Tight Junction Status Research Articles

Diabetic db/db mouse presents intravitreal vessels protruding from the optic disc

AbstractPurposeIt is widely known that the main lesions that occur in diabetic retinopathy are oedema and neovascularization, but, still, there is not an accurate model that can reproduce both. In this study, we used a well‐established model of type 2 diabetes, the db/db mice, to assess the presence of retinal oedema and/or neovascularization.Methods12 weeks old db/+ and db/db mouse (n=8) were used in this study. We studied blood retinal barrier integrity by means of vascular leakage in fundus angiography by fluorescein injection and immunocytochemistry against albumin. Furthermore, we used transmitted electron microscopy to observe tight junction status. Finally, to visualize and assess retinal vascularization, eye fundus angiography and histology were performed.ResultsFluorescein and albumin remained inside the vessels proving the integrity of blood retinal barrier, in both controls and db/db retinas. However, retinal fundus angiography revealed that 50% of db/db mice presented intravitreal vessels protruding from the optic disc. Moreover, histological analysis showed a significant increment in the number of vessels in the db/db vitreous compared to db/+, in accordance with the angiography findings.ConclusionsRetinal fundus angiography and histological analysis showed the presence of intravitreal vessels in diabetic db/db mice, suggesting the existence of a current angiogenic process. However, apparently, twelve old week’s db/db mice did not show signs of retinal oedema.

Blood Brain Barrier Permeability and Brain Capillary Endothelial Cell Tight Junctions in the Dahl S Model of Spontaneous Superimposed Preeclampsia

Previous studies in our lab have shown increases in brain water content and Evan’s Blue extravasation in the brains of pregnant Dahl salt‐sensitive (Dahl S) rats compared to Sprague Dawley (SD) rats. The goal of this study was to test the hypothesis that pregnant Dahl S rats display increased blood brain barrier (BBB) permeability and disruption of brain capillary endothelial cell tight junctions.Female (20 week old) Dahl S and SD virgin and pregnant rats were used. Female rats were mated with males of the same strain, and gestational day 1 (G1) of pregnancy was defined by the detection of sperm histologically from a vaginal swab. On G20, rats were perfused with a 1.0% paraformaldehyde/2.0% glutaraldehyde solution and brains were collected for electron microscopy. Transmission electron microscopic images were used to assess the following parameters of brain capillaries: basement membrane status (% dissolved), basement membrane thickness, vesicle formation, tight junction status (% open), luminal membrane protrusions, and vessel diameter. In a separate group, rats were perfused with 5mg/kg of 3kDa Texas Red dextran (1 hour circulation) followed by 5mg/kg injection of 500kDa FITC (3 minute circulation) on G20, brains were collected, and 5μm sections from the posterior cortex were used for confocal microscopy. A ratio of extra‐:intra‐vascular Texas Red mean intensity values was determined to portray a percentage of Texas Red extravasation. All image analyses were conducted blindly.Dahl S pregnant rats showed significant differences in basement membrane status (23% increase of dissolution) and vesicle formation (11±1 vs. 4±1) compared to SD pregnant rats (p < 0.01, 2way ANOVA for multiple comparisons – Tukey post‐hoc). Significant differences were observed in tight junction status among the groups (% open: SD Virgin = 3.03, SD Pregnant = 11.77, Dahl S Virgin = 11.77, Dahl S Pregnant = 26.47, p = 0.04, Fisher’s Exact Test of Independence). No statistical differences were found for measurements of basement membrane thickness, luminal membrane protrusions, or vessel diameter. Dahl S pregnant rats displayed a 17% increase in Texas Red extravasation compared to Dahl S virgin rats (p = 0.01, unpaired t‐test with Welch’s correction).Our results indicate that pregnant Dahl S rats display increases in BBB permeability to low molecular weight fluorescent dextran and increased disruption of brain capillary endothelial cell tight junctions as well as decreased basement membrane integrity. In addition, the increase in vesicle formation may suggest increases in BBB permeability as it relates to transcellular transport across the BBB.Support or Funding InformationKJM was supported by 5T32HL105324‐10 and 20PRE352000021. Work was supported by 1R01HL137673 to MRG and R01HL134711 to JMS.

Decreased Expression of Cystathionine β-Synthase Exacerbates Intestinal Barrier Injury in Ulcerative Colitis.

Endogenous H2S regulates multiple physiological and pathological processes in colon epithelial tissues. The current study investigated the role of cystathionine β-synthase [CBS], a major producer of H2S in colon epithelial cells, in the pathogenesis of ulcerative colitis [UC]-related intestinal barrier injury. The expression and DNA methylation level of CBS were investigated in inflamed and non-inflamed colon tissues collected from UC patients, and the effect of decreased CBS levels on Caco-2 monolayer barrier injury and altered status of tight junctions elicited by tumour necrosis factor/interferon [TNF/IFN] was determined. The expression of CBS and the methylation level of the CBS promoter were assessed in non-inflamed and inflamed colon epithelial tissue samples collected from UC patients. Barrier function, status of tight junction proteins and activation of the NF-κB p65-mediated MLCK-P-MLC signalling pathway were further investigated in Caco-2 monolayers. Decreased expression of CBS and elevated methylation levels of the CBS promoter were observed in inflamed sites compared with in non-inflamed sites in the colon epithelial samples from UC patients. In Caco-2 monolayers, decreased expression of CBS exacerbated TNF/IFN-induced barrier injury and altered localization of tight junction proteins. Decreased expression of CBS predisposed Caco-2 monolayers to injury elicited by TNF/IFN via augmentation of the NF-κB p65-mediated MLCK-P-MLC signalling pathway. Decreased expression of CBS propagates the pathogenesis of UC by exacerbating inflammation-induced intestinal barrier injury. Elevated methylation of the CBS promoter might be one of the mechanisms underlying the decreased expression of CBS in inflamed sites of colon epithelial tissues from UC patients.

In the absence of overt urothelial damage, chondroitinase ABC digestion of the GAG layer increases bladder permeability in ovariectomized female rats.

Loss of integrity of the protective impermeability barrier in the urothelium has been identified as significant in bladder dysfunction. In this study, we tested the theory that the luminal layer of glycosaminoglycans (GAG) serves as an important component of barrier function. The peptide polycation protamine sulfate (PS), 1 mg/ml, was instilled intravesically for 10 min into rat bladders. Chondroitinase ABC (ChABC), 63 IU/ml, was instilled into an additional six rats for 30 min to digest the GAG layer. Unmanipulated controls and sham-injected controls were also performed. After 24 h, the rats were euthanized, the bladders were removed, and permeability was assessed in the Ussing chamber and by diffusion of FITC-labeled dextran (4 kDa) to measure macromolecular permeability. The status of tight junctions was assessed by immunofluorescence and electron microscopy. In control and sham treated rat bladders, the transepithelial electrical resistance were means of 2.5 ± 1.1 vs. 2.6 ± 1.1 vs 1.2 ± 0.5 and 1.01 ± 0.7 kΩ·cm(2) in the PS-treated and ChABC-treated rat bladders (P = 0.0016 and P = 0.0039, respectively). Similar differences were seen in dextran permeability. Histopathology showed a mild inflammation following PS treatment, but the ChABC-treated bladders were indistinguishable from controls. Tight junctions generally remained intact. ChABC digestion alone induced bladder permeability, confirming the importance of the GAG layer to bladder barrier function and supports that loss of the GAG layer seen in bladder biopsies of interstitial cystitis patients could be a significant factor producing symptoms for at least some interstitial cystitis/painful bladder syndrome patients.

Toll-Like Receptor Signalling and the Control of Intestinal Barrier Function.

Epithelial barrier function and innate immunity are fundamental to the pathogenesis of inflammatory and infectious disease. Along with plasma membranes, epithelial cells are the primary cellular determinant of epithelial barrier function. The mechanism by which polarized epithelia form a permeability barrier is of fundamental importance to the prevention of many infectious and inflammatory diseases. Moreover, epithelial cells express Toll-like receptors (TLRs) which upon recognition of conserved microbial factors such as lipopolysaccharide (LPS) induce epithelial responses including epithelial cell proliferation, secretion of secretory IgA into the lumen and production mucins and antimicrobial peptides, thereby promoting intestinal barrier function. Understanding gut barrier integrity and regulation of permeability is crucial to increase our understanding of the pathogenesis of intestinal disease. A variety of tests have been developed to assess this barrier, including assessing intestinal epithelial cell proliferation or death, intestinal tight junction status and the consequence of intestinal barrier integrity loss such as increased intestinal permeability and susceptibility to bacterial infection. Using a mouse model, this chapter describes some of the methods to assess the functional integrity of this epithelial barrier and the part played by a TLR signalling pathway.

Tu1374 1,25-Dihydroxyvitamin D3 Preserves Intestinal Epithelial Barrier Function From TNF-α Induced Injury via Suppression of MLCK-Mlc Phosphorylation Signaling Pathway

Intestinal barrier dysfunction occurs inmultiple enteropathies. Some known proinflammatory cytokines, such as TNF-a, INF-γ, play a pivotal role as mediator between various pathogens and gut epithelium. In our previous work, TNF-α injured Caco-2 monolayers were used to screen potential barrier protective reagents. 1,25-Dihydroxy vitamin D3 (1,25(OH)2D3) turned out to be a very effective protector. This is consistent with literatures which report that 1,25(OH)2D3 has anti-inflammation effect in bowel diseases. To elucidate the underlying mechanism, we performed this work. Given the fact that tight junction (TJ) is the important intercellular structure related with intestinal epithelial barrier function, and that the key role of myosin light chain (MLC) phosphorylation in regulating TJ. The TJ status and MLC phosphorylation was studied.Methods Experiments were conducted on Caco-2 monolayers. Barrier dysfunction was generated by 10ng/ml TNF-α treatment for up to 48h. Barrier function was represented by trans-epithelial electrical resistance (TEER), and macro molecule tracer, FITC-Dextran 40,000 Da (FD-40) trans-membrane flux. Morphological changes of intercellular TJ were visualized by immunofluorescence staining of ZO-1 and Occludin. The expression level of phosphorylated MLC (P-MLC), and its upstream MLC kinase (MLCK) were determined by immunoblotting. The nuclear translocation and activation of NF-kB p65 was analyzed by Electrophoretic Mobility Shift Assay (EMSA) and immunofluorescent staining. Results TNF-α decreased TEER by ~60% after 48h treatment, 10-7M 1,25(OH)2D3 reversed TEER to ~80% of original level. Consistently, TNF-α increased FD-40 flux through monolay by nearly 5 folds, 1,25(OH)2D3 repressed the number to about2 folds. TJ structure altered significantly following TNF-α treatment. 1,25(OH)2D3 ameliorated TNF-α-induced TJ damage. The MLC phosphorylation, as well as MLCK protein expression were induced by TNF-α. These increases were significantly inhibited by 1,25(OH)2D3 treatment. Additionally, 1,25(OH)2D3 suppressed the nuclear translocation and activation of NF-kB p65.Conclusion 1,25(OH)2D3 attenuates TNF-α-induced intestinal epithelial barrier dysfunction by inhibiting the signaling pathway ofMLCK-dependentMLCphosphorylation via NF-kB p65 pathway.

Colostrogenesis during an induced lactation in dairy cattle.

Colostrum immunoglobulin G (IgG) is of major importance for the newborn calf because epitheliochorial placentae do not provide transport in utero. The formation of colostrum occurs in the later stages of pregnancy. Our objectives were to induce lactation in non-pregnant dairy cows and (i) to determine the changes of IgG in serum and mammary secretions during the induction process and (ii) to establish α-lactalbumin (αLA) and prolactin (Prl) alterations to monitor the changing mammary epithelial tight junction status and development pattern. Estradiol-17β (E2) and progesterone (P4) injections in a 1-7days series were combined with a 3-day injection series (day 21-23) of dexamethasone (DEX). Blood and both front quarter secretion samples were collected daily. Milking started 24days after the start of the experiment. Results show that the mammary secretory IgG1 was increased at >7days after the start of steroid injections and depicted a bimodal pattern reaching a high of 16mg/ml at 21day compared with 3.2mg/ml in the serum. There was a small increase in secretory IgG2 that did not correlate with tight junction status, but never reached the serum concentration. The injections of DEX resulted in constriction of tight junctions. Secretory αLA was immediately increased with steroid injections, dropped precipitously after 7days and then began a steady increase until the start of milking. Changes in serum αLA are related to mammary tight junctions while serum Prl gradually increased from 30 to >60ng/ml after the steroid injections stopped. These results provide insights into the mechanisms and timing of colostrogenesis during an induced lactation protocol.

Non-invasive assessment of barrier integrity and function of the human gut

Over the past decades evidence has been accumulating that intestinal barrier integrity loss plays a key role in the development and perpetuation of a variety of disease states including inflammatory bowel disease and celiac disease, and is a key player in the onset of sepsis and multiple organ failure in situations of intestinal hypoperfusion, including trauma and major surgery. Insight into gut barrier integrity and function loss is important to improve our knowledge on disease etiology and pathophysiology and contributes to early detection and/or secondary prevention of disease. A variety of tests have been developed to assess intestinal epithelial cell damage, intestinal tight junction status and consequences of intestinal barrier integrity loss, i.e. increased intestinal permeability. This review discusses currently available methods for evaluating loss of human intestinal barrier integrity and function.

Chronic hypoperfusion increases claudin-3 immunoreactivity in rat brain

Chronic hypoperfusion-induced changes in blood–brain barrier (BBB) tight junction components have not been well studied. In the present study, we investigated the temporal profiles of claudin-3 (a BBB tight junction element) and myleoperoxidase (MPO, a marker of neutrophil infiltration) in the cortical and thalamic regions of rat brain subjected to chronic cerebral hypoperfusion. Chronic cerebral hypoperfusion was induced by an occlusion of two common carotid arteries and the immunoreactivity of claudin-3 or MPO was determined at 1, 2, 3, or 6 weeks after the occlusion. A typical pattern of BBB breakdown was observed from 2 weeks of the occlusion in cortical and thalamic regions based on Evans Blue leakage. Claudin-3 immunoreactivity was increased only in cortical regions after 2 weeks of occlusion. However, after 3 weeks of occlusion, marked increases in claudin-3 immunoreactivity were observed in both cortical and thalamic regions ( P < 0.05), which persisted for at least 6 weeks after the occlusion despite a slight reduction. In contrast, MPO immunoreactivity was increased only in the thalamic regions after 2 weeks of occlusion. But the pattern of MPO immunoreactivity at 3 and 6 weeks after the occlusion was same as claudin-3. At these time points, MPO immunoreactivity was significantly increased in both cortical and thalamic regions ( P < 0.05). These results show that chronic cerebral hypoperfusion increases the immunoreactivity of claudin-3 and neutrophil infiltration in cortical and thalamic regions of the brain, and demonstrate changes in BBB tight junction status during chronic cerebral hypoperfusion.

Time course of milk accumulation-induced opening of mammary tight junctions, and blood clearance of milk components.

Eight cows in early lactation were used to study the effect of milk accumulation on the state of mammary tight junctions and to examine alpha-lactalbumin as an indicator of tight junction permeability in vivo. During three successive periods, the cows were milked twice (4 days), once (6 days), and twice daily (4 days). Plasma lactose, alpha-lactalbumin, and milk sodium concentrations were used as indicators of tight junction permeability. Furthermore, four cows were used to study the clearance of lactose and alpha-lactalbumin from the blood. Milk yield during once-daily milking decreased by 15.4% (P < 0.001). All indicators of mammary tight junction patency increased (P < 0.05) transiently during once-daily milking and indicated that tight junctions opened after approximately 18 h. Plasma alpha-lactalbumin and lactose were highly correlated (r = 0.82, P < 0.001), indicating the suitability of plasma alpha-lactalbumin as an indicator of tight junction status in vivo. Clearance of alpha-lactalbumin and lactose from the blood was best described by a biexponential model. Elimination half-lives for lactose and alpha-lactalbumin were 44 and 40 min, respectively. This study showed that milk stasis during early established lactation induces tight junctions to switch to a leaky state after approximately 18 h and to revert to the closed state shortly after milking.

Evidence that growth hormone stimulates milk synthesis by direct action on the mammary gland and that prolactin exerts effects on milk secretion by maintenance of mammary deoxyribonucleic acid content and tight junction status.

Both PRL and GH play a role in maintaining lactation in the rat, although GH can only maintain pup weight gain at around 50% of the control value, whereas PRL can maintain weight gain close to 90% in the absence of GH. In this study we examined the effects of PRL and GH deficiency (using bromocriptine and an antiserum to rat GH) on milk yield and composition in lactating rats. Treatment with bromocriptine to suppress PRL secretion for 48 h led to a 57% decrease in milk yield with a concomitant decrease in milk protein and lactose yields, but no decrease in fat output. This led to the production of milk with a lower lactose concentration but increased concentrations of protein and particularly fat (increased 100%), which suggests that GH serves an auxiliary role by maintaining an energy-rich milk for the neonate when PRL secretion is reduced. This decrease in milk synthesis was accompanied by decreases in total mammary DNA content and increased milk sodium concentrations. The latter indicates the opening of tight junctions between mammary epithelial cells, which normally occurs during dedifferentiation and involution of the mammary gland. This suggests that PRL maintains milk synthesis at least in part by inhibiting epithelial cell loss and maintaining cellular differentiation. A deficiency in GH, by contrast, caused only a small decrease (24%) in milk yield and had no effect on the major constituents of milk or on milk sodium concentrations or total mammary DNA content. When animals were made deficient in both PRL and GH, however, there was a further marked decrease (88%) in milk volume along with the yields of all major milk constituents, confirming our previous findings that PRL and GH are the major regulators of milk synthesis. Recent studies have indicated that GH exerts direct effects on mammary gland growth, but its actions on milk secretion have been proposed to be mediated indirectly via insulin-like growth factor-I (IGF-I). We, therefore, inhibited lactation by inducing PRL and GH deficiency for 48 h and then attempted to reinitiate it by administering GH either systemically or by local oil-based implants into the mammary gland. Oil-based GH implants were as effective in stimulating milk secretion in the treated (but not contralateral, control) gland as was systemic GH treatment. Thus, GH does act directly on the mammary gland to stimulate milk synthesis, although this does not rule out the possibility that GH acts by stimulating local production of IGF-I.(ABSTRACT TRUNCATED AT 400 WORDS)

Evidence that growth hormone stimulates milk synthesis by direct action on the mammary gland and that prolactin exerts effects on milk secretion by maintenance of mammary deoxyribonucleic acid content and tight junction status

Both PRL and GH play a role in maintaining lactation in the rat, although GH can only maintain pup weight gain at around 50% of the control value, whereas PRL can maintain weight gain close to 90% in the absence of GH. In this study we examined the effects of PRL and GH deficiency (using bromocriptine and an antiserum to rat GH) on milk yield and composition in lactating rats. Treatment with bromocriptine to suppress PRL secretion for 48 h led to a 57% decrease in milk yield with a concomitant decrease in milk protein and lactose yields, but no decrease in fat output. This led to the production of milk with a lower lactose concentration but increased concentrations of protein and particularly fat (increased 100%), which suggests that GH serves an auxiliary role by maintaining an energy-rich milk for the neonate when PRL secretion is reduced. This decrease in milk synthesis was accompanied by decreases in total mammary DNA content and increased milk sodium concentrations. The latter indicates the opening of tight junctions between mammary epithelial cells, which normally occurs during dedifferentiation and involution of the mammary gland. This suggests that PRL maintains milk synthesis at least in part by inhibiting epithelial cell loss and maintaining cellular differentiation. A deficiency in GH, by contrast, caused only a small decrease (24%) in milk yield and had no effect on the major constituents of milk or on milk sodium concentrations or total mammary DNA content. When animals were made deficient in both PRL and GH, however, there was a further marked decrease (88%) in milk volume along with the yields of all major milk constituents, confirming our previous findings that PRL and GH are the major regulators of milk synthesis. Recent studies have indicated that GH exerts direct effects on mammary gland growth, but its actions on milk secretion have been proposed to be mediated indirectly via insulin-like growth factor-I (IGF-I). We, therefore, inhibited lactation by inducing PRL and GH deficiency for 48 h and then attempted to reinitiate it by administering GH either systemically or by local oil-based implants into the mammary gland. Oil-based GH implants were as effective in stimulating milk secretion in the treated (but not contralateral, control) gland as was systemic GH treatment. Thus, GH does act directly on the mammary gland to stimulate milk synthesis, although this does not rule out the possibility that GH acts by stimulating local production of IGF-I.

5-methyltetrahydrofolate urinary excretion: modeling by cultured human kidney cells.

Initial attempts to model urinary folate reabsorption using cultures of HPT cells on porous filter inserts produced disappointing results in that large amounts of 5M were transported across the epithelial monolayer in a nonspecific manner. Since the impermeable molecule inulin was also transported, there apparently existed a significant leakage pathway in the way that the cultured cells were used for transport studies. 5M was bound to the AP membrane and taken up into the HPT cell by specific processes, while inulin was excluded, suggesting that the HPT cells were nevertheless operating functionally. TER values from cultured HPT cells plateaued at a high level when cells became confluent, suggesting an epithelial layer with functional tight junctions. However, when growth media were removed and replaced with transport buffers, there was an immediate loss of TER that fully recovered if the transport buffers were preincubated for 60 min. Under these conditions, transport studies showed the expected results--no movement of inulin through the cell layer and much reduced transfer of folate through the paracellular pathway. These results suggest that a transient opening of tight junctions occurs when growth media are replaced with biological buffers (or with fresh growth media), but that recovery of tight junction function occurs with time.

Effect of cadmium chloride on transepithelial electrical resistance of sertoli cell monolayers in two-compartment cultures—A new model for toxicological investigations of the “Blood-testis” barrier in vitro

The effect of various doses (0.75–24 μ m) of cadmium chloride (CdCl 2) on the development of intercellular tight junctions by immature rat Sertoli cells (Sc) was investigated in vitro using the two-compartment culture system. The status of tight junctions was monitored by repeated measurements of the transepithelial electrical resistance (TER). For defining the specificity of CdCl 2 effects, the TER changes were correlated with Sc secretory activity (immunoactive inhibin), the cell number (DNA content), and viability (MTT test). The effects of CdCl 2 depended on the concentration of the toxicant as well as on the onset and duration of exposure (4 and 18 hr on Day 1 or 5 of culture). The observed effects could be divided into four categories: (1) At highest doses employed, the TER values decreased significantly and irreversibly during 13 days of culture, and the decrease was accompanied by a significant and irreversible drop in inhibin secretion, cell viability, and cell number. (2) Within a narrow range of doses, the irreversible, or only partially reversible, decrease of TER was accompanied by a transient decrease, or no change, of the secretory activity and no significant changes in Sc cell number and/or viability. (3) With still lower doses, the TER values rapidly decreased and then returned to control level within 3–4 days. In this group, no changes in either inhibin secretion or cell viability were observed. (4) Exposure to the lowest doses of CdCl 2 caused a delayed, but significant increase in TER. This increase was not accompanied by noticeable changes in other parameters evaluated. These data suggest that CdCl 2 may selectively compromise, at least in vitro, the development and maintenance of the inter-Sc tight junctions, without affecting the secretory activity or the cell number and viability. However, increasing cumulative doses of CdCl 2 (concentration multiplied by the time of exposure) led to decreased inhibin secretion and cell viability and then, finally, to irreversible cell damage and death. We believe that the experimental model and approach reported in this paper should be very useful for investigating the mechanism of action of known or potential testicular toxicants, particularly those suspected to compromise the integrity of the “blood-testis” barrier.

Effects of cyclic AMP and phorbol ester on transepithelial electrical resistance of Sertoli cell monolayers in two-compartment culture

The effects of dibutyryl cyclic AMP ((Bu) 2cAMP) and phorbol ester (TPA), in the absence or presence of follicle-stimulating hormone (FSH) and/or testosterone, on the development of tight junctions by immature rat Sertoli cells (Sc) were investigated in vitro using the two-compartment culture system. The tight junction status was evaluated by repeated measurements of transepithelial electrical resistance (TER). Untreated cell monolayers developed stable TER of approximately 120 Ωcm 2 during 3 days of culture. Continuous presence of FSH (200 ng/ml) from day 1 onward significantly increased the TER up to approximately 300 Ωcm 2 after a transient (24–36 h) delay. The initial delay was prolonged to 3–4 days by the addition of 1-methyl-3-isobutylxanthine (MIX) (0.2 mM), whereas the subsequent increase of TER was significantly potentiated by the concomitant presence of testosterone (10 μM). Cholera toxin (CHT; 10 ng/ml) and forskolin (FR; 50 μM) mimicked these FSH effects. (Bu) 2cAMP, at concentrations which maximally stimulated immunoactive inhibin secretion (100–500 μM), inhibited the initial TER increase and significantly decreased the TER level when added on days 1 and 5 of culture, respectively. In contrast, low concentrations of (Bu) 2cAMP (4–20 μM) consistently stimulated the TER development, mimicking the stimulatory phase of FSH action. TPA (100 nM) alone had no effect on TER development, but potentiated the stimulatory effect of testosterone in a manner similar to FSH, CHT, FR or low concentrations of (Bu) 2cAMP. These results demonstrate, for the first time, a concentration-dependent, dual effect of exogenous cAMP on the Sc function. Moreover, our data suggest that, at least in vitro, FSH regulates the formation of inter-Sc tight junctions via a cAMP-mediated mechanism and that protein kinase C may modulate the testosterone-, but not FSH-dependent development of tight junctions by immature Sc.