Epigenetic control of variation and stochasticity in metabolic disease

Abstract

Background: The alarming rise of obesity and its associated comorbidities represents a medical burden and a major global health and economic issue. Understanding etiological mechanisms underpinning susceptibility and therapeutic response is of primary importance. Obesity, diabetes, and metabolic diseases are complex trait disorders with only partial genetic heritability, indicating important roles for environmental programing and epigenetic effects.

Scope of the review: We will highlight some of the reasons for the scarce predictability of metabolic diseases. We will outline how genetic variants generate phenotypic variation in disease susceptibility across populations. We will then focus on recent conclusions about epigenetic mechanisms playing a fundamental role in increasing variability and subsequently disease triggering.

Major conclusions: Currently, we are unable to predict or mechanistically define how "missing heritability" drives disease. Unravelling this black box of regulatory processes will allow us to move towards a truly personalized and precision medicine.

Keywords: Epigenetics; Inheritance; Metabolic diseases; Phenotypic variation.

Figure 1

Modes of variation impacting on phenotype. Mendelian variation, non-Mendelian variation and the environment concur to modify the phenotypic outcome of an individual, often intermingling into each-other. The contribution of Mendelian variation to height determination is estimated around 80–90%, whereas for obesity/T2D estimates are lower (∼35% heritability).

Figure 2

Distributions of phenotypic variation. Non-Mendelian phenotypic variation acts during development and can be described either as a discrete, combined or as a continuous distribution. In the first scenario the traits subject to the variation are termed ‘polyphenic’ (here exemplified by BMI, whereas height is exemplifying continuously varying traits).

Figure 3

Windows of sensitivity during development. From the moment of conception throughout adulthood, the environment modulates and defines the phenotypic outcome of an individual, particularly impacting highly sensitive windows. These windows of sensitivity parallel the major morphological and epigenetic sub-divisions of development. Blue line: paternally derived alleles; light blue line: re-establishment of DNA methylation at paternally derived alleles after gametogenesis; purple line: maternally derived alleles; pink line: re-establishment of DNA methylation at maternally derived alleles after gametogenesis; dark line: epigenetic reprogramming in Primordial Germ Cells (PGCs); dotted dark line: alleles that are protected from epigenetic reprogramming at fertilization (imprinted genes, repetitive sequences, heterocromatin near centromeres); dotted yellow line: Endogenous RetroViruses (ERVs) and metastable epialleles can fully or partially escape epigenetic reprogramming. Epigenetic inheritance through generations can occur from incomplete epigenetic reprogramming (Cantone and Fisher, 2013).

Figure 4

Intergenerational epigenetic inheritance. (A) In C. elegans, dsRNAs in food can silence an endogenous reporter. This repression is maintained for more than 20 generations in absence of the initial silencer. This maintenance of silencing requires nuclear RNAi factors and chromatin modifying proteins (Ashe et al., 2012). (B) In D. melanogaster (below), adult fed an obesogenic high-sugar exhibit a typical U-shaped obesity response with low- and high-sugar sired individuals showing exaggerated triglycerides and the latter also characterized by increased body weight. This intergenerational metabolic reprogramming results from global alterations in chromatin state integrity, particularly from reduced H3K27me3 and H3K9me3 domains (Öst, Lempradl et al., 2014).