INTRODUCTION

Top Abstract Introduction Materials & Methods Results Discussion References

Changes in the availability of nitrogen result in altered gene expression in microorganisms. When bacterial growth is limited by the supply of nitrogen, the expression of genes required for the transport and catabolism of nitrogen-containing compounds is elevated. This altered gene expression, referred to as nitrogen regulation, increases the degradation of nitrogen-containing compounds and results in the production of glutamate and glutamine, the major nitrogen donors in cellular biosynthesis (19).

In enteric bacteria, the activation of gene expression during nitrogen restriction is mediated by the two-component Ntr regulatory system (18). A nitrogen regulatory system analogous to the enteric Ntr system is not present in the gram-positive sporulating soil bacterium Bacillus subtilis (31); instead, the TnrA regulatory protein activates the expression of many genes during nitrogen-limited growth in B. subtilis (38). These nitrogen-regulated gene products include a putative ammonium permease located in the nrgAB operon, the -aminobutyrate permease (gabP), the nitrate assimilatory proteins (nasA, nasBCDEFG), urease (ureABC), and Kipl, an inhibitor of kinase A that lies within the ycsFGI-kipIAR-ycsK operon (36-38). In addition, TnrA represses expression of the glnA gene, which encodes glutamine synthetase, during nitrogen-limited growth (38).

The TnrA-regulated nrgAB, gabP P2, and nas promoters all contain a common inverted repeat (TGTNAN₇TNACA; TnrA site) centered 49 to 51 bp upstream of their transcriptional start sites (Fig. 1) (7, 22, 38, 39). This same palindromic DNA sequence is centered 90 bp upstream of the nitrogen-regulated ureABC P3 promoter (37). Mutational analysis indicates that this conserved sequence is required for TnrA-dependent activation of the gabP and nas genes. Deletion of the TnrA site upstream of the gabP P2 promoter region prevents high-level gabP expression during nitrogen-limited growth (7). Replacement of two of the conserved nucleotides in the TnrA site of the divergently transcribed nas promoter region abolishes nitrogen regulation of the nasA and nasBC genes (22).

View larger version (16K): [in this window] [in a new window]   FIG. 1.   Alignment of nitrogen-regulated promoters. Nucleotides corresponding to the conserved upstream inverted repeat are boxed. The 35 region of the nrgAB promoter is overlined.

A second regulatory protein, GlnR, also contributes to nitrogen regulation in B. subtilis. During growth in the presence of excess nitrogen, GlnR represses the expression of the glnRA and ureABC operons (32, 37). The TnrA and GlnR proteins are homologs that have extensive sequence similarity within their proposed DNA-binding domains (38). Moreover, the two GlnR operators in the glnRA promoter contain the conserved TnrA-binding-site sequence (38). Expression of glnRA is negatively regulated by TnrA (38). These observations suggest that TnrA and GlnR bind to DNA sites with similar sequences. GlnR does not regulate expression of the nrgAB, gabP, or nas genes (3, 22).

In Salmonella typhimurium, the first signal of nitrogen limitation is a drop in the internal glutamine pool (13). The signals regulating the activity of the TnrA and GlnR proteins have not been identified. Several experimental observations argue that this signal does not involve the intracellular levels of glutamine. First, the B. subtilis glnA22 mutant, which expresses the glnRA operon and all other known nitrogen-regulated genes constitutively during growth on excess nitrogen (3, 8), has intracellular glutamine pools which are sixfold higher than those of wild-type cells (8). In addition, glutamine does not affect the in vitro binding of GlnR to its operators (5). Because constitutive expression of TnrA- and GlnR-regulated genes occurs in glnA mutants (3, 22, 33, 37, 38), glutamine synthetase is required for the synthesis and/or transduction of the nitrogen regulatory signal(s) to the TnrA and GlnR proteins.

In this study, mutational analysis and gel mobility shift DNA-binding assays were used to demonstrate that the conserved inverted repeats located upstream of the nrgAB and gabP P2 promoters function as TnrA-binding sites.

	MATERIALS AND METHODS

Top Abstract Introduction Materials & Methods Results Discussion References

Bacterial strains and plasmids. The B. subtilis strains used in this study are all derivatives of strain 168 (trpC2). Strain SF15W contains a chloramphenicol-resistant mini-Tn10 insertion in the B. subtilis ptsI gene. This insertion was isolated during a search for mutations that prevented the high-level expression of nrgAB during nitrogen-limited growth. This mutant was isolated by the same procedure used to isolate the tnrA62::Tn917 mutation (38) except that Tn10 insertion libraries constructed with plasmid pHV1249 (28) were used in place of Tn917 libraries. DNA adjacent to the Tn10 insertion was cloned by plasmid rescue (35) and sequenced by using an oligonucleotide primer complementary to the ends of the Tn10 transposon.

Cell growth, media, and enzyme assays. The methods used for bacterial cultivation in minimal medium and NSM (25) have been previously described (4). Overnight NSM cultures were always grown for four generations in fresh medium, and the resulting logarithmic NSM culture was used to inoculate NSM cultures harvested for enzyme assays. Liquid minimal cultures were grown in the morpholinepropanesulfonic acid (MOPS) minimal medium of Neidhardt et al. (23). Glucose was added at 0.5% to MOPS minimal medium and 1% to NSM medium. All nitrogen sources were added at 0.2% to MOPS minimal medium. The production of heat-resistant spores was examined as previously described (25).

Oligonucleotide mutagenesis. Plasmid pNRG406 contains the EcoRI-HindIII nrgAB promoter fragment from pNRG401 cloned into the oligonucleotide mutagenesis vector pALTER-1 (Promega Corp.). The EcoRI-HindIII gabP promoter fragment from pNRG704 was inserted into pALTER-1 to give plasmid pNRG710 (7). Oligonucleotide mutagenesis of pNRG406 and pNRG710 was conducted by using a protocol designed by the supplier of pALTER-1 (Promega Corp.). EcoRI-HindIII DNA fragments of the mutated promoters were cloned into pSFL6 for functional analysis.

Overexpression of TnrA. Plasmid pTNR14 was constructed by cloning the BsaJI-AseI DNA fragment from pJVK75 (14) containing the tnrA coding sequence into the EcoRV site of pBluescriptSK (Stratagene). The EcoRI-HindIII tnrA DNA fragment from pTNR14 was cloned into pET23+ (Novagen, Inc.) to give plasmid pTNR16. E. coli BL21(DE3) containing pTNR16 was grown in Luria-Bertani broth at 30°C to mid-log phase, where the expression of TnrA was induced by the addition of isopropyl--D-thiogalactopyranoside to 0.4 mM. Cells were harvested by centrifugation 3 h after induction, washed with TKE buffer (50 mM Tris [pH 7.5], 50 mM KCl, 1 mM EDTA), and stored frozen at 70°C. To prepare extracts, the frozen cells were thawed and resuspended in TKE buffer containing 1 mM phenylmethylsulfonyl fluoride, 1 mM dithiothreitol, and 10% glycerol. Cells were lysed by sonication, and a clarified extract was obtained by centrifugation at 16,000 × g for 15 min. Aliquots of the supernatant were frozen with liquid nitrogen and stored at 70°C. The same procedure with BL21(DE3) cells containing plasmid pET23+ was used to prepare an E. coli extract lacking TnrA.

Gel mobility shift DNA-binding assays. The DNA probe for DNA-binding assays was the 164-bp EcoRI-HindIII fragment from pNRG401 (containing the wild-type nrgAB promoter) or pNRG406G42 (containing the nrgAB promoter 42G mutation). The gel-purified DNA fragments were end labeled by filling in the sticky ends with the Klenow fragment of DNA polymerase I in the presence of [-³²P]dATP. The binding reaction volume of 20 µl contained 0.1 nM DNA probe, 50 mM HEPES (pH 7.5), 50 mM sodium acetate, 1 mM dithiothreitol, 1 mM EDTA, 100 ng of poly(dI-dC) · poly(dI-dC), 100 µg of bovine serum albumin per ml, 0.025% Triton X-100, 5% glycerol, and the indicated amount of protein from the TnrA overexpression extract. The binding reaction mixtures were incubated at 30°C for 20 min and loaded onto an 8% polyacrylamide gel (50 mM Tris [pH 8.9], 50 mM glycine, 1 mM EDTA). The gel was subjected to electrophoresis at 110 V for 2.5 h and dried, and bands were visualized by autoradiography.

	RESULTS

Top Abstract Introduction Materials & Methods Results Discussion References

Mutational analysis of the nrgAB inverted repeat. -Galactosidase expression from the single-copy chromosomal (nrgA-lacZ)416 transcriptional fusion increases over 2,000-fold during nitrogen-restricted growth (Table 1). In the presence of excess nitrogen, only background levels of -galactosidase expression occur. Nutritional regulation of nrgAB expression requires only the tnrA gene product (38). A large inverted repeat is located upstream of the nrgAB promoter between positions 59 and 39 (Table 1). Embedded within this upstream inverted repeat is the dyad symmetry element (TGTNAN₇TNACA) found upstream of other TnrA-regulated promoters (Fig. 1). The symmetric half-sites of these TnrA sites are interrupted by a 7-bp A+T-rich sequence. In addition, a 5- to 6-bp A+T-rich nucleotide tract lies upstream of these sites.

Effect of spacing mutations on nrgAB expression. The distance between the two half-sites of the TnrA inverted repeat was altered to determine whether this spacing is important for nrgAB regulation. Nitrogen regulation of nrgAB expression was almost completely eliminated by either the deletion of a single A nucleotide at position 48 (NRG441) or the insertion of a single A nucleotide at position 48 (NRG442) within the central region of the TnrA site (Table 2).

Deletion analysis of the nrgAB promoter region. A 5- to 6-bp A+T-rich sequence is located upstream of the TnrA sites in the nas, gabP P2, and nrgAB promoters (Fig. 1). Deletion analysis of the gabP P2 promoter indicated that this upstream sequence is required for high-level gabP P2 transcription in nitrogen-restricted cultures (7). To determine whether the tract of five A nucleotides located between 61 and 65 in the nrgAB promoter region is required for high-level nrgAB expression during nitrogen-limited growth, various truncated versions of the nrgAB regulatory region were constructed. In these constructs, increasing amounts of the sequence upstream of the nrgAB TnrA site were deleted and replaced with the sequence CCCGGG. Deletion to position 67 (NRG477) leaves the upstream A sequence intact and has no effect on nrgAB regulation (Table 2). In contrast, deletions to 65 (NRG475) and 63 (NRG474) remove a portion of the A tract and reduce the level of nrgAB activation two- and fivefold, respectively (Table 2). The remaining deletions (NRG467, NRG468, NRG469, NRG470, and NRG471) all completely remove the upstream A tract and cause a 12- to 20-fold reduction in the level of nrgAB activation (Table 2). These results indicate that the A tract upstream of the TnrA site is required for optimal nrgAB expression during nitrogen-limited growth.

Gel mobility shift DNA-binding assays. To determine if the nrgAB promoter is the direct target of the TnrA protein, crude protein extracts from an E. coli strain engineered to overproduce TnrA were used to conduct gel mobility shift DNA-binding experiments with DNA fragments containing either the wild-type or mutated TnrA site of the nrgAB promoter. While DNA containing the wild-type nrgAB promoter had reduced mobility in the presence of the TnrA overexpression extract (Fig. 2), no retardation was observed with an extract that lacked TnrA (data not shown). The DNA fragment containing a C-to-G mutation at 42 (NRG416-42G) exhibited no significant binding even in the presence of 5 µg of the TnrA extract (Fig. 2). Since binding of TnrA to the wild-type nrgAB promoter fragment could be observed with 8 ng of extract, these results indicate that TnrA has at least a 625-fold-higher affinity for the wild-type TnrA site than the mutated site. The binding of TnrA to the wild-type nrgAB DNA fragment was not altered in the presence of either 5 mM glutamine, 5 mM glutamate, or 5 mM NH₄Cl (data not shown). This result suggests that these compounds are unlikely to function as a nitrogen signal that regulates the DNA-binding activity of TnrA.

View larger version (48K): [in this window] [in a new window]   FIG. 2.   Retardation analysis of wild-type and mutant nrgAB promoter regions by TnrA. A 164-bp nrgAB promoter DNA fragment containing the B. subtilis wild-type TnrA-binding site (lanes 1 to 6) and the mutant TnrA-binding site from NRG416-42G (lanes 7 to 12) was used in the binding reactions. Total amounts of protein from TnrA overexpression extracts present in the reaction mixtures: lanes 1 and 7, 0 µg; lanes 2 and 8, 8 ng; lanes 3 and 9, 40 ng; lanes 4 and 10, 200 ng; lanes 5 and 11, 1.0 µg; lanes 6 and 12, 5.0 µg.

Mutational analysis of the gabP P2 upstream inverted repeat. High-level expression of gabP during nitrogen restriction primarily results from TnrA-dependent activation of transcription from the gabP P2 promoter (7). To demonstrate that the putative TnrA site in the gabP P2 region is required for TnrA regulation of gabP expression, single and double symmetric mutations were introduced into conserved nucleotides in both half-sites of the gabP P2 TnrA site. Expression of the wild-type gabP P2 promoter (NRG706) increases 13-fold during nitrogen-limited growth (Table 3). A single nucleotide replacement at position 58 (NRG715) reduces nitrogen regulation of the gabP P2 promoter twofold, while no nitrogen regulation is observed when a T is introduced into position 44 (NRG732 and NRG733) (Table 3). Single nucleotide substitutions at positions 55 (NRG735) and 45 (NRG714) reduced the level of gabP P2 activation during nitrogen limitation 6.5- and 3-fold, respectively (Table 3). A double symmetrical mutation at these positions (NRG731) abolished nitrogen regulation (Table 3).

Expression of the nrgAB-lacZ fusion in a ptsl mutant strain. During a search for gene products that regulate nrgAB expression, we isolated a Tn10 transposon insertion mutant, SF15W, which was unable to activate the expression of the nrgAB-lacZ fusion during nitrogen-limited growth with glucose as the carbon source. Cloning and sequencing experiments revealed that the transposon insertion occurred in the 65th codon of the ptsl gene that encodes the enzyme I component of the phosphoenolpyruvate:carbohydrate phosphotransferase transport system (PTS) (10). Since enzyme I is required for the PTS-dependent phosphorylation and transport of PTS sugars across the membrane, ptsl mutants are completely deficient in the PTS-mediated transport of glucose (24). As expected, the ptsl15::Tn10 mutant cultures grow slowly in glucose minimal medium containing excess nitrogen (glutamate plus ammonium) compared to wild-type cultures (Table 4). The residual growth of the ptsl mutant in the glucose minimal medium presumably results from PTS-independent transport of glucose (27).

View this table: [in this window] [in a new window]   TABLE 4.   -Galactosidase levels in extracts of wild-type and pstI mutant strains grown in various media

Expression of the nrgAB-lacZ fusion in nutrient broth sporulation medium. Wang et al. (36) have reported that expression of the TnrA-regulated operon encoding Kipl is induced by glucose when cells are grown in NSM. To determine if this mode of regulation is also exhibited by other TnrA-regulated genes, nrgAB expression was examined by determining -galactosidase levels from strain SF407 [(nrgA-lacZ)407] harvested at various times during growth in NSM. No expression of nrgAB was observed in wild-type cells during exponential- or early stationary-phase growth in NSM (Fig. 3). In contrast, when the NSM contained 1% glucose, the culture grew to a higher cell density and nrgAB expression increased 25-fold during late exponential-phase growth (Fig. 3). No activation of nrgAB expression is observed when glutamine, one of the best nitrogen sources for B. subtilis, is added to NSM cultures containing glucose (data not shown). Expression of the (nrgA-lacZ)407 lacZ fusion could not be detected in a tnrA mutant strain grown in NSM containing 1% glucose (Fig. 3). This result indicates that the activation of nrgAB expression in NSM containing 1% glucose requires the TnrA protein.

View larger version (20K): [in this window] [in a new window]   FIG. 3.   Growth and -galactosidase expression from the (nrgA-lacZ)407 fusion in wild-type and tnrA cells during growth in either NSM or NSM containing 1% glucose. Samples were removed periodically, and -galactosidase activity was assayed in cell extracts. Data from a typical experiment are shown. Triangles, SF407 [(nrgA-lacZ)407] cells grown in NSM plus glucose; squares, SF407T [(nrgA-lacZ)407 tnrA62] cells grown in NSM plus glucose; circles, SF407 [(nrgA-lacZ)407] cells grown in NSM. Open symbols, Klett units; closed symbols, -galactosidase specific activity.

	DISCUSSION

Top Abstract Introduction Materials & Methods Results Discussion References

Several lines of evidence indicate that TnrA-dependent activation of the nrgAB, nasA, nasBC, and gabP genes requires the common palindromic sequence located upstream of their promoters. Mutational analysis has shown that conserved nucleotides within this sequence are required for high-level expression of the nrgAB, gabP P2, and nas promoters during nitrogen-limited growth. Nucleotide replacements at two of the four conserved positions in the nrgAB half-sites reduced the level of nrgAB activation during nitrogen-limited growth. Mutations at the corresponding positions in the TnrA sites for the gabP P2 and nas promoters also abolish nitrogen regulation (22). Interestingly, single mutations at the innermost conserved nucleotide in the TnrA half-sites for the nas and gabP P2 promoters reduced transcriptional activation, while double symmetric mutations at these positions were required to inhibit nitrogen regulation of nrgAB expression. One explanation for this result is that a transcriptional activation assay was used to examine the in vivo effect of TnrA site mutations. Since transcriptional activation involves two TnrA-dependent steps, DNA binding and stimulation of RNA polymerase, mutations in an optimal TnrA site which only partially reduces TnrA binding may still be capable of activating transcription to wild-type levels. Thus, if the nrgAB TnrA site is an optimal site (compared to the nas and gabP P2 TnrA sites), then only those mutations in the nrgAB TnrA site with a significantly lower affinity for TnrA may reduce the level of nrgAB activation observed in vivo. In this regard, it is worth noting that Morett et al. (21) identified mutations in the NifA-binding site of the nifH promoter that have reduced affinity for NifA but are still capable of activating nifH expression.

A gel mobility shift DNA-binding assay was used to show that the TnrA protein in crude protein extracts has a 625-fold-higher affinity for the wild-type TnrA site than for a mutant TnrA site with a C-to-G mutation at position 42. The reduced in vitro affinity of the TnrA protein for the mutant TnrA site correlates well with the low level of nrgAB activation by this mutant site in vivo. The twofold symmetry of the TnrA site suggests that TnrA most likely binds to this site as a dimeric protein. Since the DNA sequence between the two TnrA half-sites (TGTNA and TNACA) is always 7 bp, this spacing must position the two half-sites into the alignment required for recognition by TnrA. Indeed, mutations which increase or reduce the spacing between the half-sites by a single nucleotide abolished the activation of nrgAB expression. Although this 7-bp central sequence is A+T rich, there is no conserved nucleotide sequence between the TnrA half-sites of the nrgAB, gabP P2, or nas promoters. Since replacement of the A+T-rich central base pairs in the nrgAB TnrA site with a G+C-rich sequence reduced the level of nrgAB activation, this A+T-rich central sequence appears to enhance TnrA binding in vivo. Similar results have been observed for the 434 repressor, which binds to an operator with an A+T-rich central sequence (15). In this case, because there is no evidence for direct interaction between the central nucleotides and the 434 repressor protein, the A+T-rich central base pairs most likely increase the affinity of the 434 repressor for its binding site by influencing DNA conformation.

Deletion analysis showed that an upstream A+T-rich sequence is required for optimal expression of the nrgAB and gabP P2 promoters during nitrogen-limited growth. Similar A+T-rich sequences adjacent to the binding sites for the catabolite gene activator and Lex proteins have been shown to increase affinity for these sites by promoting protein-induced DNA bending (9, 16). It is possible that the A+T-rich sequences upstream of the nrgAB and gabP P2 promoters play a similar role in TnrA activation of the transcription from these promoters. An alternative explanation is that maximal nrgAB transcription requires both TnrA binding and an interaction between the alpha subunit of RNA polymerase and the A+T-rich sequences located upstream of the nrgAB promoter (29).

TnrA belongs to the MerR family of DNA-binding positive regulatory proteins which includes MerR, SoxR, TipA_L, and BmrR (1, 11, 12, 20). These four proteins activate transcription by binding to inverted repeats located between the 35 and 10 regions of promoters that have suboptimal 19-bp (rather than 17-bp) spacing between the 35 and 10 regions. It has been shown in vitro that binding of the MerR protein to the mer P_T promoter causes an alteration of the DNA structure located between the 35 and 10 promoter regions (26). This resulting distortion of the DNA reorients the conserved promoter regions to optimal spatial positions that facilitate RNA polymerase binding (2). A number of observations indicate that TnrA utilizes a different mechanism for transcriptional activation than does MerR. First, unlike the MerR DNA-binding site, the TnrA sites of the nrgAB, nasA, nasBC, and gabP P2 promoters are all located upstream of the 35 region for these promoters. Second, the observation that high-level activation of the nrgAB expression occurs only when the TnrA site is centered 49 to 51 bp upstream of the transcriptional start site argues that the relative alignment of the TnrA site and the 35 region of the nrgAB promoter is critical for transcriptional activation. Finally, the nucleotide sequence of the 35 and 10 regions of the mer P_T promoter are highly similar to the consensus sequences for these regions. In contrast, the promoters for TnrA-regulated genes generally contain multiple mismatches within the 35 and 10 consensus sequences for promoters transcribed by the ^A form of RNA polymerase. Taken together, these observations suggest that TnrA activates transcription by directly interacting with and facilitating the binding of RNA polymerase to the promoter region.

The ptsl mutant strain is able to activate nrgAB expression during nitrogen-limited growth on malate, a carbon source that is not transported by the PTS. This result indicates that the PTS is not required for the activation of nrgAB expression and suggests that the ptsl mutation indirectly affects nrgAB expression in glucose-grown cultures. Because growth of the ptsl mutant is carbon limited in glucose minimal medium, depletion of intracellular nitrogen pools may not occur when the ptsl mutant is grown in glucose minimal medium containing glutamate as the nitrogen source. This hypothesis would explain why no activation of nrgAB expression is observed in the ptsl mutant grown on this medium.

Examination of the -galactosidase expression from the nrgAB-lacZ fusion showed that nrgAB expression is activated by TnrA in cultures grown in NSM containing glucose but not in NSM cultures grown in the absence of glucose. Since nrgAB is expressed at similar levels in minimal media with glutamine as the nitrogen source and either glucose or citrate as the carbon source (Table 4), nrgAB expression is not regulated in response to carbon availability. Thus, it seems unlikely that glucose directly induces nrgAB expression in cultures grown in NSM containing glucose. Because the glucose-grown NSM cultures reach a higher final cell density than NSM cultures, the simplest explanation for the activation of nrgAB expression in NSM containing glucose is that the increased cell growth depletes the nitrogen compounds from the growth medium and that this results in nitrogen-limited growth and the activation of nrgAB expression. This hypothesis is supported by the observations that the activation of nrgAB expression in NSM containing glucose requires TnrA and that this activation is abolished by the addition of glutamine to this growth medium.