Abstract. The interaction of homologous and heterol-

Plasmalemmal vesicles (PVs) or caveolae are plasma membrane invaginations and associated vesicles of regular size and shape found in most mammalian cell types. They are particularly numerous in the continuous endothelium of certain microvascular beds (e.g., heart, lung, and muscles) in which they have been identified as transcytotic vesicular carriers. Their chemistry and function have been extensively studied in the last years by various means, including several attempts to isolate them by cell fractionation from different cell types. The methods so far used rely on nonspecific physical parameters of the caveolae and their membrane (e.g., size-specific gravity and solubility in detergents) which do not rule out contamination from other membrane sources, especially the plasmalemma proper. We report here a different method for the isolation of PVs from plasmalemmal fragments obtained by a silica-coating procedure from the rat lung vasculature. The method includes sonication and flotation of a mixed vesicle fraction, as the first step, followed by specific immunoisolation of PVs on anticaveolin-coated magnetic microspheres, as the second step. The mixed vesicle fraction is thereby resolved into a bound subfraction (B), which consists primarily of PVs or caveolae, and a nonbound subfraction (NB) enriched in vesicles derived from the plasmalemma proper. The results so far obtained indicate that some specific endothelial membrane proteins (e.g., thrombomodulin, functional thrombin receptor) are distributed about evenly between the B and NB subfractions, whereas others are restricted to the NB subfraction (e.g., angiotensin converting enzyme, podocalyxin). Glycoproteins distribute unevenly between the two subfractions and antigens involved in signal transduction [e.g., annexin II, protein kinase Ca, the Ga subunits of heterotrimeric G proteins (as, aq, ai2, ai3)

Luminal membranes of the vascular endothelium were isolated from brain, heart and lungs by modification of their density. The presence of P-glycoprotein (P-gp) was detected by Western blotting in luminal membranes from the endothelium of the three tissues. Strong enrichment in brain capillary luminal membranes, compared with brain capillaries (17-fold) and whole membranes (400-500-fold), indicates that P-gp is mainly located on the luminal side of the brain endothelium. Western blotting was also performed with antibodies directed against GLUT1, glial fibrillary acidic protein, adaptin, IP3R-3, integrins alphav and collagen IV as controls to determine whether the preparations were contaminated by other membranes. Strong enrichment of GLUT1 in brain capillary luminal membranes (9.9-fold) showed that the preparation consisted mainly of endothelial cell plasma membranes. Poor enrichment of glial fibrillary acidic protein (1.4-fold) and adaptin (2.4-fold) and a decreased level of IP3R-3, integrins alphav and collagen IV excludes the possibility of major contamination by astrocytes or internal and anti-luminal membranes. High levels of P-gp in the luminal membranes of brain capillary endothelial cells suggests that it may play an important role in limiting the access of anti-cancer drugs to the brain.

By using an immunoisolation procedure (Stan, R.-V., W.G. Roberts, K. Ihida, D. Predescu, L. Saucan, L. Ghitescu, and G.E. Palade. 1997. Mol. Biol. Cell. 8:595–605) developed in our laboratory, we have isolated a caveolar subfraction from rat lung endothelium and we have partially characterized the proteins of this subfraction which include an apparently caveolae-specific glycoprotein we propose to call PV-1 (formerly known as gp68). The isolation and partial sequencing of PV-1, combined with the cloning of the full length PV-1 cDNA led to the following conclusions: (a) PV-1 is a novel single span type II integral membrane protein (438 amino acids long) which forms homodimers in situ; (b) the transmembrane domain of PV-1 is near the NH2 terminus defining a short cytoplasmic endodomain and a large COOH-terminal ectodomain exposed to the blood plasma; (c) PV-1 is N-glycosylated and its glycan antennae bear terminal nonreducing galactosyl residues in α1-3 linkage. PV-1 is expressed mostly in the lung but both the messenger RNA and the protein can be detected at lower levels also in kidney, spleen, liver, heart, muscle, and brain. No signal could be detected in testis and two lower molecular weight forms were detected in brain. Immunocytochemical studies carried out by immunodiffusion on rat lung with an anti–PV-1 polyclonal antibody directed against a COOH-terminal epitope reveal a specific localization of PV-1 to the stomatal diaphragms of rat lung endothelial caveolae and confirm the extracellular orientation of the PV-1 COOH terminus.

Abstract. The steady-state distribution of endogenous albumin in mouse diaphragm was determined by quantitative postembedding protein A-gold immunocytochemistry using a specific anti-mouse albumin antibody: Labeling density was recorded over vascular lumen, endothelium, junctions, and subendothelial space. At equilibrium, the volume density of interstitial albumin was 18% of that in circulation. Despite this large difference in albumin concentration between capillary lumen and interstitium, ptasmalemmal vesicles labeling was uniformly distributed across the endothelial profile. 68 % of the junctions displayed labeling for albumin, which was however low and confined to the luminal and abluminal sides. The scarce labeling of the endothelial cell surface did not confirm the fiber matrix theory. The kinetics of albumin transcytosis was evaluated by injecting radioiodinated and DNPtagged BSA. At 3, 10, 30, and 60 min, and 3, 5, and 24 h circulation time, blood radioactivity was measured and diaphragms were fixed and embedded. Anti-DNP antibodies were used to map the tracer in aforementioned compartments. A linear relationship between blood radioactivity and vascular labeling density was found, with a detection sensitivity approaching 1 gold particle per DNP-BSA molecule. Tracer presence over endothelial vesicles reached rapidly (10 min) a saturation value; initially localized near the luminal front, it evolved towards a uniform distribution across endothelium during the first hour. An hour was also needed to reach the saturation limit within the subendothelial space. Labeling of the junctions increased slowly, out of phase with the inferred transendothelial albumin fluxes. This suggests that they play little, if any, role in albumin transcytosis, which rather seems to proceed through the vesicular way.central point of interest in the biology of the cardiovascular system concerns the transendothelial transport (transcytosis) of plasma molecules. A large amount of data was generated by experiments in which radioactively tagged molecules of different sizes and physiological significance have been injected or perfused in segments of the body or in isolated organs. The vascular permeability was evaluated by measuring the radioactivity collected in lymph, retained in the tissue, or cleared from the circulating fluids (42). The interpretation of these data suggested some models in which the endothelial layer was considered as a passive sieve endowed with pores of different dimensions. Although widely accepted by physiologists, these models, based solely on diffusional and convectional processes, did not completely explain the experimental data regarding the vascular permeability to macromolecules, leaving room for other possible mechanisms of endothelial transport (35). Purely conceptual, these models prompted a considerable number of morphological studies on vascular endothelium in an attempt to directly visualize the transendothelial pathways followed by macromolecules. However, the putative endothelial pore system has no...