Search

The role of plasma membrane lipids in regulating the passage of ions and other solutes through the paracellular pathway remains controversial. In this study we explore the contribution of cholesterol (CH) in maintaining the barrier function of an epithelial cell line using the CH-solubilizing agent methyl beta-cyclodextrin (MBCD) to stimulate CH efflux. Inclusion of 20 mM MBCD in both apical and basolateral media reduced CH levels by 70-80% with no significant effect on cell viability. Most of that decrease occurred during the first 30 min of incubation. Recovery of CH content to initial values was nearly complete 22 h after removal of MBCD. Within 30 min of adding MBCD to the culture medium, transepithelial electrical resistance (TER) increased, reaching maximum values 30-40% above controls. This early rise in TER occurred when MBCD was added to either side of the monolayer. The later rapid decline in TER was observed only when MBCD bathed the basolateral surface from which, coincidentally, CH efflux was most rapid. Freeze fracture replicas and transmission electron microscopy of monolayers exposed to MBCD for only 30 min revealed no increase in either the average tight junction (TJ) strand number or the dimensions of the lateral intercellular space. There was a statistically significant increase in the number of TJ particles associated with the E fracture face at this time. This raises the interesting possibility that during CH efflux there is a change in the interaction between TJ particles and underlying cytoskeletal elements. There was no change in staining for occludin and ZO-1. After exposing the basolateral surface to MBCD for 2 h, TER fell below control levels. The accompanying increase in mannitol flux suggests strongly that the decrease in TER resulted from an increase in the permeability of the paracellular and not the transcellular pathway. A decrease in immuno-staining for occludin and ZO-1 at TJs, a striking accumulation of actin at tri-cellular areas as well as a decline in the number of parallel strands, as seen in freeze fracture replicas, suggest that changes in cytoskeletal organization during long incubations with MBCD had physically disrupted the TJ network. Data are presented which suggest that the observed changes in paracellular permeability during CH efflux may be related to increased levels of lipid-derived second messengers, some of which may trigger changes in the phosphorylation status of TJ proteins.

Dendritic cells (DCs) play a critical role in capturing and presenting inhaled antigens to T lymphocytes. We report that pulmonary DCs in the Lewis rat are normally located in the lung in immediate proximity to nerve fibers that contain immunoreactive substance P (SP). Functionally, pulmonary DCs bound 125I-SP and displayed increased motility in vitro in response to graded concentrations of SP. However, SP had no effect on the accessory cell activities of DCs. To examine the role of neural influences on the pulmonary immune response to inhaled antigen, Lewis rats were pretreated with capsaicin (CAP), which damages small nerves and depletes neuropeptide stores, and then challenged intratracheally (i.t.) with hen egg lysozyme (HEL). The number and antigen-presenting cell activities of pulmonary DCs in the CAP-treated rats were comparable to those of controls up to day 14. T lymphocytes harvested from the regional lymph nodes draining the lung were effectively sensitized to HEL in both groups. However, when CAP-treated rats sensitized to HEL i.t. at day 0 were rechallenged with HEL i.t. at day 14, the lungs showed decreased numbers of OX-6+ DCs and diminished pulmonary lymphoid infiltrates compared with controls. We suggest that CAP interferes with a neural-mediated response that contributes to the accumulation of inflammatory cells during the efferent limb of the pulmonary-cell-mediated immune response in vivo.

In the presence of Ca2+, application of trypsin to the basolateral surface of confluent MDCK cell monolayers with formed tight junctions (TJ), induces the formation of basolaterally oriented aberrant TJ strands. Induction of aberrant TJ strands is accompanied by an increase in transepithelial electrical resistance (TER), up to 90%, which upon addition of trypsin inhibitor is maintained for up to 1 h. Thereafter TER returns slowly to baseline values. Under similar conditions, application of trypsin to the apical surface has little or no effect on either TER or the number of aberrant TJ strands. Confocal microscopy of monolayers, immunostained for ZO-1, revealed that this TJ associated cytoplasmic protein, extended below the TJ along the basolateral surface following brief exposure to trypsin. Removing Ca2+ after treatment of the monolayer with basolaterally applied trypsin resulted, after 20 min, in the increased partitioning of TJ particles onto the E fracture face, of both normal and aberrant TJ strands. Like the TJ strands themselves, therefore, aberrant strands may be linked to cytoskeletal elements. Aberrant TJ strands do not form when monolayers, maintained in low Ca2+ medium, are exposed to trypsin, suggesting that under these conditions TJ precursors, and/or trypsin-sensitive proteins regulating TJ strand assembly, are sequestered in a vesicular compartment that is inaccessible to exogenous trypsin. Prolonged exposure of the apical surface of an established, polarized epithelium with intact TJ to trypsin, had little effect on TJ integrity and did not induce aberrant strands.

A substantial body of evidence indicates that the primary sensitization of naive T cells to inhaled antigens occurs in the regional lymph nodes, whereas secondary responses may be generated directly within lung tissue. Ia+ pulmonary dendritic cells are widely distributed within the rat lung where they can participate in the induction of the immune response to inhaled antigens. Recently, two subsets of Ia+ pulmonary dendritic cells have been distinguished based on their expression of Fc receptors (FcR), but little is known concerning their abilities to support the responses of naive or sensitized T cells. In order to address this question, pulmonary FcR+/- dendritic cells have been purified from enzymatic digests of Lewis rat lungs, based on their differential binding to heat-aggregated immunoglobulin. The FcR+/- dendritic cell subsets differed with respect to their light microscopic appearance and in their expression of non-specific esterase. Only the FcR+ subset was able to phagocytize latex beads and showed intracellular phagolysosomes by electron microscopy. Both of the FcR+/- subsets rapidly formed clusters with naive (OX-22+) and sensitized (OX-22-) T cells. However, the clusters yielded by the FcR+ subset were substantially smaller, possibly reflecting their diminished surface membrane expression of the intercellular adhesion molecule-1. The FcR+/- subsets were capable of presenting soluble and particulate antigens to OX-22- T cells. FcR+ cells were less effective than FcR- cells in promoting the proliferative response of OX-22+ T cells to concanavalin A and in the primary mixed leukocyte reaction. We conclude that the FcR+/- pulmonary dendritic cells differ in their abilities to support the responses of naive and sensitized T lymphocytes. This observation may have significance for how primary and secondary pulmonary cell-mediated immune responses are generated.

A role for lipids in the formation of tight junctions (TJ) has been proposed. Attempts to relate changes in whole cell phospholipid composition to the formation of TJs, however, have yielded equivocal results. The object in the present study was to relate changes in TJ of MDCK cells more specifically to alterations in plasma membrane lipids. Cholesterol, which resides primarily in the plasma membrane, was reduced by 25% after incubation of cell monolayers for 24 h in a low Ca2+ medium supplemented with (1-2 microM) Lovastatin, an inhibitor of hydroxymethylglutarylcoenzyme A (HMG-CoA) reductase. This was associated with a halving of the time required for Ca2+ to induce TJ formation as monitored by transepithelial electrical resistance (TER). [3H]Mannitol flux, and morphometric measurements made on freeze fracture replicas confirm that the effects on TER reflect changes in the characteristics of the paracellular pathway. Peak and steady state values of TER were also elevated over control values. The changes in cholesterol content and the time course for TJ assembly were apparent at levels of Lovastatin which do not affect prenylation of proteins, and were prevented if 5 mM mevalonate was present along with Lovastatin. Paradoxically, despite a decrease of approximately 1/3 in the Ca concentration required to yield maximum rates of TJ assembly, 45Ca2+ uptake was actually reduced after cholesterol depletion. The data suggest that cholesterol may modulate the properties of membrane proteins and/or phospholipids which interact with Ca2+, possibly on the exoplasmic leaflet, during TJ assembly.

The formation and maintenance of tight junctions as a barrier to the diffusion of ions and other water-soluble across epithelia is an energy-dependent process. The administration of N-formyl-hydroxyaminoacetic acid (Hadacidin), an analog of aspartate and a competitive inhibitor of adenylosuccinate synthetase, has been shown to inhibit the multiplication of clone 4 MDCK cells and concomitantly reduce the levels of ATP and cAMP (J. Cell. Physiol. 140, 186-194 (1989)). When added to mitotically quiescent confluent cultures of clone 4 MDCK cells, millimolar concentrations of Hadacidin inhibited the generation of transepithelial electrical resistance (TER). In such cultures passive Na+ permeability was similar to controls indicating that the effect of Hadacidin was not on the transcellular pathway. That these cells were viable was demonstrated by their ability to exclude Trypan Blue, and the fact that they remained competent to develop steady state TER upon removal of the inhibitor. Suppression of TER was completely reversed within 48 h of replacing the Hadacidin-supplemented medium with one containing aspartate. Adenosine, but not aspartate, when added simultaneously with the drug, obviated the latter's effect on TER. A mixture of dibutyryl cAMP (db-cAMP) and theophylline was only partially effective in overcoming the effects of Hadacidin on the development of TER and, in fact, markedly delayed its development in control cultures not treated with the drug. When monolayers with established steady state TER were exposed to Hadacidin, no change was noted during the first 24 h. By 48 h, however, TER had decreased to very low values.(ABSTRACT TRUNCATED AT 250 WORDS)

A distinctive glomerular lesion of renal allografts, acute allograft glomerulopathy (AAG), is characterized by hypercellularity and endothelial cell hypertrophy and injury. For elucidating the pathogenesis of this lesion, the infiltrating cells were analyzed by light- and electron-microscopic immunoperoxidase techniques with monoclonal antibodies and compared with those in cellular rejection without AAG (non-AAG). Substantial numbers of T lymphocytes (CD3+,Leu-4+) were detected in the glomeruli in AAG (11.4 +/- 2.4 cells per glomerular cross-section), whereas very few were found in non-AAG (1.4 +/- 1.8, P less than 0.001). In AAG the CD8+ (Leu-2a) subset accounted for essentially all of the intraglomerular T cells, which had a lower CD4/CD8 ratio than the corresponding peripheral blood lymphocytes (P less than 0.03). AAG also differed from non-AAG by the accumulation of intraglomerular mononuclear cells that expressed HNK-1 antigen (Leu-7) and mononuclear phagocytes, which were identified by the presence of the monocyte fibronectin receptor (A6F10). Intraglomerular mononuclear cells were positive for the interleukin-2 receptor (IL2R) and HLA Class II antigens (HLA-DR). The glomeruli in AAG stained more intensely for HLA Class I antigens than the tubules, in contrast to non-AAG cases (P less than 0.03). The interstitial infiltrates in AAG grafts were less intense than in non-AAG of similar duration (P less than 0.01) and had a lower CD4/CD8 ratio, whereas arterial intimal infiltrates were more severe in AAG (P less than 0.03) and consisted of CD8+, but not CD4+, cells. Pathologic features that correlated with poor graft survival were increased numbers of intraglomerular CD8+ cells (both AAG and non-AAG), fewer interstitial T cells (AAG), and more interstitial CD8+ cells (non-AAG) (all P less than 0.05). Cytomegalovirus (CMV) infection was detected in all (8/8) patients with AAG who were studied for infection before or within 3 weeks after the biopsy, compared with 3 of 9 patients with non-AAG. It is proposed that acute allograft glomerulopathy is a distinct form of T-cell-mediated allograft rejection, sometimes associated with CMV infection, in which the glomerular endothelium, often with the arterial endothelium, becomes the principal target of CD8+ T cells.

The effects of systemic cell-mediated hypersensitivity reactions on glomeruli and lungs were investigated in rats. The animals were given an intravenous injection of antigen 7 days after sensitization or were given an intravenous injection of lymph node cells from sensitized syngeneic donors 1 day after antigen injection. Control animals were given an irrelevant antigen or saline. All animals received three injections of 3H-thymidine during the course of the experiments. The animals were sacrificed 2 or 3 days after antigen injection. Autoradiographs of renal and pulmonary tissue showed significantly more labeled mononuclear cells in glomeruli and pulmonary alveolar walls in the experimental groups than in the control groups. Immunofluorescence studies did not reveal antigen, rat IgG, or C3 in glomeruli. The results indicate that systemic cell-mediated reactions can lead to an accumulation of mononuclear cells in glomeruli and lungs, an effect that may contribute to tissue injury.

The relationship between glomerular C3 receptor activity and intraglomerular C3 deposition was studied in 73 cases of various forms of renal disease. C3 receptor activity was measured by enumeration of complement-coated sheep red blood cells (IgM EAC) that adhered to glomeruli in frozen sections and expressed as a percentage of the mean number present in control kidneys. Adjacent sections were studied for the presence and distribution of C3 deposits. The precise location of corresponding dense deposits was determined through 1-mu sections or electron micrographs. Marked depression of C3 receptor activity was found in most cases in which there was accumulation of C3 along the glomerular basement membrane, irrespective of whether the deposits were in a subepithelial, intramembranous, or subendothelial location. In contrast, when C3 was not detectable or was found exclusively in the mesangium, C3 receptor activity was affected only moderately, if at all. These data are consistent with the hypothesis that the interaction of C3 with podocytes is a major determinant in the loss of C3 receptor activity in glomerular disease, but do not exclude the possibility that interaction with mesangial or endothelial cells may also influence C3 receptor activity.

Cellular expression and extracellular deposition of fibronectin and fibrin/fibrinogen in experimental allergic encephalomyelitis (EAE) in guinea pigs (GP) were studied by light and electron microscopic immunohistochemistry of the central nervous system (CNS). Normal GPs had few fibronectin+ and fibrin/fibrinogen+ CNS vessels. Positively stained vessels were more numberous in sensitized GPs prior to and after onset of EAE, and there were extracellular deposits of these molecules in inflammatory foci. Staining was greater in EAE-susceptible strain 13 GPs than in sensitized EAE-resistant strain 2 GP. On ultrastructural analysis both molecules were localized on endothelial cell luminal surfaces and on intravascular mononuclear cells. Fibronectin and the macrophage fibronectin receptor (A6F10) were found on mononuclear cells in parenchyma. Endothelial cell luminal surface fibronectin and fibrin/fibrinogen may mark and/or contribute to early immune events in EAE, including mononuclear cell margination and emigration in inflammatory sites. Genetic factors, cytokines, and components of coagulation may regulate these interactions in vivo.

The substructure of the human glomerular slit diaphragm is examined after perfusion fixation of a human kidney with tannic acid-glutaraldehyde. The present study shows that the substructure of the slit diaphragm in man is very similar to that described in rats and mice, and that the dimensions of the contained pores (50 times 120 A) in the human differ only slightly from those reported for the murine kidney. It is unlikely that rectangular pores of the dimensions of those measured in the human kidney would permit, under physiological conditions, the penetration of significant quantities of serum protein molecules. It is possible, therefore, that the slit diaphragm in the human glomerulus could function as a fine filter for serum proteins, although this remains to be definitely proved.