Our laboratory is dedicated to solving important problems in bacterial mechanotransduction related to pathogenesis. Bacterial pathogenesis involves the programmed changes in gene expression to evade host defense systems through sensing the host environment both physically and chemically. In bacteria, signal transduction occurs largely via two-component regulatory systems. These systems employ an inner membrane histidine kinase (the sensor) and a cytoplasmic response regulator (Fig. 1). Environmental sensing is in most cases directly coupled to regulation of gene expression. The two components communicate via a series of phosphorylation/phosphotransfer reactions. The sensor kinase is autophosphorylated by ATP at a conserved histidine residue upon activation by environmental stress. The sensor kinase then transfers the phosphoryl group to a conserved aspartate on the response regulator. Phosphorylation of the response regulator often increases its affinity for DNA leading to transcriptional activation or repression, depending on the mode of action. The EnvZ/OmpR system that we study responds to osmotic changes and controls the expression of the outer membrane proteins OmpF and OmpC. OmpR is an important global regulator and plays a major role in regulating virulence genes in many pathogens, including Salmonella, Yersinia, Shigella and E. coli.
Fig. 1. Osmoregulation in E. coli. Left. At low osmolality, (Low osm) OmpR~P is low, either because the EnvZ kinase activity is low or the EnvZ phosphatase activity is high and OmpF is the major porin in the outer membrane. Right. At high osmolality, OmpR~P increases (high EnvZ kinase or low EnvZ phosphatase), activating ompC and repressing ompF. The question marks indicate that the signal was unknown, but our recent advances indicate that high osmolality drives a coil to helix transition in the cytoplasmic domain of EnvZ, promoting autophosphorylation (see below). OM = outer membrane, PP = periplasm, IM = inner membrane.
A central question in the EnvZ/OmpR two-component system that we study (and an unanswered question in most other sensor kinases as well) is how does a change in the osmolality of the external environment (the signal) result in the differential regulation of outer membrane porin gene expression (the output)?
The Mechanism of Osmosensing by EnvZ and other HKs (Jeremy Wang, Soumya Ranganathan)
We used amide hydrogen/deuterium exchange mass spectrometry in collaboration with Dr. Ganesh Anand in the Department of Biological Sciences at NUS to map conformational changes in EnvZ associated with osmotic stress. We discovered that the osmosensing apparatus is localized to a four-helix bundle subdomain (Wang et al., 2012). Specifically, a helical segment containing the EnvZ auto-phosphorylation site (His243) exhibits decreased exchange in response to increasing osmolality (NaCl, sucrose), while an adjacent helical peptide exists in multiple conformations in solution. ATP binding to the kinase subdomain is independent and uncoupled from osmosensing. Surprisingly, the inner membrane protein EnvZ does not need to be in the membrane for osmosensing. Our results support a model where osmolytes promote intrahelical H-bonding that enhances helix stabilization and increase the rate of autophosphorylation (Fig. 2). This model combines coupling of osmolyte-induced helix stabilization with downstream signaling and likely provides a conserved mechanism for signaling proteins that respond to diverse physical and mechanical stimuli. Further tests of the model are underway in our laboratory.
Fig. 2. Coil to helix transcription is the mechanism of osmosensing by the EnvZ histidine kinase. At low osmolality, the histidine-containing helix (shown with the His side-chain sticking out) is slightly unfolded and histidine autophosphorylation is low. At high osmolality, this peptide shows increased hydrogen bonding leading to helix stabilization and increased autophosphorylation.
Porin Gene Transcription in Single Cells (Dr. Yunfeng Gao)
In order to understand how the increased phosphorylation of EnvZ causes increased expression of OmpF and OmpC, we are analyzing ompF- or ompC- promoter fusions to the green fluorescent protein gene in single cells. The transcription of GFP from the fusions will enable us to see how expression levels change in response to osmolality. Our results indicate that the noise is intrinsic to the promoters and not dependent on copy number or location. Identical measurements in an flhDC null strain identify the master flagellar regulator FlhDC as an indirect regulator of ompC. This connection between FlhDC and the EnvZ/OmpR system has been noted by others studying mouse models of infection (Leatham et al., 2005, De Paepe et al., 2011). We are working to further understand this connection.
Fig. 3. MG1655 cells that contain an ompC-gfp promoter fusion exhibit varying degrees of fluorescence in response to osmotic stress. Cells were grown overnight in LB and then sub-cultured in A media + 400 mM NaCl. The cells were placed on agar pads and their fluorescence was monitored over time in a Deltavision deconvolution microscope.
Measurements of Intracellular pH in E. coli and S. Typhimurium (Dr. Smarajit Chakraborty)
Both pH and osmolality have been implicated in the activation of pathogenesis, but it is difficult to know what the pH is within bacteria after invading host cells. A recently developed DNA nanomachine (I-switch) can measure pH changes using fluorescence resonance energy transfer (Krishnan & Simmel, 2011). It enables spatiotemporal measurement of intrabacterial pH. We have adapted this approach to a prokaryotic system to measure intracellular pH under different physiological conditions in bacteria. Osmotic stress in bacteria can also alter the intracellular pH, which in turn, can regulate expression of virulence genes. We are using the I-switch to map the spatiotemporal pH changes in E. coli during osmotic signaling. Similarly, we are using this technology to measure pH changes in Salmonella enterica serovar Typhimurium, where acidic pH is known to induce expression of virulence genes on Pathogenicity Island 2. Our long-term goal is to measure the internal pH of Salmonella in the macrophage vacuole during systemic infection.
Silencing and anti-silencing in Salmonella pathogenesis (Drs. Priyanka Das, Stuti Desai)
There are many important questions related to the control of gene expression, since the molecular mechanisms are not well understood. When Salmonella resides in a macrophage vacuole, the EnvZ/OmpR two-component system activates transcription of the SsrA/B system located on pathogenicity island 2 (SPI-2), in response to acid stress (Fig. 4). The response regulator SsrB activates expression of genes encoded within and outside of SPI-2, which are required for systemic infection. SsrB binds upstream of the sifA, sifB, and sseJ effector genes and directly regulates transcription (Walthers et al., 2011). SsrB also relieves gene silencing by the nucleoid protein H-NS. Single molecule experiments with magnetic tweezers demonstrated that H-NS has two binding modes that are influenced by Mg2+ or other divalent cations (Liu et al., 2010). SsrB displaces H-NS from DNA only when it is bound in a polymerization (stiffening) mode and not when H-NS is bound to DNA in the bridging mode (Walthers et al., 2011). Thus, in contrast to previous views, the polymerization-binding mode of H-NS is the relevant form for counter-silencing by SsrB (Fig. 5). Our results reveal that response regulators can directly activate transcription and also relieve H-NS silencing. These experiments are part of an ongoing collaboration with Dr. Jie Yan in the Department of Physics at NUS. We have recently characterized the binding properties of the H-NS homologue StpA (Lim et al., 2012b) as well as H-NS mutants that are unable to polymerize (Lim et al., 2012a).
Fig. 4. A summary of our current view of OmpR/EnvZ regulation of ssrA/B. EnvZ phosphorylates OmpR (step 1) in response to an unknown stress (acid pH?). OmpR~P binds with high affinity upstream of ssrA and activates transcription (step 2). SsrA phosphorylates SsrB (step 3) and SsrB~P activates transcription of genes both within and outside of SPI-2 (step 4). OmpR also regulates additional genes such as sseI (Feng et al., 2004) and sifA (Pickard et al., 1994).
Fig. 5. AFM image of H-NS bound to DNA from (Liu et al., 2010). At low Mg2+ concentration, H-NS binds to DNA in a polymerization mode, which leads to stiffening and elongation. This is the form that is de-repressed by SsrB, leading to transcriptional activation (Liu et al., 2010, Walthers et al., 2011).
Screening a Small Molecule Library for Anti-silencing Inhibitors (Dr. Smarajit Chakraborty)
There are a number of clinically relevant bacteria that generally are not treated with antibiotics, such as Salmonella enterica serovar Typhimurium. It has the ability to form tubular structures known as Salmonella–induced filaments (Sifs) for rapid replication in host cells. SifA is an important effector protein secreted by the Salmonella pathogenicity island 2-encoded type III secretion system that is responsible for the maintenance of the vacuolar membrane enclosing the pathogen. sifA mutant strains show high attenuation for systemic infection in mice with a dramatic replication defect. We are screening a chemical library (in collaboration with Young-tae Chang in the Department of Chemistry at NUS) that can prevent SifA expression, rendering bacteria highly attenuated. Two compounds with strong inhibitory activity are now being characterized.
Super-resolution microscopy of EnvZ, OmpR, H-NS and chromosomal imaging (Drs. Hong Fang Zheng, Stuti Desai)
We have constructed photoactivatable fusions to EnvZ, OmpR and H-NS and are now imaging these proteins in various strain backgrounds and in response to numerous environmental stresses, including high osmolality and acid pH. We have also been examining OmpR/DNA interactions using PALM.
Fig. 7. STORM image of pSL5 (EnvZ-PAmCherry) in E.coli MG1655. Cells were grown in A medium with 1 mM IPTG induction for 8 h. Scale bar is 2 μm.
Structural Analysis of SPI-2 needles (Dr. Hideaki Mizusaki)
Salmonella Pathogenicity Island-2 (SPI-2) plays a central role in the intracellular survival of Salmonella by encoding a type III secretion system, whose structure is currently unknown. SPI-2 gene expression is controlled by the two-component regulatory system SsrA/B, whose expression is in turn controlled by additional regulatory networks. The type III secretion system that is encoded by SPI-2 has not been well-studied, in part because it is not abundant and also because other surface appendages such as pili and flagellae obscure its visualization. Yet without SPI-2 type III secretion, Salmonella is unable to launch a systemic infection and becomes avirulent. We are using cryoelectron microscopy (cryoEM) in collaboration with Dr. Wah Chiu (Baylor College of Medicine) to perform a structural analysis of the macromolecular complex of the SPI-2 secretory apparatus in order to fully study structural and assembly aspects of SPI-2 secretion.
Fig. 8. Phase contrast and fluorescence images of SsaC-GFP fusion strain, S. Typhimurium (14028s). The outer ring protein SsaC-GFP fusion was expressed from an arabinose-induced plasmid (pMPM-K6Ω) in a ΔssaC background. The strain was grown in PCN (pH 5.8) at 37˚C for 4h. 2 μm scale bars.