Our Scientists

Bacterial pathogens that infect mucosal surfaces in the respiratory and intestinal tracts are the leading causes of disease and death worldwide. Our laboratory investigates bacterial determinants that are important in the transmission and virulence of mucosal pathogens and how expression of these factors is regulated. For facultative pathogens, i.e., those that spend portions of their life cycles both inside and outside of the human body, major shifts in gene expression occur upon entry into and exit from the human host. We are using Vibrio cholerae, the causative agent of the severe diarrheal disease cholera, as a model facultative pathogen. We are investigating the spatial and temporal patterns of gene expression upon entry into and exit from the intestinal tract, and the factors important in transmission during cholera outbreaks.

Human commensal bacteria are also a frequent source of disease in both immunocompromised and healthy people. When commensal bacteria invade protected mucosal surfaces such as in the lung, they too undergo a shift in gene expression that contributes to disease. We are studying Streptococcus pneumoniae as a model commensal pathogen. S. pneumoniae normally resides in the nasopharynx of healthy people but occasionally invades the lung or bloodstream, leading to pneumonia, bacteremia, or meningitis. Understanding these shifts in gene expression and how they are orchestrated, as well as deciphering the roles of the induced virulence factors, will lead to a better understanding of transmission and disease by V. cholerae and S. pneumoniae and will aid in the design of improved diagnostics, therapeutics, and vaccines.

The Shift in Gene Expression when V. cholerae Enters the Host Intestine
Bacteria are extremely efficient organisms and generally do not express genes unnecessarily. This makes the discovery and study of virulence factors challenging, since many such factors are only expressed during infection of the host. We have developed and refined genetic tools to identify and study bacterial virulence factors in the context of the host. One such tool is recombination-based in vivo expression technology (RIVET). This method uses a site-specific DNA recombinase as a reporter gene. When expressed, the recombinase mediates the excision and loss of a selectable marker from the genome. In this way, gene induction is coupled to a heritable phenotypic change in the bacteria, allowing us to know if, when, and where a gene is turned on during infection.

Using RIVET both in a mouse model of cholera and in experimentally infected human volunteers, we identified numerous genes that are turned on when V. cholerae enters the host intestine, and these genes have provided insight into important host-pathogen interactions. The products encoded by these genes play diverse roles in nutrient acquisition, motility, gene regulation, and resistance to chemical stresses. We have also used RIVET to investigate the expression of known virulence genes, including those for cholera toxin (CT) and the toxin-coregulated pilus (TCP). TCP and CT are critical virulence factors required for colonization of the intestinal wall and for eliciting profuse watery diarrhea. Although CT and TCP are coexpressed during in vitro growth under conditions that mimic the host, we found that during infection TCP is expressed several hours earlier than CT. In addition, we discovered that TCP-mediated colonization is required for the later expression of CT. These and other data show that V. cholerae expresses virulence factors in an orchestrated manner, with specific times of induction and functional interdependencies that could never be observed in vitro.

We investigated the requirements for the timing of CT induction during infection and identified a signal transduction phosphorelay system composed of three proteins—VieS, VieA, and VieB—that is necessary for correct expression. VieS is a membrane-localized sensor kinase that detects certain amino acids in the small intestine. Upon binding these amino acids, VieS activates VieA by phosphorylation. VieA is a response regulator that uses its DNA-binding domain to amplify its own expression. Importantly, VieA harbors an independently functioning phosphodiesterase domain that cleaves a cyclic dinucleotide called cyclic di-guanylate (c-di-GMP). By lowering the cellular concentration of c-di-GMP, VieS and VieA turn off environmental survival genes (specifically biofilm formation genes) while simultaneously turning on virulence genes. Because amino acid stimulation of VieS results in a reduced c-di-GMP concentration, amino acids are the first messengers and c-di-GMP is the second messenger in this signal transduction pathway. Among the many unanswered questions associated with this model are these: What factors in V. cholerae sense c-di-GMP levels and impart changes in gene expression? Why are certain free amino acids a specific host signal for the bacteria? How does VieB, which imposes a shut-off signal for VieS, function mechanistically?

The Shift in Gene Expression when V. cholerae Exits the Host
Our observation that genes involved in resistance to chemical stress are turned on during infection led us to hypothesize that a preinduction of such genes might increase the infectivity of V. cholerae. Indeed, we found that such a preinduction increases infectivity by an order of magnitude; i.e., 10-fold fewer bacteria need be ingested to cause a full-blown infection. In an independent line of investigation, we found that V. cholerae shed by cholera patients in their watery stools are also more infectious, having a similar 10-fold reduction in infectious dose. To test the hypothesis that this heightened state of infectivity may have evolved to aid in transmission of cholera during outbreaks, we are investigating the properties of stool V. cholerae that lead to increased infectivity. One clue, which came from our determining the transcriptome of stool V. cholerae using DNA microarrays, is that chemotaxis in the shed bacteria is temporarily altered in a way that increases infectivity. We have confirmed this model, and we are investigating additional factors that contribute to transmission of cholera.

To learn more about transmission and dissemination of V. cholerae, we developed a genetic selection to identify genes that are expressed at the late stage of infection just prior to exit from the host. We identified several genes with specific roles in survival in the watery diarrhea and/or in the aquatic environment once shed. These include genes involved in nutritional and chemical stress tolerance as well as genes needed for growth on chitin, which is the major carbon source in aquatic environments. These data show that V. cholerae has evolved the trick of preinducing genes needed for life in the environment prior to exit from the host. We continue to investigate the role of these genes in dissemination and transmission.

Identification of New Virulence Factors of S. pneumoniae
In a separate project, we are addressing basic questions concerning the virulence properties of S. pneumoniae. Using a genetic screen, we identified more than 200 genes that are required for infection of the lung. Seven of these genes are contiguous in the genome and form a "pathogenicity islet," a mobile cluster of genes endowing bacteria with a particular virulence trait. In this case, the trait is the formation of a multiprotein pilus that is covalently attached to the bacterial surface. We later found that this pilus plays an important role in biofilm formation on mucosal surfaces. We dissected the regulation of the genes in this islet and identified a cis-encoded transcriptional regulator (RlrA), as well as an unlinked regulator (MgrA). MgrA controls the expression of rlrA in response to unknown signals. Recent findings suggest that uptake and metabolism of host sugars by S. pneumoniae may serve as a signal to regulate the activity of MgrA. This prompted us to study the mechanisms and regulation of sugar metabolism in S. pneumoniae, and this ongoing work has established a strong connection between sugar metabolism, gene regulation, and virulence.

Sugar metabolism in bacteria is complex, particularly the process of choosing which sugar to metabolize when multiple sugar types are available, such as on mucosal surfaces. We identified one regulator of sugar metabolism (CcpA), which is needed for virulence in mouse models of infection. However, our results indicated that there are additional regulators that work in conjunction with CcpA in a partially redundant manner to regulate sugar metabolism. To identify these additional regulators and explore the link between sugar metabolism and virulence in greater depth, we screened for S. pneumoniae genes that have synthetic genetic interactions with ccpA. To identify genetic interactions in a high-throughput, genome-wide manner, we invented Tn-seq. Using this method, we identified several previously unknown transcriptional regulators and numerous sugar hydrolases and transporters that play important roles in sugar metabolism. We are applying Tn-seq to identify genes important for other facets of virulence and transmission of both S. pneumoniae and V. cholerae.

Grants from the National Institutes of Health, the Pew Charitable Trusts, and the Bill and Melinda Gates Foundation provided partial support for these projects.

As of March 19, 2012