Fenestrated Capillaries Research Articles

A light and fluorescence cytochemical and electron microscopic study of granule-containing cells in the intrapulmonary ganglia of Pseudemys scripta elegans.

In the lung of the red-eared turtle, large numbers of intramural ganglia located in the intraparenchymal connective tissue are demonstrated. Numerous cells in close proximity to the principal ganglionic neurons displayed a bright blue-white formaldehyde-induced fluorescence. Microspectrofluorometric analysis revealed the presence of dopamine (DA) in all cells measured. Subsequent light histochemical staining of the fluorescent sections showed the DA-containing cells to display argentaffinity. Electron microscopy of serial sections revealed cells characterized by dense-cored vesicles corresponding to the intensely formaldehyde-induced fluorescent cells. The argentaffin technique performed directly on ultrathin sections selectively stained the dense-cored vesicles. After fixation with glutaraldehyde followed by dichromate, x-ray microanalysis showed the chromium to be incorporated into the dense granules. Cholinergic-type nerve endings formed axosomatic synaptic contacts with the DA-containing cells, which can therefore be considered as intrinsic postganglionic elements. No efferent synapses from the granule-containing cells to the principal ganglionic neurons could be observed. The granule-containing cells occurred solitarily and in clusters, partially invested with satellite cells, and usually located near fenestrated capillaries; they displayed cytoplasmic processes and indicated emiocytotic granule release. Adjacent granule-containing cells were separated by spaces about 20 nm wide, gradually widening to form intercellular channels with apically projecting microvilli and primary cilia. It is concluded that the intrapulmonary granule-containing cells of the red-eared turtle belong to the APUD system. Furthermore, morphologically these cells appeared to possess a special sensory apparatus which designates them as paraneurons. The possible physiological significance of these intrapulmonary granule-containing cells is discussed.

Fluid and protein clearance in the rat endometrium. Part II: Ultrastructural evidence for the presence of alternative, non-lymphatic clearance mechanisms in the rat endometrium.

In the rat endometrium, resident macrophages and exudate phagocytes ensure proteolysis by means of phagocytosis, macro- and micropinocytosis. Using exogenous tracer particles no ultrastructural evidence could be obtained for the occurrence of endometrial prelymphatics. It is suggested that the free tissue fluid may be drained via the fenestrated (probably venous) blood capillaries.

Partial chemical characterization of the anionic sites in the basal lamina of fenestrated capillaries

The distribution of anionic sites in the basal laminae of the blood capillaries of the murine pancreas was studied in specimens fixed in ruthenium red (RR)-glutaraldehyde mixtures. The sites appeared as discrete, small (6 to 18 nm) particles distributed throughout the three laminae but concentrated primarily in the lamina rara externa, in which—spaced 80–100 nm apart—they formed a planar, partially ordered lattice comparable to that revealed by cationized ferritin in previous studies ( M. Simionescu, N. Simionescu, and G. E. Palade, 1982, J. Cell Biol. 95, 425–434). The chemical nature of the anionic sites was explored by incubating fresh tissue specimens in solutions of selected enzymes before fixation in RR-glutaraldehyde mixtures. Pronase P and papain removed completely the anionic sites and left behind an extensively degraded and disorganized basal lamina. Trypsin caused the removal of anionic sites only, did not degrade the rest of the basal lamina, but detached it completely from the endothelium. Chondroitinase ABC reduced slightly the size and the surface density of RR-stainable particles, and detached focally the rest of the basal lamina from the endothelium and pericytes. Crude heparinase caused a nearly complete removal of anionic sites, and pure heparitinase gave comparable but less extensive results. Similar effects were recorded on the basal laminae of smooth muscle fibers and pancreatic acini and ducts. The results indicate that the anionic sites of all basal laminae examined are contributed primarily by heparan sulfate proteoglycans and trace amounts of chondroitin sulfate proteolycans.

The effects of tranverse cuts caudal to the preoptic recess on the hypothalamo-neurohypophyseal neurosecretomy system

Electrolytic lesions of tissue surrounding the preoptic recess (AV3V region) appear to cause loss of stimulatory input to the supraoptic nuclei from angiotensin receptors and osmoreceptors. To investigate the pathways affected by AV3V lesions, we observed the ultrastructural effects of coronal cuts in a plane caudal to the organum vasculosum lamina terminalis upon supraoptic nuclei and neural lobes of rats. Like AV3V lesions, these cuts caused degeneration of axons and terminals in the supraoptic nuclei. Degenerating terminals lay in axodendritic synapses and in axosomatic synapses on neurosecretory cells. Unlike AV3V lesions, the cuts did not result in an appearance of decreased secretory activity in the supraoptic nuclei or decreased release of hormone from the neural lobe. On the contrary, terminals in the neural lobe tended to be depleted of neurosecretory material, and glial cell processes tended to be withdrawn from the secretory interface at the basal lamina surrounding fenestrated capillaries; both are changes which have been associated with enhanced hormone release. We suggest that inhibitory input to the supraoptic nuclei is lost as a result of these cuts.

Long term effects on the supraoptic nuclei and neural lobe produced by ablation of the tissue surrounding the preoptic recess (AV3V)

In previous investigations, lesions of the AV3V region surrounding the preoptic recess in the rat caused acute loss of the antidiuretic response to severe dehydration. Ultrastructural study of the supraoptic nuclei and neural lobes following several days of adipsia revealed a lack of response in neurosecretory cell bodies, which would normally show hypertrophy, and engorgement of axon terminals in the neural lobe with unreleased hormone, which would normally be depleted. When rats with such lesions were sustained through the acute phase, they regained their ability to concentrate their urine maximally in response to water deprivation, but their antidiuretic responses to angiotensin II or centrally injected hypertonic saline were still impaired. In this study we compared the fine structure of supraoptic nuclei and neural lobes of rats with AV3V lesions and rats with sham lesions after a recovery period of 5 weeks. We also compared responses to 5 days of water deprivation in lesioned and sham lesioned rats. Neurosecretory cell bodies in lesioned rats were smaller and contained fewer neurosecretory granulated vesicles, and axon terminals in the neural lobe contained more neurosecretory granulated vesicles compared to controls. In addition, the basal lamina surrounding fenestrated capillaries was covered to a greater degree by pituicyte processes in neural lobes of lesioned rats. Lesioned rats deprived of water for 5 days had changes in supraoptic nuclei and neural lobes which in general were qualitatively similar to those of controls. However, AV3V lesions blocked the increase in neurosecretory granulated vesicles in cell bodies in supraoptic nuclei and significantly inhibited depletion of neurosecretory granulated vesicles from axon terminals in the neural lobe of water deprived rats. We conclude that after 5 weeks of recovery the neurosecretory system had regained its ability to respond to dehydration, but the response was chronically affected by AV3V lesions.

Microvasculature of the mouse epididymis, with special reference to fenestrated capillaries localized in the initial segment.

The blood supply, microvasculature, and ultrastructure of the capillaries in the epididymis in adult mice were regionally examined. The epididymal duct of the initial segment is surrounded with a dense network of fenestrated capillaries running just under the epithelium. The other segments have loose networks of nonfenestrated capillaries running in the interductal connective tissue. The fenestration of capillaries in the initial segment was markedly reduced in frequency immediately after cutting the efferent duct. In adult mice which were subjected to cutting of the efferent duct neonatally, the dense capillary network did not develop, and fenestrated capillaries were absent in the initial segment. We interpret our results to indicate that the fenestrated capillaries in the initial segment provide for absorption of the testicular fluid and that their development is dependent upon the testicular fluid entering the epididymal duct.

Ultrastructure of oral mucous membrane lesions in psoriasis

The ultrahistopathology of six patients with psoriatic oral mucosal membrane lesions is described. Four of the patients had tongue lesions (geographic or fissured tongue) and two had lesions of the buccal mucosa. The same features were found in both types of lesions. The most prominent feature was the large number of neutrophils. Neutrophil pustules caused flattening of the keratinocytes, resulting in a sponge-like network of epithelial cells. Other features were the dilated fenestrated capillaries surrounded by neutrophils and oedema of the upper lamina propria, basal lamina gaps, basal spongiosis and keratinocyte activation. Psoriasis lesions of the oral mucosa belong to the group of diseases characterized by epithelial non-bacterial pustules.

An analysis of the effect of synovial capillary distribution upon trans-synovial concentration profiles and exchange.

Recent morphometric studies of synovium lining the rabbit knee showed that capillaries extend to greater than 200 microns below the synovial surface; that a proportion of the superficial capillaries are fenestrated; and that the numerical density of the capillaries varies with depth ( Knight & Levick , 1983, 1984). Such observations raised the question 'Down to what depth do capillaries contribute significantly to the diffusional flux of a small extracellular solute (e.g. sulphate) into the joint cavity?' This problem was addressed by a mathematical model of trans-synovial diffusion. The periarticular tissue was considered as a stack of thin slices (delta chi = 2.5 microns), each of known depth and capillary population. An expression for interstitial concentration normal to the surface was derived from the Fick diffusion equations and the mass-balance equation for each slice. Morphometric values were substituted into the area and distance terms of the Fick equation. A range of values was explored for two permeability terms, namely Df/Di (diffusivity across unit fenestral matrix relative to that across unit interstitial matrix) and PS/Q (permeability-surface area product/blood flow for synovial fenestrated capillaries); the ranges lay between the estimated extreme values. Steady-state concentrations and transcapillary fluxes at each depth were obtained by iterative calculation on a microcomputer. The thickness of tissue from which 80% of the joint influx arose (T0.8) was 7-13 microns for diffusion-limited transcapillary exchange (PS/Q = 0.05; Df/Di = 1-0.3); or 9-17 microns for flow-limited exchange (PS/Q = 5; Df/Di = 1-0.03) in areolar synovium, which is the most extensive tissue. The relative independence of the results from the permeability conditions was due to the dominant effect of a peak in capillary density just below the synovial surface. The resistance of the capillary endothelial wall as a fraction of total blood-joint barrier resistance was calculated to be 0.18-0.80 in areolar synovium under conditions of diffusion-limited exchange.

Permeability characteristics of microvascular walls.

Blood to lymph transport of macromolecules has been modeled by assuming a rather numerous population of small pores and a considerably smaller population of large pores across the microvascular walls. Such "black box" studies, however, are inherently incapable of identifying the precise pathways of movement. Electron microscopy has shown a variety of structures which might be identified as the "pores". These include the intercellular junctions; vesicles which may shuttle across the endothelial cells; chains of vesicles forming open pathways; and, in the case of fenestrated capillaries, the fenestrae. With the exception of the shuttling vesicles, these structures have generally been thought to be relatively static, representing a part of the architecture of the normal cell, albeit varying from organ to organ. Arterioles, capillaries and venules have been shown to have differing transport properties. Three possible barriers to transport are now recognized; a fibrous layer on the surface of the endothelial cells; the endothelial cells themselves; and the basement membrane. Three properties of the macromolecule appear to be important in determining its ease of transmural passage; its size; its charge; and its chemical constitution. The simple "shuttling" model of transport by means of vesicles has recently been seriously questioned, and does not appear to be adequate to explain macromolecular transport. More and more evidence is accumulating to indicate that the chemical nature of the macromolecule is of considerable importance in determining whether or not it will transit the microvascular wall. Evidence will be shown that macromolecular transport is largely, if not entirely, a biochemically stimulated set of dynamic events, rather than being explicable on the basis of passive diffusion processes. The current state of our knowledge of these aspects of the problem of macromolecular transport are discussed, and suggestions made as to how to resolve some of the questions still unanswered.

The neuroanatomical and neurovascular organization of normal fetal hypothalamic explants in the third cerebral ventricle of Brattleboro rats with homozygous diabetes insipidus

This investigation has combined microangiography, immunocytochemistry, coupled with transmission and scanning electron microscopy to discuss the neuroanatomical interactions that occur in the brains of Brattleboro rats with diabetes insipidus, following stereotaxic placement of normal fetal hypothalamic fragments into the third cerebral ventricle. Following surgical placement of 17 day post-coitus hypothalamic fragments, host rats with chronic autosomal homozygous diabetes insipidus were killed and their brains were prepared for analysis. A significant degree of explants (68%) flourished and grew in the lumen of the third cerebral ventricle of recipient hosts. Explants were rapidly invaded by host vessels from two routes. (1) Vessels arose from the underlying mantle plexus of portal capillaries which remained fenestrated in the lower one-third of the explants and developed neurovascular (neurohemal) zones. (2) The second source of vessels arose from bed capillaries of the adjacent paraventricular nucleus and adjacent hypothalamus. In contrast to vessels arising from the contact zone, these latter vessels remained unfenestrated. Small clusters of immunocytochemically positive neurons (neurophysin positive) were seen throughout the explants. Numerous healthy magnocellular neurons harboring numerous dense core vesicles and exhibiting multiple axosomatic and axodendritic synapses were seen throughout the neuropil of explants. Axon profiles were noted to terminate upon the abluminal basal lamina of perivascular spaces surrounding fenestrated capillaries in the lower one third of explants. None of the host animals exhibited physiological return to normal parameters of urine output, drinking behavior, and/or urine osmolarity. However the growth and development of explants in the third cerebral ventricle of DI hosts coupled with the emergence of bonafide neurovascular zones supports a potential anatomical substrate for the central delivery of neuropeptide hormones in this experimental model.

Microvascular architecture of the enamel organ in the rat-incisor maturation zone. Scanning and transmission electron microscopic studies.

In order to clarify the microvascular architecture and ultrastructural features of the capillary vessels related to transendothelial transport of metabolites, scanning electron microscopy of tissues digested by HCl-collagenase and of vascular corrosion casts as well as thin-section, tracer, and freeze-fracture replications were employed to study the maturation zone of rat-incisor enamel organ. The enamel-organ maturation zone was shown to have a well-developed, dense capillary plexus. The capillary vessels were distributed along furrows formed by the enamel-organ papillary ridges. In central regions they formed a regular, blindlike network; in the peripheral regions, however, they formed an irregular, circular network. Everywhere except in the nuclear and perinuclear regions, the very thin capillary-vessel endothelial walls were pierced with numerous fenestrations. Such fenestrations were evident in endothelial walls facing the ameloblast-layer site. In tracer experiments, intravenously injected horseradish peroxidase passed through the fenestrations in the endothelial walls to diffuse throughout the enamel-organ extracellular spaces. It did not, however, pass through intercellular spaces or transendothelial channels. The dense, regular distribution of highly fenestrated capillaries in the enamel organ is thought to make possible the rapid transcapillary exchange of various metabolites between the vascular system and the ameloblast and papillary layers that is necessary for enamel maturation.

Anionic sites in basement membranes. Differences in their electrostatic properties in continuous and fenestrated capillaries

We have used ruthenium red, a cationic dye, to detect at the electron microscopic level the presence of anionic sites in various murine basement membranes, with particular emphasis on those of the microvasculature. We have observed anionic sites in all continuous and fenestrated capillaries examined. Terminal lymphatics, which have a discontinuous basement membrane, have sites only where the basement membrane is present. Anionic sites are not present beneath sinusoidal lining cells of the liver which lack a basement membrane. Basement membranes of epithelial cells and those surrounding striated and smooth muscle cells, pericytes, fat cells, and Schwann cells also exhibit anionic sites. We compared the electrostatic properties of anionic sites in basement membranes of continuous and fenestrated capillaries by determining the salt concentration (critical electrolyte concentration, Scott and Dorling, 1965) required to displace ruthenium red from the sites. A concentration of 0.5 M Na + was required to displace ruthenium red from the basement membrane of continuous capillaries of muscle whereas 1.3 M Na + was required to displace ruthenium red from the basement membrane of fenestrated peritubular capillaries of the renal cortex. Our results suggest that anionic sites in the basement membrane of fenestrated peritubular capillaries are more strongly negatively charged than those in the basement membrane of continuous capillaries of muscle. We conclude from this study, first, that anionic sites are a general property of vascular, epithelial, and pericellular basement membranes and, second, that the electrostatic properties of the sites differ in different vascular basement membranes. We speculate that the anionic sites in vascular basement membranes and the variation in their electrostatic properties in different types of capillaries may have important implications for exchange of substances across the capillary wall.

Relationship between granule-containing cells and blood vessels in the rat autonomic ganglia.

Granule-containing (GC) cells and related blood vessels in the superior cervical ganglion and the pelvic plexus of the rat were examined by light and electron microscopy of serial thin sections. In the superior cervical ganglion, GC cells formed many clusters of more than 20 cells. These clusters were supplied with dense networks of fenestrated capillaries, while the ordinary ganglion cells had continuous capillaries distributed more sparsely. Several continuous capillaries diverged from the networks in the GC cell clusters to join with capillaries around the ganglion cells. In the pelvic plexus, continuous capillaries with well-developed pericytes were distributed similarly around the GC cells and the ganglion cells. Dense networks of fenestrated capillaries in the clusters were never seen. Based on these observations, the physiological significance of the GC cells was discussed.

Three-dimensional fine structure of elastic fibers in the perivascular space of some circumventricular organs as revealed by high-voltage electron microscopy

Detailed fine structure of elastic fibers and their distribution in the brains of rats, guinea pigs, and cats were examined by conventional and high-voltage (HVEM) electron microscopes. Small elastic fibers were observed in the perivascular space of fenestrated capillaries in circumventricular organs such as the area postrema, the subfornical organ, and the pineal body. The amorphous component of elastic fibers was identified as elastin by its affinity for orcein and specific susceptibility to elastase. Stereoscopic observation by HVEM revealed the right-turned double-stranded helical structure of microfibrils of elastic fibers. Some physiological roles of these elastic fibers are discussed as related to a possible cerebrospinal fluid absorptive system in the central nervous system.

Permeability of fenestrated capillaries in the isolated pig pancreas, with effects of bradykinin and histamine, as studied by simultaneous registration of filtration and diffusion capacities

Combining an isogravimetric technique and a colorimetric 'on-line' method (Rippe & Stage 1978), filtration capacity (CFC) and diffusion capacity (PS) were simultaneously measured in the maximally vasodilated 'fenestrated' capillary bed of isolated, artificially perfused pancreatic glands in 12 juvenile pigs. Both CFC and PS for Cr-EDTA were about 20 times greater than in the 'continuous' capillary bed of skeletal muscle. With perfusate flow rates of 250 ml/min x 100 g during isogravimetry, PS-Cr-EDTA averaged 110 +/- 10.0 (S.E.) ml/min X 100 g, and diffusion limitation occurred first at flow rates above 300 ml/min X 100 g. CFC was independent of flow rate and averaged 0.641 +/- 0.027 ml/min X 100 g X mmHg. The parallel augmentation of PS-Cr-EDTA and CFC in the fenestrated capillary bed compared with continuous ones seems to reflect both a higher number of capillaries per unit tissue and an increased number of 'small pores' per unit capillary surface, whilst the 'large pore system' appears to be similar. Following bradykinin or histamine infusion, results were similar to those for continuous capillaries (e.g. Rippe, Kamiya & Folkow 1978). Thus, without further vasodilatation CFC increased 3-fold While PS-Cr-EDTA increased only some 25%, and subsequent isoprenaline infusion reversed these effects. Previous studies on continuous capillaries indicate that histamine-type agents act by opening additional 'large pores' in the venular exchange sections (cf. Rippe & Grega (1978, Svensjö 1978), while beta-adrenergic agonists block this effect. The results further suggest that the fenestrae are not involved in these bradykinin-histamine effects, but rather function as a high-density, small pore population.

Die Veränderungen der Butyrylcholinesterase-Aktivität der fenestrierten Capillaren in der Area postrema während der postnatalen Entwicklung

Ontogenetic changes in butyrylcholinesterase (BuChE) activity of fenestrated capillaries in the rat Area postrema lying outside of the blood-brain barrier were investigated by means of light and electron microscopic histochemical methods. Marked enzyme activity could be demonstrated in the capillary endothelial cells of the neonatal rat. The intensity of the light microscopically detectable BuChE activity in the capillaries decreased gradually during the postnatal ontogenesis, but the capacity of the enzyme synthesis of the endothelium remained unchanged in the adult rat, too. Our results do not indicate any direct correlation between BuChE positivity of capillaries and the existence of the blood-brain barrier.

Differentiated microdomains on the luminal surface of the capillary endothelium. I. Preferential distribution of anionic sites.

Cationized ferritin (CF), introduced systemically in vivo or by perfusion in situ, binds preferentially to certain microdomains of the luminal plasmalemma of fenestrated capillaries (mouse pancreas and jejunum). The density and affinity of binding decrease in the following order: fenestral diaphragms greater than coated pits greater than plasmalemma proper. CF binds neither to the membrane of plasmalemmal vesicles and transendothelial channels nor to the corresponding stomatal diaphragms. The distribution pattern is the same when glutaraldehyde fixation precedes the administration of the tracer by perfusion, provided fixation is followed by quenching of residual free aldehyde groups. A much smaller cationic probe (alcian blue) perfused together with the fixative reveals a similar distribution pattern. The functional implications of the association of these microdomains with structures involved in capillary permeability are discussed.

Cells capable of uptake of horseradish peroxidase in some circumventricular organs of the cat and rat.

The structure of mesenchymal cells distributed in some of the "hypendymal organs" of the circumventricular system in the cat and rat was demonstrated after intravenous injection of high doses of horseradish peroxidase. These cellular elements were observed in the vicinity of blood vessels of the organon vasculosum laminae terminalis, subfornical organ and area postrema. Electron-microscopically, these cells located between the basal laminae of the brain parenchyma and the blood capillaries show long cellular processes encircling fenestrated capillaries. Light- and electron-microscopic examination revealed that this cell type is identical with the "horseradish peroxidase-uptake cells,", previously reported in the vicinity of the hypophysial portal system. Such phagocytic cells may be considered as a cellular component intervening between the brain parenchyma and the blood stream, playing a role in selective barrier functions in the above-mentioned circumventricular organs where a blood-brain barrier in the classical sense of the definition is lacking.

Vascular permeability to proteins and peptides in the mouse pineal gland.

The permeability of fenestrated capillaries in the mouse pineal gland to proteins and peptides was demonstrated by means of ultrastructural tracers. Horseradish peroxidase (HRP) and microperoxidase (MP) were injected intravenously and allowed to circulate for approximately 30 s, 1 min, 5 min, 1 or 2 h. The tissue was then fixed by vascular perfusion or by immersion with aldehydes. In all experiments a pronounced extravasation of HRP and MP occurred. Transendothelial vesicular transport seemed to have occurred across the fenestrated capillaries. The most pronounced tracer labeling of vesicles was found after 1 min of MP- or HRP-circulation. The vesicles were uncoated and more than 70% of the HRP- and MP-containing vesicles exhibited diameters between 50 and 110 nm. Furthermore, three other transcapillary pathways taken by the tracers are suggested: 1) via intercellular junctions, 2) through fenestrae and 3) via channels formed by fusion of vesicles with the luminal and abluminal cell membranes. Based on these results, it is assumed that the capillaries in the mouse pineal gland also permeable to peptides synthesized and secreted by the pineal gland.

Endothelial cell junctions in the ciliary body microvasculature

Endothelial cells of the microvasculature (arterioles, fenestrated capillaries, and small venules) of the rabbit ciliary body have been examined in freeze-fracture replicas in ultrahigh vacuum and at very low temperatures, with special regard to their junctional boundaries. The junctional complexes of fenestrated capillaries and venular endothelium are discontinuous and do not seal the interendothelial clefts. They cannot be compared with zonulae occludentes or tight junctions, while the junctions in the arteriolar endothelium are continuous and well organized, suggesting a very low degree of paracellular permeability. It would be of interest to know if these structural differences at the junctional level in each segment of the microvasculature correspond to different paracellular permeability properties.