Prohibitin Overexpression in Adipocytes Induces Mitochondrial Biogenesis, Leads to Obesity Development, and Affects Glucose Homeostasis in a Sex-Specific Manner

In addition to the classic role in energy production, mitochondria have tissue-specific functions, especially in white adipose tissue (WAT), which relies on mitochondria for its specialized function (1). Mitochondrial biogenesis is an important process in adipocyte differentiation (2,3). However, the role of adipocyte mitochondria in adipose tissue homeostasis and metabolic regulation is not entirely understood. Obesity is characterized by an increase in adipose tissue mass, and this requires corresponding changes in adipocyte biology, including mitochondrial biogenesis and function, to synthesize and store excess fat. However, obese animal models based on primary alterations in adipocyte mitochondria are not available.

We have reported that prohibitin (PHB [also known as PHB1]) is important in adipocyte differentiation (4). PHB is an evolutionarily conserved protein that functions as a mitochondrial chaperone and plays an important role in mitochondrial biology (5). In humans, the PHB gene has been mapped to chromosome 17q12–q21 (6), and locus 17q21 harbors genes that influence the propensity to accumulate fat in the intra-abdominal area (7). Phb knockdown in Caenorhabditis elegans results in a significant reduction in intestinal fat content (8). Moreover, PHB overexpression in preadipocytes facilitates adipogenesis, whereas PHB silencing attenuates adipogenesis and mitochondrial function (4,9,10). Collectively, this evidence points to a role of PHB in adipose tissue biology; however, the precise mechanism involved remains to be determined.

We hypothesized that the role of PHB in adipogenesis is mediated through mitochondrial biogenesis. To test this hypothesis, we generated transgenic mice overexpressing PHB in adipocytes under the control of the adipocyte protein-2 (aP2) gene promoter. PHB overexpression in adipocytes increased mitochondrial biogenesis and caused obesity in mice. We named these mice Mito-Ob because they develop obesity as a result of adipocyte mitochondrial remodeling. This article reports on the metabolic phenotype of Mito-Ob mice.

The aP2 promoter-containing vector [pBS-aP2 promoter (5.4 kb) polyA] was obtained from Addgene (Cambridge, MA). A full-length PHB clone was obtained from OriGene Technologies (Rockville, MD). PHB clone was digested with Not1 to release the full-length cDNA of PHB, and then PHB cDNA was subcloned into the Not1 site of the pBS-aP2 promoter vector.

The aP2 promoter containing PHB clone was digested with Sac1 and Kpn1 restriction enzymes to release the fragment containing aP2 promoter with full-length PHB. This DNA was purified and used for pronuclear transfer into zygotes from CD1 mice. Founder animals were identified by PCR amplification of tail genomic DNA using forward primer 5′-GCAGCCCGGGGGATCCACTA-3′ and reverse primer 5′-GCACACGCTCATCAAAGTCCTCTCCGATGCTG-3′. Founder animals were mated with CD1 female mice to establish PHB transgenic mouse lines as per standard procedures. Experiments involving animals were performed as per the study protocols approved by the Institutional Animal Care and Use Committee.

The animals were given normal chow (LabDiet, St. Louis, MO). Body weight of Mito-Ob and wild-type mice was recorded weekly after weaning, and food intake during 3–6 months of age was determined (11).

Horizontal activity levels were measured using a metabolic cage system (AccuScan, Columbus, OH), and data were collected every 5 min for 24 h (12).

Adipose tissues from 6-month-old Mito-Ob and wild-type littermates were fixed in buffered formaldehyde and subsequently dehydrated and embedded in paraffin; 5-μm sections were stained with hematoxylin-eosin (13).

Adipose tissue lysates from Mito-Ob and wild-type mice containing equal amounts of proteins (∼15 μg/lane) were separated by SDS-PAGE and subsequently analyzed by immunoblotting (4,13). MitoBiogenesis Western blotting cocktail (Abcam Inc.) was used to determine succinate dehydrogenase-A (SDH-A) and cytochrome c oxidase-I (COX-I) protein levels in adipose tissue.

Adipose tissues were excised into small pieces (<1 mm³) and fixed with 2.5% glutaraldehyde in 0.1 mol/L PBS buffer (pH 7.4) for 3 h. Each specimen was postfixed in 1% osmium tetroxide for 1 h before embedding in epon resin. Transmission electron microscopy was performed with a Philips CM10 at 80 kV on ultra-thin sections (100 nm) and stained with uranyl acetate and counterstained with lead citrate.

Mitochondrial DNA (mtDNA) copy number in adipose tissue was determined by real-time PCR (14).

Serum adipokines and hormones were measured with a Bio-Plex Pro Mouse Diabetes Assay and Bio-Plex 200 3D Multiplex Suspension Array System (Bio-Rad, Hercules, CA) per the manufacturer’s protocols.

Glucose and insulin tolerance tests in 6-month-old mice were performed as previously described (13).

Serum cholesterol (Molecular Probes, Eugene, OR), triglyceride, and free fatty acid (FFA) levels were measured by using kits from BioAssay Systems (Hayward, CA). FFA uptake and oxidation were measured by using kits from Abcam Inc.

All statistical analyses were performed with GraphPad Prism 6 software. Experimental results are shown as mean ± SEM. Two-tailed Student unpaired t tests were performed to compare sex-matched Mito-Ob and wild-type littermates unless otherwise indicated. P < 0.05 was considered significant.

Both male and female Mito-Ob mice initially weighed similarly to their wild-type littermates; however, they started to gain weight beginning at ∼1 month of age and became significantly obese (P < 0.001) by 2–3 months of age (Fig. 1A and B) without any founder effect (Fig. 1A). No difference in food intake was found between Mito-Ob and wild-type mice (Fig. 1B), but a reduction in horizontal activity levels was observed in Mito-Ob mice (Fig. 1B). These results suggest that PHB overexpression in mouse adipocytes leads to obesity development independent of food intake.

Female Mito-Ob mice accumulated more visceral fat, but less brown adipose tissue (BAT) than male Mito-Ob mice, whereas subcutaneous fat mass increased similarly in Mito-Ob female and male mice (Fig. 1C). Collectively, these data suggest that PHB overexpression in mouse adipocytes leads to increased adipose tissue mass, with sex-related differential effects on brown and visceral fat depots.

Consistent with an increase in the adipose tissue mass in Mito-Ob mice, adipocyte hypertrophy was apparent in the WAT from Mito-Ob mice (Fig. 1D). In addition, WAT from male Mito-Ob mice showed increased distribution of crown-like structures, a sign of macrophage infiltration (Fig. 1D). Furthermore, male Mito-Ob mice had increased fat accumulation in BAT (Fig. 1D). These data suggest that PHB overexpression in adipocytes leads to adipocyte hypertrophy, implying increased triglyceride synthesis and/or storage in WAT of Mito-Ob mice.

Transmission electron microscopy showed an increase in mitochondrial number and distribution throughout the cytoplasmic rim of white adipocytes in Mito-Ob mice versus wild-type littermates (Fig. 2A). The mitochondria in these adipocytes were predominantly elongated in Mito-Ob mice but round in wild-type mice (Fig. 2A). In contrast, BAT had similar mitochondrial content in Mito-Ob and wild-type mice (Fig. 2B). However, a reduction in mitochondrial size was observed in BAT from Mito-Ob male mice (Fig. 2B).

To confirm mitochondrial biogenesis in the WAT of Mito-Ob mice, we determined mtDNA copy numbers and mitochondrial protein (SDH-A, COX-I, PGC-1α, DNA poly-γ A, mitochondrial transcription factor A [Tfam], etc.) levels. Both were significantly upregulated (P < 0.05–0.001) in Mito-Ob mice (Fig. 2C–E). Collectively, these results suggest that PHB overexpression in adipocytes induces mitochondrial biogenesis.

Only Mito-Ob males were found to have significantly impaired glucose (P < 0.05–0.001) and insulin (P < 0.05–0.01) tolerance (Fig. 3A). Mito-Ob female mice had insulin sensitivity similar to wild-type animals (Fig. 3A). A sign of hepatic steatosis was also found in the liver of Mito-Ob mice (Fig. 3B). The data suggest that obesity in Mito-Ob mice have sex-specific metabolic impairment. Among lipid parameters, serum triglyceride, glycerol, and cholesterol levels were significantly decreased (P < 0.001) in Mito-Ob female mice, whereas FFA levels were significantly increased (P < 0.01) in Mito-Ob male mice (Table 1).

Serum insulin levels were significantly higher (P < 0.01) in Mito-Ob male mice versus wild-type littermates (Table 1). However, Mito-Ob female mice had insulin levels similar to wild-type littermates (Table 1). Among adipokines, adiponectin levels were increased (P < 0.05) only in Mito-Ob female mice, whereas leptin levels were higher in Mito-Ob male mice (Table 1). Resistin levels were unchanged in Mito-Ob male mice (Table 1).

We next analyzed the expression levels of adipose triglyceride lipase, hormone-sensitive lipase, and lipoprotein lipase in WAT. Adipose triglyceride lipase and hormone-sensitive lipase were upregulated in both male and female Mito-Ob mice versus wild-type mice, but their levels were higher in Mito-Ob male mice than in Mio-Ob female mice (Fig. 3C). Lipoprotein lipase protein levels were decreased in Mito-Ob male mice but remained unchanged in Mito-Ob female mice (Fig. 3C). FFA uptake was significantly decreased in Mito-Ob male mice (Fig. 3C) without any significant change in fatty acid oxidation (data not shown).

The obese phenotype of Mito-Ob mice affirms the emerging notion that PHB and adipocyte mitochondria play a role in the regulation of adipose tissue homeostasis and metabolic regulation (1–4,8–10). However, the metabolic consequences of obesity in Mito-Ob mice are sex specific, suggesting that intrinsic differences exist in the way adipose tissues are regulated and/or respond to obesity development in male and female mice despite a similar underlying cause. The aP2 promoter used in this study is primarily expressed in adipocytes and has been used in similar work (15). In addition, it is expressed in immune cells, such as macrophages (16). However, a similar effect of PHB manipulation in adipocytes in vitro and in vivo (4,8–10) suggests that the phenotype we have observed is most likely due to the role of PHB in adipocytes.

PHB translocates from mitochondria to the nucleus in response to estrogen (17). In addition, PHB has been associated with the function of Tfam and nuclear factor-like 2 (Nrf-2) (18,19). Tfam plays an important role in mitochondrial biology, and Nrf-2 regulates transcription of nuclear-encoded mitochondrial proteins (18,20). This evidence along with the upregulation of mitochondrial biogenesis found in the WAT of Mito-Ob mice suggest a potential role of PHB in the mitonuclear cross talk required for mitochondrial biogenesis. Furthermore, the upregulation mitochondrial biogenesis markers in the WAT of Mito-Ob mice suggests that PHB indeed plays a role in this process. An enhanced mitonuclear cross talk may be the underlying mechanism behind increased mitochondrial biogenesis in the WAT of Mito-Ob mice.

Men and women have distinct distributions of body fat, where subcutaneous fat tends to be predominant in females and visceral fat is predominant in males (21). It is the visceral fat that is related to obesity and its complications (21). However, the paradox of normal insulin sensitivity along with increased visceral obesity in the Mito-Ob female mice suggests that it is not the visceral obesity per se but the functional status of the adipose tissue, such as the dynamics of lipid handling and adipokine secretion, that leads to obesity-associated disorders. For example, a differential regulation of adiponectin in Mito-Ob male and female mice could be related to the different metabolic status because adiponectin promotes adipocyte differentiation, insulin sensitivity, and lipid accumulation (22), and adiponectin secretion has been reported to correlate with mitochondrial function (23). In this context, it should be noted that estrogens, which have protective effects against obesity, also play a role in mitochondrial biogenesis (21,24,25). However, it is not known whether the role of estrogen in mitochondrial biology is linked to its protective effect against obesity. The Mito-Ob female mice provide a unique opportunity to dissect the relationship among estrogen, PHB, and mitochondria in adipose tissue biology and metabolic regulation.

In summary, the obese phenotype and sex-specific metabolic dysregulation in Mito-Ob mice establish for the first time to our knowledge an important role of PHB in adipose tissue biology in mammals. It is expected that Mito-Ob mice will prove a valuable tool for obesity research. We anticipate that Mito-Ob mice will help to better define the sex differences in obesity and associated health problems and the role of adipose tissue mitochondria.

Acknowledgments. The authors thank Tooru M. Mizuno and Pei San Lew, Department of Physiology, University of Manitoba, for help with horizontal activity measurement.

Funding. The authors were supported by the Natural Sciences and Engineering Research Council of Canada, Manitoba Health Research Council (MHRC), and Canada Foundation for Innovation. G.P.P.-M. is a recipient of an MHRC Postdoctoral Fellowship Award.

Duality of Interest. No potential conflicts of interest relevant to this article were reported.

Authors Contributions. S.R.A. generated the aP2-PHB construct, performed genotyping, measured hormones and adipokines levels in serum, and contributed to all experiments involving animals. K.H.N. performed real-time PCR, measured lipid parameters, and contributed to all experiments involving animals. G.P.P.-M. analyzed protein samples. W.W. performed histological analyses. B.L.G.N. contributed to the data interpretation and writing of the manuscript. S.M. contributed to the experimental design, data analysis and interpretation, and writing of the manuscript. S.M. is the guarantor of this work and, as such, had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data.

Prior Presentation. This study was presented in abstract form at the 3rd Scientific Sessions of the Society for Mitochondrial Research and Medicine, Bengaluru, India, 19–20 December 2013.