Network-based analysis of the sphingolipid metabolism in hypertension.

Mogens Fenger,Jørgen Jeppesen,Allan Linneberg

doi:10.3389/fgene.2015.00084

Mogens Fenger, Jørgen Jeppesen + Show 1 more

Open Access

https://doi.org/10.3389/fgene.2015.00084

Copy DOI

Abstract

Common diseases like essential hypertension or diabetes mellitus are complex as they are polygenic in nature, such that each genetic variation only has a small influence on the disease. Genes operates in integrated networks providing the blue-print for all biological processes and conditional of the complex genotype determines the state and dynamics of any trait, which may be modified to various extent by non-genetic factors. Thus, diseases are heterogenous ensembles of conditions with a common endpoint. Numerous studies have been performed to define genes of importance for a trait or disease, but only a few genes with small effect have been identified. The major reasons for this modest progress is the unresolved heterogeneity of the regulation of blood pressure and the shortcomings of the prevailing monogenic approach to capture genetic effects in a polygenic condition. Here, a two-step procedure is presented in which physiological heterogeneity is disentangled and genetic effects are analyzed by variance decomposition of genetic interactions and by an information theoretical approach including 162 single nucleotide polymorphisms (SNP) in 84 genes in the sphingolipid metabolism and related networks in blood pressure regulation. As expected, almost no genetic main effects were detected. In contrast, two-gene interactions established the entire sphingolipid metabolic and related genetic network to be highly involved in the regulation of blood pressure. The pattern of interaction clearly revealed that epistasis does not necessarily reflects the topology of the metabolic pathways i.e., the flow of metabolites. Rather, the enzymes and proteins are integrated in complex cellular substructures where communication flows between the components of the networks, which may be composite in structure. The heritabilities for diastolic and systolic blood pressure were estimated to be 0.63 and 0.01, which may in fact be the maximum heritabilities of these traits. This procedure provide a platform for studying and capturing the genetic networks of any polygenic trait, condition, or disease.

Highlights

Essential hypertension refers to hypertension with no known cause, affects approximately 30% of the adult population, and is a major risk factor for stroke, coronary incidences, and end-stageGenetics of hypertension renal disease (Kearney et al, 2005)
Blood pressure levels in a population constitutes an ensemble of polygenic conditions (Shih and O’Connor, 2008) meaning that the blood pressure is regulated by a plethora of integrated biochemical and physiological processes that are blue-printed in the genome (Fenger et al, 2011)
General Description of the Population Fourteen subpopulations were previously identified by latent class/structural equation modeling (Fenger et al, 2011) using blood pressure measurements as the outcome variables

Summary

Introduction

Essential hypertension refers to hypertension with no known cause, affects approximately 30% of the adult population, and is a major risk factor for stroke, coronary incidences, and end-stage. In particular the large GWA studies have been somewhat disappointing as only sketchy information about the networks regulating the blood pressure have been provided This may not be that surprising as the vast majority of analysis have been done as a search of monogenic effects in a case-control framework. We addressed the problem of resolving physiological heterogeneity of a population by implementing a latent class/structural equation modeling (LCA/SEM) framework using common physiological variables generally assumed to be related to cardiovascular conditions (Fenger et al, 2011) This approach revealed 14 distinct subpopulations with different propensity to develop hypertension embracing subpopulations with no hypertensive cases at all to subpopulations where the majority or all the subjects presented themself with hypertension. Even with the relatively small number of genetic variations included here the analytical procedures are highly involved, but most importantly, despite the genetic complexity of blood pressure regulation, almost all affected subjects could be captured by genotyping a few genes, the predictive value of the compound genotypes varies considerably

Materials and Methods

Results

Discussion