Caveolin-1 Provides Palliation for Adverse Hepatic Reactions in Hypercholesterolemic Rabbits

Caveolins are an essential component of cholesterol-rich invaginations of the plasma membrane known as caveolae. These flask-shaped, invaginated structures participate in a number of important cellular processes, including vesicular transport, cholesterol homeostasis, and signal transduction. We investigated the effects of CAV-1 on mitochondrial biogenesis and antioxidant enzymes in hypercholesterolemia-affected target organs. A total of eighteen male New Zealand white rabbits were divided into three groups: a normal-diet group, an untreated hypercholesterolemia-induced group, and a hypercholesterolemia-induced group that received intravenous administration of antennapedia-CAV-1 (AP-CAV-1) peptide every 2 days for 2 weeks. Serum biochemistry, CAV-1 distribution, neutral lipid distribution, mitochondrial morphology, biogenesis-mediated protein content, oxidative stress balance, antioxidant enzyme levels, and apoptotic cell death of liver tissue were analysed. Hepatic and circulating cholesterol and low-density lipoprotein cholesterol (LDL-C) levels differed significantly between the three groups (P<0.05). Immunohistochemical staining intensity of CAV-1 was greater in AP-CAV-1-treated rabbits than in untreated rabbits, especially in the vicinity of the liver vasculature. The high levels of neutral lipids, malondialdehyde, peroxisome proliferator-activated receptor-γ coactive 1α (PGC-1α), and nuclear respiratory factor-1 (NRF-1) seen in untreated hypercholesteremic animals were attenuated by administration of AP-CAV-1 (P<0.05). In addition, mitochondria in animals that received treatment exhibited darker electron-dense matrix and integrated cristae. Furthermore, the levels of ROS modulator 1 (Romo1) and superoxide dismutase (SOD)-2, as well as catalase activity were significantly lower in CAV-1-treated hypercholesterolemic rabbits (P<0.05). AP-CAV-1 treatment also restored mitochondrial respiratory chain subunit protein content (OXPHOS complexes I–V), thereby preserving mitochondrial function (P<0.05). Furthermore, AP-CAV-1 treatment significantly suppressed apoptotic cell death, as evidenced by a reduction in the number of TUNEL-positive cells. Our results indirectly indicate that CAV-1 mediates the negative effects of PGC-1α on hepatic mitochondrial respiratory chain function, promotes the antioxidant enzyme defence system, and maintains mitochondrial biogenesis.


Introduction
Caveolae are omega-shaped invaginations of the plasma membrane with a diameter of approximately 50-100 nm. These flask-shaped, invaginated structures participate in a number of important cellular processes including vesicular transport, cholesterol homeostasis, signal transduction and tumour suppression [1,2]. They also play a prominent role in various human pathobiological conditions, particularly in the cardiovascular system [3][4][5]. The key structural and functional protein in caveolae in most cell types is caveolin (CAV). Deficiency in CAV-1, an important protein marker of caveolae, has been shown to result in cholesterol ester uptake associated with increased acyl-CoA:cholesterol acyl-transferase activity and scavenger receptor class B type I (SR-BI) overexpression [6][7][8]. Frank et al., found that caveolin-1-deficient (CAV-1 2/2 ) mice exhibited enhanced plasma triglyceride (TG) levels, increased high-density lipoprotein (HDL) levels, and reduced hepatic very low-density lipoprotein (VLDL) secretion. They also found that caveolin-1 deficiency prevented transcytosis of LDL across endothelial cells, suggesting that CAV-1 may be implicated in the regulation of plasma LDL levels and, hence, may play an important role in the development of atherosclerosis [2]. In CAV-1-overexpressing animal models, plasma HDL-cholesterol (HDL-C) levels and hepatic free cholesterol levels were elevated, but TG and total cholesterol levels remained unchanged. In addition, CAV-1 has been shown to stimulate HDL-mediated cellular cholesterol efflux, which further prevents ox-LDL uptake into human endothelial cells [9][10][11][12]. The effect of CAV-1 expression on cholesterol efflux remains controversial. A number of studies have shown that CAV-1 deficiency results in accumulation of free cholesterol in mitochondrial membranes, which causes mitochondrial dysfunction and eventually cell death [1,[13][14][15][16]. Studies have also shown that CAV-1 negatively regulates endothelial nitric oxide synthase (eNOS), the enzyme responsible for synthesizing nitric oxide [4,[17][18][19].
More than 90% of cellular ATP generation occurs in the mitochondria via oxidative phosphorylation (OXPHOS). In addition, mitochondria are principally responsible for the generation of reactive oxygen species (ROS), calcium homeostasis, stimulation of apoptosis, and ageing [20,21]. The peroxisome proliferator-activated receptor-c coactive 1a (PGC-1a) is a major regulator of mitochondrial biogenesis and function-related gene expression [i.e. nuclear respiratory factor-1 (NRF-1) and mitochondrial transcription factor A]. Valle et al. demonstrated that PGC-1a overexpression resulted in a notable reduction in ROS accumulation, upregulation of mitochondrial number and activity and reduction in apoptotic cell death, even under oxidative stress conditions, leading to protection against vascular endothelial cell dysfunction, particularly in various cardiovascular diseases [22,23]. Studies have shown that changes in mitochondrial biogenesis are associated with diabetic retinopathy, diabetic neuropathy, chronic exercise, inflammatory processes, chronic cholestasis, and cholestatic liver diseases [24][25][26][27][28][29][30][31]. However, there is no clear therapeutic strategy for improving mitochondrial dysfunction and enhancing clinical outcome [32].
Antennapedia-CAV-1 peptide, a homeodomain of antennapedia, is a Drosophila transcription factor that can be internalized by cells via a receptor-independent pathway. It possesses translocating properties and is capable of carrying peptides comprising 20-30 amino acid residues across the plasma membrane [33][34]. Coupling of the CAV-1 C-terminal scaffolding domain (amino acids 82-101) to the AP peptide was shown to facilitate its translocation across the cell membrane in vivo. After in vivo delivery, AP-CAV-1 attenuated eNOS-dependent NO production and reduced inflammation in mice. AP-CAV-1 also inhibited smooth muscle proliferation in mice with pulmonary artery hypertension [35][36]. Our previous study revealed that CAV-1 may be important in the prevention and treatment of atherosclerosis [37]. In addition, CAV-1 is associated with the HDLmediated cholesterol efflux pathway in aortic endothelial cells [38]. Whether this effect of CAV-1 is related to the disturbance of oxidative and antioxidative homeostasis, mitochondrial biogenesis, and mitochondrial activity remains unknown. Therefore, in the present study, we investigated the effects of CAV-1 on cholesterol efflux into the circulation, neutral lipid distribution, mitochondrial function, mitochondrial morphology, mitochondrial biogenesis, apoptotic cell death, and oxidative stress balance in a rabbit model of hypercholesterolemia.

Animals and DNA Extraction
A total of eighteen male New Zealand white rabbits (weighing approximately 4 kg) were divided into three groups. Rabbits in the first group (normal-diet group, n = 6) were fed a standard rabbit chow diet (Fu-Shou Co.; Taichung, Taiwan). Animals in the second group (untreated hypercholesterolemia-induced group, n = 6) were fed a 2% cholesterol diet for 8 weeks. Rabbits in the third group (treated hypercholesterolaemia-induced group, n = 6) were fed a 2% cholesterol diet for 8 weeks and then, beginning on the sixth week, received intravenous administration of antennapedia-CAV-1 (AP-CAV-1) peptide [RQPKIWEFPNRRKPWK-K-DGIWKASFTTFVTKYWFYR-(OH); GU-Yuan Biotech Services Corp., TW; 1 mg/kg] every 2 days for 2 weeks. The rabbits in each group were sacrificed after 8 weeks. Blood and liver tissue were collected for analysis. Hepatic DNA was isolated using GenoMarker reagent according to the manufacturer's recommendations. The animal facilities and protocols were reviewed and approved by the Institutional Animal Care and Use Committee (IACUC) of the Taichung Veterans General Hospital, Taichung, Taiwan (Approval No: La-95278).

Analysis of Serum Biochemistry
Blood (3.0 ml) was collected from rabbit ears into vacutainerH vials and centrifuged at 3000 rpm for 10 min. Total serum glucose, cholesterol, TG, AST, ALT, and free fatty acid levels were measured by enzymatic colorimetric methods using an automatic analyser (ADVIA 1800; Siemens, Tarrytown, NY, USA). HDL-C and LDL-C were measured by elimination/ catalase methods using an automatic analyser (ADVIA 1800; Siemens). Insulin was detected by a chemiluminescence method using an autoanalyser (ADVA Centaur; Siemens).

Primer Design and Real-Time PCR
Sequences of rabbit mtDNA (GeneBank accession number, NC_001913) and b-actin (BA) (GeneBank accession number, NW_003159464) were used to design species-specific primers. The rabbit mtDNA D-loop (DL)-specific primers used in this study are as follows: forward, 59-gg ttc tta cct cag ggc cat ga-39and reverse, 59-gat tag tca tta gtc cat cga gat-39. The rabbit BA-specific primers used in this study are as follows: forward, 59-atc gtg cgc gac atc aag gag aag c-39 and reverse, 59-g cag ctc gta gct ctt ctc cag-39. PCR amplification was performed in 50 ml reaction volumes containing 56 PlatinumH SYBRH Green qPCR SuperMix UDG, 10 mM each of DL or BA forward and reverse primers, 16 ROX reference dye, and approximately 50 ng of sample DNA. The cycling conditions included an initial phase of 2 min at 50uC, followed by 2 min at 95uC and 40 cycles of 15 s at 95uC and 30 s at 60uC. Each sample was assayed in duplicate, and fluorescence spectra were continuously monitored using the 7700 Sequence Detection System (Applied Biosystems). Data analysis was based on measurement of the cycle threshold (C T ). The difference in C T values was used as the measure of relative abundance, i.e. C T (DL)2C T (BA) was used to measure the abundance of the mitochondrial genome.

Electron Microscopy
Liver slices (1-mm 3 thick) were prepared manually and were immediately placed in cold fixative consisting of a mixture of 4% formaldehyde and 1% glutaraldehyde in 0.2 M cacodylate buffer (pH 7.4) for 6 h, post-fixed with osmium tetraoxide, and embedded in Spurr's resin. Ultrathin sections were double stained with uranyl acetate and lead citrate and examined using a JEM-1230 electron microscope (JEOL Ltd., Japan).

Apoptosis Determination
Liver sections (4-mm thick) were deparaffinized with xylene, rehydrated in ethanol, and washed with PBS buffer. The sections were incubated with proteinase K (25 mg/ml) for 30 min at room temperature to expose DNA for end labelling. A positive control liver section was exposed to DNase I solution for 30 min at room temperature. After incubation, cell death was detected in liver sections by a terminal deoxynucleotidyl transferase dUTP nick end-labelling (TUNEL) assay kit (Invitrogen, Eugene, OR, USA) according to the manufacturer's instructions. The sections were examined by fluorescence microscopy using an Olympus BX61 microscope and TUNEL-positive cells were quantified in 7 fields per animal at 6400, N = 6 rabbits for each group per time point (Image-Pro Plus 4.5).

Neutral Lipid Staining
Fresh liver tissue was frozen in optimal cutting temperature compound (Sakura Finetek, Torrance, CA, USA) and cryostat sections (7 mm-thick) were fixed in ice-cold 10% formalin for 10 min. Fresh oil-red O solution was prepared by dilution of the oil-red O stock solution (Sigma, 3.5 g/l in 99.5% isopropyl alcohol) in distilled water to a final concentration of 0.2%. This solution was mixed thoroughly and filtered through a 0.2-mm filter before use. Next, slides were placed in absolute isopropyl alcohol for 2 min. Subsequently, the sections were stained in pre-warmed oil-red O solution for 20 min in an oven at 60uC, rinsed in distilled water and counterstained with haematoxylin, followed by washing with running tap water. The slides were mounted with Clear Mount TM Mounting solution (Invitrogen, Eugene) and examined using an Olympus BX61 microscope.

Malondialdehyde (MDA) Measurement
MDA formation, a measure of lipid peroxidation, was assayed by the reaction with thiobarbituric acid (TBA). MDA concentration was detected in liver supernatant by a NWLSS TM Malondialdehyde assay kit (Northwest, Vancouver, WA, USA).
In brief, we prepared a 10% w/v homogenate in cold assay buffer via centrifugation and stored the resultant supernatant on ice. Then, BHT reagent (5 ml), acid reagent (125 ml), and TBA reagent (125 ml) were added to 125 ml of tissue supernatant and the mixture was incubated at 60uC for 1 h. After cooling the vials at 4uC for 10 min the mixture was centrifuged at 10,0006g for 2-3 min. A 200-ml reaction mixture was then transferred to a 96-well microplate and absorbance was measured at 532 nm. The results are expressed as nanomoles of malondialdehyde per milligram of protein.

Statistical Analysis
Statistical analyses were performed using one-way ANOVA and SigmaPlot t-tests. Data are presented as mean 6 SD from 3 independent experiments. A P value,0.05 was considered to represent statistical significance.

Serum Biochemistry
The serum glucose, cholesterol, and LDL-C levels differed significantly among each group (P,0.05; Table 1). Blood glucose and cholesterol levels were significantly higher in untreated hypercholesterolemic rabbits (untreated control group) and in caveolin-1-treated rabbits (CAV-1 group) than in rabbits that received a diet of normal rat chow (normal-diet group). LDL-C levels were also significantly higher in the hypercholesterolemic rabbits than in rabbits that were fed a normal diet (P,0.05). In addition, glucose, cholesterol, TG, AST, ALT, LDL-C, insulin, and free fatty acid levels were significantly lower in the CAV-1 group than in the untreated group.

Distribution of CAV-1 and Neutral Lipids in Hepatic Cells of Hypercholesterolemic rabbits
Analysis of liver sections revealed that the staining intensity of CAV-1 was higher in hypercholesterolemic rabbits than in the normal-diet group (14.965.9 and 3.361.48 positive cells/4006 field, P,0.05). Our previous study indicated that CAV-1 expression in the thoracic aorta of rabbits fed a high-cholesterol diet reaches its highest level at 5 weeks and then gradually decreases after 8 weeks [37]. Thus, hypercholesterolemic rabbits were administered intravenous injections of AP-CAV-1, 1 mg/kg,   Figure 1A and 1B. Hepatic cells in the control group showed a markedly higher degree of oil-red O staining and significantly higher malondialdehyde levels (6.1760.51 mM/mg protein, P,0.05), a reflection of neutral lipid accumulation, than hepatic cells in the AP-CAV-1treated group and the normal-diet group ( Figure 1C-1E).

Mitochondrial Function, Biogenesis, and Morphology
The mtDNA copy numbers, PGC-1a and NRF-1 expression levels, as well as mitochondrial morphology in rabbit liver were examined. The real-time PCR assay revealed that the mtDNA copy number/cell in the control and CAV-1 groups was significantly lower than that in the normal-diet group (P,0.05) ( Figure 2). Immunohistochemistry revealed markedly higher levels of PGC-1a and NRF-1 in the livers of the hypercholesterolemic rabbits than in livers of the normal-diet group. However, at the end of the experiment, expression levels of PGC-1 a and NRF-1 were 0.4-and 0.5-fold lower, respectively, in the AP-CAV-1treated group than in the untreated group ( Figure 3A-3C). In addition, electron microscopy revealed a more condensed cytoplasm, with a dark electron-dense matrix and integrated cristae after CAV-1 treatment ( Figure 3D).

Romo1-derived ROS and Antioxidant Enzymes Expression
We then examined whether AP-CAV-1 treatment has an attenuation effect on the increased levels of mitochondrial-derived ROS seen in hypercholesterolemic rabbits. We found that the level of Romo-1, a novel source of ROS in mitochondria, was 1.1-fold lower in the CAV-1 group than in untreated hypercholesterolemic animals. In addition, AP-CAV-1 treatment resulted in a 1.1-fold reduction in SOD2 and a 0.7-fold reduction in catalase activity (P,0.05) ( Figure 4A-4D).

Apoptotic Cell Death
We measured apoptotic cell death in hepatic cells using the TUNEL assay to investigate whether CAV-1 activity is related to cell survival. We found that the number of TUNEL-positive cells per high power field (4006) was significantly higher in the control group (4.3361.14) than in the normal-diet group (2.7961.26) or in the AP-CAV-1-treated group (3.0760.98) (P,0.01) as shown in Figure 6A, indicating that treatment with CAV-1 significantly suppressed apoptosis. Hoechst staining showed the appearance of the nucleus. Quantification of the results is shown in Figure 6B.

Discussion
In our novel study, we found that CAV-1-treated hypercholesterolemic rabbits had markedly less fat acumination in the liver than untreated animals. In addition, we also found that CAV-1 affects the molecular mechanisms governing mitochondrial biogenesis and the OXPHOS chain, as well as oxidative and anti-oxidative homeostasis, fatty acid oxidation, and apoptosis of hepatic cells.
Hypercholesterolemia is an important inducer of atherosclerosis and exclusion of excess cholesterol from cells and tissues can prevent the development of the disease. Cholesterol efflux involves a number of transporters including CAV-1, ATP-binding cassette (ABC) A1, ABCG1, SR-B1, and CTP 27A1 [38]. In 2001, Frank et al. demonstrated that CAV-1 downregulation enhanced cellular cholesterol efflux to HDL [10]. Similarly, our previous study provided evidence that CAV-1 levels in plaques located in coronary arteries and aortic arches of rabbits fed a high-cholesterol diet decreased after 8 weeks [37]. We also found that CAV-1   [38]. In this study, we revealed that AP-CAV-1 treatment helps restore LDL-C to normal levels and may provide palliation for adverse hepatic reactions by decreasing AST and ALT levels in hypercholesterolemic rabbits.
Although the effects of CAV-1 on intracellular cholesterol metabolism and cholesterol efflux have been studied previously, the results were inconsistent. Earlier studies revealed that significant increases in CAV-1-mediated transport of newly synthesized cholesterol to plasma membrane caveolae resulted in enhanced cholesterol efflux in L1210-JF cells [40], human fibroblasts [41], and THP-1 cells [42]. In contrast, other studies showed that changes in CAV-1 levels did not result in changes in cholesterol efflux in J774, RAW [43], HEK-293T, FRT [6,44], and RAW 264.7 cells [12]. The reason for the observed differences is that the role of CAV-1 in cholesterol metabolism and efflux are cell-type specific. Studies have shown that the liver is a major source of HDL-C in plasma and that it may regulate reverse cholesterol transport by being both the origin and destination of the pathway [45]. CAVs are cholesterol-binding proteins that participate in many important cellular processes, including cholesterol transport. However, only a few studies have explored the role of CAVs in the liver. Moreno et al., for example, found that hepatic CAV overexpression increases bile flow, bile saltdependent biliary lipid secretion, plasma HDL-C levels, and hepatic free cholesterol levels in mice [11]. Fu et al., also demonstrated that CAV-1 expression enhances cholesterol efflux in hepatic cells [12]. However, the molecular mechanisms governing these effects remain unclear.
Mitochondrial dysfunction plays a central role in the development of cholestatic liver diseases because mitochondria regulate cell death signalling -triggered by inflammatory cytokines or bile acid secretion -and contribute to oxidative damage and metabolic disorder [31]. Bosch et al., have shown that CAV-1 participates in the regulation of mitochondrial cholesterol levels and that there is a connection between CAV, cholesterol metabolism, and mitochondrial diseases [13,14]. In this study, we provide evidence that CAV-1 overexpression protects against fatty acid oxidation, increases mitochondrial OXPHOS chain function, suppresses the expression of Romo1 without increasing SOD2 and catalase activities, attenuates PGC-1a and NRF-1 expression, and protects against apoptosis.
There are some limitations in our study. First, the hypercholesterolemic rabbit model used in this study might not accurately reflect the relevant pathways in humans. Further studies should be conducted using other model species. Second, we did not perform time-dependent studies to evaluate whether mitochondria are the cause of the observed effects.
In summary, our results indirectly indicate that CAV-1 mediates the effects of PGC-1a on hepatic mitochondrial OXPHOS chain function, the antioxidant defence system, and mitochondrial biogenesis. Thus, CAV-1 may provide palliation for adverse hepatic reactions in hypercholesterolemic rabbits by attenuating the PGC-1a/NRF-1 pathway.  , which were upregulated when the rabbits were fed a high-cholesterol diet for 8 weeks, were higher in the control group than in the normal and CAV-1 groups. TUNEL staining has green fluorescence (Alexa Fluror 488nm), and the nuclei are stained with Hoechst dye (blue). (B) Quantification of TUNEL IHF stain in rabbit liver tissue per 6400 field/7 fields per animal, n = 6 rabbits for each group. A significantly suppressed number of TUNEL-positive cells were observed in the CAV-1 group. * and { are significantly different from the normal and control groups, respectively, P,.01. Values are mean 6 SD from at least 3 independent experiments, n = 6 for each group. doi:10.1371/journal.pone.0071862.g006