RNA-based Identification of Genes Expressed by Microbial Pathogens in Response to Host Interaction

RNA-based Identification of Genes Expressed by Bacterial Pathogens in Response to Host Interactions

Research on infectious disease microorganisms in the laboratory of James E. Graham
at the University of Louisville School of Medicine

My research here in the Department of Microbiology and Immunology involves the identification of gene expression patterns and corresponding metabolic pathways that allow bacteria to adapt to environments encountered during infection. These environments are not readily simulated in laboratory cultures, as complex combinations of environmental cues and specific interactions with host cells typically signal adaptive responses. To identify gene expression patterns in small numbers of bacteria present during natural infections, I have developed a novel PCR-based RNA analysis method called SCOTS (selective capture of transcribed sequences). This approach has provided insights into several important microbial pathogens, and allowed the first global characterizations of bacterial gene expression patterns in naturally infected human tissues.

Although substantial developments have recently occurred in the determination of numerous complete bacterial genome sequences, analysis of bacterial genome function by direct examination of bacterial RNA is inherently difficult due to the nature of prokaryotic mRNA. Formidable obstacles include short message half-lives, limited polyadenylation, and scarcity of material, particularly for pathogens growing in the natural environments of host cells and tissues. These barriers continue to slow progress in terms of global RNA-based analysis of in vivo gene expression in bacterial pathogens. As a postdoctoral fellow in the laboratory of J. E. Clark-Curtiss at Washington University in St. Louis in 1997, I developed a method for the identification of cDNAs for RNAs produced by bacterial pathogens only during growth in association with the host. This method, which I call SCOTS (selective capture of transcribed sequences), directly identifies bacterial genes specifically expressed in different environments, without specialized genetic techniques, DNA microarrays, libraries, or species-specific cloning vectors, and is applicable to any microorganism from which genomic DNA can be obtained. Unlike other recently described methods (e.g. IVET, DFI, STM) for analysis of in vivo gene expression, SCOTS identifies potentially important genes, rather than promoter regions, and is not confounded by polar effects when genes are arranged in prokaryotic polycistronic operons. SCOTS also has the distinct advantage of being able to determine differential expression from very small numbers of pathogens, allowing for the first time analysis of samples taken from human tissues in natural disease states (see below).

Initial application of SCOTS to the leading killer among infectious microorganisms, Mycobacterium tuberculosis, identified nine genes which appear to activated following phagocytosis by cultured human macrophages (see Graham and Clark-Curtiss, 1999.). In collaboration with several other researchers, SCOTS has also been applied to M. avium, Salmonella typhi and S. typhimurium in vivo gene expression, and to the identification of Listeria monocytogenes RNAs for genes expressed in response to growth at low temperature (Liu, et. al. 2002.).

Recently, I used SCOTS for the global analysis of Helicobacter pylori gene expression in human gastric mucosa. SCOTS analyses of single human tissue biopsies obtained global H. pylori in vivo gene expression profiles from tissue specimens containing fewer than 10,000 bacteria (Graham et al., 2002.). These analyses identified several large putative H. pylori-specific operons that appear to be differentially expressed by bacteria in the human gastric mucosa relative to those grown in broth culture (Graham, 2003). Work is also now continuing on the characterization of the role of previously identified M. tuberculosis genes in pathogenesis, and the global analysis of M. tuberculosis gene expression in murine and human cells and lung tissues. Ongoing collaborative efforts of the laboratory include development of a promising dual antigen subunit tuberculosis vaccine with Haval Shirwan and ApoImmune of Louisville, and animal modeling of tuberculosis infections with Michael Cynamon, and Carolyn Shoen at Upstate Medical Center in Syracuse. Other studies involve gene expression profiling in Listeria and Staphylococci with Brian Wilkinson of Illinois State University and John Gustafson at New Mexico State University. Together with surgeon Sergo Vashakidze in Tbilisi, Georgia we are analyzing M. tuberculosis gene expression during naturally occurring human respiratory infections.

Identification of Unique Sequences in Genomes and Metagenomic DNA Pools by Genomic Fragment Enrichment (GFE)

In 2004 I was contacted by Jorge Santo Domingo of the U.S. Environmental Protection Agency in Cincinnati about the possibility of developing an approach to identifying unique sequences within DNA pools from different complex microbial communities. These discussions, and the expertise of the EPA's Orin Shanks, led to the developement in 2005 of a second nucleic acid sorting method designated 'GFE'. GFE has since been shown to be a simple and efficient approach to a variety of problems, including identification of the sources of fecal contamination in water, and the direct cloning of unique regions of chromosomal DNA from closely related microbial isolates (Shanks et al. 2006a, 2006b, 2007) . This proprietary approach is now being applied to identify differences in the microbial flora in health and disease (with Pascale Alard from the UofL Department of Microbiology), and disturbances in the intestinal flora of cancer prone chemokine receptor knockout mice with Hari Bodduluri of the J. G. Brown Cancer Center.