Rapid emergence of drug resistant infectious parasites represents an ever increasing global health burden. Malaria, caused by protozoa of the genus Plasmodia, is characterized as endemic in 101 countries and annually results in 300-500 million clinical cases with over 2 million deaths. Resistance to common anti-malarial drugs is reported globally, and is cited as the major contributing factor to the growing global malaria problem. Addressing this threat in 1970's, Walter-Reed Army Institute of Research (WRAIR) developed a 4-quinoline methanol compound named Mefloquine (trade name LariamTM), a potent, long lasting blood schizontocide, effective against drug-resistant Plasmodium falciparum [1]. Mefloquine (MFQ) is administered to individuals traveling to malaria endemic areas, and is useful for both therapy and chemoprophylaxis. However, MFQ is known to cause adverse neurological or psychiatric symptoms in as many as 25% percent of individuals taking MFQ at prophylactic doses and 70% of individuals taking MFQ at treatment doses. Adverse neurological effects include ataxia and mood changes [2], thereby offsetting the positive prophylactic aspects of MFQ. Recent research strongly indicates that neurotoxicity associated with MFQ can be attributed to specific single nucleotide polymorphisms (SNPs) [3]. Identification of a critical set of predictive SNPs could form the basis of a clinical test to enable identification of at-risk individuals. Therefore, the objective of the current work is to develop and apply a novel systems biology-centered approach to construct a pathway model of cellular response to mefloquine, and systematically identify a critical subset of species that could serve as the basis for a genetic screen. Our primary hypothesis is that there exists a small subset of cellular components whose action is critical in determining the cellular response to MFQ. Small perturbations caused by genetic variation will alter the function of these critical species, and result in a change in the cellular response, ultimately manifesting itself as an altered observable physiological response. Rational (in contrast with trial-and-error) identification of optimal intracellular candidates for a robust genetic screen, therefore, demands the development of a novel pathway-centric approach. Specific project objectives include · Quantify Transcriptional Response of primary neurons exposed to MFQ. · Construct an Intracellular Pathway Model of cellular response to MFQ utilizing microarray data, relevant databases, scientific literature, and expert guidance. · Generate a Prioritized List of Cellular Processes associated with MFQ exposure using novel computational analysis of the intracellular pathway model. · Verify and Validate the intracellular pathway model utilizing gene suppression/expression techniques in order to quantify the impact of genes identified as critical to cellular response to MFQ. Transcriptional response of rat primary cortical neurons exposed to 10 microM MFQ for 3 or 12 hours was measured using the Applied Biosystems Rat Genome Survey Array. Two additional sets of data for primary rat forebrain neuronal response to 80 microM MFQ exposure were measured using the Affymetrix Rat Genome U34A array [4,5] and graciously provided to CFDRC by WRAIR through a CRADA/MTA agreement. Significantly expressed genes were identified from both sets of data using the Significance Analysis of Microarray (SAM) technique [6]. The transcriptional profiles indicated increased expression of genes associated with the disruption of Ca2+ homeostasis and endoplasmic reticulum stress, in addition to genes associated with apoptosis mediated by endoplasmic reticulum stress. Significant genes were imported into the Database for Annotation, Visualization and Integrated Discovery (DAVID) tool [7], which incorporates ontological information and KEGG canonical pathway databases to simultaneously identify and visualize statistically significant canonical pathways from a the list of biologically relevant genes. Canonical pathway files (KGML) identified by DAVID were obtained from the KEGG PATHWAY database and converted to SBML file format using a file conversion software module developed at CFDRC. Multiple SBML files were automatically concatenated to construct a detailed pathway model describing the cellular response to MFQ exposure. An analysis of the resultant pathway model was performed using a suite of computational tools collectively known as the Boolean Network Dynamics Target Identification algorithm (BNDTI). This computational suite was developed to objectively identify critical nodes within a pathway using a Boolean pseudodynamics-based approach. The genes PTK2B and MDR1 were consistently identified by BNDTI as the most probable SNP target across multiple data sets, and multiple toxicity markers. The PTK2B gene codes for a protein tyrosine kinase (Pyk2) activated by elevated cytoplasmic Ca2+. The MDR1 gene codes for a p-glycoprotein (ABCB1), an ATP-binding cassette (ABC) transporter functioning as part of the blood brain barrier. Both genes were selected as initial targets for further investigation into the development of a genetic screen. An experimental validation of PTK2B function is currently underway using gene-silencing techniques (siRNA).

References: 1. World Health Organization. (2001) The use of antimalarial drugs. Report [Online] Accessed April 2006. 2. van Riemsdijk et al. (1997) Eur J Clin Pharmacol. 52:1 3. Aarnoudse et al. (2006) Clin Pharmacol Ther. 80:367 4. Dow et al. (2003b) Malar J. 2:14 5. Dow et al. (2005) Genomics. 86:539 6. Tusher et al. (2001) Proc Natl Acad Sci USA. 98:5116 7. Dennis et al (2003) Genome Biol. 4:R60

Extended Abstract Status: File Uploaded