Evidence for involvement of microtubules in the action of vasopressin in toad urinary bladder. III. Morphological studies on the content and distribution of microtubules in bladder epithelial cells

Structural correlates of the transepithelial water transport

Transepithelial permeability is one of the fundamental problems in cell biology. Epithelial cell layers protect the organism from its environment and form a selective barrier to the exchange of molecules between the lumen of an organ and an underlying tissue. This chapter discusses some problems and analyzes the participation of intercellular junctions in the paracellular transport of water, migration of intramembrane particles in the apical membrane during its permeability changes for isotonic fluid in cells of leaky epithelia, insertion of water channels into the apical membrane and their cytoplasmic sources in cells of tight epithelia under ADH (antidiuretic hormone)-induced water flows, the osmoregulating function of giant vacuoles in the transcellular fluxes of hypotonic fluid across tight epithelia, and the role of actin filaments and microtubules in the transcellular transport of water across epithelia.

Effect of a dynein inhibitor on vasopressin action in toad urinary bladder

1. The effect of the dynein inhibitor erythro-9-[3-(2-hydroxynonyl)] adenine (EHNA) on the osmotic water flow response to vasopressin or exogenous cAMP has been investigated in isolated toad urinary bladders. 2. Pretreatment with serosal EHNA had no effect on basal water flow, but inhibited the development and maintenance of the hydrosmotic response to vasopressin (20 mU ml-1) or 8-(4-parachlorophenylthio)-adenosine 3',5'-cyclic monophosphate (8 CPT-cAMP; 0.1 mM). 3. The inhibitory effect of EHNA on vasopressin-induced water flow was dose dependent. Inhibition occurred in the dose range in which EHNA inhibits the ATPase and motor activities of dynein in vitro. 4. EHNA also inhibited the maintenance of the high rate of water flow established by prior exposure to vasopressin. 5. The inhibitory effect of EHNA on the onset phase of the vasopressin response was attenuated after exposure of the tissue to the microtubule-disruptive drug nocodazole but was fully additive with that of cytochalasin B. 6. EHNA inhibited basal and vasopressin-stimulated transepithelial sodium transport. 7. The findings support the view that EHNA inhibits hormone-induced water flow through an action on a cytoplasmic dynein. The results are consistent with the hypothesis that dynein is involved in the microtubule-based delivery of water channel-containing vesicles to the apical membrane of the granular epithelial cells during both the onset and maintenance of the water permeability response to vasopressin.

Fura-2 transport in toad urinary bladder epithelium: effects of antidiuretic hormone, colchicine and osmotic gradients

Fluorescence is transferred across the toad urinary bladder when fura-2/AM is added to the mucosal or serosal sides of the epithelium. It was now observed that: (1) Oxytocin (20 nM, serosal) increased fluorescence transfer from the mucosal to the serosal but not from the serosal to the mucosal baths. The ratio between the fluorescence intensities recorded with excitation wavelengths of 340 and 380 nm indicates that the calcium sensitive probe (free fura-2) was transferred to the serosal but not to the mucosal compartment by an oxytocin sensitive transport. (2) Preincubation with probenecid did not change fluorescence transfer in basal conditions but significantly reduced the oxytocin induced increase in free fura-2 transport. (3) Fluorescence accumulation inside the tissue was strongly reduced by oxytocin, but only when fura-2/AM was added to the mucosal side. (4) An osmotic gradient, in the presence of oxytocin, further increased the transfer of fluorescence at 380 nm but not at 340 nm. This indicated that the transfer of a calcium-insensitive fraction was being stimulated. (5) Preincubation with colchicine strongly inhibited fluorescence transfer across the tissue, at both 340 and 380 nm (the 340/380 ratio did not change). (6) Tissue accumulation was increased by colchicine. (7) Vanadate did not inhibit fura-2 transfer in the toad urinary bladder. We conclude that intracellularly-generated free fura-2 is only transported across the basolateral border, and that this transfer is stimulated by ADH. The calcium-insensitive fraction is transferred by a temperature-dependent process, sensitive to an osmotic gradient and colchicine.

Effect of colcemid on the water permeability response to vasopressin in isolated perfused rabbit collecting tubules

1. The effect of the microtubule-disruptive agent, colcemid (N-deacetyl-N-methyl-colchicine), on the water permeability response to vasopressin has been investigated in isolated cortical collecting tubules from the rabbit kidney perfused in vitro. 2. Pretreatment of collecting tubules with colcemid inhibited the increase in water permeability elicited by vasopressin, 50 microU ml-1, in a time- and dose-dependent manner. After 75 min exposure to the drug, inhibition of the response to the hormone averaged 72 +/- 6% (n = 4, P < 0.01) at a colcemid concentration of 7.2 x 10(-5) M. Inhibition was estimated to be half-maximal at a colcemid concentration of 1.9 x 10(-6) M. 3. Colcemid, 2.7 x 10(-7) to 7.2 x 10(-5) M, had no effect on basal water permeability nor on the increase in lumen negative potential difference (PD) induced by the hormone. 4. Lumicolcemid, an isomer of colcemid that does not disrupt microtubules, had no influence on the water permeability response to vasopressin. 5. Pretreatment with colcemid, 2.7 x 10(-5) M, for 45 min inhibited the water permeability response to 8-CPT-cAMP, 1.8 x 10(-5) M, by 38 +/- 4% (n = 5, P < 0.01). 6. When collecting tubules were exposed to colcemid, 5.5 x 10(-5) M, for 45 min after the hydrosmotic response to vasopressin had been established, the drug had no influence on the maintenance of the raised water permeability. 7. The results provide further evidence that cytoplasmic microtubules play a role in the initiation of the hydrosmotic response to vasopressin in the mammalian collecting tubule at a site distal to the generation of cyclic AMP.

Effect of nocodazole on the water permeability response to vasopressin in rabbit collecting tubules perfused in vitro

1. The effect of the microtubule-disruptive agent, nocodazole (methyl [5-(2-thienylcarbonyl)-1H-benzimidazol-2-yl] carbamate), on the water permeability response to vasopressin or the synthetic cyclic AMP analogue, 8-parachlorophenylthio-cyclic AMP (8-CPT-cAMP), has been investigated in isolated cortical collecting tubules from rabbit kidneys, perfused in vitro. 2. Pre-treatment with nocodazole, 1-4 micrograms ml-1, had no significant effect on basal water permeability, but inhibited the increase in hydraulic conductivity elicited by vasopressin, 50 microU ml-1, in a dose-dependent manner. Inhibition of the response to the hormone averaged 65 +/- 6% (n = 8, P less than 0.001) at a nocodazole concentration of 4 micrograms ml-1. 3. Nocodazole, 1-4 micrograms ml-1, had no effect on the increase in lumen-negative potential difference (PD) induced by the hormone. 4. Pre-treatment with nocodazole, 4 micrograms ml-1, inhibited the development of the water permeability response to 8-CPT-cAMP, 1.8 x 10(-5) M, by 45 +/- 7% (n = 7, P less than 0.001). 5. When collecting tubules were exposed to nocodazole, 4 micrograms ml-1, after the hydrosmotic response to vasopressin had been fully established, the drug had no inhibitory effect on the maintenance of a high water permeability. 6. The results are consistent with the view that cytoplasmic microtubules play a role in the initiation of the water permeability response to vasopressin in the mammalian cortical collecting tubule at a cellular site beyond the generation of cyclic AMP.

The role of microtubules and microfilaments in the hydrosmotic response to antidiuretic hormone

To test the effects of colchicine and cytochalasin B on the ADH-induced response, unidirectional and net water fluxes were measured at one or two minutes intervals in frog urinary bladder. The action of these agents on the appearance of intramembrane particles aggregates in the luminal membrane of target cells under oxytocin stimulation and the changes in the tissue ultrastructure induced by cytochalasin B were also studied. It was observed that: the time-course of the response to oxytocin was strongly slowed by colchicine while the washout was not affected; the time-course of the 'on and off' of the response to oxytocin was not modified by cytochalasin B; cytochalasin B pretreatment proportionally reduced unidirectional and net water fluxes measured after glutaraldehyde fixation; the combined action of colchicine and cytochalasin B proportionally reduced the net water flux and the number of intramembrane particles aggregates, observed in freeze-fracture studies; after cytochalasin B action the dilation of intercellular spaces classically observed under oxytocin stimulation is strongly reduced. It is concluded that: microtubules probably play an important role in the water channels plug-in, but not in their removal; microfilaments integrity is necessary for the mechanisms inducing intercellular space dilation and the observed results confirm that water permeability is controlled by the number of permeation units present in the luminal border of granular cells and probably represented by the intramembrane particle aggregates.

Differences in c-AMP levels in epithelial cells from pelvic and pectoral regions of the toad skin

Arginine vasopressin (AVP) stimulated active Na+ transport (JNa+) and osmotic water flow (JH2O) across the pelvic skin but only JNa+ across the pectoral skin of the toad, Bufo woodhouseii. Isolated epithelial cells from the pelvic skin had a maximal c-AMP level of 11.16 pmoles/mg protein after 5 min of AVP treatment while that of pectoral skin was 3.64 pmoles/mg protein. The c-AMP level of both skin areas fell to unstimulated values after 20 min of AVP treatment; however, JH2O (pelvic skin) and JNa+ (pelvic and pectoral skin) remained elevated during 3 hr of treatment. Dibutyryl c-AMP and theophylline stimulated JH2O across the pelvic but not the pectoral skin. Maintaining toads in water for 12-24 hr resulted in a substantial lowering of JH2O across the pectoral skin which was not reversible by treatment with c-AMP and theophylline.

Role of membrane fusion in CO2 stimulation of proton secretion by turtle bladder

The changes in cell structure produced during stimulation of proton secretion by CO2 in turtle bladder were examined using ultrastructural morphometric methods. One hour after CO2 addition, the area of the luminal membrane of the carbonic anhydrase-containing (CA) cell population was increased 2.5-fold and the volume percent of electron-lucent cytoplasmic vesicles in these CA cells was decreased by 61%. No changes were observed in the epithelial granular cells. These results suggest that during CO2 stimulation the vesicles fuse with the luminal membrane. CO2 stimulation of proton secretion is inhibited by the cytoskeleton-disrupting drugs colchicine and cytochalasin B and by 99% deuterium oxide as the Ringer solvent. Deuterium oxide also inhibits the decrease in cytoplasmic vesicles. Thus stimulation of proton secretion by turtle bladder CA cells depends to a large extent on vesicle fusion and the resultant increase in luminal surface area.

Ultrastructural characterization of cholesterol distribution in toad bladder using filipin

The polyene antibiotic filipin was used to characterize the cholesterol distribution in the membranes of the toad bladder epithelium in freeze-fracture replicas. The apical membranes of granular and mitochondria-rich cells incorporate moderate amounts of filipin while the basolateral membranes of both cell types incorporate substantially greater amounts. Intracellular membranes, in general, take up very little filipin. The major exception to this is the granule membrane, which appears to be rich in cholesterol. An inverse correlation was found between the density of filipin-sterol complexes in the apical membrane and the incidence of granules in the cytoplasm. This suggests that fusion of granules with the apical membrane may be responsible for variation in the concentration of cholesterol in the apical membrane. Thirty minutes following vasopressin exposure, there is no consistent change in the cholesterol content of the apical membrane of granular cells as measured by the incidence of filipin-sterol complexes. The lack of change in the amount of membrane cholesterol indicates that the vasopressin-induced increase in transepithelial water permeability is not mediated by a change in cholesterol content of the apical membrane.

Effect of antimicrotubule agents on microtubules and steroidogenesis in luteal cells

This report represents an effort to reinvestigate the relationship of steroid hormone processing with microtubule protein, using the highly active progesterone-producing cells of superovulated immature rats as a model steroid hormone system. For the most part, gonadotropin-primed rats were used at 6 days and were reinjected with saline or various doses of colchicine and vinblastine: after 3 h the animals were injected with human chorionic gonadotropin (hCG, 10 IU) or saline for an additional hour. Subsequently, the ovaries were perfused with fixative for morphological studies or luteal cells were isolated for in vitro incubation with various additives. The results are as follows: 1) luteal cell progesterone synthesis is reduced in a dose- and time-dependent manner after treatment with antimicrotubule drugs; 2) additional stimulating agents given in vitro (hCG, Bt2cAMP, epinephrine) do not overcome the reduction produced by treatment with antimicrotubule drugs; 3) luteal cells show normal protein synthesis after treatment with antimicrotubule drugs; 4) luteal cells have microtubules of unusual appearance that are insensitive to the action of antimicrotubule agents. These results show that in the luteinized ovary, both colchicine and vinblastine interfere with hormone-stimulated steroidogenesis. Although the mechanism for the effect is not yet understood, it does not appear to be related to the content of the intact microtubules in luteal cells. Whether the drugs exert their action by binding to the unassembled form of microtubule protein in the cells remains to be determined.

Evaluation by capacitance measurements of antidiuretic hormone induced membrane area changes in toad bladder

A technique for estimating effective transepithelial capacitance in vitro was used to investigate changes in epithelial cell membrane area in response to antidiuretic hormone (ADH) exposure in toad bladder. The results indicate that transepithelial capacitance increases by about 30% within 30 min after serosal ADH addition and decreases with ADH removal. This capacitance change is not blocked by amiloride and occurs whether or not there is a transepithelial osmotic gradient. It is blocked by methohexital, a drug which specifically inhibits the hydro-osmotic response of toad bladder to ADH. We conclude that the hydro-osmotic response of toad bladder to ADH is accompanied by addition of membrane to the plasmalemma of epithelial cells. This new membrane may contain channels that are permeable to water. Stimulation of Na+ transport by ADH is not related to membrane area changes, but appears to reflect activation of Na+ channels already present in the cell membrane before ADH challenge.

Nocodazole inhibition of the vasopressin-induced water permeability increase in toad urinary bladder

Nocodazole is a synthetic antitumor drug that binds rapidly to tubulin. When this drug is applied to toad bladder prior to vasopressin stimulation it inhibits the vasopressin response. A maximum inhibition (68%) is reached with a dose level of 10 micrograms/ml applied one-half hour prior to vasopressin stimulation (20 mU/ml). This compares with an inhibition of 50% seen with a 3-h exposure of the tissue to colchicine (0.1 mM) prior to stimulation with vasopressin. Application of nocodazole (1 microgram/ml) 3 min after hormonal stimulation shows no inhibition of the response at one-half hour past stimulation. These data support the view that microtubules are involved in the vasopressin-induced increase in water permeability in toad bladder and also indicate that this involvement is limited to the period prior to or directly after stimulation.

Quantitative analysis of exocytosis and endocytosis in the hydroosmotic response of toad bladder

This study concerns the timing and magnitude of exocytosis and endocytosis in the granular cells of toad bladder during the hydroosmotic response to antidiuretic hormone. Granule exocytosis at the luminal cell surface is extensive within 5 min of the administration of a physiological dose of hormone. Hydroosmosis becomes detectable during this time period. The amount of membrane added to the luminal surface by exocytosis during 60 min of exposure to hormone can be of the same order of magnitude as the extent of the luminal plasma membrane. Endocytosis, demonstrated by peroxidase uptake from the luminal surface, becomes extensive during the period 15-45 min after hormone administration. Thus, maximal endocytic activity occurs later than the period of most extensive exocytosis and seems to correlate with the onset of the decline in water movement. The amount of membrane retrieved from the luminal surface by endocytosis during 60 min of stimulation is at least three quarters of that added by exocytosis. The bulk membrane movement in ADH stimulated preparations does not require the presence of an osmotic gradient. Colchicine inhibits the hydroosmotic response, the exocytosis of granules, and endocytosis at the luminal surface. These results strengthen our view that the bulk circulation of membrane at the cell surface, via exocytosis and endocytosis, is closely related to the permeability changes occurring at the surface.

Nonelectrolyte fluxes across gastric mucosa in relation to gastric stimulation. Is gastric juice secreted by osmosis or exocytosis?

The effects of histamine and thiocyanate, added to the serosal bathing solution, on unidirectional fluxes of some nonelectrolytes (thiourea, methylated thiourea derivates, mannitol), and on H+, pepsinogen and mucous secretion were investigated in frog (Rana esculenta) fundic gastric mucosa. Histamine (10(-4) M) increases significantly the outfluxes (serosa to mucosa fluxes) of only thiourea and its derivates (but not mannitol) and the stimulation is the greater the more lipidsoluble the nonelectrolyte is. Influxes (mucosa to serosa fluxes) of the same molecules are not affected. In parallel histamine stimulates H+-secretion but does not modify pepsinogen and mucous secretion. SCN- (10(-2) M) inhibits the histamine effect on thiourea outfluxes and on H+-secretion, while pepsinogen and mucous secretion are not affected. Colchicine (10(-4) M) pretreatment inhibits the histamine effect on outfluxes and H+-secretion. It is concluded that: (1) histamine induces a secretion of nonelectrolytes towards the lumen; (2) such secretion is correlated with the hormone-induced secretion of HCl and fluid; (3) this process is mediated by an exocytotic mechanism.

Evidence that ADH-stimulated intramembrane particle aggregates are transferred from cytoplasmic to luminal membranes in toad bladder epithelial cells

In freeze-fracture (FF) preparations of ADH-stimulated toad urinary bladder, characteristic intramembrane particle (IMP) aggregates are seen on the protoplasmic (P) face of the luminal membrane of granular cells while complementary parallel grooves are found on the exoplasmic (E) face. These IMP aggregates specifically correlate with ADH-induced changes in water permeability. Tubular cytoplasmic structures whose membranes contain IMP aggregates which look identical to the IMP aggregates in the luminal membrane have also been described in granular cells from unstimulated and ADH-stimulated bladders. The diameter of these cytoplasmic structures (0.11 +/- 0.004 micrometers) corresponds to that of tubular invaginations of the luminal membrane seen in thin sections of ADH-treated bladders (0.13 +/- 0.005 micrometers). Continuity between the membranes of these cytoplasmic structures (which are not granules) and the luminal membrane has been directly observed in favorable cross-fractures. In FF preparations of the luminal membrane, these apparent fusion events are seen as round, ice-filled invaginations (0.13 +/- 0.01 micrometer Diam), of which about half have the characteristic ADH-associated aggregates near the point of membrane fusion. They are less numerous than, but linearly related to, the number of aggregates counted in the same preparations (n = 78, r = 0.71, P less than 0.01). These observations suggest that the IMP aggregates seen in luminal membrane after ADH stimulation are transferred preformed by fusion of cytoplasmic with luminal membrane.

Evidence that microtubules play a permissive role in hepatocyte very low density lipoprotein secretion

To determine whether a minimum number of assembled microtubules is required for very low density lipoprotein (VLDL) triglyceride TG) secretion in hepatocytes, antimicrotubule drugs of different concentrations were given to rats. Hepatic VLDL-TG release was subsequently measured by a liver perfusion system, and hepatocyte ultrastructural changes were analyzed by quantitative ultrastructural methods. The results demonstrate a tight coupling between the reduction in hepatocyte microtubule content and the reduction in hepatic VLDL-TG secretion which is related to the dose of colchicine or vinblastine administered. The various estimates imply that a minimum number of microtubules is necessary for hepatic VLDL secretion to proceed normally and that hepatic VLDL secretion rates reach their nadir (10--30% of control) when microtubules comprise less than 0.005% of the cytoplasm (or less than 10% of control values) when microtubules comprise less than 0.005% of the cytoplasm (or less than 10% of control values). At this point, hepatocyte Golgi complexes are also greatly altered; Golgi complexes with recognizable dictyosomal membranes are reduced to 15% of control values and the region is filled with large numbers of electron-dense bodies which appear to be lysosomes in the process of digesting VLDL. There is a predilection for the remaining Golgi complexes to be associated with a few segments of microtubules, even when no microtubules can be measured in random samplings of hepatocytes. Clusters of vacuoles containing VLDL are also present throughout the cytoplasm; the limiting membranes of 25% of these vacuoles are studded with ribosomes. These findings demonstrate that the administration of antimicrotubule agents results in decreases in hepatic VLDL-TG secretion which are associated with loss of microtubules and alteration of existing Golgi complexes.

Evidence for involvement of microtubules in the action of vasopressin in toad urinary bladder. II. Colchicine binding properties of toad bladder epithelial cell tubulin

Colchicine, podophyllotoxin and vinblastine have been found to inhibit the action of vasopressin on water movement in the toad urinary bladder. Tubulin is the major colchicine binding component of toad bladder epithelial cells, accounting for approximately 3.3% of the total cell protein. More than 99% of the tubulin is found in the soluble fraction after sonication, the remainder is in the particulate fraction. Similar to the characteristics of the binding of colchicine to tubulins from other sources, the binding of colchicine to toad bladder tubulin is temperature- and time-dependent, is inhibited competitively by podophyllotoxin (Ki= 5.5 x 10(-7)m), and has a binding constant of 1 X 10(6) liters/mole at 37 degrees. Binding activity decays according to first-order kinetics and is stabilized by vinblastine. The characteristics of the interactions of colchicine and podophyllotoxin with epithelial cell tubulin in vitro closely parallel the ability of these drugs to inhibit the response to vasopressin in vivo. These results, coupled with those of functional and morphological studies, support the view that the ability of these drugs to affect vasopressin-induced water movement across toad bladder epithelial cells is related to the depolymerization of cytoplasmic microtubules.

Evidence for involvement of microtubules in the action of vasopressin in toad urinary bladder. I. Functional studies on the effects of antimitotic agents on the response to vasopressin

The antimitotic agents colchicine, podophyllotoxin, and vinblastine inhibit the action of vasopressin and cyclic AMP on osmotic water movement in the toad urinary bladder. The alkaloids have no effect on either basal or vasopressin-stimulated sodium transport or urea flux across the tissue. Inhibition of vasopressin-induced water movement is half-maximal at the following alkaloid concentrations: colchicine, 1.8 X 10(-6) M; podophyllotoxin, 5 X 10(-7)M; and vinblastine, 1 X 10(-7)M. The characteristics of the specificity, time-dependence and temperature-dependence of the inhibitory effect of colchicine are similar to the characteristics of the interaction of this drug with tubulin in vitro, and they differ from those of its effect on nucleoside transport. Inhibition of the vasopressin response by colchicine, podophyllotoxin, and vinblastine is not readily reversed. The findings support the view that the inhibition of vasopressin-induced water movement by the antimitotic agents is due to the interaction of these agents with tubulin and consequent interference with microtubule integrity and function. Taken together with the results of biochemical and morphological studies, the findings provide evidence that cytoplasmic microtubules play a critical role in the action of vasopressin on transcellular water movement in the toad bladder.