Selection Protocol for Candidate Gene SNPs

Fig 1: Flowchart of the sequential selection criteria to obtain 5 SNPs per candidate gene. Order of selection is from base of the pyramid (the width represents, schematically, the number of SNPs) to the top. All SNPs involved have been validated to occur at a prevalence of at least 10% of the human population.

			Sample Tables Candidate Genes (Excel file) EDL SNPs (Excel file) Sample Patient Forms Patient Questionnaire forms (PDF file)

Introduction

In this proposal we will use a candidate gene approach to identify genetic determinants which predispose toward the development of hemodynamically significant atherosclerosis in the peripheral arteries. We have chosen candidate genes based upon the potential for their involvement in the pathophysiology of Peripheral Arterial Disease (PAD), and based upon our own ability to fully characterize polymorphisms of these genes at molecular and clinical levels, given a promising association. The candidate genes were identified in the original proposal, as well as potential SNPs for these genes. However, for most of our candidate genes, there are many more SNPs than we can investigate (we propose on average 5 SNPs per candidate gene). In the body of the grant we propose a strategy for selection of SNPs that we elaborate on below. Furthermore, we provide below an example of the selection strategy applied to endothelial lipase gene, which was first cloned and characterized by one of us (TQ). In the grant proposal, we have built in a 5 month start-up period, which will permit us to finalize the list of SNPs, a process that will require pilot phenotyping of 28 patients, so as to determine allele frequency.

Selection Criteria for SNPs within individual genes

The search for a list of all potential SNPs was achieved through a number of websites such as the UCSC Genome Browser , the SNP consortium and the Human Genome Project . Only validated SNPs will be selected, that is SNPs that are known to exist at a frequency of more than 10% in the general population. Such a list of SNPs was shown in Appendix I of the original PAD-RFA. Our objective is to choose five SNPs from this list for clinical analysis for each candidate gene.

Initially, we shall search the literature and utilize SNPs that have already been studied and have been associated with relevant states of vascular dysfunction (e.g. G894T or E298D in human eNOS is associated with preclampsia and placental disruption [1] . In addition, we shall employ the algorithm below (Fig 1) to complete the selection process.

As mentioned all SNPs selected will have been validated (by the literature and/or our pilot genotyping study) to exist in at least 10% prevalence in the human population,. [2] [3] For the purposes of simplicity we shall select only those SNPs that reside in exonic regions of the candidate genes and then choose SNPs that should lead to non-synonymous, non-conservative, polymorphisms. It has been shown that such SNPs while low in frequency have a moderate to high probability of being associated with a diseased phenotype. [2] The subsequent set of screening criteria will be focused upon the functional consequences (potential or realized) of the SNPs in question. Again, we shall use the existing literature to search for SNPs fitting these criteria but failing that we shall select SNPs that reside in catalytic sites of the candidate genes (e.g. His72Tyr for NADPH oxidase. [4, 5] Other selection criteria will be the SNPs proximity to regulatory subunits, or to sites of post-translational modification (phosphorylation, glycosylation and farnesylation sites). SNPs that could potentially alter secondary and tertiary structures of proteins (junctions of bends or at hydrophobic or hydrophilic pockets) will also be considered. Although the focus will be on exon-located SNPs, those SNPs located at exon-intron junctions or at intron sites important for RNA stability (such as the 5'- and 3'- untranslated regions) may also be considered. Whenever possible, literature studies whereby relevant SNPs have been associated with ethnic subgroups, family or twin studies or clustered with other SNPs that have been linked to diseased states will be utilized to direct the winnowing process. Since submission of the proposal, we documented the feasibility of this approach for the gene human Endothelial Lipase (EDL) which had previously been cloned and characterized by Dr. Quertermous [6] .

In a pilot genotyping study Drs. Quertermous and Myers identified SNPs of EDL and determined their prevalence (n=24 individuals). This exercise revealed 49 alleles. Only 13 of these alleles had the required prevalence in the population (>10%) for this genomic association study. Only 6 of these alleles were in exons. Accordingly these SNPs will be used in candidate gene association study. There is also one interesting non-synonymous allele (EDL_1.2) which results in a Gly to Ser substitution. Because there is a higher likelihood that this allele could result in a change in EDL function, we intend to include this allele in our association study, even though its prevalence (2.08%) falls below our general criteria. 

References:

1. Yoshimura, T., et al., The missense Glu298Asp variant of the endothelial nitric oxide synthase gene is strongly associated with placental abruption. Hum Genet, 2001. 108(3): p. 181-183.

 2. Tabor, H.K., N.J. Risch, and R.M. Myers, Opinion: Candidate-gene approaches for studying complex genetic traits: practical considerations. Nat Rev Genet, 2002. 3(5): p. 391-397.

 3. Xie, H.G., et al., Allelic, genotypic and phenotypic distributions of S-mephenytoin 4'-hydroxylase (CYP2C19) in healthy Caucasian populations of European descent throughout the world. Pharmacogenetics, 1999. 9(5): p. 539-549.

 4. French, J.K., et al., Potential thrombophilic mutations/polymorphisms in patients with no flow-limiting stenosis after myocardial infarction. Am Heart J, 2003. 145(1): p. 118-124.

 5. Stanger, O., et al., NADH/NADPH oxidase p22 phox C242T polymorphism and lipid peroxidation in coronary artery disease. Clin Physiol 2001 Nov;21(6):718-22, 2001. 21(6): p. 718-722.

 6. Hirata, K., et al., Cloning of a unique lipase from endothelial cells extends the lipase gene family. J Biol Chem 1999 May 14;274(20):14170-5, 1999. 274(20): p. 14170-14175.