- Email: [email protected]
Nongenomic effects of aldosterone on Ca2+ in M-1 cortical collecting duct cells
Kidney International, Vol. 57 (2000), pp. 1395–1403
Nongenomic effects of aldosterone on Ca2⫹ in M-1 cortical collecting duct cells BRIAN J. HARVEY and MARIA HIGGINS Wellcome Trust Cellular Physiology Research Unit, Department of Physiology, University College, Cork, Ireland
Nongenomic effects of aldosterone on Ca2ⴙ in M-1 cortical collecting duct cells. Background. Aldosterone at physiological levels induces rapid (⬍5 min) increases in intracellular protein kinase C (PKC) activity and a rise in calcium and pH in mineralocorticoid hormone target epithelia, such as distal colon and sweat gland. The end targets of these rapid responses in epithelia are Na⫹/H⫹ exchange and K⫹ channels. Methods. The mouse cortical collecting duct (CCD) M-1 cell line was grown to confluency and loaded with Fura-2 for spectrofluorescence measurements of intracellular free calcium at 37⬚C bathed in Krebs solution. Results. Aldosterone (1 nmol/L) produced a rapid, transient peak increase in [Ca2⫹]i in M-1 cells. This effect was abolished upon removal of extracellular Ca2⫹, but was unaffected by pretreatment with spironolactone (10 mol/L) or actinomycin D (10 mol/L). However, pretreatment with the specific PKC inhibitor chelerythrine chloride (1 mol/L) prevented the aldosterone-induced rise in [Ca2⫹]i. Dexamethasone, at a concentration 10,000-fold higher than aldosterone (10 mol/L), also produced a transient increase in [Ca2⫹]i, but this response was significantly smaller than that of aldosterone. In contrast, hydrocortisone had no effect on [Ca2⫹]i at either nmol/L or mol/L concentrations. Both of the sex steroids, 17-estradiol (10 nmol/L) and progesterone (10 nmol/L), induced protein kinase C-dependent increases in [Ca2⫹]i. Conclusions. Aldosterone and sex steroid hormones activate intracellular calcium signaling in CCD cells via a nongenomic PKC-dependent pathway, which may have important implications for renal transport.
The mineralocorticoid hormone aldosterone plays an important role in body fluid, electrolyte, and pH homeostasis. Congenital hyposecretion of aldosterone is incompatible with life, while hypersecretion is associated with hypertension, hypokalemia, and alkalosis. Abnormalities of the renin-angiotensin-aldosterone system are associated with hypertension and cardiac failure. The distal nephron is an important target site for aldosterone. In the cortical collecting duct (CCD), aldosterone promotes electrogenic salt reabsorption by a numKey words: Sex steroids, nongenomic pathway, M-1 cells, CCD cells, calcium signaling, protein kinase C, renal transport.
2000 by the International Society of Nephrology
ber of mechanisms, which target all of the component membrane transport steps: apical Na⫹ channel, basolateral Na⫹/K⫹ pump, and K⫹ channels [1]. These effects are considered to be latent (onset 1 to 2 h) and involve binding to an intracellular type I mineralocorticoid receptor with subsequent induction of protein synthesis. The recently discovered nongenomic mechanism for aldosterone action has created much interest. This response contrasts greatly in its time course (seconds to minutes) and signaling mechanisms to that of the genomic regulation of epithelial transport, which occurs over a period of hours to days. The nongenomic response has been identified in classic aldosterone-responsive epithelia such as distal colon [2], sweat gland [3], and amphibian skin [4, 5]. Although the aldosterone signal transduction pathways (genomic and nongenomic) are vague, paradoxically, our knowledge of fast aldosterone signaling in nonrenal cells is at a relatively advanced level [6]. Multiple intracellular signaling pathways for aldosterone exist in vascular smooth muscle cells, endothelial cells, and leukocytes, which involve Ca2⫹, protein kinase C (PKC), and inositol triphosphate (IP3) [7, 8]. The receptors and transport targets may also be multiple, some involving IP3 and Na⫹/H⫹ exchange. Aldosterone may bind to a specific receptor located close to or on the cell membrane [9, 10], but this novel receptor remains as yet unidentified. In aldosterone-responsive epithelia such as human and rat distal colon, aldosterone has recently been shown to act via a mechanism that does not require activation of the mineralocorticoid type I receptor (MR) and to activate intracellular signaling pathways such as those involving calcium [11] and PKC [12, 13]. The end target of the nongenomic effects of aldosterone in the distal colon is basolateral membrane ion transporters, particularly Na⫹/H⫹ exchange [14, 15] and K⫹ channels [2]. The physiological response is to shift the balance toward net absorption in a pluripotential epithelium by stimulating a class of K-ATP channels involved in sodium absorption while inhibiting the activity of calcium-dependent K⫹ channels necessary for chloride secretion [2, 14, 16].
1395
1396
Harvey and Higgins: Aldosterone in M-1 cells
In this study, we report a novel, nonclassic pathway for aldosterone action in the mouse CCD principal cell line, which does not express the MR receptor. Aldosterone, at physiologic concentrations, induced a transient rise in intracellular calcium after a latent period of one to two minutes. This response appears to be specific for mineralocorticoids, since the glucocorticoid hormone dexamethasone required a 10,000-fold higher concentration than that of aldosterone to achieve a similar response. Hydrocortisone, which also acts on the MR, did not produce a significant increase in intracellular calcium. The response to aldosterone does not require the induction of new protein synthesis, as is observed in the genomic pathway, because it is unaffected by preincubation with actinomycin D, an inhibitor of transcription processes. In turn, the addition of the specific type I mineralocorticoid receptor antagonist, spironolactone, did not affect the rise in intracellular calcium, suggesting the involvement of an unidentified receptor molecule. PKC plays a pivotal role in the activation of this calcium signaling pathway, in turn affecting basolateral Na⫹/H⫹ exchange. Preincubation of M-1 cells with the specific protein kinase inhibitor chelerythrine chloride completely abolished the response to aldosterone. This calcium response may be induced by activation of a PKClinked, aldosterone-specific receptor or a PKC-dependent calcium entry pathway. Finally, in agreement with the recently acknowledged role of sex steroids on salt and water reabsorption and their possible nongenomic mechanism of action, we examined the effects of both estradiol and progesterone on intracellular calcium in M-1 cells. Both of these steroids produced significant increases in intracellular calcium concentrations. In the absence of extracellular calcium, the responses to both estradiol and progesterone were abolished, indicating a requirement for calcium entry in these responses. Preincubation with the protein kinase inhibitors, chelerythrine chloride and bisindolylylmaleimide, completely inhibited the calcium responses, implicating PKC as a mediator in the nongenomic effects of sex steroid hormones in the CCD. Since the effects of aldosterone and the sex steroids on a major intracellular regulator of CCD ion transport are apparent at physiological concentrations of these hormones, these nongenomic responses have important implications for the reassessment of steroid hormonal regulation of renal transport. METHODS Tissue culture M-1 cells were purchased from the American Type Culture Collection (ATCC No. CRL-2038; Batch F-11321). The M-1 cell line is derived from CCD microdissected from a mouse transgenic for the early region of SV40
[strain Tg(SV40E)Bri7]. M-1 cells were grown in monolayers on permeable supports, and exhibited a high transepithelial resistance and developed a lumen negative potential difference. The associated short-circuit current (Isc) was amiloride sensitive. Cultured cells responded to AVP with a significant increase in Isc. The M-1 cell line displays the properties of the native tissue. making it a useful model to study CCD function. Moreover, M-1 cells, like amphibian A6 cells, do not express a functional mineralocorticoid receptor, making them an ideal model to explore nonclassic, nongenomic rapid responses to physiological concentrations of aldosterone, which do not activate the glucocorticoid receptor. The cells were grown in 25 or 75 cm2 polystyrene culture flasks (Gruinn Technology, Dublin, Ireland) containing a 1:1 mixture of Ham’s F-12 medium and Dulbecco’s modified Eagles medium (Flow Laboratories, Irvine, Scotland, UK) containing 5 mol/L dexamethasone (95%) and fetal bovine serum (5%). This medium also contained 1% HEPES and 0.5% penicillin/streptomycin and was maintained at pH 7.4. Cells were equilibrated with 5% CO2/95% O2 and kept at 37⬚C. Fluid was renewed two to three times weekly. Subconfluent cultures were subcultured by rinsing with 0.25% trypsin-0.03% ethylenediaminetetraacetic acid (EDTA) solution. Most of the trypsin solution was removed, and cells were incubated at 37⬚C for 15 to 20 minutes until cells detached. The cell suspension was centrifuged at 1500 r.p.m. for 10 minutes, and the supernatant was carefully removed. Three to 4 mL of fresh medium were added and aspirated, and cells were dispensed into new flasks. The subcultivation ratio was 1:3 to 1:4. Following aspiration of the supernatant, the cell pellet was resuspended at approximately 2 ⫻ 106 cells per mL of freezing medium containing 95% culture medium and 5% dimethyl sulfoxide (DMSO). One milliliter of the suspension was added per freezing vial. Vials were placed in storage and lowered into liquid nitrogen over a threehour period to achieve a gradual freezing process. To revive cells from prolonged storage, the cells were thawed rapidly in a waterbath at 37⬚C. The cell suspension was centrifuged at 1500 r.p.m. for 10 minutes, and the supernatant was removed. The cell pellets were resuspended with 1 mL of culture medium and transferred to culture flasks. Fresh culture medium was added, and flasks were incubated at 37⬚C. The cells studied experimentally were grown to confluent monolayers with typical dome formation. Experiments were performed on single isolated cells or on monolayers, between passages 15 and 30. M-1 cell monolayers were bathed in normal Krebs solution of composition (in mmol/L): NaCl 140, KCl 5, CaCl2 1, MgCl2 1, glucose 5, HEPES 10, at a pH 7.4. Spectrofluorescence imaging Intracellular calcium activity was measured in M-1 cells loaded with Fura-2 using a spectrofluorescence
Harvey and Higgins: Aldosterone in M-1 cells
video microscopy (Imstar, Paris, France) as described in detail previously [11]. The equation that relates the measured Ca2⫹-bound/Ca2⫹-free fluorescence ratio (R) to [Ca2⫹]i is as follows: [Ca2⫹]i ⫽ Kd ⫻ (R ⫺ Rmin)/(Rmax ⫺ R) ⫻ Sf2/Sb2 where Kd is the product of the dissociation constant of the Ca2⫹:Fura-2 complex and a constant related to the optical characteristics of the system; R is the ratio F340/ F380 of the fluorescence signals measured at 340 nm and 380 nm; Rmin and Rmax are the limiting values of R in the presence of zero and saturating Ca2⫹ concentrations; and Sf2/Sb2 is the ratio of fluorescence values for Ca2⫹bound/Ca2⫹-free indicator measured at the wavelengths used to monitor the Ca2⫹-free indicator (the denominator wavelength of R). Intracellular Fura-2 signals were calibrated after each experimental protocol by imaging cells loaded with the ester derivative of the probe under permeabilized conditions (exposure to 2 mol/L ionomycin). Rmax was measured at an external concentration of 10 mmol/L CaCl2 and Rmin in calcium-free solutions buffered in 2 mmol/L EGTA. Drugs All drugs used in experiments were purchased from Sigma Chemical Co. Ltd. (Fancy Road, Poole, Dorset, UK). In all cases, stock solutions were made up in Krebs solution when water soluble. Otherwise, stock solutions were prepared in an methanol or DMSO, and serial dilutions were made such that the final concentrations of methanol or DMSO were less than 0.1% without any significant effect on results. Data analysis All data values are expressed as mean ⫾ SEM. A minimum of six experiments was performed for each experimental protocol. In experiments in which inhibition of a particular pathway was being investigated, cells treated with inhibitor were matched with an untreated control. Significance was calculated using the Students unpaired t-test. RESULTS Source of [Ca2ⴙ]i in M-1 cells Evidence for a luminal Ca2⫹ entry pathway in M-1 cells. To examine the effects of external calcium on [Ca2⫹]i, M-1 monolayers were first incubated for 10 minutes in a nominally Ca2⫹-free medium on the luminal side, with calcium present in the basolateral solution. A subsequent addition of 1 mmol/L CaCl2 to the luminal side produced a significant yet transient increase in intracellular calcium ([Ca2⫹]i changed from 29 ⫾ 9 to 189 ⫾ 66 nmol/L; ⌬[Ca2⫹]i ⫽ 160 ⫾ 64 nmol/L, N ⫽ 5, P ⬍ 0.05). This increase in [Ca2⫹]i was inhibited when verapamil (10 mol/L)
1397
was present in the luminal bathing solution. Removal of calcium from the basolateral side alone did not affect [Ca2⫹]i over this time period. The influence of intracellular calcium stores on [Ca2⫹]i, was examined using the endoplasmic Ca2⫹-ATPase inhibitors thapsigargin (TG) and 2,5-di-(tert-butyl)-1,4benzohydroquinone (TBQ). The addition of TG, at a concentration of 1 mol/L, resulted in a significant increase in [Ca2⫹]i from 110 ⫾ 22 to 388 ⫾ 31 nmol/L, representing a change in [Ca2⫹]i of 290 ⫾ 46 nmol/L (N ⫽ 9, P ⬍ 0.0001). Surprisingly, the addition of TBQ, at a concentration of 1 mol/L, failed to increase [Ca2⫹]i significantly, which showed an apparent change from 70 ⫾ 16 to 83 ⫾ 19 nmol/L, ⌬[Ca2⫹]i ⫽ 14 ⫾ 5 nmol/L (N ⫽ 10, P ⫽ 0.59). These data indicate that calcium can be mobilized from IP3-sensitive calcium pools in M-1 cells; however, the less selective Ca2⫹-pump antagonist TBQ does not appear to affect calcium mobilization from this pool over a range of concentrations. Effects of Na⫹ transport inhibition on [Ca2⫹]i. The effect of sodium transport inhibitors on intracellular calcium was tested in polarized M-1 cell monolayers. The removal of extracellular sodium on the basolateral side produced a rapid increase in calcium (from 230 ⫾ 20 to 780 ⫾ 80 nmol/L, N ⫽ 6). This effect may be due to a reversal of the Na⫹/Ca2⫹ exchange. The inhibition of Na⫹/K⫹-ATPase pump with ouabain failed to produce any change in intracellular calcium, indicating that small changes in the transmembrane Na⫹ gradient arising from pump inhibition are insufficient to influence the activity of sodium-coupled calcium transporters. The addition of amiloride (1 mol/L) to the apical side of an M-1 cell confluent monolayer produced a sustained decrease in intracellular calcium (from 210 ⫾ 30 to 75 ⫾ 18 nmol/L, N ⫽ 6). A possible mechanism for this effect is an increase in basolateral Na⫹/Ca2⫹ exchange activity following a reduction in intracellular sodium arising from inhibition of the luminal sodium channel. The addition of amiloride to the Krebs solution bathing the basolateral side of M-1 cell monolayers produced an opposite effect on [Ca2⫹]i, as was seen for apical addition of the diuretic. Under these conditions, basolateral amiloride produced an increase in from 180 ⫾ 32 to 720 ⫾ 50 nmol/L (N ⫽ 6). At a concentration of 1 mol/L, amiloride is also a blocker of Na⫹/Ca2⫹ exchange, and these nonspecific effects of the diuretic could explain the rise in calcium. Aldosterone action on [Ca2⫹]i. Aldosterone, at a concentration of 1 nmol/L, produced a rapid, transient peak increase in [Ca2⫹]i from 132 ⫾ 25 to 499 ⫾ 82 nmol/L (mean increase in [Ca2⫹]i ⫽ 367 ⫾ 75 nmol/L, N ⫽ 9, P ⬍ 0.001) in M-1 cell monolayers. This dose of aldosterone was sufficient to produce a maximal increase in [Ca2⫹]i. The mean half-maximal response was produced at 0.1 nmol/L aldosterone, and the minimum concentration of hormone required to produce a significant rise
1398
Harvey and Higgins: Aldosterone in M-1 cells
Fig. 1. Representative Fura-2 spectrofluorescence recording of the increase in intracellular calcium stimulated by 1 nmol/L aldosterone (Aldo) in confluent M-1 cell monolayers (䊉). This increase was completely inhibited by pretreatment (15 min) with the protein kinase C inhibitor chelerythrine chloride (䊏).
in [Ca2⫹]i was 0.08 nmol/L. A typical experiment is shown in Figure 1. The aldosterone response was characterized by a latency period of one to two minutes, which may be indicative of receptor binding or a response that occurs secondarily to a primary signal, such as PKC. The aldosterone effect was abolished under nominally Ca2⫹-free conditions only when calcium was removed from the luminal side. Under the latter conditions, aldosterone produced a nonsignificant change in baseline [Ca2⫹]i levels from 90 ⫾ 7 to 114 ⫾ 10 nmol/L, representing a mean change in [Ca2⫹]i of 24 ⫾ 4 nmol/L (N ⫽ 6, P ⬍ 0.01). The removal of calcium from the basolateral side did not significantly attenuate the [Ca2⫹]i response to hormone. Pretreatment of M-1 cells with TG (10 mol/L) caused a 75% reduced calcium response to aldosterone (N ⫽ 6), indicating that the response requires release from an intracellular stores. Since exposure to an external calcium-free solution for one second prior to aldosterone abolished the response, the data indicate that aldosterone-induced calcium release from intracellular stores is indirect via a calcium entry-induced calcium release mechanism. Effects of actinomycin D and spironolactone. In the presence of actinomycin D (10 mol/L), which inhibits gene transcription, aldosterone also induced a calcium transient (Fig. 2) from a baseline [Ca2⫹]i of 104 ⫾ 31 to 419 ⫾ 44 nmol/L (⌬[Ca2⫹]i ⫽ 315 ⫾ 56 nmol/L, N ⫽ 5, P ⬍ 0.0005). This increase was not significantly different from the aldosterone control response (P ⬍ 0.643). Actino-
Fig. 2. Effect of aldosterone (1 nmol/L) on Ca2ⴙ in M-1 cells in the presence of inhibitors of the genomic pathway, actinomycin D and spironolactone, and the protein kinase C (PKC) inhibitor chelerythrine chloride. Note that the PKC activator, phorbol myristyl acetate (PMA), produced a similar Ca2⫹ response to aldosterone.
mycin D did not affect intracellular calcium levels. The rapid increase in [Ca2⫹]i induced by aldosterone was also unaffected by pretreatment with the specific mineralocorticoid type I receptor antagonist spironolactone (10 mol/L). Aldosterone in the presence of spironolactone induced a rise in [Ca2⫹]i from 115 ⫾ 20 to 332 ⫾ 67 nmol/L (⌬[Ca2⫹]i ⫽ 217 ⫾ 54 nmol/L, N ⫽ 8, P ⬍ 0.001). This increase in [Ca2⫹]i was not significantly different from the aldosterone control response (P ⬍ 0.133). Effect of protein kinase C inhibition. Pretreatment with the specific PKC inhibitor chelerythrine chloride (1 mol/L) completely abolished the aldosterone-induced rise in [Ca2⫹]i (Fig. 1). In the presence of chelerythrine chloride, aldosterone addition resulted in an increase in [Ca2⫹]i from 130 ⫾ 13 to 144 ⫾ 11 nmol/L (⌬[Ca2⫹]i ⫽ 14 ⫾ 11 nmol/L, N ⫽ 6, P ⫽ 0.375). The aldosterone response in the presence of the PKC inhibitor was significantly different from control responses (P ⬍ 0.005). These data indicate that aldosterone increased intracellular calcium in M-1 cells by first activating PKC. In agreement with this conclusion, direct activation of PKC with the phorbol ester phorbol myristyl acetate (PMA;
Harvey and Higgins: Aldosterone in M-1 cells
Fig. 3. Steroid hormone effects on intracellular calcium in confluent M-1 cells. Aldosterone (1 nmol/L) induced a rapid increase in [Ca2⫹]i that was significantly different from that produced by dexamethasone (10 mol/L), 17-estradiol (10 nmol/L), or progesterone (10 nmol/L). Hydrocortisone failed to increase [Ca2⫹]i significantly even at a concentration of 10 mol/L.
1399
Sex steroids. The estrogen analogue 17-estradiol at a concentration of 10 nmol/L raised [Ca2⫹]i significantly from a baseline level of 120 ⫾ 12 to 320 ⫾ 53 nmol/L (⌬[Ca2⫹]i ⫽ 200 ⫾ 44, N ⫽ 7, P ⬍ 0.005). Another female sex steroid, progesterone (10 nmol/L), also generated a significant increase in [Ca2⫹]i from 114 ⫾ 24 to 332 ⫾ 42 nmol/L (⌬[Ca2⫹]i ⫽ 188 ⫾ 23, N ⫽ 4, P ⬍ 0.01). Estradiol and progesterone failed to produce a significant rise in [Ca2⫹]i under nominally Ca2⫹-free conditions. In the presence of zero external calcium, estradiol addition resulted in a slight increase in [Ca2⫹]i from 50 ⫾ 8 to 69 ⫾ 13 nmol/L (⌬[Ca2⫹]i ⫽ 19 ⫾ 6, N ⫽ 4, P ⫽ 0.27), whereas progesterone addition produced an insignificant increase in [Ca2⫹]i from 59 ⫾ 7 to 62 ⫾ 6 nmol/L (⌬[Ca2⫹]i ⫽ 6 ⫾ 2, N ⫽ 8, P ⫽ 0.69). Responses to 17-estradiol and progesterone were prevented by pretreatment (15 min) with the PKC inhibitor chelerythrine chloride (1 mol/L). Estradiol in the presence of chelerythrine chloride induced an increase in [Ca2⫹]i from 112 ⫾ 11 to 121 ⫾ 10 nmol/L (⌬[Ca2⫹]i ⫽ 14 ⫾ 2, N ⫽ 8, P ⫽ 0.518). Similarly, progesterone in the presence of chelerythrine chloride induced an insignificant increase in [Ca2⫹]i from 93 ⫾ 11 to 101 ⫾ 10 nmol/L (⌬[Ca2⫹]i ⫽ 8 ⫾ 2, N ⫽ 8, P ⫽ 0.63). Therefore, like aldosterone, the Ca2⫹ signal transduction pathway for estradiol and progesterone appears to involve PKC. DISCUSSION
100 nmol/L) induced an increase in [Ca2⫹]i from 137 ⫾ 33 to 407 ⫾ 67 nmol/L (⌬[Ca2⫹]i ⫽ 269 ⫾ 44 nmol/L, N ⫽ 7, P ⬍ 0.005). The increases in intracellular calcium following the addition of aldosterone, in the presence or absence of specific inhibitors, are summarized in Figure 2. Inhibitors of the classic genomic pathway for steroid action have little or no effect on the aldosterone response, indicating a nonclassic pathway. PKC inhibition abolishes the response, and activation of the PKC pathway is capable of increasing [Ca2⫹]i in the M-1 cell line, similar to that observed with aldosterone. Rapid Ca2⫹ responses to other steroid hormones. The effects of other salt-retaining steroid hormones on intracellular Ca2⫹ were compared with the aldosterone response (Fig. 3). Glucocorticoids. Dexamethasone (10 mol/L) induced a transient increase in [Ca2⫹]i that was smaller in magnitude than the response observed following aldosterone addition. Dexamethasone increased [Ca2⫹]i from 27 ⫾ 9 to 211 ⫾ 57 nmol/L (⌬[Ca2⫹]i ⫽ 185 ⫾ 55, N ⫽ 4, P ⬍ 0.05). Another glucocorticoid hormone, hydrocortisone, had no significant effect on [Ca2⫹]i, even at micromolar concentrations. Hydrocortisone changed [Ca2⫹]i from 89 ⫾ 10 to 108 ⫾ 6 nmol/L (⌬[Ca2⫹]i ⫽ 20 ⫾ 7, N ⫽ 6, P ⫽ 0.14).
Aldosterone is the most important mineralocorticoid released from the adrenal glands. It regulates solute reabsorption, acting in a delayed manner to influence ion transport across tight junction epithelia such as distal nephron and colon. The classic mechanism of aldosterone stimulation of Na⫹ reabsorption involves the activation of a genomic pathway. It is well documented that aldosterone binds to a cytosolic MR, which is spironolactone sensitive. The active hormone-receptor complex translocates to the nucleus, where it binds to hormoneresponsive elements in the genome and initiates the production of mRNA, which is later translated into new proteins, termed aldosterone-induced proteins (AIPs). These proteins are released into the cytoplasm, where they may be relocated to the apical border and act to alter membrane permeability. Several theories exist as to how aldosterone increases apical Na⫹ reabsorption, and some have proposed that the AIPs themselves are newly synthesized Na⫹ channels, while others argue that these proteins act at a site close to the channel protein and induce a conformational change of the channels [17]. In recent years, a new model for aldosterone action in target epithelia, which introduces a novel specificityconferring mechanism for mineralocorticoid action on epithelial ion transport via protein kinase and calcium signaling has been postulated [2]. In this system of rapid,
1400
Harvey and Higgins: Aldosterone in M-1 cells
Fig. 4. Cortical collecting duct (CCD) cell model of signals involved in the nongenomic pathway for steroid hormone action. Aldosterone binds to cytosolic or membrane-associated elements that may include PKC or a spironolactone-insensitive “receptor” to induce a nongenomic rise in intracellular calcium via a PKC-sensitive luminal entry of calcium and calcium release from intracellular stores. The calcium-kinase complex activates basolateral Na/H exchange, and the resultant intracellular alkalinization activates basolateral K⫹ channels. The increased K⫹ conductance primes the cell for the later genomic up-regulation of sodium absorption.
nongenomic steroid action, aldosterone selectively targets basolateral Na⫹/H⫹ exchange and K⫹ channels (Fig. 4). In the distal colon, for example, aldosterone rapidly up-regulates a K-ATP class channel necessary for sodium absorption while down-regulating a K-Ca channel required for chloride secretion [2, 16]. Thus, the physiological role of the rapid nongenomic response to aldosterone is to shift a pluripotential epithelium into a net absorptive transport mode. These rapid nongenomic actions are unaffected by inhibitors of the type I MR or of protein synthesis (nongenomic) and show a high selectivity for aldosterone over cortisol. This model offers an additional explanation for the clinical evidence for differential effects of glucocorticoids and mineralocorticoids in Na⫹ homeostasis. There is increasing evidence for a rapid action of aldosterone on renal Na⫹/H⫹ exchange. In amphibian renal tubule, aldosterone produced a spironolactone-sensitive activation of Na/H exchange [18], which reflects an alkaline shift in the exchanger’s set point of activation [19], but these effects may be part of the classic genomic effect requiring new protein synthesis [20]. In Madin Darby canine kidney cells, a rapid response to aldosterone on pHi regulation and Na⫹/H⫹ exchange has been reported [21], which appears to be secondarily dependent on a rapid nongenomic action of aldosterone to stimulate a plasma membrane proton conductance [22]. In human and rat distal colon and human sweat gland, aldosterone produces a rapid (within ⬍5 sec) activation of basolateral N/H exchange [3, 15]. This activation appears to be the result of pleiotropic convergent signaling pathways in-
volving Ca2⫹, PKC-␣, arachidonic acid metabolism, and a pertussis-sensitive G-protein activation [15]. If the rapid effects of aldosterone on Na/H exchange are prevented, then the hormone fails to activate the K-ATP channel or inhibit the K-Ca channel [2]. These K⫹ channels show opposite pH sensitivity [2, 23], and it is possible that an intracellular alkalinization alone can explain the different effects of aldosterone on these channels in different cell types in a pluripotential epithelium containing absorptive (surface) cells and secretory (crypt) cells. A direct effect of the protein kinase on channel opening may also be involved in the early response. A rapid calcium transient is a common feature of the nongenomic aldosterone response in colon and sweat gland. In this study, we have shown evidence for a luminal calcium entry pathway in mouse collecting duct M-1 cells, which is stimulated by aldosterone. The reintroduction of calcium into a previous calcium-free bathing solution caused a rapid rise in Ca2⫹i. Also, these cells have the capacity to mobilize calcium from intracellular stores, most likely an IP3-sensitive pool. Inhibitors of sodium reabsorption were also shown to affect intracellular calcium. The removal of external sodium from the basolateral side produced a transient rise in calcium, possibly mediated by reversal of a Na⫹/Ca2⫹ exchanger, whereas removal of sodium from the luminal side caused a fall in intracellular calcium. Likewise, the sodium transport inhibitor amiloride had the opposite effects on calcium, depending on its side of addition. Exposure of monolayers to apical amiloride caused a rapid and sustained fall in calcium. This effect may be due to an increased
Harvey and Higgins: Aldosterone in M-1 cells
driving force for Na⫹/Ca2⫹ exchange following a reduction in intracellular sodium concentration (as a result of inhibition of Na⫹ uptake through ENaC). When the diuretic was added on the basolateral side, however, a transient increase in calcium was observed. Amiloride is also known to inhibit Na⫹/Ca2⫹ exchange at micromolar concentrations, and this latter effect may account for the rise in calcium observed where the basolateral membrane is exposed to the diuretic. The relatively fast action of aldosterone on intracellular Ca2⫹ signaling in mouse CCD cells is incompatible with the classic view of mineralocorticoid hormone action in renal epithelia for the following reasons: (1) Aldosterone produced a rapid transient peak increase in [Ca2⫹]i. (2) The response was abolished in the absence of Ca2⫹ in the extracellular solution, indicating that the rise in [Ca2⫹]i requires entry of Ca2⫹ from the external environment. (3) Pretreatment with spironolactone or actinomycin D failed to abolish the aldosterone effect, indicating a nonclassic pathway. (4) Pretreatment with chelerythrine chloride, a PKC inhibitor, completely inhibited the induction of the calcium response, providing evidence that PKC is involved in this nonclassic increase in [Ca2⫹]i. (5) The phorbol ester PMA induced a calcium transient of similar magnitude to aldosterone, indicating that PKC can mediate fast increases in [Ca2⫹]i. The rapid effects of aldosterone to increase cytosolic calcium are difficult to reconcile with previous reports suggesting that chronic increases in [Ca2⫹]i and PKC down-regulate apical Na⫹ channels [(abstract; Palmer et al, Fed Proc 46:495, 1987); 23–25]. However, there is evidence for a role of Ca2⫹ in mediating the latent genomic aldosterone response of enhanced sodium absorption in renal A6 cells (abstract; Petzel et al, Kidney Int 37:358, 1990). The temporal coincidence of the increases in Ca2⫹, IP3, and Isc and the apparent dependence of both the Ca2⫹ transient and Isc on transcriptional and translational processes would seem to indicate a causal relationship [26]. In tight epithelia in which the rapid effects of aldosterone have been identified, the transport target appears to be the Na⫹/H⫹ exchanger and K⫹ channel activity [2, 14, 18, 19]. Thus, aldosterone may have a rapid action to enhance pHi regulation and affects sodium absorption indirectly via modulation of pHi-sensitive transport pathways (Fig. 4). In the case of stimulation of Na⫹/H⫹ exchange, the resultant intracellular alkalinization can up-regulate K⫹ conductance [14]. An enhanced K⫹ conductance would serve to limit membrane depolarization resulting from chronic aldosterone effects on Na⫹ reabsorption. Thus, the nongenomic phase of aldosterone on Ca2⫹/ PKC, Na⫹/H⫹ exchange, and K⫹ conductance may serve to prime the CCD to maintain membrane potential close to the K⫹ equilibrium potential and ensure a favorable electrochemical membrane potential driving force for genomic effects on Na⫹ uptake (Fig. 4). An enhanced
1401
K⫹ conductance would serve to limit membrane depolarization resulting from chronic aldosterone effects on Na⫹ reabsorption [23]. Protein kinase C may play a role in rapid activation of Na⫹/H⫹ exchange by aldosterone. Serine kinases are important regulators of Na⫹/H⫹ exchange, and protein kinases are involved in activating Na⫹/H⫹ exchange by prostaglandins in renal proximal cells [27, 28]. Shortterm activation of PKC leads to a transient increase in Na⫹/H⫹ exchange activity independent of transcription and translation, whereas long-term activation causes a persistent increase in activity that is dependent on these processes and associated with increased mRNANa/H. This effect may mediate increased Na⫹/H⫹ exchanger activity in chronic conditions such as acidosis or K⫹ deficiency. In the distal colon, the activation of PKC with phorbol esters stimulates Na⫹/H⫹ exchange [15]. Recent work in our laboratory has shown that aldosterone rapidly activates the PKC-␣ isoform in rat colon. In our experiments here, we found that specific inhibition of PKC completely prevents the rapid aldosterone effect on [Ca2⫹]i in M-1 cells. Together, these results suggest that activation of PKC is a key step in the signal transduction of the rapid aldosterone effect in epithelia. The physiological significance of a nongenomic effect of aldosterone is underlined by the low concentrations of hormone required for the rapid calcium response effects, which concur with the physiological concentration of free aldosterone (0.1 nmol/L). Inhibition of aldosterone action is important in the treatment of disturbances in extracellular fluid (ECF) balance, for example, hypertension, heart failure. The type I MR antagonist spironolactone is used pharmacologically to inhibit aldosterone action, but is much less effective at reducing hypertension and ECF volume than angiotensin-converting enzyme (ACE) inhibitors or angiotensin II receptor antagonists. This may be due to the effects of angiotensin II on vascular smooth muscle and salt-absorbing epithelia, but may also be explained by the existence of two classes of aldosterone receptor. There are a number of steps in the rapid nongenomic aldosterone pathway that may serve as targets for pharmacological manipulation. Specific inhibition of aldosterone binding to the unidentified nongenomic receptor may be a means of controlling the rapid aldosterone effect and may be an important target in controlling blood volume and pressure. This could be the basis of the development of a novel mineralocorticoid receptor antagonist with potential new strategies for treatment of human cardiovascular disease. Also, the specific inhibition of PKC isoforms may control nongenomic effects. The rapid estradiol response may involve a similar mechanism to that of aldosterone, independent of protein synthesis and involving PKC. The renal estrogen receptor may be a substrate for PKC, and its properties
1402
Harvey and Higgins: Aldosterone in M-1 cells
can be modified by PKC similarly to the vitamin D receptor. Renal binding of 17-estradiol has been primarily localized to the proximal tubule, while the role of the estradiol receptor in the distal tubule is unknown. Estrogens are allegedly indirectly involved in proximal tubular Ca2⫹ homeostasis via stimulation of 1-␣-hydroxylase, which is responsible for the synthesis of 1,25(OH)2D3, the biologically active form of vitamin D3. In the mouse CCD cells, progesterone and estradiol both increase [Ca2⫹]i, which might explain some of the nongenomic effects of these hormones. The glucocorticoids dexamethasone and hydrocortisone had differing effects on [Ca2⫹]i. Dexamethasone produced a transient increase in [Ca2⫹]i at a concentration 10,000-fold higher than that of aldosterone, whereas hydrocortisone had no effect on [Ca2⫹]i. The response to dexamethasone was statistically different from that of aldosterone. These data indicate that this response is mineralocorticoid specific but does not involve type I receptor activation, since hydrocortisone binds to this receptor at high concentrations. M-1 cells express a functional glucocorticoid receptor but not a mineralocorticoid receptor. Thus, the fast calcium effects of aldosterone at a concentration that does not activate GR and the absence of functional MR in M-1 cells argue strongly in favor of a novel receptor mediating the nongenomic rapid responses to physiological concentrations of aldosterone. CONCLUSIONS AND PERSPECTIVES Our results provide new evidence for a nonclassic, nongenomic mechanism for aldosterone action in CCD, independent of type I mineralocorticoid receptors and involving a PKC-linked, aldosterone-specific receptor or a PKC-dependent Ca2⫹ entry pathway. Furthermore, we have provided evidence that sex steroids also produce rapid increases in [Ca2⫹]i that are PKC dependent and characteristic of recently described nongenomic mechanisms of steroid action. The physiological role of sex steroid targeting to the CCD may be important in determining the degree of fluid retention seen at high circulating concentrations of these hormones. The identity of the putative cyclooxygenase metabolite involved in the nongenomic response remains unknown, and the precise PKC isoform and its phosphorylation targets also remain to be identified. The cellular location of the nongenomic “receptor” is unclear, and although evidence points toward a membrane receptor [9] or PKC itself [12], this has yet to be determined. The rapid (minutes) effects of aldosterone on basolateral Na⫹/H⫹ exchange and K⫹ channels must be reconciled with the delay (hours) before effects on transepithelial sodium absorption become apparent.
ACKNOWLEDGMENTS This work was supported by the Wellcome Trust (programme grant #040067/Z/93) and the Health Research Board of Ireland. Reprint requests to Dr. Brian J. Harvey, Wellcome Trust Cellular Physiology Research Unit, Department of Physiology, University College, Cork, Ireland. E-mail: [email protected]
REFERENCES 1. Rossier BC, Paccolat MP, Verrey F, Kraehenbuhl J-P, Geering K: Mechanism of action of aldosterone: A pleiotropic response. Horm Cell Regul 9:209–225, 1985 2. Maguire D, MacNamara B, Cuffe JE, Winter D, Doolan CM, Urbach V, O’Sullivan GC, Harvey BJ: Rapid responses to aldosterone in human distal colon. Steroids 64:51–63, 1999 3. Hegarty JM, Harvey BJ: Aldosterone accelerates a Na⫹/H⫹exchange dependent pH recovery after acid loading in cultured human eccrine sweat gland epithelial cells. J Physiol (Lond) 517P:20P, 1999 4. Harvey BJ: Energization of sodium absorption by the H(⫹)ATPase pump in mitochondria-rich cells of frog skin. J Exp Biol 172:289–309, 1992 5. Urbach V, Van Kerkhove E, Maguire D, Harvey BJ: Rapid activation of KATP channels by aldosterone in principal cells of frog skin. J Physiol (Lond) 491(Pt 1):111–120, 1996 6. Wehling M: Specific, nongenomic actions of steroid hormones. Annu Rev Physiol 59:365–393, 1997 7. Christ M, Meyer C, Sippel K, Wehling M: Rapid aldosterone signaling in vascular smooth muscle cells: Involvement of phosholipase C, diacylglycerol and protein kinase C. Biochem Biophys Res Commun 213:123–129, 1995 8. Christ M, Eisen C, Aktas J, Theisen K, Wehling M: The inositol1,4,5-trisphosphate system is involved in rapid effects of aldosterone in human mononuclear leukocytes. J Clin Endocrinol 77: 1452–1457, 1993 9. Wehling M, Eisen C, Christ M: Membrane receptors for aldosterone: A new concept of nongenomic mineralocorticoid action. NIPS 8:241–244, 1993 10. Wehling M: Novel aldosterone receptors: Specificity-conferring mechanism at the level of the cell membrane. Steroids 59:160–163, 1994 11. Doolan CM, Harvey BJ: Rapid effects of steroid hormones on free intracellular calcium in T84 colonic epithelial cells. Am J Physiol 271(6 Pt 1):C1935–C1941, 1996 12. Doolan CM, Harvey BJ: Modulation of cytosolic protein kinase C and calcium ion activity by steroid hormones in rat distal colon. J Biol Chem 271:8763–8767, 1996 13. Doolan CM, Harvey BJ: Rapid effects of steroid hormones on cytosolic protein kinase C and calcium ion activity in human distal colon. Mol Cell Endocrinol 138:71–79, 1998 14. Harvey BJ, Ehrenfeld J: Role of Na⫹/H⫹ exchange in the control of intracellular pH and cell membrane conductances in frog skin epithelium. J Gen Physiol 92:793–810, 1988 15. Winter DC, Schneider MF, O’Sullivan GC, Harvey BJ, Geibel JP: Rapid activation of sodium-hydrogen exchange in colonic crypts by aldosterone. J Membr Biol 170:17–26, 1999 16. MacNamara B, Winter DC, Cuffe JE, O’Sullivan GC, Harvey BJ: Basolateral potassium channel involvement in Forskolin (cAMP)-activated secretion in human colon. J Physio (Lond) 519(Pt 1):25–260, 1999 17. Rossier BC, Verrey F, Kraehenbuhl J-P: Transepithelial sodium transport and its control by aldosterone: A molecular approach. Curr Top Membr Transplant 34:167–183, 1989 18. Oberleithner H, Weight M, Westphale J, Wang WH: Aldosterone activates Na⫹/H⫹ exchange and raises cytoplasmic pH in target cells of the amphibian kidney. Proc Natl Acad Sci USA 84:1464–1468, 1987 19. Cooper GJ, Hunter M: Na⫹-H⫹ exchange in frog early distal tubule: Effect of aldosterone on the set-point. J Physiol (Lond) 479(Pt 3):423–432, 1994
Harvey and Higgins: Aldosterone in M-1 cells 20. Cooper GJ, Hunter M: Role of de novo protein synthesis and calmodulin in rapid activation of Na⫹-H⫹ exchange by aldosterone in frog diluting segment. J Physiol (Lond) 491(Pt 1):219–223, 1996 21. Gekle M, Golenhofen N, Oberleithner H, Silbernagl S: Rapid activation of Na⫹/H⫹ exchange by aldosterone in renal epithelial cells requires Ca2⫹ and stimulation of a plasma membrane proton conductance. Proc Natl Acad Sci USA 93:10500–10504, 1996 22. Gekle M, Silbernagl S, Oberleithner H: The mineralocorticoid aldosterone activates a proton conductance in cultured kidney cells. Am J Physiol 273(5 Pt 1):C1673–C1678, 1997 23. Harvey BJ: Cross-talk between sodium and potassium channels in tight epithelia. Kidney Int 48:1191–1199, 1995 24. Palmer LG, Frindt G: Effects of cell Ca and pH on Na channels from rat CCT. Am J Physiol 253:F333–F339, 1987
1403
25. Ling BN, Kokko KE, Eaton DC: Inhibition of apical Na⫹ channels in rabbit cortical collecting tubules by basolateral prostaglandin E2 is modulated by protein kinase C. J Clin Invest 90:1328–1334, 1992 26. Petzel D, Ganz MB, Nestler EJ, Lewis JL, Goldenring J, Akcicek F, Hayslett JP: Correlates of aldosterone-induced increases in Ca2⫹i and Isc suggest that Ca2⫹i is the second messenger for stimulation of apical membrane conductance. J Clin Invest 89:150– 156, 1992 27. Weinman EJ, Shenolikar S: Protein kinase C activates the renal apical membrane Na⫹-H⫹ exchanger. J Membr Biol 93:133–139, 1986 28. Weinman EJ, Dubinsky W, Shenolikar SL: Regulation of the renal Na⫹-H⫹ exchanger by protein phosphorylation. Kidney Int 36:519–525, 1989
Nongenomic effects of aldosterone on Ca2⫹ in M-1 cortical collecting duct cells BRIAN J. HARVEY and MARIA HIGGINS Wellcome Trust Cellular Physiology Research Unit, Department of Physiology, University College, Cork, Ireland
Nongenomic effects of aldosterone on Ca2ⴙ in M-1 cortical collecting duct cells. Background. Aldosterone at physiological levels induces rapid (⬍5 min) increases in intracellular protein kinase C (PKC) activity and a rise in calcium and pH in mineralocorticoid hormone target epithelia, such as distal colon and sweat gland. The end targets of these rapid responses in epithelia are Na⫹/H⫹ exchange and K⫹ channels. Methods. The mouse cortical collecting duct (CCD) M-1 cell line was grown to confluency and loaded with Fura-2 for spectrofluorescence measurements of intracellular free calcium at 37⬚C bathed in Krebs solution. Results. Aldosterone (1 nmol/L) produced a rapid, transient peak increase in [Ca2⫹]i in M-1 cells. This effect was abolished upon removal of extracellular Ca2⫹, but was unaffected by pretreatment with spironolactone (10 mol/L) or actinomycin D (10 mol/L). However, pretreatment with the specific PKC inhibitor chelerythrine chloride (1 mol/L) prevented the aldosterone-induced rise in [Ca2⫹]i. Dexamethasone, at a concentration 10,000-fold higher than aldosterone (10 mol/L), also produced a transient increase in [Ca2⫹]i, but this response was significantly smaller than that of aldosterone. In contrast, hydrocortisone had no effect on [Ca2⫹]i at either nmol/L or mol/L concentrations. Both of the sex steroids, 17-estradiol (10 nmol/L) and progesterone (10 nmol/L), induced protein kinase C-dependent increases in [Ca2⫹]i. Conclusions. Aldosterone and sex steroid hormones activate intracellular calcium signaling in CCD cells via a nongenomic PKC-dependent pathway, which may have important implications for renal transport.
The mineralocorticoid hormone aldosterone plays an important role in body fluid, electrolyte, and pH homeostasis. Congenital hyposecretion of aldosterone is incompatible with life, while hypersecretion is associated with hypertension, hypokalemia, and alkalosis. Abnormalities of the renin-angiotensin-aldosterone system are associated with hypertension and cardiac failure. The distal nephron is an important target site for aldosterone. In the cortical collecting duct (CCD), aldosterone promotes electrogenic salt reabsorption by a numKey words: Sex steroids, nongenomic pathway, M-1 cells, CCD cells, calcium signaling, protein kinase C, renal transport.
2000 by the International Society of Nephrology
ber of mechanisms, which target all of the component membrane transport steps: apical Na⫹ channel, basolateral Na⫹/K⫹ pump, and K⫹ channels [1]. These effects are considered to be latent (onset 1 to 2 h) and involve binding to an intracellular type I mineralocorticoid receptor with subsequent induction of protein synthesis. The recently discovered nongenomic mechanism for aldosterone action has created much interest. This response contrasts greatly in its time course (seconds to minutes) and signaling mechanisms to that of the genomic regulation of epithelial transport, which occurs over a period of hours to days. The nongenomic response has been identified in classic aldosterone-responsive epithelia such as distal colon [2], sweat gland [3], and amphibian skin [4, 5]. Although the aldosterone signal transduction pathways (genomic and nongenomic) are vague, paradoxically, our knowledge of fast aldosterone signaling in nonrenal cells is at a relatively advanced level [6]. Multiple intracellular signaling pathways for aldosterone exist in vascular smooth muscle cells, endothelial cells, and leukocytes, which involve Ca2⫹, protein kinase C (PKC), and inositol triphosphate (IP3) [7, 8]. The receptors and transport targets may also be multiple, some involving IP3 and Na⫹/H⫹ exchange. Aldosterone may bind to a specific receptor located close to or on the cell membrane [9, 10], but this novel receptor remains as yet unidentified. In aldosterone-responsive epithelia such as human and rat distal colon, aldosterone has recently been shown to act via a mechanism that does not require activation of the mineralocorticoid type I receptor (MR) and to activate intracellular signaling pathways such as those involving calcium [11] and PKC [12, 13]. The end target of the nongenomic effects of aldosterone in the distal colon is basolateral membrane ion transporters, particularly Na⫹/H⫹ exchange [14, 15] and K⫹ channels [2]. The physiological response is to shift the balance toward net absorption in a pluripotential epithelium by stimulating a class of K-ATP channels involved in sodium absorption while inhibiting the activity of calcium-dependent K⫹ channels necessary for chloride secretion [2, 14, 16].
1395
1396
Harvey and Higgins: Aldosterone in M-1 cells
In this study, we report a novel, nonclassic pathway for aldosterone action in the mouse CCD principal cell line, which does not express the MR receptor. Aldosterone, at physiologic concentrations, induced a transient rise in intracellular calcium after a latent period of one to two minutes. This response appears to be specific for mineralocorticoids, since the glucocorticoid hormone dexamethasone required a 10,000-fold higher concentration than that of aldosterone to achieve a similar response. Hydrocortisone, which also acts on the MR, did not produce a significant increase in intracellular calcium. The response to aldosterone does not require the induction of new protein synthesis, as is observed in the genomic pathway, because it is unaffected by preincubation with actinomycin D, an inhibitor of transcription processes. In turn, the addition of the specific type I mineralocorticoid receptor antagonist, spironolactone, did not affect the rise in intracellular calcium, suggesting the involvement of an unidentified receptor molecule. PKC plays a pivotal role in the activation of this calcium signaling pathway, in turn affecting basolateral Na⫹/H⫹ exchange. Preincubation of M-1 cells with the specific protein kinase inhibitor chelerythrine chloride completely abolished the response to aldosterone. This calcium response may be induced by activation of a PKClinked, aldosterone-specific receptor or a PKC-dependent calcium entry pathway. Finally, in agreement with the recently acknowledged role of sex steroids on salt and water reabsorption and their possible nongenomic mechanism of action, we examined the effects of both estradiol and progesterone on intracellular calcium in M-1 cells. Both of these steroids produced significant increases in intracellular calcium concentrations. In the absence of extracellular calcium, the responses to both estradiol and progesterone were abolished, indicating a requirement for calcium entry in these responses. Preincubation with the protein kinase inhibitors, chelerythrine chloride and bisindolylylmaleimide, completely inhibited the calcium responses, implicating PKC as a mediator in the nongenomic effects of sex steroid hormones in the CCD. Since the effects of aldosterone and the sex steroids on a major intracellular regulator of CCD ion transport are apparent at physiological concentrations of these hormones, these nongenomic responses have important implications for the reassessment of steroid hormonal regulation of renal transport. METHODS Tissue culture M-1 cells were purchased from the American Type Culture Collection (ATCC No. CRL-2038; Batch F-11321). The M-1 cell line is derived from CCD microdissected from a mouse transgenic for the early region of SV40
[strain Tg(SV40E)Bri7]. M-1 cells were grown in monolayers on permeable supports, and exhibited a high transepithelial resistance and developed a lumen negative potential difference. The associated short-circuit current (Isc) was amiloride sensitive. Cultured cells responded to AVP with a significant increase in Isc. The M-1 cell line displays the properties of the native tissue. making it a useful model to study CCD function. Moreover, M-1 cells, like amphibian A6 cells, do not express a functional mineralocorticoid receptor, making them an ideal model to explore nonclassic, nongenomic rapid responses to physiological concentrations of aldosterone, which do not activate the glucocorticoid receptor. The cells were grown in 25 or 75 cm2 polystyrene culture flasks (Gruinn Technology, Dublin, Ireland) containing a 1:1 mixture of Ham’s F-12 medium and Dulbecco’s modified Eagles medium (Flow Laboratories, Irvine, Scotland, UK) containing 5 mol/L dexamethasone (95%) and fetal bovine serum (5%). This medium also contained 1% HEPES and 0.5% penicillin/streptomycin and was maintained at pH 7.4. Cells were equilibrated with 5% CO2/95% O2 and kept at 37⬚C. Fluid was renewed two to three times weekly. Subconfluent cultures were subcultured by rinsing with 0.25% trypsin-0.03% ethylenediaminetetraacetic acid (EDTA) solution. Most of the trypsin solution was removed, and cells were incubated at 37⬚C for 15 to 20 minutes until cells detached. The cell suspension was centrifuged at 1500 r.p.m. for 10 minutes, and the supernatant was carefully removed. Three to 4 mL of fresh medium were added and aspirated, and cells were dispensed into new flasks. The subcultivation ratio was 1:3 to 1:4. Following aspiration of the supernatant, the cell pellet was resuspended at approximately 2 ⫻ 106 cells per mL of freezing medium containing 95% culture medium and 5% dimethyl sulfoxide (DMSO). One milliliter of the suspension was added per freezing vial. Vials were placed in storage and lowered into liquid nitrogen over a threehour period to achieve a gradual freezing process. To revive cells from prolonged storage, the cells were thawed rapidly in a waterbath at 37⬚C. The cell suspension was centrifuged at 1500 r.p.m. for 10 minutes, and the supernatant was removed. The cell pellets were resuspended with 1 mL of culture medium and transferred to culture flasks. Fresh culture medium was added, and flasks were incubated at 37⬚C. The cells studied experimentally were grown to confluent monolayers with typical dome formation. Experiments were performed on single isolated cells or on monolayers, between passages 15 and 30. M-1 cell monolayers were bathed in normal Krebs solution of composition (in mmol/L): NaCl 140, KCl 5, CaCl2 1, MgCl2 1, glucose 5, HEPES 10, at a pH 7.4. Spectrofluorescence imaging Intracellular calcium activity was measured in M-1 cells loaded with Fura-2 using a spectrofluorescence
Harvey and Higgins: Aldosterone in M-1 cells
video microscopy (Imstar, Paris, France) as described in detail previously [11]. The equation that relates the measured Ca2⫹-bound/Ca2⫹-free fluorescence ratio (R) to [Ca2⫹]i is as follows: [Ca2⫹]i ⫽ Kd ⫻ (R ⫺ Rmin)/(Rmax ⫺ R) ⫻ Sf2/Sb2 where Kd is the product of the dissociation constant of the Ca2⫹:Fura-2 complex and a constant related to the optical characteristics of the system; R is the ratio F340/ F380 of the fluorescence signals measured at 340 nm and 380 nm; Rmin and Rmax are the limiting values of R in the presence of zero and saturating Ca2⫹ concentrations; and Sf2/Sb2 is the ratio of fluorescence values for Ca2⫹bound/Ca2⫹-free indicator measured at the wavelengths used to monitor the Ca2⫹-free indicator (the denominator wavelength of R). Intracellular Fura-2 signals were calibrated after each experimental protocol by imaging cells loaded with the ester derivative of the probe under permeabilized conditions (exposure to 2 mol/L ionomycin). Rmax was measured at an external concentration of 10 mmol/L CaCl2 and Rmin in calcium-free solutions buffered in 2 mmol/L EGTA. Drugs All drugs used in experiments were purchased from Sigma Chemical Co. Ltd. (Fancy Road, Poole, Dorset, UK). In all cases, stock solutions were made up in Krebs solution when water soluble. Otherwise, stock solutions were prepared in an methanol or DMSO, and serial dilutions were made such that the final concentrations of methanol or DMSO were less than 0.1% without any significant effect on results. Data analysis All data values are expressed as mean ⫾ SEM. A minimum of six experiments was performed for each experimental protocol. In experiments in which inhibition of a particular pathway was being investigated, cells treated with inhibitor were matched with an untreated control. Significance was calculated using the Students unpaired t-test. RESULTS Source of [Ca2ⴙ]i in M-1 cells Evidence for a luminal Ca2⫹ entry pathway in M-1 cells. To examine the effects of external calcium on [Ca2⫹]i, M-1 monolayers were first incubated for 10 minutes in a nominally Ca2⫹-free medium on the luminal side, with calcium present in the basolateral solution. A subsequent addition of 1 mmol/L CaCl2 to the luminal side produced a significant yet transient increase in intracellular calcium ([Ca2⫹]i changed from 29 ⫾ 9 to 189 ⫾ 66 nmol/L; ⌬[Ca2⫹]i ⫽ 160 ⫾ 64 nmol/L, N ⫽ 5, P ⬍ 0.05). This increase in [Ca2⫹]i was inhibited when verapamil (10 mol/L)
1397
was present in the luminal bathing solution. Removal of calcium from the basolateral side alone did not affect [Ca2⫹]i over this time period. The influence of intracellular calcium stores on [Ca2⫹]i, was examined using the endoplasmic Ca2⫹-ATPase inhibitors thapsigargin (TG) and 2,5-di-(tert-butyl)-1,4benzohydroquinone (TBQ). The addition of TG, at a concentration of 1 mol/L, resulted in a significant increase in [Ca2⫹]i from 110 ⫾ 22 to 388 ⫾ 31 nmol/L, representing a change in [Ca2⫹]i of 290 ⫾ 46 nmol/L (N ⫽ 9, P ⬍ 0.0001). Surprisingly, the addition of TBQ, at a concentration of 1 mol/L, failed to increase [Ca2⫹]i significantly, which showed an apparent change from 70 ⫾ 16 to 83 ⫾ 19 nmol/L, ⌬[Ca2⫹]i ⫽ 14 ⫾ 5 nmol/L (N ⫽ 10, P ⫽ 0.59). These data indicate that calcium can be mobilized from IP3-sensitive calcium pools in M-1 cells; however, the less selective Ca2⫹-pump antagonist TBQ does not appear to affect calcium mobilization from this pool over a range of concentrations. Effects of Na⫹ transport inhibition on [Ca2⫹]i. The effect of sodium transport inhibitors on intracellular calcium was tested in polarized M-1 cell monolayers. The removal of extracellular sodium on the basolateral side produced a rapid increase in calcium (from 230 ⫾ 20 to 780 ⫾ 80 nmol/L, N ⫽ 6). This effect may be due to a reversal of the Na⫹/Ca2⫹ exchange. The inhibition of Na⫹/K⫹-ATPase pump with ouabain failed to produce any change in intracellular calcium, indicating that small changes in the transmembrane Na⫹ gradient arising from pump inhibition are insufficient to influence the activity of sodium-coupled calcium transporters. The addition of amiloride (1 mol/L) to the apical side of an M-1 cell confluent monolayer produced a sustained decrease in intracellular calcium (from 210 ⫾ 30 to 75 ⫾ 18 nmol/L, N ⫽ 6). A possible mechanism for this effect is an increase in basolateral Na⫹/Ca2⫹ exchange activity following a reduction in intracellular sodium arising from inhibition of the luminal sodium channel. The addition of amiloride to the Krebs solution bathing the basolateral side of M-1 cell monolayers produced an opposite effect on [Ca2⫹]i, as was seen for apical addition of the diuretic. Under these conditions, basolateral amiloride produced an increase in from 180 ⫾ 32 to 720 ⫾ 50 nmol/L (N ⫽ 6). At a concentration of 1 mol/L, amiloride is also a blocker of Na⫹/Ca2⫹ exchange, and these nonspecific effects of the diuretic could explain the rise in calcium. Aldosterone action on [Ca2⫹]i. Aldosterone, at a concentration of 1 nmol/L, produced a rapid, transient peak increase in [Ca2⫹]i from 132 ⫾ 25 to 499 ⫾ 82 nmol/L (mean increase in [Ca2⫹]i ⫽ 367 ⫾ 75 nmol/L, N ⫽ 9, P ⬍ 0.001) in M-1 cell monolayers. This dose of aldosterone was sufficient to produce a maximal increase in [Ca2⫹]i. The mean half-maximal response was produced at 0.1 nmol/L aldosterone, and the minimum concentration of hormone required to produce a significant rise
1398
Harvey and Higgins: Aldosterone in M-1 cells
Fig. 1. Representative Fura-2 spectrofluorescence recording of the increase in intracellular calcium stimulated by 1 nmol/L aldosterone (Aldo) in confluent M-1 cell monolayers (䊉). This increase was completely inhibited by pretreatment (15 min) with the protein kinase C inhibitor chelerythrine chloride (䊏).
in [Ca2⫹]i was 0.08 nmol/L. A typical experiment is shown in Figure 1. The aldosterone response was characterized by a latency period of one to two minutes, which may be indicative of receptor binding or a response that occurs secondarily to a primary signal, such as PKC. The aldosterone effect was abolished under nominally Ca2⫹-free conditions only when calcium was removed from the luminal side. Under the latter conditions, aldosterone produced a nonsignificant change in baseline [Ca2⫹]i levels from 90 ⫾ 7 to 114 ⫾ 10 nmol/L, representing a mean change in [Ca2⫹]i of 24 ⫾ 4 nmol/L (N ⫽ 6, P ⬍ 0.01). The removal of calcium from the basolateral side did not significantly attenuate the [Ca2⫹]i response to hormone. Pretreatment of M-1 cells with TG (10 mol/L) caused a 75% reduced calcium response to aldosterone (N ⫽ 6), indicating that the response requires release from an intracellular stores. Since exposure to an external calcium-free solution for one second prior to aldosterone abolished the response, the data indicate that aldosterone-induced calcium release from intracellular stores is indirect via a calcium entry-induced calcium release mechanism. Effects of actinomycin D and spironolactone. In the presence of actinomycin D (10 mol/L), which inhibits gene transcription, aldosterone also induced a calcium transient (Fig. 2) from a baseline [Ca2⫹]i of 104 ⫾ 31 to 419 ⫾ 44 nmol/L (⌬[Ca2⫹]i ⫽ 315 ⫾ 56 nmol/L, N ⫽ 5, P ⬍ 0.0005). This increase was not significantly different from the aldosterone control response (P ⬍ 0.643). Actino-
Fig. 2. Effect of aldosterone (1 nmol/L) on Ca2ⴙ in M-1 cells in the presence of inhibitors of the genomic pathway, actinomycin D and spironolactone, and the protein kinase C (PKC) inhibitor chelerythrine chloride. Note that the PKC activator, phorbol myristyl acetate (PMA), produced a similar Ca2⫹ response to aldosterone.
mycin D did not affect intracellular calcium levels. The rapid increase in [Ca2⫹]i induced by aldosterone was also unaffected by pretreatment with the specific mineralocorticoid type I receptor antagonist spironolactone (10 mol/L). Aldosterone in the presence of spironolactone induced a rise in [Ca2⫹]i from 115 ⫾ 20 to 332 ⫾ 67 nmol/L (⌬[Ca2⫹]i ⫽ 217 ⫾ 54 nmol/L, N ⫽ 8, P ⬍ 0.001). This increase in [Ca2⫹]i was not significantly different from the aldosterone control response (P ⬍ 0.133). Effect of protein kinase C inhibition. Pretreatment with the specific PKC inhibitor chelerythrine chloride (1 mol/L) completely abolished the aldosterone-induced rise in [Ca2⫹]i (Fig. 1). In the presence of chelerythrine chloride, aldosterone addition resulted in an increase in [Ca2⫹]i from 130 ⫾ 13 to 144 ⫾ 11 nmol/L (⌬[Ca2⫹]i ⫽ 14 ⫾ 11 nmol/L, N ⫽ 6, P ⫽ 0.375). The aldosterone response in the presence of the PKC inhibitor was significantly different from control responses (P ⬍ 0.005). These data indicate that aldosterone increased intracellular calcium in M-1 cells by first activating PKC. In agreement with this conclusion, direct activation of PKC with the phorbol ester phorbol myristyl acetate (PMA;
Harvey and Higgins: Aldosterone in M-1 cells
Fig. 3. Steroid hormone effects on intracellular calcium in confluent M-1 cells. Aldosterone (1 nmol/L) induced a rapid increase in [Ca2⫹]i that was significantly different from that produced by dexamethasone (10 mol/L), 17-estradiol (10 nmol/L), or progesterone (10 nmol/L). Hydrocortisone failed to increase [Ca2⫹]i significantly even at a concentration of 10 mol/L.
1399
Sex steroids. The estrogen analogue 17-estradiol at a concentration of 10 nmol/L raised [Ca2⫹]i significantly from a baseline level of 120 ⫾ 12 to 320 ⫾ 53 nmol/L (⌬[Ca2⫹]i ⫽ 200 ⫾ 44, N ⫽ 7, P ⬍ 0.005). Another female sex steroid, progesterone (10 nmol/L), also generated a significant increase in [Ca2⫹]i from 114 ⫾ 24 to 332 ⫾ 42 nmol/L (⌬[Ca2⫹]i ⫽ 188 ⫾ 23, N ⫽ 4, P ⬍ 0.01). Estradiol and progesterone failed to produce a significant rise in [Ca2⫹]i under nominally Ca2⫹-free conditions. In the presence of zero external calcium, estradiol addition resulted in a slight increase in [Ca2⫹]i from 50 ⫾ 8 to 69 ⫾ 13 nmol/L (⌬[Ca2⫹]i ⫽ 19 ⫾ 6, N ⫽ 4, P ⫽ 0.27), whereas progesterone addition produced an insignificant increase in [Ca2⫹]i from 59 ⫾ 7 to 62 ⫾ 6 nmol/L (⌬[Ca2⫹]i ⫽ 6 ⫾ 2, N ⫽ 8, P ⫽ 0.69). Responses to 17-estradiol and progesterone were prevented by pretreatment (15 min) with the PKC inhibitor chelerythrine chloride (1 mol/L). Estradiol in the presence of chelerythrine chloride induced an increase in [Ca2⫹]i from 112 ⫾ 11 to 121 ⫾ 10 nmol/L (⌬[Ca2⫹]i ⫽ 14 ⫾ 2, N ⫽ 8, P ⫽ 0.518). Similarly, progesterone in the presence of chelerythrine chloride induced an insignificant increase in [Ca2⫹]i from 93 ⫾ 11 to 101 ⫾ 10 nmol/L (⌬[Ca2⫹]i ⫽ 8 ⫾ 2, N ⫽ 8, P ⫽ 0.63). Therefore, like aldosterone, the Ca2⫹ signal transduction pathway for estradiol and progesterone appears to involve PKC. DISCUSSION
100 nmol/L) induced an increase in [Ca2⫹]i from 137 ⫾ 33 to 407 ⫾ 67 nmol/L (⌬[Ca2⫹]i ⫽ 269 ⫾ 44 nmol/L, N ⫽ 7, P ⬍ 0.005). The increases in intracellular calcium following the addition of aldosterone, in the presence or absence of specific inhibitors, are summarized in Figure 2. Inhibitors of the classic genomic pathway for steroid action have little or no effect on the aldosterone response, indicating a nonclassic pathway. PKC inhibition abolishes the response, and activation of the PKC pathway is capable of increasing [Ca2⫹]i in the M-1 cell line, similar to that observed with aldosterone. Rapid Ca2⫹ responses to other steroid hormones. The effects of other salt-retaining steroid hormones on intracellular Ca2⫹ were compared with the aldosterone response (Fig. 3). Glucocorticoids. Dexamethasone (10 mol/L) induced a transient increase in [Ca2⫹]i that was smaller in magnitude than the response observed following aldosterone addition. Dexamethasone increased [Ca2⫹]i from 27 ⫾ 9 to 211 ⫾ 57 nmol/L (⌬[Ca2⫹]i ⫽ 185 ⫾ 55, N ⫽ 4, P ⬍ 0.05). Another glucocorticoid hormone, hydrocortisone, had no significant effect on [Ca2⫹]i, even at micromolar concentrations. Hydrocortisone changed [Ca2⫹]i from 89 ⫾ 10 to 108 ⫾ 6 nmol/L (⌬[Ca2⫹]i ⫽ 20 ⫾ 7, N ⫽ 6, P ⫽ 0.14).
Aldosterone is the most important mineralocorticoid released from the adrenal glands. It regulates solute reabsorption, acting in a delayed manner to influence ion transport across tight junction epithelia such as distal nephron and colon. The classic mechanism of aldosterone stimulation of Na⫹ reabsorption involves the activation of a genomic pathway. It is well documented that aldosterone binds to a cytosolic MR, which is spironolactone sensitive. The active hormone-receptor complex translocates to the nucleus, where it binds to hormoneresponsive elements in the genome and initiates the production of mRNA, which is later translated into new proteins, termed aldosterone-induced proteins (AIPs). These proteins are released into the cytoplasm, where they may be relocated to the apical border and act to alter membrane permeability. Several theories exist as to how aldosterone increases apical Na⫹ reabsorption, and some have proposed that the AIPs themselves are newly synthesized Na⫹ channels, while others argue that these proteins act at a site close to the channel protein and induce a conformational change of the channels [17]. In recent years, a new model for aldosterone action in target epithelia, which introduces a novel specificityconferring mechanism for mineralocorticoid action on epithelial ion transport via protein kinase and calcium signaling has been postulated [2]. In this system of rapid,
1400
Harvey and Higgins: Aldosterone in M-1 cells
Fig. 4. Cortical collecting duct (CCD) cell model of signals involved in the nongenomic pathway for steroid hormone action. Aldosterone binds to cytosolic or membrane-associated elements that may include PKC or a spironolactone-insensitive “receptor” to induce a nongenomic rise in intracellular calcium via a PKC-sensitive luminal entry of calcium and calcium release from intracellular stores. The calcium-kinase complex activates basolateral Na/H exchange, and the resultant intracellular alkalinization activates basolateral K⫹ channels. The increased K⫹ conductance primes the cell for the later genomic up-regulation of sodium absorption.
nongenomic steroid action, aldosterone selectively targets basolateral Na⫹/H⫹ exchange and K⫹ channels (Fig. 4). In the distal colon, for example, aldosterone rapidly up-regulates a K-ATP class channel necessary for sodium absorption while down-regulating a K-Ca channel required for chloride secretion [2, 16]. Thus, the physiological role of the rapid nongenomic response to aldosterone is to shift a pluripotential epithelium into a net absorptive transport mode. These rapid nongenomic actions are unaffected by inhibitors of the type I MR or of protein synthesis (nongenomic) and show a high selectivity for aldosterone over cortisol. This model offers an additional explanation for the clinical evidence for differential effects of glucocorticoids and mineralocorticoids in Na⫹ homeostasis. There is increasing evidence for a rapid action of aldosterone on renal Na⫹/H⫹ exchange. In amphibian renal tubule, aldosterone produced a spironolactone-sensitive activation of Na/H exchange [18], which reflects an alkaline shift in the exchanger’s set point of activation [19], but these effects may be part of the classic genomic effect requiring new protein synthesis [20]. In Madin Darby canine kidney cells, a rapid response to aldosterone on pHi regulation and Na⫹/H⫹ exchange has been reported [21], which appears to be secondarily dependent on a rapid nongenomic action of aldosterone to stimulate a plasma membrane proton conductance [22]. In human and rat distal colon and human sweat gland, aldosterone produces a rapid (within ⬍5 sec) activation of basolateral N/H exchange [3, 15]. This activation appears to be the result of pleiotropic convergent signaling pathways in-
volving Ca2⫹, PKC-␣, arachidonic acid metabolism, and a pertussis-sensitive G-protein activation [15]. If the rapid effects of aldosterone on Na/H exchange are prevented, then the hormone fails to activate the K-ATP channel or inhibit the K-Ca channel [2]. These K⫹ channels show opposite pH sensitivity [2, 23], and it is possible that an intracellular alkalinization alone can explain the different effects of aldosterone on these channels in different cell types in a pluripotential epithelium containing absorptive (surface) cells and secretory (crypt) cells. A direct effect of the protein kinase on channel opening may also be involved in the early response. A rapid calcium transient is a common feature of the nongenomic aldosterone response in colon and sweat gland. In this study, we have shown evidence for a luminal calcium entry pathway in mouse collecting duct M-1 cells, which is stimulated by aldosterone. The reintroduction of calcium into a previous calcium-free bathing solution caused a rapid rise in Ca2⫹i. Also, these cells have the capacity to mobilize calcium from intracellular stores, most likely an IP3-sensitive pool. Inhibitors of sodium reabsorption were also shown to affect intracellular calcium. The removal of external sodium from the basolateral side produced a transient rise in calcium, possibly mediated by reversal of a Na⫹/Ca2⫹ exchanger, whereas removal of sodium from the luminal side caused a fall in intracellular calcium. Likewise, the sodium transport inhibitor amiloride had the opposite effects on calcium, depending on its side of addition. Exposure of monolayers to apical amiloride caused a rapid and sustained fall in calcium. This effect may be due to an increased
Harvey and Higgins: Aldosterone in M-1 cells
driving force for Na⫹/Ca2⫹ exchange following a reduction in intracellular sodium concentration (as a result of inhibition of Na⫹ uptake through ENaC). When the diuretic was added on the basolateral side, however, a transient increase in calcium was observed. Amiloride is also known to inhibit Na⫹/Ca2⫹ exchange at micromolar concentrations, and this latter effect may account for the rise in calcium observed where the basolateral membrane is exposed to the diuretic. The relatively fast action of aldosterone on intracellular Ca2⫹ signaling in mouse CCD cells is incompatible with the classic view of mineralocorticoid hormone action in renal epithelia for the following reasons: (1) Aldosterone produced a rapid transient peak increase in [Ca2⫹]i. (2) The response was abolished in the absence of Ca2⫹ in the extracellular solution, indicating that the rise in [Ca2⫹]i requires entry of Ca2⫹ from the external environment. (3) Pretreatment with spironolactone or actinomycin D failed to abolish the aldosterone effect, indicating a nonclassic pathway. (4) Pretreatment with chelerythrine chloride, a PKC inhibitor, completely inhibited the induction of the calcium response, providing evidence that PKC is involved in this nonclassic increase in [Ca2⫹]i. (5) The phorbol ester PMA induced a calcium transient of similar magnitude to aldosterone, indicating that PKC can mediate fast increases in [Ca2⫹]i. The rapid effects of aldosterone to increase cytosolic calcium are difficult to reconcile with previous reports suggesting that chronic increases in [Ca2⫹]i and PKC down-regulate apical Na⫹ channels [(abstract; Palmer et al, Fed Proc 46:495, 1987); 23–25]. However, there is evidence for a role of Ca2⫹ in mediating the latent genomic aldosterone response of enhanced sodium absorption in renal A6 cells (abstract; Petzel et al, Kidney Int 37:358, 1990). The temporal coincidence of the increases in Ca2⫹, IP3, and Isc and the apparent dependence of both the Ca2⫹ transient and Isc on transcriptional and translational processes would seem to indicate a causal relationship [26]. In tight epithelia in which the rapid effects of aldosterone have been identified, the transport target appears to be the Na⫹/H⫹ exchanger and K⫹ channel activity [2, 14, 18, 19]. Thus, aldosterone may have a rapid action to enhance pHi regulation and affects sodium absorption indirectly via modulation of pHi-sensitive transport pathways (Fig. 4). In the case of stimulation of Na⫹/H⫹ exchange, the resultant intracellular alkalinization can up-regulate K⫹ conductance [14]. An enhanced K⫹ conductance would serve to limit membrane depolarization resulting from chronic aldosterone effects on Na⫹ reabsorption. Thus, the nongenomic phase of aldosterone on Ca2⫹/ PKC, Na⫹/H⫹ exchange, and K⫹ conductance may serve to prime the CCD to maintain membrane potential close to the K⫹ equilibrium potential and ensure a favorable electrochemical membrane potential driving force for genomic effects on Na⫹ uptake (Fig. 4). An enhanced
1401
K⫹ conductance would serve to limit membrane depolarization resulting from chronic aldosterone effects on Na⫹ reabsorption [23]. Protein kinase C may play a role in rapid activation of Na⫹/H⫹ exchange by aldosterone. Serine kinases are important regulators of Na⫹/H⫹ exchange, and protein kinases are involved in activating Na⫹/H⫹ exchange by prostaglandins in renal proximal cells [27, 28]. Shortterm activation of PKC leads to a transient increase in Na⫹/H⫹ exchange activity independent of transcription and translation, whereas long-term activation causes a persistent increase in activity that is dependent on these processes and associated with increased mRNANa/H. This effect may mediate increased Na⫹/H⫹ exchanger activity in chronic conditions such as acidosis or K⫹ deficiency. In the distal colon, the activation of PKC with phorbol esters stimulates Na⫹/H⫹ exchange [15]. Recent work in our laboratory has shown that aldosterone rapidly activates the PKC-␣ isoform in rat colon. In our experiments here, we found that specific inhibition of PKC completely prevents the rapid aldosterone effect on [Ca2⫹]i in M-1 cells. Together, these results suggest that activation of PKC is a key step in the signal transduction of the rapid aldosterone effect in epithelia. The physiological significance of a nongenomic effect of aldosterone is underlined by the low concentrations of hormone required for the rapid calcium response effects, which concur with the physiological concentration of free aldosterone (0.1 nmol/L). Inhibition of aldosterone action is important in the treatment of disturbances in extracellular fluid (ECF) balance, for example, hypertension, heart failure. The type I MR antagonist spironolactone is used pharmacologically to inhibit aldosterone action, but is much less effective at reducing hypertension and ECF volume than angiotensin-converting enzyme (ACE) inhibitors or angiotensin II receptor antagonists. This may be due to the effects of angiotensin II on vascular smooth muscle and salt-absorbing epithelia, but may also be explained by the existence of two classes of aldosterone receptor. There are a number of steps in the rapid nongenomic aldosterone pathway that may serve as targets for pharmacological manipulation. Specific inhibition of aldosterone binding to the unidentified nongenomic receptor may be a means of controlling the rapid aldosterone effect and may be an important target in controlling blood volume and pressure. This could be the basis of the development of a novel mineralocorticoid receptor antagonist with potential new strategies for treatment of human cardiovascular disease. Also, the specific inhibition of PKC isoforms may control nongenomic effects. The rapid estradiol response may involve a similar mechanism to that of aldosterone, independent of protein synthesis and involving PKC. The renal estrogen receptor may be a substrate for PKC, and its properties
1402
Harvey and Higgins: Aldosterone in M-1 cells
can be modified by PKC similarly to the vitamin D receptor. Renal binding of 17-estradiol has been primarily localized to the proximal tubule, while the role of the estradiol receptor in the distal tubule is unknown. Estrogens are allegedly indirectly involved in proximal tubular Ca2⫹ homeostasis via stimulation of 1-␣-hydroxylase, which is responsible for the synthesis of 1,25(OH)2D3, the biologically active form of vitamin D3. In the mouse CCD cells, progesterone and estradiol both increase [Ca2⫹]i, which might explain some of the nongenomic effects of these hormones. The glucocorticoids dexamethasone and hydrocortisone had differing effects on [Ca2⫹]i. Dexamethasone produced a transient increase in [Ca2⫹]i at a concentration 10,000-fold higher than that of aldosterone, whereas hydrocortisone had no effect on [Ca2⫹]i. The response to dexamethasone was statistically different from that of aldosterone. These data indicate that this response is mineralocorticoid specific but does not involve type I receptor activation, since hydrocortisone binds to this receptor at high concentrations. M-1 cells express a functional glucocorticoid receptor but not a mineralocorticoid receptor. Thus, the fast calcium effects of aldosterone at a concentration that does not activate GR and the absence of functional MR in M-1 cells argue strongly in favor of a novel receptor mediating the nongenomic rapid responses to physiological concentrations of aldosterone. CONCLUSIONS AND PERSPECTIVES Our results provide new evidence for a nonclassic, nongenomic mechanism for aldosterone action in CCD, independent of type I mineralocorticoid receptors and involving a PKC-linked, aldosterone-specific receptor or a PKC-dependent Ca2⫹ entry pathway. Furthermore, we have provided evidence that sex steroids also produce rapid increases in [Ca2⫹]i that are PKC dependent and characteristic of recently described nongenomic mechanisms of steroid action. The physiological role of sex steroid targeting to the CCD may be important in determining the degree of fluid retention seen at high circulating concentrations of these hormones. The identity of the putative cyclooxygenase metabolite involved in the nongenomic response remains unknown, and the precise PKC isoform and its phosphorylation targets also remain to be identified. The cellular location of the nongenomic “receptor” is unclear, and although evidence points toward a membrane receptor [9] or PKC itself [12], this has yet to be determined. The rapid (minutes) effects of aldosterone on basolateral Na⫹/H⫹ exchange and K⫹ channels must be reconciled with the delay (hours) before effects on transepithelial sodium absorption become apparent.
ACKNOWLEDGMENTS This work was supported by the Wellcome Trust (programme grant #040067/Z/93) and the Health Research Board of Ireland. Reprint requests to Dr. Brian J. Harvey, Wellcome Trust Cellular Physiology Research Unit, Department of Physiology, University College, Cork, Ireland. E-mail: [email protected]
REFERENCES 1. Rossier BC, Paccolat MP, Verrey F, Kraehenbuhl J-P, Geering K: Mechanism of action of aldosterone: A pleiotropic response. Horm Cell Regul 9:209–225, 1985 2. Maguire D, MacNamara B, Cuffe JE, Winter D, Doolan CM, Urbach V, O’Sullivan GC, Harvey BJ: Rapid responses to aldosterone in human distal colon. Steroids 64:51–63, 1999 3. Hegarty JM, Harvey BJ: Aldosterone accelerates a Na⫹/H⫹exchange dependent pH recovery after acid loading in cultured human eccrine sweat gland epithelial cells. J Physiol (Lond) 517P:20P, 1999 4. Harvey BJ: Energization of sodium absorption by the H(⫹)ATPase pump in mitochondria-rich cells of frog skin. J Exp Biol 172:289–309, 1992 5. Urbach V, Van Kerkhove E, Maguire D, Harvey BJ: Rapid activation of KATP channels by aldosterone in principal cells of frog skin. J Physiol (Lond) 491(Pt 1):111–120, 1996 6. Wehling M: Specific, nongenomic actions of steroid hormones. Annu Rev Physiol 59:365–393, 1997 7. Christ M, Meyer C, Sippel K, Wehling M: Rapid aldosterone signaling in vascular smooth muscle cells: Involvement of phosholipase C, diacylglycerol and protein kinase C. Biochem Biophys Res Commun 213:123–129, 1995 8. Christ M, Eisen C, Aktas J, Theisen K, Wehling M: The inositol1,4,5-trisphosphate system is involved in rapid effects of aldosterone in human mononuclear leukocytes. J Clin Endocrinol 77: 1452–1457, 1993 9. Wehling M, Eisen C, Christ M: Membrane receptors for aldosterone: A new concept of nongenomic mineralocorticoid action. NIPS 8:241–244, 1993 10. Wehling M: Novel aldosterone receptors: Specificity-conferring mechanism at the level of the cell membrane. Steroids 59:160–163, 1994 11. Doolan CM, Harvey BJ: Rapid effects of steroid hormones on free intracellular calcium in T84 colonic epithelial cells. Am J Physiol 271(6 Pt 1):C1935–C1941, 1996 12. Doolan CM, Harvey BJ: Modulation of cytosolic protein kinase C and calcium ion activity by steroid hormones in rat distal colon. J Biol Chem 271:8763–8767, 1996 13. Doolan CM, Harvey BJ: Rapid effects of steroid hormones on cytosolic protein kinase C and calcium ion activity in human distal colon. Mol Cell Endocrinol 138:71–79, 1998 14. Harvey BJ, Ehrenfeld J: Role of Na⫹/H⫹ exchange in the control of intracellular pH and cell membrane conductances in frog skin epithelium. J Gen Physiol 92:793–810, 1988 15. Winter DC, Schneider MF, O’Sullivan GC, Harvey BJ, Geibel JP: Rapid activation of sodium-hydrogen exchange in colonic crypts by aldosterone. J Membr Biol 170:17–26, 1999 16. MacNamara B, Winter DC, Cuffe JE, O’Sullivan GC, Harvey BJ: Basolateral potassium channel involvement in Forskolin (cAMP)-activated secretion in human colon. J Physio (Lond) 519(Pt 1):25–260, 1999 17. Rossier BC, Verrey F, Kraehenbuhl J-P: Transepithelial sodium transport and its control by aldosterone: A molecular approach. Curr Top Membr Transplant 34:167–183, 1989 18. Oberleithner H, Weight M, Westphale J, Wang WH: Aldosterone activates Na⫹/H⫹ exchange and raises cytoplasmic pH in target cells of the amphibian kidney. Proc Natl Acad Sci USA 84:1464–1468, 1987 19. Cooper GJ, Hunter M: Na⫹-H⫹ exchange in frog early distal tubule: Effect of aldosterone on the set-point. J Physiol (Lond) 479(Pt 3):423–432, 1994
Harvey and Higgins: Aldosterone in M-1 cells 20. Cooper GJ, Hunter M: Role of de novo protein synthesis and calmodulin in rapid activation of Na⫹-H⫹ exchange by aldosterone in frog diluting segment. J Physiol (Lond) 491(Pt 1):219–223, 1996 21. Gekle M, Golenhofen N, Oberleithner H, Silbernagl S: Rapid activation of Na⫹/H⫹ exchange by aldosterone in renal epithelial cells requires Ca2⫹ and stimulation of a plasma membrane proton conductance. Proc Natl Acad Sci USA 93:10500–10504, 1996 22. Gekle M, Silbernagl S, Oberleithner H: The mineralocorticoid aldosterone activates a proton conductance in cultured kidney cells. Am J Physiol 273(5 Pt 1):C1673–C1678, 1997 23. Harvey BJ: Cross-talk between sodium and potassium channels in tight epithelia. Kidney Int 48:1191–1199, 1995 24. Palmer LG, Frindt G: Effects of cell Ca and pH on Na channels from rat CCT. Am J Physiol 253:F333–F339, 1987
1403
25. Ling BN, Kokko KE, Eaton DC: Inhibition of apical Na⫹ channels in rabbit cortical collecting tubules by basolateral prostaglandin E2 is modulated by protein kinase C. J Clin Invest 90:1328–1334, 1992 26. Petzel D, Ganz MB, Nestler EJ, Lewis JL, Goldenring J, Akcicek F, Hayslett JP: Correlates of aldosterone-induced increases in Ca2⫹i and Isc suggest that Ca2⫹i is the second messenger for stimulation of apical membrane conductance. J Clin Invest 89:150– 156, 1992 27. Weinman EJ, Shenolikar S: Protein kinase C activates the renal apical membrane Na⫹-H⫹ exchanger. J Membr Biol 93:133–139, 1986 28. Weinman EJ, Dubinsky W, Shenolikar SL: Regulation of the renal Na⫹-H⫹ exchanger by protein phosphorylation. Kidney Int 36:519–525, 1989