Deletion of Cavin/PTRF causes global loss of caveolae, dyslipidemia, and glucose intolerance

Abstract

Caveolae are specialized invaginations of the plasma membrane found in numerous cell types. They have been implicated as playing a role in a variety of physiological processes and are typically characterized by their association with the caveolin family of proteins. We show here by means of targeted gene disruption in mice that a distinct caveolae-associated protein, Cavin/PTRF, is an essential component of caveolae. Animals lacking Cavin have no morphologically detectable caveolae in any cell type examined and have markedly diminished protein expression of all three caveolin isoforms while retaining normal or above normal caveolin mRNA expression. Cavin-knockout mice are viable and of normal weight but have higher circulating triglyceride levels, significantly reduced adipose tissue mass, glucose intolerance, and hyperinsulinemia--characteristics that constitute a lipodystrophic phenotype. Our results underscore the multiorgan role of caveolae in metabolic regulation and the obligate presence of Cavin for caveolae formation.

Figure 1

Generation of Cavin knockout mice. The Cavin knockout mice were generated using a previously described strategy involving lacZ insertion (Yang et al., 2006). (A) Schematic representation of the structure of the targeting vector and restriction enzyme sites of the Cavin gene locus before and after homologous recombination. Exon 1 in Cavin gene was replaced by the gene encoding prokaryotic β-galactosidase, designated here as LacZ. (B) Genotype analysis by Southern blotting of offspring from Cavin heterozygote (Cavin^+/−) mice intercrosses to generate homozygote (Cavin^−/−) mice is described in Experimental Procedures. Probing of EcoRI and NdeI digested genomic DNA revealed 8.4 Kb and 5.3 Kb fragments for wild-type and knockout genes, respectively. (C) PCR analysis was also devised as a 3-primer PCR-based screening strategy. The common forward primer (p1) was derived from the Cavin promoter region before the recombination site, and the wild-type specific reverse primer was derived from Cavin exon 1 (p2). The knockout-specific reverse primer was derived from the LacZ cassette (p3). Primer sequences are provided in Supplemental Experimental Procedures. (D) Total protein from lung and adipocytes of wild type (Cavin^+/+), heterozygote (Cavin^+/−) and knockout (Cavin^−/−) mice was examined by Western blotting with anti-Cavin, anti-Cav1 and anti-tubulin antibodies. See Supplemental Experimental Procedures for details.

Figure 2

The absence of Cavin leads to global loss of caveolae, down-regulation of protein and up-regulation of mRNA expression levels for caveolin isoforms and caveolae resident proteins. Electron micrographs of lung (A), smooth muscle (B) and skeletal muscle (C) tissues in wild type (Cavin^+/+) and Cavin knockout mice (Cavin^−/−). Analysis was performed on 8 weeks old male Cavin^+/+ or Cavin^−/− mice of identical strain. The caveolae from wild type mice showed the normal small invaginations (arrows). The caveolae structures have disappeared from lung, skeletal muscle and smooth muscle. (D) Whole cell lysates from wild type (Cavin^+/+) and Cavin knockout mice (Cavin^−/−) (10 weeks old strain-matched males) from the indicated tissues were prepared in RIPA buffer. Protein (10–50 µg) was subjected to SDS-PAGE, and immunoblotted with the indicated antibodies. Abbreviations are: Ad, Lu, Li, Br, He, Mu, St, SB, Ki, Sp, and Te, which represent adipocytes, lung, liver, brain, heart, muscle (mixed hind limb skeletal), stomach, small bowel, kidney, spleen and testis, respectively. (E) Quantitative RT-PCR (Supplemental Experimental Procedures) was performed in adipocytes, lung and muscle tissue by amplification with primers specific for Cav1, Cav2 and Cav3. Data were normalized to 18S rRNA and expressed as fold change relative to the mRNA levels in wild type samples. The statistical significance of differences between wild type and Cavin knockout mice (asterisks) was determined with Student's t test, *P < 0.01. Each result represents the average value and SE of at least three independent experiments.

Figure 3

Analysis of metabolic parameters in cavin null and wild type mice. (A) Body weight, body composition, plasma insulin, triglyceride, free fatty acid, adiponectin, leptin levels were measured in Cavin^+/+ and Cavin^−/− mice as described in Experimental Procedures. Mice were individually housed in metabolic chambers under fed and fasted conditions and oxygen consumption and carbon dioxide production were measured for determination of (B) respiratory exchange ratio (RER), (C) basal metabolic rate and (D) infrared beam breaks were used to quantify ambulatory activity. (E) Glucose tolerance was assessed in Cavin^+/+ and Cavin^−/− mice and values are expressed relative to basal concentrations. Results are presented as means ± standard error. The statistical significance of differences was determined by Student's t-test. The effects of group and time were determined by two-way analysis of variance. When group effects were significant (p < 0.05), Tukey-Kramer sub-analyses were performed to identify statistically significant differences between Cavin^+/+ and Cavin^−/− mice. (4-male and 4-female per group, *P < 0.05, **P < 0.01, ***P < 0.001).

Figure 4

GLUT4 expression is reduced and Insulin signaling is impaired in Cavin null mice. After insulin injection (i.p. 0.75 U/kg), animals were sacrificed and whole cell lysates of wild type (Cavin^+/+) and Cavin knockout mice (Cavin^−/−) (16 weeks old strain-matched males) from adipocyte, muscle and liver tissues were prepared in RIPA buffer. Representative western blots and quantitative bar graphs are shown for the detection of (A) GLUT4, (B) IR-beta protein levels and (C) total Akt and Akt S473 phosphorylation levels. The statistical values are displayed as means and SE of 3 independent experiments. (*P < 0.05, **P < 0.01, ***P < 0.001).