Yersinia enterocolitica Type III Secretion Depends on the Proton Motive Force but Not on the Flagellar Motor Components MotA and MotB

Materials.The protonophore carbonyl cyanide m-chlorophenylhydrazone (CCCP) and the sodium motive force (SMF) inhibitors 2-n-heptyl-4-hydroxyquinoline N-oxide (HQNO) and phenamil were purchased from Sigma. Ingredients for Luria-Bertani (LB) and brain heart infusion (BHI) media were purchased from Difco.

Nucleic acid manipulations.An isogenic motAB mutant of Y. enterocolitica WA-314 (15) was constructed by the novel phage lambda Red recombinase cloning procedure (7). The entire coding regions of the motA and motB genes were replaced by a kanamycin resistance cassette. This was achieved by homologous recombination mediated by λ phage Redα and Redβ recombinases, which were expressed in Yersinia from a curable plasmid pKD46. For recombination, a PCR product harboring a kanamycin cassette with 50-nucleotide-homology arms was amplified from plasmid pACYC177 (New England Biolabs) with the following primers: MotABfor 5′-GGGTGCTAACGCCTCACCTTCCACCCACACACTTTGCTAAGGATATCGCGTCACTGACACCCTCATCAGTG-3′, and MotABrev 5′-GGCCGCAAAGCTTTCAGTCGCGGGTCTGATTCAATTAATTGATCCAGTTTCGTCAAGTCAGCGTAATGCTC-3′.

Transformation of this PCR product into the wild type resulted in WA-314ΔmotAB, which was verified by PCR and a motility assay.

For complementation of WA-314ΔmotAB, a 2.2-kb fragment encompassing the complete motAB genes, including the putative promotor region, was amplified by PCR with primers 5′-TCTAGAGTGCTAACGCCTCACCTTCCAC-3′ and 5′-GTCGACTACTGTGCGTCGCGGCTTG-3′, introducing XbaI and SalI restriction sites at the ends. The PCR product was ligated into vector pGEM-T (Promega). The insert was cut out with XbaI and SalI and ligated into pACYC184 (New England Biolabs) at the same sites. The resulting plasmid, pmotAB, was transformed into WA-314ΔmotAB.

Motility assay.Motility of Y. enterocolitica WA-314 and WA-314ΔmotAB was assessed by growing strains on 0.3% floating agar (LB medium) after overnight incubation at 28°C. Optionally, agar was supplemented with CCCP (30 μM), which was added immediately before agar (45°C) was poured into petri dishes.

Analysis of secreted and translocated proteins.Yop secretion was analyzed as described previously (28). In brief, yersiniae were cultured overnight at 27°C in BHI medium. Cultures were diluted 1:40 and incubated for 2 h at 37°C in BHI medium supplemented with 200 μM CaCl₂. Then, Yop secretion was induced by Ca²⁺ depletion with 5 mM EGTA (16). Unless otherwise indicated, bacteria were pelleted (15 min, 10,000 × g) after 1.5 h of continued incubation, supernatants were precipitated with trichloroacetic acid (TCA), and precipitates were subjected to sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE). CCCP was freshly dissolved in dimethyl sulfoxide (DMSO) and added to the cultures as indicated for the specific experiments. DMSO was added to controls not treated with CCCP to eliminate DMSO-induced side effects. The final concentration of DMSO in cultures was usually 0.5 to 1%.

Translocation of Yops was analyzed as described previously (29). HeLa cells were infected at a multiplicity of infection of 50 for 2 h. Subsequently, cells were detached by treatment with proteinase K (30 μg/ml), which was also applied to digest extracellular Yops. HeLa cells were washed twice in phosphate-buffered saline (PBS) supplemented with 1 mM phenylmethylsulfonyl fluoride (PMSF) and lysed by osmotic shock (resuspension in distilled H₂O with 1 mM PMSF). This treatment leaves adhering bacteria intact. Cell debris was pelleted, and the supernatant was precipitated with TCA. The precipitate was subjected to SDS-PAGE, followed by Western blotting to analyze translocated Yops.

Determination of cytosolic ATP.The ATP Bioluminescence Assay kit CLS II (Roche) was used according to the manufacturer's recommendations to analyze the influence of CCCP on intracellular ATP levels. Bacterial cultures (2 ml) were centrifuged, and the pellets were resuspended in 100 mM Tris-HCl (pH 7.75)-4 mM EDTA, with adjustment of the optical density at 600 nm (OD₆₀₀) of the cell suspension to 1.0. The cell suspensions (100 μl) were boiled for 2 min at 100°C. Samples were centrifuged, and 50 μl of each supernatant was transferred to a microtiter plate that was kept on ice until measurement. Luciferase reagent (50 μl) was injected, and luminescence read with a MicroLumatPlus LB 96V luminometer (Berthold Technologies) at 20°C. ATP standards provided with the kit were diluted in the range from 10⁻⁵ to 10⁻¹⁰ M ATP.

Immunofluorescent labeling of flagella.Bacteria grown on minimal medium (M9 minimal salts, 1% glucose, 0.1% Casamino Acids, 100 μg of thiamine/ml) agar plates overnight at 28°C were gently harvested with PBS and processed for immunofluorescent staining with the use of an anti-Y. enterocolitica flagellum serum and rhodamine-conjugated secondary anti-rabbit immunoglobulin (IgG) antibodies (Sigma-Aldrich, Munich, Germany). The anti-flagellum antibody was raised against Y. enterocolitica serogroup 0:9 flagella by immunizing a New Zealand rabbit with flagella that were purified from motile bacteria by shearing through vigorous shaking. The presence of flagella in the Yersinia strains was analyzed by immunofluorescence microscopy.

Mouse infection experiments.Virulence of Y. enterocolitica WA-314 and WA-314ΔmotAB was tested in the mouse orogastric infection model as described previously (26). Prior to infection of 6- to 8-week-old C57BL/6 mice, Yersinia frozen stock suspensions (bacteria grown to late exponential phase at 27°C in LB medium) were thawed and washed twice in PBS. After appropriate dilutions, 10⁸ bacteria were fed to groups of five mice with a microliter pipette. At 5 days postinfection, mice were sacrificed and the number of bacteria surviving in the lumina of the small intestine, Peyer's patches, spleens, and livers were determined.

PMF-dependence of Yop secretion.The E. coli flagellar motor is known to require the PMF (24), whereas other flagellar motors, such as that of Vibrio cholerae, are driven by SMF (14). Since Yersinia MotA and MotB exhibit high degrees of homology to Escherichia coli MotA and MotB (73 and 80% identity, respectively), it was expected that the Yersinia flagellar motor would require PMF. If MotAB engagement for Yop delivery would hold, then both motility and Yop delivery should be sensitive to the protonophore CCCP. Therefore, motility of Y. enterocolitica WA-314 was analyzed on floating agar in the absence or presence of 30 μM CCCP. As expected, motility was inhibited by CCCP at 28°C (Fig. 1A, left panel).

• Open in new tab
• Download powerpoint

FIG. 1.

Motility of Y. enterocolitica WA-314 (wild type [WT]) and WA-314ΔmotAB on floating agar. (A) Yersiniae were grown on 0.3% agar (LB medium) with or without 30 μM CCCP and incubated overnight at 28°C. (B) Transcomplementation of WA-314ΔmotAB with plasmid pmotAB and influence on motility.

Next, we wondered whether Yop secretion would also be sensitive to CCCP. Yersiniae were preincubated at 37°C in the presence of Ca²⁺ to induce expression of the TTSS. Then, CCCP was added (concentration range of 0, 1, 5, 10, and 15 μM) and simultaneously secretion of Yops was induced by addition of EGTA, which depletes the medium of Ca²⁺ ions (Fig. 2A). Alternatively, cultures were preincubated with CCCP for 10 min prior to induction of Yop secretion (Fig. 2B). Accumulation of Yops in the culture supernatant was analyzed by SDS-PAGE 2 h after induction of Yop secretion. As can be seen in Fig. 2A, 10 μM CCCP is sufficient to completely abolish Yop secretion immediately. In order to allow a better assessment of the specificity of this phenomenon, we analyzed the influence of CCCP on growth restriction (Fig. 3A), Yop secretion and expression (Fig. 3B), and ATP level (Fig. 3C), in parallel. Yersiniae (six cultures in parallel) were preincubated in the presence of Ca²⁺ as described above. At an OD₆₀₀ of approximately 1.0, CCCP (10 μM) was added to three cultures. DMSO, the solvent for CCCP, was added to the other three cultures at a final concentration of 0.5%. Five minutes later, secretion of Yops was induced in all six cultures by the addition of 5 mM EGTA and 10 mM MgCl₂. Before addition of CCCP-DMSO, 30 and 60 min thereafter, 2-ml aliquots of each culture were removed for analysis of the supernatant for Yop secretion and the pellet for intracellular Yops and ATP levels. Treatment with 10 μM CCCP caused an almost immediate growth arrest (Fig. 3A), accompanied by a complete inhibition of Yop secretion (Fig. 3B, upper panel). The intracellular level of Yops, as exemplified by YopE and YopH (Fig. 3B, lower panel), did not decrease dramatically with CCCP treatment. Thus, the inhibition of Yop secretion cannot be explained by a lack of secretion substrates. Since Yersinia TTSS function requires ATPase activity (31), a lack of ATP after collapse of the proton gradient could account for the abolished secretion. We therefore analyzed intracellular ATP levels by a luciferase assay (Fig. 3C). We found that the ATP levels of yersiniae treated with CCCP were comparable to those of untreated cells within an hour. Hence, CCCP treatment was not accompanied by a sudden decrease in intracellular ATP. We observed a 10 to 15% decrease in ATP levels upon CCCP treatment within minutes when CCCP was added to early- or late-logarithmic cultures (OD₆₀₀ of <0.8 or >1.4; data not shown), indicating a dependence of this reaction on the fitness of the cultures. Vigorous shaking of the cultures was also crucial for maintaining ATP levels of CCCP-treated cells comparable to ATP levels of untreated cells. Altogether, from these experiments we can conclude that Yop secretion requires PMF.

• Open in new tab
• Download powerpoint

FIG. 2.

Y. enterocolitica type III secretion is sensitive to the protonophore CCCP. (A) Yersiniae were grown at 37°C for 2 h. Then, secretion of Yops was started by complexing Ca²⁺ with EGTA; simultaneously, CCCP was added (B) Alternatively, yersiniae were grown at 37°C for 2 h, preincubated with CCCP for 10 min, and subsequently stimulated with EGTA. Secretion was analyzed after 2 h of EGTA stimulation by TCA precipitation of the culture supernatant and subsequent SDS-PAGE (Coomassie staining).

• Open in new tab
• Download powerpoint

• Open in new tab
• Download powerpoint

• Open in new tab
• Download powerpoint

FIG. 3.

Influence of CCCP on Y. enterocolitica growth (A), Yop secretion and expression (B), and ATP level (C). Six Yersinia cultures were grown in parallel for 2 h at 37°C in BHI. CCCP (10 μM) was added to three cultures, and DMSO (solvent for CCCP) was added to the other three cultures as a control. After 5 min, Yop secretion was induced by the addition of 5 mM EGTA and 10 mM MgCl₂. Samples of supernatant and cell pellets were taken at 0, 30, and 60 min relative to the addition of CCCP or DMSO. The OD₆₀₀ of cultures was monitored. (B) Culture samples with OD of >1 were diluted. Supernatant was precipitated with TCA, and equal amounts adjusted by OD (corresponding to approximately 1 ml of 1 OD₆₀₀) were loaded onto an SDS gel and stained with Coomassie (upper panel). One fifth of corresponding cell pellets was loaded onto an SDS gel, electroblotted, and immunostained by using anti-YopE and anti-YopH antisera (lower panel). (C) For determination of intracellular ATP levels, 2 ml of bacterial cultures was centrifuged, and pellets were resuspended in 100 mM Tris-HCl (pH 7.75)-4 mM EDTA with adjustment of the OD₆₀₀ of the cell suspension to 1.0. The cell suspensions (100 μl) were boiled for 2 min at 100°C. Samples were centrifuged, and 50 μl of each supernatant was transferred to a microtiter plate. Luciferase reagent (50 μl) was injected, and luminescence read by a luminometer at 20°C.

It is known that acetate can cross the membrane in its protonated form thereby acidifying the cytoplasm and accordingly reducing ΔpH but not membrane potential Δψ. In addition, it was shown that treatment of E. coli with acetate (34 mM and pH 5) did not dramatically decrease the total PMF but reduced motility effectively (25). We found that Yop secretion was abolished upon pretreatment with 30 mM potassium acetate (pH 5.1) (data not shown). This suggests that it is the ΔpH component of the PMF which is required for Yop secretion.

Motility and Yop secretion of Y. enterocolitica were not affected by substances that inhibit SMF driven motors (data not shown), such as HQNO and phenamil (14).

Yop secretion and mouse virulence are independent of MotAB.To test whether PMF for Yop secretion is deployed via the flagellar motor, we constructed an isogenic deletion mutant of Y. enterocolitica lacking the motor components MotA and MotB. This was accomplished by a one-step inactivation procedure based on homologous recombination between a PCR product and the target gene. Recombination by just 50-nucleotide-homology arms was made possible by expressing λ phage recombinases from a curable plasmid in Yersinia. Successful mutagenesis was asserted by testing motility on floating agar (Fig. 1A, right panel) and by PCR (data not shown). The motAB mutant was successfully complemented in trans by using plasmid pmotAB, a construct based on the moderate-copy-number vector pACYC184 (Fig. 1B). To ensure that the immotile phenotype was not due to a lack of flagella, caused, e.g., by interference of mutagenesis with the flagellar regulatory network, we visualized flagella. Immunofluorescence microscopy of antiserum against flagella revealed that deletion of motAB did not abolish flagellar synthesis (Fig. 4). Thus, assembly of flagella that is based on flagellar TTSS secretion does not depend on the presence of the motor components MotAB. Consistent with this finding, it has been shown that secretion of Yp1A by the Yersinia flagellar TTSS was not affected in a motA mutant (33).

• Open in new tab
• Download powerpoint

FIG. 4.

Immunofluorescence microscopy of yersinia (wild-type strain WA-314 left) and mutant WA-314ΔmotAB (right) flagella. Yersiniae were grown on minimal medium agar plates overnight at 28°C, gently harvested with PBS, and processed for immunofluorescent staining with anti-Y. enterocolitica flagellum serum and rhodamine-conjugated secondary anti-rabbit IgG antibodies.

Next, the influence of the motAB mutation on Yop secretion and translocation, was analyzed. Figure 5 shows that the motAB mutant secreted amounts of Yops to the culture supernatant similar to those of the wild type. The amount of intracellular Yops was also not affected (data not shown). Furthermore, no differences in cytopathic effects on infected HeLa cells or translocation of YopE and YopH into HeLa cells as determined by Western blotting of intracellular fractions could be detected (Fig. 6).

• Open in new tab
• Download powerpoint

FIG. 5.

Yop secretion of the Yersinia ΔmotAB mutant. Secretion of Yops by the wild-type strain and its motAB mutant was analyzed on a Coomassie-stained gel after secretion for 90 min.

• Open in new tab
• Download powerpoint

FIG. 6.

Translocation of Yop effectors to HeLa cells is not affected by the deletion of motAB. The translocation-deficient yscV (lcrD) mutant served as a control. HeLa cells were infected at a multiplicity of infection of 50 for 2 h. Then, cells were detached by treatment with proteinase K, washed, and lysed by osmotic shock. Samples were centrifuged. S refers to supernatant after HeLa cell lysis and subsequent TCA precipitation thus corresponding to the fraction containing translocated Yops. P refers to the pellet fraction after HeLa cell lysis including adherent bacteria and cell debris. Fractions were loaded on an SDS gel, which was electroblotted. YopE and YopH were detected with appropriate antisera.

To rule out that a possible interaction of MotAB with the injectisome could become manifest only in vivo, we orally infected C57BL/6 mice with 10⁸ CFU of the motAB mutant or the wild-type strain. Five days later, mice were sacrificed and colony counts of surviving bacteria in organs were determined. The course of infection with the mutant was progressive, with bacteria colonizing not only the small intestine and Peyer's patches but also the spleen and liver, as is the case for the wild-type strain. No significant difference in bacterial colony counts between the wild type and the mutant could be detected in Peyer's patches (5.42 ± 1.13 and 5.58 ± 0.73 mean log CFU ± standard deviations, respectively) and the spleen (6.43 ± 0.79 and 6.21 ± 1.22 mean log CFU, respectively). In summary, no evidence was found that MotAB function is linked to type III-dependent transport of Yops.