Second messenger-specific protein kinases in a salt-absorbing intestinal epithelium.

Calcium, adenosine 3',5'-cyclic monophosphate (cAMP), and guanosine 3',5'-cyclic monophosphate (cGMP) can regulate the same or different ion transport processes within an epithelium, presumably via independent protein phosphorylation mechanisms. Because there have been few detailed studies characterizing these processes in epithelia, we examined the distribution of Ca-, cAMP-, and cGMP-specific protein kinases and substrates in vitro in a homogenous salt-absorbing epithelium, the winter flounder intestine. In this tissue cGMP and Ca inhibit Na-K-2Cl cotransport, cAMP increases anion permeability, and phorbol esters do not affect ion transport. The Ca-specific kinases are calmodulin (CaM) dependent. The tissue possesses type III Ca-CaM protein kinase and its specific substrate elongation factor 2 and type II but not type I Ca-CaM kinase. Addition of phosphatidylserine (PS) and Ca to crude or DEAE-cellulose-purified cytosol neither increased the phosphorylation of exogenous histone H1 substrate nor that of any endogenous substrates. Although these suggest the absence of Ca-phospholipid-dependent kinase (PKC), the cytosol has a 78-kDa protein recognizable by a highly specific polyclonal sheep antibody to rat brain PKC. Both the particulate and cytosolic fractions possess cAMP-specific binding proteins and cAMP-specific phosphoprotein substrates. The particulate fraction cAMP-binding proteins are of molecular mass 50 kDa (pI 5.2) and 48 kDa with multiple isoforms (pI 5.6-6.2); these proteins generate different peptide maps. The cytosol chiefly contains a 50-kDa (pI 5.2) cAMP binding protein that is similar to the particulate 50-kDa protein on peptide mapping. The flounder cAMP binding proteins have the same pI but lower molecular mass and different peptide profiles than the rat brain RII (54/52 kDa) and RI (50 kDa) cAMP regulatory proteins. The cGMP-specific protein kinase was less prominent, very low levels of cGMP-specific binding proteins being detected either by equilibrium binding or by photoaffinity labeling. A prominent kinase substrate in homogenates is a 50-kDa protein, the phosphorylation of which is increased by Ca and cGMP but decreased by cAMP. When intact tissue was prelabeled with 32Pi and then exposed to cGMP, the phosphorylation of a number of substrates including that of a 50-kDa protein was increased. In summary, the flounder intestine possesses the necessary protein phosphorylation mechanisms to account for the regulation of its ion transport processes by second messengers.

8-BrcAMP does not affect Na-K-2Cl cotransport in winter flounder intestine.

The flounder intestinal epithelium possesses luminal Na-K-2Cl cotransport and K secretory mechanisms that account for the short-circuit current (Isc) across the tissue. This epithelium is also highly cation selective. Bumetanide or 8-bromoguanosine 3',5'-cyclic monophosphate (8-BrcGMP) completely inhibit Isc, Na-K-2Cl cotransport, and K secretion, whereas 8-bromoadenosine 3',5'-cyclic monophosphate (8-BrcAMP) inhibits K secretion, partially inhibits Isc, and greatly increases Cl permeability. Although the direct effects of adenosine 3',5'-cyclic monophosphate (cAMP) on other Na-K-2Cl cotransport systems have been examined, the effects of 8-BrcAMP on flounder intestinal Na-K-2Cl cotransport have not been directly measured. In this study, the effects of 8-BrcAMP and bumetanide, either alone or in combination, on the influx (initial rates of uptake) of Cl, Rb, and Na across the luminal surface of the flounder intestine were examined. Bumetanide significantly inhibited Na (50%), Cl, and Rb influx (70% each). In contrast, 8-BrcAMP significantly increased Cl influx in the presence or absence of bumetanide and thereby did not effect the bumetanide-sensitive Cl influx. The nucleotide neither altered bumetanide-sensitive Rb influx nor net transpithelial Na absorption measured under short-circuit conditions but caused a 51-60% decrease in Isc. Measurements of bumetanide-sensitive Na influx exhibited large experimental variability but showed no statistically significant effects of 8-BrcAMP. Prostaglandin E1 (PGE1, 10 microM) and forskolin (10 microM) but not atrial natriuretic factor (1 microM) increased flounder intestinal [cAMP] 200 and 237%, respectively, and, like 8-BrcAMP, increased tissue conductance.(ABSTRACT TRUNCATED AT 250 WORDS)

Rectum has abnormal ion transport but normal cAMP-binding proteins in cystic fibrosis.

The luminal membranes of involved tissues in cystic fibrosis (CF) are relatively impermeable to Cl and the regulation of Cl transport by adenosine 3',5'-cyclic monophosphate (cAMP)-mediated hormones is abnormal. We investigated the human rectum as a putative model for CF. We compared in vivo transrectal potential difference (PD) in CF and in normal subjects in response to sequential perfusions with various test solutions. The base-line PD was different in normal (-35.5 +/- 4.0 mV; lumen negative; mean +/- SE; n = 9) and CF subjects (-23.4 +/- 3.1 mV; n = 6; P less than 0.025) and was eliminated by amiloride (10(-4) M) perfusion in both groups by 3 min. However, in response to a Cl-free solution with amiloride, all six CF subjects exhibit less of a change in PD (PD, -2.2 +/- 1.2 mV vs. -11.7 +/- 1.5 mV in 6 controls; P less than 0.01). Furthermore, normal subjects (n = 7) respond to a 5 mM theophylline + amiloride perfusion with an increase in lumen-negative PD, whereas, CF subjects (n = 6) show no increase in lumen-negative PD. Rectal biopsy specimens from four normal and four CF subjects exhibit similar (2- to 3-fold) increases in theophylline-induced cAMP content and have similar cAMP-binding proteins (CF, n = 3; control, n = 3). We conclude that the rectum is an involved epithelium in CF in which the aberration may lie at a point beyond the binding of cAMP to its protein kinase.

Effects of estradiol on calcium-specific protein phosphorylation in the rat corpus luteum.

Estradiol regulates progesterone synthesis by the corpus luteum of several species. The molecular mechanism(s) of the action of estradiol is unknown, but appears to be independent of exogenous gonadotropins and tissue cAMP. A role for Ca2+ in steroidogenesis has been implicated, and we demonstrate here that estradiol alters Ca-specific protein phosphorylation in rat luteal cells. Corpora lutea obtained from rats hypophysectomized and hysterectomized on day 12 of pregnancy and subsequently treated for 3 days with vehicle or estradiol were assayed in vitro for Ca-specific phosphorylation. Estradiol elevated the content of a number of cytosolic proteins (mol wt X 10(-3), 58, 82, 100, 166, and 183); however, phosphorylation of the 100K protein alone appeared to be specifically enhanced by estradiol. In the presence of Ca2+, phosphorylation of the luteal 100K protein increased 2.33 +/- 0.2-fold in estradiol-treated vs. 1.42 +/- 0.2-fold in vehicle-treated rats (n = 6; P less than 0.05). Phosphorylation of 100K was inhibited by trifluoperazine and stimulated by calmodulin (CaM). Phosphopeptide maps revealed that the 100K luteal protein is identical to the cytosolic CaM-protein kinase III substrate, termed 100K, present in a number of tissues. Estradiol increased both 100K content and CaM-kinase III activity in luteal cells. Immunochemical analysis using antibodies prepared against pancreatic 100K revealed that estradiol treatment increased by at least 5 to 7-fold the luteal content of 100K. In addition, luteal cytosol of estradiol-treated rats enhanced phosphorylation of purified pancreatic 100K 3-fold, whereas that of vehicle-treated rats caused a 1.8-fold stimulation. The effects of estradiol on cytosolic proteins appear to be specific for 100K, since it does not after the activities of CaM-kinases I and II or cAMP-PK. In summary, results of this investigation demonstrate for the first time that estradiol increases the content of several proteins in the corpus luteum; estradiol enhances specifically the Ca-CaM-dependent phosphorylation of a 100K cytosolic protein; and the CaM-kinase III-100K substrate system is hormonally regulated.

Atrial natriuretic factor inhibits Na-K-Cl cotransport in teleost intestine.

Addition of atrial natriuretic factor (ANF) to the contraluminal side of the intestinal mucosa of a marine teleost, the winter flounder Pseudopleuronectes americanus, inhibits short-circuit current, net transepithelial fluxes of Na and Cl, and the unidirectional influx of Rb across the brush border membrane. This action of ANF is closely mimicked by addition of 8-bromo-guanosine 3',5'-cyclic monophosphate (8-BrcGMP). In contrast to the intestine, the opercular epithelium of the flounder did not respond to the in vitro addition of either ANF or 8-BrcGMP. Because intestinal salt and water absorption diminishes when marine fish enter water of lower salinity, ANF may be an important hormonal regulator through which euryhaline fish adapt to varying salinities.

Differing effects of cGMP and cAMP on ion transport across flounder intestine.

The intestinal epithelium of the winter flounder is highly cation selective and actively absorbs NaCl via a bumetanide-sensitive (Na + K + 2Cl) cotransport system; it also actively secretes K+. Combined addition of adenosine 3',5'-cyclic monophosphate (cAMP) and theophylline was previously shown to partially inhibit NaCl absorption and to increase passive Cl- permeability. Because theophylline increases cyclic GMP (cGMP) and cAMP concentrations, we compared the effects of the 8-Bromo analogues of these two nucleotides on ion transport. cGMP inhibits Cl- absorption, K+ secretion, and Cl- and K+ influx across the brush border as effectively as do furosemide and bumetanide. Even at maximal doses, cAMP is less effective than cGMP in inhibiting ion transport; however, unlike cGMP, it abolishes the cation selectivity of the epithelium by greatly increasing Cl- permeability. The effects of the two nucleotides are not additive with each other or with those of bumetanide, although cGMP or bumetanide can further inhibit transport in cAMP-treated tissues.