Regulation of Adipocyte Lipolysis by Degradation of the Perilipin Protein

A decrease in the lipid droplet-associated protein perilipin may constitute a mechanism for enhanced adipocyte lipolysis under nonstimulated (basal) conditions, and increased basal lipolysis has been linked to whole body metabolic dysregulation. Here we investigated whether the lipolytic actions of the human immunodeficiency virus protease inhibitor, nelfinavir, are mediated by decreased perilipin protein content and studied the mechanisms by which it occurs. Time course analysis revealed that the decrease in perilipin protein content preceded the increase in lipolysis. A causative relationship was suggested by demonstrating that nelfinavir potently increased lipolysis in adipocytes derived from mouse embryonal fibroblasts expressing perilipin but not in mouse embryonal fibroblast adipocytes devoid of perilipin and that adenoviral mediated overexpression of perilipin in 3T3-L1 adipocytes blocked the lipolytic actions of nelfinavir. Nelfinavir did not alter mRNA content of perilipin but rather decreased perilipin proteins t½ from >70 to 12 h. Protein degradation of perilipin in both control and nelfinavir-treated adipocytes could be prevented by inhibiting lysosomal proteolysis using leupeptin or NH4Cl but not by the proteasome inhibitor MG-132. We propose that proteolysis of perilipin involving the lysosomal protein degradation machinery may constitute a novel mechanism for enhancing adipocyte lipolysis.

The lipid droplet-associated protein perilipin is thought to function as a critical regulator of basal (nonstimulated) adipocyte lipolysis (1, 2), a process central to maintain normal energy balance and utilization (3,4). In response to lipolytic stimulation by ␤-adrenergic agents, cAMP-dependent protein kinasemediated phosphorylation of perilipin renders it a facilitator of lipolysis, at least partly by altering the subcellular localization of cellular lipases (5,6). Conversely, in the absence of a lipolytic stimulus (basal or constitutive lipolysis), perilipin inhibits lipase(s) activity, an action commonly described as the "barrier function" between cellular lipases and their substrate (triglycerides comprising the lipid droplets) (7,8). The regulatory role of this function of perilipin was suggested by demonstrating that tumor necrosis factor ␣ treatment of murine adipocytes reduced perilipin expression while increasing lipolysis, whereas overexpression of perilipin A blocked the lipolytic effect of tumor necrosis factor (9). The physiological role of perilipin as an inhibitor of basal lipolytic rate was confirmed in perilipin knock-out mice, which exhibit a lean phenotype attributed to elevated basal lipolysis (10,11). In humans, recent evidence suggests that polymorphisms in the perilipin gene may predict vulnerability to obesity (12)(13)(14)(15). In studies where perilipin protein expression was examined in isolated adipocytes, perilipin was found to be decreased in obese individuals, suggesting a mechanism for the elevated levels of circulating fatty acids typically observed in obesity (16,17). Intriguingly, protein expression of perilipin does not always correlate with the mRNA level (17)(18)(19), suggesting that post-transcriptional mechanisms participate in regulating perilipin protein content and thereby regulate the adipocyte lipolytic rate.
An enhanced rate of lipolysis is likely a common metabolic side effect of highly active anti-retroviral therapy (HAART). 2 Patients receiving HAART develop a dyslipidemia characterized by enhanced lipid mobilization and elevated levels of circulating free fatty acids (FFA) (20). In addition, the lipolysis rate was increased when assessed directly in vivo (21). Furthermore, dyslipidemia in HAART patients, particularly in patients receiving certain HIV protease inhibitors, is frequently associated with fat tissue redistribution (loss of subcutaneous "peripheral" fat and accumulation of intra-abdominal fat). Thus, enhanced lipolysis may contribute to both the frequently observed loss of subcutaneous fat mass, as well as to the dyslipidemia (22).
Investigating the cellular mechanisms for enhanced adipocyte lipolysis in response to HIV protease inhibitors revealed that the nelfinavir-induced lipolysis rate in 3T3-L1 adipocytes was associated with decreased protein levels of perilipin (23). Furthermore, the HIV protease inhibitor ritonavir had no effect on cAMP levels nor on cAMP-dependent protein kinase activity in 3T3-L1 adipocytes (24). This suggests that the increase in lipolysis rate in response to HIV protease inhibitors is mediated by alterations downstream of cAMP-dependent protein kinase (24). In the present study we investigated whether the nelfinavir-associated decrease in perilipin plays a causative role in the increased rates of basal lipolysis. Furthermore, we investigate the underlying cellular mechanisms by which this HIV protease inhibitor reduces adipocyte perilipin protein content.

EXPERIMENTAL PROCEDURES
Materials-Tissue culture medium, serum, and antibiotic solutions were obtained from Biological Industries (Beit-Haeemek, Israel). Recombinant human insulin was from Novo Nordisk (Bagsvaerd, Denmark). Anti-perilipin, anti-adipocyte differentiation-related protein (ADRP), and anti-hormonesensitive lipase (HSL) antibodies and recombinant adenoviruses expressing perilipin A or green fluorescent protein (GFP) were generated as previously described (8,9,25). Peroxidaseconjugated anti-rabbit IgG, anti-mouse IgG, protein G-Sepharose, protein A-Sepharose, and [ 35 S]methionine were from Amersham Biosciences. Alexa 488-conjugated anti-rabbit IgG antibody was from Molecular Probes (Eugene, OR). The FFA measurement kit was obtained from Roche Applied Science. Two sources of nelfinavir were tested: Roche Pharmaceuticals (through the branch in Tel Aviv, Israel) and Shanghai 21CEC Pharma Ltd. (London, UK). All other chemicals (including a glycerol measurement kit) were obtained from Sigma.
Adipocyte Differentiated from Mouse Embryonic Fibroblasts (MEF)-Stable lines of MEF preadipocytes overexpressing peroxisome proliferator-activated receptor ␥ were generated from immortalized MEFs of perilipin-null mice (MEFϪ/Ϫ) or wild type controls (MEFϩ/ϩ) as described (6,28). Peroxisome proliferator-activated receptor ␥ was introduced by retrovirally mediated transduction followed by puromycin selection. MEF were cultured as described (6) in DMEM with 10% fetal bovine serum, 1 mM L-glutamine, 100 units/ml penicillin, 0.1 mg/ml streptomycin, and 0.25 g/ml amphotericin B. At confluence the cells were exposed to a pro-differentiation regimen consisting of dexamethasone (1 M), insulin (5 g/ml), isobutylmethylxanthine (0.5 mmol/liter), and rosiglitazone (5 M). After 2 days, the medium was changed, and the cells were maintained in medium containing insulin for 4 days and then in medium without insulin for an additional 4 -5 days before use. The cells used for all experiments exhibited Ͼ90% adipocyte morphology by light microscopy.
For nelfinavir treatment, differentiated adipose cells were incubated in serum-free DMEM supplemented with 0.5% radioimmunoassay grade bovine serum albumin, with or without nelfinavir, and in the presence or absence of the indicated inhibitors. The final nelfinavir medium concentrations were achieved by the appropriate dilutions of a 100 mM stock solution prepared in 100% ethanol. Final ethanol concentrations of up to 0.04% were without measurable effects on the parameters measured in this study. As in previous publications (23,26,29), cell viability was monitored by protein recovery and 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide assays, excluding compromised cell viability under the conditions used.
Adenovirus-mediated Perilipin Overexpression-The cells were infected as previously described (8). Briefly, 3T3-L1 fibroblasts placed in 12-well plates were cultured in DMEM containing 10% bovine calf serum and differentiated as described above. The adipocytes were infected on day 3 of differentiation with a multiplicity of infection of ϳ100 plaque-forming units/ cell, by incubation with Lipofectamine Plus reagent (Invitrogen) for 3 h at 37°C. Eight days later, infected adipocytes were used for the experiments, with at least 70 -80% of cells exhibiting increased perilipin content. As a control, recombinant adenovirus expressing GFP was used.
Cell Lysates and Western Blots-After treatments, the cells were rinsed three times with PBS and scraped into ice-cold lysis buffer containing 50 mmol/liter Tris-HCl, pH 7.5, 0.1% (w/v) Triton X-100, 1 mmol/liter EDTA, 1 mmol/liter EGTA, 50 mmol/liter NaF, 10 mmol/liter sodium ␤-glycerophosphate, 5 mmol/liter sodium pyrophosphate, 1 mmol/liter sodium vanadate, and 0.1% (v/v) 2-mercaptoethanol, and inhibitors (a 1:1,000 dilution of protease inhibitor mixture; Sigma). The lysates were gently shaken for 15 min at 4°C, centrifuged (12,000 ϫ g, 15 min at 4°C), and the fraction between the pellet and adipose cake was collected. Protein concentration was determined using the Bio-Rad Bradford method procedure (Munich, Germany). Protein samples were resolved on 10% SDS-polyacrylamide gel electrophoresis and subjected to Western blot, followed by quantitation using video densitometry analysis, as described previously (27,30). In each experiment the intensity of the band derived from control cells was assigned a value of 1 arbitrary unit, and the intensity of all of the treatment groups was expressed as the fold value of control.
Lipolysis-For basal lipolysis measurements, the cells were rinsed three times with PBS (after the indicated period of nelfinavir treatment) and further incubated for 1 h with Krebs-Ringer phosphate buffer (50 mmol/liter HEPES, pH 7.4, 128 mmol/liter NaCl, 4.7 mmol/liter KCl, 1.25 mmol/liter CaCl 2 , 1.25 mmol/liter MgSO 4 , and 10 mmol/liter sodium phosphate) supplemented with 1% radioimmunoassay grade bovine serum albumin, as previously described (31). FFA concentrations in Krebs-Ringer phosphate buffer were determined using a commercial kit (Roche Applied Science) following the instructions of the manufacturer and calculated according to a palmitic acid standard curve. Glycerol was measured using a glycerol 3-phosphate oxidase trinder kit (Sigma), as described (31).
Quantitative Real Time Polymerase Chain Reaction-cDNA was prepared using a Reverse-iT TM 1st strand synthesis kit (ABgene, UK) and quantitative real time PCR assays were carried out for perilipin A and for 18 S rRNA exactly as previously described (32). The following primers were used: murine perilipin A, sense, GGCCTGGACGACAAAACC, and antisense, CAGGATGGGCTCCATGAC; murine 18 S rRNA, sense, CGCCGCTAGAGGTGAAATTCT, and antisense, CATT-CTTGGCAAATGCTTTCG.
Pulse-Chase Experiments-Prior to [ 35 S]methionine incorporation the cells were preincubated for 1 h with methioninefree medium to deplete the cells from intracellular methionine. Next, the cells were pulse-labeled for 5 h with 100 Ci/ml [ 35 S]methionine added to methionine-free medium. Pulse-labeling reaction was terminated by three washes with PBS supplemented with 1,000-fold unlabeled methionine. The cells were subsequently chased for the indicated time periods in serum-free DMEM, then washed, and lysed in ice-cold radioimmune precipitation assay lysis buffer (50 mM Tris-HCl, pH 7.4, 150 mM NaCl, 1% Nonidet P-40, 0.5% sodium deoxycholate, 0.1% SDS, and 1 mM EDTA) supplemented with proteases inhibitors (a 1:100 dilution of protease inhibitor mixture; Sigma). The cell lysates were prepared as described above. The protein concentration was determined by BCA assay (Pierce), and 0.5 mg of cell lysate protein was used for immunoprecipitation. Anti-perilipin A antibody was linked to prewashed Sepharose AϩG beads mix (1:1). Ninety l of beads mixture were incubated (by gentle shaking) with anti-perilipin A antibody at 4°C overnight, after which the samples were centrifuged for 1 min at 12,000 rpm, and the supernatant was discarded. Prior to immunoprecipitation, the lysates were precleared by incubation (shaking) with 90 l of Sepharose-protein AϩG mixture for 20 min at 4°C. Then the samples were centrifuged for 1 min at 12,000 rpm, and the supernatants were added to the Sepharose bead-bound perilipin antibodies. The immunoprecipitation reaction was performed by gentle shaking at 4°C for 3 h. The immunoprecipitated complex was pelleted by centrifugation at 5000 rpm at 4°C for 2 min, washed three times with cold radioimmune precipitation assay buffer, and mixed with a 10% SDS Laemmli sample buffer (final SDS concentration, 2%). The samples were boiled at 100°C for 5 min and centrifuged, and the liquid fraction was resolved by SDS-PAGE electrophoresis. The gel was dried using a vacuum gel drier, and [ 35 S]methionine-labeled perilipin A in immunoprecipitates was detected directly by autoradiography using a phosphorimaging device (Fujifilm FLA-5100).
Immunofluorescence-The cells were grown on glass coverslips. After treatment, the cells were fixed in 3% paraformaldehyde for 30 min, washed, and incubated with 100 mM glycine for 10 min. Following another wash with PBS, the cells were permeabilized with 0.1% (v/v) Triton X-100 in PBS for 15 min followed by 1 h of incubation in 5% goat serum-supplemented PBS. Next, the cells were incubated with primary anti-perilipin A antibody (dilution of 1:100 in the above blocking solution) for 1 h at room temperature. The cells were washed three times with PBS and incubated with Alexa 488-conjugated anti-rabbit IgG antibody (1:750 dilution in blocking solution) for 1 h, at room temperature, washed, and mounted on glass slides. The images were taken at mid-z-axis height of the cells, using a Zeiss laser confocal IF microscope. The acquisition parameters were kept constant within the experiment to allow comparison between resulting signal intensities.
Statistical Analysis-The data are expressed as the means Ϯ S.E. Each treatment was compared with control, and statistical significance of differences between two groups was evaluated using the Student's t test. The criterion for significance was set at p Ͻ 0.05.

Decreased Perilipin Content in 3T3-L1 Adipocytes Precedes the Increase in Basal Lipolysis Rate Induced by Nelfinavir-We
have previously demonstrated that 3T3-L1 adipocytes exposed to nelfinavir exhibit increased constitutive (basal) lipolysis along with decreased expression of the lipid droplet-associated protein, perilipin (23). As a first step in elucidating the role of perilipin in nelfinavir-mediated increase in lipolysis, we compared the chronology of the reduction in perilipin protein with measurements of lipolysis. Time course analysis was performed by assessing perilipin content and glycerol release to the medium following incubation of 3T3-L1 adipocytes with 30 M nelfinavir (Fig. 1). Two hours of nelfinavir treatment were sufficient to induce a statistically significant 15.5 Ϯ 3.1% decrease in perilipin level compared with time 0 (Fig. 1B). The progressive decrease in perilipin content over time (reaching 76.9 Ϯ 11.8% at 24 h) was a specific response to nelfinavir, because HSL and ␤-actin protein content remained stable (Fig. 1A). In contrast to the early reduction in perilipin content to nelfinavir, a significant increase in basal lipolysis rate appeared only after 8 h of nelfinavir treatment (Fig. 1B), corresponding to a 50% decrease in perilipin. We utilized indirect immunofluorescence and laser confocal microscopy to determine whether gross alterations in cellular morphology and subcellular localization of perilipin occurred in response to nelfinavir. Nelfinavir treatment resulted in a decrease in perilipin signal intensity, without apparent alterations in its distribution (Fig. 1C) or in overall cellular morphology (not shown). Collectively, these findings indicate that a decrease in perilipin cell content temporally precedes nelfinavir-induced lipolysis and that nelfinavir treatment results in a progressive reduction of perilipin protein expression.
A Causative Role for Decreased Perilipin Expression in Nelfinavir-induced Lipolysis-To establish a causative relationship between a decrease in perilipin content and an increase in lipolysis induced by nelfinavir, we undertook two complementary approaches. First, we investigated whether nelfinavir increases basal lipolysis in an adipocyte cellular model devoid of perilipin. Second, we determined whether overexpression of perilipin, using an adenovirus expression system, inhibits nelfinavir-induced lipolysis.
For the first set of experiments, we used adipocytes differentiated from MEFs of perilipin knock-out (MEFϪ/Ϫ) or wild type (MEFϩ/ϩ) mice. As reported recently (6), MEFϪ/Ϫ adipocytes exhibit similar adipogenesis potential to that of MEFϩ/ϩ. Furthermore, consistent with the phenotype of perilipin knock-out mice, differentiated MEFϪ/Ϫ were observed to have an increased basal lipolysis rate compared with MEFϩ/ϩ ( Fig. 2A). In response to 18 h of nelfinavir treatment, MEFϩ/ϩ adipocytes exhibited a decrease in perilipin content, whereas the expression of HSL was not significantly altered (Fig. 2B), consistent with the effects observed in 3T3-L1 adipocytes (Ref. 23 and Fig. 1). Thus, MEF-derived adipocytes represent a reliable adipocyte model system to examine the relationship between perilipin and lipolysis in response to nelfinavir. Subsequently, we investigated whether the absence of perilipin altered the ability of nelfinavir to increase basal lipolysis. MEFϪ/Ϫ and MEFϩ/ϩ adipocytes were treated with nelfinavir for 18 h, after which glycerol release to the medium was assessed. MEFϩ/ϩ cells exhibited a robust increase in basal lipolysis after nelfinavir treatment (nelfinavir concentrations of 30 and 45 M increased lipolysis by 13.2 Ϯ 3.1-and 24.5 Ϯ 5.3-fold, respectively), whereas only a minor increase was observed in MEFϪ/Ϫ (Fig. 2C). Similar results were observed when measuring FFA release (data not shown). This finding is consistent with the notion that the effect of nelfinavir on basal lipolysis requires the presence of perilipin.
If the decrease in perilipin levels is a required step in the induction of lipolysis by nelfinavir (Fig. 1), overexpression of perilipin should prevent the lipolytic response to the drug. To overexpress perilipin in 3T3-L1 adipocytes, we utilized adenoviruses expressing perilipin A under the control of the cytomegalovirus promoter (Ad-Peri) or a control adenovirus encoding GFP under the regulation of the same promoter (Ad-GFP) (9). Cells infected by Ad-Peri exhibited increased perilipin content compared with cells infected with Ad-GFP (Fig. 3, A and B). In cells where perilipin was overexpressed, as in those infected with the control virus, nelfinavir decreased perilipin protein content in a dose-dependent manner (Fig. 3, A and B). How- . Representative blots of five to nine independent experiments are shown. B, perilipin content (Ⅺ) and lipolysis (glycerol release, F) were followed for up to 24 h incubation with nelfinavir and expressed relative to time 0 (which was assigned a value of 100%). The results are expressed as the means Ϯ S.E. *, p Ͻ 0.05 versus time 0. C, following incubation with or without nelfinavir (30 M, 18 h), 3T3-L1 adipocytes were fixed, and indirect immunofluorescence detection of perilipin was performed, using anti-perilipin A antibody, followed by laser confocal microscopy, as described under "Experimental Procedures." Shown are panels from two independent experiments. Scale bar, 10 m. ever, nelfinavir did not alter the expression of either HSL or ␤-actin, confirming the specific effect of nelfinavir on perilipin protein content. Although nelfinavir decreased perilipin expression in the perilipin overexpressing cells, the remaining perilipin content after 18 h of nelfinavir treatment was still increased as compared with cells infected by the control adenovirus (Fig. 3, A and B). Importantly, perilipin overexpression, as compared with adipocytes transduced with Ad-GFP, significantly blocked the action of nelfinavir to increase adipocyte lipolysis (Fig. 3C).
Enhanced Lysosomal Degradation of Perilipin Is Involved in Decreasing Its Protein Content by Nelfinavir-Given the causative role of decreased perilipin expression in nelfinavir-mediated increase in basal lipolysis, we assessed the mechanism by which this reduction occurs. The amount of perilipin A mRNA (controlled for 18 S rRNA) was unaltered by nelfinavir treatment, as assessed by quantitative real time PCR using primers specific for this isoform of perilipin (Fig. 4A). There was also no significant change in perilipin B mRNA content (data not shown). This observation suggested that nelfinavir regulated perilipin through post-transcriptional mechanisms. To assess perilipin protein half-life, we performed cycloheximide (CHX) chase studies in which cells were incubated with 5 g/ml CHX. This concentration of the protein synthesis inhibitor resulted in over 95% inhibition of [ 35 S]methionine incorporation to total protein (data not shown). Under such conditions the amount of perilipin over time reflects its degradation rate. In control cells, the estimated perilipin protein half-life was over 70 h (Fig. 4B), suggesting that it is a highly stable protein. In contrast, nelfinavir-treated 3T3-L1 adipocytes, in which protein synthesis was inhibited by CHX, revealed a 50% decrease in perilipin protein content by 12 h (Fig. 4B). To further verify that nelfinavir enhanced the protein degradation of perilipin, we used a pulsechase approach. The cells were pulsed for 5 h with [ 35 S]methionine in methionine-free medium, after which radioactive methionine was thoroughly washed, and the cells were incubated in regular medium supplemented with unlabeled methionine in the presence or absence of nelfinavir for 12 and 24 h. To determine radioactive methionine content in perilipin, the cell lysates were prepared and subjected to immunoprecipitation using anti-perilipin antibodies. The immunoprecipitate was subjected to SDS-PAGE followed by autoradiography to determine the amount of radiolabeled perilipin. Consistent with the CHX chase approach, after 24 h of chase the loss of radioactivity in perilipin was undetectable in control cells (Fig.  4C). In contrast, cells incubated with nelfinavir exhibited greatly enhanced loss in [ 35 S]methionine-containing perilipin. The calculated t1 ⁄ 2 of perilipin using this approach was 12 h, remarkably similar to that determined by CHX chase.
To assess which protein degradation system is involved in the enhanced perilipin proteolysis induced by nelfinavir, we used inhibitors of proteasomal or of lysosomal proteolysis. MG-132 is considered a specific 26 S proteasome inhibitor, whereas leupeptin is a reversible and competitive intralysosomal proteolysis inhibitor that specifically inhibits serine and some cysteine proteases such as cathepsin B, trypsin, endoproteinase Lys-C, Kallikrein, papain, and thrombin. Consistent with a previous report in adipocytes, MG-132 had no effect on perilipin content in control adipocytes (33) and did not affect its levels in nelfinavir-treated cells (Fig. 5, A and B). In contrast, inhibiting the proteasome resulted in a marked increase in ADRP (Fig. 5, A  and C), a protein known to be proteasomally degraded (34). Leupeptin treatment of control cells resulted in 40% increase in perilipin content (p ϭ 0.015), and when present during incubation of 3T3-L1 adipocytes with nelfinavir, almost completely prevented the decrease in perilipin content (Fig. 5, A and B). To further corroborate these findings, the cells were treated with NH 4 Cl, which blocks lysosomal proteolysis by inhibiting lysosomal acidification. Like leupeptin, preventing lysosomal acidity tended to increase perilipin content in control cells and significantly prevented the decrease in perilipin protein content in nelfinavir-treated cells (Fig. 5, A and B). Finally, consistent with the causative role of decreased perilipin in nelfinavir-induced lipolysis, leupeptin and NH 4 Cl resulted in 35.3 Ϯ 10.7% (p ϭ

DISCUSSION
Two novel observations emerge form the present study: 1) the decrease in adipocyte perilipin content plays a causative role in the enhanced basal lipolytic rate induced by nelfinavir and 2) lysosomal protein degradation process(es) are likely involved in the regulation of perilipin content and, hence, in the regulation of lipolysis.
Regulation of Adipocyte Lipolysis by Proteolytic Degradation of Perilipin-Enhanced adipocyte basal lipolytic flux is a common feature in conditions associated with insulin resistance, including obesity, type 2 diabetes, and the lipodystrophy syndrome associated with HAART (3,4,21). At least two studies suggested that obesity was indeed associated with decreased perilipin protein content (17,18). Because perilipin has been shown to inhibit the basal lipolysis rate, reduced expression of this protein likely contributes to enhanced basal lipolytic flux associated with obesity. This is supported by the increased basal lipolytic rate observed in perilipin knock-out mice (10,11) and in adipocytes derived from this mouse model ( Fig. 2A), as well as by the demonstration that a decrease in perilipin induced by either nelfinavir or by tumor necrosis factor ␣ plays a causative role in the induction of lipolysis (Figs. 1-3 and Ref. 9, respectively). It is interesting to note that the reduction in perilipin protein content caused by tumor necrosis factor ␣ treatment is associated with reduced perilipin mRNA levels, whereas nelfinavir does not decrease perilipin mRNA. Mechanisms controlling the protein stability of perilipin in adipocytes are largely unknown but may be functionally important, because changes in the protein content of perilipin are not always associated with alterations in mRNA levels (17)(18)(19). In the present study, direct measurements of the half-life of perilipin protein in adipocytes revealed that it is a highly stable protein under control conditions, with a t1 ⁄ 2 exceeding 70 h (Fig. 4, B and C), and that enhanced protein degradation, not a change in its transcription rate, accounts for the decreased expression of perilipin in response to nelfinavir. Thus, an important area for future research will be to determine whether enhanced proteolytic degradation of perilipin constitutes a major process for altering perilipin expression in response to endogenous cues, thereby regulating the basal adipocyte lipolysis rate.
Involvement of Lysosomal Proteolysis of Perilipin in Regulating Lipolysis-If regulating perilipin content is indeed a functional means of controlling lipolysis in adipocytes, it is important to determine how perilipin expression is regulated. A small but statistically significant increase in perilipin protein content was observed when adipocytes were treated with the lysosomal proteases inhibitor leupeptin for 18 h (Fig. 5, A and B). In contrast, the proteasomal inhibitor MG-132 did not alter perilipin content while markedly preventing the proteolytic degradation of ADRP, another lipid droplet associated protein known to be degraded by the proteasome (34). This observation seemingly contradicts a recently published finding (35). Yet, this study, which assessed perilipin protein stability and degradation, utilized nonadipocyte cells (Chinese hamster ovary cells) exogenously expressing perilipin. Therefore, we hypothesize that in FIGURE4.Nelfinavirdecreasesperilipincontentthroughenhancedproteindegradation. A, total RNA was isolated from control or nelfinavir treated cells (30 M, 18 h). After RT-PCR, perilipin cDNA amount was quantified by real time PCR using primers specific for perilipin A, as described under "Experimental Procedures." The resultswerenormalizedtotheamountof18SrRNAandarethemeansϮS.E.ofthree independent experiments. B, 3T3-L1 adipocytes were incubated with 5 g/ml CHX alone or in the presence of 30 M nelfinavir. At each time point the cells were lysed, and the amount of perilipin protein in control and nelfinavir-treated cells was quantified by Western blot. Shown is a representative blot and densitometry analysis of five independent experiments. C, 3T3-L1 adipocytes were pulsed for 5 h in the presence of 100 Ci/ml [ 35 S]methionine, as described under "Experimental Procedures." Thenthecellswerechasedfor12and24hwithorwithout30Mnelfinavir.PerilipinA was immunoprecipitated, and the amount of associated [ 35 S]methionine was detectedbyautoradiography.Arepresentativeautoradiogramandthemeans ϮS.E. of three independent experiments are shown. *, p Ͻ 0.05 versus respective cells untreated with nelfinavir.
adipocytes perilipin protein degradation occurs in lysosomes and that this process is enhanced by nelfinavir. Furthermore, perilipin degradation through the lysosome versus ADRP degradation by the proteasome provides a novel mechanism to explain the reciprocal regulation of these two lipid droplet-associated proteins.
Lysosomal degradation of intracellular proteins can occur as part of cellular microautophagy or macroautophagy (36,37). Intriguingly, autophagy can be triggered by oxidative stress (38,39), which has been shown to be induced in nelfinavir-treated adipocytes (32,40). This could be mediated by oxidative stress-induced impairment in Akt/PKB activation (27), because this kinase was shown to inhibit autophagy (41). Nevertheless, micro-and macro-autophagy are nonspecific mechanisms, whereas some degree of selectivity was observed in the enhanced lysosomal degradation of perilipin induced by nelfinavir (Figs. 1 and 5). Although chaperone-mediated autophagy is a selective proteolytic mechanism (42, 43), its described substrates are all soluble cytosolic proteins, whereas perilipin is lipid-anchored. Thus, whether perilipin is targeted to the lysosomes through chaperone-mediated autophagy and, if so, whether it becomes detached from the lipid droplet is worth investigating in future studies. Finally, a recent study suggests that perilipin may undergo regulated translocation to the plasma membrane (44), raising the theoretical possibility that perilipin traffics between the phospholipids monolayer of the lipid droplet, the plasma membrane, and the lysosome.
In conclusion, studying the regulation of lipolysis by the HIV protease inhibitor nelfinavir has provided new insights into the protein degradation of perilipin as a cellular mechanism that likely involves the lysosomal degradation machinery. Future studies will determine whether lysosomal degradation of perilipin is important in physiological states such as obesity and in response to inflammatory cytokines.