Lymphatic system of the pancreas.

A network of lymphatic vessels exists within the pancreas. The majority of vessels forming this network lie in the interlobular septa of connective tissue that subdivide the pancreas into lobes and lobules. Peripheral extensions of these interlobular lymphatics can be found within the lobules, but these intralobular lymphatics are relatively sparse. In the main, the intimate relationships of these internal pancreatic lymphatics are with the blood vessels and associated connective tissue. However in random areas, both intra- and interlobular lymphatics come into close relationship with acinar cells. Rarely are there lymphatics associated with islets of Langerhans, and then only where lymphatic vessels in connective tissue septa pass close to a pancreatic lobule that contains an islet at its periphery. Intra- and interlobular lymphatics are similar in structure. Both are thin walled having an endothelial lining and a delicate component of connective tissue. The pattern of interendothelial cell contacts and the sparsity of gaps between adjacent cells suggest that fluid movement through the intracytoplasmic system of vesicles is important in lymph formation in the pancreas. However intercellular transport is also likely to occur by a dynamic process involving fluid movement through dilatations between cells from interstitium to lymphatic lumen. Both exocrine and endocrine secretions of the pancreas may enter thoracic duct lymph directly in pancreatic lymph, but in normal circumstances this route of entry is not quantitatively important. The structural relationships between lymphatics and pancreatic parenchymal cells also make clear that lymph is not a significant pathway for their secretory products. Rather, the arrangement of lymphatics in the pancreas supports the view that lymph is primarily the drainage medium for substances that, for whatever reason, enter the interstitium. In addition, the low flow of lymph compared with that of plasma lends credence to the view that lymph is not a functionally important pathway for endocrine secretions from the pancreas to reach the blood. Both structural and functional evidence suggests that the proper functioning of the lymphatic system is of critical importance in the homeostasis of the pancreas. The lymphatic system of the pancreas, like that in other organs, is essential in the removal of excess fluid from the interstitium. In this sense, the lymphatics may be considered to serve as an overflow, protective, or safety system. When the system is inadequate or its capacity is exceeded, as in inflammation of the pancreas, exocrine secretions entering the interstitium are not cleared and the proteolytic enzymes cause major damage to the tissue. This, in turn, exacerbates the edema, accentuates the inability of lymphatics to drain the fluid, and results in further damage. The fibrosis that ensues damages the lymphatics either directly or through stricture of the surrounding connective tissue. In consequence, they become inadequate at an even earlier stage in subsequent attacks of inflammation and thereby predispose to chronic and recurrent pancreatitis. The larger interlobular lymphatics formed by the junction of their tributaries emerge upon the surface of the pancreas. There they travel primarily with blood vessels and stream toward a ring of lymph nodes that intimately surrounds the pancreas. A second system of nodes extensively involved in drainage from the pancreas is related to the front and sides of the aorta from the level of the celiac trunk to the origin of the superior mesenteric artery. This second set of nodes receives lymph either directly from the pancreas or indirectly from the first echelon of nodes that rings the organ. Although there is general agreement on the disposition of the groups within these sets of nodes, confusion results from the different classifications used by various authors. These classifications range from being purely descriptive, through an alpha and num

Lymphatic system of the rat pancreas.

Light and electron microscopy combined with morphometric analysis were used to investigate the distribution, extent and structure of lymphatic vessels in the head, body and tail of the rat pancreas. Serial sections 3-4 microns in thickness were cut from tissue fixed by perfusion. Alternate sections were processed for light microscopy. Intervening sections were left uncovered to be re-embedded and sectioned for electron microscopy as needed. Vessels with valves were tentatively identified as lymphatics using the light microscope, with final identification being made on adjacent sections by electron microscopy. The ultrastructure of the pancreatic lymphatic vessels was typical of lymphatics generally. Interlobular lymphatic vessels were present throughout the pancreas and were found to be associated primarily with blood vessels lying in connective tissue septa. Intralobular lymphatics were also seen but were comparatively rare. Only about 19% of the wall of the lymphatic system of the pancreas was in close relationship to acinar cells--none was closely related to the endocrine islets. The mean volume density of the system was 0.0012 microns3/microns3 and the profile density of lymphatics was 3.24/mm2. Special attention was paid to the areas of contact between adjacent endothelial cells. Open gaps of more than 30 nm in width were rare. Dilatations and associated cytoplasmic processes, suggestive of a type of intercellular transport, were seen in addition to the intracellular cytoplasmic vesicular system. The findings are consistent with the view that the lymphatic system of the pancreas does not have a specific role in the transport of pancreatic secretions other than the removal of macromolecules that may escape to the interstitium in small amounts under normal circumstances. The fine structure of the endothelial wall suggests that the mechanism of lymph formation in the pancreas is more comparable to that in other encapsulated organs such as the kidney and liver than to that in the dermis or diaphragm where fluid appears to enter lymphatics primarily by way of gaps between adjacent cells.

Lymphatic valves of the rat pancreas.

The detailed structure of pancreatic lymphatic valves in rats was examined in an attempt to identify ultrastructural features that could be correlated with the ability of these delicate structures to withstand retrograde flow. Sprague-Dawley rats were perfusion-fixed and the pancreas processed for light and electron microscopy. Lymphatic vessels were identified by their typical appearance coupled with the presence of valves within their lumen. These valves were consistently formed of cuspid leaflets joined to the lymphatic wall at the bases and sides enclosing valvular pockets or sinuses between cusp and wall. Each cusp or leaflet consisted of two simple squamous endothelial layers separated by a connective tissue core and thus appeared, at first sight, as a simple infolding of the lymphatic vessel lining with its underlying connective tissue. However, certain differences were seen. Frequently the free margins of the cusps, instead of being smooth as might be expected, exhibited endothelial extensions or processes which were arranged in such a way that they could interdigitate with similar extensions on the opposing cusps and thus aid in closure of the valves. A striking difference between the endothelial lining of the vessel and that of the cusp was the presence of a distinct and almost continuous basal lamina underlying the endothelial cells which lined the surface of the cusp facing the valve pocket. The opposing surface of the cusp, which faced the central lumen was similar to the typical lining of lymphatics in showing little or no basal lamina.(ABSTRACT TRUNCATED AT 250 WORDS)

Differences in lymphatic and blood capillary permeability: ultrastructural-functional correlations.

The major structural features of lymphatic capillaries, as they contrast with blood capillaries and as they pertain to endothelial permeability, are reviewed briefly with special emphasis on intrarenal vessels. The most characteristic structural feature of lymphatic endothelium is the discontinuity of the basal lamina. Basal laminae of blood vessels, such as renal glomerular capillaries, are prominent and are known to play a role in preventing extravasation of plasma proteins. By analogy, the lack of a basal lamina around lymphatic capillaries can be considered to be of major functional importance in facilitating access of interstitial macromolecules to the abluminal surface of endothelial cells and thus to the transport pathways that provide entry to the lymph. Tracer studies with horseradish peroxidase, for example, reveal that the protein enters the intraendothelial cytoplasmic vesicular system suggesting that this system may provide a transport pathway. Tracer is also seen between adjacent endothelial cells but in the kidney, liver and thyroid these intercellular channels comprise relatively narrow spaces of about 20 nanometers or less and do not form prominent gaps such as are seen in lymphatics of the diaphragm and skin. Evidence that macromolecular transport across endothelial cells may be asymmetric, favoring movement from interstitium to lymph, is derived from 1) studies using isolated perfused lymphatics, 2) differential luminal and abluminal membrane staining with cationic stains, 3) the presence of charged microdomains on lymphatic endothelial cell surfaces revealed with macromolecules of different charges, and 4) studies on cultured monolayers of porcine arterial endothelial cells.(ABSTRACT TRUNCATED AT 250 WORDS)

Lymphatic system of the thyroid gland in the rat.

Intraglandular thyroid lymph vessels in the rat were studied by qualitative and quantitative analyses in order to obtain information regarding their structure, distribution, relationships, and possible mechanisms of lymph formation. Ultrastructurally, the lymphatic vessels were similar to those described in other organs. The volume density of the lymphatic vessels was 0.007, the profile density 5.68 mm2 and the maximum diameter 17.87 nm. Ultrastructurally, visible transport pathways across the vessels appear to be represented by intracytoplasmic vesicles and channels between endothelial cells. The mean maximum diameter of the vesicles was 96 nm and they occupied 6.9% of the cytoplasm. They were equally distributed between luminal, abluminal, and intracytoplasmic positions. Open junctions (greater than 30 nm) were not seen between endothelial cells, but dilations along part of the length of interdigitating and overlapping contacts were frequent. It was concluded that the mechanism of lymph formation in the thyroid is similar to that in the kidney and liver, but differs from that in the dermis or diaphragm. However, the volume density of the vesicles of the thyroid was twice that of the liver and more than twice that of the kidney. This finding is consistent with an increase in transendothelial vesicular transport of macromolecules.

The renal cortical lymphatic system in the rat, hamster, and rabbit.

Rat, hamster, and rabbit renal cortical lymphatics were examined by light and electron microscopy. Rat and hamster kidneys possessed both intra- and interlobular lymphatics that were structurally similar at the light microscopic level. Ultrastructural examination of the hamster lymphatic endothelium, however, revealed an unusual arrangement of cytoplasmic extensions not seen in the other two species. The intralobular lymphatics were related primarily to tubules, afferent arterioles, and renal corpuscles and were consistent with lymph formation from both plasma filtrate and tubular reabsorbate. Interlobular lymphatics were seen in connective tissue associated with the interlobular blood vessels. Rabbit cortex contained only interlobular lymphatics. Cross-sectional area, maximum diameter, volume density, and profile density were determined by stereological measurements using a computer-based image analyzer. The morphological data from the rat were used, in combination with published values for lymph flow, to calculate the rate of lymph formation per unit area of endothelium in lymphatics of the renal cortex. Among kidneys fixed by retrograde perfusion, the cortical lymphatic system was most extensive in maximum diameter, volume density, and profile density. It was smallest in the rabbit and intermediate in the rat. Lower volume and profile density were found for rat kidneys fixed by the dripping technique. It was concluded that: tubular reabsorbate probably contributes to renal lymph in the rat and hamster, but not in the rabbit; significant differences exist in the extent of the renal lymphatic systems among the three species, with the hamster kidney having the richest network and the rabbit the poorest; the method of fixation influences the measured size and density of renal cortical lymphatics; and the estimated rate of lymph formation in the kidney of the rat is roughly comparable to that in the dog.

Pattern and distribution of intrahepatic lymph vessels in the rat.

The pattern and distribution of intrahepatic lymph vessels were examined by light and electron microscopy in rat livers fixed by perfusion through the portal vein. Lymph vessels were found in the connective tissue of the larger portal canals, where they coursed in close association with branches of the hepatic artery. The smallest portal canals contained no lymphatics. Of the portal canals that lacked a lymphatic, over 50% also lacked an arterial component. Direct connections between the lymphatic lumen and the spaces of Disse or Mall were not observed but lymphatics were found close to Mall's space, separated by only a sparse connective tissue space containing a few collagen fibrils. Lymphatics were neither seen within the parenchyma, nor associated with intercalated (sublobular) veins. Cross-sectional area (223.2 +/- 48.7 micron2 SEM), maximum diameter (20.5 +/- 2.0 microns), volume density (0.00098 +/- 0.00046 micron3/micron3) and profile density (1.8 +/- 0.3 lymphatics per 1 mm2) of hepatic lymph vessels were determined by stereological measurement by a computer-based image analyzer. These data were used to estimate the rate of lymph formation in the liver. It was concluded that 1) initial lymphatics probably originate in the portal canals; 2) the concept that fluid in the space of Disse can be regarded as the principal source of fluid-forming hepatic lymph is questioned, since initial lymphatics appear to be separated from the space of Disse by hepatocytes and the space of Mall; and 3) the rate of lymph formation in the liver of the rat is approximately 0.06-0.08 microliter/min/cm2 of lymphatic endothelium.

Distribution of charged sites on lymphatic endothelium.

The charge distribution on the luminal and abluminal aspects of fixed and living lymphatic endothelium was examined with particular emphasis on the endocytotic vesicular system and interendothelial junctions. Native ferritin (NF; pl = 4.5), when administered abluminally to perfused lymphatics, entered endocytotic vesicles and abluminal and luminal caveolae; NF was also found in intercellular channels, in contrast, NF when applied luminally was largely excluded from both luminal caveolae and intercellular channels. Cationic ferritin (CF; pl = 8.4) bound to the discontinuous basal lamina and to the abluminal plasma membrane, clustering preferentially around the stomata of abluminal caveolae. CF did not, however, bind to the plasma membrane of, or enter, either the vesicular system or intercellular channels, when administered abluminally. When added to the perfusion fluid CF bound to the luminal membrane and to the infundibula of intercellular channels. Ruthenium red (RR) and alcian blue (AB), both cationic stains, bound intensely to the luminal membrane and much less so to the abluminal surface, thus simulating the binding pattern of CF. Unlike CF, however, RR and AB bound to the membranes of abluminal and luminal caveolae with the same level of staining as to the plasma membrane to which they were attached. These results reflect a marked asymmetry in the membrane charge characteristics of endothelial cells.

An ultrastructural study of transport pathways across rat hepatic lymph vessels.

Hepatic lymph vessels in the rat were examined by qualitative and quantitative analyses in order to obtain data pertinent to the mechanism of lymph formation. The ultrastructually visible transport pathways across these vessels appeared to be by way of intracytoplasmic vesicles (89.6 micron mean diameter) and normal channels (22.6 micron wide) between endothelial cells. Three types of intercellular contacts were seen, end-to-end, overlapping, and interdigitating. Only one open junction (greater than 30 nm) was seen in 226 contacts examined. Specialized junctional complexes, either fasciae occludentes or fasciae adherentes, were seen in 65% of the contacts. Approximately one-third of the contacts had a dilatation along part of their length separating the opposing endothelial cells. Vesicles occupied 3.5% of the endothelial cytoplasmic volume and were distributed as follows: 40% opening onto or touching the luminal membrane, 34% without visible connection to either surface, 23% opening onto or touching the abluminal membrane, and less than 3% associated with membranes forming intercellular contacts. It was concluded that the mechanism of lymph formation in the liver is similar to that in the kidney and different from that in the dermis or diaphragm.

Temperature dependence of protein transport across lymphatic endothelium in vitro.

The purpose of the work was to develop an in vitro model for the study of lymphatic endothelium and to determine, using this model, whether or not a cytoplasmic process may be involved in transendothelial transport. Segments of canine renal hilar lymphatics were dissected clean, cannulated at both ends, and transferred to a perfusion chamber for measurement of transendothelial protein transport and for ultrastructural tracer studies. The segments were subsequently processed for light and electron microscopy. By both structural and functional criteria the lymphatics were judged to have retained their integrity. At 37 degrees C, 36 lymphatics showed a mean rate of protein transport of 3.51 +/- 0.45 (SEM) micrograms/min per cm2 of lymphatic endothelium. The rate was influenced by the temperature of the system, being significantly reduced by 49% +/- 4.8, 31% +/- 5.3, and 29% +/- 3.9 when the temperature was lowered to 4 degrees, 24 degrees, and 30 degrees C, respectively. When the temperature was raised to 40 degrees C, the rate was significantly increased by 48% +/- 12.2. The vesicular system and the intercellular regions in vessels with increased or reduced rates of transport were analyzed quantitatively to ascertain whether the rate changes could be correlated with ultrastructurally demonstrable changes in either of these postulated pathways. No significant changes in junctional or vesicular parameters were found between the control lymphatics and those perfused at 24 degrees, 30 degrees, and 40 degrees C. At 4 degrees C, the temperature at which the rate of protein transport was maximally reduced, vesicular size decreased, and the number of free cytoplasmic vesicles increased, whereas the number associated with the abluminal and luminal surfaces decreased. We concluded that isolated perfused lymphatic segments transport protein at a relatively constant rate under control conditions, and that this transendothelial transport comprises both temperature-dependent and temperature-independent mechanisms. The findings were considered in terms of the different theories of lymph formation and were interpreted as providing support for the vesicular theory.

Morphometric analysis of thymic medullary non-lymphoid cell changes during postnatal development.

Thymic non-lymphoid cells have been shown to influence the differentiation of T-lymphocytes. Large numbers of non-lymphoid cells are concentrated in the thymic medullary zone. A morphometric analysis of these cells in the medulla identified little change in their absolute number and volume density during thymic development in the rat. Expansion of thymic medullary volume with age was documented by planimetry. Several subtypes of medullary non-lymphoid cell were identified by electron microscopy, including both squamous and cystic epithelial cells, interdigitating cells and macrophages. Changes in their relative proportion during development were determined with the percentage of squamous epithelial cells declining, interdigitating cells increasing and the other cell types exhibiting little change. Subcellular evidence of secretory activity was sought by analysis of changes in the volume density of organelles in each of the non-lymphoid classes. Cystic enlargement in epithelial cells and a decrease in the volume density of macrophage inclusion bodies were noted but no clear morphometric trend could be correlated with increased secretory activity expected in the developing endocrine thymus.

Ultrastructural evidence for a reabsorptive role by intrarenal veins.

Intrarenal veins, although known to have thin walls, are considered to be simple conducting vessels. Using light- and electron microscopy, the distribution and structure of named intrarenal veins was examined qualitatively and quantitatively in rat kidneys fixed by retrograde arterial perfusion. Although the venous system follows the pattern of arterial branching in general, a class of intracortical veins similar in appearance to interlobular veins but without a companion artery was found in the present study. It is suggested that these vessels be designated intralobular veins. Structurally, we have found intrarenal veins to be surprisingly similar to peritubular capillaries both in respect to their spatial relations to renal tubules as well as ultrastructurally. Like peritubular capillaries, the majority of the wall of intrarenal veins is intimately apposed to renal tubules. This fraction decreases centrally, values of 0.83 being obtained in intralobular veins, 0.69 in interlobular veins, and 0.56 in arcuate veins. The walls of intrarenal veins are comprised of little more than an endothelium, which, like peritubular capillaries, is remarkable for its thinness, high density of fenestrae, and lack of extraintimal elements. Endothelial thickness was not significantly greater in either interlobular or arcuate veins than in peritubular capillaries; the fenestrae were, however, about twice as frequent in peritubular capillaries as in interlobular or arcuate veins and 35 times more frequent than in interlobar veins. The size and numerical and volume densities of uncoated endocytotic vesicles did not differ significantly between peritubular capillaries and any of the intrarenal veins. Based on their marked qualitative and quantitative similarities to peritubular capillaries, we conclude that, like the latter, intrarenal veins are capable of sustaining passive transport between plasma and interstitium.

The organization of endocytotic vesicles in lymphatic endothelium.

The lymphatic endothelium from renal hilar lymphatics in the rat was subjected to qualitative and quantitative ultrastructural analysis with emphasis on the extent and disposition of its vesicular component. The uncoated endocytotic vesicles had an average maximum diameter of 0.073 micron and occupied 7% of the cytoplasm. There were approximately 21 vesicles in each cubic micrometer of cytoplasm. In standard electron microscopic preparations of glutaraldehyde-fixed endothelium 50% of the vesicles appeared to lie free within the cytoplasm. The remainder were seen to touch or open onto the luminal or abluminal surface of the endothelium. The degree to which intracytoplasmic endocytotic vesicles remained discrete or communicated with the plasma membrane was examined using tannic acid and ruthenium red. These substances specifically bind to charged molecules on the cell surface and identify membranes continuous with it. When this technique was applied to aldehyde-fixed tissue nearly 90% of the vesicles that were apparently free within the cell could be shown to retain a connection with the surface, approximately equal numbers communicating with either the luminal or abluminal surface. At least 15% of these vesicles existed as intercommunicating clusters. These results suggest that many vesicles are not simple discrete units, but rather form parts of chains that reach either luminal or abluminal surface. Thus, apparently discrete vesicles may be parts of vesicular chains cut in cross section. The possible relation between chemical fixation and this plan of vesicular organization is discussed and it is concluded that while chemical fixation may result in an overestimation of the numbers of intercommunicating vesicles, the qualitative aspects of vesicle disposition seem largely unaffected. Although the functional significance of our observations has yet to be determined, should this plan of vesicular organization apply to initial lymphatics as well, the concept of vesicular transport solely by random movement of discrete vesicles across lymphatic endothelium should be modified.

Inhibition of artificially induced decidual cell reaction by indomethacin in the mature oophorectomized rat.

Prostaglandin F2 alpha (PGF2 alpha) is capable of inducing a decidual cell reaction (DCR) in the hormonally prepared rat. In the present work indomethacin, a PG synthetase inhibitor, was used to determine whether PGF2 alpha is involved in the DCR induced by artificial stimulation of the endometrium. Thirty-seven animals were oophorectomized and subsequently given daily injections of progesterone for 6 days and one injection of estradiol 17 beta on the fourth day. Later on the fourth day, one of several experimental maneuvers was carried out on the right uterine horn of each animal; these included: 1) introduction of phosphate-buffered saline (PBS) twice into the uterus, 2) intrauterine injection of PGF2 alpha with no subsequent application or manipulation, 3) intrauterine injection of indomethacin followed by subsequent injection of PGF2 alpha, 4) intrauterine injection of indomethacin with subsequent artificial stimulation (scratch), 5) intrauterine injection of PBS with subsequent scratch, 6) scratch followed by injection of PBS, and 7) scratch followed by a second scratch. The extent of the ensuing DCR was assessed 48 h later by measurement of horn weight, by light and electron microscopy, by ranking the DCR, and by the mitotic index. Indomethacin significantly reduced the horn weight in animals treated with scratch but had a much less marked effect on animals treated with PGF2 alpha. Similarly the rank of the DCR and the mitotic index were significantly less in endometria treated by indomethacin with scratch than those treated by indomethacin with PGF2 alpha. From these findings it was concluded that the DCR induced by scratch was inhibited, but not abolished, when preceded by indomethacin. Conversely the DCR induced by PGF2 alpha was not inhibited by indomethacin, thus demonstrating that when local generation of PG is reduced or abolished, PGF2 alpha can sustain the decidual cell response.

Effect of experimental glomerulonephritis on the cells in canine renal lymph with special reference to the veiled cell.

An immunological glomerulonephritis was induced in dogs by the administration of rabbit anti-canine glomerular serum (1 mg/kg) and the effects on white cells in blood, thoracic duct lymph and peripheral renal lymph were observed over 14 days. The principal response was seen in the renal lymph which showed, within 1 hr of the injection of antiserum, a significant increase in its cellular content that lasted for approximately 16 hr. The cells affected by this early response were phagocytic, being both mononuclear and polymorphonuclear. Although the number of lymphocytes leaving the kidney did not increase, the character of these cells changed. Thus, during the first day there was a comparative increase in the number of large lymphocytes, and cells undergoing mitosis appeared in the lymph. These changes suggested the presence of local blast transformation. Later in the course of the response the proportion of small lymphocytes increased and that of large lymphocytes declined. Non-lymphocytic mononuclear cells with characteristics of 'veiled' and 'frilly' cells appeared in renal lymph on the first day and persisted throughout the 14 days. Many of these cells formed the centres of lymphocyte rosettes. The presence of these cells, which have previously been associated with Langerhans cells in the skin, in renal lymph suggests that they have a wide distribution in the body and that they are important during the immunological response to glomerulonephritis.