Modulation of Na and Cl transport by mineralocorticoids

Coping with excess salt: adaptive functions of extrarenal osmoregulatory organs in vertebrates

In all organisms, changing environmental conditions require appropriate regulatory measures to physiologically adjust to the altered situation. Uptake of excess salt in non-mammalian vertebrates having limited or no access to freshwater is balanced by extrarenal salt excretion through specialized structures called 'salt glands'. Nasal salt glands of marine birds are usually fully developed in very early stages of their lives since individuals of these species are exposed to salt soon after hatching. In individuals of other bird species, salt uptake may occur infrequently. In these animals, glands are usually quiescent and glandular cells are kept in a fairly undifferentiated state. This is the situation in 'naive' ducklings, Anas platyrhynchos, which have never been exposed to excess salt. When these animals become initially osmotically stressed, the nasal glands start to secrete a moderately hypertonic sodium chloride solution but secretory performance is meager. Within 48 h after the initial stimulus, however, the number of cells per gland is elevated by a factor of 2-3, the secretory cells differentiate and acquire full secretory capacity. During this differentiation process, extensive surface specializations are formed. The number of mitochondria is increased and metabolic enzymes and transporters are upregulated. These adaptive growth and differentiation processes result in a much higher efficiency of salt excretion in acclimated ducklings compared with naive animals. Receptors and signal transduction pathways in salt gland cells controling the adaptive processes seem to be the same as those controling salt secretion, namely muscarinic acetylcholine receptors and receptors for vasoactive intestinal peptide. This review focusses on signal transduction pathways activated by muscarinic receptors which seem to fine-tune salt secretion in salt-adapted ducklings and may control adaptive growth and differentiation processes in the nasal gland of naive animals.

General overview of mineralocorticoid hormone action

The adrenal cortex elaborates two major groups of steroids that have been arbitrarily classified as glucocorticoids and mineralocorticoids, despite the fact that carbohydrate metabolism is intimately linked to mineral balance in mammals. In fact, glucocorticoids assured both of these functions in all living cells, animal and photosynthetic, prior to the appearance of aldosterone in teleosts at the dawn of terrestrial colonization. The evolutionary drive for a hormone specifically designed for hydromineral regulation led to zonation for the conversion of 18-hydroxycorticosterone into aldosterone through the catalytic action of a synthase in the secluded compartment of the adrenal zona glomerulosa. Corticoid hormones exert their physiological action by binding to receptors that belong to a transcription factor superfamily, which also includes some of the proteins regulating steroid synthesis. Steroids stimulate sodium absorption by the activation and/or de novo synthesis of the ion-gated, amiloride-sensitive sodium channel in the apical membrane and that of the Na+/K+-ATPase in the basolateral membrane. Receptors, channels, and pumps apparently are linked to the cytoskeleton and are further regulated variously by methylation, phosphorylation, ubiquination, and glycosylation, suggesting a complex system of control at multiple checkpoints. Mutations in genes for many of these different proteins have been described and are known to cause clinical disease.

In birds, NHE2 is major brush-border Na+/H+ exchanger in colon and is increased by a low-NaCl diet

We previously reported that mammalian small intestinal and colonic brush borders (BBs) contained both epithelial Na+/H+ exchangers NHE2 and NHE3. We now show that, in the avian (chicken) colon, NHE2 is the major functional isoform under basal conditions and when stimulated by a low-NaCl diet. Hubbard chickens were maintained for 2 wk on a high- or low-NaCl diet. After the chickens were killed, the ileum and colon were removed, and BBs were prepared by Mg2+ precipitation and 22Na and D-[14C]glucose uptake determined in the BB vesicles. NHE2 and NHE3 were separated by differential sensitivity to HOE-694 (NHE2 defined as Na+/H+ exchange inhibited by 50 microM HOE-694). Chickens on a low-Na+ diet have increased plasma aldosterone (10 vs. 207 pg/ml). On the high-NaCl diet, both NHE2 and NHE3 contributed to ileal and colonic apical Na+/H+ exchange, contributing equally in ileum, but NHE2 being the major component in colon (86%). Low-NaCl diet significantly increased ileal and colonic BB Na+/H+ exchange; the increase in BB Na+/H+ exchange in both ileum and colon was entirely due to an increase in NHE2 with no change in NHE3 activity. In contrast, low-NaCl diet decreased ileal and colonic Na+-dependent D-glucose uptake. Western analysis showed that low-Na+ diet increased the amount of NHE2 in the ileal and colonic BB and decreased the amount of ileal Na+-dependent glucose transporter SGLT1. Both NHE2 and NHE3 were present in the apical but not basolateral membranes (BLM) of ileal and colonic epithelial cells. In summary, 1) NHE2 and NHE3 are both present in the BB and not BLM of chicken ileum and colon; 2) NHE2 is the major physiological colonic BB Na+/H+ exchanger under basal conditions; 3) low-NaCl diet, which increases plasma aldosterone, increases ileal and colonic BB Na+/H+ exchange and decreases Na+-dependent D-glucose uptake; 4) the stimulation of colonic BB Na+/H+ exchange is due to increased activity and amount of NHE2; and 5) the inhibition of ileal D-glucose uptake is associated with a decrease in SGLT1 amount. NHE2 is the major chicken colonic BB Na+/H+ exchanger.

Metabolism of corticosterone in mammalian and avian intestine

11beta-hydroxysteroid dehydrogenase (11betaHSD) catalyzes the conversion of the glucocorticoids, corticosterone and cortisol, to the respective derivatives 11-dehydrocorticosterone and cortisone. The recent findings underline the importance of this enzyme in excluding glucocorticoids from mineralocorticoid receptors. In the present study, 11betaHSD activity was compared in the intestine of herbivorous (guinea pig), omnivorous (rat), and granivorous (hen) animals, i.e., in animals in which the Na+ transport either is or is not regulated by aldosterone under normal conditions and in which the plasma levels of individual glucocorticoids are different. Slices of various intestinal segments were incubated in the presence of corticosterone or 11-dehydrocorticosterone, and the steroids were extracted and analyzed by HPLC. In the mammalian intestine, the activity of 11betaHSD was very low (approaching zero) in aldosterone-insensitive segments (duodenum, jejunum) but significant activity was revealed in aldosterone-sensitive segments (ileum, cecum, and proximal and distal colon). In comparison with the rat, the guinea pig large intestine exhibited significantly higher activity of 11betaHSD. There was no detectable reductase activity (conversion of 11-dehydrocorticosterone to corticosterone) in any intestinal segments of either species. Unexpectedly, no 11betaHSD activity was observed in the avian intestine. It was found that, in contrast to the mammalian intestine, corticosterone was metabolized to 20-dihydrocorticosterone while 11-dehydrocorticosterone was converted to 11-dehydro-20-dihydrocorticosterone. The distribution of 20-hydroxysteroid dehydrogenase (20HSD) activity in the avian intestine was homogenous along the intestine and did not correlate with the mineralocorticoid sensitivity of intestinal segments. To trace different cosubstrate dependence of 11betaHSD and 20HSD, homogenates of ileum and distal colon were incubated with NAD+/NADH or NADP+/NADPH, respectively. In accordance with slice experiments mammalian intestine displayed only oxidation of corticosterone to 11-dehydrocorticosterone and NAD+ preference. In avian intestine, the metabolite formed from corticosterone was 11-dehydrocorticosterone in the presence of NAD+ or NADP+ whereas in the presence of NADPH 11-dehydro-20-dihydrocorticosterone and 20-dihydrocorticosterone were formed. Given the wide similarity between mineralocorticoid regulation of epithelial transport in mammals and birds, the unexpected finding of differences in the metabolism of corticosterone suggests that role of 20HSD is to allow aldosterone occupancy of mineralocorticoid receptors.

An aldosterone regulated chicken intestine protein with high affinity to amiloride

The pattern of chicken intestine amiloride-binding proteins was determined using the photoreactive amiloride analogue 2'-methoxy-5'-nitrobenzamil (NMBA) and a polyclonal anti-amiloride antibody. At 10(-7)M, NMBA inhibits approximately 62% of the Na+ channel activity. At this concentration the amiloride analogue labels a number of membrane proteins, and in particular a 40-45 kDa polypeptide denoted ABP40. Incorporation of NMBA into ABP40 could be prevented by a 100-fold excess of benzamil, but not by a 1000-fold excess of 5-(N-ethyl-N-isopropyl)-amiloride. Labeling of ABP40 was intense in membranes derived from salt-deprived chickens and approximately 5-fold weaker in membranes from salt-repleted animals. Because of its small size, ABP40 is not likely to be an avian Na+ channel subunit, yet this amiloride-binding protein could be involved in the response to aldosterone.

Cloning and induction by low NaCl intake of avian intestine Na+ channel subunits

The alpha-subunit of the highly Na(+)-selective amiloride-blockable channel (ENaC) was cloned from chicken lower intestine. The deduced amino acid sequence of the avian clone exhibits -60% identity to the previously cloned mammalian and amphibian alpha-subunits. It also maintains the same hydropathy profile and structural motifs. These include two transmembrane domains separated by a large extracellular loop, four extracellular N-glycosylation sites, a cysteine-rich box in the extracellular domain, and a proline-rich stretch at the carboxy terminus. Xenopus oocytes injected with cRNA transcribed from this clone express a small amiloride-blockable Na+ conductance. Degenerate primers have been used to amplify two other related products. Sequence homology indicates that one of them is the beta-subunit, whereas the other appears to represent a closely related but different transcript. Regulation of the mRNA corresponding to these clones was examined in chickens fed normal and low-NaCl rations. The low-salt diet evoked an approximately fourfold increase in the abundance of mRNA coding for the alpha-subunit, presumably through an increase in plasma aldosterone. The beta- and "beta-like" transcripts were even more strongly affected. The current data provide additional information on sequence conservation in the growing ENaC family and demonstrate that the avian intestine channel is strongly induced by varying NaCl intake.

cAMP-activation of amiloride-sensitive Na+ channels from guinea-pig colon expressed in Xenopus oocytes

Guinea-pig distal colonic mRNA injection into Xenopus laevis oocytes resulted in expression of functional active epithelial Na+ channels in the oocyte plasma membrane. Poly(A)+ RNA was extracted from distal colonic mucosa of animals fed either a high-salt (HS) or a low-salt (LS) diet. The electrophysiological properties of the expressed amiloride-sensitive Na+ conductances were investigated by conventional two-electrode voltage-clamp and patch-clamp measurements. Injection of poly(A)+ RNA from HS-fed animals [from hereon referred to as HS-poly(A)+ RNA] into oocytes induced the expression of amiloride-sensitive Na+ conductances. On the other hand, oocytes injected with poly(A)+ RNA from LS-fed animals [LS-poly(A)+ RNA] expressed a markedly larger amount of amiloride-blockable Na+ conductances. LS-poly(A)+ RNA-induced conductances were completely inhibitable by amiloride with a Ki of 77 nM, and were also blocked by benzamil with a Ki of 1.8 nM. 5-(N-Ethyl-N-isopropyl)-amiloride (EIPA), even in high doses (25 "mu"M), had no detectable effect on the Na+ conductances. Expressed amiloride-sensitive Na+ channels could be further activated by cAMP leading to nearly doubled clamp currents. When Na+ was replaced by K+, amiloride (1 "mu"M) showed no effect on the clamp current. Single-channel analysis revealed slow gating behaviour, open probabilities (Po) between 0.4 and 0.9, and slope conductances of 3. 8 pS for Na+ and 5.6 pS for Li+. The expressed channels showed to be highly selective for Na+ over K+ with a permeability ratio PNa/PK > 20. Amiloride (500 nM) reduced channel Po to values < 0.05. All these features make the guinea-pig distal colon of LS-fed animals an interesting mRNA source for the expression of highly amiloride-sensitive Na+ channels in Xenopus oocytes, which could provide new insights in the regulatory mechanism of these channels.

Immunocytochemical localization of amiloride-sensitive sodium channels in the lower intestine of the hen

We have used polyclonal antibodies generated against purified bovine renal amiloride-sensitive Na+ channels to localize amiloride-sensitive Na+ channels within the lower intestine (colon and coprodeum) of the hen. These antibodies cross-reacted with two polypeptides exhibiting M(r)'s of 235 and 150 kDa on immunoblots of detergent-solubilized apical membrane fractions from both the colon and coprodeum. The apparent molecular masses of theses polypeptides are in agreement with the M(r)'s of 2 of the subunits of the renal high amiloride-affinity Na+ channel, namely the alpha and the beta (= amiloride binding) subunits. The cellular distribution of Na+ channels was determined by immunoperoxidase and indirect immunofluorescence cytochemical techniques. The apical (luminal) membrane and cytoplasm of villar principal cells in both colon and coprodeum exhibited immunoreactivity, whereas goblet cells were negative. Both principal and goblet cells of the crypts were also negative. We conclude that the amiloride-sensitive Na+ channels are localized to the principal cells of the intestinal villi and that these cells are responsible for intestinal Na+ absorption.

Expression of amiloride-sensitive Na+ channels of hen lower intestine in Xenopus oocytes: electrophysiological studies on the dependence of varying NaCl intake

Epithelial Na+ channels were incorporated into the plasma membrane of Xenopus laevis oocytes after micro-injection of RNA from hen lower intestinal epithelium (colon and coprodeum). The animals were fed either a normal poultry food which contained NaCl (HS), or a similar food devoid of NaCl (LS). Oocytes were monitored for the expression of amiloride-sensitive sodium channels by measuring membrane potentials and currents. Oocytes injected with poly(A)+RNA prepared from HS animals or non-injected control oocytes showed no detectable sodium currents, whereas oocytes injected with LS-poly(A)+RNA had large amiloride-blockable sodium currents. These currents were almost completely saturated by sodium concentrations of 20 mM with a Km of about 2.6 mM sodium. Amiloride (10 microM) inhibits the expressed sodium channels entirely and examination of dose response relationships yielded a half-maximal inhibition concentration (Ki) of 120 nM amiloride. I-V difference curves in the presence or absence of sodium or amiloride (10 microM) indicate a potential dependence of the sodium transport which can be described by the Goldman equation. When Na+ is replaced by K+, no amiloride response was detected indicating a high selectivity for Na+ over K+. These results provide strong evidence that intestinal Na+ channels are regulated by dietary salt intake on the RNA level.

Na-transport during long-term incubation of the hen lower intestine: No aldosterone effect

1. In order to test the aldosterone effect in vitro, Na-transport of the coprodeal epithelium from hens on low-NaCl diet was measured in the Ussing chamber for up to 8 hr. Short-circuit current (SCC, near equal to the amiloride inhibitable Na-transport) was recorded. 2. Incubation media were either Krebs-phosphate or bicarbonate buffer with and without addition of beta-hydroxybutyrate, glutamine and mannose as "metabolic fuels". The media were replaced every hour. The Krebs-phosphate buffer was further tested with and without indomethacin and media replacement. Na-transport was best maintained in this buffer with replacement: SCC at 4 hr: 156 +/- 21 microA/cm2, 8 hr: 73 +/- 14 microA/cm2. 3. The aldosterone experiments were carried out on tissues from hens resalinated for 24 hr. No effects were demonstrated at concentrations up to 10(-5) M. The SCC showed an unexpected raise within 2-4 hr to a very high level (4 hr: 221 +/- 61 microA/cm2) both in the control and in all aldosterone-treated tissues. This SCC decreased slowly to 210 +/- 29 microA/cm2 at 8 hr. It was abolished by amiloride. 4. No increase in SCC was observed in tissues from hens after 48 and 72 hr of resalination either after aldosterone or on chronic high-NaCl diet.

Different modes of electrogenic Na+ absorption in the coprodeum of the chicken embryo: role of extracellular Ca2+

Transepithelial electrogenic Na+ transport (INa) was investigated in the coprodeum of 20-days-old chicken embryos in Ussing chambers. Short circuit current (Isc) and transepithelial resistance (Rt) were 14.7 +/- 4.8 microA.cm-2 (n = 12) and 0.53 +/- 0.09 k omega.cm-2 (n = 12), respectively. INa was calculated from changes in Isc by substitution of mucosal Na+ by (N-methyl-D-glucamine) (NMDG). Isc inversed during Na+ removal, and INa was found to be 27.8 +/- 4.7 microA.cm-2 (n = 12). Amiloride (100 mumol.l-1) inhibited only about 60% of INa. Analysis of Isc fluctuations revealed a Lorentzian component in the power density spectrum with a corner frequency of about 57 Hz. This component was not correlated to INa, and its origin is still unclear. Removal of mucosal Ca2+ increased INa about 2.5-fold due to an increase of the amiloride-insensitive component of INa in additionally investigated adult tissues. The results clearly show that this is due to a non-selective cation channel with an "apparent" order of selectivity Cs+ greater than Na+ = K+ greater than Rb+ greater than Li+. The Ca2+ concentration required to block 50% of the Isc was about 18 mumol.l-1. The IscCa could also be suppressed by other divalent cations such as Mg2+ and Ba2+. Additionally, an INa-linked Lorentzian component occurred which dominated the control spectrum with a significantly higher corner frequency (about 88 Hz). The results indicate that Na+ absorption in the coprodeum of the chicken embryo is more complex than in adult hens.(ABSTRACT TRUNCATED AT 250 WORDS)