Genetic modification of the Salmonella membrane physical state alters the pattern of heat shock response

Abstract

It is now recognized that membranes are not simple physical barriers but represent a complex and dynamic environment that affects membrane protein structures and their functions. Recent data emphasize the role of membranes in sensing temperature changes, and it has been shown that the physical state of the plasma membrane influences the expression of a variety of genes such as heat shock genes. It has been widely shown that minor alterations in lipid membranes are critically involved in the conversion of signals from the environment to the transcriptional activation of heat shock genes. Previously, we have proposed that the composition, molecular arrangement, and physical state of lipid membranes and their organization have crucial roles in cellular responses during stress caused by physical and chemical factors as well as in pathological states. Here, we show that transformation of Salmonella enterica serovar Typhimurium LT2 (Salmonella Typhimurium) with a heterologous Delta(12)-desaturase (or with its trans-membrane regions) causes major changes in the pathogen's membrane dynamic. In addition, this pathogen is strongly impaired in the synthesis of major stress proteins (heat shock proteins) under heat shock. These data support the hypothesis that the perception of temperature in Salmonella is strictly controlled by membrane order and by a specific membrane lipid/protein ratio that ultimately causes transcriptional activation of heat shock genes. These results represent a previously unrecognized mode of sensing temperature variation used by this pathogen at the onset of infection.

FIG. 1.

Expression of the desA gene of Synechocystis PCC6803 in S. Typhimurium. desA is strongly expressed in Stm(pΔ¹²) (Δ¹²) heat shocked for 15 min at 41°, 45°, and 47°C while it is not expressed in Stm(pNir) (nir)containing the empty vector.

FIG. 2.

SDS-PAGE and Western blotting of S. Typhimurium strains. (A) Proteins of the membrane fraction were separated by SDS-PAGE and stained with Coomassie brilliant blue. (B) Western blot analysis using a monoclonal antibody against Synechocystis Δ¹²-desaturase shows that it was present in the membrane fraction of Stm(pΔ¹²). Lanes kDa, protein markers; lanes 1, membrane proteins from Stm(pNir) cells; lanes 2, membrane proteins from Stm(pΔ¹²). The arrow indicates the position of the Δ¹²-desaturase, with an apparent molecular mass of 36 kDa.

FIG. 3.

NPN fluorescence of Stm(pNir) and Stm(pΔ¹²) membranes. Overproduction of Δ¹²-desaturase induced membrane destabilization, which resulted in a permanent leakiness of the outer membrane of Stm(pΔ¹²) compared to Stm(pNir). The effect is very significant within the physiological range of 25° to 40°C while it is not present at higher temperatures. Bars represent the standard deviation. All statistical differences were evaluated by a two-tailed Student's t test. P values less than 0.05 were considered statistically significant. All experiments were performed at least in quadruplicates.

FIG. 4.

Differential scanning calorimetry of the isolated outer membrane of Stm(pNir) and Stm(pΔ¹²). (A) The overproduction of Δ¹²-desaturase lowered the transition temperature of lipid domains in the outer membranes of bacteria. Between 10° and 65°C, one major endothermic peak was observed in the first and in the second up-scans. The reversible endothermic peaks in the temperature range of 15° to 45°C corresponds to the phase transition of membrane lipids. (B) The midpoints of the phase transition were at 34.1° and 30.8°C for outer membranes of Stm(pNir) and of Stm(pΔ¹²), respectively.

FIG. 5.

Bacterial growth. Overnight cultures at 30° and 37°C of Stm(pΔ¹²) and Stm(pNir) were grown anaerobically in LB medium. A slight difference in growth was detected at these temperatures. Bars represent the standard deviations. All statistical differences were evaluated by a two-tailed Student's t test. P values less than 0.05 were considered statistically significant. All experiments were performed at least in quadruplicate.

FIG. 6.

Bacterial growth. Overnight cultures of Stm(pBAD-200), Stm(pBAD-212), and Stm(pBAD) grown at 37°C in RM minimal medium containing 0.4% glucose as a sole carbon source were inoculated in fresh RM minimal medium containing 2% arabinose to induce transcription of ORF200 and ORF212 from the P_BAD promoter. Expression of ORF200 strongly inhibited growth of Stm(pBAD-200) compared to the control strain Stm(pBAD) up to 9 h; expression of ORF212 had no significant effect on bacterial growth.

FIG. 7.

dnaK mRNA expression in Stm(pNir) (nir) and Stm(pΔ¹²) (Δ¹²). (A) dnaK was expressed at 30°C in Stm(pΔ¹²) but was scarcely detectable in the control strain Stm(pNir). dnaK expression increased in Stm(pΔ¹²) when the strain was heat shocked for 15 min at 37° or 39°C. dnaK expression decreased in Stm(pΔ¹²) when it was heat shocked at 43°C. (B and C) When Stm(pΔ¹²) was grown at the non-heat shock temperature of 37°C, dnaK was highly expressed. When cells were heat shocked between 41° and 47°C, dnaK expression was not detectable.

FIG. 8.

ibpB mRNA expression in Stm(pNir) and Stm(pΔ¹²). A strong decrease in ibpB expression was observed in Stm(pΔ¹²) under heat shock conditions compared to Stm(pNir).

FIG. 9.

SDS-PAGE analysis of outer membrane proteins stained with Coomassie brilliant blue. A significant accumulation of IbpA and IbpB in strain Stm(pΔ¹²) grown at 30°C is detectable.

FIG. 10.

Expression of dnaK in BA-treated Stm(pNir) cells. dnaK expression was absent in Stm(pNir) grown at 30°C under anaerobic conditions and incubated for 30 min with 50 mM BA. Both Stm(pΔ¹²) and BA-treated Stm(pNir) showed high levels of dnaK expression at 30°C.

FIG. 11.

Reversible expression of dnaK in BA-treated Stm(pNir) cells. dnaK is still detectable in Stm(pΔ¹²) after a shift to aerobic conditions; in BA-treated Stm(pNir) cells dnaK is no longer present when BA is removed from the medium.

FIG. 12.

Intracellular persistence inside MΦ of Stm(pNir), BA-treated Stm(wt), and Stm(pΔ¹²). At different time points, MΦwere lysed, and recovered bacteria were plated on LB agar or LB Amp¹⁰⁰ agar with similar plating efficiencies. Survival is expressed as a percentage of the number of CFU at each time point (0, 5, 10, 15, 20, and 30 min) compared to the number of CFU present at time zero. The efficiency of MΦ infection of Salmonella wt treated for 30 min with 50 mM BA within the first 5 min of infection was similar to that measured with Stm(pΔ¹²). However, within 15 min, BA-treated Stm(wt) recovered the ability to survive and multiply inside MΦ. Genetically modified Salmonella Stm(pΔ¹²) was unable to survive inside MΦ. Values are representative of three independent experiments, each performed in duplicate. Statistical differences were evaluated by a two-tailed Student's t test. P values less than 0.05 were considered significant.