Poorly predictable response variability to many commonly-used drugs may impose important clinical and economical constraints, sometimes leading to unwanted side-effects or showing at best little relief. Amid critical factors that contribute to such drawbacks, common genetic variants may modulate not only disease but also drug action and response. Yet, the identification of pharmacogenetic determinants in humans is complicated by limitations (genetic heterogeneity, sample size, compliance to drug prescription, poly-medication, placebo effect) that can be more easily addressed in controlled animal models. To date, most murine cardiovascular-associated quantitative trait loci (QTL) map concordant human QTLs. Besides, genetic determinants for both mono- and multigenic traits have been successfully characterised in such models. Thus, we reasoned that identifying pharmacogenetic variants in mice may advantageously contribute to the understanding of drug response variability in humans. As most inbred laboratory mice descend from a very limited number of progenitors, each strain can be considered as an individual member of a rather close family of mice. Moreover, the sequencing of the mouse genome, together with the availability of ever denser single nucleotide polymorphism (SNP) maps for diverse inbred strains provide major tools to enhance and accelerate mouse genetic trait mapping by correlating in vivo phenotypes with SNP alleles.
Here, we propose to develop a systemic and unbiased genome-wide screening model for cardiovascular pharmacogenetic variants in mice. Specifically, we will phenotype age- and sex-matched inbred mice of at least 20 strains for parameters like systolic blood pressure, heart rate and electrocardiogram, under basal conditions and in response to isoproterenol (a Beta-adrenergic agonist) or atenolol (a Beta-blocker). At the end of the 2-week treatment, complementary phenotypes such as cardiac mass, as well as lipids, electrolytes and drug metabolites in urine and plasma will be measured. Mice will then be sacrificed and several biological samples will be collected and stored for later studies (functional genomics, transcription profiling, proteomics). To map the genetic loci that modulate the cardiovascular response in untreated and drug-exposed mice by bio-statistical correlation methods, we will use publicly available multi-strain SNP information to identify SNP alleles that co-segregate with the divergent phenotypes.
Apart from providing information on the genetic architecture of the trait studied, these studies will allow the identification of the etiologic variants underlying these differences, some of which will map in previously unsuspected genes. Also, they may shed new light on the normal and patho-physiological mechanisms of cardiac diseases. In the long term, these results may open up onto a whole new field of functional, physiological or pharmacological projects, eventually allowing the testing of orthologous human loci for their involvement in differential drug responses. Once validated, these could be introduced into clinical practice as prognostic biomarkers.