Abstract
Mg2+ homeostasis is important for Salmonella pathogenesis. In Salmonella enterica, the transcription of the mgtA gene, which encodes a Mg2+ transporter, is regulated by a Mg2+-sensing riboswitch [Cromie MJ, Shi Y, Latifi T, Groisman EA (2006) Cell 125:71–84]. In a genetic analysis of the determinants of thermotolerance in S. enterica serovar Typhimurium, we isolated the chr-1 mutation that increased the resistance of exponential phase cells to killing by high temperature. This mutation is a single base change in the mgtA riboswitch that causes high-level constitutive expression of mgtA. We showed that another mgtA riboswitch mutation, ΔUTRre-100, which had been constructed by Cromie et al., also confers similar increased thermotolerance. Surprisingly, the chr-1 mutation is located at a position that would not be predicted to be important for the regulatory function of the riboswitch. We obtained physiological evidence suggesting that the chr-1 mutation increases the cytosolic free Mg2+ concentration. High-level expression of the heterologous MgtE Mg2+ transport protein of Bacillus subtilis also enhanced the thermotolerance of S. enterica. We hypothesize that increased Mg2+ accumulation might enhance thermotolerance by protecting the integrity of proteins or membranes, by mitigating oxidative damage or acting as an inducer of thermoprotective functions.
Keywords: heat tolerance, riboswitch, Mg2+ homeostasis, heat shock
As a pathogen that can grow in a free living state in various environments and in hosts ranging from ectothermic reptiles to warm-blooded mammals and birds whose temperature may be elevated by fever (1), Salmonella possesses sophisticated mechanisms for thermal adaptation. Salmonella is the most frequent causative agent of foodborne bacterial infections in the United States (2), and because heat treatment is the most common and economical means for inactivating food pathogens better understanding of the regulation of thermotolerance in this organism has important applications in food safety.
The best-studied mechanism of thermal adaptation is the heat shock response. This response involves the transient induction of thermoprotective proteins by exposure to sublethal high temperatures, which enables organisms to survive subsequent treatments with higher temperatures that would be otherwise lethal (3). Thermotolerance can be regulated by a number of other factors, including stationary phase (4) and acidic or alkaline pH (5, 6).
Another environmental condition that regulates thermoresistance is water activity (aw). Exposure to elevated osmolality enhances the thermotolerance of bacteria in two ways: by elevating their upper limit of growth temperature (7, 8) and increasing their viability at lethal high temperatures (9, 10). Our laboratory observed that the thermoprotective effect of high osmolality on the survival of Salmonella enterica serovar Typhimurium is reversed by glycine betaine (11). Glycine betaine is an osmoprotectant, which can alleviate the growth inhibitory effects of high osmolality (12). Along with ameliorating the osmotic inhibition of growth, this compound suppresses a number of other responses brought about by osmotic stress, such as the accumulation of K+, the synthesis of trehalose, and the shrinkage of the cytoplasmic volume (13).
To identify components of the apparatus mediating the osmotic control of thermotolerance, we wanted to isolate mutants of S. enterica in which the high osmolality-dependent stimulation of thermotolerance was no longer antagonized by glycine betaine. In this selection, we obtained chr (constitutively heat resistant) mutations that conferred increased thermotolerance not only at high osmolality in the presence of glycine betaine but also at low osmolality. Surprisingly, one of these mutations, chr-1, resulted in high-level constitutive expression of the mgtA (Mg2+ transport) gene. Thus, these results revealed that Mg2+ has a hitherto unknown function in the regulation of thermotolerance.
Mg2+, the second-most abundant cation in bacteria, has multiple functions: it is a cofactor in ATP-dependent phosphorylations and numerous other enzymatic reactions, it stabilizes ribosomes, it influences nucleic acid-protein interactions and RNA folding, and it is required for membrane integrity (14). Typical of the exquisite homeostatic control observed for the intracellular concentrations of most cations (15), the Mg2+ content of the cells is maintained at an essentially invariant level by tight regulation of uptake, efflux, and binding to macromolecules. As a result, there is <5-fold change in the intracellular Mg2+ content in response to five orders of magnitude variation in the external Mg2+ concentration (16, 17).
In S. enterica, Mg2+ is taken up by three transporters: MgtA and MgtB, which are P-type ATPases, and CorA, which is a channel (18). The mgtA and mgtB genes can be induced >2,000-fold by limitation for Mg2+ (19). The transcriptional activation of these two genes depends on the PhoPQ two-component system, which influences the expression of ≈5% of the genes of S. enterica (20). PhoQ is a membrane-bound signal sensor kinase, which phosphorylates PhoP protein at low external concentrations of Mg2+, Ca2+, and Mn2+ (21–23). The phosphorylated PhoP is the transcriptional activator of genes in the PhoPQ regulon. At high concentrations of Mg2+, Ca2+, or Mn2+, PhoQ promotes the dephosphorylation of PhoP and down-regulates the transcription of genes in the PhoPQ regulon. PhoQ responds to other regulatory signals in addition to divalent cations, because the repression caused by high concentrations of Mg2+, Ca2+, or Mn2+ is strongly antagonized by low concentrations of certain antimicrobial peptides (21, 24). Acidic pH can also overcome the repression by divalent cations, although there is a discrepancy in the literature concerning whether pH signal is sensed by PhoQ (24) or by a PhoPQ-independent mechanism (22). Mg2+ homeostasis is important for Salmonella pathogenesis, as suggested by the fact that the PhoPQ (25, 26) and the CorA proteins (16) are essential virulence determinants and the mgtA and mgtB genes are induced during infection in mice (27) and in cultured epithelial cells (28). Because macrophages have an acidic cytoplasm that is believed to be low in Mg2+, it has been proposed that S. enterica uses the PhoPQ system to detect that it is in an intracellular environment and needs to induce virulence genes (24, 29). Recently, it has been shown that high-level expression of the Rob protein, which is a transcriptional activator of genes involved in resistance to antibiotics, heavy metals, superoxide, and organic solvents, results in the induction of the mgtA gene from a PhoP-independent promoter that is 43 nt downstream of the major PhoP-dependent promoter. However, the role of this protein in the regulation of expression of mgtA is unclear, because deletion of rob does not affect the expression of mgtA at low or high Mg2+ concentrations (30).
Superimposed on the regulation of transcription initiation by the PhoPQ system, which is responsive to the external Mg2+ concentration, the mgtA gene is subject to an additional control mechanism that is mediated by an Mg2+-sensing riboswitch. This regulatory element, which is specified by a 264-nt UTR at the 5′ end of the mgtA mRNA, has been proposed to adapt two different stem and loop conformations at high vs. low intracellular Mg2+, where the structure formed at high Mg2+ acts as a transcription terminator at a site between the promoter and the mgtA coding sequences (31). Transcription from the mgtA promoter requires the Mg2+-regulated PhoPQ system for initiation (23, 27), but the riboswitch amplifies the response to Mg2+ by directing the RNA polymerase to stop or to proceed at the terminator at adequate intracellular or limiting Mg2+ concentrations, respectively. Mutations that destabilize the terminator structure reduce the efficiency of transcription termination at high Mg2+ concentrations, but they do not eliminate the need for an upstream promoter for transcription initiation (31).
In this work, we show riboswitch mutations resulting in increased expression of the mgtA gene confer increased heat tolerance in S. enterica.
Results
Isolation of the Mutations That Confer Increased Heat Resistance.
The basis for the isolation of the heat tolerant mutants was the observation that glycine betaine impairs the high osmolality-dependent stimulation of thermotolerance. This result is reproduced in Fig. 1A. Osmotic stress imposed by 0.3 M NaCl caused a dramatic increase in survival of the WT strain at 53 °C, which was antagonized by glycine betaine. To isolate the mutants with enhanced heat tolerance, the WT was mutagenized with ethylmethane sulfonate, and derivatives were selected that acquired increased resistance to 53 °C in minimal medium containing 0.3 M NaCl + glycine betaine. This procedure yielded two independent mutants, TL3380 (chr-1) and TL3360 (chr-2), that exhibited similar enhanced thermotolerance not only at high osmolality in the presence of glycine betaine, which was the selected phenotype, but also at low osmolality, which was demonstrated on subsequent characterization (Fig. S1). The two mutations do not have a detectable effect on growth rate at any temperature from 30–45 °C in the presence or absence of 0.3 M NaCl or glycine betaine. So far, we characterized only the chr-1 mutation, and we do not have additional information about chr-2 other than that it is not linked to any of the Mg2+ transport genes.
Fig. 1.
The thermotolerance of the strains TL1 (wild type; A) and TL3290 (chr-1; B) was determined at 53 °C in exponentially growing cells in M63 + 10 mM glucose containing the indicated supplements, as described in Materials and Methods. GB, 1 mM glycine betaine.
Because it is possible that the initial heat-resistant isolates may contain multiple mutations, for further characterization, the chr-1 mutation was transduced via linkage with the STM4446::Tn10dCm insertion into the WT strain. Fig. 1B documents that in this new background the chr-1 mutation resulted in a dramatic increase in thermotolerance in exponential phase cells both in medium of low osmolality and medium containing 0.3 M NaCl + glycine betaine, and it also conferred slightly increased heat tolerance in medium augmented with 0.3 M NaCl without glycine betaine.
The chr-1 Mutation Is in the Regulatory Riboswitch RNA for the mgtA.
The steps for the identification and the confirmation of the sequence of the chr-1 mutation are described in detail in SI Materials and Methods. Briefly, this procedure involved the isolation of the linked STM4446::Tn10dCm insertion, three-factor transductional mapping against flanking treC1::KmR and yjgF2::Tn10 insertions, and amplification and sequencing of a 4881-bp fragment containing the treR (trehalose repressor), mgtA, and STM4457 (predicted transposase) genes. The sequence analysis revealed that chr-1 mutation is a C to T change at position 4,699,521 of S. enterica serovar Typhimurium LT2 genome. The affected nucleotide is in the riboswitch of the mgtA gene at position +98 of the mRNA (Fig. 2).
Fig. 2.
The chr-1 mutation caused a C to U change at position +98 of the mgtA riboswitch. The stem and loop structures formed at low and high Mg2+ were proposed by Cromie et al. (31). The beginning and endpoints of the interval from positions 148 to 247 that was replaced by an unrelated sequence of 84 nt in the ΔUTRre-100 mutation (31) are indicated by the arrows. [Reproduced with permission from ref. 31 (Copyright 2006, Elsevier).]
The ΔUTRre-100 mgtA Riboswitch Mutation also Results in Increased Thermotolerance.
Because the chr-1 mutation was obtained by a positive selection for increased thermotolerance, it might be exceptional in this regard. For the characterization of the mgtA riboswitch, Cromie et al. (31) constructed the site-directed mutation ΔUTRre-100, which involved the replacement of nucleotides 148-247 of this regulatory element with an unrelated 84-bp “scar.” We transduced this mutation into our WT strain TL1 (Table S3) and determined its effects on thermotolerance. As shown in Fig. 3, the ΔUTRre-100 mutation was similar to chr-1 in conferring increased thermotolerance in M63 and in M63 + 0.3 M NaCl + glycine betaine. This result demonstrates that the chr-1 allele is not a special or fortuitous example of a riboswitch mutation that can confer increased thermotolerance. It is noteworthy that the two mutations were obtained by different approaches: chr-1 was isolated in our laboratory by positive selection, whereas ΔUTRre-100 was constructed by Cromie et al. (31) by reverse genetics, and has not been suspected previously to affect thermotolerance.
Fig. 3.
The effect of the ΔUTRre-100 mutation on thermotolerance. This experiment with strain TL4391 was carried out at the same time and under the same conditions as those shown in Fig. 1. GB is 1 mM glycine betaine.
The chr-1 Mutation Renders mgtA Expression Constitutive.
To test whether the chr-1 mutation altered the transcriptional control of mgtA, we introduced it upstream of a mgtA-lacZ fusion and determined its effect on the β-galactosidase activities of cells grown with various concentrations of Mg2+ (Table 1). As controls, we used isogenic strains that carried the ΔUTRre-100 mutation and the chr+ allele in combination with the mgtA-lacZ fusion. In accord with previous observations (19, 27, 31), in the chr+ background, the mgtA-lacZ fusion was expressed maximally at 0.016 mM Mg2+ (the lowest concentration tested) and was progressively down-regulated as the concentration of this cation was raised. The chr-1 mutation resulted in 8- to 45-fold increase in the expression of mgtA-lacZ in the presence of 0.016-16 mM Mg2+ compared with that seen in the chr+ background. In the ΔUTRre-100 mutant, mgtA was expressed at 30-35% higher level than in the chr-1 mutant at all Mg2+ concentrations, but the two mutations were similar in that they both rendered the expression of the gene essentially insensitive to the external Mg2+ concentration. These results suggest that the enhanced heat tolerance of the chr-1 mutant is the consequence of increased expression of the mgtA+ gene. We can rule out the possibility that the riboswitch mutations result in enhanced thermotolerance by elevating the transcription of a gene that is downstream from mgtA, because the next downstream gene, STM4457, is transcribed in the opposite direction than mgtA.
Table 1.
The effect of riboswitch mutations on the expression of mgtA-lacZ
| MgSO4, mM | β-galactosidase specific activity ± SD, Miller units |
||
|---|---|---|---|
| chr+mgtA-lacZ | chr-1 mgtA-lacZ | ΔUTRre-100 mgtA-lacZ | |
| 0.016 | 83 ± 4 | 628 ± 62 | 1,010 ± 126 |
| 0.16 | 72 ± 1 | 646 ± 18 | 1,000 ± 18 |
| 1.6 | 28 ± 3 | 727 ± 25 | 940 ± 33 |
| 16 | 12 ± 1 | 538 ± 8 | 798 ± 102 |
Strains TL4295 (chr+ mgtA-lacZ), TL4293 (chr-1 mgtA-lacZ), and TL4385 (ΔUTRre-100 mgtA-lacZ) were grown in Mg2+-free M63 containing the indicated concentrations of MgSO4, and the β-galactosidase activities were determined as described in Materials and Methods.
The result that the expression of the mgtA-lacZ fusion is essentially insensitive to the Mg2+ concentration in the riboswitch mutants is unexpected, because the initiation of transcription at the mgtA promoter would be predicted to be subject to residual control by this cation via the PhoPQ system. Different members of the PhoPQ regulon display a wide range of regulation, from 4- to >250-fold induction by Mg2+ limitation (23). It is possible that the PhoP-dependent transcription initiation at the mgtA promoter might not be very sensitive to the Mg2+ concentration, and most of the transcriptional control of the mgtA gene is provided by the riboswitch.
Effect of the chr-1 Mutation on Mg2+ Homeostasis.
To determine whether the accumulation of Mg2+ could be elevated by increasing the expression of mgtA, we determined the effect of the ΔUTRre-100 mutation on the total cellular concentration of this cation (Table S1). The increased expression of mgtA did not result in a statistically significant increase in the total Mg2+ content of the cells compared with that measured in the WT. However, because Mg2+ inside the cells is largely bound to macromolecules whose levels are determined by growth rate, it may not be unexpected that higher expression of mgtA did not result in a detectable increase in the total Mg2+ content. There were also no statistically significant changes in total cellular content of K, Cu, Zn, Fe, Mn, Ni, Mo, and Co in the riboswitch mutant (Table S1). These results indicate that elevated expression of the mgtA+ gene did not affect the levels of any of the biologically relevant cations, including Mn2+, which has been suggested to be beneficial for oxidative stress tolerance (32).
Although the total Mg2+ content of the cells is largely invariant in response to the external concentration of this cation (16, 17), it has been hypothesized that even submillimolar changes in the Mg2+ concentration of the medium perturb the free cytoplasmic concentration of this cation sufficiently to be detected by the mgtA riboswitch (31). We carried out an experiment in which we used the riboswitch to test whether elevated expression of the mgtA gene increased the free cytoplasmic concentration of Mg2+. A plasmid carrying the lacZ gene fused to the WT (chr+) mgtA promoter was introduced into two strains, one of which was chr-1 mgtA+ on the chromosome and the other was chr+ mgtA+. The chr-1 mutant had a 2.3-fold lower β-galactosidase activity than the chr+ counterpart (Table 2). This result is consistent with the conclusion that the higher level of expression of the mgtA+ gene in the chr-1 mutant resulted in increased accumulation of free Mg2+, which down-regulated the expression of the plasmid-borne mgtA-lacZ reporter, compared with that seen in the chr+ background.
Table 2.
Increased expression of the chromosomal mgtA+ gene by the chr-1 mutation down-regulates the expression of a plasmid-borne mgtA-lacZ fusion
| Plasmid | Chromosomal genotype | β-Galactosidase-specific activity, Miller units |
|---|---|---|
| chr+mgtA-lacZ | chr-1 mgtA+ | 460 ± 2 |
| chr+mgtA+ | 1,066 ± 31 |
The β-galactosidase specific activities of strains TL4335 (chr+ mgtA+ / chr+ mgtA-lacZ) and TL4337 (chr-1 mgtA+ / chr+ mgtA-lacZ) were determined as described in Materials and Methods.
Overproduction of the Bacillus subtilis Mg2+ Transporter MgtE also Increases Thermotolerance.
In addition to MgtA, MgtB, and CorA, there is a third type of Mg2+ transporter, MgtE, which is found in organisms in all three biological kingdoms, although not in Salmonella and Escherichia coli (33). To test whether the increased thermotolerance in the mgtA constitutive mutants was the result of increased Mg2+ transport activity, we determined the effect of overexpression of an MgtE protein. The strain used for this experiment harbored a plasmid carrying the mgtE+ gene of B. subtilis expressed from the arabinose-inducible pBAD promoter (34). The induction of mgtE+ by arabinose increased the heat resistance of the strain to a level that was comparable with that seen in the mgtA riboswitch mutants (Fig. 4). Because the B. subtilis MgtE does not have significant sequence similarity to the S. typhimurium MgtA, MgtB, and CorA proteins (blastp Expect >0.18), this experiment supports the conclusion that the increased thermotolerance seen on the high-level expression of the Mg2+ transporters MgtA and MgtE is the result of increased uptake of this cation.
Fig. 4.
High-level expression of the MgtE protein also confers increased thermotolerance. Strain TL4445 carrying the empty cloning vector pBADMycHis(a) (A) and strain TL4449 carrying the B. subtilis mgtE gene under the control of the arabinose-inducible pBAD promoter (B) were grown in M63 + 20 mM glycerol without (−Ara) or with 1 mM L-arabinose (+Ara) and their heat resistance determined at 53 °C, as described in Materials and Methods.
Additional Characterization of the Role of Mg2+ Transport in Thermotolerance.
Because the chr-1 and the ΔUTRre-100 mutations resulted in similar increased thermotolerance and in similar loss of sensitivity of the regulation mgtA transcription by Mg2+, we used them interchangeably for some of the experiments. However, in all of the experiments that were done with both mutations, we always obtained consistent results. The ability of the riboswitch mutations to confer increased thermotolerance in M63 and M63 + 0.3 M NaCl + glycine betaine was abolished on the disruption of the mgtA gene with a MudJ insertion (Fig. S2). This result corroborates the conclusion that the enhanced thermotolerance in the mgtA-constitutive mutants is caused by increased Mg2+ uptake and not by some other effect of the riboswitch mutation (which is present in these mgtA::MudJ mutants). Interestingly, 0.3 M NaCl in the absence of glycine betaine still stimulated thermotolerance in the riboswitch mutants carrying the mgtA::MudJ insertion. The latter result implies that high osmolality increases heat tolerance by a different mechanism than the Mg2+ transport-dependent response. Inactivation of mgtA with the MudJ insertion in the chr+ background did not change the thermotolerance of the strain, indicating that this gene is normally not involved in thermal adaptation in the WT.
Although increased expression of mgtA confers enhanced thermotolerance, this gene is not induced by high temperature. The β-galactosidase-specific activities of the chr+ mgtA-lacZ strain grown exponentially with 0.16 mM Mg2+ were 59 ± 2 and 55 ± 2 Miller units at 30 °C and 42 °C, respectively. The β-galactosidase activities at these two temperatures in the chr-1 mgtA-lacZ strain were 1,377 ± 84 and 695 ± 28 Miller units, and in the ΔUTRre−100 mgtA-lacZ strain they were 833 ± 40 and 454 ± 8 Miller units, respectively. The ≈2-fold repression of the mgtA-lacZ fusion at 42 °C vs. 30 °C in the riboswitch mutants may be caused by a down-regulation of transcription initiation by the PhoPQ system or to a temperature-dependent functioning of a vestigial riboswitch, but we have not examined this effect further.
The treR and the mgtA genes are transcribed from partially overlapping divergent promoters (35). To test whether the chr-1 mutation might confer increased thermotolerance by altering the expression of treR, we constructed treR-lacZ fusions using DNA fragments that carried the chr-1 or chr+ alleles. The β-galactosidase assays indicated the chr-1 mutation did not alter the transcription from this promoter (Table S2), demonstrating that the increased thermotolerance is not caused by change in the expression of treR. We confirmed in S. enterica the observation made in E. coli (10, 36) that loss of the RpoS protein causes a drastic reduction in thermotolerance in both exponential and stationary phase. However, the chr-1 mutation still conferred increased thermotolerance in an rpoS::amp mutant background in both growth phases (Fig. S3), demonstrating that increased Mg2+ transport confers increased thermotolerance by a mechanism that does not depend on RpoS function.
The chr-1 mutation did not stimulate the growth rate of the cells near the upper limit of temperature in the absence or the presence of methionine, which is required above 43 °C because of the lability of homoserine transsuccinylase (37). This result suggests that the growth cessation at nonlethal high temperatures and the loss of viability at lethal high temperatures are the consequences of different deleterious effects of thermal stress. As was the case with the ΔUTRre-100 mutant (19, 27, 31), expression of the mgtA-lacZ fusion in the chr-1 mutant was reduced by a phoP::Tn10 insertion to background levels (to 9 and 7 Miller units at 0.016 and 16 mM Mg2+, respectively), demonstrating that transcription from the mgtA promoter in this mutant still depends on the PhoPQ system. The phoP insertions also abolished the increased thermotolerance in the chr-1 mgtA+ background.
Last, increasing the Mg2+ concentration of the medium did not increase the thermotolerance of the WT or the chr-1 mutant. However, in view of the relative insensitivity of the internal Mg2+ pool to the external concentration (16, 17), high-level expression of Mg2+ transport systems might be more effective in altering the control of the intracellular Mg2+ homeostasis than by increasing the external concentration.
Discussion
Why Does Increased Mg2+ Transport Confer Increased Thermotolerance?
The isolation of the chr-1 mutation provided the insight that increased expression of Mg2+ transport proteins dramatically increases the thermotolerance of Salmonella. Richmond et al. (38) reported that corA is induced by heat shock in E. coli, providing a prior line of evidence for the beneficial effects of increased Mg2+ uptake at high temperature. Although we cannot propose a definitive model for the connection between thermotolerance and Mg2+ homeostasis, we can offer a number of hypotheses.
It is generally believed that protein denaturation is the primary cause of death at high temperature, which is supported by the fact that prominent members of the heat shock regulon are chaperonins and proteases (3). Increased cytoplasmic concentrations of free Mg2+ might enhance thermotolerance by stabilizing proteins or protein–nucleic acid interactions. A second possibility is that membrane damage may contribute to the thermal death. There are data indicating that Mg2+ is needed for membrane integrity (14). High external concentrations of this cation increase the resistance of yeast to both high temperature and ethanol by counteracting increased membrane leakage (39). It is possible to isolate E. coli mutants that are almost completely devoid of phosphatidylethanolamine by disrupting the pss (phosphatidylserine synthase) gene (40). In a pss mutant, phosphatidylethanolamine, which is zwitterioninc, is replaced by the anionic lipids phosphatidylglycerol and cardiolipin. A pss null mutant is inviable in media containing normal concentrations of Mg2+, but it can grow in media augmented with ≥10 mM Mg2+, Ca2+, or Sr2+. High levels of these divalent cations may be able to repair the growth defect of the pss mutant by neutralizing the excess negative charge of the phospholipids (41). Lusk and Kennedy (42) isolated an E. coli mutant with an uncharacterized defect in lipid biosynthesis that was also inhibited at low Mg2+, but could be rescued by high concentrations of this cation. Induction of mgtA by the overproduction of the Rob protein has been found to increase the resistance of S. enterica to cyclohexane (30), which could be a manifestation of increased membrane stability resulting from increased Mg2+ accumulation. If high temperature causes membrane damage, it is possible that increased cytoplasmic concentrations of Mg2+ resulting from higher transport activity might improve the thermotolerance of the cells by stabilizing the phospholipid bilayer from the inside.
A third conceivable reason for the thermoprotective effect of Mg2+ could be that it might mitigate oxidative stress at high temperature. The production and the reactivity of oxygen radicals increases with increasing temperature, suggesting that oxidative damage could contribute to the lethal effects of thermal stress (43). Limitation for Mg2+ increases the accumulation of Fe2+ (44, 45), and thereby, renders the cells hypersensitive to oxidative damage. Increased accumulation of free Mg2+ may result in improved thermotolerance by lessening oxidative damage preventing the excessive uptake of Fe2+, or competing with Fe2+ for binding to macromolecules.
Last, it is possible that Mg2+ might be a metabolic signal that up-regulates the thermotolerance response. Mg2+ has been proposed to be an important regulatory signal for the activation of mitosis in animal cells (46), and it might have a regulatory role in inducing or activating some thermoprotective functions in bacteria. If Mg2+ functions as a regulatory signal, it might exert this role as the free cation or in a complex with some other metabolite, such as ATP or ADP (46). Because the binding of proteins to nucleic acids is exquisitely sensitive to the ion concentration (47), small changes in the free Mg2+ concentration could have profound effects on gene expression. It will be of interest to carry out transcriptomic or proteomic studies to determine the effects of increased Mg2+ accumulation on global gene expression.
Why Does the chr-1 Mutation Increase the Expression of the mgtA Gene?
According to the model proposed by Cromie et al. (31), the mgtA riboswitch RNA can shift between two different conformations in regulating transcription termination upstream of the mgtA gene in response to the cytoplasmic Mg2+ concentration. However, surprisingly, the nucleotide at position +98 that was changed by the chr-1 mutation is in a single-stranded region that would not be predicted to be crucial for the formation of stem structures of the riboswitch (Fig. 2). This result suggests that the regulatory mechanism of the riboswitch may be more intricate than proposed by Cromie et al. (31). It is possible that regulation by the riboswitch might involve a protein that recognizes the base altered by the chr-1 mutation or it may be mediated by some higher-order structure in the RNA, such as a pseudoknot, that is affected by the mutation. Another interesting possibility could be that the chr-1 mutation affects a nucleotide that is part of the Mg2+ “binding site” of the riboswitch.
Does Mg2+ Transport Have a Role in the Regulation of Thermotolerance in Other Organisms?
Our finding that overproduction of the B. subtilis MgtE protein resulted in increased thermotolerance of Salmonella (Fig. 4) suggests that Mg2+ transport may be involved in the regulation of thermotolerance in other bacteria. Using the database SEED (www.theseed.org/wiki/Main_Page), we found that in Yersinia, the mgtE ortholog is 136 nt upstream of and in the same orientation as the Hsp70 family heat shock gene yegD, and in Vibrio cholerae, mgtE is 186 nt upstream of and in the same orientation as the Hsp60 chaperonin genes groES and groEL. Because in prokaryotes genes that have related functions or share regulatory signals are often located in close proximity to each other (48, 49), the clustering of an mgtE ortholog with heat shock genes raises the possibility that this Mg2+ transport gene may be part of the heat shock regulon in these bacteria. Although at present this is only speculative, if mgtE were induced by heat shock in these organisms, this result would provide additional support for the conclusion that increased Mg2+ uptake is generally beneficial for high temperature tolerance.
Materials and Methods
Media, Growth Conditions, and Bacterial Strains.
All bacterial strains used in this work were derived from strain TL1, which is our laboratory's line of WT S. enterica serovar Typhimurium LT2; the construction of the strains is summarized in Table S3. The experiments were carried with strains grown aerobically at 37 °C in minimal medium M63 (50), which contains 0.16 mM Mg2+.
Heat Killing Experiments.
Thermotolerance measurements were carried out at 53 °C with early exponential- or stationary-phase cultures grown in M63, as described in SI Materials and Methods. The data shown are typical of results obtained in at least five independent measurements.
Isolation of the Thermotolerant Mutants.
The isolation of the heat resistant mutants is described in detail in SI Materials and Methods, but briefly, the chr-1 and chr-2 mutants were obtained from two independent cultures of strain TL1 that were mutagenized with ethylmethane sulfonate and subjected to three cycles of heat treatment at 53 °C for 4 min M63 + glucose + 0.3 M NaCl + glycine betaine, each followed by outgrowth at 30 °C in the same medium. Because the isolation of the mutants involved three cycles of amplification of the survivors of the 53 °C treatment, we cannot estimate of the frequency of the heat resistance mutants in the mutagenized population.
Supplementary Material
Acknowledgments.
We thank B. J. Gasper, C. A. Lindquist, and D. F. Ready for critical comments on the manuscript; A. D. Hanson for insightful discussions; D. M. Downs, E. A. Groisman, M. E. Maguire, M. J. Mahan, and J. R. Roth for bacterial strains; D. L. Court and E. Kofoid for advice on recombineering; and E. A. Groisman for advice on sequencing from total chromosomal DNA. This work was supported by United States Department of Agriculture National Research Initiative Food Safety Program Grant 2004-04337.
Footnotes
The authors declare no conflict of interest.
This article is a PNAS Direct Submission.
This article contains supporting information online at www.pnas.org/cgi/content/full/0906160106/DCSupplemental.
References
- 1.Uzzau S, et al. Host adapted serotypes of Salmonella enterica. Epidemiol Infect. 2000;125:229–255. doi: 10.1017/s0950268899004379. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.CDC. Preliminary FoodNet Data on the incidence of infection with pathogens transmitted commonly through food–10 states, 2008. Morbid Mortal Wkly Rep. 2009;58:333–337. [PubMed] [Google Scholar]
- 3.Guisbert E, Yura T, Rhodius VA, Gross CA. Convergence of molecular, modeling, and systems approaches for an understanding of the Escherichia coli heat shock response. Microbiol Mol Biol Rev. 2008;72:545–554. doi: 10.1128/MMBR.00007-08. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Klauck E, Typas A, Hengge R. The σS subunit of RNA polymerase as a signal integrator and network master regulator in the general stress response in Escherichia coli. Sci Prog. 2007;90:103–127. doi: 10.3184/003685007X215922. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Alvarez-Ordóñez A, Fernández A, López M, Arenas R, Bernardo A. Modifications in membrane fatty acid composition of Salmonella typhimurium in response to growth conditions and their effect on heat resistance. Int J Food Microbiol. 2008;123:212–219. doi: 10.1016/j.ijfoodmicro.2008.01.015. [DOI] [PubMed] [Google Scholar]
- 6.Taglicht D, Padan E, Oppenheim AB, Schuldiner S. An alkaline shift induces the heat shock response in Escherichia coli. J Bacteriol. 1987;169:885–887. doi: 10.1128/jb.169.2.885-887.1987. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Cánovas D, Fletcher SA, Hayashi M, Csonka LN. The role of trehalose in growth at high temperature in Salmonella typhimurium. J Bacteriol. 2001;183:3365–3371. doi: 10.1128/JB.183.11.3365-3371.2001. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Tesone S, Hughes A, Hurst A. Salt extends the upper temperature limit for growth of food-poisoning bacteria. Can J Microbiol. 1981;27:970–972. doi: 10.1139/m81-154. [DOI] [PubMed] [Google Scholar]
- 9.Calhoun CL, Frazier WC. Effect of available water on thermal resistance of three nonsporeforming species of bacteria. Appl Microbiol. 1966;14:416–420. doi: 10.1128/am.14.3.416-420.1966. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Hengge-Aronis R, Lange R, Henneberg N, Fischer D. Osmotic regulation of rpoS-dependent genes in Escherichia coli. J Bacteriol. 1993;175:259–265. doi: 10.1128/jb.175.1.259-265.1993. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Fletcher SA, Csonka LN. Characterization of the induction of increased thermotolerance by high osmolality in Salmonella typhimurium. Food Microbiol. 1998;15:307–317. [Google Scholar]
- 12.Csonka LN, Epstein W. Osmoregulation. In: Neidhardt FC, et al., editors. Escherichia coli and Salmonella typhimurium: Cellular and Molecular Biology. 2nd. Ed. Washington, DC: Am Soc Microbiol; 1996. pp. 1210–1223. [Google Scholar]
- 13.Cayley S, Lewis BA, Record MT., Jr Origins of the osmoprotective properties of betaine and proline in Escherichia coli K-12. J Bacteriol. 1992;174:1586–1595. doi: 10.1128/jb.174.5.1586-1595.1992. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Silver S. (1996) Transport of inorganic cations. In: Neidhardt FC, et al., editors. Escherichia coli and Salmonella typhimurium: Cellular and Molecular Biology. 2nd. Ed. Washington, DC: Am Soc Microbiol; 1996. pp. 1091–1102. [Google Scholar]
- 15.Outten CE, O'Halloran TV. Femtomolar sensitivity of metalloregulatory proteins controlling zinc homeostasis. Science. 2001;276:30670–30677. doi: 10.1126/science.1060331. [DOI] [PubMed] [Google Scholar]
- 16.Papp-Wallace KM, et al. The CorA Mg2+ channel is required for the virulence of Salmonella enterica serovar Typhimurium. J Bacteriol. 2008;190:6517–6523. doi: 10.1128/JB.00772-08. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Silver S, Clark D. Magnesium transport in Escherichia coli. J Biol Chem. 1971;246:569–576. [PubMed] [Google Scholar]
- 18.Kehres DG, Maguire ME. Structure, properties and regulation of magnesium transport proteins. BioMetals. 2002;15:261–270. doi: 10.1023/a:1016078832697. [DOI] [PubMed] [Google Scholar]
- 19.Tao T, Grulich PF, Kucharski LM, Smith RL, Maguire ME. Magnesium transport in Salmonella typhimurium: Biphasic magnesium and time dependence of the transcription of the mgtA and mgtCB loci. Microbiology. 1998;144:655–664. doi: 10.1099/00221287-144-3-655. [DOI] [PubMed] [Google Scholar]
- 20.Monsieurs P, et al. Comparison of the PhoPQ regulon in Escherichia coli and Salmonella typhimurium. J Mol Evol. 2005;60:462–474. doi: 10.1007/s00239-004-0212-7. [DOI] [PubMed] [Google Scholar]
- 21.Bader MW, et al. Recognition of antimicrobial peptides by a bacterial sensor kinase. Cell. 2005;122:461–472. doi: 10.1016/j.cell.2005.05.030. [DOI] [PubMed] [Google Scholar]
- 22.García Véscovi E, Soncini FC, Groisman EA. Mg2+ as an extracellular signal: Environmental regulation of Salmonella virulence. Cell. 1996;84:165–174. doi: 10.1016/s0092-8674(00)81003-x. [DOI] [PubMed] [Google Scholar]
- 23.Soncini FC, García Véscovi E, Solomon F, Groisman EA. Molecular basis of the magnesium deprivation response in Salmonella typhimurium: Identification of PhoP-regulated genes. J Bacteriol. 1996;178:5092–5099. doi: 10.1128/jb.178.17.5092-5099.1996. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Prost LR, Miller SI. The Salmonellae PhoQ sensor: Mechanisms of detection of phagosome signals. Cell Microbiol. 2008;10:576–582. doi: 10.1111/j.1462-5822.2007.01111.x. [DOI] [PubMed] [Google Scholar]
- 25.Fields PI, Groisman EA, Heffron F. A Salmonella locus that controls resistance to microbicidal proteins from phagocytic cells. Science. 1989;243:1059–1062. doi: 10.1126/science.2646710. [DOI] [PubMed] [Google Scholar]
- 26.Miller SI, Kukral AM, Mekalanos JJ. A two-component regulatory system (phoP phoQ) controls Salmonella typhimurium virulence. Proc Natl Acad Sci USA. 1989;86:5054–5058. doi: 10.1073/pnas.86.13.5054. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Heithoff DM, et al. Coordinate intracellular expression of Salmonella genes induced during infection. J Bacteriol. 1999;181:799–807. doi: 10.1128/jb.181.3.799-807.1999. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Smith RL, Kaczmarek MT, Kucharski LM, Maguire ME. Magnesium transport in Salmonella typhimurium: Regulation of mgtA and mgtCB during invasion of epithelial and macrophage cells. Microbiology. 1998;144:1835–1843. doi: 10.1099/00221287-144-7-1835. [DOI] [PubMed] [Google Scholar]
- 29.Groisman EA. The ins and outs of virulence gene expression: Mg2+ as a regulatory signal. BioEssays. 1998;20:96–101. doi: 10.1002/(SICI)1521-1878(199801)20:1<96::AID-BIES13>3.0.CO;2-3. [DOI] [PubMed] [Google Scholar]
- 30.Barchiesi J, Castelli ME, Soncini FC, Véscovi EG. mgtA expression is induced by Rob overexpression and mediates a Salmonella enterica resistance phenotype. J Bacteriol. 2008;190:4951–4958. doi: 10.1128/JB.00195-08. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Cromie MJ, Shi Y, Latifi T, Groisman EA. An RNA sensor for intracellular Mg2+ Cell. 2006;125:71–84. doi: 10.1016/j.cell.2006.01.043. [DOI] [PubMed] [Google Scholar]
- 32.Daly MJ. A new perspective on radiation resistance based on Deinococcus radiodurans. Nat Rev Microbiol. 2009;7:237–245. doi: 10.1038/nrmicro2073. [DOI] [PubMed] [Google Scholar]
- 33.Moomaw AS, Maguire ME. The unique nature of Mg2+ channels. Physiology. 2008;23:275–285. doi: 10.1152/physiol.00019.2008. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Papp-Wallace KM, Maguire ME. Regulation of CorA Mg2+ channel function affects the virulence of Salmonella enterica serovar Typhimurium. J Bacteriol. 2008;190:6509–6516. doi: 10.1128/JB.00144-08. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Yamamoto K, Ogasawara H, Fujita N, Utsumi R, Ishihama A. Novel mode of transcription regulation of divergently overlapping promoters by PhoP, the regulator of two-component system sensing external magnesium availability. Mol Microbiol. 2002;45:423–438. doi: 10.1046/j.1365-2958.2002.03017.x. [DOI] [PubMed] [Google Scholar]
- 36.Lange R, Hengge-Aronis R. Identification of a central regulator of stationary-phase gene expression in Escherichia coli. Mol Microbiol. 1991;5:49–59. doi: 10.1111/j.1365-2958.1991.tb01825.x. [DOI] [PubMed] [Google Scholar]
- 37.Ron EZ, Shani M. Growth rate of Escherichia coli at elevated temperatures: Reversible inhibition of homoserine trans-succinylase. J Bacteriol. 1971;107:397–400. doi: 10.1128/jb.107.2.397-400.1971. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Richmond CS, Glasner JD, Mau R, Jin H, Blattner FR. Genome-wide expression profiling in Escherichia coli K-12. Nucleic Acids Res. 1999;27:3821–3835. doi: 10.1093/nar/27.19.3821. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39.Birch RM, Walker GM. Influence of magnesium ions on heat shock and ethanol stress responses of Saccharomyces cerevisiae. Enzyme Microb Technol. 2000;26:678–687. doi: 10.1016/s0141-0229(00)00159-9. [DOI] [PubMed] [Google Scholar]
- 40.DeChavigny A, Heacock PN, Dowhan W. Sequence and inactivation of the pss gene of Escherichia coli: Phosphatidylethanolamine may not be essential for cell viability. J Biol Chem. 1991;266:5323–5332. [PubMed] [Google Scholar]
- 41.Killian JA, et al. Effect of divalent cations on lipid organization of cardiolipin isolated from Escherichia coli strain AH930. Biochim Biophys Acta. 1994;1189:225–232. doi: 10.1016/0005-2736(94)90069-8. [DOI] [PubMed] [Google Scholar]
- 42.Lusk JA, Kennedy EP. Altered phospholipid metabolism in a sodium-sensitive mutant of Escherichia coli. J Bacteriol. 1972;109:1034–1046. doi: 10.1128/jb.109.3.1034-1046.1972. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 43.Benov L, Fridovich I. Superoxide dismutase protects against aerobic heat shock in Escherichia coli. J Bacteriol. 1995;177:3344–3346. doi: 10.1128/jb.177.11.3344-3346.1995. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 44.Hantke K. Ferrous iron uptake by a magnesium transport system is toxic for Escherichia coli and Salmonella typhimurium. J Bacteriol. 1997;179:6201–6204. doi: 10.1128/jb.179.19.6201-6204.1997. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 45.Chamnongpol S, Groisman EA. Mg2+ homeostasis and avoidance of metal toxicity. Mol Microbiol. 2002;44:561–571. doi: 10.1046/j.1365-2958.2002.02917.x. [DOI] [PubMed] [Google Scholar]
- 46.Rubin H. The logic of the membrane, magnesium, mitosis (MMM) model for the regulation of animal cell proliferation. Arch Biochem Biophys. 2007;458:16–23. doi: 10.1016/j.abb.2006.03.026. [DOI] [PubMed] [Google Scholar]
- 47.Richey B, et al. Variability in the intracellular ionic environment of Escherichia coli: Differences between in vitro and in vivo effects of ion concentrations on protein-DNA interactions and gene expression. J Biol Chem. 1987;262:7157–7164. [PubMed] [Google Scholar]
- 48.Lawrence JG, Roth JR. Selfish operons - horizontal transfer may drive the evolution of gene clusters. Genetics. 1996;143:1843–1860. doi: 10.1093/genetics/143.4.1843. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 49.Overbeek R, Fonstein M, D'Souza M, Pusch GD, Maltsev N. The use of gene clusters to infer functional coupling. Proc Natl Acad Sci USA. 1999;93:12637–12642. doi: 10.1073/pnas.96.6.2896. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 50.Cohen GN, Rickenberg RH. Specific reversible concentration of amino acids in E. coli. Ann Inst Pasteur Paris. 1956;91:693–720. (in French) [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.




