The distal convoluted tubule of rabbit kidney does not express a functional sodium channel.

We sought to assess whether the distal convoluted tubule (DCT) segment of the rabbit nephron expresses a functional epithelial sodium channel. First, the transepithelial voltage (V(te), lumen vs. bath) was measured in isolated perfused DCT segments (assessed separately in the upstream half and the downstream half of the DCT). V(te) was zero and not affected by amiloride or barium in the upstream DCT. V(te) was sometimes negative in the downstream DCT and depolarized by amiloride and hyperpolarized by barium, suggesting inclusion of connecting tubule (CNT) cells. To determine expression of epithelial sodium channel (ENaC) mRNA subunits by the upstream DCT, rabbit alpha-, beta-, and gamma-ENaC cDNA fragments were cloned and primers were selected for single-nephron RT-PCR analysis. Although alpha-ENaC was expressed by the DCT, beta- and gamma-ENaC were not detected in the DCT. In contrast, the CNT, CCD, and outer medullary collecting duct (OMCD) expressed all three subunits. Nedd4 was also not detected in the DCT but was expressed by the CNT, CCD, and OMCD. When upstream DCT fragments were grown to confluent monolayers in primary culture, the epithelia exhibited negative voltages and high transepithelial resistances and expressed mRNA for all three ENaC subunits as well as for Nedd4. The absence of a negative voltage and failure to detect transcript for beta- and gamma-ENaC and Nedd4 in the native rabbit DCT suggest that the sodium channel is not a significant pathway for sodium absorption by this segment. The phenotype conversion observed when DCT cells are grown in culture does not rule out the possibility that there may be conditions in which the DCT in the intact kidney expresses sodium channel activity. The results are consistent with the notion that DCT sodium transport is predominantly, if not exclusively, electroneutral.

KCNA10: a novel ion channel functionally related to both voltage-gated potassium and CNG cation channels.

Our laboratory previously cloned a novel rabbit gene (Kcn1), expressed in kidney, heart, and aorta, and predicted to encode a protein with 58% amino acid identity with the K channel Shaker Kv1.3 (Yao X et al. Proc Natl Acad Sci USA 92: 11711-11715, 1995). Because Kcn1 did not express well (peak current in Xenopus laevis oocytes of 0.3 microA at +60 mV), the human homolog (KCNA10) was isolated, and its expression was optimized in oocytes. KCNA10 mediates voltage-gated K(+) currents that exhibit minimal steady-state inactivation. Ensemble currents of 5-10 microA at +40 mV were consistently recorded from injected oocytes. Channels are closed at the holding potential of -80 mV but are progressively activated by depolarizations more positive than -30 mV, with half-activation at +3.5 +/- 2.5 mV. The channel displays an unusual inhibitor profile because, in addition to being blocked by classical K channel blockers (barium tetraethylammonium and 4-aminopyridine), it is also sensitive to inhibitors of cyclic nucleotide-gated (CNG) cation channels (verapamil and pimozide). Tail-current analysis shows a reversal potential shift of 47 mV/decade change in K concentration, indicating a K-to-Na selectivity ratio of at least 15:1. The phorbol ester phorbol 12-myristate 13-acetate, an activator of protein kinase C, inhibited whole cell current by 42%. Analysis of single-channel currents reveals a conductance of approximately 11 pS. We conclude KCNA10 is a novel human voltage-gated K channel with features common to both K-selective and CNG cation channels. Given its distribution in renal blood vessels and heart, we speculate that KCNA10 may be involved in regulating the tone of renal vascular smooth muscle and may also participate in the cardiac action potential.

Close association of the N terminus of Kv1.3 with the pore region.

The Shaker superfamily encodes voltage-gated potassium (Kv) channels. The N termini of Shaker proteins are located intracellularly and contain several domains shown to regulate important aspects of channel function, such as speed of inactivation, channel assembly (T1 domain), and steady state protein level (T0 domain, amino acids 3-39 in rabbit). Mutations and/or deletion of certain amino acids in the T0 domain lead to a 13-fold amplification of Kv current as compared with wild type channels, primarily by increasing the absolute number of channel proteins present in the membrane (Segal, A. S., Yao, X., and Desir, G. V. (1999) Biochem. Biophys. Res. Commun. 254, 54-64). Although T0 mutants have kinetic properties virtually indistinguishable from wild type, they were noted to have a slightly larger single channel conductance, suggesting that the T0 domain might also interact with the pore region. In the present study we show that although T0 does not affect pore selectivity, it does modulate the binding affinity of the pore blocker, charybdotoxin. These results suggest that the N terminus of Kv1.3 is closely associated with the pore region.

Defective processing and expression of thiazide-sensitive Na-Cl cotransporter as a cause of Gitelman's syndrome.

Gitelman's syndrome is an autosomal recessive disorder of salt wasting and hypokalemia caused by mutations in the thiazide-sensitive Na-Cl cotransporter. To investigate the pathogenesis of Gitelman's syndrome, eight disease mutations were introduced into the mouse thiazide-sensitive Na-Cl cotransporter and studied by functional expression in Xenopus oocytes. Sodium uptake into oocytes that expressed the wild-type clone was more than sevenfold greater than uptake into control oocytes. Uptake into oocytes that expressed the mutated transporters was not different from control. Hydrochlorothiazide reduced Na uptake by oocytes expressing the wild-type gene to control values but had no effect on oocytes expressing the mutant clones. Western blots of oocytes injected with the wild-type clone showed bands representing glycosylated (125 kDa) and unglycosylated (110 kDa) forms of the transport protein. Immunoblot of oocytes expressing the mutated clones showed only the unglycosylated protein, indicating that protein processing was disrupted. Immunocytochemistry with an antibody against the transport protein showed intense membrane staining of oocytes expressing the wild-type protein. Membrane staining was completely absent from oocytes expressing mNCC(R948X); instead, diffuse cytoplasmic staining was evident. In summary, the results show that several mutations that cause Gitelman's syndrome are nonfunctional because the mutant thiazide-sensitive Na-Cl cotransporter is not processed normally, probably activating the "quality control" mechanism of the endoplasmic reticulum.

The T0 domain of rabbit KV1.3 regulates steady state channel protein level.

The Shaker superfamily encodes voltage-gated potassium (Kv) channels. The amino (N) terminus is important for channel assembly and mediates fast inactivation. We recently isolated a Kv channel from rabbit kidney, denoted rabKv1.3 (Yao et al., J. Clin. Invest. 97, 2525-2533, 1996) and found that deleting a region (T0 domain, amino acids 3-39) proximal to the T1 recognition domain (a.a 42-185) leads to a 13-fold amplification of Kv current as compared to wild type channels (Yao et al., BBRC 249, 492-498). Here we show that deleting the T0 domain affects neither single channel conductance nor channel open probability. Instead, it increases the absolute number of channel proteins present in the membrane. We conclude that the T0 domain is a previously unrecognized Shaker Kv1.3, N-terminal regulatory region that modulates steady state channel protein density in the plasma membrane.

Rabbit distal convoluted tubule coexpresses NaCl cotransporter and 11 beta-hydroxysteroid dehydrogenase II mRNA.

BACKGROUND: Although the renal cortical collecting duct (CCD) is a principal target for aldosterone, recent evidence suggests that salt transport by other nephron segments may also be regulated by aldosterone. Electroneutral and thiazide-sensitive NaCl cotransport by the distal convoluted tubule (DCT) of the rat is increased in animals deprived of dietary NaCl. We tested the hypothesis that the DCT of the rabbit is an aldosterone target tissue. METHODS: The single-nephron reverse-transcriptase/polymerase chain reaction (RT-PCR) technique was used to determine mRNA expression of NaCl cotransporter and 11 beta-HSD 2 in dissected nephron segments. The rabbit NaCl cotransporter was first cloned and rabbit-specific primers selected. A micro-assay was developed to assess 11 beta-HSD 2 enzyme activity in 0.5 mm samples of the same nephron segments. RESULTS: NaCl cotransporter was expressed in 0 of 6 proximal tubule (PT), 6 of 6 DCT and 3 of 6 CCD samples, while 11 beta-HSD was found in 0 of 7 PT, 7 of 7 DCT and 9 of 9 CCD samples. Corticosterone was converted to 11-dehydrocorticosterone at a high rate and to a similar extent by both the DCT and CCD, but not the PT. CONCLUSIONS: We conclude that the DCT is a target tissue for the action of aldosterone. Axial heterogeneity of electroneutral (in DCT) and electrogenic (in CCD) Na transporters along the distal nephron may improve sodium recovery in low salt and volume states.

Cloning and localization of a double-pore K channel, KCNK1: exclusive expression in distal nephron segments.

The K-selective channel, TOK1, recently identified in yeast, displays the unusual structural feature of having two putative pore regions, in contrast to all previously cloned K channels. Using the TOK1 pore regions as probes, we identified a human kidney cDNA encoding a 337-amino acid protein (hKCNK1) with four transmembrane segments and two pore regions containing the signature sequence of K channels. Amino acid identity to TOK1 is only 15% overall but 40% at the pores. Northern analysis indicates high expression of a 1.9-kb message in brain > kidney >> heart. Nephron segment localization, carried out in rabbit by reverse transcription-polymerase chain reaction, reveals that KCNK1 is expressed in cortical thick ascending limb, connecting tubule, and cortical collecting duct. It was not detected in the proximal tubule, medullary thick ascending limb, distal convoluted tubule, and glomerulus. We conclude that KCNK1 is a unique, double-pore, mammalian K channel, distantly related to the yeast channel TOK1, that is expressed in distal tubule and is a candidate to participate in renal K homeostasis.

Genomic localization of the human gene for KCNA10, a cGMP-activated K channel.

Potassium (K) channels are important components of virtually all cells, and they play critical roles in many cellular functions. KCNA10 represents a new class of K channel specifically regulated by cGMP and postulated to mediate the effects of substances that increase intracellular cGMP. Since KCNA10 has the potential to be useful in candidate gene analysis of inherited diseases, the human gene for KCNA10 was characterized. Fluorescence in situ hybridization indicates that human KCNA10 maps to chromosome 1 at p13.1-->p22.1. Finer mapping of the gene was achieved by PCR of a set of CEPH YAC clones that spanned the region of interest. We found that YAC 818b9 contains human KCNA10. These data indicate human KCNA10 maps to 1p13.1 and resides within the genetic interval defined by microsatellite loci D1S2809 and D1S2726. That region of chromosome 1 contains another K channel gene, KCNA3.

Genomic structure and regulation of Kcn1, a cGMP-gated potassium channel.

We recently cloned a novel rabbit gene that encodes a 725-amino acid protein (Kcn1) (Y. Yao, A. S. Segal, P. Welling, X. Zhang, C. M. McNicholas, D. Engel, E. L. Boulpaep, and G. Desir. Proc. Natl. Acad. Sci. USA 92: 11711-11715, 1995). Kcn1 RNA injected in Xenopus oocytes leads to the expression of potassium channels that are specifically activated by guanosine 3',5'-cyclic monophosphate (cGMP). Northern blot and ribonuclease (RNase) protection analysis show that Kcn1 is differentially expressed in kidney, aorta, brain, and heart. The purpose of present study is to determine the structure of Kcn1 gene, analyze the promoter region, and identify cis-regulatory elements responsible for transcription. We find that the coding region of Kcn is intronless. The major transcription initiation site was identified by primer extension. Sequence analysis of the 5'-flanking region indicates that, although the gene lacks a typical TATA box, it does have a TATA-box-like region (-TAT-). Using luciferase reporter constructs transfected in the porcine kidney cell line (LLC-PK1), the promoter region and a 5' enhancer element were identified by deletion analysis. Phorbol esters (12-O-tetradecanoylphorbol-13-acetate) and forskolin stimulated Kcn1 gene expression 2.5- and 3.5-fold, respectively. In conclusion, we have identified the region of the novel potassium channel gene, Kcn1, that contains the promoter, a 5' enhancer, and several cis-regulatory elements and shown that gene transcription is stimulated by cAMP and phorbol esters.

Molecular cloning of a glibenclamide-sensitive, voltage-gated potassium channel expressed in rabbit kidney.

Shaker genes encode voltage-gated potassium channels (Kv). We have shown previously that genes from Shaker subfamilies Kv1.1, 1.2, 1.4 are expressed in rabbit kidney. Recent functional and molecular evidence indicate that the predominant potassium conductance of the kidney medullary cell line GRB-PAP1 is composed of Shaker-like potassium channels. We now report the molecular cloning and functional expression of a new Shaker-related voltage-gated potassium channel, rabKv1.3, that is expressed in rabbit brain and kidney medulla. The protein, predicted to be 513 amino acids long, is most closely related to the Kv1.3 family although it differs significantly from other members of that family at the amino terminus. In Xenopus oocytes, rabKv1.3 cRNA expresses a voltage activated K current with kinetic characteristics similar to other members of the Kv1.3 family. However, unlike previously described Shaker channels, it is sensitive to glibenclamide and its single channel conductance saturates. This is the first report of the functional expression of a voltage-gated K channel clone expressed in kidney. We conclude that rabKv1.3 is a novel member of the Shaker superfamily that may play an important role in renal potassium transport.

Primary structure and functional expression of a cGMP-gated potassium channel.

Cyclic nucleotides modulate potassium (K) channel activity in many cells and are thought to act indirectly by inducing channel protein phosphorylation. Herein we report the isolation from rabbit of a gene encoding a K channel (Kcn1) that is specifically activated by cGMP and not by cAMP. Analysis of the deduced amino acid sequence (725 amino acids) indicates that, in addition to a core region that is highly homologous to Shaker K channels, Kcn1 also contains a cysteine-rich region similar to that of ligand-gated ion channels and a cyclic nucleotide-binding region. Northern blot analysis detects gene expression in kidney, aorta, and brain. Kcn1 represents a class of K channels that may be specifically regulated by cGMP and could play an important role in mediating the effects of substances, such as nitric oxide, that increase intracellular cGMP.

The structure, regulation and pathophysiology of potassium channels.

Potassium channels are membrane proteins that allow the passive diffusion of potassium down its electrochemical gradient. They play a critical role in many cellular processes and have, therefore, been extensively studied. Over the past several years, the molecular structures of several types of potassium channels have been elucidated. This review investigates the most recent findings regarding the structure, regulation and pathophysiology of potassium channels.

Identification of a novel K-channel gene (KC22) that is highly expressed in distal tubule of rabbit kidney.

The Shaker gene family encodes voltage-gated K channels. Five partial-length Shaker-like cDNAs (KC2, 4, 10, 19, and 22) were previously isolated from rabbit kidney using polymerase chain reaction (PCR) [G. V. Desir, E. Hamlin, A.H. Puente, R.F. Reilly, F. Hiledebrandt, and P. Igarashi. Am. J. Physiol. 262 (Renal Fluid Electrolyte Physiol. 31): F151-F157, 1992]. We now report the cloning of another Shaker-like cDNA (KC6) from rabbit kidney and the identification of one isoform that is highly expressed in rabbit distal tubule cells grown in culture. A partial-length cDNA (859 bp) for KC6 was isolated by PCR amplification of rabbit kidney cDNA using Shaker-specific degenerate primers. KC6 was most similar to the rat brain clone RBK2 (77% amino acid identity) and to the rabbit clone KC19 (78% amino acid identity). Transcript levels for KC2, 4, 6, 10, 19, and 22 were quantified using the ribonuclease protection assay. Transcripts for all six isoforms were detected in renal tissues. KC22 was the most abundant isoform in kidney cortex and medulla (20- to 40-fold greater than the other isoforms). Furthermore, KC22 expression levels were fivefold higher in primary cultures of rabbit distal convoluted tubules and connecting tubules than in whole kidney cortex. Although the partial-length sequence for KC22 represents the most conserved regions in the Shaker gene family it only has 35-88% amino acid identity with other Shaker channels, suggesting that KC22 represents a novel isoform. In contrast, KC4 and KC19 (less abundant in kidney than KC22) are highly homologous to the rat brain clones RBK1 and RBK2, respectively (97% amino acid identity).(ABSTRACT TRUNCATED AT 250 WORDS)

Immunocytochemical characterization of the high-affinity thiazide diuretic receptor in rabbit renal cortex.

Thiazide diuretics increase urinary NaCl excretion primarily by inhibiting Na and Cl transport across the apical membrane of cells in the renal distal tubule. Although these diuretics bind to a membrane protein that couples transport of Na and Cl directly, the molecular nature of this transporter and its localization in the mammalian kidney remain controversial. The present experiments were designed to develop monoclonal antibodies to the high-affinity thiazide diuretic receptor to investigate its molecular characteristics and its cellular and subcellular localization in rabbit kidney. Mice were immunized with high-affinity thiazide diuretic receptors that had been partially purified from rabbit kidney cortex. Resulting hybridomas were screened for the ability to immunoprecipitate thiazide diuretic receptors that were labeled with the thiazide-like diuretic [3H]metolazone. A single hybridoma (MAb JM5) produced antibodies capable of immunoprecipitating up to 80% of the labeled thiazide receptors from solubilized renal cortical membranes. MAb JM5 reacted with a 125-kDa protein on Western blots of solubilized renal cortical apical membranes. It stained the apical membrane of cells in the distal convoluted and connecting tubule but did not stain proximal tubules, glomeruli, or interstitial structures. Less intense staining of apical membranes of principal cells in the collecting tubule and a subpopulation of cells in the thick ascending limb were also present. These results indicate that the high-affinity thiazide diuretic receptor comprises a 125-kDa protein that localizes to the apical membrane of cells in the renal distal tubule.

Inhibition of Ca-activated K+ channels from renal microvillus membrane vesicles by amiloride analogs.

The effect of the K(+)-sparing diuretic amiloride and two of its hydrophobic analogs, methylisobutyl amiloride (MIA) and ethylisopropyl amiloride (EIPA), on Ca-activated K+ channels from renal microvillus membrane vesicles incorporated into planar lipid bilayers was investigated. Amiloride did not inhibit currents through Ca-activated K+ channels. MIA and EIPA, however, inhibited channel currents when added to both the internal and external solutions in concentrations between 10 and 250 microM. Furthermore, when dose-response data for channel inhibition were examined using Hill plots, Hill numbers of approximately 1.5 were found for both blockers from both sides, suggesting that the mechanism of block involves multiple inhibitory binding sites. A simple kinetic scheme is proposed that can account for the results.

Isolation of putative voltage-gated epithelial K-channel isoforms from rabbit kidney and LLC-PK1 cells.

Epithelial voltage-gated potassium (K) channels have been well studied using electrophysiological methods, but little is known about their structures. We tested the hypothesis that some of these channels belong to the Shaker gene family, which encodes voltage-gated K channels in excitable tissues. From published sequences of Shaker proteins in Drosophila, rat, and mouse brain, we chose regions that were conserved between species. Based on these protein sequences, degenerate oligonucleotides flanking the putative voltage sensor (S4) were synthesized and used as primers for the polymerase chain reaction. Five Shaker-like cDNAs were amplified from rabbit kidney cortex and three from LLC-PK1, an epithelial cell line derived from pig kidney. Each partial-length rabbit kidney cDNA is approximately 850 base pairs (bp) long. The deduced amino acid sequences contain five putative transmembrane segments and are 79-97% identical to two Shaker isoforms expressed in rat brain (RBK1 and RBK2). Sequence similarity is greatest in the putative transmembrane segments S1-S5. Importantly, the S4 segment, the putative voltage gate is highly conserved in all 5 cDNAs. Southern analysis of rabbit genomic DNA suggests that each isoform is encoded by a different gene. The partial length LLC-PK1 cDNAs are 450-bp long, and the deduced amino acid sequences are 77-99% identical to the rabbit cDNAs. This is, to our knowledge, the first demonstration that Shaker-like genes are expressed in renal epithelial cells. These genes most likely encode voltage-gated K channels involved in renal epithelial K transport.

Solubilization and partial purification of the thiazide diuretic receptor from rabbit renal cortex.

This study was designed to solubilize, characterize and begin to purify the thiazide-sensitive Na/Cl transporter from mammalian kidney. Metolazone, a thiazide-like diuretic drug, binds to receptors in rat renal cortex closely related to the thiazide-sensitive Na/Cl transport pathway of the renal distal tubule. In the current study, [3H]metolazone bound to receptors in rabbit renal cortical microsomes. The portion of [3H]metolazone binding that was inhibited by hydrochlorothiazide reflected binding to a high-affinity class of receptor. The affinity (Kd 2.0 +/- 0.1 nM) and number (Bmax = 0.9 +/- 0.4 pmol/mg protein) of high-affinity receptors in rabbit renal cortical membranes were similar to values reported previously for rat. When proximal and distal tubule fragments were separated by Percoll gradient centrifugation, receptors were restricted to the fraction that contained distal tubules. When compared with cortical homogenates, receptor density was enriched 12-fold by magnesium precipitation and differential centrifugation. The zwitterionic detergent CHAPS solubilized 25-35% of the receptors (at 6 mM). Chloride inhibited and Na stimulated binding of [3H]metolazone to solubilized high-affinity receptors. The receptors could be purified significantly by hydroxyapatite chromatography and size exclusion high performance liquid chromatography (HPLC). The combination of magnesium precipitation and differential centrifugation, hydroxyapatite chromatography, and size exclusion HPLC resulted in a 213-fold enrichment of receptors, compared to renal cortical homogenate. The current results indicate that thiazide receptors from rabbit kidney share characteristics with receptors from rat, and that rabbit receptors can be solubilized in CHAPS and purified significantly by hydroxyapatite chromatography and size exclusion HPLC.

Reconstitution and partial purification of an amiloride-sensitive, cation channel from rabbit kidney.

The aim of the present study was to reconstitute and purify an epithelial potassium channel from rabbit kidney. Renal brush border membrane vesicles (BBMV) were found to contain a potassium conductance which was inhibited by amiloride, 5-(N-methyl-N-isobutyl)amiloride (MIA) and by barium. Membrane vesicle proteins were solubilized and reconstituted in proteoliposomes. Channel activity was assayed using Acridine orange and the voltage sensitive dye, 3,3'-diethylthiadicarbocyanine iodide (DiSC2(5)). Both methods yielded similar results which indicated the presence of an amiloride-sensitive, cation channel in the proteoliposomes. This channel was more permeable to K than to Na and its activity was increased in reconstituted proteoliposomes as compared to native brush border membranes. We conclude that rabbit BBMV possess an amiloride sensitive cation channel. Channel activity was successfully reconstituted in proteoliposomes and the protein was partially purified during reconstitution.