The central focus of our research program is how dysregulation of lipid uptake and trafficking contributes to human diseases. We use human genetics to identify genes that contribute to disorders of lipid metabolism, and perform functional studies to define the underlying mechanisms. Since the optimal strategy to identify functional sequence variations contributing to disease depends on the underlying genetic architecture of the trait, we have used three complementary genetic approaches:

Resequencing extremes: To identify alleles of large phenotypic effect, we resequenced candidate genes in individuals with extreme phenotypes in the general population (1). This strategy has revealed that severe loss-of-function alleles are more common than previously recognized, and provided the first direct evidence that rare variants play an important role in complex traits in the general population (1; 2). The power of the “resequencing extremes” strategy is best illustrated by identification of loss-of-function alleles in PCSK9, which reduced plasma low density lipoprotein-cholesterol (LDL-C) levels and protect against coronary heart disease (CHD) (3, 4).

Resequencing populations: Our “resequencing extremes” strategy suggested that severe loss-of-function alleles may be sufficiently common to allow a reverse genetics approach to determine the roles of genes in human physiology. We hypothesized that the phenotypic consequences of sequence variations in genes can be defined by sequencing large cohorts of well-characterized individuals. This approach was used to determine the effects of loss-of-function mutations in angiopoietin-like proteins (ANGPTL3-5) in humans (5; 6). We showed that mutations in all three proteins are associated with reductions in plasma triglyceride levels.  We also identified a new member of the ANGPTL family, ANGPTL8, that plays a key role in triglyceride trafficking to tissues in response to nutrient intake (7, 8). We have used whole exome and genome sequencing in informative families, individuals and populations to identify new genes contributing to metabolic diseases (910 11).

Genome-wide association studies (GWAS): To identify novel genes, we have performed unbiased large-scale association studies in a large population with uniform, comprehensive phenotyping, the Dallas Heart Study. This approach yielded the first genetic locus (9p21) that is directly associated with CHD independent of known risk factors (12) and the first sequence variant that confer susceptibility to nonalcoholic fatty liver disease (NAFLD), including steatosis, steatohepatitis, cirrhosis, and hepatocellular carcinoma (13). This variant, PNPLA3 (I148M), also magnifies the risk of alcoholic liver disease.  Most recently, we have identified a new risk variant in a gene of unknown function, TM6SF2 (14), that is involved in the secretion of lipoproteins from the liver (15), resulting in lower plasma levels of triglyceride and cholesterol.

Identification of these genes has provided us with molecular handles for mechanistic studies to define key pathways in lipid metabolism in humans. A major focus of our ongoing research program is to elucidate the roles of ANGPTLs, PNPLA3, and TM6SF2 in the trafficking and processing of lipids and lipoproteins. We are also identifying new molecular pathways contributing to metabolic health and disease. To perform these studies we have established the following methodologies in our lab:

  • Mass spectrometry to comprehensively profile lipids in cells, sub-cellular compartments, and tissues (16).
  • Stable isotope labeling and isotopomer analysis to monitor lipid biosynthesis, partitioning and transport in cultured cells and in vivo in mice (16).
  • X-ray crystallography and cryo-EM to determine the structure of a sterol transporter, ABCG5/ABCG8 (17).
  • Flow cytometry to assay inter-individual variability of cholesterol accessibility in the membranes of red blood cells (RBCs) in humans (18).

The overall goal of our laboratory is to elucidate the key processes that maintain lipid and energy homeostasis, and when disrupted cause human disease.