The multifunctional role of hormone-sensitive lipase in lipid metabolism

Pergamon

Advan. Enzyme Regul. Vol. 34, pp. 355-37(I, 1994 Copyright ~) 1994 Elsevier Science Ltd Printed in Great Britain. All rights reserved 0065-2571/94/$26.00

0065-2571 (93) E0009-D THE MULTIFUNCTIONAL ROLE HORMONE-SENSITIVE LIPASE LIPID METABOLISM

OF IN

S T E P H E N J. Y E A M A N , G A B R I E L E M. SMITH, C A T H E R I N E A. J E P S O N , S T E V E N L. W O O D and N E l L E M M I S O N D e p a r t m e n t of Biochemistry and Genetics, The Medical School, University of Newcastle upon Tyne NE2 4HH, U.K. INTRODUCTION

Adipose tissue triacylglycerol is quantitatively the most important energy store in mammals. Hydrolysis of this triacylglyceroi is tightly controlled by a variety of hormonal and neural influences allowing release of this energy to be regulated in response to changing energy requirements of the animal. The key enzyme in this release (i.e., lipolysis) is hormone-sensitive lipase (HSL) (EC 3.1.1.3). HSL catalyzes the rate-limiting step in the breakdown of triacylglycerol and its activity is regulated acutely via reversible phosphorylation. HSL is phosphorylated and activated in response to a variety of lipolytic hormones, with insulin exerting a major anti-lipolytic action (1). HSL was long considered to be an adipose-specific enzyme possessing a major activity of hydrolyzing triacyiglycerol. A significant advance however was the successful purification of the enzyme and the observation that HSL, in addition to its activity against triacylglycerol, also has a significant activity against diacylglycerol, monoacylglycerol and against long chain esters of cholesterol (2). Indeed, the activity against cholesterol esters is approximately equal to that against triacylglycerol, at least in vitro. Subsequent to this observation, evidence was obtained that HSL is present, not only in adipose tissue, but also in tissues in which cholesterol esters are used as a storage form of cholesterol, to be released for utilization in steroid hormone biosynthesis (3). In particular, enzymological and immunological data lead to the realization that HSL is expressed and active in adrenal cortex and ovaries, where the activity is regulated acutely by the appropriate hormone controlling the pathway of steroidogenesis (4, 5). Subsequently, HSL has also been shown to be present in several other tissues, including heart and muscle (6, 7), but perhaps most importantly in macrophages (8). In this tissue it apparently acts as a cholesterol ester hydrolase and this, as discussed below, has important implications in atherogenesis.

355

356

S.J. YEAMAN et al. MATERIALS

AND METHODS

HSL was purified from bovine perirenal adipose tissue, essentially as in (9), as modified in (10). Cyclic AMP-dependent protein kinase was purified from bovine heart as in (11) and the catalytic subunits of protein phosphatases 1 and 2A from rabbit skeletal muscle (12). AMP-activated protein kinase from rat liver was a generous gift from the laboratory of Dr D. G. Hardie, University of Dundee.

RESULTS

AND DISCUSSION

Phosphorylation of HSL From a combination of work using isolated rat adipocytes and purified HSL, protein kinases and phosphatases, it has been established that there are two phosphorylation sites on the HSL polypeptide, which are phosphorylated in vitro and in vivo (1, 13). Site 1 (the regulatory site) is responsible for the activation of HSL which occurs in response to lipolytic stimuli (13). It has been shown that this site is phosphorylated by the cyclic AMP-dependent protein kinase (14, 15) and in conjunction with cloning studies (16) and phosphopeptide sequencing (15) this serine residue has been identified as position 563 in the rat HSL sequence. A second phosphorylation site has also been identified on HSL. It is termed Site 2 or the basal site, indicating that it is phosphorylated within adipocytes under basal conditions of lipolysis (13). Several protein kinases have been shown to act on this site in vitro, the most likely of which to be of physiological importance being the AMP-activated protein kinase (17). The basal phosphorylation site has been located at position 565 in the rat HSL sequence. An interesting observation resulting from mapping studies on HSL phosphorylated in vitro is that the two sites are mutually exclusive, i.e., phosphorylation of Site 1 blocks subsequent phosphorylation of Site 2 and vice versa (17, 18). It has yet to be established whether this also occurs in vivo but the implication is that, if this does occur, Site 2 can exert an anti-lipolytic action in that its phosphorylation will prevent activation of HSL via phosphorylation at Site 1. It has been shown that both sites are phosphorylated in vivo but it is not yet known which factors affect the phosphorylation state of Site 2 and whether this state alters, either acutely or chronically. Dephosphorylation of HSL HSL is a potential target for regulation by protein phosphatases and there is evidence that activation of protein phosphatases may be involved in the anti-lipolytic effect of insulin (19). In eukaryotic cells four protein phosphatase (PP) catalytic subunits, termed PP1, PP2A, PP2B and PP2C

HORMONE-SENSITIVELIPASE

357

account for most of the Ser/Thr phosphatase activity towards substrates examined to date (20). HSL can be dephosphorylated in vitro by PP1, PP2A and PP2C, but not by the Ca2+-sensitive phosphatase PP2B (calcineurin) (21). PP2A and PP2C dephosphorylate HSL in vitro with rates comparable to those against glycogen phosphorylase and phosphorylase kinase ((x-subunit) respectively, whereas in comparison PP1 exhibits a relatively low HSL phosphatase activity (21). Preferential dephosphorylation of the basal site of HSL is observed in vitro, PP1 displaying a 20% higher activity towards the basal site than the regulatory site, PP2A and PP2C being 80% more active towards the basal site. A major role for PP2A in the dephosphorylation of HSL has been proposed (21) by combining the measured specific activities of the purified protein phosphatases towards HSL with the relative levels of the four phosphatases within adipose tissue (22). However, this suggestion may have limited physiological relevance as it is based on phosphatase specific activity measurements obtained using non-physiological levels of purified HSL and phosphatases. Furthermore, scapula fat (which is partially brown adipose tissue) was used as the source of the adipose tissue data for this study (22) and this may not reflect the situation relevant to the regulation of HSL within white fat cells. A more recent study (10) using the selective protein phosphatase inhibitor okadaic acid (23) to quantify the HSL phosphatases present within extracts of freshly isolated adipocytes suggested that PP2A is a major phosphatase for both sites of HSL (48 and 59% of the total phosphatase activity towards the regulatory and basal phosphorylation sites, respectively). No significant PP2B activity was detected towards either site of HSL, whilst, in contrast with the conclusions of the previous study (21), PP2C displayed a high activity towards both sites (50 and 32% of the total phosphatase activity towards the regulatory and basal sites of HSL, respectively). The significant PP2C activity towards HSL may allow co-ordinate regulation of HSL with other enzymes involved in regulating lipid metabolism, e.g., HMG-CoA reductase (24) and AMP-activated protein kinase (25), for which PP2C is also a significant phosphatase activity. PP1 within adipocyte extracts displays a great preference for the basal site of HSL against which it has a 30-fold greater activity compared to the regulatory site and it is possible that marked site preference of PP1 may enable the selective dephosphorylation of the regulatory site of HSL in vivo. Preference for basal site dephosphorylation is observed with all adipocyte phosphatases, PP2A and PP2C dephosphorylating the basal site with a 2-3 fold higher rate than the regulatory site (a greater discrepancy than observed with purified protein phosphatases). Studies involving the hormonal pretreatment of isolated adipocytes may further help elucidate the role of the individual protein phosphatases in the regulation of HSL activity.

358

s.J. YEAMANet al.

Domain Structure o f Hormone-Sensitive Lipase

As a result of cDNA cloning and sequencing studies it has been established that several mammalian lipases are members of a gene family. These members include lipoprotein lipase, hepatic lipase and pancreatic lipase (26). However, from the predicted protein sequence of both rat (16) and human HSL (27), it is clear that HSL is not a member of that family, although it does show some homology with a lipase from an antarctic bacterium. Properties of HSL which distinguish it from these lipases include its regulation by reversible phosphorylation and its relatively broad substrate specificity, particularly the high activity against cholesterol esters (2) (bile-salt stimulated lipase has some activity against cholesterol esters but this is only 1% of that against triacylglycerol (28)). Considerable structural information is available for several lipases. The recent determination of human pancreatic lipase (PL) crystal structure has provided an important framework for more directed approaches to lipase structure/function studies. PL is made up of two well-defined domains; a large N-terminal catalytic domain corresponding to 66% of the molecule and a small C-terminal domain (29). The crystal structures for two fungal lipases have also been solved and, although apparently unrelated in sequence, they have a strong structural homology to the PL N-terminal domain, which includes the [3-pleated sheet structure and a catalytic triad similar to that of serine proteases (30, 31). These surprising structural similarities between otherwise unrelated proteins may establish a general structure for the entire lipase gene family and certain features are likely to extend to all lipases (reviewed in 32, 33). This may include HSL whose primary sequence failed to reveal any significant homology with either the members of the lipase gene family (26) or any other lipase for which structural information is available (16). Thus, the sequence revealed little in terms of the structural organization of the HSL enzyme protein and has clearly limited our understanding of the molecular basis of its many unique properties. In addition to its lipid substrates, HSL also has significant esterase activity against the water-soluble p-nitrophenyl butyrate (PNPB) (34). Limited tryptic digestion of native bovine HSL changes the kinetic properties of the enzyme; the activity against lipid substrates is dramatically reduced with increasing trypsin concentration whilst activity against PNPB is relatively trypsin-insensitive (Fig. 1). This retention of esterase activity following tryptic digestion indicates that a major catalytic component of the polypeptide chain must remain intact under these conditions. The presence of regions of the HSL protein which are resistant to cleavage upon proteolytic digestion of native protein suggests separately folded domains joined by short, protease-sensitive sequences, consistent with lipase crystallographic data. These results, in which mild proteolysis

HORMONE-SENSITIVE LIPASE

359

1 O0 -eL

80-

.m ",~. 0

60 -

m 0 t+.,

O

40-

20

I

-

l

0t

lo

2o

-t

3o

ao

so

Trypsin concentration (pg/ml) FIG. 1. Effect of limited tryptic digestion on the lipase and esterase activities of bovine HSL. HSL was subjected to proteolytic digestion at pH 7.0 for 15 min at 25°C with addition of 0.2 vol of the indicated concentration of trypsin. Digestion was terminated by the addition of an equal vol of trypsin inhibitor (100/~g/ml). Aliquots were then assayed in duplicate against tri[aH]oleoylglycerol(0) and PNPB (11). Results are expressed as a percentage of the activity observed in the absence of trypsin and are means + SEM (n = 5).

differentially affects the esterase activity and the lipolytic activities of HSL, confirm the findings in (34) and are similar to those reported for a n u m b e r of other lipases including lipoprotein lipase (35), hepatic lipase (36, 37) and gastric lipase (38). Investigation of the effect of such limited tryptic digestion on the enzyme protein revealed that it is possible to cleave H S L into several fragments. Subsequent [3H]DFP-labelling of any digestion fragments retaining catalytic activity revealed the presence of a stable domain of approximately 17.6 k D a which contains the active site serine residue (Fig. 2). The production of this stable band corresponded to complete loss of triacylglycerol hydrolase activity whilst almost 80% of the PNPB hydrolase activity remained. Therefore, the active site of the enzyme still remained in an active conformation, able to catalyze the hydrolysis of a water-soluble substrate and react with DFP. Hence, we have proposed

360

s.J. YEAMAN et al.

that this fragment represents a catalytic domain within the HSL enzyme protein (39). N-terminal sequence analysis of this putative catalytic domain has led to its identification as between residues 333--499 of rat HSL. The C-terminus of this domain is identified solely on the basis of its molecular weight. Interestingly, this putative catalytic domain of HSL contains the GXSXG motif (residues 421-425 of rat and human HSL) which is conserved in all lipases, and is also found around the essential serine of esterases and proteinases (32). There has been a great deal of controversy over whether this pentapeptide represents part of a lipid binding or active site sequence in lipases. However, the crystal structures of a number of lipases have confirmed that the Ser within this motif is part of the catalytic triad, and a direct covalent bond has been identified between this serine and a substrate analogue (40). Consistent with this, the ability of the bovine HSL fragment described here to hydrolyze PNPB but not triacylglycerol indicates that the serine residue (Ser423) in this sequence represents the active site serine of HSL rather than being involved in lipid binding. Furthermore, the sequence surrounding this Ser in human HSL is predicted to be in a structure resembling the 13-eSer-ot motif (27) which is proposed to surround the catalytic serine residue of triacylglycerol lipases and esterases (41). One characteristic feature of lipolytic catalysis is the substantial increase in activity in the presence of a lipid-water interface, a phenomenon known as interracial activation. The structural basis for this interfacial activation has been revealed by the crystal structures of three neutral triacylglycerol lipases (reviewed in (33)). The catalytic triad of these enzymes is buried within the protein molecule protected by a helical lid and is therefore inaccessible to the surrounding solvent. Following binding of the enzyme to a lipid interface, this lid is displaced to reveal the active site, a movement which greatly enlarges the non-polar surface at the active surfaces and buries previously exposed polar residues. The identification of a lid structure in a number of unrelated lipases implies that the general stereochemistry of lipase activation at the lipid-water interface is likely to apply to the entire family of lipases. The conformational change induced by interfacial activation exposes the active site and thereby increases the catalytic power for a monomeric substrate. Such an enhancement of esterase activity by interaction of the enzyme with a lipid interface has been shown for a number of lipases including hepatic lipase (42) and lipoprotein lipase (43). Consistent with the findings for these other triacylglycerol lipases, we have shown that the HSL-catalyzed hydrolysis of the water-soluble substrate PNPB is enhanced by the addition of phospholipid vesicles or emulsified trioleoylglycerol (Smith and Yeaman, unpublished). This may be an indication that HSL possesses a comparable lid structure covering the catalytic triad

-1706kD

0 O-5 1

2

4

6

lb 50

TRYPSJN CONCENTRATION (pg/mll FIG. 2. Effect of limited tryptic digestion on the active site of HSL. HSL was subjected to limited digestion with the indicated concentrations of trypsin as described in Figure I. The resultant fragments were incubated with [‘H]DFP for 30 min at 30°C. The fragments were then separated by SDS-PAGE, electroblotted and visualized by direct autoradiography.

84kD

TRYPSIN CONCENTRATION (pg/ml) FIG. 3. Limited tryptic digestion of HSL phosphorylated at Site I. HSL was phosphorylated by incubation for IS min at 30°C with [+P]ATP-Mg and the catalytic subunit of cyclic AMP-dependent protein kinase. jzP-HSL was subjected to limited tryptic digestion with the indicated concentrations of enzyme, as described in Figure 1. and the radiolabelled fragments were visualized by SDS-PAGE and autoradiography.

HORMONE-SENSITIVE LIPASE

361

residues which is displaced upon interaction with a lipid-water interface. In comparison, the esterase activity of trypsin-treated HSL is not enhanced by the presence of a lipid-water interface, a result similar to that found for digested hepatic lipase (36). Therefore, in addition to losing lipid hydrolyzing activity following tryptic digestion, the ability of a lipid interface to activate the esterase activity of HSL is lost. These results suggest that HSL has a lipid binding site that is sensitive to proteolytic digestion and is distinct from the catalytic site which remains intact and capable of hydrolyzing water-soluble substrates. As previously mentioned, HSL is regulated by reversible phosphorylation which occurs at two distinct sites within the enzyme protein. These phosphorylation sites, termed regulatory (Site 1) and basal (Site 2), have been identified as Ser563 and Ser565 respectively in rat HSL (18). Limited tryptic digestion of 32p-HSL labelled at Site 1 generates a stable domain of M r approximately 11.5 kDa which contains both phosphorylation sites (Fig. 3). This fragment may represent a regulatory domain within the HSL enzyme protein. Exact localization of this domain within the HSL enzyme protein by N-terminal sequencing has not yet been possible; however, evidence suggests that the catalytic domain is produced together with this domain, so locating the region of this N-terminus to between residues 499-563. Native HSL phosphorylated at Site 1 can therefore be digested by trypsin to produce a catalytic domain containing the active site residue and a regulatory domain containing the phosphorylation sites. Attempts to identify this sequence may have been complicated by the observation that the most heterologous region between the rat HSL (16) and the recently determined human HSL sequence (27) is present immediately upstream of the phosphorylation sites. A 12 amino acid deletion is observed in human HSL, flanked by a number of non-conserved residues. This deletion in human HSL modifies the HSL secondary structure by deleting a connecting loop present in rat HSL between a hydrophilic region and the a-helix containing the phosphorylation sites which is not found in human HSL (27). Furthermore, the deleted sequence is encoded by the 3' end of exon 7 whereas the phosphorylation sites are encoded by exon 8. Therefore, this deletion in human HSL occurs in a region that could represent a link between two structural domains of the protein and so may represent the N-terminal region of the regulatory domain. In contrast to limited tryptic digestion of site 1 phosphorylated HSL, no phosphorylated domain was generated following digestion of 3zP-HSL phosphorylated at site 2, instead the label was recovered in a small phosphopeptide (Smith and Yeaman, unpublished). The 17.6 kDa catalytic domain was, however, generated following digestion of HSL phosphorylated at site 2. Thus, a conformational difference must exist between the two phosphorylated forms of the enzyme. The conformation

362

S.J. YEAMANet al.

induced by phosphorylation of HSL at site 2 may "destabilize" the regulatory domain by bringing the region surrounding the phosphorylation sites into a conformation which is highly susceptible to proteolysis. We have provided evidence which indicates that HSL is made up of at least three functional domains; a regulatory domain which contains the phosphorylation sites which control activity, a catalytic domain which contains the active site responsible for catalytic activity and a lipid-binding domain which contains the site responsible for anchoring the lipase at the lipid-water interface. HSL may have a similar structural organization to that of other lipases which includes a catalytic triad, superimposable in all lipases and related proteins so far identified and in serine proteases, and a central 13-sheet structure which seems to serve as a scaffold for this triad. The larger size of HSL compared to other lipases may reflect a more complex domain structure which is essential to cope with the broad substrate specificity and regulation by reversible phosphorylation. The catalytic and regulatory domains identified constitute only approximately 35% of the total enzyme protein; the remainder presumably incorporates structural elements necessary for recognition and binding of the various lipid substrates hydrolyzed by HSL. Role of HSL in Foam Cell Development in Atherogenesis Atherosclerosis is the commonest form of arterial disease, characterized by the thickening, loss of elasticity and obstruction of arteries following formation of fatty plaques in the intima of the vessels. The resulting impaired blood supply to the tissues increases the occurrence of thrombosis, coronary heart disease, stroke and several other diseases of middle and old age (44). The fatty plaques of atheroma contain lipid-laden foam cells which are derived, at least in part, from macrophages which have become overloaded with cholesterol esters (44). The atherosclerotic plaque has a complex structure (45); however, the component of the lesion that is chiefly responsible for its pathogenicity is the cholesterol esters present in foam cells (46). Clearly the accumulation of cholesterol esters in macrophages has serious implications and an understanding of the control of cholesterol ester metabolism in these cells is of great importance in the prevention and management of atherosclerosis. Macrophages accumulate large amounts of cholesterol ester by uptake of plasma lipoproteins that have leaked through damaged endothelium and penetrated into the arterial intima. Atherogenesis is associated with elevated plasma concentrations of low density lipoprotein (LDL), which is the principal lipoprotein responsible for the transport of cholesterol to the peripheral tissues (44). Macrophages possess too few receptors for

HORMONE-SENSITIVE LIPASE

363

native LDL for uptake of this lipoprotein to lead to foam cell formation; however, they express a high affinity "scavenger" receptor for modified LDLs (47). Unlike the LDL receptor expressed by other cell types, the macrophage scavenger receptor is not subject to down-regulation, facilitating accumulation of vast quantities of cholesterol ester in the cell. Lipoproteins internalized via the scavenger receptor are delivered to macrophage lysosomes where the protein and cholesterol ester components are hydrolyzed (47, 48). The liberated cholesterol is released into the cytoplasm and is either re-esterified by microsomal acyl CoA:cholesterol acyl transferase (ACAT), or released from the cell if an external acceptor such as high density lipoprotein (HDL) is present (49). Under normal conditions, cholesterol esters stored in macrophages undergo a constant cycle of hydrolysis and re-esterification, catalyzed by a cytoplasmic cholesterol ester hydrolase and ACAT respectively. If the rate of esterification exceeds that of hydrolysis, cholesterol esters will accumulate and foam cells develop (48). The relative activities and regulation of the enzymes involved in this "cholesterol ester cycle" therefore play an important role in the pathogenesis of atherosclerosis. We have previously demonstrated that hormone-sensitive lipase (HSL) is solely responsible for the neutral hydrolysis of cholesterol esters in macrophages, by the selective and total inhibition of the neutral cholesterol ester hydrolase activity in soluble extracts of the mouse macrophage cell-line WEHI-3b and mouse peritoneal macrophages by antibodies raised against purified HSL (8). The anti-HSL antibodies also specifically immunoprecipitated from macrophage homogenates a phosphoprotein of the same molecular weight and with several properties identical to HSL (8). We also demonstrated that endogenous HSL is phosphorylated in intact macrophages (50), by immunoprecipitation of HSL from cells pre-incubated in medium containing [32p]-phosphate and treated with the protein phosphatase inhibitor okadaic acid, or with agents which increase intracellular levels of cyclic AMP (such as the phosphodiesterase inhibitor, 3-isobutyl-l-methylxanthine and the cyclic AMP analogue dibutyryl cyclic AMP). Treatment of macrophages with these agents also resulted in an increase in neutral cholesterol ester hydrolase activity, consistent with phosphorylation of HSL enzyme protein by cyclic AMP-dependent protein kinase (50). To investigate the role of HSL in foam cell formation, we have generated foam cells from WEHI-3b macrophages by incubation with a commerciallyavailable bovine lipoprotein-cholesterol concentrate (BLCC). Lipid accumulation in these cells was confirmed qualitatively by staining the cells for neutral lipids with oil red O and quantitatively by HPLC analysis of intraceUular free cholesterol and cholesterol esters. The activity of HSL was found to be greatly reduced in BLCC-treated macrophages, with a total loss

364

s.J. YEAMANet al.

of enzyme activity in cells maximally loaded with cholesterol esters (Jepson and Yeaman, unpublished). This loss of HSL activity contributes to the imbalance in the "cholesterol ester cycle" which leads to lipid accumulation in the macrophages. Possible mechanisms by which the activity of HSL might be reduced in macrophage foam cells were subsequently examined. As previously mentioned, a key feature of HSL is its ability to be activated by phosphorylation by cyclic AMP-dependent protein kinase (14); however, the diacylglycerol hydrolase activity of the enzyme is apparently unaffected by this modification (51). The possibility that reduction of HSL activity in foam cells results from dephosphorylation of the enzyme protein was therefore excluded when a complete loss of HSL activity against the diacylglycerol analogue (3)-mono-oleoyl-2-O-oleylglycerol was observed in macrophages maximally loaded with cholesterol esters. In addition, the large reduction in HSL activity measured in BLCC-treated cells could not be accounted for by the relatively small changes in enzyme activity which result from phosphorylation effects measured under in vitro assay conditions (2, 5). We have previously determined that HSL activity is inhibited by various lipid metabolites, including fatty acyl CoAs (52); since ACAT generates cholesterol ester by the esterification of free cholesterol to fatty acyl CoA, the increased levels of free cholesterol observed in foam cells may be accompanied by increased levels of fatty acyl CoA to supply the esterification reaction. However, no significant change in the levels of fatty acyl CoA in foam cells as compared to control macrophages was observed. It therefore seems unlikely that the reduction of HSL activity in foam cells is due to feedback inhibition by fatty acyl CoA (9). As mentioned, levels of free cholesterol also increase in foam cells; however, cholesterol has no effect on HSL activity (52). Recently, we have observed that WEHI-3b macrophage cytosol and its acetone-diethyl ether powder inhibit both purified bovine HSL and HSL in WEHI-3b macrophage extracts (Jepson and Yeaman, unpublished). Inhibition is both heat- and trypsin-sensitive, indicating the probable involvement of a protein component. The inhibitor appears to be relatively specific for HSL, having no effect on the cholesterol esterase activities of rat liver cytosol and P s e u d o m o n a s fluorescens; this suggests that the inhibitor does not act by simply interacting with the lipid substrate. Cytosol prepared from macrophage foam cells has a significantly greater inhibitory effect than that prepared from control cells and this increase in inhibitory activity correlates with loss of HSL activity, suggesting that the reduction in HSL activity observed in cholesterol ester-laden macrophages is related to the corresponding increase in activity of the cytosolic inhibitor protein (9). Although it appears to have a major effect, it is not at present clear whether the inhibitor protein is solely responsible for the loss of HSL

HORMONE-SENSITIVELIPASE

365

activity observed in foam cells. However, as the inhibitor protein appears to have an important role in the development of macrophage foam cells it has the potential to be targeted in anti-atherogenic therapy; this will first require the purification of the protein and full characterization of its mode of action and regulation. Another possibility is that cholesterol ester accumulation also affects the level of HSL polypeptide in macrophages; however, this remains to be fully investigated. In preliminary studies, we have identified a 3.3 kb species in cytoplasmic RNA preparations from control WEHI-3b macrophages using a 1.9kb cDNA probe corresponding to nucleotides 595-2489 of rat HSL DNA. This corresponds to the size of HSL mRNA identified in several other tissues (7). However, further work must be performed to determine whether the expression of HSL is reduced in macrophage foam cells. Possible Role of Translocation in the Regulation of HSL When adipocytes in the basal or resting state are maximally stimulated by lipolytic agents the observed rate of lipolysis increases 50 to 100 fold (53). However, when non-phosphorylated HSL is phosphorylated stoichiometrically in vitro using cyclic AMP-dependent protein kinase, a procedure which should fully maximize HSL activity, the observed increase in HSL activity is routinely in the order of 2- to 3-fold (2, 4, 14). The apparent discrepancy between the activation observed in the isolated adipocyte and using the purified enzyme has been explained as being due to the fact that the assay of the purified enzyme is carried out under non-physiological conditions using artificial substrates and hence it is not expected that the physiological activation will be observed (54). Recently, however, two pieces of evidence have indicated a possible explanation to account for this discrepancy. Firstly, evidence has been presented that when HSL is activated within the fat cell in response to lipolytic stimulation, this is associated with a translocation of the HSL polypeptide from the cytoplasmic compartment to the fat globule (55, 56). The implication of this is that phosphorylation not only activates HSL but also targets it towards the triacylglycerol substrate. Independent work has also studied the phosphorylation of fat droplet associated proteins in response to lipolytic hormones. In particular, a protein of M r approximately 65 kDa has been identified whose phosphorylation and dephosphorylation in response to lipolytic hormones parallels that of HSL (57, 58). It has been hypothesized that this protein, termed perilipin, may serve an anchoring role and that, in response to iipolytic hormones, perilipin becomes phosphorylated and the phosphorylated protein then binds HSL and localizes it to the triacylglycerol droplet. However, direct evidence in support of this hypothesis is currently lacking but clearly this will be an important area for future study. In

366

S.J. YEAMANet al.

particular, it will be of interest whether perilipin, despite being reported to be adipocyte-specific (58), is co-expressed with HSL in tissues other than adipose tissue. SUMMARY Hormone sensitive lipase (HSL) is an enzyme of relatively broad specificity, having the ability to hydrolyze tri-, di- and mono-acylglycerols as well as cholesterol esters and small water-soluble substrates. This broad specificity allows HSL to perform a variety of functions in several tissues. A key feature of HSL is its ability to be activated via phosphorylation by cyclic AMP-dependent protein kinase. In addition it is phosphorylated at a second site by several kinases, notably AMP-activated protein kinase. Phosphorylation of this site apparently plays a role in rendering the enzyme hormone-insensitive, in that prior phosphorylation at site 2 prevents phosphorylation and activation at site 1 by cyclic AMP-dependent protein kinase. Investigation of the protein phosphatases responsible for dephosphorylation of these sites has indicated that phosphatase 2A plays a predominant role but also that protein phosphatase 2C is a significant phosphatase targeted against both phosphorylation sites. Evidence indicates that HSL has at least three functional domains which contain (a) the phosphorylation sites which control activity, (b) the active site responsible for the catalytic activity and (c) a lipid binding site responsible for anchoring the lipase at the water-lipid interface. Using limited proteolytic studies we have found that it is possible to cleave HSL into several fragments including a stable domain of M r approximately 17.6 kDa which contains the active site serine residue. Digestion under similar conditions also generates a stable domain o f M r approximately 11.5 kDa containing both phosphorylation sites. Furthermore, under appropriate conditions it is possible to digest HSL and retain activity against water-soluble substrates but with the concomitant loss of activity against triacylglycerol, implying that a lipid binding domain is lost during this procedure. HSL is responsible for the neutral cholesterol esterase activity in macrophages and it may play a role in the accumulation of cholesterol esters which occur during the development of foam cells. HSL activity is reduced in macrophage foam cells, at least partly due to increased activity of a cytosolic HSL inhibitor protein. A finding unexplained for many years has been that, although lipolysis can be stimulated 50-100-fold in adipocytes by lipolytic hormones, HSL can apparently only be activated 2-3-fold via phosphorylation in vitro by cyclic AMP-dependent protein kinase. One possibility to explain this discrepancy is that an additional anchoring protein is missing from the in vitro system and indirect evidence is now accumulating for such a protein.

HORMONE-SENSITIVE LIPASE

367

ACKNOWLEDGEMENTS W o r k in t h e a u t h o r s ' l a b o r a t o r y h a s b e e n s u p p o r t e d b y g r a n t s f r o m t h e M e d i c a l R e s e a r c h C o u n c i l , U , K . a n d t h e B r i t i s h H e a r t F o u n d a t i o n a n d by studentships from the MRC and the Science and Engineering Research C o u n c i l , U . K . W e t h a n k D r A . J. G a r t o n f o r his i n v a l u a b l e c o n t r i b u t i o n to m a n y a s p e c t s o f t h e s e s t u d i e s . REFERENCES 1. S. J. YEAMAN, Hormone-sensitive lipase - - a multipurpose enzyme in lipid metabolism, Biochim. Biophys. Acta 1052. 128-132 (1990). 2. G. FREDRIKSON, P. STRALFORS, N. 0. NILSSON and P. BELFRAGE, Hormone-sensitive lipase of adipose tissue. Purification and some properties, J. Biol. Chem. 256, 6311-6320 (1981). 3. K . G . COOK, F-T. LEE AND S. J. YEAMAN, Hormone-sensitive cholesterol ester hydrolase of bovine adrenal cortex, FEBS. Lett. 132, 10-14 (1981). 4. K. G. COOK, S. J. YEAMAN, P. STRA,LFORS, G. FREDRIKSON and P. BELFRAGE, Direct evidence that cholesterol ester hydrolase is the same enzyme as hormone-sensitive lipase from adipose tissue, Eur. J. Biochem. 125, 245-249 (1981). 5. K. G. COOK, R. J. COLBRAN, J. SNEE and S. J. YEAMAN, Cytosolic cholesterol ester hydrolase from bovine corpus luteum, Biochim. Biophys. Acta 752, 46-53 (1983). 6. C. A. SMALL, A. J. GARTON and S. J. YEAMAN, The presence and role of hormone-sensitive lipase in heart muscle, Biochem. J. 25g, 67-72 (1989). 7. C. HOLM, P. BELFRAGE and G. FREDR1KSON, Immunological evidence for the presence of hormone-sensitive lipase in rat tissues other than adipose tissue, Biochem. Biophys. Res. Commun. 148, 99-105 (1987). 8. C . A . SMALL, J.A. G O O D A C R E and S. J. YEAMAN, Hormone-sensitive lipase is responsible for the neutral cholesterol ester hydrolase activity in macrophages, FEBS Left. 247, 205-208 (1989). 9. S. R. CORDLE, R. J. COLBRAN and S. J. YEAMAN, Hormone-sensitive lipase from bovine adipose tissue, Biochim. Biophys. Acta 887, 51-57 (1986). 10. S.L. WOOD, N. EMMISON, A. C. BORTHWICK and S. J. YEAMAN, The protein phosphatases responsible for dephosphorylation of hormone-sensitive lipase in isolated rat adipocytes, Biochem. J. 295, 531-535 (1993). 11. E . M . REIMANN and R. A. BEHAM, Catalytic subunit of cAMP-dependent protein kinase, Meth. Enzymol. 99, 51-55 (1983). 12. P. COHEN, S. ALEMANY, B. A. HEMMINGS, T. J. RESINK, P. STRALFORS and H. Y. L. TUNG, Protein phosphatase-1 and protein phosphatase-2A from rabbit skeletal muscle, Meth. Enzymol. 159, 390-408 (1988). 13. P. STRALFORS, P. BJORGELL and P. BELFRAGE, Hormonal regulation of hormone-sensitive lipase in intact adipocytes: Identification of phosphorylated sites and effects on the phosphorylation by lipolytic hormones and insulin, Proc. Natl. Acad. Sci. USA 81, 3317-3321 (1984). 14. P. STRALFORS and P. BELFRAGE, Phosphorylation of hormone-sensitive lipase by cyclic AMP-dependent protein kinase, J. Biol. Chem. 258, 15146-15152 (1983). 15. A. J. GARTON, D. G. CAMPBELL, P. COHEN and S. J. YEAMAN, Primary structure of the site on bovine hormone sensitive lipase phosphorylated by cyclic-AMP-dependent protein kinase, FEBS Lett. 229, 68-72 (1988). 16. C. HOLM, T. G. KIRCHGESSNER, K. L. SVENSON, G. FREDRIKSON, S. NILSSON, C. G. MILLER, J. E. SHIVELY, C. HEINZMANN, R. S. SPARKES, T. MOHANDAS, A. J. LUSIS, P. BELFRAGE and M. C. SCHOTZ, Hormonesensitive lipase: sequence, expression and chromosomal location to 19 cent-q13..3, Science 241, 1503-1506 (1988).

368

s . J . YEAMAN et al.

17. A. J. GARTON, D. G. CAMPBELL, D. CARLING, D. G. HARDIE, R. J. COLBRAN and S. J. YEAMAN, Phosphorylation of bovine hormone-sensitive lipase by the AMP-activated protein kinase, Eur. J. Biochem. 179, 249-254 (1989). 18. A. J. GARTON and S. J. YEAMAN, Identification and role of the basal phosphorylation site on hormone-sensitive lipase, Eur. J. Biochem. 191, 245-250 (1990). 19. P. STR/~LFORS and R. C. HONNOR, Insulin-induced dephosphorylation of hormone-sensitive lipase, Eur. J. Biochem. 182, 379-385 (1989). 20. P. COHEN, The structure and regulation of protein phosphatases, Ann. Rev. Biochem. 58,453-508 (1989). 21. H. OLSSON and P. BELFRAGE, The regulatory and basal phosphorylation sites of hormone-sensitive lipase are dephosphorylated by protein phosphatase-1, 2A and 2C but not by protein phosphatase-2B. Eur. J. Biochem. 168, 399-405 (1987). 22. T. INGEBRITSEN and P. COHEN, The protein phosphatases involved in cellular regulation. 1. Classification and substrate specificities. Eur. J. Biochem. 132, 255-261 (1983). 23. P. COHEN, S. KLUMPP and D. L. SCHELL1NG, An improved procedure for identifying and quantitating protein phosphatases in mammalian tissues, FEBS. Lett. 250, 596-600 (1989). 24. T. S. INGEBRITSEN, A. A. STEWART and P. COHEN, The protein phosphatases involved in cellular regulation. 6. Measurement of type-1 and type-2 protein phosphatases in extracts of mammalian tissues; an assessment of their physiological roles, Eur. J. Biochem. 132, 297-307 (1983). 25. D. G. HARDIE, Regulation of fatty acid and cholesterol metabolism by the AMP-activated protein kinase, Biochim. Biophys. Acta 1123, 231-238 (1992). 26. T. G. KIRCHGESSNER, K. L. SVENSON, A. J. LUSIS and M. C. SCHOTZ, The sequence of cDNA encoding lipoprotein lipase. A member of a lipase gene family, J. Biol. Chem. 262, 8463-8466 (1987). 27. D. LANGIN, H. LAURELL, L. S. HOLST, P. BELFRAGE and C. HOLM, Gene organisation and primary structure of human hormone-sensitive lipase: possible significance of a sequence homology with a lipase from Moraxella TA144, an antarctic bacterium, Proc. Natl. Acad. Sci. USA 90, 4897-4901 (1993). 28. L. BL,~CKBERG and O. HERNELL, The bile salt-stimulated lipase in human milk. Purification and characterisation, Eur. J. Biochem, 116, 221-225 (1981). 29. F . K . WlNKLER, A. D'ARCY and W. HUNZIKER~ Structure of human pancreatic lipase, Nature 343, 771-774 (1990). 30. L. BRADY, A. M. BRZOZOWSKI, Z. S. DEREWENDA, E. DODSON, G. DODSON, S. TOLLEY, J. P. TURKENBURG, L. CHRISTIANSEN, B. HUGE-JENSEN, L. NORSKOV, L. THIM and U. MENGE, A serine protease triad forms the catalytic centre of a triacyiglycerol lipase, Nature 343, 767-770 (1990). 31. J . D . SCHRAG, Y. LI, S. WU and M. CYGLER, Ser-His-Glu triad forms the catalytic site of a lipase from Geotrichum candidum, Nature 351,761-764 (1991). 32. Z. S. D E R E W E N D A and A. M. SHARP, News from the interface: the molecular structures of triaclyglycerol lipases, Trends Biochem. Sci. 18, 20-25 (1993). 33. G . G . DODSON, D. M. LAWSON and F. K. WlNKLER, Structural and evolutionary relationships in lipase mechanism and activation, Faraday Discuss. 93, 95-105 (1992). 34. T. TSUJITA, H. NINOMIYA and H. OKUDA, p-Nitrophenyl butyrate hydrolysing activity of hormone-sensitive lipase from bovine adipose tissue, J. Lipid Res. 30, 997-1004 (1989). 35. G. BENGTSSON and T. OL1VECRONA, Lipoprotein lipase: Modification of its kinetic properties by mild tryptic digestion, Eur. J. Biochem. 113, 547-554 (1981). 36. K. SHIRAI, Y. SAITO and S. YOSHIDA, Post-heparin plasma hepatic triacylglycerol lipase-catalysed tributyrin hydrolysis. Effect of trypsin treatment, Bioehim. Biophys. Acta 795, 9-14 (1984). 37. T. TSUJITA, A. NAKAGAWA, K. SHIRAI, Y. SAITO and H. OKUDA, Methyl

HORMONE-SENSITIVE LIPASE

38. 39. 40.

41. 42. 43. 44.

45. 46. 47.

48. 49. 50. 51. 52. 53.

54. 55.

369

butyrate hydrolysing activity of hepatic triglyceride lipase from rat post-heparin plasma, J. Biol. Chem. 259, 11215-11220 (1984). B. BERNB,~.CK and L. BLA,CKBERG, Human gastric lipase. The N-terminal tetrapeptide is essential for lipid binding and lipase activity, Eur. J. Biochem. 182, 495-499 (1989). G . M . SMITH, A. J. GARTON, A. AITKEN and S. J. YEAMAN, Domain structure of hormone-sensitive lipase, Biochem. Soc. Trans. 21,233S (1993). A . M . BRZOZOWSKI, U. DEREWENDA, Z. S. DEREWENDA, G. G. DODSON, D. M. LAWSON, J. P. TURKENBURG, F. BJORKLING, B. HUGE-JENSEN, S. A. PATKAR and L. THIM, A model for interracial activation in lipases from the structure of fungal lipase-inhibitor complex, Nature 351,491-494 (1991). Z.S. DEREWENDA and U. DEREWENDA, Relationships among serine hydrolases: evidence for a common structural motif in triacylglycerol lipase and esterases, Biochem. Cell Biol. 69, 842-851 (1991). K. SHIRAI, N. MATSUOKA, Y. SAITO and S. YOSHIDA, Post-heparin plasma hepatic triaclyglycerol lipase-catalyzed hydrolysis of tributyrin. Effect of lipid interface, Biochim. Biophys. Acta 795, 1-8 (1984). K. SHIRAI and R. L. JACKSON, Lipoprotein lipase-catalysed hydrolysis of p-nitrophenyl butyrate. Interracial activation by phospholipid vesicles, J. Biol. Chem. 257, 1253-1258 (1982). D. STEINBERG, Current theories of the pathogenesis of atherosclerosis, pp. 5-23, in Hypercholesterolaemia and Atherosclerosis: Pathogenesis and Prevention (D. STEINBERG and J. M. OLEFSKY, eds.), Churchill Livingstone Inc., New York (1987). R. ROSS and J. A. GLOMSET, The pathogenesis of atherosclerosis, New Engl J. Med. 295, 369-376 (1976). R. W. WlSSLER, Development of the atherosclerotic plaque, pp. 155-166 in The Myocardium: Failure and Infarction (E. BRAUNWALD, ed.), H.P. Publ. Co., New York (1973). J. L. GOLDSTEIN, Y. K. HO, S. K. BASU and M. S. BROWN, Binding site on macrophages that mediates uptake and degradation of acetylated low density lipoprotein, producing massive cholesterol deposition, Proc. Natl. Acad. Sci. USA 76, 333-337 (1979). M. S. BROWN, Y. K. HO and J. L. GOLDSTEIN, The cholesteryl ester cycle in macrophage foam cells, J. Biol. Chem. 255, 9344-9352 (1980). M. S. BROWN, J. L. GOLDSTEIN, M. KREIGER, Y. K. HO and R. G. W. ANDERSON, Reversible accumulation of cholesteryl esters in macrophages incubated with acetylated lipoproteins, J. Cell Biol. 82, 597-613 (1979). C. A. SMALL, M. P. ROGERS, J. A. G O O D A C R E and S. J. YEAMAN, Phosphorylation and activation of hormone-sensitive lipase in isolated macrophages, FEBS Lett. 279, 323-326 (1991). R. A HELLER and D. STEINBERG, Partial glyceridase activity of a protein kinase activatable triglyceride lipase from rat adipose tissue, Biochim. Biophys. Acta 270, 65-73 (1972). C. A. JEPSON and S. J. YEAMAN, Inhibition of hormone-sensitive lipase by intermediary lipid metabolites, FEBS Lett. 310, 197-200 (1992). N. O. NILSSON, P. STR~LFORS, G. FREDRIKSON and P. BELFRAGE, Regulation of adipose tissue lipolysis: Effects of noradrenaline and insulin on phosphorylation of hormone-sensitive lipase and on lipolysis in intact rat adipocytes, FEBS Lett. 111,125-130 (1980). P. STRA,LFORS, H. OLSSON and P. BELFRAGE, Hormone-sensitive lipase, pp. 147-177 in The Enzymes X V I I (P. D. BOYER and E. G. KREBS, eds.) Academic Press, Orlando, FL (1987). A. H. HIRSCH and O. M. ROSEN, Lipolytic stimulation modulates the subcellular distribution of hormone-sensitive lipase in 3T3-L1 cells, J. Lipid Res. 25,665-667 (1984).

370

S.J. YEAMAN et al.

56. J.J. EGAN, A. S. GREENBERG, M-K. CHANG, S. A. WEK, M. C. MOOS, JR and C. LONDOS, Mechanism of hormone-stimulated iipolysis in adipocytes: Translocation of hormone-sensitive lipase to the lipid storage droplet, Proc. Natl. Acad. Sci., USA 89, 8537-8541 (1992). 57. R. A. MOONEY and K. L. BORDWELL, Counter-regulation by insulin and isoprenaline of a prominent fat-associated phosphoprotein doublet in rat adipocytes, Biochem. J. 274, 433-438 (1991). 58. A.S. GREENBERG, J. J. EGAN, S. A. WEK, N. B. GARTY, E. J. BLANCHETTEMACKIE and C. LONDOS, Perilipin, a major hormonally regulated adipocyte-specific phosphoprotein associated with the periphery of lipid storage droplets, J. Biol. Chem. 266, 11341-11346 (1991).