Home > Research > Research Areas > Metabolism

Metabolism broadly refers to all biochemical processes that sustain a living organism. More narrowly, metabolism can refer to the biochemical processes that regulate a particular substance or a group of molecules. Our work focuses on energy metabolism; we aim to identify the genes necessary for proper utilization and storage of carbohydrates and lipids.

Diabetes, obesity, and hypercholesterolemia

Genetic and environmental factors both strongly influence energy metabolism, and our studies aim to identify heritable metabolic defects that may predispose individuals to obesity and diabetes. This research is of particular importance in light of the growing prevalence of metabolic syndrome in the industrialized world, where currently 20-30% of the population is affected by the syndrome. Identifying the genes required for metabolic processes of cells and tissues will enable the development of drug treatments that modulate gene function to promote healthy energy metabolism. Genetic screens in this research area include screens that measure deviations from normal of body weight, blood pressure, and levels of blood glucose and cholesterol in mutagenized mice. Screens which challenge mice with insulin or glucose are also carried out to identify insulin resistance and impairments of glucose metabolism.

Whereas many heritable obesity phenotypes are known, lean phenotypes are comparatively uncommon.  Yet they can reveal critical checkpoints regulating energy balance.  The teeny mutation of Kbtbd2 and the supermodel mutation of Samd4 (described below) caused such phenotypes.

An unusually small body size defined the teeny phenotype, ascribed to a mutation of Kbtbd2.  Although smaller in size and weight throughout life, homozygous teeny mice showed normal levels of IGF-1 but consumed more food and water per gram of body weight than wild type mice.  Detailed phenotypic analysis further revealed hyperglycemia, hyperinsulinemia, insulin resistance, shrunken adipose tissue and adipocytes, and fatty liver.  These defects were found to stem from impaired insulin signaling exclusively in adipose tissue, since implantation of wild type adipose tissue into KBTBD2-deficient mice was by itself sufficient to rescue hyperglycemia, hyperinsulinemia, and hepatic steatosis.  The function of KBTBD2 is substrate recognition for the Cullin 3- based ubiquitin ligase complex that marks proteins for proteasome-mediated degradation, and we found that p85α, the regulatory subunit of phosphoinositol-3-kinase (PI3K), is one of its targets.  Thus, in the absence of KBTBD2, p85α accumulated to 30-fold greater levels than in wild type adipocytes, and excessive p110-free p85α blocked the binding of p85α-p110 heterodimers to insulin receptor substrate (IRS) 1, interrupting the insulin signal (Figure 1).  Inactivation of the p85α-encoding gene Pik3r1 rescued diabetes and hepatic steatosis phenotypes of Kbtbd2−/− mice.

KBTBD2 deficiency defined a new form of insulin resistance leading to severe diabetes.  Strikingly, we found that diet-induced obesity led to downregulation of KBTBD2 and upregulation of p85α in adipose tissue of wild type mice.  Our findings shed light on the role of insulin sensitivity in fat, where KBTBD2 may be modulated by diet-induced obesity to impact systemic glucose and insulin regulation.

The teeny phenotype and causative Kbtbd2 mutation are reported in (Zhang et al. Proc.Natl.Acad.Sci.U.S.A. 113, E6418-E6426).

Figure 1. Regulation of p85α by KBTBD2 in wild type (WT) and homozygous teeny mice. (Left) PI3K is a heterodimer of a p85 regulatory subunit and a p110 catalytic subunit. In the absence of insulin, p85α binds to p110 and conformationally inhibits p110 catalytic activity. Insulin induces phosphorylated IRS-1 to recruit p85α-p110 heterodimers, relieving the inhibitory conformation of p85α and resulting in activation of p110 catalytic activity and downstream signaling. In WT mice, KBTBD2 regulates p85α protein abundance by promoting K48-linked ubiquitination of p85α, leading to proteasome-mediated degradation of p85α. (Right) In KBTBD2-deficient teeny mice, p85α accumulates to high levels, and excess p85α by itself binds to tyrosine phosphorylated IRS-1. The constitutive association of p110-free p85α with phosphorylated IRS-1 prevents recruitment of p85α-p110 heterodimers, resulting in impaired PI3K signaling in response to insulin. Figure modified from Zhang et al. Proc.Natl.Acad.Sci.U.S.A. 113, E6418-E6426.


The supermodel phenotype was characterized by leanness and thoracic kyphosis, reduced adipose and muscle tissue, and abnormalities in the morphology of myofibers and adipocytes.  Altered glucose metabolism was evident in homozygous supermodel mice, in which diminished insulin release resulted in impaired glucose tolerance; however, fasting glucose levels were reduced and insulin sensitivity was elevated in these animals.  Supermodel mice were resistant to high fat diet-induced obesity, likely because of excessive energy expenditure due to uncoupled mitochondrial respiration.  The causative mutation was identified in Samd4, a gene encoding an RNA-binding protein with no previously known function in mammals.  Samd4 is a mammalian homolog of Drosophila Smaug, a translational repressor in fly embryos. 

The phenotype of supermodel homozygotes was similar to that of mice with muscle- and adipose-specific deficiency of Raptor, an mTORC1 complex component, and we found reduced phosphorylation of 4E-BP1 (an mTORC1 target) and S6 (a substrate of the mTORC1 target S6K1) in muscle and adipose tissues of homozygous supermodel mice.  Although Samd4 failed to interact with mTORC1 complex components, it interacted with several 14-3-3 proteins and was subject to phosphorylation by Akt in vitro.  We suggest that Samd4 fulfills a nonredundant function in muscle and adipose tissue to promote signaling by mTORC1, a key regulator of cellular energy metabolism (Figure 2).

The supermodel phenotype and causative Samd4 mutation are reported in Chen et al. Proc.Natl.Acad.Sci.U.S.A. 111, 7367-7372.

Figure 2. Samd4 may promote mTORC1 signaling that promotes glucose metabolism in muscle and adipose tissue. The mTORC1 complex regulates numerous cellular processes (left). mTORC1 positively regulates glucose metabolism, and we hypothesize that Samd4 is an upstream regulator of mTORC1 in this process (right). Both 14-3-3 and Akt are positive regulators of mTORC1 through inhibiting PRAS40 and TSC2. Samd4 interacts with 14-3-3 proteins in a muscle cell line and is phosphorylated by Akt in vitro.