Interendothelial Clefts Research Articles

Protein diffusion barriers in the anterior uvea of the rabbit eye.

The distribution in the anterior uvea of the protein tracer horseradish peroxidase, has been studied in eyes of albino rabbits. The tracer was injected intravenously 30 or 60 min before enucleation. Light microscopy of frozen sections and of semithin sections of Epon‐embedded material revealed: 1. The stroma of the iridial and ciliary processes, and the choroid contained considerable amounts of peroxidase reaction product. 2. In the iris and the ciliary body apart from the processes (the ciliary body base), the peroxidase reaction product was confined to the vessels. 3. A sharp demarcation of peroxidase activity was observed between the iridial processes and the iris stroma, and between the ciliary processes and the ciliary body base.By electron microscopy, peroxidase reaction product was found in the vessels and the stroma of the ciliary processes. In the ciliary body base, reaction product was found within the vessels, but not in the stroma. The endothelium of these vessels had no fenestrations, and there was no sign of passage of peroxidase across the endothelium either by the interendothelial clefts or by vesicular transport. The study indicates that there is a functional barrier against diffusion of proteins from the iridial and ciliary processes across the iris and the ciliary body base.

A light and electron microscopic study of the permeability of the rabbit iris vessels to horseradish peroxidase in experimental uveitis.

Experimental uveitis has been produced in two groups of albino rabbits by a single intravitreal injection of antigen. The animals in group I were immunized by injection of 10 mg of human serum albumin, whereas those belonging to group II received 50 mg. To study the blood‐aqueous barrier to proteins in the iris vessels of these eyes, horseradish peroxidase has been used as a protein tracer by light and electron microscopy. The tracer was injected intravenously at different time intervals (1–30 min) before enucleation.In group I, no leakage of peroxidase from the iris vessels was demonstrated in any eyes except one. The diffusion of peroxidase was halted by the zonulae occludentes of the vascular endothelium, and there was no evidence of vesicular transport of peroxidase through the endothelial cells. In the more heavy inflamed eyes belonging to group II, a prominent inflammatory cell infiltration was found in the iris. In these eyes leakage of peroxidase through the interendothelial clefts of the iris vessels was demonstrated.

Traffic of metabolites between blood and tissue

A hypothesis of blood-tissue exchange is presented comprising the following elements: 1. 1. Interendothelial clefts function as osmotic siphons. They are involved mainly in the Starling-type transport of water and participate in the regulation of blood volume. 2. 2. Vesicular transport involves both water and solutes and is thus nutritive in nature. 3. 3. The intermittent activity of precapillary sphincters establishes transient fugacity gradients of water, extending from open to closed capillaries. Flow of water along these gradients and diffusion are instrumental in the transport of metabolites to and from parenchymal cells. This nutritive flux persists even when the maximal intravascular hydraulic pressure in a tissue region falls below the oncotic pressure of blood plasma.

Relationship between capillary wall structure and “effective” transcapillary osmotic pressure

The colloid osmotic pressure difference across the capillary wall should be determined by the protein concentration difference ( ΔC R ) across just the narrow restricted region of the capillary interendothelial cleft and not the concentration difference ( ΔC T ) across the whole wall as is usually assumed. During capillary filtration, ΔC R should not equal ΔC T and the assumption that the colloid osmotic pressure difference is proportional to ΔC T will be in error. The theoretical magnitude of this error for an idealized cleft is determined.

Transcapillary transport of D- and L-glucose in isolated skeletal muscle.

AbstractThe study was performed to clarify whether D‐glucose traverses muscle capillary walls faster than L‐glucose. The transcapillary transport of D‐glucose was slightly faster than that of L‐glucose in experiments at 37° C. The difference in transport rate persisted when the experiments were carried out at 4° C. This indicates that the faster transport of D‐glucose is not due to facilitated transport involving enzymatic steps. It is proposed that the ground matrix of the interendothelial cleft (and of the extravascular space) gives the D‐form of glucose a slight diffusional advantage over the L‐form. A pore effect cannot be completely ruled out.

Permeability of ovarian blood vessels and follicles of juvenile rats

Enzymatic and heme-protein tracers (including, cytochrome c, horseradish peroxidase, hemoglobin; mol wt 12,000–60,000) and electron-opaque protein and non-protein particles (ferritin, Thorotrast, carbon) have been used to demonstrate the ultrastructural basis of vascular permeability and the route by which blood-borne tracers reach the oocyte in ovaries of juvenile rats. Tracers were injected intravascularly, subsequent to an injection of histamine plus serotonin antagonists (triprolidine, methysergide maleate) into both Wistar and Sprague-Dawley rats. In other experiments tracers were injected alone or into animals treated with histamine and/or serotonin. The results indicate that cytochrome c and peroxidase readily passed through interendothelial clefts of capillaries and venules of the theca. Neither did the concentration or purity of peroxidase (i.e., 1–10 mg Sigma VI or Worthington HPOFF; 10–30 mg Sigma II) greatly affect junctional transport. Treatment with methysergide and triprolidine depressed leakage of peroxidase via gaps in venules of the theca. Transport by means of vesicles was minute. Ovarian capillary permeability to hemoglobin, Thorotrast, ferritin, and carbon was not significantly affected by histamine or serotonin or their antagonists. The junctions were impermeable to these tracers and vesicular transport remained minute. Leakage via gaps in venules accounted, in part, for transport of Thorotrast or ferritin, or a combination of these two tracers. Ferritin, larger than hemoglobin or Thorotrast, was transported faster than either tracer into the tissue spaces. Treatment with histamine plus serotonin antagonists depressed leakage of Thorotrast via gaps, but failed to suppress ferritin build-up in the perivascular space. Movement of cytochrome c and horseradish peroxidase from the blood to the oocyte occurred between 30 and 120 sec after injection of the tracer into the saphenous vein or aorta. The tracers permeated the perivascular basement lamina, passed between cells of the theca, and infiltrated the basement lamina surrounding the ovarian follicles. Cytochrome c and peroxidase surrounded follicle cells and filled the interfollicular cell spaces and the zona of preantrum follicles and, saturated the liquor of antrum-containing follicles. Cytochrome c and peroxidase accumulated within the perivitelline space of follicles at all stages of development and surrounded the first polar body of preovulatory follicles. Under the conditions of our experiments neither the oocyte, polar body, nor follicle cells actively incorporated exogenous cytochrome c or peroxidase by pinocytosis. Larger tracers like Thorotrast and ferritin appeared to accumulate beneath the perifollicular basement lamina and the mesothelium of ovaries of juvenile rats. They did not pass into ovarian follicles and appeared to accumulate beneath the mesothelium and in lymphatic vessels surrounding follicles.

Modified filtration-permeability model of transcapillary transport—a solution of the Pappenheimer pore puzzle?

The filtration-permeability pore model of Pappenheimer and co-workers 1951 has been reexamined by the existing irreversible thermodynamic theory of parallel pathways, pores, and cells. An additional effect is introduced, namely, the constricted nature of the interendothelial cleft (Karnovsky), assumed as the pore. This constriction gives an effective pathlength for filtration about one-fifth of that for diffusion. The parallel pathways effect is small. The reflection coefficient effect and the constriction effect are large but in opposite directions. The various effects largely cancel themselves out and preserve the original model result of a 37 Å slit width, now interpreted as the narrow width (40 Å) of the interendothelial cleft. Alternatively, introducing the Karnovsky electron microscopic dimensions of the interendothelial cleft as given data into the modified model, enables calculation from Pappenheimer et al.'s 1951, isogravimetric data, (permeability)/(reflection coefficient), of permeability and reflection coefficient separately for each test solute. These data are compatible with existing independently measured transcapillary permeabilities and reflection coefficients.