Nonalcoholic fatty liver disease (NAFLD) is marked by accumulation of fat in liver cells with accompanying inflammation and variable degrees of cell injury and fibrosis. When cell injury and fibrosis are present, the disease has a potential to progress and is referred to as nonalcoholic steatohepatitis (NASH), which can lead to cirrhosis, liver cancer, morbidity and mortality. The etiology of NASH is not clear nor is there an approved treatment modality for it. NAFLD has become an extremely common disorder, estimated to affect up to 30% of individuals in the US. Unfortunately, it commonly goes unrecognized, as demonstrated by our recent finding that over a 5-year period, 28% of subjects enrolled as healthy volunteers to studies at the NIH Clinical Center were likely to have underlying NAFLD. As such, there is a clear need to understand the pathophysiology of the disease and its treatment. Our focus on NAFLD is three-fold: first, we aim to identify and characterize key genes that play a role in the pathogenesis of NAFLD through the use of genetic studies and cell- or animal models and identify their suitability as therapeutic target. Second, we aim to identify and refine effective treatments for the disorder. Our third focus is on the physiology of fat accumulation and injury in the liver, especially as it relates to handling of oral caloric load. Genome wide association (GWA) studies identified single nucleotide polymorphisms (SNPs) that are associated with increased hepatic fat or elevated liver enzymes, presumably reflecting nonalcoholic fatty liver disease (NAFLD). We initiated a study to investigate whether these SNPs are associated with histological severity in a large cohort of NAFLD patients. We confirmed the association of the rs738409G allele in the PNPLA3 gene with steatosis and were first to describe its association with histological severity. In pediatric patients, the high-risk rs738409G allele was associated with an earlier presentation of disease. Similarly, we discovered an association of several SNPs near or in the gene for hydroxysteroid (17) dehydrogenase 13 (HSD17B13) with histological features of NAFLD. In-depth genotyping of the gene region demonstrated associations of coding and splice-site SNPs in the gene with NAFLD, confirming a possible role for the protein in the pathogenesis of NASH. HSD17B13 is an enzyme that we found to be predominantly expressed in the liver and to colocalize with lipid droplets. We identified that HSD17B13 is involved in retinoid metabolism and that genetic variants that lead to loss of its enzymatic function are genetically associated with decreased severity of NAFLD. These findings generated remarkable interest in the pharmaceutical industry and several HSD17B13 inactivating agents are being explored as potential therapies for NASH. Despite the promising genetic association and pharma interest, the actual physiological role of HSD17B13 in vivo and the mechanism by which its inactivation leads to protection from NASH-associated injury are still unknown. To improve our understanding of HSD17B13s role, we established an Hsd17b13 knock-out mouse model and tested it under several dietary conditions mimicking NAFL, NASH and alcoholic liver disease. Curiously, we did not identify protection from injury for any of those models, suggesting that inter-species differences may exist. We are currently exploring those differences to further shed light on the potential therapeutic targeting of HSD17B13. In a similar genetic analysis, we identified associations of SNPs near RAR-related orphan receptor (RORA) with NAFLD histology. In-depth genotyping near RORA, a nuclear receptor involved in control of circadian rhythm and metabolic functions, showed that SNPs that are associated with NAFLD are located in the putative promoter region of 2 of the 4 splice variants (variants 2 and 3) as opposed to SNPs upstream of other variants, suggesting that alternation in the relative expression of the different isoforms affects fat accumulation in the liver. RORA knock-out in cell lines did not affect the degree of hepatic lipid accumulation under standard conditions. However, when cells were overloaded with nutrients (high glucose, high-fatty acids medium), the amount of intracellular fat significantly decreased with knock-down, predominantly through a decrease in the average size of the lipid droplets. A similar effect was seen in adipocytes. We have established a murine liver-specific Rora knock-out and through that demonstrate the important role of RORA on the generation of liver fat. Vitamin E has been shown in a randomized placebo-controlled trials to be an effective therapy for NASH. Curiously, treatment with vitamin E resulted not only in a decrease in injury (thought to reflect its antioxidant effect) but was also associated with a decrease in liver fat, through an unknown mechanism. We combined a clinical mechanistic trial with in vitro experiments to determine the mechanism of action of vitamin E. We identified a key cellular pathway by which oxidative stress (a common feature of NAFLD) increases liver fat through upregulation of hepatic de novo lipogenesis and confirmed that vitamin E blocks the activation of this pathway through its antioxidant activity. In collaboration with colleagues at NCI, we were able to establish an automated method to quantify hepatic 4-hydroxynonenal (4-HNE) adducts, a marker of oxidative stress-induced damage, and confirmed its applicability with samples from the vitamin E trial. As NAFLD is intricately related to food intake and energy metabolism, we are undertaking clinical trials to evaluate the handling and fate of nutrients by the fatty liver. We used the BreathID real-time breath test device in combination with a labeled fatty acid to demonstrate a decrease in the rate of fatty acid oxidation in subjects with NAFLD compared to controls. Similarly, we utilized a metabolomic approach to identify the response of NAFLD subjects to a standardized meal challenge. Finally, we are currently performing a clinical trial, aimed to elucidate the hepatic response to an oral carbohydrate load at the transcriptomic, lipidomic levels, and to identify prediction rules for response to Semaglutied, a GLP-1 receptor agonist.