Uncoupling Caveolae From Intracellular Signaling In Vivo

Abstract

Rationale: Caveolin-1 (Cav-1) negatively regulates endothelial nitric oxide (NO) synthase-derived NO production, and this has been mapped to several residues on Cav-1, including F92. Herein, we reasoned that endothelial expression of an F92ACav-1 transgene would let us decipher the mechanisms and relationships between caveolae structure and intracellular signaling.

Objective: This study was designed to separate caveolae formation from its downstream signaling effects.

Methods and results: An endothelial-specific doxycycline-regulated mouse model for the expression of Cav-1-F92A was developed. Blood pressure by telemetry and nitric oxide bioavailability by electron paramagnetic resonance and phosphorylation of vasodilator-stimulated phosphoprotein were determined. Caveolae integrity in the presence of Cav-1-F92A was measured by stabilization of caveolin-2, sucrose gradient, and electron microscopy. Histological analysis of heart and lung, echocardiography, and signaling were performed.

Conclusions: This study shows that mutant Cav-1-F92A forms caveolae structures similar to WT but leads to increases in NO bioavailability in vivo, thereby demonstrating that caveolae formation and downstream signaling events occur through independent mechanisms.

Keywords: caveolin-1; cell endothelial cell; eNOS; mice; nitric oxide; vascular function.

Figure 1. Expression of the Cav-1-F92A-HA transgene

A, Whole lung protein from control and Cav-1-F92A mice was analyzed for the expression of the transgene. B, MLECs were isolated with CD31-dynabeads. The fraction bound (CD31⁺) and the non-bound fraction (CD31⁻) was separated and immunoblotted for HA. The endothelial specific protein VECAD was used as a marker for the enrichment of endothelial cells by using the CD31-beads. C, Top panel, Cross sections of the thoracic aorta. Sections were stained for the nucleus (blue), α-SMA (red) and HA (green) [L, lumen]. Bottom panel, Whole-mount staining of mesenteric artery. En-face preparations were stained for the nucleus (blue), PECAM-1 (red) and HA (green). Scale bars are 50 μm.

Figure 2. Effects of suppression of Cav-1-F92A-HA expression and nitric oxide measurements

A, Left panel, Whole lung protein from double transgenic mice was analyzed for the expression of the transgene at day 0, day 3 and day 7 after treatment with doxycycline (2 mg/ml) with 5 % sucrose in the drinking water. Each time point was repeated in three animals. Right panel, Densitometry analysis of the HA intensity and the HSP90 loading control. B, Left panels, Averaged systolic blood pressure from 8 control and 8 Cav-1-F92A animals. The first two days baseline was recorded, before drinking water was switched to doxycycline (2 mg/ml) with 5 % sucrose. Dotted lines show the changes of the systolic blood pressure (slopes of the curves are listed in Table I). Right panels, bar graph presentation of the same results. The bars present the average systolic blood pressure without and with doxycycline. C, Left panels, Averaged systolic blood pressure from 3 control and 3 Cav-1-F92A animals. The first two days baseline was recorded, before drinking water was switched to L-NAME (1 mg/ml). Dotted lines show the changes of the systolic blood pressure (slopes of the curves are listed in Table II). Right panels, bar graph presentation of the same results. The bars present the average systolic blood pressure without and with L-NAME. D, EPR-quantification of the peak to trough length of 4 mice each (representative traces are shown in Figure IV). E, Quantification of the p-VASP/t-VASP ratio (immunoblot is shown in Figure V).

Figure 3. Stability of caveolae in the presence of Cav-1-F92A mutant

A, Left panel, Whole lung protein from single and double transgenic mice was analyzed for the expression of Cav-2 and Cavin-1. Cav-1 and HA immunoblotting as a control for the expression of the transgene. HSP90 was used as loading control. Right panel, Quantification of the relative intensity of Cav-2 and Cavin-1 normalized to HSP90. B, Adenoviral reconstitution of immortalized Cav-1 KO MLECs with either GFP, Cav-1 WT or Cav-1-F92A mutant. Immortalized WT cells are loaded as control for the expected expression level of Cav-1 and Cav-2. Cav-2 immunoblotting as a readout for the stability of caveolae. C, Top panel, Immunoblot analysis of Cav-1 (WT, endogenous) and (HA, transgene) of the sucrose gradient fractions. HSP90 was used as marker for bulk fractions. Lower panel, Quantification of the immunoblot represented as % band intensity per fraction. D, TEM images for Cav-1 KO MEFs adenoviral reconstituted with WT or Cav-1-F92A mutant. Scale bar is 500 nm. At least 35 individual images were analyzed per group for the quantification of caveolae/μm of plasma membrane.

Figure 4. Histology, Echocardiography and Signaling

A, Histology of single (WT) and double transgenic (WT + Tg) mice. First panel, gross morphology of heart cross section through the ventricles. Second panel, Wall thickness of main coronary arteries. Third panel, Lung alveolar area. Fourth panel, Lung large bronchioles (B) and arteries (A). Scale bars: first panel: 1 mm, second to fourth panel: 200 μm. B, Echocardiography. Top panel, Representative M-mode images. Bottom panel, Quantification of the left ventricle diameter in diastole (LVD,d) and systole (LVD,s), the thickness of the intraventricular septum wall (IVSW) and the posterior wall (PW). Bar graph present the mean and the SEM for 5 WT and 5 WT + Tg animals. C, Signaling in heart and lung. Top panel shows representative immunoblots. Bottom panel shows the quantification for Akt and ERK activation by measuring the phospho/total ratio.