PROJECT SUMMARY
Dogma teaches that an individual’s phenotype (and disease) results from genetics, the environment, and their
interactions. Yet numerous studies of monozygotic human twins and isogenic animal models indicate that
significant portions of disease variability cannot be explained by genetic and environmental inputs. For example,
genetics accounts for ~50% and environment accounts for <1% of metabolic disease in monozygotic twins,
leaving a striking unexplained variance (discordance) of ~50%. Similar results are reported for many other human
diseases and complex traits. The long-term goal of this project is to understand the origins and regulatory
mechanisms underlying this unexplained phenotypic and disease variation. The operating hypotheses are that
phenotypic variation itself is a quantitative trait, and there are probabilistic, intracellular processes regulated by
epigenetic mechanisms that are responsible for significant portions of unexplained phenotypic variation. The
hypothesis is based on prior work with haploinsufficient Trim28+/D9 mice, where genetically and environmentally
identical littermates emerge as either lean or obese, with few intermediates (i.e., an epigenetically driven obesity
polyphenism). TRIM28 mRNA expression levels also predict obesity in human children. This body of work
suggests that epigenetic silencers are master regulators of probabilistic processes in humans, and may also be
responsible for regulating phenotypic variation. However, the epigenetic mechanisms and genomic loci
responsible for this remarkable, probabilistic, and bi-stable disease potential are unknown. Before the field can
even begin deciphering the (epi)genetic architecture that regulates probabilistic process and variability, we need
to first determine which type of epigenetic silencers are involved in the bistable switch from one development
trajectory to the other, and which genetic loci respond to the switch. We will meet this objective by performing a
focused gene-gene and gene-environment epistasis experiment with Trim28+/D9 mice, and score the offspring
for stability, severity, and frequency (i.e. the variability) of bistable metabolic disease. For crosses showing
additive effects on disease variability, we will perform total RNAseq and RELACS in precursor and mature
adipocytes to identify genomic loci associated with metabolic disease switches and variation. With this
knowledge, we will be able to generate specific hypotheses about the genes, pathways, and physiological
mechanisms that not only regulate metabolic disease, but control phenotypic variation as a quantitative trait.