Transcriptional Expression of Escherichia coli Glutamate-Dependent Acid Resistance Genes gadA and gadBC in an hns rpoS Mutant

Abstract

Resistance to being killed by acidic environments with pH values lower than 3 is an important feature of both pathogenic and nonpathogenic Escherichia coli. The most potent E. coli acid resistance system utilizes two isoforms of glutamate decarboxylase encoded by gadA and gadB and a putative glutamate:γ-aminobutyric acid antiporter encoded by gadC. The gad system is controlled by two repressors (H-NS and CRP), one activator (GadX), one repressor-activator (GadW), and two sigma factors (σ^S and σ⁷⁰). In contrast to results of previous reports, we demonstrate that gad transcription can be detected in an hns rpoS mutant strain of E. coli K-12, indicating that gad promoters can be initiated by σ⁷⁰ in the absence of H-NS.

Commensal and pathogenic Escherichia coli must survive transit through the acidic conditions of the stomach, where the pH is normally between 2 and 3, before they can colonize a mammalian host.

The most effective E. coli acid resistance system is dependent upon glutamate and involves two isoforms of glutamate decarboxylase (GAD) encoded by gadA and gadB that convert intracellular glutamate to γ-aminobutyrate, consuming one intracellular proton in the reaction. gadB is transcribed in an operon with gadC, which encodes an antiporter that is proposed to import glutamate inside the cell while simultaneously exporting γ-aminobutyrate (9, 15, 20).

The regulation of the gad system is extremely complex. The expression of all of the gad genes is regulated by separate σ^S-dependent and -independent pathways, despite only one transcriptional start site being identified for each locus (4, 5, 6). The alternate sigma factor, σ^S, encoded by rpoS, is associated with the stationary-phase expression of gad in cells grown in rich medium, whereas σ^S-independent regulation occurs in cells grown in minimal-glucose medium.

The levels of expression of the two GAD isoforms and gadC are greatly enhanced in an hns deletion background compared to levels found in wild-type cells of E. coli (6, 17, 24, 25). The histone-like protein H-NS is a major component of the bacterial nucleoid which influences a variety of cellular processes, such as transcription, recombination, and replication (19, 26). The mechanism underlying gene regulation by H-NS is due to either transcriptional silencing through preferential binding to AT-rich curved DNA sequences often found upstream of E. coli promoters or to changes in DNA supercoiling (23, 24). H-NS silencing of gene expression is relieved by environmental signals, such as changes in osmolarity, growth phase, low temperature, anaerobiosis, and pH (2). H-NS has also been implicated in the posttranscriptional regulation of σ^S expression, as σ^S accumulation is observed in exponential phase in an hns mutant (3, 23). This suggested that H-NS has an indirect effect on gad expression. However, it has now been shown that H-NS represses gad expression directly (11).

It has been established that the expression of the gad system is mediated by the GadX protein, a member of the AraC/XylS family of transcriptional regulators, encoded by gadX (located downstream of gadA) (10, 13, 17). Expression of gadX is primarily driven by σ^S (11, 17). GadX was originally shown to play a central role in the H-NS control of genes required in glutamate-dependent acid resistance and to bind to the gadA and gadBC promoters (13, 17). However, repression of the gad genes by H-NS has now been demonstrated to be independent of GadX (11). GadW, another AraC-like regulator, mediates control of the gadA and gadBC genes by repressing gadX and can also activate the gad genes directly in rich medium at pH 8 in stationary phase (11). The cyclic AMP receptor protein (CRP) represses the system by inhibiting the production of σ^S. This complex regulatory network maintains tight control over expression of the gad system but also provides flexibility for inducing acid resistance under a variety of conditions that precede exposure to acid.

Transcription of gadA and gadBC in an hns rpoS background.

De Biase et al. (6) have reported that gad transcription is abrogated in an hns rpoS double mutant. They suggest, however, that gad promoters can be recognized by H-NS and transcribed by σ⁷⁰, as GAD expression driven by multicopy plasmids was not completely shut off in an rpoS mutant. Consistent with this observation, the gad genes are expressed in Salmonella enterica serovar Typhimurium LT2 (6), which carries a weak rpoS allele with a rare UUG start codon (21). More recently it has been implied that gad expression can be driven by σ⁷⁰ in cells grown to mid-exponential phase in minimal medium (4).

Because of this inconsistency we investigated the expression of the gad genes in isogenic hns (22), rpoS (15), and hns rpoS (23) mutants of E. coli K-12 MC4100 (14) by performing Northern hybridization with the internal fragments of both gadB and gadC (Fig. 1). Due to the high DNA sequence identity (98%) between gadB and its isoform gadA (16), the gadB probe was used to detect the transcripts of both genes. MC4100 is a commonly studied laboratory strain and can be observed to be still undergoing rapid division at an A₆₀₀ of 1.0 (see Fig. 2).

FIG. 1.

Northern analysis of gadB and gadC transcription in E. coli MC4100 and isogenic hns, rpoS, and hns rpoS mutants. Total RNA was extracted from cultures of each strain at the growth curve points of A₆₀₀ of 1.0, 1.5, and 1.8. The membrane was probed with the ³²P-labeled internal fragments of gadB (A) and gadC (B).

FIG. 2.

Effect of hns mutation on the expression of acid resistance in E. coli MC4100. Cultures were diluted 1:1,000 and were grown in LB medium. Results were taken from a single experiment performed in triplicate. The percentage of acid resistance was determined along the growth curve for strains MC4100 (circles), MC4100 hns (squares), MC4100 rpoS (diamonds), and MC4100 hns rpoS (triangles). Specific acid resistance and optical densities (OD) at 600 nm are indicated by open and closed symbols, respectively.

Total RNA was prepared from cultures grown in Luria-Bertani (LB) medium at various stages of the growth curve (A₆₀₀ of 1.0, 1.5, and 1.8) by using TRIzol reagent as described by the manufacturers (Gibco) and by 1 h of DNase treatment (Boehringer Mannheim). RNA samples (30 μg) were heated for 5 min at 65°C, fractionated on 1.2% formaldehyde-agarose gels, and blotted onto nylon membranes (Amersham). Membranes were hybridized with the internal fragments of gadB and gadC and were nick translated with [α-³²P]deoxynucleoside triphosphates (Amersham).

Primers used for generating the 1.0-kb internal fragment of gadB were 5′ TCC GCT GCA CGA AAT GCG CGA CGA TGT CGC A 3′ and 5′ AAC CTG GTA AGA GGC GTT CTG TAC TTT GGT 3′. Primers used for generating the 1.2-kb internal fragment of gadC were 5′ TCT GGG TCC GAG ATG GGG ATT TGC AGC GAT 3′ and 5′ TGG TGA ACG TCG ACG CGG GTG CAG GAA GAA 3′. PCR amplification was carried out as described previously (20).

Transcripts of all gad genes were absent in the rpoS mutant as well as in the exponentially grown parental strain, whereas transcripts of gad genes were detected in stationary-phase cultures (A₆₀₀ = 1.8) of the parental strain (Fig. 1). The sizes of the mRNA obtained corresponded to those predicted from the open reading frames of both gadA/B and gadC (data not shown). We also observed a higher-molecular-weight transcript that corresponded with the deduced length of a polycistronic gadBC message, but this transcript was weaker than either the gadA/B or gadC transcript (data not shown). As previously reported by De Biase et al. (6), transcription of the gadA and gadBC genes was strongly enhanced in the hns mutant and occurred during exponential growth.

In contrast to De Biase et al. (6), however, we were able to detect transcripts of both gadA and gadBC in an hns rpoS mutant. The presence of transcripts in an hns rpoS double mutant suggests that the promoters of all gad genes can be recognized by σ⁷⁰ in vivo. This has been demonstrated for other σ^S-dependent genes (1, 12, 23). The stronger signal present at an A₆₀₀ of 1.0 in the hns mutant compared to that of the hns rpoS strain may be explained by the accumulation of σ^S in mid-exponential phase in this strain (3, 23).

The hns rpoS mutant used in this study was provided by T. Mizuno (Nagoya University), and we have confirmed that it is defective at the rpoS locus as acid resistance can be restored by complementation with the wild-type rpoS gene (data not shown). The detection of gad transcripts in this strain is compatible with the observation of a previous study which showed the expression of transcripts and proteins from other σ^S-dependent genes (csgBA and hdeAB) in similar E. coli K-12 hns rpoS mutant strains (1).

Possible explanations for the differences between the observations of De Biase et al. (6) and those reported here are the different growth conditions and the strains used in the two studies. De Biase et al. (6) examined a different E. coli K-12 hns rpoS mutant, YK4122, and grew it to early exponential phase for 2 h (A₆₀₀ = ∼0.4) and to stationary phase for 5 h (A₆₀₀ = ∼2.1). For stationary-phase induction we grew our cultures overnight for 16 h (A₆₀₀ = 1.8). It is possible that their incubation times were too short to detect significant levels of gad transcription in this strain. Under these same growth conditions they were only able to detect extremely weak signals of gad transcription in the MC4100 K-12 strain.

Expression of acid resistance in an hns mutant of E. coli.

To determine whether the increased transcriptional expression of the gad genes in hns and hns rpoS mutants correlated with relief of growth phase dependence of acid resistance, the expression of acid resistance was assayed along the growth curve for the isogenic hns, rpoS, and hns rpoS mutants and the wild-type parent of E. coli K-12 grown at 37°C in LB medium (Fig. 2). LB medium for extreme acid exposure (pH 2.5) was adjusted for pH by using HCl. Cultures were adjusted to the appropriate concentrations (1 × 10⁵ to 5 × 10⁵ CFU/ml) by serial dilutions in phosphate-buffered saline, and an aliquot was added to the acidified LB medium and was incubated for 2 h at 37°C with aeration. Control dilutions from the original culture were plated on LB agar and were grown overnight at 37°C. Colony counts were then compared to those obtained from the same culture exposed to the acidified medium to determine the percentage of survival. Acid resistance was defined as the percentage of the number of bacteria surviving the acid treatment compared to the initial number of inoculum exposed. Values shown for percentage of survival represent the mean of a single experiment performed in triplicate.

The hns mutant exhibited induction of acid resistance in mid-exponential phase (A₆₀₀ = 1.2) some 3 h earlier than the parent strain. This acid resistance was not constitutive and implies that σ^S is still required for the induction of expression. This finding is supported by the observation that CRP-dependent repression of gadA and gadBC is not relieved in an hns mutant (11), indicating that CRP is a master regulator of glutamate-dependent acid resistance. The hns rpoS double mutant did not exhibit the acid resistance phenotype, reflecting previous observations by De Biase et al. (6). This was despite the fact that we have shown that this strain produces transcripts of gadA and gadBC as well as other acid resistance genes (hdeAB) (20), which are also expressed much earlier in the growth phase by this strain (24). The hns rpoS mutant still expressed GAD enzyme activity when cells were permeabilized with Triton X-100 (data not shown).

A repressing action of H-NS on gadA and gadBC promoters, preventing a successful transcription-inducing complex with σ⁷⁰ during the exponential growth, can be predicted. In stationary phase the σ^S levels increase, and this can overcome H-NS repression, probably by more efficient recognition of curved promoter regions required for H-NS binding. The region upstream of many σ^S-dependent promoters is often AT rich and shows intrinsic DNA curvature (8). Thus, σ^S may relieve H-NS repression at these promoters. A −10 consensus sequence (CTATACT) was determined for σ^S-dependent promoters based on the comparison of characteristic promoters known to be under the control of σ^S and those that can be recognized in vitro by both σ^S and σ⁷⁰ (7). Sequences similar to this consensus were identified by primer extension mapping of the transcriptional start points of gadA and gadBC (4, 6).

σ^S appears to be the major sigma factor used for directing gad transcription for cells grown in complex, but not minimal, medium. Minimal-medium cultures, however, must be able to utilize a different sigma factor under acidic conditions (4). Since only one promoter for each gene appears to be involved regardless of the inducing conditions, the other sigma factor is suggested to be σ⁷⁰ (4). In accordance with this, a crp gadX gadW mutant which produced no GAD when grown to exponential phase in LB medium made copious amounts of GAD once an hns mutation was introduced (11).

Together these results suggest that additional genes that are likely to be dependent upon σ^S for their expression contribute to glutamate-dependent acid resistance, a conclusion endorsed in previous reports (5, 6, 18). Some of these reports have also indicated that the gad decarboxylase/antiporter system would create a futile proton cycle, since transport of glutamic acid brings into the cell a proton that is consumed by its decarboxylase (5, 18). To compensate, a mechanism for generating endogenous glutamate would be required. Glutaminases have been previously hypothesized to play a role in glutamate-dependent acid resistance (5). Identification of additional genes and further investigation of the interactions between regulatory components of the gad system will be important for understanding the complex mechanism of glutamate-dependent acid resistance.