Lipid droplet-associated proteins in alcoholic and non-alcoholic steatohepatitis in patients with polymorphisms in PNPLA3

Lipid droplets (LDs) play a crucial role in lipid homeostasis by tightly regulating fatty acid storage and release while preventing a toxic accumulation of cellular lipids. The LD-associated proteins of the perilipin family are key players at LDs. They differentially regulate biogenesis, structural integrity, as well as degradation of LDs by interacting with lipases. In particular, perilipin 1 and 5 were shown to be essential proteins in lipolysis, the stepwise breakdown of lipids, in different tissues. Malfunctions within this process lead to an accumulation of cellular lipids. Perilipins thereby play a pivotal role in liver diseases, especially alcoholic (ASH) and non-alcoholic steatohepatitis (NASH).

The I148 M single nucleotide polymorphism of the human patatin-like phospholipase domain containing protein 3 (PNPLA3) gene has been shown to be associated with hepatic steatosis, increased inflammation, fibrosis, cirrhosis and the development of hepatocellular carcinoma and is thereby an important prognostic marker in patients with NASH.

In contrast to the role of PNPLA3 as a prognostic marker, the underlying molecular mechanism that leads to steatogenesis and inflammation remains largely unexplored.

In order to unravel the role of the lipase PNPLA3 in regulating LDs in a perilipin-dependent manner, (immuno)-histochemical analysis of a collective of 47 ASH- and 25 NASH-patients with known PNPLA3-status was undertaken. Histologically, livers of I148 M carriers showed enhanced ballooning, acinar inflammation, microgranulomas and increased fibrosis. In addition, hepatocytes of livers that carry the I148 M polymorphism showed a more prominent staining for perilipin 2 at an increased number of ballooned hepatocytes. Perilipin 5 localized less to LDs but instead showed a more cytoplasmic and partially nuclear localization. Interestingly, hepatic zones that were strongly positive for PNPLA3 showed diminished perilipin 1 expression.

In in vitro co-immunoprecipitation experiments, we could identify perilipin 5 and the lipase PNPLA2, a key enzyme in lipolysis, as novel interaction partners of PNPLA3. Both proteins strongly co-localized with PNPLA3 at LDs when overexpressed in human hepatocellular carcinoma cells. In addition, we could show that perilipin 5, PNPLA3, and PNPLA2 are part of the same complex whereas perilipin 5 is enhancing the binding of PNPLA3 with PNPLA2 dramatically. Strikingly, PNPLA2 forms homo-dimers or -oligomers which is inhibited in the presence of PNPLA3. When we addressed the impact of PNPLA3 as well as PNPLA3(I148 M) on the lipolytic activity of PNPLA2, an inhibitory effect of PNPLA3 was detected, which is strongest for PNPLA3 (I148 M).

In summary, our data indicate that PNPLA3 regulates lipolysis by repressing the lipolytic activity of PNPLA2 in a perilipin 5-dependent manner most likely via disrupting active PNPLA2 homo-dimers/oligomers. The enzymatically inactive PNPLA3(I148 M) even further reduces PNPLA2-mediated lipolysis and increases toxic accumulation of lipids. We have thereby unraveled a mechanism of PNPLA3 in the progression of steatotic liver diseases. Concerning the long-standing debate on why and when bland steatosis progresses to steatohepatitis, our data point to a critical step in lipolysis rather than in lipogenesis itself.

This contribution contains data of an unfinished doctoral thesis of Inga Schwittai.