Biochimica et Biophysica Acta 1771 (2007) 66 – 74 www.elsevier.com/locate/bbalip
Heat shock protein 70 is translocated to lipid droplets in rat adipocytes upon heat stimulation Hongfeng Jiang a , Jinhan He a , Shenshen Pu a,b , Chaoshu Tang a,b , Guoheng Xu a,b,⁎ a b
Department of Physiology and Pathophysiology, Peking University Health Science Center, Beijing 100083, China Key Laboratory of Molecular Cardiovascular Sciences, The Ministry of Education of China, Beijing 100083, China Received 13 June 2006; received in revised form 16 October 2006; accepted 31 October 2006 Available online 15 November 2006
Abstract In mammalian cells, lipid storage droplets contain a triacylglycerol and cholesterol ester core surrounded by a phospholipid monolayer into which a number of proteins are imbedded. These proteins are thought to be involved in modulating the formation and metabolic functions of the lipid droplet. In this study, we show that heat stress upregulates several heat shock proteins (Hsps), including Hsp27, Hsp60, Hsp70, Hsp90, and Grp78, in primary and differentiated adipocytes. Immunostaining and immunoblotting data indicate that among the Hsps examined, only Hsp70 is induced to redirect to the lipid droplet surface in heat-stressed adipocytes. The thermal induction of Hsp70 translocation to lipid droplet does not typically happen in a temperature- or time-dependent manner and occurs abruptly at 30–40 min and rapidly achieves a steady state within 60 min after 40 °C stress of adipocytes. Though Hsp70 is co-localized with perilipin on the lipid droplets in stressed adipocytes, immunoprecipitation experiments suggest that Hsp70 does not directly interact with perilipin. Alkaline treatments indicate that Hsp70 associates with the droplet surface through non-hydrophobic interactions. We speculate that Hsp70 might noncovalently associate with monolayer microdomains of the lipid droplet in a manner similar to its interaction with lipid bilayer moieties composed of specific fatty acids. As an acute and specific cellular response to the heat stimulation, accumulation of Hsp70 on adipocytes lipid droplets might be involved in stabilizing the droplet monolayer, transferring nascent proteins to the lipid droplets, or chaperoning denatured proteins on the droplet for subsequent refolding. © 2006 Elsevier B.V. All rights reserved. Keywords: Heat shock protein; Lipid droplet; Lipid body; Thermal stress; Heat shock; Translocation; Perilipin; Adipocyte; Lipid membrane
1. Introduction Lipid storage droplets in animal cells are metabolically active organelles, which serve as an energy reservoir and provide precursors for membrane biogenesis and steroid hormone synthesis [1]. Depending on the type of cell, the lipid droplets Abbreviations: ADRP, adipose differentiation-related protein; BSA, defatted bovine serum albumin; Cy5, Cyanine-5; fDMEM, phenol red-free and serumfree Dulbecco's modified Eagle's medium; FITC, fluorescein isothiocyanate; Grp, glucose-regulated protein; HRP, horseradish peroxidase; Hsc, heat shock cognate protein; Hsp, heat shock protein; HSL, hormone-sensitive lipase; PAT, Perilipin/ADRP/TIP47; PBS, phosphate-buffered saline; SDS-PAGE, SDS polyacrylamide gel electrophoresis ⁎ Corresponding author. Department of Physiology and Pathophysiology, Peking University Health Science Center, 38 Xueyuan Road, Beijing 100083, China. Tel./fax: +86 10 8280 2916. E-mail address:
[email protected] (G. Xu). 1388-1981/$ - see front matter © 2006 Elsevier B.V. All rights reserved. doi:10.1016/j.bbalip.2006.10.004
vary in size and number, but their structure is rather simple in all eukaryotes. They consist of a hydrophobic triacylglycerol and cholesterol ester core surrounded by a phospholipid monolayer into which a number of proteins are embedded [1,2]. Major proteins on the lipid droplet surface are perilipin [3,4], adipose differentiation-related protein (ADRP, also termed adipophilin) [5], TIP47 [6,7], and S3–12 [6]. These proteins belong to a PAT (Perilipin/ADRP/TIP47) family, share sequence homology within their PAT domains [2], and derive from a common ancestral gene [8]. To date, perilipin is the only family member whose function in lipid metabolism has been firmly established [2,9–11]. Native perilipin may serve as a barrier to prevent access of lipases to the lipid droplet core, thus protecting triacylglycerols from hydrolysis and promoting lipid droplet formation [12]. By contrast, perilipin phosphorylated by protein kinase A governs translocation of hormone-sensitive lipase (HSL) from the cytosol to the lipid droplet surface [9], hence
H. Jiang et al. / Biochimica et Biophysica Acta 1771 (2007) 66–74
enhancing adipocyte lipolysis [9–11]. Most recently, we have revealed that hyperphosphorylation of perilipin independent on protein kinase A activation also results in considerable lipolysis without inducing the HSL translocation, and suggested that protein kinase A and a phosphatase cooperatively modulate the steady state of perilipin phosphorylation and dephosphorylation and hence provide a two-step strategy in adipocytes to regulate the lipolysis response to hormonal stimulation [10]. The proteins on the lipid droplet surface may virtually control the formation and metabolic function of the droplets [1,2,6,12,13]. Recent proteomic analyses have proposed that the protein components of the lipid droplet may be more than what we know [14–18]. These proteins can be classified into approximately 5 groups: (i) the PAT family proteins that are well characterized to associate with the lipid droplets; (ii) lipid metabolic enzymes such as HSL, fatty acid CoA ligase, acetylCoA carboxylase, acyl-CoA synthetases, lanosterol synthase, 17β-hydroxysteroid dehydrogenase type 7 and 11, and NADPdependent steroid dehydrogenase; (iii) membrane and vesicle trafficking and targeting proteins such as GTPase-related Rab family proteins; (iv) cytoskeletins, including tubulin, β-actin, cytokeratin, vimentin; and (v) molecular chaperones, including heat shock proteins (Hsps) and glucose-regulated proteins (Grp). By far, only a small proportion of these proteins have been cytochemistrically confirmed to be associated with the lipid droplets; these include perilipin [3,4,19], ADRP [5,13], TIP47 [6,7], S3–12 [6], CGI-58 [20], HSL [9,10,21], caveolins [22,23], vimentin [24], Rab18 GTPase [25], and NADP-dependent steroid dehydrogenase [26]. Because the lipid droplet preparations for the proteomic analysis are easily contaminated with plasma membrane fragments or other organelles [14], the above proteomically identified proteins might not be intrinsic components of the lipid droplet but, rather, artificially derived from the contamination. For example, though unconfirmed yet, it is perhaps unsurprising if some of these proteins such as lipid metabolic enzymes, localize on the lipid droplet surface, where they manipulate metabolic reactions. Nevertheless, lipid droplet association with other components such as Hsp60 [16], Hsp70 [14,17], Hsp73 [16], Hsp90 [16], Grp78/BiP [14–16], or Grp94 [16], as identified by proteamic analysis [14–17], remains to be clarified, because these chaperones are well known to locate predominately in cytoplasm. During our earlier experiments in examining heat stressinduced lipolysis (Xu G., in preparation), we observed that Hsp70 was redirected from the cytosol to the lipid droplets of rat adipocytes in response to heat stimulation. Further study of this observation was then encouraged by recent proteomic data [14]. In this report, we provide the first direct evidence to show that the cytosolic Hsp70 is markedly translocated to the surface of the lipid droplets in heat-stressed rat adipocytes. 2. Materials and methods
67
U.S. National Institutes of Health. Affinity purified polyclonal antibody against Hsp70 (catalog #sc-1060), monoclonal antibody against Hsp70 and Hsc70 (catalog #sc-24), and other primary antibodies were purchased from Santa Cruz Biotechnology (Santa Cruz, CA, USA). AffiniPure second antibodies conjugated with fluorescein isothiocyanate (FITC) or Cyanine-5 (Cy5) or horseradish peroxidase (HRP) were obtained from Jackson ImmunoResearch Laboratories Inc. (West Grove, PA, USA). Nitrocellulose membrane, prestained protein molecular weight markers, and Super Enhanced chemiluminescence detection reagents were from Applygen Technologies Inc. (Beijing, China).
2.2. Isolation of rat primary adipocytes and preadipocytes Adipocytes were isolated from epididymal fat pads of Sprague–Dawley rats (140–180 g) [10]. The fat pads were minced and digested in 5 ml Krebs–Ringer solution containing 25 mM HEPES, pH 7.4, 0.75 mg/ml type I collagenase and 1% defatted bovine serum albumin (BSA). After incubation for 60 min at 37 °C in a water bath shaken at 100 cycles/min, cells were filtered through a 250-μm nylon mesh and washed 3 times in pre-warmed fDMEM. The suspension tube was incubated at 37 °C for 20 min to allow primary adipocytes to float on the top of the tube. The infrantant stromal vascular cells served as preadipocytes and were collected through a sterile syringe with a long needle. The primary adipocytes left in the original tube were resuspended with 1% BSA-fDMEM, and centrifuged at 200×g for 3 min. Packed adipocytes were diluted in 1% BSAfDMEM to generate a 10% (v/v) cell suspension, in which the packed cell volume (PCV) was determined as 10% [10]. The fat cells were preincubated for 1 h at 37 °C and shaken at 30 cycles/min, prior to treatment.
2.3. Differentiation of rat preadipocytes into adipocytes The infrantant containing adipose precursor cells was obtained from epididymal fat pads as described above. The infrantant was filtered through a 25-μm nylon mesh and centrifuged at 600×g for 10 min. The pelletted preadipocytes were resuspended in DMEM containing 10% fetal bovine serum, plated, and cultured for 24 h at 37 °C in an atmosphere of 5% CO2. The preadipocytes were differentiated into adipocytes for 3 days in serum-free DMEM-F12 (1:1) medium supplemented with 5 μg/ml insulin, 33 μM biotin, and 200 pM triiodothyronine [10]. The differentiated adipocytes received the heat stress at day 5 of differentiation.
2.4. Heat stress of adipocytes Primary adipocytes were cultured in soft plastic tubes and differentiated adipocytes were grown on glass slides placed in 24-well Costar plates. The individual adipocyte cultures were incubated in the covered water incubators. The virtual (stress) temperature of the culture medium was monitored with use of a thermometer, and indicated in the text.
2.5. Isolation of intact endogenous lipid droplets from primary adipocytes Primary rat adipocytes were lysed in 4 ml pre-warmed (37 °C) hypoosmotic buffer (5 mM Tris–Cl buffer, pH 7.4) [27]. The suspension was gently inverted 3 times and centrifuged for 3 min at 200×g, at room temperature. Endogenous lipid droplets floating on top were collected and washed once in 4 ml pre-warmed (37 °C) buffer (5 mM Tris–Cl, pH 7.4 and 0.025% Triton X-100) by slowly swinging the tube 3 times. The lipid droplets were collected after centrifugation for 3 min at 200×g, and washed twice in the pre-warmed buffer (20 mM Tris–Cl, pH 7.4, 135 mM NaCl, 5 mM KCl, and 1 mM MgCl2). Hoechst 33258 nuclear staining was performed to identify the contaminated intact adipocytes in the lipid droplet preparations. Isolated intact lipid droplets were free of nucleoli and cannot be stained by Hoechst 33258 dye, hence distinguished form intact adipocytes whose nucleoli were positively stained in blue.
2.1. Materials 2.6. Immunostaining Phenol red-free Dulbecco's modified Eagle's medium (fDMEM) containing 5 mM glucose was from Sigma (St. Louis, MO, USA). Rabbit polyclonal antibodies against rat perilipin or HSL were generous gifts from C. Londos at the
Immunostaining was performed as described previously [10,19]. The adipocytes or isolated intact lipid droplets were fixed for 15 min with 4%
68
H. Jiang et al. / Biochimica et Biophysica Acta 1771 (2007) 66–74
paraformaldehyde and 0.01% Triton X-100 in phosphate-buffered saline (PBS, pH 7.2) at room temperature, followed by 3 rinses for 10 min each with PBS. Nonspecific binding sites in the samples were blocked for 60 min with 2% serum derived from the same species as the secondary antibody, and washed for 3 × 5 min; the samples were incubated with the indicated primary antibodies (1:200–1:500) overnight at 4 °C, washed, and subsequently incubated with FITC- or Cy5-conjugated second antibodies at 1:500 for 1 h in the dark at room temperature. Immunofluorescent signal was observed and imaged with use of a Nikon Eclipse TE2000-U microscope.
2.7. Fractionation of the cytosol and fat cake for immunoblotting Following the heat stress, adipocytes were packed by centrifuging for 3 min at 200×g, and homogenized in ice-cold fractionation buffer (20 mM Tris–Cl, pH 7.4, 255 mM sucrose, 1 mM EDTA, 0.1 mM sodium orthovanadate, and 5 mM sodium fluoride), as we described [10]. The tube of the lysate was incubated on ice for 15 min, centrifuged at 20,000×g for 5 min at 4 °C, incubated at 37 °C for 1 min, and then centrifuged at 20,000×g for 30 min at 4 °C. The fat cake containing the lipid droplets floated on top of the tube, with the cytosolic fraction below the layer of fat cake. Both fractions were collected, mixed with concentrated Laemmli sample buffer (62 mM Tris–Cl, pH 6.8, 5% SDS, 1.7% beta-mercaptoethanol, 0.04% bromophenol blue, and 10% glycerol, in final) and adjusted equivalently against protein concentrations [10]. The samples were denatured at 95 °C for 5 min and cleaned at 12,000×g for 10 min, prior to loading on SDS-PAGE.
2.8. Alkaline carbonate treatments of isolated lipid droplets The stressed and unstressed primary adipocytes were lysed in the prewarmed (37 °C) lysis buffer (10 mM Tris–Cl, pH 7.4, 1 mM EDTA), and the lipid droplets were prepared and washed once with use of PBS buffer at room temperature [27]. Alkaline treatment was performed as described previously [28,29] with modification. Briefly, the isolated lipid droplets were diluted 10fold volume in an alkaline solution of 100 mM sodium carbonate (Na2CO3, pH 11.5), or in the lysis buffer (10 mM Tris–Cl, pH 7.4, 1 mM EDTA) that served as the control for the alkaline treatment. The lipid droplets were treated on ice for 30 min and centrifuged for 20 min at 15,000×g at 4 °C. Next, the oil phase floating on the top of the tubes and the aqueous phase below were collected and transferred to new tubes. Both fractions were adjusted and extracted with use of the Laemmli sample buffer containing 5% SDS. The extracts were denatured at 95 °C for 5 min and cleaned at 12,000×g for 5 min, prior to loading on SDSPAGE.
2.9. Immunoblotting Protein extracts were prepared from whole adipocytes or from the cytosol and the lipid droplet fractions. Protein content in the extracts was determined by use of a bicinchoninic acid protein assay kit (Applygen Technologies Inc., Beijing, China). Equal amounts of the proteins were loaded and separated by 10% SDS-PAGE. After electrophoresis, the proteins were transferred to nitrocellulose membranes and immunoblotted as described [13,19,30]. Briefly, the membranes were blocked with 5% non-fat milk in TBS-T buffer (150 mM NaCl, 20 mM Tris–Cl, pH 7.4, 0.05% Tween-20) and incubated at room temperature for 1 h with the indicated primary antibodies. After being washed for 3 × 10 min in TBS-T buffer, the membranes were probed for 1 h with HRPconjugated secondary antibodies. The blots were washed and then developed by use of a Super Enhanced chemiluminescence detection kit (Applygen Technologies Inc., Beijing, China). The protein bands were visualized after exposure of the membranes to Kodak X-ray film.
and centrifuged for 10 min at 15,000×g at 4 °C. This procedure was repeated once. The lower extract was transferred to a new tube and adjusted to 1-ml volume by addition of the above lysis buffer. The supernatant was precleaned by addition and incubation with 15 μl of Protein A-Sepharose mixture for 30 min, followed by a brief centrifugation [13]. Next, rabbit anti-perilipin (1:500) or mouse anti-Hsp70 (1:200) antibodies were added in the supernatant, and then 30 μl of agarose-protein A/G beads was added. Pre-immune serum from rabbit or mouse was used as a control. The immunoprecipitation was performed at 4 °C overnight on a rotator. The immunoprecipitates were centrifuged at 12,000 rpm at 4 °C for 1 min, washed 4 times in the above lysis buffer, and finally reconstituted in Laemmli sample buffer containing 4% SDS and 10 mM dithiothreitol. After being boiled at 95 °C for 5 min, the extract was loaded and separated by SDS-PAGE, and then underwent immunoblot analysis.
3. Results 3.1. Heat stress promotes the expression of HSPs in adipocytes Primary rat adipocytes were stressed and the whole cell lysates were prepared for immunoblot analysis. Heat stress at 40 °C for 60 or 90 min readily promoted the expression of Hsp27, Hsp60, Hsp70, and Hsp90 in adipocytes (Fig. 1). At a higher stress temperature, 41.5 °C, the expression of these Hsps was regulated rapidly within a relatively short (30 or 60 min) period (Fig. 1). Grp78 was only modestly elevated in expression, whereas that of Hsp105 seemed unchanged (data not shown). Hsp70 was induced more markedly as compared with the other Hsps examined. In subsequent experiments, we stimulated adipocytes at 40 °C for 60 min, which significantly induced Hsp expression but did not obviously affect cell viability. 3.2. Hsp70 is translocated to the lipid droplets Primary adipocyte cytosol is occupied by a unilocular lipid droplet that is tightly close to the cell membrane, thus unsuitable for use in immunofluorescent detection of protein association with the lipid droplet. As an alternative, we differentiated rat preadipocytes into adipocytes [10], which have multiple lipid droplets well separated from the cell membranes. The differentiated adipocytes were incubated at 37 °C or heat-
2.10. Immunoprecipitation After the heat treatments, 500 μl of primary adipocytes were packed by centrifuging at 800×g, then lysed on ice for 20 min in 800 μl lysis buffer (50 mM Tris–Cl, pH 7.4, 135 NaCl, 1 mM EDTA, 0.5% Triton X-100, 5 μg/ ml leupeptin, and 5 μg/ml aprotinin, 0.1 mM sodium orthovanadate, and 10 mM sodium fluoride). The lysate was heated at 37 °C for 10 s, vortexed,
Fig. 1. Heat stress induces the expression of HSPs in primary rat adipocytes. After the stress, equal amount of extracts prepared from whole primary adipocytes were separated by SDS-PAGE, followed by immunoblot analysis. The blots shown represent results from 4 separate experiments. HSP, heat shock protein; GADPH, glyceraldehyde-3-phosphate dehydrogenase.
H. Jiang et al. / Biochimica et Biophysica Acta 1771 (2007) 66–74
69
Fig. 2. Heat stress induces the redirection of Hsp70 to the surfaces of the lipid droplets in adipocytes. The differentiated rat adipocytes were stressed at 37 °C or 40 °C for 60 min. The cells were fixed and incubated with use of several primary antibodies against their respective Hsps. After being washed 3 times, the cells were incubated with FITC-conjugated second antibodies. Immunofluorescent signal was observed under a microscope. Hsp27, Hsp60, Hsp70, Hsp90, but not Hsp105, were significantly elevated whereas Grp78 was slightly increased in expression in heat-stressed adipocytes. Only Hsp70 was largely recruited to the lipid droplet surface, which is indicated as the bright ring-loops surrounding the droplets.
stressed at 40 °C for 60 min and then immunofluorescently stained. Fig. 2 shows that Hsp27, Hsp60, Hsp70, and Hsp90 were significantly increased in expression, whereas fluorescence signal of Grp78 was slightly elevated but that of Hsp105 was almost unchanged in the adipocytes upon stimulation with heat stress. Surprisingly, the fluorescence of Hsp70 appeared as bright ring-loops surrounding the lipid droplet surface (Fig. 2), which indicates that the Hsp70 was thermally induced to translocate to the lipid droplets. Grp78, a constitutively expressed member of Hsp70 family [31], was previously proposed to associate with the lipid droplets isolated from the differentiated 3T3-F422A adipocytes [32]. However, our immunostaining (Fig. 2) and immunoblotting (data not shown) failed to detect an association of Grp78 with the lipid droplets in primary and differentiated rat adipocytes under normal or thermal stressed conditions. 3.3. Characterization of Hsp70 translocation to the lipid droplets The lipid droplet (fat cake) and cytosol fractionations of primary adipocytes were prepared for immunoblot analysis. Primary adipocytes incubated for 60 min at different temperatures showed that thermal stress at 38.5 °C only slightly elevated the cytosolic Hsp70 but did not cause the redistribution of Hsp70 to the lipid droplets (Fig. 3A). Hsp70 was markedly translocated to the fat cake fraction of adipocytes stressed at 40 °C; however, the magnitude of the Hsp70 translocation seemed no further enhanced at a higher temperature of 41.5 °C (Fig. 3A). When adipocytes were stressed at 40 °C for different periods, Hsp70 abruptly appeared in the lipid droplet fraction at 30 min and the translocation quickly reached a steady state at approximately 60 min after the thermal stress; further prolonging the stress period to 90 or 120 min only flatly promoted the proportion of Hsp70 in the fat cake fraction (Fig. 3B). Apparently, the thermal induction of the Hsp70 translocation to the lipid droplets was not typically in a temperature- or timedependent manner but occurred rapidly and significantly after stress at 40 °C for 30–60 min. In addition, overexposure of the
Fig. 3. Characterization of Hsp70 translocation to the lipid droplets. Panels A and B, rat primary adipocytes were stressed at different temperatures (A) or for different periods (B). The lipid droplet (Fat cake) and cytosol (Cytosol) fractions as well as whole adipocyte lysates (Whole) were prepared for immunoblot analysis. On the bottom of the panel B, the fat cake blot was overexposed (dark) to show small amounts of Hsp70 detectable in the fat cake fraction of unstressed adipocytes. Panel C, the differentiated rat adipocytes were stressed at 40 °C for 0–100 min and then immunostained with use of anti-Hsp70 antibodies and FITC-conjugated secondary antibodies. The data represent results of 3 separate experiments.
70
H. Jiang et al. / Biochimica et Biophysica Acta 1771 (2007) 66–74
immunoblots indicated that small amounts of Hsp70 were readily detectable in the fat cake fractions of unstressed (37 °C) adipocytes (Fig. 3A, B), which suggests that Hsp70 is a constitutive protein component on the lipid droplets. For better illustration of the dynamic assembly of Hsp70 to the lipid droplets, we immunostained the differentiated adipocytes stressed at 40 °C for 0–100 min. Fig. 3C shows that the fluorescence of Hsp70 surrounding the lipid droplet surface remained minimal during 20-min stress but abruptly increased as ring-loops in differentiated adipocytes stressed for 40–60 min at 40 °C. Comparably, when the stress period was prolonged to 80 or 100 min, the increase in the intensity of Hsp70 fluorescence on the lipid droplets seemed limited (Fig. 3C), which is consistent with our immunoblotting observations (Fig. 3A, B). Thus, the immunoblotting and immunostaining data suggest that the translocation of Hsp70 to the lipid droplets is likely a sharp cellular response to heat stimulation. In addition, when adipocytes were stressed at 40 °C for longer than ∼ 4 h, Hsp70 could be redirected to the nucleoli but not to the cell membranes (data not shown).
3.4. Hsp70 associates with intact lipid droplets isolated from stressed primary adipocytes and is co-localized with perilipin Next, we tried to directly examine the translocation of Hsp70 to the lipid droplets in rat primary epididymal adipocytes. We isolated the intact endogenous lipid droplets from primary adipocytes incubated for 60 min at 37 °C or 40 °C. The isolated lipid droplets and adipocytes were stained with Hoechst 33258 dye and simultaneously immunostained with use of primary antibodies against Hsp70 or perilipin. As shown in Fig. 4A, nucleoli (blue) that were tightly squeezed to the membranes of intact adipocytes were stained but not the isolated lipid droplets that were free of the nucleoli. Perilipin was shown to constitutively localize on the surface of the isolated intact lipid droplets and primary adipocytes under stressed or unstressed conditions. In contrast, Hsp70 appeared as bright ring-loops and was co-localized with perilipin on the surface of intact lipid droplets separated only from stressed (40 °C) but not unstressed (37 °C) adipocytes (Fig. 4A). In stressed intact primary adipocytes, Hsp70 and perilipin were co-localized on
Fig. 4. Hsp70 associates with the isolated intact lipid droplets and is co-localized with perilipin. (A) Immunostaining of intact endogenous lipid droplets isolated from rat primary adipocytes stressed for 60 min at 37 °C or 40 °C. Hsp70 (green, FITC-label) tightly associates with the intact lipid droplets isolated from the stressed (40 °C) but not unstressed (37 °C) adipocytes. Hsp70 (green) and perilipin (red, Cy5-label) were co-localized (yellow, merge) on the surface of the separated lipid droplets. The bottom image on the panel A shows 2 stressed intact primary adipocytes with the lipid droplets and nucleoli close to the cell membranes. Hoechst 33258 dye stained only the nucleoli (blue) of the intact adipocytes but not the isolated endogenous lipid droplets free of the nucleoli. (B) Rat differentiated adipocytes were stressed for 60 min at 40 °C and underwent immunostaining. Hsp70 and perilipin were co-localized on the surface of intracellular lipid droplets under stress conditions. The data represent results of 3 separate experiments. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
H. Jiang et al. / Biochimica et Biophysica Acta 1771 (2007) 66–74
71
the lipid droplets tightly associated with the cell membranes, near which the nucleoli were stained in blue (Fig. 4A). Similarly, in stressed differentiated rat adipocytes, Hsp70 was also clearly co-localized with perilipin on the lipid droplet surface in response to the heat stimulation (Fig. 4B). 3.5. Hsp70 does not directly interact with perilipin Perilipin is the most abundant protein localizing on the lipid droplets in adipocytes [4,33]. The co-localization of Hsp70 and perilipin on the lipid droplets (Fig. 4A, B) implies that Hsp70 might associate with the droplets via interacting with perilipin. Nevertheless, our co-immunoprecipitation experiments did not support the direct interaction between Hsp70 and perilipin in heat-stressed primary adipocytes (Fig. 5). We also examined but failed to detect a co-immunoprecipitated complex of Hsp70 with HSL (data not shown). 3.6. Hsp70 associates with lipid droplets through non-hydrophobic interactions We next examined whether the association of Hsp70 with the lipid droplet is mediated through hydrophobic interactions. Alkaline carbonate treatment is a classic technique for determining hydrophobic interactions between proteins and lipid droplets [28]. The proteins that peripherally associate with lipid droplets via hydrophobic interactions cannot be stripped from the lipid droplets treated by alkaline carbonate [14,23,29,34]. We treated lipid droplets isolated from primary adipocytes stimulated at 40 °C for 60 min in an alkaline solution of 100 mM sodium carbonate (Na2CO3, pH 11.5). After the treatments, the proteins remaining in association with the lipid droplets (oil phase) were extracted and immunoblotted. As shown in Fig. 6, the alkaline treatment did not strip the perilipins from the lipid droplets, which confirms that perilipins tightly bind to lipid droplets through hydrophobic interactions [14,29,34]. By contrast, only a small amount of Hsp70 remained in the oil phase of alkalinally treated lipid droplets, the remaining amount appearing in the aqueous phase (Fig. 6). These data suggest that unlike perilipin, Hsp70 associates with the lipid droplets through noncovalent and non-hydrophobic interactions.
Fig. 5. Hsp70 does not directly interact with perilipin. Stressed primary adipocytes were lysed for immunoprecipitation (IP) assay with use of antiHsp70 (α-Hsp70) or anti-perilipin (α-Peri) antibodies. The immunoprecipitates underwent immunoblotting (IB) analysis. Hsp70 could not co-immunoprecipitate with perilipin. Similarly, Hsp70 did not co-immunoprecipitate with HSL (data not shown). The data represent results of 3 separate experiments. Peri, perilipin; S1, control sera relevant to anti-Hsp70 antibodies; S2, control sera relevant to anti-perilipin polyclonal sera.
Fig. 6. Association of HSP70 with the lipid droplet surface is mediated through non-hydrophobic interactions. Primary adipocytes were stressed for 60 min at 37 °C or 40 °C, and then lysed in buffer (10 mM Tris–HCl, pH 7.4, 1 mM EDTA). The lipid droplets were isolated and treated for 30 min in the carbonate solution (100 mM Na2CO3, pH 11.5) or the above lysis buffer (untreated control). After centrifugation, the floating oil phase of the lipid droplets and the below aqueous phase were transferred to the new tubes and equivalently adjusted in Laemmli sample buffer containing 5% SDS, respectively. Protein extracts were separated by SDS-PAGE and underwent immunoblot analysis. The data represent results of 2 separate experiments.
We also observed that under heat-stressed conditions, a small amount of perilipins may be released from the non-alkalinally treated lipid droplets (oil phase) and appeared in the aqueous phase. This observation implies that the thermal stress could change the capacity of the perilipin association with the lipid droplet surface. 4. Discussion Heat shock proteins are molecular chaperones that are involved in many cellular processes such as protein folding, transportation, and translocation [31,35]. Hsp70 normally localizes in cytosol but may be redirected to the cell membrane and nucleolus in several types of fibroblasts in response to heat stress [36–38]. Using the models of rat primary and differentiated adipocytes, we show that the heat-stress upregulates several Hsps including Hsp27, Hsp60, Hsp70, Hsp90, and Grp78 but not Hsp105; however, only Hsp70 is redirected to the surface or fraction of adipocyte lipid droplets. Approximately one third of the total Hsp70 in adipocytes was quickly assembled to the lipid droplet surface, which occurred abruptly at 30–60 min after the thermal stress at 40 °C. Further increasing stress period and temperature only slightly promoted this translocation, which indicates that the thermal induction of Hsp70 translocation did not occur typically in a temperature- or time-dependent manner. By contrast, translocation of Hsp70 to the cell membrane or nucleolus was a weak and delayed action and occurred when adipocytes were stressed at 40 °C only at ∼4 h or more. Likely, a large assembly of Hsp70 to the lipid droplet surface in adipocytes is an acute and specific cellular response to the heat stimulation. Recent proteomic studies imply that intracellular lipid droplets might associate with chaperone proteins such as Hsp60 [16], Hsp70 [14,17], Hsp73 [16], Hsp90 [16], Grp78/BiP [14–16], or Grp94 [16]. However, except for Hsp70, the other
72
H. Jiang et al. / Biochimica et Biophysica Acta 1771 (2007) 66–74
Hsp proteins were not fluorescently confirmed to associate with adipocyte lipid droplets in our study, which suggests that their proteomic detections in the lipid droplet compartments may be false, probably due to contamination in the sample preparations of proteomic analyses. Overexposure of the immunoblots shows that a small amount of Hsp70 appears in the lipid droplet fraction prepared from unstressed adipocytes, which confirms the previous proteomic findings that Hsp70 is a constitutive protein component on intracellular lipid droplets [14,17]. Nevertheless, inconsistent with our immunoblotting results, immunostaining results failed to detect Hsp70 on lipid droplets in unstressed adipocytes. This discrepancy may be because the level of constitutive Hsp70 on the lipid droplet, in contrast to plentiful cytosolic Hsp70, is too little to be fluorescently detected. Under thermal stress conditions, Hsp70 is responsively elevated in expression and gains considerable strength on the lipid droplet surface in adipocytes. Perilipin is the most abundant protein embedded in phospholipid monolayer of lipid droplets in adipocytes [4,33], which is proposed to serve as a barrier to prevent lipase access to triacylglycerol substrates [2,12]. Our immunostaining results indicate that Hsp70 tightly associated with the intact endogenous lipid droplets isolated from the stressed but not unstressed adipocytes. Moreover, Hsp70 is co-localized with perilipins on the lipid droplet surface in response to the heat stimulation. Nevertheless, immunoprecipitation experiments suggest that Hsp70 does not directly interact with perilipin. Further, the alkaline treatment can effectively strip off most of the Hsp70 but not perilipins from the lipid droplet surface. These observations indicate that the association of Hsp70 with the lipid droplet is neither bridged through perilipin nor mediated through covalent and hydrophobic interactions. Earlier studies reveal that Hsp70 and its cognate protein Hsc70 purified from heat-shocked rat liver are noncovalently bound with endogenous saturated fatty acids, predominantly palmitic and stearic acids and to a lesser extent myristic acid [39,40]. Also, Hsp70 and Hsc70 can directly be incorporated into artificial lipid bilayers via interacting with the membrane moieties consisting of palmitic and oleic acids [41–43]. Moreover, two small Hsps, α-crystallin and HSP17, were shown to associate in vitro with the lipid bilayers and antagonize hyperfluidization of specific membrane microdomains at elevated temperature [44,45]. We speculate that Hsp70 might noncovalently associate with the monolayer microdomains on intracellular lipid droplets, in a manner similar to its interaction with the lipid bilayer moieties composed of specific fatty acids [41–45]. We and others have previously revealed that departure [46,47] or phosphorylation [9–11,48] of perilipins on intracellular lipid droplets might cause conformational changes of the droplet surface and leads to an exposure of the droplet neutral lipids to HSL [9] and non-HSL lipase [10,48], hence enhancing lipolysis. We have also reported that perilipin, likely its homologue ADRP, is stabilized by neutral storage lipids and, in turn, is degraded by proteasome when they become unstable [13,19]. High temperature can increase the transition temperature of the lipid membranes [45,49], thereby causing conformational and functional changes of the lipid layer or the droplet-
associated protein or both. The present study shows that the heat stress might facilitate the departure of perilipins, though low in level, from the lipid droplet surface (Fig. 6). In additional experiments, we observed that the heat stress destabilizes perilipin but slightly increases the translocation of HSL from the cytosol to the lipid droplets in rat primary adipocytes (Xu G., in preparation). We speculate that Hsp70 might temporarily fill in the vacancy of the droplet surface because of the departure of the perilipins, thus stabilizing the lipid droplet monolayer. Alternatively, Hsp70 may chaperone the transportation of HSL or other proteins to, and thereby reside on, the lipid droplet surface. As previously observed in vitro, lipid membrane association of HSP17 and α-crystallin can stabilize the liquid-crystalline state, bind and transfer denatured proteins for subsequent multichaperone-mediated refolding during thermal fluctuations [44,45]. Similarly, the accumulation of Hsp70 on the lipid droplet may function to chaperone the denatured proteins on the droplet for subsequent refolding, and preserve conformational and functional integrities of the droplet membranes and the droplet associated proteins under the thermal stress conditions. Though the functions are unclear, the translocation of Hsp70 to the lipid droplet is an intriguing new aspect. Adipose tissue serves as the major energy reservior. Hyperthermia is a common symptom in many mammal diseases, under which adipose tissues are thermally stressed and must release more free fatty acids for use as energy supplements and consumptive thermogenesis. We recently observed that heat stress directly stimulates adipocyte lipolysis (Xu G., in preparation), which may be a cellular basis for elevated circulating fatty acids under hyperthermal conditions. Our current efforts aim to explore the signal pathway of the Hsp70 translocation to the lipid droplets and examine whether this translocation is linked to accelerated lipolysis response to heat stress. Acknowledgements This work was supported by grants 30670779 and 30270506 from the National Natural Science Foundation of China and by grants 5042015 and 5072030 from the Natural Science Foundation of Beijing Province of China. This work was supported by the Program for New Century Excellent Talents in the University, of the Education Ministry of China (#NECT-040023), and by the Major State Basic Research Development Program of China (#2006CB503903). We thank Dr. Constantine Londos at the U.S. National Institute of Diabetes and Digestive and Kidney Diseases for kind gifts of anti-perilipin and anti-HSL antibodies. References [1] D. Zweytick, K. Athenstaedt, G. Daum, Intracellular lipid particles of eukaryotic cells, Biochim. Biophys. Acta 1469 (2000) 101–120. [2] C. Londos, C. Sztalryd, J.T. Tansey, A.R. Kimmel, Role of PAT proteins in lipid metabolism, Biochimie 87 (2005) 45–49. [3] D.A. Servetnick, D.L. Brasaemle, J. Gruia-Gray, A.R. Kimmel, J. Wolff, C. Londos, Perilipins are associated with cholesteryl ester droplets in steroidogenic adrenal cortical and Leydig cells, J. Biol. Chem. 270 (1995) 16970–16973.
H. Jiang et al. / Biochimica et Biophysica Acta 1771 (2007) 66–74 [4] E.J. Blanchette-Mackie, N.K. Dwyer, T. Barber, R.A. Coxey, T. Takeda, C. M. Rondinone, J.L. Theodorakis, A.S. Greenberg, C. Londos, Perilipin is located on the surface layer of intracellular lipid droplets in adipocytes, J. Lipid Res. 36 (1995) 1211–1226. [5] D.L. Brasaemle, T. Barber, N.E. Wolins, G. Serrero, E.J. BlanchetteMackie, C. Londos, Adipose differentiation-related protein is an ubiquitously expressed lipid storage droplet-associated protein, J. Lipid Res. 38 (1997) 2249–2263. [6] N.E. Wolins, B.K. Quaynor, J.R. Skinner, M.J. Schoenfish, A. Tzekov, P.E. Bickel, S3-12, Adipophilin, and TIP47 package lipid in adipocytes, J. Biol. Chem. 280 (2005) 19146–19155. [7] S. Miura, J.W. Gan, J. Brzostowski, M.J. Parisi, C.J. Schultz, C. Londos, B. Oliver, A.R. Kimmel, Functional conservation for lipid storage droplet association among perilipin, ADRP, and TIP47 (PAT)related proteins in mammals, drosophila, and dictyostelium, J. Biol. Chem. 277 (2002) 32253–32257. [8] X. Lu, J. Gruia-Gray, N.G. Copeland, D.J. Gilbert, N.A. Jenkins, C. Londos, A.R. Kimmel, The murine perilipin gene: the lipid dropletassociated perilipins derive from tissue-specific, mRNA splice variants and define a gene family of ancient origin, Mamm. Genome. 12 (2001) 741–749. [9] C. Sztalryd, G. Xu, H. Dorward, J.T. Tansey, J.A. Contreras, A.R. Kimmel, C. Londos, Perilipin A is essential for the translocation of hormonesensitive lipase during lipolytic activation, J. Cell Biol. 161 (2003) 1093–1103. [10] J. He, H. Jiang, J.T. Tansey, T. Tang, S. Pu, G. Xu, Calyculin and okadaic acid promote perilipin phosphorylation and increase lipolysis in primary rat adipocytes, Biochim. Biophys. Acta 1761 (2006) 247–255. [11] J.T. Tansey, A.M. Huml, R. Vogt, K.E. Davis, J.M. Jones, K.A. Fraser, D. L. Brasaemle, A.R. Kimmel, C. Londos, Functional studies on native and mutated forms of perilipins. A role in protein kinase A-mediated lipolysis of triacylglycerols, J. Biol. Chem. 278 (2003) 8401–8406. [12] D.L. Brasaemle, B. Rubin, I.A. Harten, J. Gruia-Gray, A.R. Kimmel, C. Londos, Perilipin A increases triacylglycerol storage by decreasing the rate of triacylglycerol hydrolysis, J. Biol. Chem. 275 (2000) 38486–38493. [13] G. Xu, C. Sztalryd, X. Lu, J.T. Tansey, J.W. Gan, H. Dorward, A.R. Kimmel, C. Londos, Post-translational regulation of adipose differentiation-related protein by the ubiquitin/proteasome pathway, J. Biol. Chem. 280 (2005) 42841–42847. [14] D.L. Brasaemle, G. Dolios, L. Shapiro, R. Wang, Proteomic analysis of proteins associated with lipid droplets of basal and lipolytically stimulated 3T3-L1 adipocytes, J. Biol. Chem. 279 (2004) 46835–46842. [15] P. Liu, Y. Ying, Y. Zhao, D.I. Mundy, M. Zhu, R.G. Anderson, Chinese hamster ovary K2 cell lipid droplets appear to be metabolic organelles involved in membrane traffic, J. Biol. Chem. 279 (2004) 3787–3792. [16] E. Umlauf, E. Csaszar, M. Moertelmaier, G.J. Schuetz, R.G. Parton, R. Prohaska, Association of stomatin with lipid bodies, J. Biol. Chem. 279 (2004) 23699–23709. [17] C.C. Wu, K.E. Howell, M.C. Neville, J.R. Yates III, J.L. McManaman, Proteomics reveal a link between the endoplasmic reticulum and lipid secretory mechanisms in mammary epithelial cells, Electrophoresis 21 (2000) 3470–3482. [18] Y. Fujimoto, H. Itabe, J. Sakai, M. Makita, J. Noda, M. Mori, Y. Higashi, S. Kojima, T. Takano, Identification of major proteins in the lipid dropletenriched fraction isolated from the human hepatocyte cell line HuH7, Biochim. Biophys. Acta 1644 (2004) 47–59. [19] G. Xu, C. Sztalryd, C. Londos, Degradation of perilipin is mediated through ubiquitination-proteasome pathway, Biochim. Biophys. Acta 1761 (2006) 83–90. [20] V. Subramanian, A. Rothenberg, C. Gomez, A.W. Cohen, A. Garcia, S. Bhattacharyya, L. Shapiro, G. Dolios, R. Wang, M.P. Lisanti, D.L. Brasaemle, Perilipin A mediates the reversible binding of CGI-58 to lipid droplets in 3T3-L1 adipocytes, J. Biol. Chem. 279 (2004) 42062–42071. [21] D.L. Brasaemle, D.M. Levin, D.C. Adler-Wailes, C. Londos, The lipolytic stimulation of 3T3-L1 adipocytes promotes the translocation of cytosolic hormone-sensitive lipase to the surfaces of lipid storage droplets, Biochim. Biophys. Acta 1483 (2000) 251–262. [22] A.G. Ostermeyer, J.M. Paci, Y. Zeng, D.M. Lublin, S. Munro, D.A.
[23]
[24]
[25]
[26]
[27] [28]
[29]
[30]
[31] [32]
[33]
[34]
[35] [36]
[37]
[38]
[39]
[40]
[41]
[42]
73
Brown, Accumulation of caveolin in the endoplasmic reticulum redirects the protein to lipid storage droplets, J. Cell Biol. 152 (2001) 1071–1078. K. Tauchi-Sato, S. Ozeki, T. Houjou, R. Taguchi, T. Fujimoto, The surface of lipid droplets is a phospholipid monolayer with a unique fatty acid composition, J. Biol. Chem. 277 (2002) 44507–44512. S.M. Wang, T.H. Fong, S.Y. Hsu, C.L. Chien, J.C. Wu, Reorganization of a novel vimentin-associated protein in 3T3-L1 cells during adipose conversion, J. Cell. Biochem. 67 (1997) 84–91. S. Martin, K. Driessen, S.J. Nixon, M. Zerial, R.G. Parton, Regulated localization of Rab18 to lipid droplets: effects of lipolytic stimulation and inhibition of lipid droplet catabolism, J. Biol. Chem. 280 (2005) 42325–42335. M. Ohashi, N. Mizushima, Y. Kabeya, T. Yoshimori, Localization of mammalian NAD(P)H steroid dehydrogenase-like protein on lipid droplets, J. Biol. Chem. 278 (2003) 36819–36829. H. Okuda, C. Morimoto, T. Tsujita, Role of endogenous lipid droplets in lipolysis in rat adipocytes, J. Lipid Res. 35 (1994) 36–44. Y. Fujiki, A.L. Hubbard, S. Fowler, P.B. Lazarow, Isolation of intracellular membranes by means of sodium carbonate treatment: application to endoplasmic reticulum, J. Cell Biol. 93 (1982) 97–102. V. Subramanian, A. Garcia, A. Sekowski, D.L. Brasaemle, Hydrophobic sequences target and anchor perilipin A to lipid droplets, J. Lipid Res. 45 (2004) 1983–1991. G. Xu, C. Rabadan-Diehl, M. Nikodemova, P. Wynn, J. Spiess, G. Aguilera, Inhibition of corticotropin releasing hormone type-1 receptor translation by an upstream AUG triplet in the 5′ untranslated region, Mol. Pharmacol. 59 (2001) 485–492. V.R. Agashe, F.U. Hartl, Roles of molecular chaperones in cytoplasmic protein folding, Semin. Cell Dev. Biol. 11 (2000) 15–25. S. Prattes, G. Horl, A. Hammer, A. Blaschitz, W.F. Graier, W. Sattler, R. Zechner, E. Steyrer, Intracellular distribution and mobilization of unesterified cholesterol in adipocytes: triglyceride droplets are surrounded by cholesterol-rich ER-like surface layer structures, J. Cell Sci. 113 (Pt 17) (2000) 2977–2989. A.S. Greenberg, J.J. Egan, S.A. Wek, N.B. Garty, E.J. Blanchette-Mackie, C. Londos, Perilipin, a major hormonally regulated adipocyte-specific phosphoprotein associated with the periphery of lipid storage droplets, J. Biol. Chem. 266 (1991) 11341–11346. A. Garcia, A. Sekowski, V. Subramanian, D.L. Brasaemle, The central domain is required to target and anchor perilipin A to lipid droplets, J. Biol. Chem. 278 (2003) 625–635. J.G. Kiang, G.C. Tsokos, Heat shock protein 70 kDa: molecular biology, biochemistry, and physiology, Pharmacol. Ther. 80 (1998) 183–201. W.J. Welch, J.P. Suhan, Morphological study of the mammalian stress response: characterization of changes in cytoplasmic organelles, cytoskeleton, and nucleoli, and appearance of intranuclear actin filaments in rat fibroblasts after heat-shock treatment, J. Cell Biol. 101 (1985) 1198–1211. S. Ellis, M. Killender, R.L. Anderson, Heat-induced alterations in the localization of HSP72 and HSP73 as measured by indirect immunohistochemistry and immunogold electron microscopy, J. Histochem. Cytochem. 48 (2000) 321–332. A.H. Broquet, G. Thomas, J. Masliah, G. Trugnan, M. Bachelet, Expression of the molecular chaperone Hsp70 in detergent-resistant microdomains correlates with its membrane delivery and release, J. Biol. Chem. 278 (2003) 21601–21606. P.T. Guidon Jr., L.E. Hightower, Purification and initial characterization of the 71-kilodalton rat heat-shock protein and its cognate as fatty acid binding proteins, Biochemistry 25 (1986) 3231–3239. P.T. Guidon Jr., L.E. Hightower, The 73 kilodalton heat shock cognate protein purified from rat brain contains nonesterified palmitic and stearic acids, J. Cell. Physiol. 128 (1986) 239–245. N. Arispe, A. De Maio, ATP and ADP modulate a cation channel formed by Hsc70 in acidic phospholipid membranes, J. Biol. Chem. 275 (2000) 30839–30843. N. Arispe, M. Doh, O. Simakova, B. Kurganov, A. De Maio, Hsc70 and Hsp70 interact with phosphatidylserine on the surface of PC12 cells resulting in a decrease of viability, FASEB J. 18 (2004) 1636–1645.
74
H. Jiang et al. / Biochimica et Biophysica Acta 1771 (2007) 66–74
[43] N. Arispe, M. Doh, A. De Maio, Lipid interaction differentiates the constitutive and stress-induced heat shock proteins Hsc70 and Hsp70, Cell Stress Chaperones 7 (2002) 330–338. [44] Z. Torok, P. Goloubinoff, I. Horvath, N.M. Tsvetkova, A. Glatz, G. Balogh, V. Varvasovszki, D.A. Los, E. Vierling, J.H. Crowe, L. Vigh, Synechocystis HSP17 is an amphitropic protein that stabilizes heatstressed membranes and binds denatured proteins for subsequent chaperone-mediated refolding, Proc. Natl. Acad. Sci. U. S. A. 98 (2001) 3098–3103. [45] N.M. Tsvetkova, I. Horvath, Z. Torok, W.F. Wolkers, Z. Balogi, N. Shigapova, L.M. Crowe, F. Tablin, E. Vierling, J.H. Crowe, L. Vigh, Small heat-shock proteins regulate membrane lipid polymorphism, Proc. Natl. Acad. Sci. U. S. A. 99 (2002) 13504–13509.
[46] T. Ren, J. He, H. Jiang, L. Zu, S. Pu, X. Guo, G. Xu, Metformin reduces lipolysis in primary rat adipocytes stimulated by tumor necrosis factoralpha or isoproterenol, J. Mol. Endocrinol. 37 (2006) 175–183. [47] S.C. Souza, L.M. de Vargas, M.T. Yamamoto, P. Lien, M.D. Franciosa, L.G. Moss, A.S. Greenberg, Overexpression of perilipin A and B blocks the ability of tumor necrosis factor alpha to increase lipolysis in 3T3-L1 adipocytes, J. Biol. Chem. 273 (1998) 24665–24669. [48] H.H. Zhang, S.C. Souza, K.V. Muliro, F.B. Kraemer, M.S. Obin, A.S. Greenberg, Lipase-selective functional domains of perilipin A differentially regulate constitutive and protein kinase A-stimulated lipolysis, J. Biol. Chem. 278 (2003) 51535–51542. [49] B. de Kruijff, Polymorphic regulation of membrane lipid composition, Nature 329 (1987) 587–588.