Dual Involvement of CbrAB and NtrBC in the Regulation of Histidine Utilization in Pseudomonas fluorescens SBW25

Abstract

Pseudomonas fluorescens SBW25 is capable of growing on histidine as a sole source of carbon and/or nitrogen. Previous work showed that the two-component regulatory system CbrAB is required for expression of the histidine utilization (hut) locus when histidine is the sole source of carbon and nitrogen. Here, using mutational analysis and transcriptional assays, we demonstrate involvement of a second two-component system, NtrBC. When histidine is the sole carbon source, transcription of the hutU operon is initiated from a σ⁵⁴-type promoter and requires CbrB (an enhancer binding protein for σ⁵⁴-recruitment). However, when histidine is the sole nitrogen source, the hutU operon is transcribed from a σ⁷⁰-type promoter and requires either CbrB or the nitrogen regulator, NtrC. No role was found for the SBW25 homolog of the nitrogen assimilation control protein (NAC). Biolog phenotypic microarray analysis of the ability of the three mutants (ΔcbrB, ΔntrC, and ΔcbrB ΔntrC) to utilize 190 carbon and 95 nitrogen substrates confirmed the central regulatory roles of CbrAB and NtrBC in cellular carbon and nitrogen catabolism: deletion of cbrB abolished growth on 20 carbon substrates; deletion of ntrC eliminated growth on 28 nitrogen substrates. A double cbrB–ntrC mutant was unable to utilize a further 14 nitrogen substrates (including histidine, proline, leucine, isoleucine, and valine). Our data show that CbrAB plays a role in regulation of both carbon and nitrogen catabolism and maintains activity of catabolic pathways under different C:N ratios.

A coordinated response for the assimilation of carbon and nitrogen ensures that organisms maintain fitness in nutritionally complex environments. Amino acids serve as good carbon and nitrogen sources for many bacteria. The regulation of amino acid catabolism can therefore provide insights into the molecular mechanisms involved in acclimation to substrates that provide both carbon and nitrogen.

Expression of amino acid catabolic enzymes is controlled by both specific induction (presence or absence of the substrate) and general induction (mediated by global regulatory proteins that sense the physiological status of the cell). Specific induction of histidine utilization (hut) in Gram-negative bacteria is mediated by the HutC repressor (Magasanik 1978; Hu et al. 1989; Zhang and Rainey 2007), which binds to the operator sites of the hut promoter region (Hu et al. 1989). Repression is relieved by urocanate, the first intermediate of the histidine degradation pathway, which interacts with the HutC repressor (Hu et al. 1989) and presumably causes HutC to dissociate from the operator site. Histidine-induced expression of hut is achieved by intracellular conversion of histidine to urocanate, which is catalyzed by histidine ammonia-lyase (the product of hutH).

General induction of hut enzymes is complex and has been studied exclusively in enteric bacteria (Magasanik 1978; Bender 1991; Janes et al. 2003). In Salmonella typhimurium and Klebsiella aerogenes, the activators of hut differ depending upon whether histidine is a source of carbon or a source of nitrogen. When histidine is utilized as a carbon source, hut transcription is activated by the catabolite-activating protein (CAP) charged with cAMP (Nieuwkoop et al. 1984). However, when histidine is utilized as a nitrogen source, transcription of hut is activated by nitrogen assimilation control protein (NAC) from a σ⁷⁰-promoter; in turn, expression of NAC is controlled by the two-component NtrBC (nitrogen regulation) system in a σ⁵⁴-dependent manner (Bender 1991; Janes et al. 2003). NtrB possesses both kinase and phosphatase activities (Pioszak and Ninfa 2004), the relative balance of which determines the activity state of the response regulator NtrC. Under conditions of nitrogen limitation intracellular glutamine levels decline relative to 2-ketoglutarate. This causes uridylation of PII, inhibition of NtrB phosphatase activity, and activation of NtrC. Phosphorylated NtrC then activates RNA polymerase carrying σ⁵⁴ leading to expression of various genes, including the LysR-type transcriptional regulator NAC (reviewed by Itoh et al. 2007).

In nonenteric Gram-negative bacteria, such as Pseudomonas, general induction of hut expression is not well understood. Succinate-induced catabolite repression (based solely on enzyme activity assays) has been reported (Hug et al. 1968; Phillips and Mulfinger 1981), although no role has been found for the global regulatory proteins Crc (catabolic repression control; Hester et al. 2000) and Vfr (virulence factor regulator, which is homologous to the E. coli CAP protein; Suh et al. 2002). Recent work has shown the involvement of a newly discovered two-component regulatory system, CbrAB (Nishijyo et al. 2001; Zhang and Rainey 2007). CbrA is a membrane-bound histidine protein kinase and CbrB is a σ⁵⁴-dependent response regulator. In P. fluorescens SBW25 growing on histidine as a sole carbon and nitrogen source transcription of the hutU operon is controlled by a σ⁵⁴-type promoter and requires σ⁵⁴ and functional CbrAB (Zhang and Rainey 2007).

To date studies on histidine catabolism in Pseudomonas have been conducted in environments where the amino acid is the sole source of carbon and nitrogen (Nishijyo et al. 2001; Li and Lu 2007; Zhang and Rainey 2007). In this study we report the regulation of hut expression in environments where histidine is either the sole carbon or the sole nitrogen source. We show that when histidine is the sole carbon source CbrAB is essential for hut activation; however, when histidine is the sole source of nitrogen, expression of hut can be activated by either CbrAB or NtrBC. We also demonstrate a global role for both CbrAB and NtrBC in the control of cellular carbon and nitrogen metabolism and provide insight into the biological significance of the dual control.

MATERIALS AND METHODS

Bacterial strains, plasmids, and growing conditions:

A summary of bacterial strains and plasmids used in this study is provided in Table 1. E. coli DH5αλ_pir was used for gene cloning. Both Pseudomonas and E. coli strains were grown in Luria–Bertani medium (LB) at 28° and 37°, respectively. Pseudomonas strains were also cultivated in minimal medium M9 (Sambrook et al. 1989): M9 salts medium (MSM) supplemented with glucose (0.4%) and ammonium chloride (1 mg ml⁻¹). Histidine was supplemented at a final concentration of 15 mm or as specifically indicated. When appropriate, antibiotics were supplemented at the following concentrations: tetracycline (Tc), 10 μg ml⁻¹; kanamycin (Km), 50 μg ml⁻¹; ampicillin (Ap), 50 μg ml⁻¹; nitrofurantoin (Nf), 100 μg ml⁻¹.

TABLE 1.

Bacterial strains, plasmids, and oligonucleotide primers used in this study

Strain, plasmid, or primer	Genotypes and relevant characteristics	Reference or application
P. fluorescensa
SBW25	Wild-type strain isolated from sugar beet	Thompson et al. (1995)
PBR809	SBW25ΔcbrA	Zhang and Rainey (2007)
PBR810	SBW25ΔcbrB	Zhang and Rainey (2007)
PBR829	SBW25ΔntrC	This study
PBR830	SBW25ΔntrC ΔcbrB∷Km, KmR	This study
PBR831	SBW25Δpflu4026	This study
PBR808	SBW25ΔrpoN	Jones et al. (2007)
PBR813	DUP(hutD-hutU)∷pUIC3, the hutU-‘lacZ fusion strain of wild-type SBW25, TcR	Zhang and Rainey (2007)
PBR833	ΔntrC DUP(hutD-hutU)∷pUIC3, hutU-‘lacZ, TcR	This study
PBR834	ΔcbrB DUP(hutD-hutU)∷pUIC3, hutU-‘lacZ, TcR	This study
PBR835	ΔntrC ΔcbrB∷Km DUP(hutD-hutU)∷pUIC3, hutU-‘lacZ, TcR KmR	This study
PBR836	Δpflu4026 DUP(hutD-hutU)∷pUIC3, hutU-‘lacZ, TcR	This study
PBR837	ΔrpoN DUP(hutD-hutU)∷pUIC3, hutU-‘lacZ, TcR	This study
PBR838	DUP(pflu5235-cbrA)∷pUIC3, the cbrA-‘lacZ fusion strain of wild-type SBW25, TcR	This study
PBR839	DUP(cbrA-cbrB)∷pUIC3, the cbrB-‘lacZ fusion strain of wild-type SBW25, TcR	This study
Plasmid
pRK2013	Helper plasmid, Tra+, KmR	Ditta et al. (1980)
pCR8/GW/TOPO	Cloning vector, SpR	Invitrogen
pUC4K	Donor of KmR gene, KmR ApR	Taylor and Rose (1988)
pUIC3	Integration vector with promoterless ‘lacZ, Mob+, TcR	Rainey (1999)
pUIC3-8	pUIC3∷ (hutD-hutU), the hutU-‘lacZ fusion plasmid, TcR	Zhang and Rainey (2007)
pUIC3-42	pUIC3∷(pflu5235-cbrA), the cbrA-‘lacZ fusion plasmid, TcR	This study
pUIC3-43	pUIC3∷(cbrA-cbrB), the cbrB-‘lacZ fusion plasmid, TcR	This study
pUIC3-46	pUIC3 containing a 1.4-kb fragment with cbrB deletion, TcR	This study
pUIC3-48	pUIC3-46 carrying KmR gene in the middle of the insert	This study
pUIC3-53	pUIC3 containing 1.6 kb DNA with ntrC deletion, TcR	This study
pUIC3-54	pUIC3 containing 1.6 kb DNA with pflu4026 deletion, TcR	This study
pCM639	ISphoA/hah-Tc delivery plasmid, TcR	Bailey and Manoil (2002)
pME6032	Broad-host-range cloning vector, TcR	Heebet al. (2000)
pME6032-cbrB	Plasmid construct for cbrB complementation, TcR	This study
pME6032-ntrC	Plasmid construct for ntrC complementation, TcR	This study
Primerb
ntrC1	GAGATCTTGCTGCTGCTGGAAGTGC	ntrC deletion
ntrC2	cagcatgcggatccgttgacggaCGACGATCCAGACGGTTTCAC	ntrC deletion
ntrC5	tccgtcaacggatccgcatgctgGAATCTCCGACCCACTGAAGA	ntrC deletion
ntrC6	GAGATCTGCAGCAGGTCGAGACTGTG	ntrC deletion
cbrB-FB	GAAGATCTGATCGTCGAAGATGAAGGT	cbrB deletion/cbrB-‘lacZ
cbrB-4	cagcatgcggatccgttgacggaCTTTGCGGATCTTGCTGTAG	cbrB deletion
cbrB-5	tccgtcaacggatccgcatgctgAGCCGCAAATGCTTATGGGAA	cbrB deletion
cbrB-6	GAAGATCTCGGCCATAACGAAGAATGAC	cbrB deletion
NAC5	GAGATCTACCACCAGGTTGATGATGGAC	pfu4026 deletion
NAC6	cagcatgcggatccgttgacggaCTTCAATCGACGCAGGTTCA	pfu4026 deletion
NAC3	tccgtcaacggatccgcatgctgTGAGCTGATCATGAACGGCC	pfu4026 deletion
NAC4	GAAGATCTATGTCCACGGCCGCCGTTGG	pfu4026 deletion
cbrA-FB	GAAGATCTACCTGCAGGAACTGCTCG	cbrA-‘lacZ
cbrA-RB	GAAGATCTGGAAAGAAAGCCATAGCCG	cbrB-‘lacZ
cbrB-RB	GAAGATCTGAAGCATCTCGTCATGGTC	cbrB-‘lacZ
cbrB-F	CTTGCACTGGTCTATTCCAT	cbrB complementation
cbrB-R	ACATCTGTGGGTGTTACCTG	cbrB complementation
ntrC-F	GAGATCTACGACCCGAGCATTCCCGAC	ntrC complementation
ntrC-R	GAGATCTGTGGCTGATATACCGTCA	ntrC complementation

^aThe ‘lacZ fusion strains were constructed by cloning the DNA fragment (∼800 bp) into the delivery vector pUIC3 followed by integration into the genome by insertion–duplication. Thus the related gene function was not affected.

^bThe primer sequences are shown from 5′ to 3′. Restriction sites incorporated into the primers are underlined. Complementary sequences designed for the SOE–PCR are shown in lowercase type.

Growth of P. fluorescens SBW25 and mutants was assessed by using a VersaMax microtiter plate reader (Molecular Devices, Sunnyvale, CA). To ensure strains in growth curve comparisons were physiologically equal, they were inoculated directly from bacterial stocks that had been stored at −80°, grown in LB broth, and then subcultured once in M9 broth. To set up the growth experiments, cells in M9 were collected by centrifugation and washed once using the same volume of sterile water. Two microliters of the washed cells were inoculated into 200 μl of the test medium per well (in a 96-well microtiter plate). Turbidity was measured at a wavelength of 450 nm and the data were collected at 5-min intervals using SOFTmax PRO software (Molecular Devices) over a period of ∼48 hr.

Strain construction:

Recombinant DNA techniques were performed according to standard protocols (Sambrook et al. 1989). Gene mutation was achieved by splicing by overlapping extension using the polymerase chain reaction (SOE–PCR) (Horton et al. 1989) in conjunction with a two-step allelic exchange strategy, using the suicide integration vector pUIC3 (Rainey 1999). Four oligonucleotide primers were designed for deletion of each gene (ntrC, cbrB, or pflu4026) and the primer sequences are given in Table 1. To construct the double deletion mutant of the ntrC and cbrB genes, a Km^R gene was retrieved from pUC4K by BamHI digestion and then inserted into the middle of pUIC3-46, which harbors a 1.4-kb fragment containing two ∼700-bp DNA regions flanking the coding region of cbrB cloned in pUIC3. The resulting recombinant plasmid pUIC3-48 was conjugated into PBR829 (ΔntrC) with the help of pRK2013 (Tra⁺) and the allelic exchange mutant was directly selected on LB agar plates supplemented with Nf (to counterselect E. coli), Km, and X-Gal.

Complementation of the cbrB and ntrC was performed by cloning the PCR-amplified coding region of cbrB or ntrC into pME6032 (Heeb et al. 2002), a plasmid vector that can replicate in Pseudomonas. Primers used for cbrB and ntrC amplification are listed in Table 1. The resulting plasmid pME6032-cbrB or pME6032-ntrC was introduced into PBR830 (ΔntrC ΔcbrB) by conjugation with the help of pRK2013 (Tra⁺).

To construct the promoterless lacZ (‘lacZ) fusions to cbrA and cbrB, the cbrA and cbrB promoter regions were PCR amplified from the genomic DNA by using primer pairs cbrA-FB/cbrA-RB and cbrB-FB/cbrB-RB, respectively (Table 1). The ∼700-bp DNA fragments were then cloned into pUIC3 by the BglII site in front of the promoterless ‘lacZ (Rainey 1999). The resulting plasmids pUIC3-42 and pUIC3-43 carrying cbrA- and cbrB-‘lacZ, respectively, were mobilized into P. fluorescens SBW25 by conjugation with the help of pRK2013. Integration into the genome by insertion–duplication was ensured by selecting transconjugants on LB plates supplemented with Nf and Tc.

β-Galactosidase activity was assayed as previously described using 4-methylumbelliferyl-β-d-galactoside (4MUG) as the enzymatic substrate (Zhang et al. 2006). The product (7-hydroxy-4-methylcoumarin, 4MU) was detected using a Hoefer DyNA Quant 200 fluorometer (Pharmacia Biotech) following the manufacturer's instructions. The reaction was monitored at 460 nm with an excitation wavelength of 365 nm. The enzyme activity was expressed as the amount of 4MU produced per minute and per cell: amol (attomole) 4MU min⁻¹ cell⁻¹ (1 amol = 10⁻¹⁸ mol).

Rapid amplification of cDNA 5′ ends:

Transcriptional start sites of the hutF and hutU-G operons were determined using the rapid amplification of cDNA 5′ ends (5′-RACE) system (Invitrogen, Carlsbad, CA) as previously described (Zhang and Rainey 2007). Total RNA was prepared from cells grown in MSM supplemented with histidine and ammonium chloride or glucose and histidine.

Transposon mutagenesis analysis:

Plasmid pCM639 containing ISphoA/hah-Tc was introduced from E. coli SM10λ_pir (Bailey and Manoil 2002) into P. fluorescens PBR830 using a standard protocol of conjugative transfer. Transconjugants were selected on MSM agar plates supplemented with glucose, histidine, Km, and Tc. The genomic location of the transposon insertions was determined by sequencing the DNA products of arbitrary primed (AP)–PCR as previously described (Manoil 2000).

Phenotype microarrays (PMs) analysis:

Growth phenotypes of the ntrC and/or cbrB mutants were assessed using the Biolog PM1, PM2, and PM3 plates (Biolog, Hayward CA). They are 96-well microtiter plates with each well containing defined medium with a unique carbon (PM1 and PM2) or nitrogen (PM3) compound plus an indicator dye for cell respiration. Excluding a carbon-free well (negative control) for each plate, the PM1 and PM2 Biolog assays assess the ability of a bacterium to utilize 190 carbon compounds as the sole carbon source (ammonium is the nitrogen source), while the PM3 assay assesses utilization of 95 nitrogen sources (succinate is the carbon source). Experiments were performed following the manufacturer's instructions. Inoculant was prepared by first growing the cells stored in a −80° freezer in LB broth with one subculture in R2A broth. Cells were collected by centrifugation and resuspended in the same volume of sterile deionized water. A total of 100 μl of this cell suspension was inoculated into the Biolog inoculating fluid (IF-0, 15 ml) and incubated at 28° for 2 hr to starve the cells; 100 μl of the starved cells were transferred into each well and incubated at 28° for 48 hr. The final and initial optical density (A₆₆₀) was measured using a VersaMax microtiter plate reader (Molecular Devices). Consistency of the absorbance data was checked by visual inspection. Two replicates were conducted for each strain.

RESULTS

The role of CbrAB in histidine utilization:

To investigate the regulatory roles of CbrAB in histidine utilization, transcription of cbrAB was measured using chromosomally integrated ‘lacZ fusions to cbrA and cbrB. β-Galactosidase assays were conducted on cells grown in MSM supplemented with: (1) glucose and ammonium, (2) histidine (histidine is the sole source of carbon and nitrogen), (3) histidine and glucose (histidine is the sole source of nitrogen), and (4) histidine and ammonium (histidine is the sole source of carbon). Results are shown in Figure 1. Transcription of cbrA, which is consistently lower than cbrB, remained at similar levels for cells growing in the four media. cbrB expression was elevated threefold in cells grown in MSM supplemented with histidine compared to cells grown on MSM supplemented with glucose and ammonium. This result is consistent with the previous finding that cbrAB is required for the utilization of histidine when histidine is the sole source of carbon and nitrogen (Zhang and Rainey 2007). Interestingly, cbrB expression was also elevated approximately fourfold in cells grown on MSM supplemented with either glucose and histidine, or histidine and ammonium, suggesting the involvement of cbrAB in hut regulation when histidine is utilized as the sole source of carbon and as the sole source of nitrogen.

Figure 1.—

Genetic organization (A) and expression (B) of cbrAB from P. fluorescens SBW25. (A) Position of the lacZ fusions is indicated by small circles and orientation is shown by the attached arrow. (B) β-galactosidase activities (amol 4MU min⁻¹ cell⁻¹) were measured in cbrA- and cbrB-‘lacZ fusion strains PBR838 and PBR839, respectively. Bacteria were grown in MSM supplemented with (1) glucose (Glc) and ammonium, (2) histidine (His), (3) histidine and ammonium, and (4) glucose and histidine. Values are means and standard errors of three independent cultures. Two-way ANOVA revealed a significant difference among means [F_1,20 = 16.54, P < 0.001]. Bars that are not connected by the same letter (shown above each) are significantly different (P < 0.05) by Tukey’s HSD test.

To test the role of cbrAB in histidine utilization when histidine is the sole source of carbon and, in separate experiments, the sole source of nitrogen, the growth of mutants PBR809 (ΔcbrA) and PBR810 (ΔcbrB) was examined in MSM supplemented with histidine (15 mm) plus various concentrations of either ammonium (where histidine is the sole carbon source) or succinate (where histidine is the sole nitrogen source).

PBR809 and PBR810 were unable to grow in MSM plus histidine; the addition of ammonium to a final concentration of 0.1, 0.5, or 1 mg ml⁻¹ did not restore the growth defect (the growth curves remained flat over a period of 48 hr). PBR809 and PBR810 were nevertheless capable of growth on glutamate (the end product of the histidine degradation pathway) where glutamate was the sole source of carbon (MSM plus glutamate and ammonium). These results show that cbrA and cbrB are required for hut activation when histidine is a sole carbon source.

When histidine is the sole nitrogen source, growth of PBR809 (ΔcbrA) was proportional to the concentration of succinate (Figure 2). PBR810 (ΔcbrB) displayed growth patterns similar to PBR809 (only data from PBR809 are shown). Analysis of the growth of PBR809 in the presence of 5 mm succinate (Figure 2) showed that the histidine-dependent growth (μ_max = 0.408 ± 0.002/hr) ceased after 12 hr, suggesting that hut operon expression is repressed when succinate levels become exhausted [growth was restored when additional succinate (5 mm) was added (at the 24-hr time point, data not shown)]. Similar results were obtained when glucose replaced succinate in the growth medium (data not shown). These data show that cbrAB is not required when histidine is the sole nitrogen source, provided the environment contains an excess of carbon, which is surprising given that cbrB transcription was elevated in MSM supplemented with histidine and glucose (Figure 1B).

Figure 2.—

Growth of mutant PBR809 (ΔcbrA) on histidine as a sole source of nitrogen. Bacteria were grown in MSM supplemented with histidine (15 mm) and succinate at a final concentration of 0, 5, 10, or 20 mm. Results are means and standard errors of six independent cultures. Data were collected at 5-min intervals, but hourly time points are shown for clarity.

NtrC is involved in hut regulation:

The fact that PBR809 (ΔcbrA) and PBR810 (ΔcbrB) are capable of growth on histidine as a sole nitrogen source suggests the presence of another unknown regulator(s) that activates the hut operon in response to nitrogen starvation. A plausible candidate is NtrBC, a two-component signal transduction system that has been shown in other bacteria to play a central regulatory role in nitrogen assimilation (Reitzer and Schneider 2001; Ninfa and Jiang 2005). In silico analysis of the P. fluorescens SBW25 genome revealed a two-gene locus (pflu0344 and pflu0343) with significant similarity to the ntrBC genes of E. coli K12: Pflu0344 shows 45% and Pflu0343 shows 69% amino acid sequence identity to NtrB and NtrC from K12, respectively. This locus is hereafter referred to as SBW25 ntrBC.

The coding region of ntrC was deleted from wild-type SBW25 and the ability of the mutant strain PBR829 (ΔntrC) to grow on histidine was examined. Growth of PBR829 was comparable to wild type in MSM plus histidine and MSM plus histidine and ammonium. However, contrary to expectation, PBR829 (ΔntrC) grew normally on histidine as the sole nitrogen source [MSM plus succinate (or glucose) and histidine] indicating no role for NtrC in hut activation.

Data presented above showed that cbrB expression is elevated when growing on histidine as a sole nitrogen source (Figure 1). This led us to consider the possibility that CbrAB might activate the hut operon in the ΔntrC mutant and that NtrBC might regulate hut expression in the ΔcbrA mutant. A double mutant defective in cbrB and ntrC (the two response regulator components) was generated and its ability to grow on histidine as a sole carbon or nitrogen source was determined: the double mutant PBR830 (ΔntrC ΔcbrB) no longer grew on histidine as a sole nitrogen source (Figure 3A), whereas its growth on glutamate as a nitrogen source was comparable with the wild type (Figure 3B). The growth defect of PBR830 on histidine as a sole nitrogen source was fully restored by the introduction of a cloned copy of either cbrB or ntrC. In addition, PBR830 displayed the same His⁻ phenotype as PBR810 (ΔcbrB) when histidine was the sole source of carbon or the sole source of carbon and nitrogen.

Figure 3.—

Growth of mutants PBR829 (ΔntrC), PBR810 (ΔcbrB), and PBR830 (ΔntrC ΔcbrB) on histidine (A) or glutamate (B) as the sole source of nitrogen. Growth was measured in MSM supplemented with succinate (15 mm) and histidine, or glutamate (15 mm). Results are means and standard errors of six independent cultures. Data were collected at 5-min intervals, but two hourly time points are shown for clarity.

To explore the effects on hut transcription, levels of hutU-‘lacZ expression were examined in the following genetic backgrounds: wild type, ΔntrC, ΔcbrB, ΔntrC ΔcbrB, and ΔrpoN. β-Galactosidase activity was measured in cells grown in MSM with glucose and ammonium or glucose and histidine (15 mm). Results presented in Table 2 show that histidine-induced expression of hutU-‘lacZ was not significantly affected by ntrC deletion (PBR833 vs. PBR813), but was reduced in the cbrB (PBR834) mutant, although the hutU-‘lacZ fusion in PBR834 was significantly induced compared to the same strain grown in MSM supplemented with glucose and ammonium. This lower level of induction of hutU by histidine in the cbrB mutant is consistent with decreased growth of PBR810 (ΔcbrB) on histidine (as a sole nitrogen source) compared with wild-type SBW25 (Figure 3A). Significantly, histidine-induced hutU expression was not observed in ΔntrC ΔcbrB and ΔrpoN (fusion strains PBR835 and PBR837, respectively, Table 2); these two fusion strains were unable to grow in MSM plus glucose and histidine.

TABLE 2.

Histidine-induced expression of the hutU-G operon in different genetic backgrounds

Fusion strain	Genetic background	β-Galactosidase activitya
				Ammonium	Histidine	Histidine + Gln
PBR813	Wild type	1.04 ± 0.13 (a)	25.67 ± 5.84 (c)	20.71 ± 1.64 (c)
PBR833	ΔntrC	0.90 ± 0.18 (a)	22.20 ± 1.15 (c)
PBR834	ΔcbrB	0.63 ± 0.08 (a)	8.29 ± 3.72 (b)
PBR835	ΔntrC ΔcbrB	0.76 ± 0.37 (a)	0.68 ± 0.18 (a)	0.96 ± 0.26 (a)
PBR836	Δpflu4026	1.20 ± 0.14 (a)	23.23 ± 2.83 (c)
PBR837	ΔrpoN	0.43 ± 0.13 (a)	0.25 ± 0.02 (a)	0.26 ± 0.01 (a)

^aβ-Galactosidase activities (amol 4MU min⁻¹ cell⁻¹) were measured for cells growing in MSM supplemented with glucose (as C) and different N source: ammonium, histidine (15 mm), or histidine (15 mm) and glutamine (Gln, 1 mm). Data are means and standard errors of three independent cultures. Two-way ANOVA revealed a highly significant difference among means [F_5,23 = 13.02, P < 0.0001]. Activities identified by different letters in parentheses are significantly different (P < 0.05) by Tukey's HSD test.

Induction of hutU-‘lacZ was also examined following the addition of 1 mm glutamine as the primary nitrogen source to support basal growth. Results (Table 2) showed that supplementation with glutamine did not change the level of histidine-induced hutU expression (PBR813, as a positive control) and hutU transcription was not elevated in ΔntrC ΔcbrB and ΔrpoN (PBR836 and PBR837).

Taken together, both the growth assays and the induction data showed that when histidine is utilized as a sole source of nitrogen the hut operon can be activated by either CbrAB or NtrBC.

Determination of hut transcriptional start sites:

Transcriptional start sites of the hut structural genes (hutF and hutU-G) were determined in minimal medium with histidine as the sole carbon source and in a separate experiment with histidine as the sole nitrogen source. To do this, the 5′ ends of the hutF and hutU-G transcripts were identified using 5′-RACE analysis in SBW25 cells grown in MSM supplemented with histidine and ammonium and MSM supplemented with glucose and histidine. Results are summarized in Figure 4.

Figure 4.—

Structure of the P. fluorescens SBW25 hut locus with details of the promoter regions of hutF and hutU-G operons. (A) The hut genes are organized in three transcriptional units that are represented by arrowed lines. The hutU-G operon is composed of 10 genes [the 5 putative transporter genes are not shown (Zhang and Rainey 2007)]. (B) hutF is transcribed from the same promoter in cells grown on histidine as carbon (MSM plus histidine and ammonium) and histidine as nitrogen (MSM plus glucose and histidine). The hutU-G operon is transcribed by a σ⁵⁴-type promoter when histidine is the sole carbon source (C), whereas hutU-G transcription is initiated under the control of a σ⁷⁰-type promoter when histidine is utilized as the sole nitrogen source (D). The putative HutC binding site (Hu et al. 1989) is highlighted in boldface type. The transcriptional start site as determined by 5′-RACE analysis is indicated by “+1” and arrowed lines.

When SBW25 was grown in MSM supplemented with histidine and ammonium (histidine as the sole carbon source), hutF was transcribed by a σ⁷⁰-type promoter and hutU by a σ⁵⁴-type promoter. Both promoters were previously identified in cells grown on histidine as a sole source of carbon and nitrogen (Zhang and Rainey 2007). This result, combined with the growth data of the ΔcbrA mutant described above consistently show that the same mechanisms of positive hut regulation operate in these two environments.

When SBW25 was grown on histidine as the sole source of nitrogen, the hutF transcriptional start site remained the same, but the hutU transcriptional start was shifted to the 59th nucleotide upstream of the putative translational GTG start. Immediately upstream of this hutU transcriptional start site is a sequence of nucleotides (TTGCAT N18 AAAGAT) that is similar to the σ⁷⁰-promoter consensus (TTGaca N16-18 TAtaaT). Notably, a critical nucleotide is mismatched in the conserved −10 region, suggesting that a positive activator is required for efficient transcription. The putative HutC binding site (Hu et al. 1989) is located between the −35 and −10 regions.

The NAC homolog is not required for hut regulation:

The output domains of both CbrB and NtrC are predicted to be enhancer-binding proteins responsible for σ⁵⁴-recruitment. The observation that histidine utilization as a nitrogen source is regulated by either CbrAB or NtrBC and the fact that the hutU-G operon is transcribed by the σ⁷⁰-type promoter, suggested the existence of an additional transcriptional factor(s), which directly activates the hut operon in a σ⁷⁰-dependent manner with its own transcription being controlled (directly or indirectly) by σ⁵⁴ and NtrC and/or CbrB.

To identify the predicted hut activator, we mutagenized PBR830 (ΔntrC ΔcbrB, His⁻) using the mini-Tn5 derivative ISphoA/hah-Tc and searched for mutants capable of growth on histidine as a sole nitrogen source (His⁺). Promoter(s) internal to the transposon can activate gene expression (Bailey and Manoil 2002) and thus this strategy had the potential to reveal mutants capable of growth on histidine via either suppression or by activation. However, despite screening more than 6 × 10⁵ mutants, no candidate regulator gene was identified (His⁺ mutants were identified, but all mapped to the promoter region of the hutU-G operon).

Previous studies of the enteric bacterium K. pneumoniae showed that NAC is a transcriptional activator which couples σ⁵⁴-dependent regulation of the Ntr system to expression of σ⁷⁰-dependent operons including hut (Bender 1991). A search of the SBW25 genome revealed the presence of a NAC homolog (Pflu4026) that shows 56% amino acid sequence identity with NAC from K. pneumoniae (Swissprot accession no. Q08597). In addition, immediately upstream of the hutU transcriptional start site resides a sequence of nucleotides (ATAtgcTtGTAT) that is similar to the consensus sequence of NAC binding sites (ATA-N6-TNGTAT) in K. aerogenes (Janes et al. 2003). The coding region of pflu4026 was deleted from the SBW25 genome and the mutant strain PBR831 (Δpflu4026) was subjected to growth assays on histidine: PBR831 grew normally on histidine as a sole source of carbon, or nitrogen, or both carbon and nitrogen (data not shown). Moreover, histidine-induced hutU-‘lacZ expressions in both wild-type and Δpflu4026 backgrounds were comparable (fusion strains PBR813 and PBR836, Table 2). These results show that Pflu4026 has no role in histidine utilization.

Phenotypic microarray analysis of cbrB and ntrC:

The commercially available Biolog phenotype microarray plates were used to assess the ability of the wild type, PBR810 (ΔcbrB), PBR829 (ΔntrC), and PBR830 (ΔcbrB ΔntrC) to catabolize 190 carbon substrates (PM1 and PM2) and 95 nitrogen substrates (PM3)—included within this array of compounds were the standard 20 amino acids (l-isomers). The accuracy of the Biolog data was confirmed by checking the histidine utilization phenotypes of each mutant (Table 4): ΔcbrB (and ΔcbrB ΔntrC) were His⁻ when grown on histidine as the sole carbon source; ΔcbrB and ΔntrC were His⁺ whereas ΔcbrB ntrC was His⁻ when grown on histidine as the sole nitrogen source. The results are summarized in Table 3; Table 4 details effects on amino acid utilization. Profiles for each substrate as a sole source of carbon or nitrogen are listed in supplemental Tables 1 and 2 at http://www.genetics/supplemental/, respectively.

TABLE 4.

Utilization of amino acids as the sole source of carbon or nitrogen by P. fluorescens SBW25 and its derived mutants PBR810 (ΔcbrB), PBR829 (ΔntrC), and PBR830 (ΔntrC ΔcbrB)

	Carbon source (PM1 and PM2)a						Nitrogen source (PM3)a
Amino acid (l-isomer)	Genotype				Involvement of		Genotype				Involvement of
		Wild type	ΔcbrB	ΔntrC	ΔcbrB ΔntrC	CbrB	NtrCb	Wild type	ΔcbrB	ΔntrC	ΔcbrB ΔntrC	CbrB	NtrCb
Histidine	+ +	−	+ +	−	Y	N	+ +	+ +	+ +	−	Y	Y
Proline	+ + +	−	+ + +	−	Y	N	+ +	+ +	+ +	−	Y	Y
Leucine	+ +	−	+ +	−	Y	N	+ +	+	+	−	Y	Y
Isoleucine	+ + +	−	+ + +	−	Y	N	+ + +	+ +	+ +	−	Y	Y
Valine	+ +	−	+ +	−	Y	N	+ +	+ +	+	−	Y	Y
Arginine	+ +	−	+ +	−	Y	N	+ +	+ +	+ +	+ +	N	N
Alanine	+ +	+	+ +	+	Y	N	+	+	+	+	N	N
Tyrosine	+ + +c	−	+ + +	−	Y	N	+	+ +	−	−	N	Y
Phenylalanine	+	−	+	−	Y	N	+ +	+	+ +	+	Y	N
Glycine	−	−	−	−	N	N	+ +	+	+ +	+	Y	N
Lysine	−	−	−	−	N	N	+ +	+ +	−	−	N	Y
Glutamic acid	+ +	+	+	+ +	N	N	+ +	+ +	+ +	+ +	N	N
Glutamine	+ +	+ +	+ +	+ +	N	N	+ +	+ +	+ +	+ +	N	N
Serine	+	+ +	+ +	+ +	N	N	+ +	+ +	+ +	+ +	N	N
Aspartic acid	+	+	+	+	N	N	+	+	+	+	N	N
Asparagine	+	+	+ +	+ +	N	N	+	+	+	+	N	N
Tryptophan	−c	−	−	−	N	N	+ +	+ +	+ +	+ +	N	N
Threonine	−	−	−	−	N	N	−	−	−	−	N	N
Methionine	−	−	−	−	N	N	−	−	−	−	N	N
Cysteine	−c	−	−	−	N	N	−	−	−	−	N	N

^aGrowth was determined by absorbance at 660 nm and categorized into four groups: + + + (>0.8), + + (0.4–0.8), + (0.2–0.4), and − (<0.2).

^bInvolvement of CbrB or NtrC in the regulation of a given substrate catabolism is predicted from comparisons of the growth data. Substrates with which minor defects were found in either “ΔcbrB” or “ΔntrC” but no growth defects were found in “ΔcbrB ΔntrC” are considered not to be regulated by CbrB or NtrC. Y, yes; N, no.

^cl-Tyrosine, l-tryptophan and l-cystein are not included in the PM1 and PM2 plates and their utilization was tested separately in M9 salt medium with ammonium chloride as nitrogen source.

TABLE 3.

Role of cbrB and ntrC in utilization of carbon and nitrogen substrates as determined by Biolog phenotype microarray analysis

No. of substrates	Carbon sourcea (PM1 and PM2)	Nitrogen sourcea (PM3)
Total	190	95
Utilized by wild-type	71	72
Growth defects observed with ΔcbrB	32 (12, 20)	8 (8, 0)
Growth defects observed with ΔntrC	3 (3, 0)	37 (9, 28)
Growth defects observed with ΔntrC ΔcbrB	36 (16, 20)	50 (8, 42)

^aFull lists of carbon and nitrogen substrates that can be utilized by SBW25 and the derived mutants are given in supplemental Tables 1 and 2, respectively. Numbers of substrates in which utilization were (reduced, abolished) are shown separately in parentheses.

The cbrB mutant (PBR810) showed growth defects on 20 carbon sources, which include seven amino acids and four carbohydrates (d-xylose, d-mannose, d-ribose and l-arabitol) and was compromised in the utilization of 12 further substrates; it was compromised in its ability to grow on 8 nitrogen sources, but was defective on none. The ntrC mutant (PBR829) was unable to utilize 28 nitrogen substrates (and was compromised in the utilization of 9 further substrates); it was compromised in its ability to grow on 3 carbon sources, but was defective on none. The phenotype of the double mutant (PBR830) was essentially identical to ΔcbrB in terms of defects on carbon-source utilization, but was unable to utilize a further 14 nitrogen sources. Together these data show that NtrC is involved in nitrogen metabolism and that CbrB is primarily a regulator of carbon-source utilization; however, CbrB is also involved in the utilization of nitrogen substrates—and in some cases utilization of substrates as both carbon and nitrogen sources.

With specific reference to amino acid (l-isomers) utilization, seven different patterns of regulation were observed (Table 4). In addition to histidine, utilization of four other amino acids (proline, leucine, isoleucine, and valine) as carbon sources is regulated by CbrB, but requires either CbrB or NtrC when utilized as nitrogen sources. Utilization of arginine and alanine as carbon sources is controlled by CbrB, but neither CbrB nor NtrC is required when they are utilized as nitrogen sources. Lysine can only be catabolized as a nitrogen source, which is regulated by NtrC. Only utilization of tyrosine is controlled in the expected manner, with its use as a carbon source being controlled by CbrB and its use as a nitrogen source by NtrC.

The nac-like gene deletion mutant PBR831 (Δpflu4026) was also subjected to Biolog analysis. Results showed that growth of PBR831 was comparable to the wild-type strain on all 190 carbon and 95 nitrogen substrates, which indicates no functional role for Pflu4026 in cellular carbon and nitrogen utilization.

Physiological role of CbrAB in nitrogen assimilation:

Armed with the knowledge that utilization of five amino acids as nitrogen sources is positively controlled by CbrAB as well as NtrBC, we sought to determine the physiological role of CbrAB in nitrogen assimilation. Results presented in Figure 2 provide a clue: mutant strain PBR809 (ΔcbrA), in which NtrBC is intact, is able to grow in MSM supplemented with succinate (5 mm) and histidine (15 mm), but growth ceases at an early stage. Initial activation must be via NtrC (because in the absence of both cbrB and ntrC there is no growth on histidine) and inactivation presumably reflects dephosphoryation of NtrC. This would be in keeping with the shift in the C:N balance (from carbon rich to nitrogen rich) as succinate is depleted and excess ammonia is produced from histidine degradation and would be consistent with the known effects of nitrogen starvation on the activity state of NtrC (Reitzer and Schneider 2001; Ninfa and Jiang 2005).

To test the hypothesis that the cessation of growth is due to inactivation of NtrBC (a consequence of the buildup of ammonia), the growth kinetics of mutant PBR810 (ΔcbrB) were examined in three media: MSM plus succinate (5 mm) and supplemented with (1) histidine, (2) urocanate, or (3) proline [proline is subject to the same dual CbrAB/NtrBC regulation as histidine (Table 4)]. Degradation of histidine, urocanate, and proline to glutamate produces 2, 1, and 0 molecules of ammonia, respectively, therefore if buildup of nitrogen is the cause of the cessation of growth we predict that cell yield will be lowest when growing on histidine, intermediate on urocanate, and highest on proline. The results shown in Figure 5 are consistent with these predictions and interestingly growth on proline did not cease even though it is likely that the proline was entirely consumed during the course of growth. Moreover, mutant PBR829 (ΔntrC), in which CbrAB is intact, grew normally (not different to the wild type) in these three media (data not shown). Together the data indicate a role of CbrAB in maintaining the activities of the catabolic pathways under different C:N ratios.

Figure 5.—

Growth dynamics of mutant PBR810 (ΔcbrB) on histidine, urocanate, and proline as nitrogen sources. Bacteria were grown in MSM supplemented with succinate (5 mm, as primary carbon) and 15 mm of histidine, urocanate, and proline. Conversion of histidine, urocanate, and proline to glutamate (per molecule) produces 2, 1, and 0 ammonia, respectively (shown in parentheses). Results are means of nine independent cultures collected at 5-min intervals, but two hourly time points are shown for clarity. Standard errors are within the symbols and thus not visible.

DISCUSSION

The genome of Pseudomonas contains a large number of two-component regulatory systems (>60 sensor-regulator pairs per genome; Kiil et al. 2005), which play a major role in detecting niche-specific signals, including nutrient availability (Monds et al. 2006; Sonawane et al. 2006). Previous studies on histidine utilization (as the sole source of carbon and nitrogen) showed that CbrAB was required for hut activation in P. aeruginosa (Nishijyo et al. 2001) and in P. fluorescens (Zhang and Rainey 2007); more recently evidence was obtained from a suppressor analysis of a cbrAB mutant for the involvement of a second two-component system, NtrBC (Itoh et al. 2007; Li and Lu 2007). These studies did not define the hut activator(s) when histidine is copresent with other carbon or nitrogen sources (as often happens in natural environments) or the extent of any interplay between the regulators (CbrAB and NtrBC). Work described here confirmed the role of NtrBC in activation of the hut operon under conditions of nitrogen starvation. But surprisingly, we found that CbrAB was sufficient to maintain activity of hut across a range of C:N ratios: from carbon rich (where histidine is the sole nitrogen source) to nitrogen rich (where histidine is the sole carbon source). Given phenotypically identical roles in the utilization of four other amino acids (proline, leucine, isoleucine, and valine), i.e., requirement of CbrAB when the amino acid is a carbon source, but either CbrAB or NtrBC when a nitrogen source, it seems reasonable to assume that insights into CbrAB function on the basis of hut are likely to hold for a variety of other carbon and nitrogen sources, thus implicating CbrAB as an important global regulator of nutrient utilization in Pseudomonas.

The shift from a σ⁵⁴-type promoter (when histidine is a sole carbon source) to a σ⁷⁰-type promoter (when histidine is a sole nitrogen source) during hutU transcription shows the complexity of hut activation by CbrAB. While DNA binding studies are required to demonstrate the direct connection between CbrB and hut transcription, the evidence gathered (experimental and in silico) implicates CbrB as the σ⁵⁴-enhancer binding protein for transcriptional activation of the hutU operon when histidine is utilized as the sole carbon source. However, the CbrB- (and NtrC-) dependent regulator(s) that directly activates the hut operon in nitrogen-starved environments remains unclear. The identification of such a regulator(s), which is expected to be functionally equivalent to the NAC protein in enteric bacteria, is critically important for further understanding of how CbrAB regulates genes for nitrogen assimilation. Data presented here have ruled out the involvement of the NAC homolog (Pflu4026).

In E. coli, utilization of carbon is determined by the PTS system via effects on the intracellular concentration of cAMP; assimilation of nitrogen is controlled by glnALG, which encodes glutamine synthase (GS) and the Ntr regulon regulators, NtrB and NtrC. Studies over many decades have led to the widely held view that these two systems independently regulate carbon and nitrogen assimilation; however, recent work shows interplay between the regulators (Tian et al. 2001; Commichau et al. 2006; Mao et al. 2007). That there is interplay makes sense, particularly for bacteria able to utilize substrates, such as amino acids that provide both carbon and nitrogen.

Our work shows that in P. fluorescens SBW25 the relationship between the two global hut regulators has evolved beyond a simple interplay where each regulator takes its cue from different aspects of the intracellular carbon and nitrogen pool. CbrAB is capable of maintaining hut activity across a wide range of C:N ratios. The role of NtrBC thus appears to be subsumed by CbrAB. The overriding control of histidine utilization by CbrAB may reflect an adaptive response by Pseudomonas to the environments in which it is commonly found. Pseudomonas is an exceptional opportunist (Stanier et al. 1966): it thrives in highly competitive, nutritionally complex, carbon-limited environments, where resources vary not only in composition, but also in availability across space and time (Zhang et al. 2006). Under such conditions a strategy that extracts maximum benefit from the widest possible range of resources—as rapidly as possible—is likely to confer an adaptive advantage.

In terms of extracting maximum benefit, the five-step histidine degradation pathway offers a significant advantage over the four-step pathway employed by enteric bacteria. The presence of FIGLU iminohydrolase (encoded by hutF in the five-step pathway) means that FIGLU is degraded to glutamate, formate, and ammonium (Hu et al. 1987; Zhang and Rainey 2007); in the absence of HutF, FIGLU is degraded by FIGLU formiminohydrolase to yield glutamate and formamide (Magasanik 1978). Formate can be utilized as a source of single carbon compounds in Pseudomonas (Revel and Magasanik 1958), but formamide cannot (it leaks from the cell in enteric bacteria, Magasanik and Bowser 1955): the five-step pathway thus offers a more efficient means of extracting carbon from histidine. But this gain in carbon comes at the cost of additional ammonium (twice as much ammonium is produced when histidine is degraded by the five-step pathway) and this is problematic on two accounts: first, intracellular ammonium is toxic at high levels; but second, ammonium sends a “replete-with-nitrogen” signal to the Ntr system, which leads to NtrC inactivation (Weiss et al. 2002; Ninfa and Jiang 2005). Were NtrBC solely responsible for hut activation when histidine is a nitrogen source then transcription of the hut locus would cease soon after degradation of histidine began, due to the buildup of ammonium. Precisely this situation was observed when cbrAB mutants were grown on histidine with limiting amounts of succinate (Figures 2 and 5): NtrBC was unable to maintain activity of the hut operon in the face of a shift in the balance of the ratio of carbon to nitrogen. As a consequence, growth ceased despite there being carbon and nitrogen (in the form of histidine) available to support growth. The involvement of cbrAB in regulation of histidine utilization, irrespective of whether the cell experiences carbon or nitrogen starvation solves the problem of NtrBC inactivation of hut in response to elevated ammonium.

While recruitment of CbrAB to regulate hut in response to nitrogen starvation solves one problem, it potentially exacerbates another. In those situations where the sole purpose of histidine degradation is the acquisition of carbon (and where there is an abundance of histidine), the cell runs the risk of poisoning itself with ammonium. In an earlier study of the hut operon (Zhang and Rainey 2007) we argued that data on the fitness and transcriptional consequences of deleting hutD indicated that HutD acts as a governor, preventing the level of transcriptional activation of hut from exceeding a critical upper limit. Although we were unaware of the dual control of hut by CbrAB and NtrBC at the time of our earlier investigations, the fact that CbrAB maintains transcription of hut even when nitrogen is abundant means that a mechanism to prevent ammonium poisoning is physiologically relevant.

The fact that CbrAB is sufficient to ensure activity of hut across a range of C:N ratios begs the question as to the nature of the signal perceived by the CbrA sensor. While it is possible that a single sensor might detect both the intracellular carbon and nitrogen status it is possible that the signal activating CbrAB lies elsewhere, for example, in the exogenous nutrient itself. A competitive opportunist must be able to respond rapidly to available nutrients, thus ensuring the maximum possible growth rate while also ensuring that competitors are deprived of resources. By responding directly to the presence of an exogenous nutrient—rather than via signals derived from the intracellular carbon and nitrogen pool—a cell is assured of a rapid response to the availability of nutrients. We therefore cautiously suggest that CbrA might sense histidine and possibly other nutrients and that via this ability Pseudomonas acquires and utilizes exogenously available substrates free from regulatory constraints imposed by a system that takes its primary cue from intracellular signals. Such regulatory control would be consistent with a gluttonous approach to nutrient acquisition: the primary goal being to bring into the cell as much of any available resource as quickly as possible.