Role of the Five RNA Helicases in the Adaptive Response of Bacillus cereus ATCC 14579 Cells to Temperature, pH, and Oxidative Stresses

Abstract

In this study, growth rates and lag times of the five RNA helicase-deleted mutants of Bacillus cereus ATCC 14579 were compared to those of the wild-type strain under thermal, oxidative, and pH stresses. Deletion of cshD and cshE had no impact under any of the tested conditions. Deletion of cshA, cshB, and cshC abolished growth at 12°C, confirming previous results. In addition, we found that each RNA helicase had a role in a specific temperature range: deletion of cshA reduced growth at all the tested temperatures up to 45°C, deletion of cshB had impact below 30°C and over 37°C, and deletion of cshC led mainly to a cold-sensitive phenotype. Under oxidative conditions, deletion of cshB and cshC reduced growth rate and increased lag time, while deletion of cshA increased lag time only with H₂O₂ and reduced growth rate at a high diamide concentration. Growth of the ΔcshA strain was affected at a basic pH independently of the temperature, while these conditions had a limited effect on ΔcshB and ΔcshC strain growth. The RNA helicases CshA, CshB, and CshC could participate in a general adaptation pathway to stressful conditions, with a stronger impact at low temperature and a wider role of CshA.

INTRODUCTION

The DEAD-box RNA helicases are encoded by viral, archaeal, eukaryotic, and prokaryotic genomes (9) and play an important role in RNA processing, transport, and degradation and in many other processes involving RNA (4, 19, 26), such as translation or ribosome biogenesis (10, 11, 22). DEAD-box RNA helicases act as molecular motors that unwind double-stranded RNA, thereby affecting the rearrangement of RNA secondary structures (9, 21). RNA helicases could also be implicated in rearrangement of ribonucleoprotein (RNP) complexes by removing protein from RNA or by the combination of both RNA-unwinding and RNA-annealing activity to promote RNA strand exchange through a potential branch migration (5, 13, 17, 24). Bacterial cells often encounter stressful conditions that tend to decrease the cellular fitness. Consequently, bacteria have to maintain RNA pathway functionalities and control their RNA turnover. Most of the synthesized mRNA is rapidly degraded to allow adaptation to environmental changes (14). RNA helicases could be involved in stress adaptation by maintaining and regulating RNA functions.

Studies reporting the involvement of prokaryotic RNA helicases in the adaptation to abiotic stress mainly deal with response to cold, light, and salt conditions (17). The RNA helicase CrhC maintains the photosynthetic capacity of the cyanobacterium Synechocystis. Its expression is regulated by the changes on the redox potential of the electron transport chain caused by variations in light, temperature, and salt concentrations (12). CrhC catalyzes the unwinding of RNA secondary structures but also ensures rearrangements in RNA complexes (5, 25). A Bacillus subtilis CshA homolog of Clostridium perfringens is involved in the adaptation to oxidative stress, with the corresponding null mutant strain showing better survival under oxidative stress conditions (2). The involvement of RNA helicases in adaptation to environmental stresses has been most often studied in response to low temperature. Thus, Escherichia coli RNA helicases CsdA and SrmB allow the correct folding of 50S ribosomal subunits at 20°C (6, 23).

Bacillus cereus is a food-borne pathogen, widely spread in the environment and in a wide range of foods, which consequently has to face many physical and chemical stresses. B. cereus RNA helicases have recently been shown to be implicated in cold adaptation (3, 18). Five open reading frames were identified by an in silico analysis as encoding the putative RNA helicases CshA to CshE in the B. cereus ATCC 14579 genome. Characterization of the five deleted mutants showed that CshA, CshB, and CshC were essential for the adaptation of B. cereus at 10°C, while CshD and CshE were not (18).

The aim of this work was to determine whether the Csh RNA helicase family of B. cereus could be involved in adaptation to a range of temperatures and oxidative and pH stress conditions that may be encountered by this pathogenic bacterium in the food chain.

MATERIALS AND METHODS

Growth experiments.

The B. cereus ATCC 14579 strain (wild type [WT]) and the five RNA helicase null mutants (18) were grown in LB broth, pH 6.6, for 16 h at 30°C and 200 rpm. All growth experiments were performed in microplates: each well was inoculated with the appropriate volume of the tested subculture to reach an initial optical density at 600 nm (OD₆₀₀) of 0.09 in a final volume of 250 μl of LB broth. Microplates were placed in the reading chamber of a Bioscreen C (Oy Growth Curves AB Ltd., Helsinki, Finland), where they were incubated at constant temperature in the range of 10°C to 45°C and continuously shaken at the medium intensity. Bioscreen C was preheated at the set point temperature for 2 h before starting growth experiments, and OD₆₀₀ was recorded every 15 min or every 30 min at low temperature. Hydrogen peroxide and diamide concentrations were adjusted to 4.5 mM and 8.1 mM, respectively, and from 0.5 mM to 1.5 mM by addition of sterile solutions after LB broth sterilization at 120°C for 20 min. To test the effect of pH on B. cereus growth, the pH of LB broth was adjusted by the addition of 1 M HCl or 1 M NaOH to 5.0, 7.0, and 8.5 and adjusted again after sterilization with 1 N NaOH if necessary. For each pH and H₂O₂ and diamide concentration, three replicate growth kinetics assays were done with each of at least two independent inocula. CFU were enumerated by plating on LB agar plates 100 μl of serial dilutions of the WT and mutant cells submitted to stress conditions. Plates were then incubated overnight at 37°C.

Catalase activity of WT and mutant cells was tested on LB agar plates, as previously described (8). H₂O₂ concentration was measured during the kinetic assay of incubation of cells with hydrogen peroxide, using the ferrous oxidation in xylenol orange (FOX2) assay, as previously described (20).

Estimation of the growth rate and of the lag time from OD₆₀₀ growth curves.

All wells were inoculated at an OD₆₀₀ of at least 0.1. For all conditions, the maximum specific growth rate, μ_max, was estimated from the OD₆₀₀ curves as a function of incubation time with the modified Gompertz equation (28)

$$\ln \left(\frac{{\mathrm{ABS}}_t}{{\mathrm{ABS}}_{t0}}\right)=A·\exp \left\{-\exp \left[\left(\frac{\mu \max ·\mathrm{e}}{A}\right)·\left(\lambda -t\right)+1\right]\right\}$$ (1)

where ABS_t is the OD₆₀₀ of the suspension at the tested time, ABS_t0 is the initial OD₆₀₀, A is the logarithmic increase of bacterial population, and λ is the lag time determined by the model. The modified Gompertz equation was described as one of the most appropriately fitting models to determine the specific growth rate (μ_max) when the initial OD₆₀₀ of the inoculum is above the Bioscreen C detection threshold (1).

Lag time was estimated by the time to detection (Ttd), corresponding to the abscissa of the point where the initial OD₆₀₀ was increased by 0.05 unit. This point was determined by the equation

$$\mathrm{Ttd}=\left(\frac{\left(\left({\mathrm{ABS}}_{t0}+0.05\right)-b\right)}{a}\right)$$ (2)

where a represents the slope of the line and b is the y intercept after determining this equation of the curve by a linear regression between the point immediately before and the point immediately after the target OD₆₀₀ + 0.05. WT and mutant strains were paired under each stress condition. The significance of the difference between the WT strain and the mutant strain growth parameters was statistically tested by a bilateral paired Student t test with 95% confidence.

Model fit and statistical analysis.

Fits were performed by using the least squares criterion. The fitting of the curves was improved by minimizing the sum of the squared residual (SSR) values using the Microsoft Excel solver function between both the growth and the fitted curves where SSR is defined as follows:

$$\mathrm{SSR}={{\displaystyle \sum _{i=1}^n\left(\mathrm{value}{\left(i\right)}_{\mathrm{observed}}-\mathrm{value}{\left(i\right)}_{\mathrm{fitted}}\right)}}^2$$ (3)

Growth parameters μ_max and lag of the five csh-deleted mutants were determined for each stress condition and were compared to the corresponding WT strain growth parameters in order to determine if the mutants were negatively, positively, or not affected by the tested stress conditions compared to the WT strain.

RESULTS AND DISCUSSION

Responses of B. cereus ATCC 14579 RNA helicase mutant strains exposed to temperature, oxidative, and pH stresses were compared to that of the wild type. The two parameters, lag time (Tdt) and growth rate (μ_max), were determined to distinguish impact of the RNA helicase on adaptation to the stress conditions upon inoculation and the subsequent growth phase, respectively.

Growth of RNA helicase mutants at different temperatures.

Growth of the WT strain occurred at 12°C after a lag time of about 74 h and a maximum growth rate value of about 0.1 h⁻¹ (Table 1 and Fig. 1), while no growth was observed for ΔcshA, ΔcshB, and ΔcshC strains at this temperature, even after a 5-day incubation (Table 1). Growth of ΔcshD and ΔcshE mutants was similar to that of the WT strain at 12°C. At 20°C, the lag time of the WT was shorter than that at 12°C and the μ_max reached 0.4 h⁻¹. The Ttd at 20°C of ΔcshA and ΔcshC strains was longer than that of the WT, while that of the ΔcshB strain was not significantly different (Table 1). The longer lag times of the ΔcshA and ΔcshC strains correspond to a phase where cells were adapting to the stress condition and were not due to cell death, as CFU remained relatively constant during the lag phase for both mutant strains (data not shown). The μ_max values of ΔcshA, ΔcshC, and ΔcshB strains were 0.1 h⁻¹, 0.2 h⁻¹, and 0.3 h⁻¹, respectively, and were significantly lower than that of the WT (0.4 h⁻¹) (Fig. 1). ΔcshD and ΔcshE strain Ttd and μ_max values were similar to that of the WT (Table 1; Fig. 1; also data not shown). At 30°C, the ΔcshA strain was the only mutant with a lower μ_max than that of the WT, with a mean value of 0.5 h⁻¹, versus 0.8 h⁻¹ for the WT (Fig. 1). ΔcshA and ΔcshC strains had lag phases of 1.1 h and 1.0 h, respectively, indicating that these mutant strains needed 0.4 h more than the WT did to adapt and start to grow at this temperature (Table 1). When strains were grown at 37°C, the ΔcshA mutant Ttd (1.1 h) and maximum growth rate (1.2 h⁻¹) were still slightly lower than those of the WT (Fig. 1 and Table 1). We also tested the growth capacity of the csh mutants beyond the optimal temperature, at 42°C and 45°C: at 42°C, the Ttds of the five csh mutants were similar to that of the WT (Table 1) but the μ_max values of the ΔcshA and ΔcshB strains, both equal to 1.1 h⁻¹, were lower than the 1.3-h⁻¹ value observed for the WT (Fig. 1). All mutants grew similarly to the WT at 45°C, except the ΔcshB strain, which had a significantly lower μ_max (1.3 h⁻¹) than did the WT (1.5 h⁻¹). Compared to the WT, the Ttds of ΔcshA, ΔcshB, and ΔcshC strains were 10 to 50% longer (Table 1).

Table 1.

Times to detection (Ttds) of B. cereus ATCC 14579 and of its RNA helicase isogenic mutants according to growth temperature

Strain	Mean Ttd ± SD (h) at tempa:
	12°C	20°C	30°C	37°C	42°C	45°C
WT	73.8 ± 3.3	3.3 ± 0.6	0.7 ± 0.3	0.8 ± 0.1	0.8 ± 0.0	0.5 ± 0.0
ΔcshA mutant	NGO	4.2 ± 0.8*	1.0 ± 0.1*	1.1 ± 0.1*	0.8 ± 0.1	0.8 ± 0.1*
ΔcshB mutant	NGO	3.1 ± 0.7	0.8 ± 0.4	ND	0.8 ± 0.1	0.6 ± 0.1*
ΔcshC mutant	NGO	5.5 ± 0.8*	1.0 ± 0.1*	ND	0.8 ± 0.0	0.6 ± 0.1*
ΔcshD mutant	70.5 ± 6	2.5 ± 0.3*	0.8 ± 0.3	ND	0.7 ± 0.1	0.5 ± 0.0

^aThe Ttd values of the mutant strains were compared to those of the WT. The difference between WT and mutant Ttd values was statistically tested by a paired Student t test with 95% confidence, and results were represented as follows: *, significant difference observed between mutant and WT. Ttd values of the ΔcshE mutant were not statistically different from that of the WT and were not listed in this table. NGO, no growth observed after 125 h; ND, not done.

Fig. 1.

Effect of temperature on the growth rates of B. cereus ATCC 14579 (WT) (□) and ΔcshA (▵), ΔcshB (○), and ΔcshC (■) mutant strains growing in LB broth. Values are the means of nine values from three independent cultures; error bars indicate standard deviations, and an asterisk indicates a significant difference (P < 0.05) from the μ_max of the B. cereus WT strain. For clarity, ΔcshD and ΔcshE strain μ_max values were not represented in this figure, as their μ_max values were not significantly different from that of the WT strain (P > 0.05).

The RNA helicases CshA, CshB, and CshC were all required for B. cereus growth at low temperatures but not over the same range of temperatures, indicating that they probably do not act all by the same mechanism. Indeed, previous experiments of cross complementation of ΔcshB and ΔcshC strains with a cshA copy in trans showed that the overexpression of cshA was able to restore the WT phenotype in the ΔcshB mutant but not in the ΔcshC mutant at low temperature (18). These experiments suggest that CshA and CshB could have overlapping functions and are consistent with a similar thermal requirement for CshA and CshB. CshC may have functionalities complementary to those of CshA. It is the only RNA helicase acting specifically during cold acclimation and may act differently. Because of its strong involvement at low temperature and its role at all other tested temperatures, CshA may have both a major role in cold stress and an important role as an essential protein for B. cereus growth. The RNA helicase requirement for B. cereus growth at different temperatures contrasted with B. subtilis, as the deletion of the cshA or cshB gene did not reduce the growth rate of these bacteria at 15°C (10). In E. coli, the deletion of the genes encoding CsdA or SrmB RNA helicases led to a cold-sensitive phenotype at 20°C (6, 7). Some RNA helicases of B. cereus are required both for adaptation, as revealed by an increase of lag time in the mutant, and for maintaining an optimal growth rate under stressful conditions, as revealed by a lower μ_max in the mutant.

Growth of csh mutants exposed to H₂O₂ and diamide.

Growth of the WT and the five csh mutant strains was tested at 30°C with 4.5 mM and 8.1 mM hydrogen peroxide. The Ttd of the WT increased from 0.7 h (control) to 4.6 h with 4.5 mM H₂O₂ and to 13.2 h with 8.5 mM H₂O₂ (Table 1 and Table 2), while there was no significant difference in μ_max values at the tested H₂O₂ concentrations (Fig. 2).

Table 2.

Times to detection (Ttds) of B. cereus ATCC 14579 and of its RNA helicase isogenic mutants according to hydrogen peroxide and diamide concentration at 30°C

Strain	Mean Ttd ± SD (h) at concn ofa:
	H2O2		Diamide
	4.5 mM	8.1 mM	0.5 mM	0.8 mM	1 mM	1.3 mM	1.5 mM
WT	4.6 ± 3.0	13.2 ± 1.0	0.8 ± 0.0	1.1 ± 0.1	1.5 ± 0.2	2.9 ± 0.6	3.5 ± 0.6
ΔcshA mutant	13.3 ± 3.7*	22.5 ± 3.8*	1.5 ± 0.1*	1.9 ± 0.2*	2.2 ± 0.2*	2.9 ± 0.3	3.2 ± 0.1
ΔcshB mutant	9.8 ± 3.7*	NGO	1.0 ± 0.0*	1.5 ± 0.2*	1.6 ± 0.1*	1.7 ± 1.1	3.5 ± 0.6
ΔcshC mutant	10.8 ± 2.9*	NGO	1.2 ± 0.1*	1.6 ± 0.1*	2.0 ± 0.1*	3.3 ± 0.3	3.5 ± 0.7

^aFor the statistical procedures and the definition of asterisks, see footnote a of Table 1. NGO, no growth observed after 48 h. Ttd values of ΔcshD and ΔcshE mutants were not statistically different from that of the WT and were not listed in this table.

Fig. 2.

Effect of H₂O₂ concentration on the growth rate of B. cereus ATCC 14579 (black bars) and ΔcshA (gray bars), ΔcshB (white bar), and ΔcshC (striped bar) mutants in LB broth at 30°C. Values are the means of six values from two independent cultures, error bars indicate standard deviations, and an asterisk indicates a significant variation of the μ_max value from that of WT. ΔcshD and ΔcshE strain μ_max values were never statistically different (P > 0.05) from that of the WT and were not represented. NGO, no growth observed after 48 h for ΔcshB and ΔcshC strains.

In the presence of 4.5 mM H₂O₂, the Ttds of ΔcshA, ΔcshB, and ΔcshC strains, equal to 13.3 h, 9.8 h, and 10.8 h, respectively, were at least 2-fold higher than the lag time of the WT (Table 2). The μ_max of ΔcshA (0.3 h⁻¹) was lower than that of the WT (0.7 h⁻¹), while the μ_max values of ΔcshB and ΔcshC strains were similar to that of the WT. We observed a 2-log decrease in ΔcshB and ΔcshC populations and a 5-log decrease for ΔcshA cells during the lag time (data not shown). This clearly shows that the growth of the ΔcshA strain, observed after a long lag phase in the presence of 4.5 mM H₂O₂, was due to a very low number of surviving cells. This suggests that CshA may have a role in resistance of B. cereus to hydrogen peroxide stress. It was previously reported that a C. perfringens RNA helicase was involved in the oxidative response, but its exact role was not described (2).

B. cereus possesses a catalase activity, which was also present in the mutant cells (data not shown). We determined that the 4.5 mM H₂O₂ concentration was rapidly reduced to 50 nM in the presence of B. cereus WT or mutant cells but remained constant after extended incubation. We checked, as a control, that H₂O₂ was not degraded during incubation at 30°C in the absence of B. cereus cells. B. cereus cells were consequently rapidly exposed to a lower concentration of H₂O₂ than initially planned, but the lag phase was clearly due to this residual H₂O₂ concentration.

When the concentration of H₂O₂ was increased to 8.1 mM, the WT strain needed 13.2 h to start its growth and the Ttd of the ΔcshA strain (22.5 h) was markedly longer than the WT Ttd. No growth of ΔcshB and ΔcshC strains was detected with 8.1 mM H₂O₂ (Table 2). Growth of the ΔcshA strain was impaired with a lower μ_max (0.3 h⁻¹) than that of the WT (0.7 h⁻¹) (Fig. 2). The μ_max values and the Ttds of ΔcshD and ΔcshE strains were not different from those of the WT for the two concentrations of H₂O₂ at 30°C (data not shown). We showed that ΔcshA growth was less affected at 37°C than at 30°C, so we tested the effect of H₂O₂ at 37°C: the ΔcshA strain had a significantly higher Ttd than did the WT in the presence of 4.5 mM H₂O₂ (Table 3).

Table 3.

Times to detection (Ttds) of B. cereus ATCC 14579 WT and ΔcshA mutant according to diamide concentration at 37°C

Strain	Mean Ttd ± SD (h) at concn ofa:
	H2O2		Diamide
	4.5 mM	8.1 mM	0.5 mM	0.8 mM	1 mM	1.3 mM	1.5 mM
WT	9.2 ± 0.3	ND	0.9 ± 0.2	1.6 ± 0.4	2.7 ± 0.2	4.4 ± 0.6	5.9 ± 0.3
ΔcshA mutant	10.2 ± 0.7*	ND	1.2 ± 0.4	2.1 ± 0.5*	2.7 ± 0.4	4.6 ± 1.2	4.4 ± 2.6

^aFor statistical procedures and definitions of abbreviations and asterisks, see footnote a of Table 1.

The effect of diamide at concentrations ranging between 0.5 mM and 1.5 mM was tested on WT and mutant strains at 30°C: the increasing concentrations of diamide increased the lag time for WT but had no significant impact on the maximum specific growth rate (Table 2 and Fig. 3 A). The Ttds of ΔcshA, ΔcshB, and ΔcshC strains were significantly higher than that of the WT for 0.5, 0.8, and 1.0 mM diamide but not for higher concentrations (Table 2), while ΔcshD and ΔcshE strain Ttd values were not modified at all (not shown). Whatever the concentration of diamide, ΔcshA, ΔcshB, and ΔcshC strains had a significantly lower μ_max than did the WT (Fig. 3A) while ΔcshD and ΔcshE strains grew similarly to the WT. When tested at 37°C, the Ttd of the ΔcshA strain was similar to that of the WT, except at the 0.8 mM concentration (Table 3). The growth rate of the ΔcshA strain was similar to that of the WT for 0.5 mM, 0.8 mM, and 1.0 mM diamide but was significantly lower at 1.3 and 1.5 mM diamide (Fig. 3B).

Fig. 3.

Effect of diamide concentration at 30°C on the μ_max values of B. cereus ATCC 14579 (WT) (□) and ΔcshA (▵), ΔcshB (○), and ΔcshC (■) mutant strains in LB broth (A) and at 37°C on the μ_max values of WT strain (□) and of the ΔcshA mutant (▵) (B). Values are the means of six values from two independent cultures, error bars indicate standard deviations, and an asterisk indicates a significant variation of the μ_max value from that of WT. μ_max values of ΔcshD and ΔcshE strains were never statistically different from the WT μ_max and were not represented.

RNA helicases act as chaperones to restore functionalities of RNA molecules altered by oxidative agents. However, CshA, CshB, and CshC do not seem to have the same impact during growth with oxidative agents. Hydrogen peroxide is an oxygen-reactive molecule that could damage biological targets by rupture of molecular bonds and then modification of their conformations and functions. The role of RNA helicases in resistance or adaptation to H₂O₂ stress had not been investigated so far. H₂O₂ can cause the dissociation of ribosomes in 30S and 50S subunits (16). Modification of ribosome properties or ribosome folding could explain why RNA helicases seem to be essential for the adaptation to hydrogen peroxide. The lesser role of RNA helicase in the presence of diamide than in the presence of H₂O₂ could be due to different modes of action of these oxidative compounds.

CshA, CshB, and CshC are required for optimal growth in both oxidative and low-temperature environments. Oxidative compounds could also prevent the action of enzymes which are involved in the adaptive modification of membrane fatty acid profiles. These modifications are required during cold adaptation, and that could partially explain the connection between these two stresses (27).

Growth of csh mutants at acid and basic pHs.

Growth of WT and the five csh mutant strains was studied at 30°C and at pH 5.0, pH 7.0, and pH 8.5. The Ttd of the WT at pH 5.0 was longer than that at pH 7.0, indicating a stress condition for B. cereus, as already described elsewhere (15). The Ttds of ΔcshA and ΔcshC strains, 4.1 h and 3.8 h, respectively, were twice as long as that for the WT (Table 4). The longer lag phases of the mutant strains were not due to cell death, as determined by CFU counts (data not shown). At pH 5.0, all the mutants had a maximum specific growth rate similar to that of the WT (Fig. 4A).

Table 4.

Times to detection (Ttds) of B. cereus ATCC 14579 and of its RNA helicase isogenic mutants according to pH values at 30°C

Strain	Mean Ttd ± SD (h) at pHa:
	5.0	7.0	8.5
WT	2.6 ± 0.4	0.7 ± 0.3	1.0 ± 0.1
ΔcshA mutant	4.1 ± 0.4*	0.9 ± 0.4	1.2 ± 0.1
ΔcshB mutant	2.2 ± 0.3	0.8 ± 0.4	1.0 ± 0.1
ΔcshC mutant	3.8 ± 0.5*	0.8 ± 0.3	1.3 ± 0.1

^aFor statistical procedures and definition of asterisks, see footnote a of Table 1. Ttd values of ΔcshD and ΔcshE mutants were not statistically different from that of the WT and were not listed in this table.

Fig. 4.

Effect of pH on μ_max at 30°C of B. cereus ATCC 14579 (WT) (□) and ΔcshA (▵), ΔcshB (○), and ΔcshC (■) mutant strains in LB broth (A) and on μ_max at 37°C of the WT strain (□) and the ΔcshA mutant (▵) (B). Values are the means of six values from two independent cultures, error bars indicate standard deviations, and an asterisk indicates a significant variation of the μ_max value from that of WT (P < 0.05). For clarity, ΔcshD and ΔcshE strain μ_max values were not represented, as they were not statistically different from the WT μ_max (P > 0.05).

WT strain growth rates at pH 8.5 and pH 7.0 were similar, indicating that such basic pH was not a stress condition for B. cereus. The Ttds of the five mutant strains were similar to that of the WT at pH 8.5 (Table 4), but the μ_max values of all mutants, except the ΔcshD and ΔcshE strains, were lower (Fig. 4A). At 37°C, the lag time of the ΔcshA strain was higher than that of the WT at both pH 5.0 and pH 8.0 (Table 5). At pH 8.0 and 37°C, the ΔcshA mutant μ_max was significantly lower than the WT μ_max, but the difference was reduced compared to that at 30°C (Fig. 4B). Lag times and μ_max values of ΔcshD and ΔcshE mutants at pH 5.0, 7.0, and 8.5 were not different from those of WT at 30°C (data not shown).

Table 5.

Times to detection (Ttds) of B. cereus ATCC 14579 WT and ΔcshA mutant according to pH values at 37°C

Strain	Mean Ttd ± SD (h) at pHa:
	5.0	7.0	8.5
WT	1.7 ± 0.2	1.1 ± 0.1	0.4 ± 0.1
ΔcshA mutant	2.6 ± 0.3*	0.9 ± 0.1*	0.7 ± 0.1*

^aFor statistical procedures and definition of asterisks, see footnote a of Table 1.

CshA is involved in basic pH adaptation, the absence of CshB impacted on the μ_max value at basic pH but not on adaptation, and CshC was required during adaptation to acid conditions and also for maintaining an optimal growth rate under basic growth conditions. Basic conditions could modify mRNA and rRNA conformations, leading to a loss of functionality and consequently requiring the action of RNA helicases as chaperones (17).

In conclusion, the RNA helicases CshA, CshB, and CshC participate in adaptation to several stressful conditions, each RNA helicase with some specificity, CshA having presumably the widest role. These three RNA helicases had the strongest impact, with a total absence of growth of the deletion mutants at low temperature and growth highly affected in the presence of H₂O₂. In contrast, CshD and CshE are not involved in adaptation to the abiotic stress tested in this work, and the function of these two RNA helicases in B. cereus is still unknown.