SEARCH

SEARCH BY CITATION

Abstract

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References
  8. Supporting Information

Caveolae are a subtype of cholesterol-enriched lipid microdomains/rafts that are routinely detected as vesicles pinching off from the plasma membrane. Caveolin-1 is an essential component of caveolae. Hepatic caveolin-1 plays an important role in liver regeneration and lipid metabolism. Expression of caveolin-1 in hepatocytes is relatively low, and it has been suggested to also reside at other subcellular locations than the plasma membrane. Recently, we found that the peroxisomal membrane contains lipid microdomains. Like caveolin-1, hepatic peroxisomes are involved in lipid metabolism. Here, we analyzed the subcellular location of caveolin-1 in rat hepatocytes. The subcellular location of rat hepatocyte caveolin-1 was analyzed by cell fractionation procedures, immunofluorescence, and immuno-electron microscopy. Green fluorescent protein (GFP)-tagged caveolin-1 was expressed in rat hepatocytes. Lipid rafts were characterized after Triton X-100 or Lubrol WX extraction of purified peroxisomes. Fenofibric acid–dependent regulation of caveolin-1 was analyzed. Peroxisome biogenesis was studied in rat hepatocytes after RNA interference–mediated silencing of caveolin-1 and caveolin-1 knockout mice. Cell fractionation and microscopic analyses reveal that caveolin-1 colocalizes with peroxisomal marker proteins (catalase, the 70 kDa peroxisomal membrane protein PMP70, the adrenoleukodystrophy protein ALDP, Pex14p, and the bile acid–coenzyme A:amino acid N-acyltransferase BAAT) in rat hepatocytes. Artificially expressed GFP–caveolin-1 accumulated in catalase-positive organelles. Peroxisomal caveolin-1 is associated with detergent-resistant microdomains. Caveolin-1 expression is strongly repressed by the peroxisome proliferator-activated receptor-α agonist fenofibric acid. Targeting of peroxisomal matrix proteins and peroxisome number and shape were not altered in rat hepatocytes with 70%-80% reduced caveolin-1 levels and in livers of caveolin-1 knockout mice. Conclusion: Caveolin-1 is enriched in peroxisomes of hepatocytes. Caveolin-1 is not required for peroxisome biogenesis, but this unique subcellular location may determine its important role in hepatocyte proliferation and lipid metabolism. (HEPATOLOGY 2010.)

Caveolae are subtypes of lipid microdomains/rafts that are morphologically recognizable as flask-like invaginations of the plasma membrane. They are particularly involved in signal transduction and endocytosis.1 The characteristic protein component of caveolae are the caveolins that interact strongly with cholesterol.2 Three different caveolins (caveolin-1, caveolin-2, and caveolin-3) have been described. Caveolin-1 is crucial for the formation of caveolae in non–muscle cells1 and is highly expressed in lung, heart, and adipose tissue.3 In accordance with this expression profile, severe pulmonary defects, abnormal cardiac function, and lipid disorders have been reported in caveolin-1 knockout mice.4–6

Caveolin-1 is only moderately expressed in the liver,7, 8 where it has been detected in hepatocytes, Kupffer cells, stellate cells, and endothelial cells.8–12 Still, hepatic caveolin-1 is involved in important metabolic pathways such as (intracellular) cholesterol trafficking13 and lipid homeostasis.14 The important function of hepatic caveolin-1 is evident from the impaired liver regeneration and low survival of caveolin-1−/− mice after partial hepatectomy.15 In hepatocytes, caveolin-1 has been detected in the plasma membrane. However, significant amounts of caveolin-1 have also been detected intracellularly, where it has been reported to localize to lipid droplets, endoplasmic reticulum, Golgi, and/or mitochondria.3, 8, 14 We became interested in a putative peroxisomal localization of caveolin-1 after we detected that the peroxisomal membrane contains lipid microdomains/rafts (J. Woudenberg et al., unpublished data).

Peroxisomes are required for important hepatocyte functions like bile acid biosynthesis and β-oxidation of very long chain fatty acids (VLCFAs).16, 17 Typical marker proteins for peroxisomes are the peroxins that are required for peroxisome biogenesis, the peroxisomal antioxidant enzyme catalase, and the peroxisomal ATP-binding cassette (ABC) transporters, the adrenoleukodystrophy protein (ALDP, ABCD1) and 70 kDa peroxisomal membrane protein (PMP70).18, 19 Recently, we found that ALDP, PMP70, and the peroxins Pex13p and Pex14p are differentially associated with lipid rafts in human peroxisomes (J. Woudenberg et al., unpublished data). Cellular depletion of cholesterol leads to defective sorting of the peroxisomal enzyme catalase, indicating that lipid rafts are required for peroxisome biogenesis.

Because caveolae are a subtype of lipid rafts, we analyzed in this study the subcellular location of caveolin-1 in rat hepatocytes using biochemical and microscopic techniques. Caveolin-1 was predominantly detected in peroxisomes. Expression of caveolin-1 is strongly reduced after exposure to fenofibric acid, a strong ligand for the peroxisome proliferator-activated receptor alpha (PPARα). Peroxisome biogenesis defects were not observed after RNA-mediated silencing of caveolin-1 in rat hepatocytes and in livers of caveolin-1 knockout mice. We discuss these findings in relation to the putative function of caveolin-1 in liver peroxisomes.

Materials and Methods

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References
  8. Supporting Information

Animals.

Specified pathogen-free male Wistar rats (220-250 g) were purchased from Charles River Laboratories Inc. (Wilmington, MA). They were housed under standard laboratory conditions with free access to standard laboratory chow and water. Experiments were performed following the guidelines of the local Committee for Care and Use of Laboratory Animals.

Caveolin-1 knockout and wild-type mice4, 15 were kept under a controlled humidity and lighting schedule with a 12-hour dark period. All animals received human care in compliance with institutional guidelines regulated by the European Community. Food and water were available ad libitum. Male mice were sacrificed at the age of 12 weeks after which livers were removed and processed for western blotting or immunofluorescence microscopy.

Primary Cells and Culture Conditions.

Hepatocytes were isolated and cultured in William's E medium as described.20 Cells were cultured in a humidified incubator at 37°C and 5% CO2. Hepatocyte viability and purity were always more than 90% as assessed by trypan blue exclusion.

Plasmids and Transient Transfection.

Full-length mouse caveolin-1 complementary DNA (cDNA) was obtained from C2C12 myoblast cells by polymerase chain reaction (PCR) and cloned into pGEM-T-easy (Promega, Leiden, Netherlands). Constructs were sequenced and a clone without errors was selected. BglII and SalI restriction sites were added to the caveolin-1 cDNA by PCR using adapter-primers (For-BglII: 5′-GGACTCAGATCT-ATGTCTGGGGGCAAATACGTGGAC-3′, Rev-SalI: 5′-TACAAGAGTCGA-CTGCGAGAGCAACTTGGAAT- TG-3′). The resulting PCR product was inserted as a BglII-SalI DNA fragment in the corresponding sites of pEGFP-C1 (Clontech, Palo Alto, CA), resulting in an expression vector with caveolin-1 fused to the C-terminus of enhanced green fluorescent protein (EGFP).

Primary rat hepatocytes were transiently transfected with EGFP–caveolin-1 using electroporation, as described.21 After 48 hours, the expression of the hybrid protein was analyzed by western blotting and the subcellular location of EGFP–caveolin-1 was analyzed using fluorescence microscopy.

Caveolin-1 RNA Interference.

Primary rat hepatocytes were plated at a density of 1.25 × 105 cells/cm2 in William's E Medium with Glutamax supplemented with 1% heat-inactivated fetal bovine serum (Invitrogen) and dexamethasone (Sigma-Aldrich, St. Louis, MO) in the presence of double-stranded siRNA duplexes (Table 1) aimed to silence caveolin-1 (siRNA-Cav, Invitrogen). Control cells were transfected with oligonucleotides directed against luciferase (siRNA-Luc, Invitrogen). Lipofectamine (Invitrogen) was used as transfection reagent according to the manufacturer's instructions. After 24 hours, cells were either fixed for immunofluorescence microscopy or lysed for quantitative PCR or western blot analysis.

Table 1. Primers Used for Caveolin-1 RNA Interference
RNASenseAntisenseCompany/Reference
luciferase5′-cuu acg cug agu acu ucg auu-3′5′-ucg aag uac uca gcg uaa guu-3′Invitrogen
caveolin-15′-aau cuc aau cag gaa gcu cuu -3′5′-gag cuu ccu gau uga gau uuu-3′Invitrogen38

Fenofibric Acid Treatments.

Primary rat hepatocytes were plated as described previously.20 Four hours after plating, cells were incubated with 50 μM fenofibric acid (Sigma-Aldrich) or 0.1% dimethyl sulfoxide (solvent for fenofibric acid) for 24 hours. Cells were lysed for quantitative PCR or western blot analysis.

RNA isolation and Quantitative PCR.

The isolation of total RNA, its conversion to cDNA and its analysis by quantitative PCR was carried out as described previously.22 Primers and probes used in this study are listed in Table 2. The expression of each gene of interest was normalized with respect to the endogenous control, 18S (ΔΔCt method).

Table 2. Sequences of Primers and Probes Used for Real-Time Quantitative PCR Analysis
GeneSenseAntisenseProbe
18S5′-cgg cta cca cat cca agg a -3′5′-cca att aca ggg cct cga aa-3′5′FAM-cgc gca aat tac cca ctc ccg a-TAMRA3′
caveolin-15′-aac cgc gac ccc aag c-3′5′-ccg caa tca cat ctt caa agt c-3′5′FAM-tct caa cga cga cgt ggt caa gat-TAMRA3′
Pex11p5′-gcc cgc cac tac tac tat ttc ct-3′5′-tct gtc'gcg tgc aac ttg tc-3′5′FAM-cat atg cag caa gac ctc ata cag atc ccg -TAMRA3′
PMP705′-ctg gtg ctg gag aaa tca tca at-3′5′-cca gat cga act tca aaa cta agg t-3′5′FAM-tga tca tgt tcc ttt agc aac acc aaa tgg-TAMRA3′
AOX5′-gcc acg gaa ctc atc ttc ga-3′5′-cca ggc cac cac tta atg ga-3′5′FAM-cca ctg cca cat atg acc cca aga ccc-TAMRA3′
Baat5′-tgt aga gtt tct cct gag aca tcc taa-3′5′-gtc caa tct ctg ctc caa tgc-3′5′FAM-tgc caa ccc ctg ggc cca g-TAMRA3′
catalase5′-gga tta tgg cct ccg aga tct-3′5′-acc ttg gtc agg tca aat gga t-3′5′FAM-atg cca tcg cca gtg gca att acc-TAMRA3′

Subcellular Fractionation and Isolation of Peroxisomes.

The subcellular fractionation and isolation of peroxisomes from rat liver was performed as described previously by Antonenkov et al. using isolation medium-3.23 Peroxisomes were further purified from the 17,000g pellet by Nycodenz density gradient centrifugation according to established methods.24 For western blot analysis, pellet fractions were resuspended in the same volume as the corresponding supernatant fractions and equal volumes were loaded.

Isolation of Detergent-Resistant Microdomains from Peroxisomes.

Purified peroxisomal fractions from the Nycodenz gradients were pooled and centrifuged in an SS34 rotor at 17,000g. The pellets (1-1.5 mg protein) were subjected to extraction by 1% Triton X-100 (Sigma-Aldrich) or 1% Lubrol WX (Sigma-Aldrich) for the isolation of Triton X-100- or Lubrol WX-resistant rafts, respectively, according to established methods.25 Equal volume fractions of the gradient were analyzed by western blotting.

SDS-PAGE and Western Blotting.

Protein samples were separated by sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) and analyzed by western blotting as described before.21 Protein concentrations were determined using the Bio-Rad Protein Assay system (Bio-Rad Laboratories, Hercules, CA) using bovine serum albumin as standard. Primary antibodies used are listed in Table 3. Horseradish peroxidase (HRP)-conjugated secondary antibodies (HRP-conjugated swine anti-rabbit, rabbit anti-goat and rabbit anti-mouse, Dako A/S, Glostrup, Denmark) and the phototope-HRP western blot Detection System (Cell Signaling Technology Inc., Danvers, MA) were used for detection according to the manufacturers' protocols. The blots were exposed in a ChemiDoc XRS system (Bio-Rad Laboratories).

Table 3. Antibody Dilutions for Protein Analysis
AntibodyWestern BlottingImmunofluorescence MicroscopyElectron MicroscopyCompany/Reference
Mouse α-ALDP1:10001:100 Clone 1D6; Euromedex, Mundolsheim France
Rabbit α-Baat1:2000  Generous gift of Prof. C. Falany, Birmingham, AL39
Rabbit α-BSEP1:2000  Vos et al.40
Rabbit α-calnexin1:2000  SPA 860D; Stressgen, Ann Arbor, MI
Rabbit α-catalase1:20001:200 Calbiochem, La Jolla, CA
Mouse α-catalase1:20001:2001:200Sigma-Aldrich, St. Louis, MO
Rabbit α-caveolin-11:10001:1001:100N20; Santa Cruz Biotechnology Inc., Santa Cruz, CA
Mouse α-cytochrome C1:2000  BD Biosciences, Franklin Lakes, NJ
Mouse α-Gapdh1:10.000  Calbiochem, La Jolla, CA
Mouse α-EGFP1:1000  Roche Diagnostic, Almere, The Netherlands
Goat α-Pex13p1:1000  Abcam, Cambridge, UK
Rabbit α-Pex14p1:2000  Generous gift of Dr. M. Fransen, Leuven, Belgium41
Rabbit α-PMP701:1000  Sigma-Aldrich, St. Louis, MO
Rabbit α-c-Src1:500  sc18; Santa Cruz Biotechnology Inc., Santa Cruz, CA

Immunofluorescence Microscopy.

Immunofluorescence microscopy was performed on 4-μm thick sections from paraffin-embedded mouse liver as well as on paraformaldehyde (4%)-fixed primary rat hepatocytes. Sections were deparaffinized in xylene followed by rehydration, and antigen retrieval was performed using ethylene diamine tetraacetic acid buffer. Liver sections and hepatocytes were labeled and analyzed as described.21 Primary antibody dilutions are listed in Table 3. Images were captured with a TCS SP2/AOBS confocal laser scanning microscope (Leica, Heidelberg, Germany).

Statistical Analysis.

All numerical results are reported as the mean of at least three independent experiments ± standard error of the mean.

Results

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References
  8. Supporting Information

Endogenous Caveolin-1 Colocalizes with Peroxisomal Markers in Rat Hepatocytes.

To study the presence and location of caveolin-1 in rat hepatocytes, we performed immunofluorescence microscopy using specific antibodies against caveolin-1 and the peroxisomal markers catalase and ALDP (Fig. 1). The anti–caveolin-1–dependent signal was low, in line with the limited expression of this protein in liver8 (Supporting Fig. 1). When observing the caveolin-1 staining alone, no distinct localization pattern is evident, with significant dotted intracellular signals, together with nuclear and plasma membrane signals (Fig. 1B). However, when combined with a specific staining for catalase (Fig. 1, overlay in C,F,I) or ALDP (Fig. 1, overlay in L,O,R), a significant colocalization of caveolin-1 with these peroxisomal markers is evident (Fig. 1, merged images, arrows). Some peroxisomes appeared to be (almost) devoid of caveolin-1, especially those at the rim of hepatocytes (Fig. 1F,O).

thumbnail image

Figure 1. Caveolin-1 colocalizes with peroxisomal markers in rat hepatocytes. Primary rat hepatocytes were analyzed by immunofluorescence microscopy to determine the subcellular location of (A,D,G) catalase, (J,M,P) ALDP, and (B,E,H and K,N,Q) caveolin-1. The merged images are displayed in (C,F,I,L,O,R). The high magnification images in (G-I) and (P-R) show strong colocalization between caveolin-1 and catalase and ALDP, respectively (arrows). Some catalase-positive peroxisomes contain little amounts of caveolin-1 (I,R; closed arrowheads). Caveolin-1 is also detected at the plasma membrane of hepatocytes (B,K; open arrowheads).

Download figure to PowerPoint

Heterologously Expressed Caveolin-1 Partly Sorts to Hepatocyte Peroxisomes.

To obtain further proof for a peroxisomal location of hepatic caveolin-1, we transiently transfected primary rat hepatocytes with an expression plasmid producing enhanced green fluorescent protein (EGFP)-tagged caveolin-1. Using antibodies against EGFP, a protein band with the expected molecular weight of 49 kDa was readily detected in transfected rat hepatocytes, whereas this band was absent in untransfected cells (Fig. 2A). Fluorescence microscopy revealed that EGFP–caveolin-1 was detected in dotted structures that also contain (endogenous) catalase, as indicated by the yellow stain in the merged images (Fig. 2E-G, arrows). EGFP–caveolin-1 was also detected at other cellular locations, in particular in cells that highly express the hybrid protein (Supporting Fig. 2).

thumbnail image

Figure 2. In rat hepatocytes, EGFP–caveolin-1 partly sorts to catalase-positive peroxisomes. Primary rat hepatocytes were transiently transfected with EGFP-caveolin-1. At 48 hours after transfection, expression of transfected (EGFP–caveolin-1, 49 kDa) and endogenous caveolin-1 (21 kDa) was analyzed by western blotting. (A) Gapdh expression was analyzed as a marker for equal protein loading. EGFP–caveolin-1–transfected hepatocytes were analyzed by immunofluorescence microscopy to determine the subcellular location of (B,E) EGFP–caveolin-1 or (C,F) catalase. The merged images are displayed in (D) and (G). In the high magnification images, clear colocalization between caveolin-1 and catalase in dotted structures (arrows) can be observed. Some catalase-positive peroxisomes contain little amounts of caveolin-1 (G, arrowheads).

Download figure to PowerPoint

Endogenous Caveolin-1 Cofractionates with Peroxisomal Proteins After Subcellular Fractionation of Rat Liver.

To confirm the presence of caveolin-1 in hepatic peroxisomes, we performed cell fractionation studies on rat liver. After differential centrifugation of osmotically-stabilized total liver homogenates, caveolin-1 was almost exclusively detected in the 17,000g pellet fraction also containing the markers for peroxisomes, mitochondria endoplasmic reticulum, and plasma membranes (Fig. 3A). The cytosolic marker glyceraldehyde 3-phosphate dehydrogenase (Gapdh) did not appear in any of the pellet fractions (Fig. 3A). Next, the organelles in the 17,000g pellet fraction were separated using Nycodenz density gradient centrifugation. Western blot analyses showed that the peroxisomal markers PMP70, Pex14p, catalase, and Baat were enriched in fractions 3-6 (Fig. 3B), the unique high density location of peroxisomes in these gradients.24 Caveolin-1 was also highly enriched in these fractions and clearly separated away from mitochondria (cytochrome C, fractions 12-14), canalicular membranes (Bsep, fractions 12-15), plasma membranes (c-Src, fractions 15,16), and the endoplasmic reticulum (calnexin, fractions 12-15). These data show that caveolin-1 predominantly copurifies with peroxisomes during cell fractionation studies of rat liver. Only minor amounts may be present in the plasma membrane, mitochondria, endoplasmic reticulum, and/or other organelles not analyzed here.

thumbnail image

Figure 3. Endogenous caveolin-1 cofractionates with peroxisomal proteins after subcellular fractionation of rat liver. Total rat liver homogenates were fractionated by differential centrifugation. (A) Equal volumes of all pellet and supernatant fractions were analyzed by western blotting, using specific antibodies against PMP70, Pex14p, catalase, Baat, caveolin-1, Bsep, calnexin, Gapdh, and cytochrome C. The organelles in the 17,000g pellet fraction were separated using Nycodenz density gradient centrifugation. (B) Equal volumes of the gradient fractions were analyzed as described in (A). In addition, the distribution of the plasma membrane protein c-Src was analyzed in the Nycodenz density gradient.

Download figure to PowerPoint

Caveolin-1 Is Localized to Detergent-Resistant Microdomains in the Peroxisomal Membrane.

In order to determine whether peroxisomal caveolin-1 is localized to detergent-resistant microdomains, we treated purified rat liver peroxisomes with Triton X-100 or Lubrol WX followed by flotation gradient centrifugation. Equal volumes of the gradient fractions were analyzed by western blotting. Caveolin-1 largely resisted extraction by Triton X-100, as indicated by its flotation to fractions 3-5. All peroxisomal marker proteins analyzed, PMP70, Pex14p, Pex13p, catalase, and Baat, were detected in the Triton X-100–solubilized fractions 8-12 (Fig. 4A). Importantly, significant amounts of Pex14p, PMP70, and caveolin-1 resisted Triton X-100 extraction when performed on total hepatocyte cell lysates (Supporting Fig. 3A) as reported before in human HepG2 cells (J. Woudenberg et al., unpublished data). PMP70 and Pex14p largely resisted Lubrol WX extraction when performed on purified peroxisomes (Fig. 4B) and total hepatocyte lysates (Supporting Fig. 3B), with a predominant flotation to fractions 4 and 5. As expected, the majority of peroxisomal caveolin-1 was also associated with these Lubrol WX–resistant microdomains. Pex13p, catalase, and Baat were detected only in fraction 12, representing the solubilized fraction in this gradient (Fig. 4B). These results demonstrate that peroxisomal caveolin-1 is localized to detergent-resistant microdomains distinct from the ones that contain the peroxisomal membrane proteins PMP70 and/or Pex14p.

thumbnail image

Figure 4. Caveolin-1 is localized to detergent-resistant microdomains in the peroxisomal membrane. Freshly isolated rat liver peroxisomes were lysed in the presence of 1% Triton X-100 or Lubrol WX followed by flotation gradient centrifugation. One-milliliter fractions (12 in total) were collected from the top and analyzed by western blotting. (A) Triton X-100 extraction; (B) Lubrol WX extraction. Equal volumes from the gradient fractions were analyzed using specific antibodies against caveolin-1; the peroxisomal membrane proteins PMP70, Pex14p, and Pex13p; and the peroxisomal matrix proteins catalase and Baat. Equal volumes of the unfractionated total protein lysate (T) as well as the pellet fraction after flotation gradient centrifugation (P) were also analyzed.

Download figure to PowerPoint

RNA Interference–Mediated Down-Regulation of Caveolin-1 Does Not Affect the Peroxisomal Location of Baat in Primary Rat Hepatocytes.

To study whether caveolin-1 may play a direct role in the sorting of peroxisomal enzymes, we inhibited its expression by RNA interference (RNAi). Transient transfection of primary rat hepatocytes with caveolin-1–specific small interfering RNA (siRNA) duplexes (siRNA-Cav) led to a 70%-80% reduction of caveolin-1 messenger RNA (mRNA) and protein levels (Fig. 5A,C). The mRNA and/or protein expression of the peroxisomal markers Pex11p, PMP70 and Baat were comparable in siRNA-Cav-1–treated and control cells (Fig. 5B,C). Immunofluorescence microscopy analysis showed that the peroxisome-specific staining of caveolin-1 was strongly reduced in siRNA-Cav–treated hepatocytes. Under these conditions, the peroxisomal matrix enzyme Baat was still predominantly present in dotted structures in the cytoplasm, indistinguishable from the staining pattern in control cells (Fig. 5D-O).

thumbnail image

Figure 5. RNA interference–mediated down-regulation of caveolin-1 does not affect the peroxisomal location of Baat in primary rat hepatocytes. Primary rat hepatocytes were transiently transfected to silence caveolin-1 expression (siRNA-Cav). Control cells were transfected with oligonucleotides directed against luciferase (siRNA-luc). After 24 hours, mRNA levels of (A) caveolin-1 and (B) Pex11p, PMP70, catalase, and Baat were analyzed by quantitative PCR. Levels of selected proteins were analyzed by western blotting, using antibodies against caveolin-1, PMP70, catalase, and Baat. (C) As a loading control, Gapdh expression was analyzed. The subcellular location of caveolin-1 (siRNA-luc: D, G; siRNA-Cav: J, M) and the peroxisomal matrix marker Baat (siRNA-luc: E, H; siRNA-Cav: K, N) were analyzed by immunofluorescence microscopy. The merged images are shown in (F,I) for siRNA-luc and (L,O) for siRNA-Cav, respectively, showing clear colocalization of caveolin-1 and Baat in luciferase-treated cells (arrows). The subcellular location of Baat is not affected by the down-regulation of caveolin-1 (O, arrows).

Download figure to PowerPoint

Catalase Sorting, Peroxisome Number, and Shape in Hepatocytes of Caveolin-1 Knockout Mice.

To obtain conclusive evidence for a putative role of caveolin-1 in peroxisome biogenesis, we analyzed the subcellular location of catalase in liver sections of caveolin-1 knockout mice compared to wild-type controls. The absence of caveolin-1 was confirmed by western blot analysis of total liver protein lysates (Fig. 6A). Catalase staining was detected throughout the liver parenchyma, with a specific dotted pattern in wild-type hepatocytes (Fig. 6B) and caveolin-1–deficient hepatocytes (Fig. 6C). No significant amount of cytosolic catalase was detected in caveolin-1–deficient hepatocytes, nor was the number or shape of peroxisomes altered compared to wild-type hepatocytes.

thumbnail image

Figure 6. Catalase sorting and peroxisome number and shape in hepatocytes of caveolin-1 knockout mice. Caveolin-1 expression in the livers of wild-type and caveolin-1 knockout mice was analyzed by western blotting. In caveolin-1 knockout mice, caveolin-1 protein expression is undetectable. (A) As a loading control, Gapdh expression was analyzed. The subcellular location of catalase in the livers of (B) wild-type and (C) caveolin-1 knockout mice was analyzed by immunofluorescence microscopy. Caveolin-1 deficiency does not affect the sorting of catalase to punctuate dots, representing peroxisomes.

Download figure to PowerPoint

Caveolin-1 Is Negatively Regulated by the PPARα Agonist Fenofibric Acid.

Finally, we determined whether expression of caveolin-1 is coregulated with genes encoding peroxisomal proteins. Primary rat hepatocytes were treated for 24 hours with the ligand for PPARα, fenofibric acid. Typical PPARα target genes, like Pex11, PMP70 and acyl coenzyme A oxidase (AOX) were strongly induced by this compound (Fig. 7A). In contrast, expression of Baat was not responsive to fenofibric acid treatment. Remarkably, fenofibric acid treatment strongly reduced the mRNA levels of caveolin-1 to 20%-30% compared to untreated cells (Fig. 7B). These data show that expression of caveolin-1 is regulated by the PPARα ligand fenofibric acid, but that its expression is reduced under conditions where peroxisomes proliferate.

thumbnail image

Figure 7. Caveolin-1 is negatively regulated by the PPARα agonist fenofibric acid. Primary rat hepatocytes were incubated with 50 μM fenofibric acid (FF) or 0.1% dimethyl sulfoxide (solvent for FF) for 24 hours. Pex11p, PMP70, AOX, Baat, catalase [all in (A)] and caveolin-1 (B) mRNA levels were analyzed by quantitative PCR.

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References
  8. Supporting Information

In this study, we show that caveolin-1 in rat hepatocytes is present in peroxisomes. Peroxisomal caveolin-1 is highly resistant to Triton X-100 extraction, more so than PMP70 and Pex14p that were previously shown to reside in peroxisomal lipid rafts (J. Woudenberg et al., unpublished data). Hepatic caveolin-1 is negatively regulated by the PPARα agonist fenofibric acid, which induces peroxisome proliferation. In line with this, peroxisome biogenesis is still intact in the absence of caveolin-1.

Expression of caveolin-1 in the liver is relatively low,7, 8 where it is particularly enriched in endothelial cells and stellate cells.9–11 Still, significant amounts of caveolin-1 are detected in hepatocytes.8, 12, 26 Previous studies have predominantly focused on the role of caveolin-1 in internalization of caveolae from the plasma membrane. However, it has also been noted that caveolin-1 may reside at other subcellular locations, including the endoplasmic reticulum, Golgi apparatus, endosomes, lipid droplets,27 and mitochondria.3, 8, 14 Our observation that significant amounts of caveolin-1 appear to reside in the peroxisomal membrane of hepatocytes therefore came as a surprise. A possible peroxisomal location of caveolin-1 in hepatocytes may, however, have been overlooked in previous studies. Caveolin-1 has been shown to cofractionate with catalase after Nycodenz gradient separation of mouse liver organelles,28 clearly separated from plasma membranes and the Golgi apparatus. Other studies excluded the analysis of peroxisomal marker proteins,29 and caveolin-1 may have been mistakenly assigned to plasma membranes or the endoplasmic reticulum, because peroxisomes also may cofractionate with these subcellular fractions.30 In addition, immunofluorescence microscopy studies revealed that caveolin-1 colocalized with PMP70 in peroxisomal structures in the rat hepatoma cell line mcA-RH7777.31 Our study, using both cell fractionation and microscopic techniques, shows that significant amounts of caveolin-1 colocalize with the peroxisomal markers catalase, Baat, ALDP, PMP70, and Pex14p in rat hepatocytes. Also, EGFP-tagged caveolin-1 expressed in rat hepatocytes was found to accumulate in cellular structures that contain the peroxisomal marker catalase. Similar to endogenous caveolin-1, EGFP-caveolin-1 also sorted to other subcellular organelles and the plasma membrane, which was especially evident in cells expressing high levels of the hybrid protein (Supporting Fig. 2). This may suggest that peroxisomes are the primary subcellular destination of caveolin-1 when the cellular levels of this protein are low. Alternatively, a cell type–specific sorting pathway may exist in hepatocytes that drives the accumulation of caveolin-1 in peroxisomes. A similar hepatocyte-specific sorting pathway has been detected for the enzyme BAAT.21

In the plasma membrane, caveolin-1 is an integral membrane protein with its hydrophilic N-terminus and C-terminus exposed to the cytosol.32 We found that also in hepatocyte peroxisomes, caveolin-1 is an integral peroxisomal membrane protein with its hydrophilic N-terminus (and most probably also C-terminus) facing the cytosol (Supporting Fig. 4). In plasma membranes, caveolin-1 interacts with cholesterol to form caveolae, which are morphologically characterized by their flask-shaped appearance. Such structures have not yet been described at the peroxisomal surface. The level of caveolin-1 in the peroxisomal membrane may be too low to form caveolae, and its function may be unrelated to vesicle traffic to or from this organelle. On the other hand, local concentrations of caveolin-1–specific staining were often observed in immunofluorescence and occasionally in immunoelectron microscopy (Supporting Fig. 5), which may suggest local caveolae formation.

Recently, we found that ALDP, PMP70, and Pex14p are associated with peroxisomal lipid rafts and that these microdomains are essential for peroxisome biogenesis in human HepG2 cells (J. Woudenberg et al., unpublished data). PMP70 and Pex14p were detected in Triton X-100–resistant lipid rafts, whereas ALDP was detected in Lubrol WX–resistant lipid rafts. Pex13p, also an integral component of the peroxisomal membrane and interacting partner of Pex14p, appeared not to be associated with either of these biochemically-defined lipid rafts. PMP70 and Pex14p are also detected in Triton X-100–resistant lipid rafts in rat liver extracts, as was caveolin-1 (Supporting Fig. 3). All the PMPs were, however, fully solubilized when purified hepatocyte peroxisomes were extracted with Triton X-100 (Fig. 4A). This is most likely due to the very low lipid:protein ratio in purified fractions of peroxisomes compared to total cell extracts. Consequently, peroxisomal membrane (lipid)s are exposed to high detergent concentrations when purified peroxisomes are used in these extraction procedures. It is well-known that lipid raft–associated proteins eventually will be extracted when the detergent:lipid ratio is increased, also in total cell lysates.33 Nevertheless, peroxisomal caveolin-1 resisted Triton X-100 extractions even when performed on purified rat hepatocyte peroxisomes. These data suggest that three biochemically distinguishable lipid rafts exist in the peroxisomal membrane that contain either (1) caveolin-1, (2) PMP70 and/or Pex14p, or (3) ALDP (Supporting Fig. 6). Pex13p does not seem to associate with either of these peroxisomal microdomains.

The function of peroxisomal caveolin-1–containing lipid rafts is currently unknown. The absence of caveolin-1 did not affect the peroxisomal location of catalase in mouse liver. This suggests that caveolin-1 is not required for peroxisome biogenesis, whereas other types of peroxisomal lipid rafts are required (J. Woudenberg et al., unpublished data). Moreover, caveolin-1 expression was strongly reduced by treating rat hepatocytes with the PPARα agonist fenofibric acid, which induces peroxisome proliferation. This observation actually suggests a negative link between caveolin-1 and peroxisome proliferation. Although such a repressive effect of caveolin-1 on peroxisome proliferation is highly speculative, it is in line with the low levels of caveolin-1 and high numbers of peroxisomes in hepatocytes.7, 34 Previously, mouse liver caveolin-1 has been shown to be positively regulated by PPARγ.35 In contrast to PPARα, which is an important regulator of (peroxisomal) fatty acid oxidation, PPARγ stimulates the storage of fatty acids and lipids.35 Caveolae and caveolin-1 are involved in intracellular cholesterol and lipid metabolism. Hepatocyte peroxisomes are involved in the conversion of cholesterol to bile acids and in β-oxidation of VLCFAs. Peroxisomal caveolin-1 may therefore be involved in transporting substrates for these metabolic activities to or from the peroxisomes. Related to this, caveolin-1 has been suggested to play a role in lipid droplet formation. Hepatocytes accumulate triacylglycerol and cholesteryl esters in these lipid droplets during liver regeneration.15, 36 The levels of hepatic cholesterol and triglycerides increase 10-fold36 during the first 24 hours after partial hepatectomy, which coincides with an increased number of peroxisomes.37 The lipid droplets are usually detected in the vicinity of peroxisomes.15 The number of hepatic lipid droplets is strongly decreased in caveolin1−/− mice and liver regeneration is strongly compromised after partial hepatectomy, resulting in increased mortality.

In conclusion, caveolin-1 resides in the peroxisomal membrane of hepatocytes. It does not seem to play a crucial role in peroxisome biogenesis. Future research needs to establish whether peroxisomal caveolin-1 is required for efficient liver regeneration and/or hepatocyte proliferation.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References
  8. Supporting Information

The authors thank E. H. Blaauw, F. Dijk, and Dr. J. J. L. van der Want for their excellent technical assistance and critical view on the electron microscopy experiments and Dr. L. Conde de la Rosa (University of Barcelona, Spain) for preparing the paraffin-embedded liver sections from caveolin-1 knockout mice.

References

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References
  8. Supporting Information

Supporting Information

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References
  8. Supporting Information

Additional Supporting Information may be found in the online version of this article.

FilenameFormatSizeDescription
HEP_23460_sm_SuppFig1.tif684KSupplemental Figure 1
HEP_23460_sm_SuppFig2.tif1955KSupplemental Figure 2
HEP_23460_sm_SuppFig3.tif3706KSupplemental Figure 3
HEP_23460_sm_SuppFig4.tif1048KSupplemental Figure 4
HEP_23460_sm_SuppFig5.tif3874KSupplemental Figure 5
HEP_23460_sm_SuppFig6.tif806KSupplemental Figure 6
HEP_23460_sm_SuppDoc.doc45KSupplemental Data

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.