Intracellular lipase and regulation of the lipid droplet

Abstract

Purpose of review:

Lipid droplets (LDs) are increasingly recognized as distinct intracellular organelles that have functions exclusive to the storage of energetic lipids. LDs modulate macrophage inflammatory phenotype, control the availability of energy for muscle function, store excess lipid, sequester toxic lipids, modulate mitochondrial actitiviy and allow transfer of fatty acids between tissues.

Recent findings:

There have been several major advances in our understanding of the formation, dissolution, and function of this organelle during the past two years. These include new information on 1) movement and partition of amphipathic proteins between the cytosol and LD surface, 2) molecular determinants of LD formation, and 3) pathways leading to LD hydrophobic lipid formation. Rapid advances in mitochondrial biology have also begun to define differences in their function and partnering with LDs to modulate lipid storage versus oxidation.

Summary:

This relationship of LDs biology and cellular function provides new understanding of an important cellular organelle that influences muscle function, adipose lipid storage, and diseases of lipotoxicity.

INTRODUCTION

All cells have the capacity to store excess lipids when incubated with large amounts of non-esterified fatty acids (NEFAs), which are converted into triglycerides (TGs) and stored within lipid droplets (LDs). In myocytes, this excess caloric supply can be utilized for energy. In adipocytes, LDs sequester the TGs that are lipolyzed during fasting or insulin deficiency, allowing systemic routing of NEFAs to liver, muscles, and brown adipose tissue. Along with allowing a continued supply of ATP generating substrates, this system preserves glucose and allows production of ketones for use by the brain.

The generation of NEFAs from LD TGs also affects overall cellular metabolism. In cardiomyocytes, loss of adipose TG lipase (ATGL), the rate-limiting enzyme for LD TG lipolysis, leads to decreased activation of peroxisomal proliferator-activator receptors (PPARs) required to drive mitochondrial fatty acid (FA) oxidation and upregulate genes required for optimal NEFA uptake (1). Loss of ATGL specifically in adipocytes reduces release of NEFAs into the blood, but allows continued loss of adipocyte FAs within exosomes (2). The reduction in circulating NEFA levels leads to protection from catecholamine-induced heart failure (3), indicating that, as was suggested by Rame (4), the stress of heart failure drives a metabolic dysregulation that leads to a form of lipid-induced toxicity. In contrast, adipocyte ATGL deficiency reduces the ability of brown adipose tissue (BAT) to provide thermogenesis (5), an example of an important physiologic need for acute stress-induced NEFA induction.

LDs can indicate lipid overload toxicity or robust supplies of energy stores. The usual synthetic pathway requires several enzymes; diacylglycerol acyl transferases (DGATs) mediate the final step in TG synthesis. Overexpression of DGATs increases TG storage in macrophages (6), cardiomyocytes (7) and skeletal muscle myocytes (8). Unlike the accumulation of LDs due to other genetic mutations, DGAT1-created LDs lead to reduced inflammation and no cardiac dysfunction, perhaps because this enzyme reduces cellular content of toxic lipid species such as diacylglycerols (DAGs) and ceramides. While many models implicate greater TG content with dysfunction, greater LDs and TG metabolism with DGAT1 overexpression reduces heart dysfunction in an ischemic model (9). In contrast, DAGs and ceramides are increased with many lipid overload models and, at least in the heart, lead to cardiac dysfunction (10). Whether the benefits are due to the formation of different structural LDs, the number of LDs, or the collateral damage due to changes in distribution of cytosolic proteins or nucleotides is not known.

The article will focus on LDs and the lipases that degrade them, highlighting conceptual and experimental advances since a previous review of this topic (11). Most LDs primarily store TGs, but cells store other hydrophobic lipids by surrounding them with amphipathic proteins and lipids. These LDs—containing retinyl esters in stellate cells and cholesteryl esters in vascular macrophage foam cells—will not be discussed in detail in this review.

REGULATION OF LD FORMATION

Not surprisingly, conditions and genetic changes that lead to increased uptake, increased storage, or decreased fat oxidation expand the cellular LD content. Increased LD TGs result from greater uptake of either NEFA or lipoprotein-derived FAs, reduced local lipolysis, or greater trapping of the NEFAs by intracellular CoA synthetases. For a series of articles highlighting each of these pathway the reviewer is referred to (12).

Creation of LDs (Figure 1A) requires uptake and esterification of non-polar acyl chains and their storage within an amphipathic layer composed of proteins and phospholipids. This process primarily occurs within the endoplasmic reticulum (ER) as fatty acyl CoAs are esterified to a glycerol backbone. Aside from the enzymatic transferases, other proteins are required to allow LD assembly and are termed the lipid droplet assembly complex, reviewed in (13). Several investigators have focused on seipin, an ER protein required for LD maturation, and its structural requirements (14). A multimer of seipin allows a newly formed TG “lens” to evolve into a discreet LD (15). Seipin is also critical in cholesteryl ester LDs as loss of seipin prevents normal LD formation in adrenal glands (16).

Figure 1. Lipid droplet biogenesis.

Lipid droplet biogenesis occurs at the endoplasmic reticulum (ER), and can be triggered by excess external lipid availability or by a variety of cellular stressors. A. External lipid: Following uptake by CD36, FAs are activated by acyl-coenzyme A synthases. The resulting fatty acyl-coenzyme A are esterified into a glycerol backbone, with DGATs catalyzing the final step of TG synthesis. Newly synthesized TGs accumulate within the two leaflets of the ER, leading to the formation of a lens. A seipin multimer allows the nascent LD to fully bud from the ER. The surface of the newly formed LD is coated by a variety of proteins, the most abundant of which are members of the perilipin (PLIN) family. PLINs are involved in the regulation of LD lipolysis by modulating the access of intracellular lipases to the LD neutral lipid core and, in the case of PLIN5, tethering LDs to mitochondria and preventing FA oxidation (FAO). B. Stress: During cellular stress, LD biogenesis can occur from intracellular lipid sources. Members of the PLA₂ family cleave FA from membrane phospholipids, which are activated by ACSL. During lipid starvation, DIESL catalyzes the DGAT-independent synthesis of TGs from phospholipid-derived FA, allowing for the formation of LDs. In the context of nutrient deprivation, phosphorylated PLIN5 at the LD surface facilitates mitochondrial FAO via its interaction with FATP4. Glucose deprivation induces the CHK2α-mediated phosphorylation of PLIN2 and 3, which promotes LD lipolysis and FAO.

LD mitochondrial interactions:

Perilipin (PLIN) 5, formerly called OXPAT, was so named because it anchors LDs to mitochondria; this was presumed to increase FA oxidation. Subsequent studies, however, showed that PLIN5 overexpression blocked FA oxidation in the heart (17). While overexpression could have blocked available sites for lipases, a more recent understanding of the variation in the subcellular locations of mitochondria involved in FA oxidation has altered our understanding of LD synthesis and turnover.

Your neighborhood affects how you live and work. This is also true for mitochondria as the localization of mitochondria within the cell dictates their function. Proximity to the endoplasmic reticulum (ER) abets the creation of LDs and, as had been suggested, PLIN5 anchors the LD to these mitochondria. Najt et al. used a proteomics approach to show that in hepatocytes, mitochondria in proximity to LDs are enriched in enzymes that drive LD formation, whereas those in the cytosol are enriched with enzymes associated with FA oxidation (18). Using isotope tracing, they went on to show that fasting routed more FAs to the cytosolic mitochondria and promoted lipophagy.

Could PLIN5 operate differently in different cells? Miner et al. reported that PLIN5 is tethered to myocyte cytosolic mitochondria via interaction with FATP4 (ACSVL4) (19). They further showed that phosphorylation of the C-terminal region of PLIN5 mediates this switch to allow LD mitochondrial interaction; this association of the LD and mitochondria promoted FA oxidation in “starved” myofibroblasts.

Stress-triggered LDs:

In addition to excess lipid availability, LD biogenesis can be triggered by a variety of stressors including inflammation, hypoxia, oxidative stress, acidic pH, and nutrient deprivation (reviewed in (20)). Although the origin and biological function of stress-triggered LDs remains poorly understood, mounting clinical and experimental evidence suggests that they constitute a cell survival strategy. In stressed cells, LDs maintain energy and redox homeostasis and protect against lipotoxicity by sequestering toxic lipids. In addition, they serve as a reservoir of bioactive lipids that are involved in the regulation of inflammation and immunity (reviewed in (21)). In the heart, LD accumulation is a hallmark of ischemic but viable tissue following ischemia/reperfusion injury (22). In mice undergoing myocardial ischemia, cardiac LD depletion due to PLIN5 deficiency led to reduced heart function and increased mortality (23), (24). Consistently, decreased cardiac expression of PLIN5 in humans was associated with reduced heart function following myocardial ischemia (24).

LDs in the necrotic core of solid tumors (where poor vascularization leads to nutrient and oxygen deprivation) give rise to nuclear magnetic resonance (NMR)-visible signals used for diagnosis (25). Their presence strongly correlates with malignancy, resistance to treatment, and poor prognosis in a variety of cancers (reviewed in (26)).

In recent years, research aimed at understanding the role of lipid metabolism in cancer cell growth and survival has illuminated key differences in the biology of stress-triggered versus lipid availability-induced LDs. A recent study by McLelland et al. identified a novel pathway for TG biosynthesis that operates independently of DGATs. Using a haploid human myelogenous leukemia cell line, the authors showed that a novel acyltransferase, which they named DIESL, synthesized TG in DGAT-deficient cells. In contrast to DGATs, which synthesize TGs using acquired acyl groups, DIESL formed TGs using acyl groups cannibalized from membrane phospholipids, in a process that maintained mitochondrial function during periods of extracellular lipid starvation (27). LDs formed from membrane-derived FA have also been reported to sustain survival of glioblastoma cell lines undergoing nutrient deprivation (28). Starvation induced the group VIA phospholipase A₂ (iPLA₂-VIA), mediating release of FA from ER phospholipids. Those FA were subsequently esterified into TG, and the resulting LDs fueled beta-oxidation and promoted cell survival. Other members of the PLA₂ family have been reported to induce LD formation and promote survival of breast cancer cells to nutrient deprivation and lipotoxic stress (29), (30).

LD lipolysis also appears to be regulated differently in the context of cellular stress. As mentioned above, instead of guarding LDs from lipolysis, PLIN5 promoted LD oxidation in starved myofibroblasts. In glioblastoma cells, glucose deprivation induced the binding of choline kinase (CHK) α2 to LDs, where it phosphorylated LD proteins PLIN2 and PLIN3. The phosphorylated PLIN2/3 dissociated from LDs, thereby promoting LD lipolysis and FA oxidation. In mice, CHKα2-mediated lipolysis promoted brain tumor growth, and expression of inactive mutant forms of CHKα2 prolonged mouse survival time and decreased tumor cell proliferation with accumulated LDs (31).

Lipid droplet composition:

LDs have a monolayer of phospholipids and amphipathic proteins surround the hydrophobic lipid core. In a parallel to circulating lipoproteins, the proteins dictate the cellular position and the fate of the LDs. The most tractable LDs to isolate have been those in adipocytes and hepatocytes. They differ in their major coat proteins and, as discussed above, the proteins associated with lipid synthesis and FA oxidation. Proteomic analyses of LDs isolated by biochemical fractionation on sucrose gradients have enabled the characterization of the LD proteome composition in numerous species, cell types, and tissues, such as Chinese hamster ovary fibroblasts (32), cultured human HuH7 hepatoma cells (33), cultured human A431 epithelial cells (34), 3T3-L1 Adipocytes (35), mouse brain (36), rat cardio myocytes (37) and mouse hepatocytes (38). These studies revealed that LD-associated proteins include enzymes involved in many aspects of lipid metabolism. In addition, the LD proteins are not obviously related to lipids and include transcription factors, chromatin components, and toxic proteins (38-41). Recent proteomic studies by Bosch et al. have shown that multiple proteins related to responses to different classes of pathogens localize to LDs (42, 43). Thus, proteomic analysis of LDs has yielded a comprehensive catalog of proteins, and provided an evolutionary perspective on the organelle. One surprising LD protein is microsomal TG transfer protein (MTP). The primary function of MTP is to associate TG, phospholipid, and cholesterol with apolipoprotein B (apoB) to create chylomicrons and VLDL. Adipose specific MTP knockout mice were reported to have a reduction in adipose tissues, which appears to be due to less MTP association with the LD and increased ATGL activation (44).

Endothelial Cell LDs:

Due to their prominence and ease of isolation, much of the biology of LDs has focused on adipocytes and hepatocytes. Recently, however, several studies have investigated the formation and function of LDs within endothelial cells (Figure 2). The pioneering studies from the Sessa laboratory using confocal microscopy of the aorta showed that postprandial hyperlipidemia led to LD accumulation (45). The aorta is not a site of TG lipolysis, and subsequent studies by Cabodevilla et al. showed that aortic endothelial cells bind chylomicrons via the scavenger receptor B1 (SR-BI) (46). The internalized chylomicrons are degraded within liposomes and some of the liberated lipids are reassembled in LDs. In brown adipose from cold acclimatized mice, a different pathway leads to EC lipoprotein uptake. Large TG-emulsions undergo partial lipolysis by lipoprotein lipase (LPL) followed by uptake via the FA transporter CD36 (47). A defect in this pathway leads to reduced BAT angiogenesis.

Figure 2. Endothelial LDs.

In addition to FA uptake by CD36, endothelial cells can internalize unhydrolyzed chylomicrons via the SR-BI receptor. Once in the cytoplasm, chylomicrons are intracellularly hydrolyzed at the lysosomal compartment, and the resulting FAs are stored within LDs. The intracellular lipase ATGL is essential for endothelial LD lipolysis, and its ablation results in excessive LD accumulation that leads to defective eNOS activation and hypertension. A recent study by Peche et al. showed that internalized CD36 and the associated acyl chains cross the endothelial cells and are released from the basolateral side of the cells within small extracellular vesicles (sEV).

Do LDs alter endothelial cell function? A recent report investigated the roles of high fat diets and endothelial cell-specific ATGL deletion on LD formation and systemic phenotype. The presence of more LDs was associated with a defect in eNOS activation and hypertension (48). This was thought to be due to association of cytosolic proteins with the expanded LD.

Feeding/fasting conditions led to major differences in the exposure of capillary endothelial cells to circulating lipids. In the postprandial period, chylomicron lipolysis via the LPL reaction primarily occurs in capillaries. During fasting, NEFA levels increase due to lipolysis of TG stored within adipose LDs, and these NEFA are internalized by endothelial cells via CD36 (49). Consistent with the primary use of glucose, changes in endothelial cell NEFA uptake do not appear to alter its metabolism but lead to transcytosis. Using click chemistry to anchor NEFAs to CD36, Peche et al. showed that the internalized CD36 and the associated acyl chains cross the endothelial cells and are released from the basolateral side of the cells within small extracellular vesicles (50).

LD DEGRADATION

LDs can be reduced or eliminated by two major pathways: lipolysis via cytoplasmic lipases (ATGL, HSL, monoglyceride hydrolase) and lipophagy that occurs when the LD is captured by the autophagy pathway.

Regulation of intracellular lipases:

ATGL is the rate-limiting enzyme required for lipolysis of LD TG. This enzyme primarily lipolyzes TG at the sn2 position (51); although acyl chain migration can occur, the 1,3 DAG is not a PKC-activating lipid. ATGL has a number of activating and inhibiting proteins. Comparative gene identification 5 (CGI-58, also known as ABHD5) is the most characterized ATGL-activating protein. Several inhibiting proteins also exist, and the structural requirements for binding of these inhibitors G0/G1 switch gene 2 (G0S2) and hypoxia-inducible gene 2 (HIG2) and their requirement for ATGL to associate with the LD differs (52).

Lipophagy:

Since its discovery by Singh et al. in 2009 (53), the autophagic degradation of LDs (termed lipophagy) has been extensively studied and shown to play a critical role in lipid homeostasis. During lipophagy, LDs are fully or partially engulfed by de novo synthesized autophagosomes that fuse with lysosomes, where LD TG and cholesterol are hydrolyzed by acidic lipases (54). In the liver, where it has been most widely studied, lipophagy imbalance always leads to a perturbation in lipid metabolism and excessive LD accumulation, with recent data suggesting that the pathology of non-alcoholic fatty liver disease (NAFLD) is associated with impaired lipophagy (55), (56). As is the case with lipolysis, lipophagy is tightly regulated by LD proteins, including members of the PLIN family. PLIN5 has been shown to regulate hepatic autophagy during fasting. Upon PKA-mediated phosphorylation, PLIN5 translocates to the nucleus where it interacts with SIRT1, promoting the transcription of autophagy-related genes. Overexpression of PLIN2 has been shown to protect hepatic LDs from autophagic degradation, whereas its ablation reduced LD content in mouse livers by enhancing lipolysis (57). In contrast, PLIN2 deficiency led to impaired lipophagy, reduced LD-lysosome association, and increased LD accumulation in mouse cardiomyocytes. This effect was not observed in cardiac fibroblasts or liver of PLIN2-deficient mice, which in fact exhibited reduced LD content, suggesting that the role of PLIN2 in promoting LD autophagic degradation is cardiomyocyte-specific (58).

In recent years, a series of studies has established a protective role for macrophage lipophagy in atherosclerosis, identifying autophagy as a potential therapeutic target. In macrophages, atherogenic lipoproteins (oxidized and aggregated LDL) led to the tagging of LDs with ubiquitin and the recruitment of autophagy markers to the LD surface. Inhibition of autophagy or lysosomal acidic lipase (LAL) markedly reduces LD catabolism and reverses cholesterol transport in vivo and in vitro. Macrophage-specific deletion of the central autophagy regulator Atg5 in Ldlr^−/− or Ape^−/− mice exacerbated atherosclerosis and resulted in increased accumulation of lipids in plaques (reviewed in (59)). A recent study using mass spectrometry to identify lipophagy factors within the macrophage foam cell LD proteome yielded 37 candidate genes, many of which were dysregulated in foam macrophages from lesions of Apoe^−/− mice fed a high cholesterol diet (60). Taken together, these findings suggest that enhancing lipophagy could be a valuable tool in the treatment of atherosclerosis. A way to do this was reported in 2021 when the Lu laboratory developed a series of autophagy-tethering compounds (ATTC) designed to interact with both LDs and the key autophagosome protein LC3. These compounds were capable of clearing LDs in cultured mouse embryonic fibroblasts (MEF) as well as in two independent mouse models for obesity and non-alcoholic steatohepatitis (NASH) (61).

Neutral LD storage disease:

The most dramatic expansion of intracellular LDs in the heart occurs with ATGL deficiency. Presumably, the exuberant metabolism of lipids by mouse cardiomyocytes makes the heart most susceptible to loss of this enzyme. Although humans also develop a cardiomyopathy with ATGL deficiency (62), the most dramatic symptoms in these patients are pain and weakness associated with skeletal muscle breakdown; circulating levels of the muscle enzyme creatine kinase (CK) are increased 10-fold (63). A recent study attempted unsuccessfully to alleviate the cardiomyopathy due to ATGL deficiency by deletion of LPL or CD36 (64). In agreement with these observations, ATGL-deficient cardiomyocytes develop LDs from de novo fatty acid synthesis when grown in lipid-depleted media.

The phospholipid monolayer of LDs harbors membrane-associated proteins that regulate LD functions. Proteins embedded within membranes are extremely diverse in structure and function. Growing evidences suggest that a unique complement of integral and peripheral proteins associates with the LD phospholipid monolayer, and significant progress has been made in understanding the specific lipid pathways that coordinate LD biogenesis and degradation (65). Current models describe LD functional states as being largely defined by their specific proteome (66), and therefore the proteomic techniques have been used to explore biological functions of LD membranes.

SUMMARY

Research into intracellular neutral lipid storage has led to major advances in a number of fundamental areas: cellular uptake of NEFAs, creation of LDs, and release of acyl groups from the LDs. These processes, in turn, regulate cellular proliferation, energetics, inflammatory reaction, and cell survival under stress. We are learning that the LD has a more global role as its surface can be a competing membrane to attract proteins with hydrophobic domains. These studies will continue to illuminate the relationships of lipids and disease of the liver, heart, adipose and circulating lipoproteins.

Bullet point summary:

• Lipid droplets affect the overall biology of cells.

• New data illustrate the molecular machinery regulating lipid droplet formation and lipolysis.

• Alterations of lipid droplet proteins control storage versus oxidation of cellular fatty acids.

• This evolving biology will provide new understands of ways to prevent and treat lipotoxic diseases including metabolic fatty liver disease, type 2 diabetes, and some forms of heart failure.