Surface sensing in Vibrio parahaemolyticus triggers a programme of gene expression that promotes colonization and virulence

Abstract

Vibrio parahaemolyticus senses surfaces via impeded rotation of its polar flagellum. We have exploited this surface-sensing mechanism to trick the organism into thinking it is on a surface when it is growing in liquid. This facilitated studies of global gene expression in a way that avoided many of the complications of surface-to-liquid comparisons, and illuminated ∼ 70 genes that respond to surface sensing per se. Almost all are surface-induced (not repressed) and encode swarming motility proteins, virulence factors or sensory enzymes involved with chemoreception and c-di-GMP signalling. Follow-up studies were performed to place the surface-responsive genes in a regulatory hierarchy. Mapping the hierarchy revealed two surprises about LafK, a transcriptional activator that until now has been considered to be the master regulator for the lateral flagellar system. First, LafK controls a more diverse set of genes than previously appreciated. Second, some laf genes are not under LafK control, which means LafK is not the master regulator after all. Additional experiments motivated by the transcriptome analyses revealed that growth on a surface lowers c-di-GMP levels and enhances cytotoxicity. Thus, we demonstrate that V. parahaemolyticus can invoke a programme of gene control upon encountering a surface and the specific identities of the surface-responsive genes are pertinent to colonization and pathogenesis.

FIGURE 1. Swimmer and swarmer cells and the lateral flagellar system of V. parahaemolyticus

LM5674 (wild type) swimmer cell grown in liquid (A) and swarmer cell grown on a surface (B) are profoundly different. Cells were fixed and examined by immunofluorescence microscopy. Both panels are of the same magnification and the bar indicates 5 μm. Cells were stained with membrane dye FM 4–64 (colored red) and anti-polar flagellin antiserum (A, colored green) or anti-lateral flagellin antiserum (B, colored green). C. Schematic of the flagellar organelle and the hierarchy of lateral flagellar (laf) gene expression. Previous work established that LafK, a σ⁵⁴-dependent regulator was required for expression of Class 2 genes, which mostly encode the components of the basal body and hook as well as a specialized sigma factor. This sigma (FliA, or σ²⁸) is required for expression of Class 3 genes, which encode the flagellin subunit and the haps, which are the adaptors for joining the propeller to the hook. The placement of the fliM operon, which encodes components of the C-ring and membrane-associated export components, within the regulatory cascade was unknown at the outset of this work.

FIGURE 2. Conditions inducing the swarmer cell program

The luminescent reporter strain LM5738 carrying a lux fusion in the laf hook operon was used as a guide for establishing the microarray growth conditions. A. Luminescence and growth of laf::lux reporter strain LM5738 during growth on a surface. Cells were spread on plates and harvested periodically. Luminescence is reported as specific light units (SLU), which are total light units per min per ml per OD_600nm. B. Growth of the laf::lux reporter strain during growth in liquid (L), liquid dipyridyl (LD, iron-limiting broth), and liquid dipyridyl phenamil (PS2, pseudosurface 2). C. Luminescence of the laf::lux reporter strain during growth in L, LD, and PS2. D. Immunoblot of samples harvested for microarray analyses. Samples: wild type LM5674 grown in liquid (L) and on a surface (S), the lafK regulatory mutant LM7789 grown on a surface [S(K−)], the wild type grown in iron-limiting broth (LD) and grown in iron-limiting broth and phenamil (PS2), and the polar flagellar mutant LM5392 grown in iron-limiting broth (PS1). This polar flagellar mutant makes no polar flagellins. The immunoblot was probed with antisera directed against the lateral flagellin subunit (Laf) and the constitutively produced polar flagellins (Fla).

FIGURE 3. Phenotypes and reporter gene expression in strains with mutations in the surface-responsive program

A. Swarming phenotypes of strains: LM5431 (wild-type swarming strain), LM5738 (flgB_L::Tn5lux), LM9889 (vpa0260::Tn5), LM9890 (vpa0275::Tn5), LM6159 (vpa1598::Tn5lux), LM6161 (vp1002::Tn5lux), LM9376 (vpa0227::Tn5lux), and LM5093 (wild-type swarming strain). Strains were inoculated on HI swarm agar and incubated for 12 h at room temperature. B. Luminescence of strain LM9376 carrying the vpa0227::lux reporter in the gene encoding the alkaline serine protease and its congenic derivative LM9620, which has a ΔlafK mutation. Strains were grown as per the microarray conditions and harvested periodically for OD_600nm and luminescence measurements. C. Luminescence of strains with reporter lux fusions in surface-regulated genes (VPA0227, VP1002, and VPA1598). The control strain LM8858 contains a fusion in a constitutively expressed gene (VP0575). Strains were grown as per the microarray conditions and harvested periodically for light measurements. Maximum achieved luminescence is reported as normalized specific light units (SLU). Error bars represent standard deviation of triplicate measurements. Each experiment has been repeated at least three times with similar result.

FIGURE 4. RTPCR analyzing surface-responsive gene expression

RNA was prepared from the following wild-type and mutant strains grown in liquid (L) or on plates (S) and harvested according to the microarray conditions: LM5674 (WT), LM7789 (lafK mutant), and LM6210 (σ²⁸ defective mutant). In each RTPCR reaction primers were included to amplify the target gene and a constitutively expressed control gene (designated with *), which was the polar flagellar σ²⁸ (VP2232). Patterns of gene expression are divided into three classes as indicated. The genes examined (and gene product): VPA1538 (LafK), VPA1540 (FliM), VPA0459 (Collagenase), VPA1598 (GlcNAc binding protein), VP2370 (hypothetical), VP1482 (c-di-GMP signaling), VPA1550 (FliD), VPA1548 (LafA), VPA1492 (Mcp), VP1293-94 (small RNA), VPA1294 (SPOR protein), and VPA1649 (Metalloprotease). Primers and sizes of products are given in Table S5. [Note: there is no RTPCR product for lafK mRNA in the lafK strain because it is a deletion mutant.] The RTPCR reaction products are shown organized in the hierarchical classes of gene control determined in part by these analyses.

FIGURE 5. Probing the lateral hierarchy of expression by using mutants and reporter fusions

The wild-type and indicated lateral flagellar mutant strains carried reporter plasmids with lafK::lacZ or flgC_L::lacZ fusions. Strains were grown on plates and harvested periodically to measure β-galactosidase. The wild-type strain carrying each reporter was also grown as indicated in liquid (L). Activity is reported as maximal activity during growth (which occurred at ~ 10 h on plates) and is normalized to the percentage of maximal reporter activity in the wild-type strain, which was ~1500 Miller units for lafK and ~300 Miller units for flgC. These experiments were replicated at least three times with similar result and error bars are the standard deviation of triplicate assays in a single representative experiment. For all comparisons, the LacZ activity in the mutant strains was significantly different from the activity in the wild type strain (p values <0.004).

FIGURE 6. Profiles of surface-responsive gene expression

The normalized expression values (log2) for selected genes as captured by the microarray analysis are plotted for each microarray condition. The profiles of gene expression represent three differentially expressed classes and one control class: A. Surface-induced, LafK-dependent; B, Surface-induced, LafK-independent; C, Surface-repressed, LafK-independent; and D, Representative control genes showing constitutive expression over these conditions. Error bars represent standard error of replicates.

Figure 7. Intracellular concentration of c-di-GMP

The wild-type strain LM5674 was grown in liquid (L) and on a surface (S) as per the microarray condition and harvested to determine the concentration of c-di-GMP. Mutant strain LM6567 (ΔscrABC; Scr−) was also extracted after growth on a surface. The average concentrations of two replicates calculated as pmol c-di-GMP per OD_600nm unit were: L, 3.09 (± 0.2); S, 1.48 (± 0.02); and S(Scr−), 2.03 (± 0.04), with the standard deviation given in parentheses. For the graph, these numbers were converted to pmole per mg bacterial protein using an experimentally derived conversion factor for each strain grown under the particular condition. The amount of c-di-GMP measured for surface-grown wild type (S) was statistically different compared to liquid-grown wild type (L) or to the surface-grown scrABC mutant [S(Scr−)], specifically p values < 0.018 and < 0.002, respectively, determined by using Student’s t test.

FIGURE 8. Growth on a surface enhances collagenase activity and cytotoxicity towards host cells in culture

A. Collagenase activity of wild-type and mutant strains. Strains were grown in liquid as per the indicated microarray condition, harvested at similar OD_600nm, centrifuged, and the supernatants were used in a fluorescence-based collagenase assay. Activity is reported as percent activity normalized to the activity of the wild-type strain grown in liquid, which was ~280 units. Purified Clostridium histolyticum Type IV collagenase (0.1 U) produced ~7000 fluorescence units in this assay. The reaction time was 120 min. Growth conditions and strains: The wild-type strain LM5674 in L, LD, and PS2; the Fla⁻ strain LM5392 in LD (PS1); and the lafK mutant LM7789 in PS2 (PS2 (K−)]. The experiments were repeated at least three times with similar results. A representative experiment is shown and error bars represent standard deviation of triplicate measurements. B. The wild-type strain was grown as per microarray conditions (S, PS2, and L) and used to infect Chinese hamster ovary (CHO) host cells. Cytotoxicity was assayed by measuring release of host cell lactate dehydrogenase over time; it is expressed as percentage of maximum WT cytotoxicity for the experiment. Maximal WT cytotoxicity was usually ~70 % of total lysis achieved with detergent Triton X-100. Multiplicity of infection was ~15. Error bars indicate standard error of the mean of 6 individual samples at each time point for a representative experiment. P values at 3, 4, and 5 h were less than 0.004 for all comparisons (i.e., S vs L, PS2 vs L, and S vs PS2). C. The wild-type and lafK strains were grown and harvested on a surface as per microarray growth conditions, used to infect host cells, and assayed as described above. The p values for the lafK mutant compared to the wild type at 3, 4 and 5 h were 0.004, 0.02, and 0.04, respectively. Each of these experiments has been repeated at least three times with similar result.

FIGURE 9. The surface sensing program of gene control in V. parahaemolyticus includes motility, virulence and sensory enzymes

A core set of ~70 genes is regulated in response to surface sensing mediated by the polar flagellar organelle. Our microarray studies coupled with RTPCR and reporter gene analysis in this and also prior work enables the construction of a working scheme of gene control organized in three hierarchical levels. Class 1 expression is LafK-independent and includes the lafK operon. Other Class 1 genes comprise the lateral fliM operon, as well as non-flagellar genes, including genes pertinent to c-di-GMP signaling, colonization (N-acetyl glucosamine binding protein), and virulence (collagenase). Although the majority of the Class 1 genes are induced by growth on a surface and pseudosurface conditions, some were surface-repressed (distinguished by the dark box). Class 2 expression is promoted by the transcription factor LafK and σ⁵⁴. LafK directs expression of the flagellar genes needed primarily for assembly of the hook basal body structure and upregulates its own operon, which includes motor, MS-ring and the flagellar ATPase genes. LafK also modulates transcription of type three secretion genes (T3SS1) and VPA0227 encoding an alkaline serine protease; however these genes do not absolutely require LafK for transcription (and so are placed a Class 2 box distinct from the lateral flagellar operons). LafK directs expression of fliA_L encoding the specialized laf σ²⁸, which in turn promotes expression of theClass 3 genes including those encoding LafA flagellin, a chemoreceptor, metalloendoprotease, CcmA, a peptidoglycan-binding protein (SPOR), and a small RNA.