Cytochalasin B inhibition of toad bladder apical membrane responses to ADH

Transcription

1 Cytochalasin B inhibition of toad bladder apical membrane responses to ADH JAMES B. WADE AND WILLIAM A. KACHADORIAN Department of Physiology, University of Maryland School of Medicine, Baltimore, 2121; and Laboratory of Cardiovascular Science, Gerontology Research Center, National Institute on Aging, Baltimore, Maryland WADE, JAMES B., AND WILLIAM A. KACHADORIAN. Cytochalasin B inhibition of toad bladder apical membrane responses to ADH. Am. J. Physiol. 255 (Cell Physiol. 24): C526C53, The possible role of actin microfilaments in antidiuretic hormone (ADH)-induced increases in apical membrane water permeability was investigated in studies that evaluate inhibition by cytochalasin B of both permeability and membrane structural responses in the toad urinary bladder. Experiments were carried out in the absence of a transepithelial osmotic gradient to eliminate possible flow-induced distortions of the response. Measurements of osmotic water permeability after a brief tissue fixation with glutaraldehyde show that cytochalasin B reduces the permeability response to ADH by approximately one-third. Freeze-fracture electron microscopy indicates that the intramembrane particle aggregates, previously found to correlate closely with ADH-induced permeability, are reduced by about the same extent (28%) under these conditions. However, the frequency of apical membrane fusion events was not affected by cytochalasin B treatment. These results suggest that cytochalasin B treatment in the absence of an osmotic gradient alters the ADH-induced permeability through an effect on apical membrane aggregate frequency. toad urinary bladder; membrane structure and function ACTIN AND ACTIN-BINDING PROTEINS are prominent components of the hormone-responsive granular cells of the toad bladder epithelium (1, 22). Evidence that actin microfilaments may play an important role in the antidiuretic hormone (ADH) response has come from multiple physiological studies which show that cytochalasin B (CB) inhibits the hydrosmotic response of the bladder (12, 2-22, 24). However, the functional role of actin microfilaments in the action of ADH and the mechanism whereby CB inhibits transepithelial osmotic permeability remains uncertain. One hypothesis is that microfilaments have a direct role in eliciting the apical membrane permeability response. Freeze-fracture electron microscope studies have shown that CB reduces the number of intramembrane particle aggregates in the apical surface membrane of ADH-treated bladders (12). Because aggregate frequency has been closely and quantitatively correlated with hormone-related changes in water permeability under a wide range of circumstances (2, 3, 7, 1, 14, 15), it is strongly suspected that these structures reflect the incidence of water channels in the apical membrane. C526 Hardy and DiBona (8) have proposed an alternative hypothesis for the role of actin microfilaments in ADH action and the mechanism whereby CB inhibits the hydrosmotic response. Although these and previous investigators observed inhibition by CB when hormone exposure occurs in the presence of an osmotic gradient, the permeability response to ADH was reported to be increased slightly by CB if bladders are stimulated in the absence of an osmotic gradient (8). This result led to the suggestion that the important effect of CB is not at the apical membrane, as indicated by the decrease in aggregates, but is instead caused be its disruptive action on cytoskeletal elements that are important for maintaining water flow when an osmotic gradient is present. Although the incidence of aggregates has not been evaluated for CB studies carried out in the absence of a gradient, such an apparent dissociation between the structures and apical membrane permeability would undermine evidence that aggregates are sites of water permeability. However, conclusions opposite to those of Hardy and DiBona were reached by Parisi et al. (21), who reported a decrease in water permeability for CB-treated frog urinary bladders exposed to hormone in the absence of a gradient. In view of these uncertainties, we have undertaken an independent evaluation of the effect of CB on water permeability and aggregate responses to ADH in the absence of an osmotic gradient. METHODS Paired urinary bladders from large female toads (Bufo marinus) from the Dominican Republic were mounted as sacs tied to plastic cannulas with the mucosal side facing inward. For the principal experiments, a modified Ringer solution [(in mm) 11 NaCl, 2.5 NaHC3, 3. KCl, 2. KH2P4, 1 CaCIZ,.5 MgSO,, and 5 glucose] was used to rinse thoroughly and fill the mucosal side. Tissues were bathed on the serosal side with Ringer solution that was vigorously bubbled with room air. Bladders were initially filled to capacity (13) to standardize tissue distension. This volume was used to calculate total bladder surface area, assuming that the space circumscribed by the mucosal surface of the distended bladder approximates a smooth sphere. Osmotic water permeability was evaluated in paired preparations (n = 6 pairs) stimulated in the absence of an osmotic gradient with 2 mu/ml ADH (arginine

2 MEMBRANE RESPONSES TO ADH C527 vasopressin, Sigma) in the serosal bath, with or without treatment with cytochalasin B (2 X 1m5 M). To assess water permeability for bladders prepared in the absence of a gradient, the fixed-sac technique of Eggena (6) and the general protocol described by Hardy and DiBona (8) were followed. Specifically, CB was present in the serosal bath of experimental bladders 4 min before and throughout the period of exposure to ADH. Dimethyl sulfoxide was used to solubilize CB and was present in serosal media of all preparations, including controls, at a concentration of.2%. After 2 min of stimulation with ADH, the mucosal Ringer solution was replaced with a 1% solution of glutaraldehyde containing.5 M cacodylate (6) to stabilize tissue water permeability. After a 5-min fixation period, the fixative was removed and bladders were rinsed twice with Ringer solution before being filled with Ringer solution diluted 1:5 with distilled water and weighed. Bladders were resuspended in a fresh Ringer solution. Tissue weight change (osmotic water flow) was measured over the first 2.5 min and at -5-min intervals beginning at the 5th minute for a total period of either 3 min (n = 3) or 6 min (n = 3) in length. For the longer experiments, mucosal and serosal media were exchanged at 6 min with fresh solutions, and weight loss was reevaluated for an additional 5 min to check, in both control and experimental bladders, the extent to which water permeability was preserved by the Eggena technique. Water movement was normalized for bladder size by dividing it by the estimated bladder surface area. Osmotic water permeability was calculated from regression analysis of measurements as described by Hardy and DiBona (8). In a separate series of experiments, the modified Eggena technique used by Hardy and DiBona (8, 9) to assess water permeability responses was followed. The essential difference between their procedure and that detailed by Eggena (6) is that the mucosal fixative is added to full-strength Ringer [in their particular case, tris(hydroxymethyl)aminomethane (Tris)-buffered], which they believe minimizes transmural water movement during the brief fixation period (9). An additional six pairs of bladders were evaluated by freeze-fracture electron microscopy to determine the effect of CB on intramembrane particle aggregates when exposure is carried out in the absence of a gradient. Methods used for this structural analysis have been described in detail elsewhere (15). For these bladders, a 2.5% solution of buffered glutaraldehyde was applied from both surfaces for 3 min after a ZO-min treatment with ADH or ADH plus CB. Electron microscopy and quantification of micrographs for apical membrane aggregates and fusion events used methods and criteria as previously described (19) and were done without knowledge of tissue status. The area of aggregates was estimated by the stereologic point counting procedure of Weibel (25) using a grid with.5 cm separation between points. Data were statistically analyzed using Student s t test for paired observations. The 5% level of confidence was used to judge whether mean differences were significant. Data are expressed as means t SE. RESULTS Water permeability response. The effect of 2 x 1D5 M CB on ADH-induced changes in water permeability was evaluated by briefly fixing bladders from the mucosal side using the strategy introduced by Eggena (6). This method allows measurement of water permeability responses in the absence of a gradient because an osmotic gradient is only applied after the bladder has been stabilized by fixation. We evaluated two fixation protocols. The first series of experiments, shown in Table 1, employed the traditional Eggena protocol. Whether estimated from regression analysis of weight changes (Fig. 1) during ation periods or estimated directly from the flow observed for the first period immediately after the imposition of an osmotic gradient, this series of experiments revealed an inhibitory effect of CB of ~25% (range Zl-28%). Although we found that permeability was not perfectly preserved in the fixed preparations, the relative stabilization of permeability as assessed by replacing the bathing fluid with fresh solution (Fig. 1) was comparable to that originally reported by Eggena (6) and not different in CB-treated vs. control tissues (83 t 9 vs %, respectively). Thus the fall in osmotic water flow with time that is shown in Fig. 1 appears to be largely caused by the decreasing osmotic gradient, caused by the previous mucosal to serosal water movement. TABLE 1. Effect of cytochalasin B on ADH-induced water permeability in absence of an osmotic gradient using Eggena fixation technique (6). Method Used Weight change at 2.5-min Regression of weight changes at 3-min Regression of weight changes at 6-min n Control Experimental WW (ADH + CB) 6 314t15 248t25 co kl6 246t3 co.1 3 3t1 217k26 co.5 Values are means t SE. Permeability measured in pm/s I I I I I I I - FRESH BATHS - I I I I I I I TIME (min) FIG. 1. Regression analysis of osmotic water flow from weight changes in control () and CB treatment (2 x 1m6 M; ) for a typical experiment. Correlation coefficient for control was.989 (P <.1) and for CB was.98 (P <.1). At 6 min (I), bathing solution on both surfaces was removed and replaced with fresh solution to reestablish osmotic gradient.

3 C528 MEMBRANE RESPONSES TO ADH An additional series of experiments was carried out using the modified Eggena fixation technique employed by Hardy and DiBona (8,9). Because a shift in permeability caused by exposure to an osmotic gradient during the fixation period might occur, we weighed bladders over the fixation period. These measurements showed that there was virtually no significant water movement across the tissue preparations as they were being fixed. There was no difference between the paired preparations and possibly a very slight net movement of water into the bladder sacs (.1 t.1 mg. min- crnd2 for both ADH and ADH + CB tissues). Upon exposure to an osmotic gradient in the ation period, the bladders fixed by this protocol also showed an inhibitory effect of CB exposure (Table 2). For this series of experiments, the estimated water permeability responses tended to be lower than in the previous series but CB exposure reduced the response by about 36%. Structural response. As shown in Table 3, evaluation of the intramembrane particle aggregates by freeze-fracture electron microscopy demonstrates that these structures are significantly reduced by CB treatment. Both the frequency of aggregates (number per area of cell surface) and the area of aggregates (cumulative area of aggregates per area of cell surface) are reduced by -28% with CB treatment. Thus the size of the aggregates (area per aggregate) does not appear to be affected by CB treatment, which is in agreement with what has been previously reported for CB-treated bladders prepared in the presence of a gradient (12). More importantly, CB significantly inhibits the incidence of ADH-induced apical membrane aggregates in the absence of an osmotic gradient; although the 28% inhibition we observed in the present studies is less striking than the 44% inhibition previously observed with osmotic gradient exposure (12). TABLE 2. Effect of cytochalasin B on ADH-induced water permeability in absence of an osmotic gradient, using modified Eggena technique (9). Method Used Control Experimental (ADW (ADH + CB) Weight change at 2.5-min 238t22 15t23 co.1 Regression of weight 193t15 123t18 co.1 changes at 6-min Values are means t SE for 5 experiments. Permeability measured in pm/s. TABLE 3. Effect of cytochalasin B on intramembrane particle aggregates and fusion events in absence of an osmotic gradient Aggregate frequency no./1 Pm2 Cumulative area occupied by aggregates, pm2/1 pm2 Mean aggregate size, low3 Pm2 Fusion events, no./1 pm2 Values are means k SE for 6 experiments k25 co kO co.1 9.7t kO.51 NS 19t1.5 25k4.5 NS NS, not significant. The frequency of fusion events as previously defined (18, 19) was also evaluated in our study. The incidence of these structural features is not affected by CB exposure (Table 3), a result consistent with findings for CB treatment in the presence of a gradient (19). Because fusion events are an order of magnitude less numerous than aggregates and vary considerably in their incidence, we evaluated the possibility that the number of micrographs sampled might be insufficient to detect a possible effect of CB. For half of the tissue samples (3 control, 3 CB treated), it was possible to obtain micrographs of apical membrane from additional cells so that 15-2 cells per bladder could be evaluated compared with the 1 cells usually examined. Despite this increase in sample size, no important shift in fusion event incidence could be detected. DISCUSSION Our measurements of osmotic water permeability after ADH exposure in the absence of an osmotic gradient show a significant decrease in permeability response in bladders exposed to CB. Although our results are in agreement with those reported by Parisi et al. (21) on the frog bladder, our findings cannot be easily reconciled with the stimulatory effect of CB reported by Hardy and DiBona (8). We evaluated two fixation protocols, one that was identical to that used by Hardy and DiBona (8) and consistently found the water permeability response to ADH reduced by about one-third with CB treatment. There is one substantive variable that might account for the difference in results. We used only fully distended bladders (lo-12 ml internal volume) to ensure comparability with our previous functional and structure studies (12). The studies of Hardy and DiBona (8) employed substantially less distension (5 ml for apparently similar sized bladders; M. Hardy, personal communication). In view of the confounding effect of tissue distension on water flow previously reported in detail (l3), we speculate that our values reflect CB effects on the apical membrane response, whereas the previously reported enhancement of water flow may result from action(s) of CB at another site (8). Our freeze-fracture analysis shows that apical membrane aggregates are reduced by CB treatment. Taken together with the other available observations (11, 19-Zl), this provides strong evidence that actin microfilaments have a role in eliciting the apical membrane permeability response to ADH. Actin microfilaments may well have other important roles and CB appears to have effects at multiple sites (4, 12, 21, 22). It should be noted that our absolute values for aggregate frequency and area (Table 2) are substantially greater than those previously observed when CB treatment is carried out in the presence of a gradient (12). Although our CB treatment differed from that of the previous structural study, both studies used conditions found to produce a maximal inhibition of the water flow response to ADH (22). The difference in the magnitude of the responses is most likely due to the difference in gradient conditions in view of previous reports showing that both the structural and physiological responses elicited in the apical membrane are about twice as great

4 MEMBRANE RESPONSES TO ADH c529 when hormone exposure is carried out in the absence of gradient driven water flow itself (5, 7, 11, 18, 23). Moreover, CB inhibited the aggregate response by 28% in the absence of a gradient, whereas an inhibition of 44% was observed in the previous study where osmotic flow occurred (12). It is possible that osmotic water flow could conceivably augment the inhibitory effect of CB on the apical membrane even though the difference in the magnitude of inhibition observed is not compellingly large. The 28% inhibition of aggregates is adequate to explain the 25-36% inhibition of permeability we observe under these conditions. Previous studies have shown that the relationship between aggregates and transbladder water flow is nonlinear in unfixed tissues consistent with a barrier to flow in series with the apical membrane (17). Our present demonstration that aggregates and flow fall proportionately in tissues that are fixed in the absence of a gradient suggests that this series barrier does not play a detectable role under these conditions. The analysis of fusion event frequency carried out in this work has implications regarding the possible importance of these structures as sites for water flow. There is strong evidence that fusion of aggregate carrying tubular vesicles ( aggrephores ) with the apical surface produces these structures (19). Levine et al. (16) have proposed that aggregates in the walls of these fused tubules may be important sites for water flow and specifically that the high values measured for osmotic permeability relative to diffusional permeability result from the geometry of fused aggrephores.as in the previous freeze-fracture study of CB action (l2), we were unable to detect a decrease in the incidence of fused aggrephores although many more occur in the absence of a gradient and much more membrane was examined. Because aggregates are reduced in the flat region of the apical membrane in proportion to the measured reduction in water permeability, our results suggest that these aggregates are more important sites for water flow than suggested by the hypothesis of Levine et al. (16) and that fused aggrephores are unlikely to represent exclusively the important structures for water permeability. It should be kept in mind, however, that current methods for assessing fused aggrephores do not assess directly the presence or area of aggregates in the wall of fused aggrephores. Thus it is possible to imagine a change in the geometry of fused aggrephores (e.g., if only very short aggrephores were to fuse with the apical membrane when CB is present) such that undetected changes in these structures occur. Because a direct evaluation of water flow through fused aggrephores does not exist, the aggregates in the apical surface membrane remain attractive as primary sites of water permeability because their incidence correlates closely with water permeability. We gratefully acknowledge Drs. Jacqueline Muller and Sherman Levine for their helpful suggestions and the excellent technical assistance of Kathy Lowe in these studies. This work was supported by National Institute of Diabetes Digestive and Kidney Diseases Grant DK and National Science Foundation Instrumentation Grant DMB Received 21 September 1987: accepted in final form 1 Mav REFERENCES 1. AUSIELLO, D. A., H. L. CORWIN, AND J. H. HARTWIG. Identification of actin-binding proteins in toad bladder epithelia. Am. J.Physiol. 246 (Renal Fluid Electrolyte Physiol. 15): FlOl-F14, BROWN, D., G. I. SHIELDS, H. VALTIN, J. F. MORRIS, AND L. ORCI. Lack of intramembranous particle clusters in collecting ducts of mice with nephrogenic diabetes insipidus. Am. J. Physiol. 249 (Renal Fluid Electrolyte Physiol. 18): F582-F589, CHEVALIER, J., J. BOURGUET, AND J. S. HUGON, Membrane associated particles: distribution in frog urinary bladder epithelium at rest and after oxytocin treatment. Cell Tissue Res. 152: , DIBONA, D. R. Cytoplasmic involvement in ADH-mediated osmosis across toad urinary bladder. Am. J. Physiol. 245 (Cell Physiol. 14): C297-C37, EGGENA, P. Osmotic regulation of toad bladder responsiveness to neurohypophyseal hormones. J. Gen. Physiol. 6: , EGGENA, P. Glutaraldehyde - fixation method for determining the permeability to water of the toad urinary bladder. Endocrinology 91: , ELLIS, S. J., W. A. KACHADORIAN, AND V. A. DISCALA. Effect of an osmotic gradient on ADH-induced intramembranous particle aggregates in toad bladder. J. 1Membr. Biol. 52: , HARDY, M. A., AND D. R. DIBONA. Microfilaments and the hydrosmotic action of vasopressin in toad urinary bladder. Am. J. Physiol. 243 (Cell Physiol. 12): c2--c24, HARDY, M. A., AND D. R. DIBONA. Extracellular Ca++ and the effect of antidiuretic hormone on the water permeability of the toad urinary bladder: an example of flow-induced alteration of flow. J. Membr. Biol. 67: 27-44, HARMANCI, M. C., P. STERN, W. A. KACHADORIAN, H. VALTIN, AND V. A. DISCALA. Vasopressin and collecting duct intramembranous particle clusters: a dose-response relationship. Am. J. Physiol. 239 (Renal Fluid Electrolyte Physiol. 8): F56-F564, HARRIS, H. W. JR., J. B. WADE, AND J. S. HANDLER. Transepithelial water flow regulates apical membrane retrieval in antidiuretic hormone-stimulated toad urinary bladder. J. Clin. Invest. 78:73-712, KACHADORIAN, W. A., S. A. ELLIS, AND J. MULLER. Possible roles for microtubules and microfilaments in ADH action on toad urinary bladder. Am. J. Physiol. 236 (Renal Fluid Electrolyte Physiol. 5): F14-F2, KACHADORIAN, W. A., AND S. D. LEVINE. Effect of distension on ADH-induced osmotic water flow in toad urinary bladder. J. Membr. Biol. 64: , KACHADORIAN, W. A., S. D. LEVINE, J. B. WADE, V. A, DISCALA, AND R. M. HAYS. Relationship of aggregated intramembranous particles to water permeability in vasopressin-treated toad urinary bladder. J. Clin. Invest. 59: , , KACHADORIAN, W. A., J. B. WADE, C. C. UITERWYK, AND V. A. DISCALA. Membrane structural and functional responses to vasopressin in toad bladder. J. Membr. Biol. 3: , LEVINE, S. D., M. JACOBY, AND A. FINKELSTEIN. The water permeability of toad urinary bladder. II. The value of PflPd (w) for the antidiuretic hormone-induued water permeation pathway. J. Gen. Physiol. 83: , LEVINE, S. D., AND W. A. KACHADORIAN. Barriers to water flow in vasopressin-treated toad urinary bladder. J. Membr. Biol. 61: ,198l. 18. MULLER, J., AND W. A. KACHADORIAN. Aggregate-carrying membranes during ADH stimulation and washout in toad bladder. Am. J. Physiol. 247 (Cell Physiol. 16): C9-C98, MULLER, J., W. A. KACHADORIAN, AND V. A. DISCALA. Evidence that ADH-stimulated intramembrane particle aggregates are transferred from cytoplasmic to luminal membranes in toad bladder epithelial cells. J. Cell Biol. 85: 83-95, PALMER, L. G., AND M. LORENZEN. Antiduretic hormone-dependent membrane capacitance and water permeability in the toad urinary bladder. Am. J. Physiol. 244 (Renal Fluid Electrolyte Physiol. 13): F195-F24, PARISI, M., M. PISAM, J. MEROT, J. CHEVALIER, AND J. BOUR- GUET. The role of microtubules and microfilaments in the hydrosmotic response to antidiuretic hormone. Biochim. Biophys. Acta 817: ,1985.

5 c53 MEMBRANE RESPONSES TO ADH 22. PEARL, M., AND A. TAYLOR. Actin filaments and vasopressin- 24. TAYLOR, A., M. MAMELAK, E. REAVEN, AND R. MAFFLY. Vasostimulated water flow in toad urinary bladder. Am. J. Physiol. 245 pressin: possible role of microtubules and microfilaments in its (CeZZ Physiol. 14): C28-C39, action. Science Wash. DC 181: , STETSON, D. L., S. A. LEWIS, W. ALLES, AND J. B. WADE. 25. WEIBEL, E. R. AND R. P. BOLENDER. Stereological technique for Evaluation by capacitance measurements of antidiuretic hormone electron microscopic morphometry. In: Principles and Techniques induced membrane area changes in toad bladder. Biochim. Biophys. of Electron Microscopy, edited by M. A. Hayat. New York: Van Acta 689: ,1982. Nostrand Reinhold, 1973, vol. 3, p