: Non-alcoholic fatty liver disease (NAFLD) is currently the most common liver disease worldwide, paralleling the epidemic of obesity and type 2 diabetes. Despite the high prevalence of NAFLD, only a minority of patients progress to NASH and advanced fibrosis. The reasons for this inter-individual variability are not completely understood but can be partially accounted for by genetic risk factors (1). Although several common genetic variants associated with liver disease have been identified, there is still a proportion of NAFLD heritability that remains unknown. The rare rs143545741 C>T variant in the autophagy related 7 (ATG7) gene (P426L) has been associated with a higher risk of progressive NAFLD (2). Interestingly, ATG7 encodes a E1-like ubiquitin activating enzyme which is involved in hepatic lipophagy (3). We hypothesized that the unknown heritability of NAFLD might be partially explained by rare genetic variants, therefore not identified in the GWAS studies. Moreover, we assumed that loss-of-function variants in ATG7 might confer an increased susceptibility to NAFLD by reducing autophagic catabolism of lipid droplets in the liver. To examine the underlying mechanism of the low-frequency V471A variant and the rare T86I, L127I, Q170E, and P426L variants in ATG7, we performed in vitro experiments of HepaRG cells overexpressing the human V5-tagged ATG7. We observed a reduction in intracellular lipid content in HepaRG cells overexpressing the ATG7 wild type and the 86I mutant protein (p=0.029, n=4) but not the 127I, 170E, 426L, and 471A mutant proteins. Cells with the ATG7 127I, 170E, 426L, and 471A mutants had higher intracellular lipid content compared to cells overexpressing the wild type protein (p=0.029, n=4). Our data suggested that the low-frequency V471A variant and the rare L127I, Q170E, and P426L variants in ATG7 are loss-of-function, resulting in defective lipophagy, reduced hepatocellular lipid droplets turnover, and excessive lipid accumulation. More experiments are needed to clarify the underlying mechanism of the T86I variant. In conclusion, we highlighted a role for ATG7 in reducing hepatocellular lipid content. Furthermore, we provided evidence showing non-synonymous variants in ATG7 increase the risk of NAFLD and that these variants are loss-of-function. We speculate that ATG7 might be a new susceptibility risk genetic locus for liver disease development and progression.