Recently, we have discovered that Porphyromonas. gingivalis, a periopathogen, alters its LPS lipid A composition in response to the hemin concentration in the growth medium. We have obtained P. gingivalis LPS preparations that are enriched in the different lipid A mass ions found at low and high hemin concentrations and have shown that both of these P. gingivalis LPS preparations interact with TLR4. However, they have opposing effects on E selectin expression from human endothelial cells. In this R21 exploratory proposal we propose to further explore the possible differential gene modulation programs by these two different P. gingivalis LPS species. Three different LPS preparations, Pgips, containing all the known lipid A species, Pg^s/uso, enriched in the lipid A species found at high hemin concentrations, and iego, enriched in the lipid A species found at low hemin concentrations wil| be added to human endothelial cells and a microarray analysis will be performed. In addition, two different preparations of Escherichia coli LPS, one wild type (ECwt)and the other a significantly less potent form obtained from an msbB mutant (EcmSbB) will be examined and serve as controls. Affymetrix full genome chips which contain more than 47,000 transcripts representing all human genes will be utilized. This approach will allow us to test our hypothesis that: P. gingivalis selectively modifies its lipid A composition to alter endothelial cell function when it's in a blood enriched environment. Alterations in endothelial cell function may contribute to vasculature ulceration in periodontal pockets and/or may facilitate systemic dissemination of the organism. In collaboration with the Center for Expression Arrays at the University of Washington the endothelial cell gene modulation programs will be elucidated for each of these LPS preparations. After triplicate analysis and statistical determinations, gene expression profiles will be analyzed for functional and pathway groups utilizing a combination of GoMiner, GenMAPP and external databases to functionally categorize the differences and similarities among the LPS preparations by cluster analysis. Based on these parameters select genes will be confirmed by Real Time PCR. In addition, select genes will be identified for their dependence upon TLR4 for activation. This information will determine if different LPS species elicit different endothelial cell activation programs and will form a data set to test hypothesis concerning P. gingivalis interactions with TLR4 or other novel LPS receptors. Furthermore, the microarray analysis results will be made public by deposition at the European Bioinformatics Institute Array Express database. ? ?
