A well-oiled machine

Abstract

Liver steatosis is characterized by an abnormal buildup of hepatic fat content. Our understanding of how this fat balance is normally regulated remains limited. Recently, autophagy has been implicated as one potential mechanism contributing to the breakdown of cytoplasmic fat storage organelles known as lipid droplets (LDs) in the hepatocyte. In our recent publication, we show that the large GTPase DNM2/dynamin 2 helps promote lipophagic turnover by facilitating the scission of nascent lysosomes from autolysosomal tubules during autophagic flux. Genetic and pharmacological perturbations of DNM2 function in cultured cells result in the generation of aberrantly long autolysosomal reformation tubules. As a consequence, hepatocytes accumulate LDs. An alleviation of DNM2 inhibition results in the scission of reformation tubules and the return of LD turnover to normal levels. DNM2 therefore plays a critical role in the regulation of the lipophagic machinery in the hepatocyte.

An important attribute of the hepatocyte is the ability to efficiently sequester triglycerides into cytoplasmic organelles known as lipid droplets. The packaging and mobilization of LDs in response to changes in nutrient availability is a tightly controlled and coordinated process. Regulatory defects can result in the excessive accumulation of hepatic lipid content (steatosis), which can ultimately progress to steatohepatitis or hepatocellular carcinoma. Importantly, the steatotic condition is thought to be entirely reversible. Therefore, a better understanding of the processes used by the cell to break down these lipid-containing structures could be vital toward the development of a therapeutic intervention in fatty liver disease. Currently, very little is known about the mechanisms underlying the normal metabolism of LDs in the liver.

For many years, LDs were viewed simply as inert depots used for the storage of excess fat. A culmination of many studies has since shown that these are actually extremely dynamic structures, continuously undergoing cycles of synthesis and breakdown. A number of reports have sought to understand the mechanisms underlying LD synthesis. In contrast, comparatively little research has been done to understand the molecular basis for how LDs are normally catabolized in the hepatocyte. Screening of the LD proteome has implicated numerous membrane trafficking proteins as being associated with the LD monolayer. We therefore set out to determine whether an important mediator of vesicle trafficking, DNM2, contributes to the regulation of LD turnover in the hepatocyte.

DNM2 is a large (100 kDa) GTPase that has been studied extensively in the scission of vesicles as they are internalized from the plasma membrane during endocytosis. In our study, we demonstrate that DNM2 also functions to control the lipophagic process at the level of the autolysosome. Using genetic and pharmacological inhibition of DNM2, we show that hepatocytes are no longer able to efficiently catabolize LDs following nutrient depletion. Knockdown experiments resulting in such metabolic defects can be rescued by re-expression of wild-type DNM2, but not a catalytically inactive mutant. Morphologically, the consequences of these various perturbations result in the formation of grossly enlarged autophagic compartments. Extending radially outward from these distended organelles are elongated tubules, occasionally greater than 20 μm in length. Electron microscopy shows that these tubules possess regularly spaced varicosities, reminiscent of the phenotype resulting from the expression of a dominant-negative form of DNM2 at the plasma membrane, which results in an accumulation of late-stage clathrin-coated pits.

These tubules appear to be exaggerations of the same membrane extensions that form during the process of autophagic lysosomal reformation. This membrane-recycling process helps repopulate the cellular lysosomal pool during enhanced autophagic flux. During autophagic lysosomal reformation, membranes are extruded from the autolysosome and “broken” into short tubular structures that bud into the cytoplasm as protolysosomes. These protolysosomes then undergo a poorly characterized maturation process. Our study localizes DNM2 to the site of action at these autolysosomal tubule structures and shows that inhibition of DNM2 function promotes their elongation and prevents the generation of new protolysosomes. Subsequent dynamin-inhibitory drug washout experiments reveal that when DNM2 function is restored, these tubules immediately begin to shorten.

It appears from the findings of others that the molecular architecture of the autolysosomal reformation tubule closely resembles that of the plasma membrane, as these structures contain clathrin, the AP2 adaptor complex, and phosphoinositide kinases, and exhibit an enrichment in PtdIns(4,5)P₂ phospholipids. These components are found in abundance at the endocytic pits of the plasma membrane, a primary site of DNM2 function. Therefore, a cytoplasmic regulatory protein like DNM2 may be recruited to the topologically equivalent reformation tubule with equal affinity. As shown in our paper, the localization of DNM2 at this autophagic endomembrane is paramount in maintaining efficient autophagic degradation of LDs (Fig. 1).

Figure 1. During lipophagy, a lipid droplet targeted for degradation is enclosed in an autophagosomal membrane. Following fusion with the lysosome and subsequent degradation of lipid by lysosomal acid hydrolases, the autolysosomal membrane is salvaged for recycling into new lysosomes. In this process, known as autophagic lysosomal reformation, membrane tubules are budded with the help of the mechanoenzyme DNM2. The newly generated protolysosomes can then undergo a maturation process to keep hepatocyte lipophagy operating at peak efficiency under nutrient-limiting conditions.

In conclusion, there are many outstanding questions regarding the regulation of lipophagy in the hepatocyte, with scores of different proteins implicated in this process. Together with our recent identification of DNM2 as an active mediator of autolysosome reformation, it is likely that the roles for additional membrane trafficking proteins can be resolved. It will be important to define how the autophagic machinery is specifically recruited to the LD surface and what signals promote the targeting of lipid droplets for lipophagic degradation. Additionally, it will be interesting to define the complementary roles of the cytoplasmic lipases present in the liver. Furthermore, more work is required to determine whether lipophagy is also common in specialized fat-storage cells such as the adipocyte. Such questions are only now beginning to be addressed.

10.4161/auto.27486