Caveolin-1 is critical for the maturation of tumor blood vessels through the regulation of both endothelial tube formation and mural cell recruitment

Abstract

In the normal microvasculature, caveolin-1, the structural protein of caveolae, modulates transcytosis and paracellular permeability. Here, we used caveolin-1-deficient mice (Cav(-/-)) to track the potential active roles of caveolin-1 down-modulation in the regulation of vascular permeability and morphogenesis in tumors. In B16 melanoma-bearing Cav(-/-) mice, we found that fibrinogen accumulated in early-stage tumors to a larger extent than in wild-type animals. These results were confirmed by the observations of a net elevation of the interstitial fluid pressure and a relative deficit in albumin extravasation in Cav(-/-) tumors (versus healthy tissues). Immunostaining analyses of Cav(-/-) tumor sections further revealed a higher density of CD31-positive vascular structures and a dramatic deficit in alpha-smooth muscle actin-stained mural cells. The increase in blood plasma volume in Cav(-/-) tumors was confirmed by dynamic contrast enhanced-magnetic resonance imaging and found to be associated with a more rapid tumor growth. Finally, an in vitro wound test and the aorta ring assay revealed that silencing caveolin expression could directly impair the migration and the outgrowth of smooth muscle cells/pericytes, particularly in response to platelet-derived growth factor. In conclusion, a decrease in caveolin abundance, by promoting angiogenesis and preventing its termination by mural cell recruitment, appears as an important control point for the formation of new tumor blood vessels. Caveolin-1 therefore has the potential to be a marker of tumor vasculature maturity that may help adjusting anticancer therapies.

Figure 1

Permeability to albumin is increased in Cav^−/− healthy tissues but not in tumors. ¹²⁵I-Albumin was injected in the tail vein of B16 melanoma-bearing Cav^+/+ or Cav^−/− mice; the animals were sacrificed 1 hour later. Bar graphs represent the extent of ¹²⁵I-albumin cleared from blood and accumulated in skeletal muscle and in tumors; data (normalized for the tissue weight) are expressed as the percent of the total amount of injected radioactivity (**P < 0.01, n = 5).

Figure 2

Parametric mapping by DCE-MRI reveals an increased blood plasma volume in Cav^−/− tumors. A: Typical parametric MR images of K_trans (influx constant from the plasma to the interstitium), K_ep (fractional rate of efflux from the interstitium), and V_p (blood plasma volume per unit volume of tumor) from 6-mm-diameter B16 melanoma established in Cav^+/+ or Cav^−/− mice. B: Quantitative analyses of the indicated pharmacokinetic parameters; data are expressed as means ± SEM (*P < 0.05, n = 5 to 6).

Figure 3

Large-size Cav^−/− tumors present an elevated IFP and an early accumulation of extravascular fibrinogen. A: B16 melanoma tumors were grown up to a 9-mm diameter and then examined with a wick-in-needle to determine IFP. Bar graph represents the IFP (mean ± SEM), as determined in size-matched tumors of both mouse genotypes (**P < 0.01, n = 6). B: B16 melanoma tumors were also collected at earlier stages (ie, when they reached 3 and 6 mm in diameter) for immunohistochemical analyses. Tumor cryosections were stained with antibodies against fibrinogen including its degradation products (red), and the tumor vasculature was co-labeled using CD31 antibodies (green). Top: Scores between 0 and 3 were attributed by two blinded investigators according to the extent of accumulated fibrinogen/fibrinogen degradation products observed in the tumor sections. Bottom: The histograms represent the frequency of observation of the different scores in 3- and 6-mm-diameter Cav^+/+ and Cav^−/− tumors (n = 4 mice, three tumor sections per mouse).

Figure 4

Caveolin deficiency reveals an inverse correlation between angiogenesis and mural cell coverage and directly impacts on tumor growth. B16 melanoma were grown for a maximum of 18 days in Cav^+/+ and Cav^−/− mice. A: Representative pictures of size-matched (6-mm) tumor sections co-stained with α-smooth muscle actin (α-SMA) and CD31 for mural cells (FITC, green) and endothelial cells (TRITC, red), respectively (n = 5 per mouse genotype). Both dual- and single-color channel staining patterns are presented for the two mouse genotypes; the top panels correspond to the center of the tumors, and the bottom panels to the peripheral staining of the tumors. B: Graph represents the evolution of the tumor diameter as measured at days 12, 15, and 18 after injection of tumor cells (**P < 0.01 versus 18-day Cav^+/+ tumors, n = 10). C: Plot represents the extent of in vivo bioluminescence (ie, number of photons) as measured 4 days after the subcutaneous injection of 10⁶ luc⁺-B16 melanoma cells. Representative images are also presented: luciferase activity was detected at earlier stages in tumors from Cav^−/− mice (*P < 0.05 versus Cav^+/+ mice, n = 5).

Figure 5

Silencing caveolin-1 reduces the extent of SMC/pericyte migration and naturally occurs in vivo. A: 10T1/2 cells were grown to confluence and after serum deprivation for 24 hours, a wound assay was performed by scrapping out a 0.5-mm-wide band of cells. Representative picture of migrating (ie, present in the wounded area 24 hours later) and nonmigrating (within the confluent monolayer far from the wound) SMC/pericyte precursor cells as observed by immunofluorescence using caveolin-1 antibodies (TRITC detection, red) and FITC-phalloidin (green). Note that the caveolin subcellular distribution is polarized at the rear of migrating cells (ie, at the opposite of the migration front, arrowheads). B: Bar graph represent the effects of a siRNA targeting caveolin (sequence 206 to 226: AAGATGTGATTGCAGAACCAG) on the protein expression of caveolin (left) and on the migration of SMC/pericyte precursors (right); data are the means ± SEM of the caveolin expression and of the migration index (see Materials and Methods) determined in three independent experiments, respectively (**P < 0.01). A representative caveolin immunoblot (IB) is also shown to validate the silencing of caveolin expression (versus control, scramble RNA) in our experimental conditions. C: Representative immunoblots and bar graph show the levels of caveolin expression in pools of 10 to 15 microvessels microdissected from B16 melanoma established (2 weeks before) by injection of tumor cells at the vicinity of the saphenous arteriole in Cav^+/+ mice; size-matched microvessels isolated from the contralateral (tumor-free) leg were used as control to validate the down-regulation of caveolin-1 in tumor microvessels. This experiment was repeated twice with similar results.

Figure 6

Caveolin deficiency promotes endothelial tube formation and prevents both basal and PDGF-stimulated SMC/myofibroblast outgrowth. A: Photomicrographs show the angiogenic response of 12-day cultured collagen-embedded thoracic aorta ring explants from Cav^+/+ and Cav^−/− mice. The linear organization of endothelial tubes allows discrimination of the corresponding DAPI-stained endothelial cell nuclei from the more dispersed ones belonging to SMCs/myofibroblasts (ie, cells not stained with endothelial cell markers such as CD31 or MECA32 antibodies, not shown). B: Bar graphs represent the numbers of endothelial branching sprouts per aortic rings (top) and the number of SMCs/myofibroblasts (bottom); *P < 0.05, **P < 0.01, n = 10 aorta rings per mouse genotype. C: Top: Photomicrographs show the effects of PDGF-BB (30 ng/ml) on the outgrowth of SMCs/myofibroblasts from Cav^+/+ and Cav^−/− aorta ring explants; note the inverse relationship between the proliferation of these cells and the extent of endothelial tube formation. Bottom: Bar graph represents the extent of SMC/myofibroblast outgrowth determined by measuring the mean distance between these cells and the explant (**P < 0.01, n = 5 aorta rings per mouse genotype). D: Representative immunoblot analysis (top) and quantification (bottom) of phospho-Akt and Akt in SMCs/myofibroblasts (isolated from Cav^+/+ and Cav^−/− aorta rings) after a 20-minute exposure to PDGF-BB (30 ng/ml) (**P < 0.01, n = 3 per mouse genotype).