Amy Decatur
Assistant Professor, Dept of Microbiology

Microbiology, Virology and Parasitology Program

Address

424 Johnson Pavilion
3610 Hamilton Walk
Philadelphia, PA 19104

Office tel.: 215 898-2392
Lab tel.: 215 898-2274
Fax: 215 898-9557
E-mail: decatura@mail.med.upenn.edu

Link(s)

Dr. Decatur's Microbiology page

Education

Vanderbilt University: BS (Chemistry and Molecular Biology), 1989.

Harvard University: PhD (Molecular and Cellular Biology), 1997.

Research Interests

• Host-pathogen interactions.

Key words: Bacterial pathogenesis, Listeria.

Search PubMed for articles

Description of Research

Establishing and maintaining a protected niche within an infected host cell is critical to the success of an intracellular pathogen. In the case of the Gram-positive bacterial pathogen Listeria monocytogenes, this protected niche is the host cell cytosol. Using L. monocytogenes as a model system, our lab is interested in determining how an intracellular pathogen interacts with its host cell to set up residence. In particular, we are exploring whether bacterial pathogens can co-opt host cell compartment-specific proteolysis to control when and where a given virulence factor acts. L. monocytogenes is an excellent model system for this study due to its rapid doubling time, genetic tractability, and the existence of both tissue culture and murine models of infection. In addition, because L. monocytogenes is a facultative pathogen, avirulent mutants are relatively easy to isolate and characterize.

Our research is focused on the secreted, bacterial pore-forming protein listeriolysin O (LLO). LLO is an essential virulence factor that allows L. monocytogenes to escape from the host cell vacuole and reach the host cell cytosol. Bacteria that are unable to produce LLO remain trapped in the host vacuole and are five orders of magnitude less virulent than wildtype bacteria in vivo. Interestingly, LLO is produced by bacteria in both the vacuole and cytosol. Yet, in the course of a normal infection, the integrity of the host plasma membrane is not compromised despite the fact that, like the vacuole, the plasma membrane contains the binding site (cholesterol) for LLO. Thus, the pore-forming activity of LLO must be compartmentalized within the host cell.

We have identified a 26 amino acid, cis-acting sequence in the amino terminus of LLO that is necessary to restrict LLO activity to the vacuole. Bacteria producing a mutant LLO protein lacking this sequence are able to escape from host vacuoles as efficiently as wild type bacteria, but subsequently permeabilize the host plasma membrane and kill the host cell. Moreover, these mutant bacteria are four orders of magnitude less virulent than wild type bacteria in vivo indicating that preservation of the bacterium's cytosolic niche is essential for the pathogenicity of L. monocytogenes. Current projects in the lab are aimed at understanding the mechanism by which this sequence restricts LLO's pore-forming activity to the host cell vacuole.

Recent Publications

Decatur, A. L. and D. A. Portnoy. (2000). A PEST-like sequence in Listeriolysin O essential for Listeria monocytogenes pathogenicity. Science 290:992-995.

Lab

Rotation Projects for 2006-2007

1. The 26 amino acid sequence described above is rich in the residues proline, glutamate, serine, and threonine and resembles a PEST sequence. PEST sequences are thought to target eukaryotic proteins for degradation. This observation has led to the hypothesis that LLO's PEST-like sequence targets this potentially toxic protein for degradation, and thus inactivation, within the host cell cytosol. In support of this model, LLO lacking the PEST-like sequence (LLOdelta26) accumulated to significantly higher levels than the wild type protein in the host cell cytosol, but not in the supernatant of bacterial broth cultures. We are currently testing whether LLO's PEST sequence acts as a degradation tag within the mammalian cytosol.
2. PEST sequences often contain recognition sites for kinases and phosphorylation at these sites often precedes degradation. We have shown that elimination of three consensus mitogen activated kinase (MAPK) sites (PXS/TP) within the PEST-like region of LLO results in a mutant protein (LLOS44A, S48A, T51A) that, similar to LLOdelta26, is toxic to host cells in vitro and 100-fold less virulent in vivo. Using a combination of genetic and biochemical approaches, we are determining whether these sites represent true sites of phosphorylation, and if so, how phosphorylation contributes to regulation of LLO activity.
3. We are also identifying residues within the PEST-like sequence of LLO that are important for its function using random, localized mutagenesis and screening for dominant negative mutants. In this way, we are taking advantage of our finding that LLO lacking the PEST sequence is dominant to the wild type protein with regard to host.

Lab personnel:

Rob Stevens, Ttechnician
Irene Godoy, Undergraduate Student