Rare genomic variants contribute to systolic blood pressure variation in the framingham heart study

Abstract

The physiology and patho-physiology of blood pressure (BP) regulation is highly complex. The identification of genomic variants associated with variation in BP would likely enhance the understanding of this regulatory network. Genome-wide association studies in ten thousands of individuals have already identified some genomic variants that might play a role. However, given the complexities of BP regulation it is highly likely that population samples are heterogeneous. Genetic factors that carry risk for BPelevation early in life might be different from those influencing high BP later in life. Physiologic BP variation in normal individuals might have different genetic contributions than those found in high risk individuals and risk factors might differ in men and women. Growth mixture modeling is a less explored method in genetic research to address unobserved heterogeneity in population samples. Here, we applied this technique to longitudinal data of the Framingham Heart Study. We examined systolic BP measures in 1060 males from 692 families at four different time points spanning thirty years of observation. We detected three subclasses, which varied significantly in their developmental trajectories over time. The first class consisted of 60 high-risk individuals characterized by elevated BPearly in life and a steep increase over time. The second group of 131 individuals displayed first normal BP, but reached high BPvalues late in their life time. The largest group of 869 individuals could be considered a normative group with normal BP on all exams. In order to identify genetic modulators for this phenotype we tested the genome-wide association between class membership probability and single nucleotide polymorphisms using the Affymetrix 500k array. In unrelated individuals (one member per family) Class 1 membership probability was significantly associated with the C allele of the SNP rs10494067 located in the gene Netrin-G1 Precursor (NTNG1) on chromosome 1p13.3 under the recessive model (correlation trend p=1.22x10 -10). All individuals homozygote for this SNP allele were found to be members of Class 1 and homozygotes for the C allele were absent in all other classes. This intronic SNP is rare in most populations (0.04 in some European and Asian populations, but absent in the Yoruba). NTNG1 is highly expressed in tubular epithelial cells of the human kidney and was found to be important in preventing renal inflammation and injury (Wang W, 2008). Some very rare SNPswere present only in members of one latent class and absent in all others. The rare C allele of SNP rs1445404 located in the third exon of the gene EYA (eyes absent homolog 1 in Drosophila)was present in only four individuals, one homozygote and three heterozygotes from four different families (6.6% of the individuals in Class 1) (correlation trend p= 1.39x10 -13, OR98.1). This miss-sense mutation in exon 3 changes an Alanine to a Proline at amino acid position 20 of the protein with likely consequences for the protein structure and folding of the protein. Mutations in EYA were found in patients with Brachio- Oto-Renal syndrome, aswell as in individuals with isolated renal malformations (Hoskins BE, 2008; Orten DJ, 2008). We demonstrate that stratification of population samples in high risk and low risk groups based on longitudinal development over time can facilitate the identification of rare genetic risk factors for common complex disorders.