Sharp (9) and described previously (2) in which mRNAs transcribed from both glnA and glnB could be detected in a single reaction. Five micrograms of total cellular RNA was used per lane. Single-stranded, 5’ end-labelled probes were synthesized by primer extension (2) using a glnA-specific 105 oligonucleotide primer described previously (2) or the glnB- specific primer GM4. The sequence of GM4 is as follows: 5’-TCCCGGCATTTCGATCA- 3’. The recombinant M13 phages used in synthesizing the single-stranded DNA hybridization probes are described in Table 1. The protected fragment from S1 analysis with the glnA probe is 125 nucleotides long. Hybridization, Sl digestion, and identification of protected products were as described previously (2). The DNA sequence ladder used for sizing the glnB transcripts was produced by primer extension of GM4 using clone M13glnB (Table 1). Computer analysis. The Pustell and Kafatos sequence analysis program (45) was used to search the upstream region for homologies to the E. coli consensus promoter, the Ntr consensus promoter, and the NtrC binding site consensus element. DNASIS and PROSIS (Hitachi Software Engineering Co.) were used to generate the predicted amino acid sequence and to optimize alignment of the amino acid sequences. Final alignment of the four amino acid sequences was done visually. RESULTS Nucleotide sequence of the B. japanicum glnB locus. The sequence of the 250 bp EcoRI-SalI fragment that lies immediately upstream of B. japanicum glnA has been reported previously ([16]; Fig. 1). It was noted that this region contains an open reading frame (ORF) that shares sequence similarity with the glnB genes from several other species (18, 29). In order to complete the sequence of this ORF and to analyze the upstream region for promoter elements I subcloned and sequenced the 106 .Hocx .x m_~mm .um m-ew .m m_moom .m m_:53m .m "mmHFm mama—ozcocco cowau_e mom .Lmuoeoca <=~m . i m # '!!."IIII‘II'WI!!PIII-Hill}(Illnlllp¢lllp«Ill? . ... " :.'.-'. . . . ._ I .5. ‘0 o - . . . . ..j.¢-l-,nunsjl!!)alloqjl-. ..__“____ “"E1 '51-; 3"“? a“ . mg“ 4 glnA Figure 4. .. .-.—‘MAd naugl— -....- ..- .. .q ‘1 .. 116 The differential expression of the t2 transcript in response to nitrogen status is typical of regulation mediated by the activator form of NtrC. I therefore looked at glnB expression in response to nitrogen status in two B. japanicum ntrC mutants, 8027147 and BJ3028 (Table l). Regardless of the culture nitrogen status, both strains produced only the longer transcript, t1 (Fig. 4, lanes 3-6). Thus, in aerobic, nitrogen-starved cultures of free-living B. japanicum, NtrC is required for expression of the t2 transcript, and apparently represses t1. i Previous observations in our laboratory, have indicated that separate i regulatory systems responsive to nitrogen or oxygen status control gene expression in free-living and symbiotic cells (39). I therefore wished . to determine how glnB expression was controlled in symbiotic cells. 51 experiments indicated that the t2 transcript was the predominant transcript in wild type (BJ110d) bacteria isolated from soybean nodules (Fig. 4, lane 7). However, in nodules infected with the NtrC null mutant, BJ27147, there was a marked decrease in the t2 transcript and an increase in the t1 transcript, similar to the pattern of expression observed with this strain under aerobic conditions (Fig. 4, lane 8). These observations indicate that expression of the glnB t2 transcript in bacteroids involves NtrC activation similar to its expression in aerobic, nitrogen-deprived cultures. In addition, I found that t1 was the predominant transcript in microaerobic, ammonia-grown cultures of wild-type B. japanicum, as it was in aerobic, ammonia-grown cultures (Fig. 4, lane 9). These results indicate that oxygen regulation is probably not involved in expression of the glnB t2 transcript under microaerobic or symbiotic conditions. Rather, expression of this 117 transcript appears primarily responsive to nitrogen status. The position of glnA downstream from glnB in B. japanicum raises the question of whether the two genes may, under some conditions, be co- transcribed. If this were the case, then a significant fraction of the glnA probe used in 51 experiments (Fig. 4) should be fully protected. In fact, the abundance of full-length glnA probe never exceeded that observed in the control digestion (Fig. 4, lane 10). Promoter mapping and analysis. The two 5’ ends of the B. japanicum glnB transcripts were mapped by 51 protection experiments (Fig. 5) using the GM4 primer-extended probe depicted in Fig. l. The transcripts, t1 and t2, were mapped to positions 261 bp and 157 bp, respectively, upstream of the putative start codon (Fig. 2, 5). Sequence analysis of the upstream region surrounding these transcriptional start sites revealed a characteristic E. coli consensus-type promoter sequence (27) preceding the t1 start site and a typical Ntr promoter sequence (6) preceding the t2 start site (Fig. 6). I designate the upstream promoter glanI and the downstream promoter glanZ. In addition, a region was found between nucleotides -86 to -102 that matches the consensus NtrC binding site (5’-TGCACCnnnnTGGTGCA-3’) from the Enterobacteriaceae in 8 of the 13 conserved positions (5, 47; Fig. 2). Interestingly, this sequence overlaps the transcriptional start site of t1, suggesting that the observed NtrC-dependent repression of t1 under conditions of nitrogen starvation may be the direct result of NtrC binding to this region. 118 Figure 5. Determination of the 5’ ends of the two glnB transcripts. A single stranded probe specific to the glnB 5’ end was generated by primer extension of the oligonucleotide GN4 using MlBglnB as a template (see Materials and Methods). This probe was hybridyzed with total bacterial RNA isolated from B. japanicum 110d cells that were grown in glutamate (XG) or ammonia (XGA) cultures. Transcripts t1 and t2 are indicated. The DNA sequence ladder was produced by primer extension of GM4. Lanes: 1 through 4 show products of the G, A, T, and C sequencing reactions, respectively; 5, BJ110d (X6); 6, BJ110d XGA . 119 123456”). 4| n‘ *L LL 4 ....E E...... E... ......r .. ...”... Figure 5. 120 Figure 6. Analysis of the promoters of the B. japanicum glnB gene. A.) The glanl promoter is aligned to maximize homology with the B. japanicum glnA promoter (16) and the E. coli consensus promoter (27). Upper case letters in the E. coli sequence indicate highly conserved bases among E. coli promoters, and upper case letters in the B. japanicum sequences indicate homologies to the E. coli consensus promoter or to the other B. japanicum promoter. 8.) The glanZ promoter is aligned to maximize homology with the Ntr consensus promoters from the enteric bacteria (6), B. japanicum (17) and the glnII promoter from B. japanicum (17). The asterisks (*) indicate points of transcriptional initiation; R denotes a purine, Y a pyrimidine, and N any nucleotide. E. coli consensus Bj glnA BJ glanI Bj Ntr consensus Enteric Ntr consensus Bj glnII 83 glanz 121 TTGACat -N10- t-tg-TAtAaT ------- * TgGACgct -N10- g-TG-TAchc ------- * TTGcCtAT -N9- T-cG-TATttT ------- * ---CC-TGGCACGCNAGTTGCT-AT ------ *** ----- CTGGYAYRNNNNTTGCA-N6_11—* ------ TGGCACGCtAaaTGCTth------* t ----c-TGGCATAgacccTGCT-N10 at? 122 DISCUSSION Several lines of evidence indicate that the open reading frame I have described encodes the B. japanicum glnB gene. First, the nucleic acid sequence is very similar to the sequences of the glnB genes from E. coli and K. pneumoniae. Second, the amino acid sequence is very highly conserved, particularly in regions that play a functional role in the GlnB protein from E. coli. Finally, a requirement for a GlnB protein in the rhizobia may be inferred from the presence of several components of an Ntr system. Members of the Rhizobiaceae contain functional homologs to the regulatory genes ntrA, ntrB, ntrC, and B. japanicum GSI is adenylylated in response to excess nitrogen in the medium (20, 24, 39, 50, 54). The physiological role of GlnB in B. japanicum remains to be established. It should be possible to clarify the function of this protein by studying a B. japanicum strain that has a glnB insertional mutation. I have constructed various mutant glnB alleles in vitro by inserting drug resistance markers into the unique SalI site of the open reading frame. However, despite numerous attempts with an established conjugative mating system, I have been unable to identify a recombinant that resulted from a double cross-over event. The difficulty in constructing a mutant may indicate that the GlnB protein plays an essential role in the regulation of nitrogen assimilation or some other process in B. japanicum. A transcriptional analysis of the glnB gene indicated that transcription is initiated from two closely spaced, but independently regulated promoters. One promoter, glanZ, is expressed under 123 conditions of nitrogen-deficiency, and its expression is NtrC-dependent. The other promoter, glanl, is expressed under conditions of nitrogen- sufficiency, and its repression is NtrC-dependent. The glanl promoter is similar to an E. coli consensus promoter (27; Fig. 6). The glanZ promoter is similar to an Ntr-regulated promoter consensus derived from several nitrogen-regulated K. pneumoniae genes (6; Fig. 6). . Transcription start sites have not been determined for the same region from R. leguminosarum, but a similar Ntr promoter consensus sequence is present (18). The regulation of the glnB genes from K. pneumoniae and E. coli has not been investigated (29, 53). It is unclear what physiological significance may be attached to the dual regulation of glnB. The transcript abundance of glnB remains relatively constant regardless of which promoter is active (Fig. 4). It is possible that the two promoters are required simply to maintain a constant level of glnB product under different nitrogen conditions. This would be necessary if, for example, GlnB is required during growth on a good source of nitrogen, since under these conditions the B. japanicum Ntr promoter is only weakly activated (Fig. 4). Tandem promoters are commonly observed in E. cali when a gene is expressed constitutively and under some global control, or when regulated by two global systems (30). For example, the glnA gene from E. coli is regulated by tandem promoters (47). The glnA upstream promoter, pl, is a consensus-type promoter and is repressed by NtrC under conditions of carbon deficiency and nitrogen excess (47). The glnA downstream promoter, p2, is a characteristic Ntr promoter and requires NtrC for expression under conditions of nitrogen limitation 124 (47). The structure and expression of the E. coli glnA tandem promoters is thus very similar to the B. japanicum glnB promoters. The presence of the glnB gene upstream of glnA in B. japanicum, however, apparently does not have physiological significance since little, if any, co- transcription can be detected. Moreover, the B. japanicum glnA has its own promoter and the transcript abundance of glnA does not change dramatically in response to nitrogen deprivation as it does in E. coli (17, 47). Rather, the NtrC-activated induction of glnA in E. coli in response to nitrogen limitation, is replaced in B. japanicum by the __. ..__.-———4-A .....- . . ._., ' . 5" ..“ ‘ . ‘ 1, NtrC-activated expression of glnII (39). The sequence upstream of glnB has two potential NtrC-binding sites. 3 One, overlaps the glnB t1 start site and is similar to a known NtrC binding sequence that overlaps the start site of ntrB in E. coli (55). A similar consensus element overlaps the E. coli glnApI start site (47). Since both ntrB and glnApl promoters are repressed by NtrC it seems reasonable to postulate that the NtrC binding consensus element overlapping the B. japanicum glanl start site is also repressed by NtrC under nitrogen deprived conditions. The appearance in the ntrC mutants, BJ27147 and BJ3028, of transcript t1 under nitrogen-deprived conditions, in contrast to wild-type B. japanicum, supports this conjecture (Fig. 4). BJ27147 does not produce NtrC protein, while BJ3028 constitutively produces an NtrC that is truncated at the carboxy-terminus (39). This truncated NtrC is thought to lack the repressor DNA binding domain (39). Another less symmetrical NtrC binding sequence element occurs at position +42 to +58 (Fig. 2). This may be indicative of multiple NtrC binding sites upstream of glnB that function in increasing the local 125 concentration of NtrC, similar to those surrounding the E. coli glnA promoters (48). It was surprising to find that NtrC participated in the regulation of glnB in B. japanicum bacteroids isolated from soybean nodules. Other genes shown to be regulated by NtrC in free-living cells of R. meliloti or B. japanicum are not so regulated in symbiotic cells (39, 54). This would make sense if the intracellular nitrogen status of bacteroids is high as is generally thought to be the case (12). This supposition is based on the results of previous workers who found a low specific activity of glutamine synthetase in extracts from rhizobial cultures grown with excess ammonia or extracts from bacteroids, but a high specific activity in extracts from nitrogen-starved cultures (12). They concluded from these observations that intracellular nitrogen in symbiotic cells is not limiting. The observation that NtrC is involved in the expression of glnB in nodules raises two contrasting possibilities. First, if the intracellular nitrogen status is indeed high in bacteroids, then the involvement of NtrC indicates that this regulatory protein can be active in this nitrogen-Sufficient environment, in contrast to its inactivity in nitrogen-sufficient free- living cells (39, 54). It would follow from this that in nodules ntrC may be expressed (or NtrC protein may be activated) in response to a signal independent of the intracellular nitrogen status. A second possibility is that intracellular nitrogen status is not high in nodules as has been thought. If this were the case then ntrC could be expressed, as it is in free-living cells, in response to nitrogen deprivation. The report that at least 94% of ammonia is exported from 126 bacteroids (42) would support the notion that bacteroid cells experience nitrogen deprivation. 127 LITERATURE CITED 1. 10. 11. 12. Adams, T.H., and B. K. Chelm. I984. The nifH and nifDK promoter regions from Rhizobium japanicum share structural homologies with each other and with nitrogen-regulated promoters from other organisms. J. Mol. Appl.Genet. 2:392-405. Adams, T.H., and B. K. Chelm. 1988. Effects of oxygen levels on the transcription of nif and gln genes in Bradyrhizobium japanicum. J. Gen. Microbiol. 134:611-618. Adams, T. H., C. R. McClung, and B. K. Chelm. 1984. Physical organization of the Bradyrhizobium japanicum nitrogenase gene region. J. Bacteriol. 159:857-862. Adler, S.P., D. Purich, and E.R. Stadtman. 1975. Cascade control of Escherichia coli glutamine synthetase. Properties of the P regulatory protein and the uridylyltransferase-uridylyl-removing enzyme. J. Biol. Chem. 250:6264-6272. Ames, G.F.-L., and K. Nikaido. 1985. Nitrogen regulation in Salmonella typhimurium. Identification of an ntrC protein-binding site and definition of a consensus binding sequence. EMBO J. 4:539- 547. Ausubel, F.M. 1984. Regulation of nitrogen fixation genes. Cell 37:5-6. Bachman, B. J. 1983. Linkage map of Escherichia coli Kr12, Edition number 7. Microbiol Rev. 47:180-230 Bancroft, S., S.C. Rhee, C. Neumann, and S. Kustu. 1978. Mutations that alter the covalent modification of glutamine synthetase in Salmonella typhimurium. J. Bacteriol. 134:1046-1055. Berk, A.J., and P.A. Sharp. 1977. Sizing and mapping early adenovirus mRNAs by gel electrophoresis of SI endonuclease- digested hybrids. Cell 12:721-732. Bloom, F.R., M. S. Levin, F. Foor, and 8. Tyler. 1978. Regulation of glutamine synthetase formation in Escherichia coli: characterization of mutants lacking the uridylyltransferase. J. Bacteriol. 134:569-577. Bozouklian, H., and C. Elmerich. I986. Nucleotide sequence of the Azospirillum brasilense Sp 7 glutamine synthetase structural gene. Biochimie 68:1181-1187. Brown, C.M., and M.J. Dilworth. 1975. Ammonia assimilation by Rhizobium cultures and bacteroids. J. Gen. Microbiol. 86:39-48. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 128 Brown, M.S., A. Segal, and E.R. Stadtman. 1971. Modulation of glutamine synthetase adenylylation and deadenylylation is mediated by metabolic transformation of the PII-regulatory protein. Proc. Natl. Acad. Sci. USA 68:2949-2953. Bueno, R., G. Pahel, and B. Magasanik. 1985. Role of glnB and glnD gene products in regulation of the glnALG operon of Escherichia coli. J. Bacteriol. 164:816-822. Carlson, T.A., and B.K. Chelm. 1986. Apparent eucaryotic origin of glutamine synthetase II from the bacterium Bradyrhizobium japanicum. Nature 322:568-570. Carlson, T.A., M.L. Guerinot, and B. K. Chelm. 1985. Characterization of the gene encoding glutamine synthetase I (glnA) from Bradyrhizobium japanicum. J.Bacteriol. 162:698-703. Carlson, T.A., G.B. Martin, and B.K. Chelm. 1987. Differential transcription of the two glutamine synthetase genes of Bradyrhizobium japonicum. J. Bacteriol. 169:5861-5866. Colonna-Romano, S., A. Riccio, M. Guida, R. Defez, A. Lamberti, M. Iaccarino, H. Arnold, U. Preifer, A. Puhler. 1987. Tight linkage of glnA and a putative regulatory gene in Rhizobium leguminosarum. Nucleic Acids Res. 15:1951-1964. Darrow, R.A., and R.R. Knotts. 1977. Two forms of glutamine synthetase in free-living root-nodule bacteria. Biochem. Biophys. Res. Comm. 78:554—559. Darrow, R.A., D. Crist, H.R. Evans, B.L. Jones, D. Keister, and R.R. Knotts. 1981. Biochemical and physiological studies on the two glutamine synthetases of Rhizobium, pp. 182-185. In: Current perspectives in nitrogen fixation, A.H. Gibson and N.E. Newton, (ed). Australian Academy of Science, Canberra. Davis, R.H, D. Botstein, and J.R. Roth. 1980. Advanced bacterial genetics. Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y. Dretzen, G., M. Bellard, P. Sassone-Corsi, and P Chambon. 1981. A reliable method for the recovery of DNA fragments from agarose and acrylamide gels. Anal. Biochem. 112:295-298. Foor, F., Z. Reuveny, and B. Magasanik. 1980. Regulation of the synthesis of glutamine synthetase by the P protein in Klebsiella aerogenes. Proc. Natl. Acad. Sci. USA 77:2g36-2640. Fuchs, R.L. D.L. Keister, 1980. Comparative properties of glutamine synthetases I and II in Rhizobium and Agrobacterium spp.. J. Bacteriol. 144:641-648. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. 36. 129 Guerinot, M.L., and B.K. Chelm. 1986. Bacterial aminolevulinic acid synthase activity is not essential for leghemoglobin formation in the soybean/Bradyrhizobium japanicum symbiosis. Proc. Natl. Acad. Sci. USA 83:1837-1841. Gussin, G.N., C.R. Ronson, and F.M. Ausubel. 1986. Regulation of nitrogen fixation genes. Ann. Rev. Genet. 20:567-592. Hawley, D.K., and R.R. McClure. 1983. Compilation and analysis of Escherichia coli promoter DNA sequences. Nucleic Acids Res. 11:2237-2255. Hirschmann, J., P.-K. Hong, K. Sei, J. Keener, and S. Kustu. 1985. Products of nitrogen regulatory gene ntrA and ntrC of enteric bacteria activate glnA transcription in vitro: evidence that the ntrA product is a sigma factor. Proc Natl. Acad. Sci. USA 82:7525- 7529. Holtel, A., and M. Merrick. 1988. Identification of the Klebsiella pneumoniae glnB gene: Nucleotide sequence of wild-type and mutant alleles. Mol. Gen. Genet. 215:134-138. Hoopes, B.C., and R.R. McClure. 1987. Stategies in regulation of transcription initiation, p. 1231-1240. In: F.C. Neidhardt (ed.), Escherichia coli and Salmonella typhimwrium, vol. 2. Amer. Soc. Microbiol., Hash. D.C. Hunt, T.P., and B. Magasanik. I985. Transcription of glnA by purified Escherichia cali components: core RNA polymerase and the products of glnF, glnG, and glnL. Proc. Natl. Acad. Sci. USA 82:8453-8457. Janssen, K.A., and B. Magasanik. I977. Glutamine synthetase of Klebsiella aerogenes: genetic and physiological properties of mutants in the adenylylation system. J. Bacteriol. 129:993-1000. Keener, J., and S. Kustu. 1988. Protein kinase and phosphoprotein phosphatase activities of nitrogen regulatory proteins NtrB and NtrC of enteric bacteria: Roles of the conserved amino-terminal domain of NtrC. Proc. Natl. Acad. Sci. USA 85:4976-4980. Kozak, M. 1983. Comparison of initiation of protein synthesis in procaryotes, eucaryotes and organelles. Microbiol. Revs. 47:1-45. MacNiel, T., G.P. Roberts, D. MacNiel, and 8. Tyler. 1982. The products of glnL and glnG are bifunctional regulatory proteins. Mol. Gen. Genet. 188:325-333. Magasanik, B. 1982. Genetic control of nitrogen assimilation in bacteria. Ann. Rev. Genet. 16:135-168. .:.- ....- 37. 38. 39. 40. 41. 42. 43. 44. 45. 46. 47. 130 Mangum, J.H., G. Magni, and E.R. Stadtman. 1973. Regulation of glutamine synthetase adenylylation and deadenylylation by the enzymatic uridylylation and deuridylylation of the PII regulatory protein. Arch. Biochem. Biophys. 158:514-525. Maniatis, T., E.F. Fritsch, and J. Sambrook. 1982. Molecular cloning. Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y. Martin, G.B., K.A. Chapman, and B.K. Chelm. 1988. Role of the Bradyrhizobium japanicum ntrC gene product in differential regulation of the glutamine synthetase II gene (glnII). J. Bacteriol. 170:5452-5459. Merrick, M.J., S. Austin, M. Buck, R. Dixon, M. Drummond, A. Holtel, and S. MacFarlane. 1987. Regulation of nitrogen assimilation in enteric bacteria, p. 277-283. In: A. Torriani-Gorini, F. Rothman, S. Silver, A. Wright, and E. Yagil (ed.), Phosphate metabolism and cellular regulation in microorganisms. American Society for Microbiology, Hash. D.C. Ninfa, A.J., and B. Magasanik. 1986. Covalent modification of the glnG product, NR1, by the glnL product, NR 1’ regulates transcription of the glnALG operon in Esc erichia coli. Proc. Natl. Acad. Sci. USA 83:5909-5913. O’Gara, F., and K.T. Shanmugam. 1976. Regulation of nitrogen fixation by Rhizobia: export of fixed N2 as NH4. Biochim. Biophys. Acta 437:313-321 Pahel, G., D.M. Rothstein, and B. Magasanik. 1982. Complex glnA- glnL-glnG operon of Escherichia coli. J.Bacteriol. 150:202-213. Pawlowski, K., P. Ratet, J. Schell, and F.J. de Bruijn. 1987. Cloning and characterization of niffl and ntrC genes of the stem nodulating bacterium ORSS7I, the nitrogen fixing symbiont of Sesbania rostrata: regulation of the nitrogen fixation (nif) genes ggstheyfree living versus symbiotic state. Mol. Gen. Genet. :20 -219. Pustell, J. and F.C. Kafatos. 1984. A convenient and adaptable package of computer programs for DNA and protein sequence management, analysis, and homology determination. Nucleic Acids Res. 12:643-655. Rao, R.N., and S.G. Rogers. 1979. Plasmid pKC7: a vector containing ten restriction endonuclease sites suitable for cloning DNA segments. Gene 7:79-82. Reitzer, L.J., and B. Magasanik. 1985. Expression of glnA in Escherichia coli is regulated at tandem promoters. Proc. Natl. Acad. Sci. USA 82:1979-1983. 48. 49. 50. 51. 52. 53. 54. 55. 56. 57. 131 Reitzer, L.J., and B. Magasanik. 1986. Transcription of glnA in Escherichia coli is stimulated by activator bound to sites far from the promoter. Cell 45:785-792. Rhee, S.G., P.B. Chock, and E.R. Stadtman. I985. Nucleotidylations involved in the regulation of glutamine synthetase in Escherichia coli, p. 273-297. In: R.B. Freedman and H.C. Hawkins (ed.), The enzymology of post-translational modification of proteins, vol II. Acad. Press, London. Ronson, C.R., B.T Nixon, L.M. Albright, and F.M. Ausubel. 1987. Rhizobium meliloti ntrA (rpoN) gene is required for diverse metabolic functions. J. Bacteriol. 169:2424-2431. Rossbach, S., J. Schell, and F.J. de Bruin. 1987. The ntrC gene of Agrobacterium tumefaciens C58 controls glutamine synthetase (GSII) activity, growth on nitrate and chromosomal but not Ti-encoded arginine catabolism pathways. Mol. Gen. Genet. 209:419-426. Sanger, F., S. Nicklen, and A.R. Coulson. 1977. DNA sequencing with chain terminating inhibitors. Proc. Natl. Acad. Sci. USA 74:5463- 5467. Son, H.S., and S.G. Rhee. 1987. Cascade control of Escherichia coli glutamine synthetase: Purification and properties of P protein and nucleotide sequence of its structural gene. J. Biol. Chem. 262:8690-8695. Szeto, H. H., B.T. Nixon, C.R. Ronson, and F.M. Ausubel. 1987. Identification and characterization of the Rhizobium meliloti ntrC gene: R. meliloti has separate regulatory pathways for activation of nitrogen fixation genes in free-living and symbiotic cells. J. Bacteriol. 169:1423-1432. Ueno-Nishio, S., S. Mango, L.J. Reitzer, and B. Magasanik. 1984. Identification and regulation of the glnL operator-promoter of the complex glnALG operon of Escherichia coli. J. Bacteriol. 160:379- 384. Upchurch, R.G., and G.H. Elkan. 1978. Ammonia assimilation in Rhizobium japanicum colonial derivatives differing in nitrogen- fixing efficiency. J. Gen. Microbiol. 104:219-225. Yanisch-Perron, C., J. Vieira, and J. Messing. 1985. Improved M13 phage cloning vectors and host strains: nucleotide sequences of the M13mp18 and pUCl9 vectors. Gene 33:103-119. SUMMARY The major conclusions of this dissertation are illustrated in Figure 1 and will be summarized here. 1. Two glutamine synthetase enzymes are present in Bradyrhizobium japonicum but either one alone provides glutamine prototrophy and enables the bacterium to establish an effective symbiosis with soybean. 2. Separate regulatory networks exist in Bradyrhizobium japanicum for activating expression of glnII in free-living and symbiotic cells. The ntrC gene product (NtrC) is involved in glnII transcription in aerobic free-living cells while another positive regulator, probably NifA, fulfills this role in oxygen-starved and symbiotic cells. 3. NtrC is also involved in nitrate utilization. Hhether this is by direct regulation of the nitrate reductase promoter (P ) or nar indirectly via a repressor is unknown. 4. The Bradyrhizobium japanicum glnA gene is not regulated in response to nitrogen status and, more specifically, is not under NtrC or NifA control. 132 133 Free-living state Nitrogen starvation Nitrogen excess I NtrC NtrA : Pglnu 1 WM —0 F | ’ .—o b I ’ '—. > I PglnA I ’ | I I I I I I I I I I I I I I.— I I I I l l I I I l I I I Symbiotic state nitro en starvation or NtrC g P InB symbiotic signal L. 29 > NtrA ' PglnII :—O > L Pnit A. NifA '°“' 02 fi —-0 activation - ° ° ° hypothetical Figure 1. Model of nitrogen regulation (Ntr) in Bradyrhizobium japonicum. 134 5. A gene sharing sequence similarity with the enteric nitrogen regulatory gene, glnB, is present in Bradyrhizobium japanicum and is regulated at tandem promoters in response to nitrogen status. This regulation involves both NtrC repression (at promoter PI) and activation (at promoter P2). 6. The glnB gene is also activated at promoter P2 by NtrC in soybean nodules. The involvement of NtrC in nodule gene expression raises the possibility that, in contrast to prevailing views, bacteroids do not sense excess nitrogen. 11111111