METHODS

Male Sprague-Dawley rats (200–250 g body wt; Charles River Laboratories, Wilmington, MA) were divided into three experimental groups: the control group was fed normal rat food that contained 4.4 g Na/kg and 9.5 g K/kg. The Na-depleted group was given a Na-free diet for 1 wk. The K-depleted group was given a K-free diet (0.6 mg K/kg) for 3 wk. All animals were allowed free access to water.

At the end of the experimental diet periods, the animals were killed and proximal colon, distal colon, ileum, jejunum, stomach, kidney, brain, lung, testis, liver, spleen, and heart from normal rats; proximal and distal colon, testis, and lung from K-depleted rats; and proximal and distal colon from Na-depleted rats were immediately removed and washed with diethyl pyrocarbonate-treated sterilized saline. Colonocytes from proximal and distal colon were obtained, as previously described (33), and all tissues were then homogenized using a Polytron homogenizer in 4 M guanidine isothiocyanate buffer for 60 s and centrifuged at 3,000 rpm for 10 min to remove any unbroken cells. Total RNA was prepared from the supernatant by ultracentrifugation through CsCl (32). Poly(A)⁺ mRNA was isolated by passing total RNA through oligo(dT) cellulose columns according to the method described by Sambrook et al. (32). Total RNA and mRNA were quantitated by absorbance at 260 nm in an ultraviolet-visible double-beam spectrophotometer (Shimadzu).

Northern blot analysis.

Northern blot analyses were performed using poly(A)⁺ mRNA, as previously described (33). A ³²P-labeled full-length HKcβ probe [1 × 10⁶counts ⋅ min⁻¹ ⋅ ml⁻¹(cpm/ml)] was added to the membrane for hybridization at 42°C in a Hybaid oven for 18 h. Blots were washed for 15 min in 0.1× SSC (1× SSC is 0.15 M NaCl and 0.015 M sodium citrate, pH 7.0) and 0.1% SDS at 65°C and were exposed to Hyperfilm (Amersham, La Jolla, CA) using a Bioplus intensifying screen at 70°C.

cDNA screening and sequencing.

A rat colon cDNA library with randomly primed cDNA inserts >1.0 kb in length was cloned in pcDNAI (Invitrogen, San Diego, CA) and was used in the cloning of the β-subunit; 300,000 colonies were plated onto 20 Luria-Bertani [LB; containing 50 μg/ml ampicillin and 7 μg/ml tetracycline (amp/tet)] plates 150 mm in diameter. A microwave colony-screening protocol from Invitrogen was strictly followed. Briefly, colonies were transferred to nylon membrane (UV Duralon membrane, Stratagene, La Jolla, CA) after nylon membrane had been placed on top of the colonies in the plate. Membrane filters were lifted carefully and were laid on a fresh LB (amp/tet) agar plate by facing the colony side up; the plates were incubated with filters at 37°C for 1 h to allow colonies to grow. To regrow the colonies on the agar plates that were lifted, plates were incubated at 37°C for about 6 h and then stored at 4°C. The filters were carefully removed from the agar plate and placed colony side up on a Whatman 3 filter paper prewetted in lysis buffer (2× SSC-5% SDS, pH 7.0). The filter paper and the membrane with colonies were placed in a turntable microwave oven and heated at high power for 6 min. The filters were then placed in prehybridization solution containing 2× SSC (pH 7.0), 1% SDS, and 0.5% nonfat dry milk at 65°C for 1 h. The filters were removed from the prehybridization solution; any cellular debris was removed. The filters were then hybridized in buffer containing 6× SSC, 1% SDS, 0.5% nonfat dry milk, 100 μg denatured salmon sperm DNA, and 1 × 10⁶ cpm/ml³²P-labeled full-length β-subunit cDNA probe from a rat astrocytoma cell line (38) labeled by random primer method (32) at 65°C for at least 16 h. The filters were then washed at room temperature for 10 min by gentle shaking in a buffer containing 2× SSC (pH 7.0) and 1% SDS, followed by 1× SSC (pH 7.0) and 1% SDS; the filters were transferred to prewarmed buffer (45°C) containing 0.1× SSC (pH 7.0) and 1% SDS and washed by gentle shaking at 45°C for 15 min. The filters were air dried, covered with Saran wrap, and exposed to Hyperfilm (Amersham) using a Bioplus intensifying screen for 18 h. After development of the film, positive colonies were identified and screened two more times to obtain positive clones. Plasmids were prepared according to the method of Morelle (27) using 2 ml of “terrific broth” (32), and the plasmid DNA containing inserts was sequenced in both strands using an automated fluorescence sequencing machine at the Yale Sequencing Facility.

Computer analysis of the nucleotide sequences was performed using Blast search [Genetics Computer Group (GCG) program, University of Wisconsin]. Hydropathy analysis was performed by the procedure of Kyte and Doolittle (19). Multiple sequence alignment of β-subunits was performed by the Pileup program (GCG software) (9).

Antibody production.

The cDNA sequence (165 bp) corresponding to the amino acid sequence between Pro-87 and Ser-142 of HKcβ was amplified using the full-length colon-derived HKcβ as a template with a sense primer (5′-GGGGGGATCCCACCGACTGCCTTGGATTATACATA-3′) to which a BamH I site was appended at the 5′ end and an antisense primer (5′-GGGGGGAATTCACTATAGTCTGGACCCTCCTG-3′) to which an EcoR I site was appended at the 5′ end. The expected size 165-bp fragment was gel purified, digested with BamH I andEcoR I, and ligated using T4 DNA ligase into PGEX-KG vector that had previously been digested withBamH I andEcoR I and dephosphorylated with calf intestine phosphatase. The ligated DNA was transformed into XL-1 blueEscherichia coli cells and spread onto LB-ampicillin plates. Plasmid DNAs were prepared from recombinant colonies, digested with BamH I andEcoR I, and electrophoresed onto 1.5% agarose gels. Positive clones containing the 165-nucleotide HKcβ fragment were selected and sequenced. Clones with correct reading frame and with correct cDNA sequence were selected to prepare glutathioneS-transferase (GST)-HKcβ fusion protein. The recombinant colonies were grown in LB-ampicillin medium to an absorbance of 0.8 at 600 nm; the GST-HKcβ fusion proteins were then overexpressed after induction with 1.0 mM isopropyl β-d-thiogalactopyranoside at 37°C for 3 h. After induction, the cells were harvested, washed with 50 mM Tris ⋅ HCl (pH 7.4) containing 10 mM EDTA, and resuspended in the lysis buffer containing 50 mM Tris ⋅ HCl (pH 7.4), 100 mM NaCl, 5 mM dithiothreitol, 2 mM EDTA, 2 mM EGTA, and 1 mg/ml lysozyme, mixed 45 min at 4°C, frozen at 80°C, and thawed. Then 1% Triton X-100, 10 mM MgCl₂, and 0.1 mg/ml DNase I were added and incubated at 4°C for an additional 30 min. After centrifugation, the GST-HKcβ fusion protein was purified from the supernatant by passage through a glutathione-agarose column; after the column was washed separately with 1× PBS and 1× PBS containing 1% Triton X-100 and with 50 mM Tris ⋅ HCl (pH 8.0) containing 1 mM EDTA, the GST-HKcβ fusion protein was eluted from the column using 10 mM reduced glutathione (pH 8.0) by incubating at 4°C with gentle nutation. The eluted GST-HKcβ fusion protein was dialyzed against 1× PBS at 4°C and concentrated using Centriprep 10. At this stage, the purified GST-HKcβ fusion proteins were run on SDS-PAGE gels and analyzed for homogeneity. The homogeneous preparation of GST-HKcβ fusion proteins was used to inject rabbits to raise polyclonal antibodies.

Antibodies were produced in a New Zealand White rabbit following primary subcutaneous injection of purified GST-HKcβ fusion protein in complete Freund’s adjuvant; subsequent boosts of fusion protein were injected in incomplete Freund’s adjuvant. The animal was killed and exsanguinated on day 11 after the last boost. Serum IgG was affinity purified using a protein A column. Antibodies to GST proteins were also removed by glutathione-Sepharose affinity chromatography. The purified antibody to HKcβ protein was used for Western blot analyses.

Specificity of HKcβ antibody.

The specificity of the HKcβ antibody was established by the following experiments. 1) In an immunodepletion experiment, GST-HKcβ fusion protein and HKcβ antibody were mixed and incubated at room temperature. The antigen-antibody complex mixture was then used as an antibody source for the Western blot analysis, which contained apical membranes, basolateral membranes, or the immunoprecipitate of apical membranes by HKcα antibody. 2) HKcβ cDNA and NaKβ₁ cDNA were subcloned into pcDNA 3.1 (+), which was transfected into COS-7 cells (13). The expressed HKcβ and NaKβ₁proteins were used for Western blot analysis with HKcβ antibody.3) Highly purified Na-K-ATPase from rabbit kidney (18) was run on SDS-PAGE gels and transferred to nitrocellulose membranes, and Western blot analysis was performed using HKcβ or NaKβ₁ antibodies.

Isolation of apical and basolateral membranes.

Apical and basolateral membranes were isolated by methods previously described in detail (29, 30).

Western blot analysis.

Western blot analyses were performed with HKcα, HKcβ, and NaKβ₁ antibodies using previously described methods (20, 33). Apical or basolateral membrane proteins (50 μg) were electrophoresed on SDS-PAGE gels and transferred to nitrocellulose membranes. Western blot analysis was then performed with dilution of 1:1,000 HKcβ antibody in Tris-buffered saline-Tween containing 5% nonfat dry milk; anti-rabbit IgG horseradish peroxidase conjugate (1:5,000 dilution) was used as the secondary antibody. HKcβ antibody-specific protein bands were visualized by the enhanced chemiluminescence procedure.

Coimmunoprecipitation.

One hundred micrograms of apical membrane proteins of rat distal colon were resuspended in 1 ml of immunoprecipitation buffer containing 10 mM Tris ⋅ HCl (pH 7.4), 150 mM NaCl, 1 mM EDTA, 1 mM EGTA, 0.2 mM phenylmethylsulfonyl fluoride, and 1% Triton X-100. One microliter of HKcα antibody was added and incubated at room temperature for 2 h in a nutator. To the above mixture, 50 μl of protein A-Sepharose (50% suspension) were added to bind antigen-antibody complex and incubated at room temperature for 1 h in a nutator. After a 1-min centrifugation, the pellet was washed three times for 5 min each in the above immunoprecipitation buffer containing 5% nonfat dry milk. The pellet was then washed three times with immunoprecipitation buffer. Finally, the pellet was resuspended in 2× SDS sample buffer containing 2-mercaptoethanol, boiled for 5 min, and centrifuged briefly, and the supernatant was loaded onto SDS-PAGE gels. After electrophoresis, proteins were transferred onto a nitrocellulose membrane and Western blot analysis was performed with HKcβ antibody diluted to 1:1,000.

RESULTS

Isolation and characterization of HKcβ cDNA clones.

We screened a rat colon cDNA library using a full-length novel β-subunit cDNA isolated from a rat astrocytoma cell line (38) as a probe and obtained positive clones that were sequenced in both directions. The full-length cDNA of HKcβ consists of 1,728 nucleotides with an open reading frame encoding 279 amino acids, an initiation codon ATG, an in-frame termination codon (TAG), a 78-nucleotide 5′ untranslated region, and an 810-nucleotide 3′ untranslated sequence. The sequence of this cDNA is identical to that previously cloned by Watanabe et al. (38) from a rat astrocytoma cell line.

A comparison of the predicted amino acid sequence of the rat HKcβ with those of the rat NaKβ₁, NaKβ₂, and HKgβ subunits is shown in Fig. 1. The predicted amino acid sequence of HKcβ exhibits 100% identity to the astrocytoma β-subunit probe we used for screening (38), 36.4% identity to rat HKgβ (34), 36.5% identity to rat NaKβ₁ (26), and 49% identity to rat NaKβ₂ (23), all previously characterized rat β-subunits. In addition, the HKcβ coding region has 80% identity to human NaKβ₃and 100% identity to the recently cloned partial β₃ cDNA from rat placenta (22). The Asn residues at positions Asn-124 and Asn-158 are potential sites of N-linked glycosylation that are not conserved with the other rat β-subunits. The six Cys residues (at positions Cys-128, Cys-144, Cys-154, Cys-170, Cys-191, and Cys-250) that might interact with an α-subunit are present in the extracellular domain of the HKcβ subunit and are highly conserved with rat HKgβ, NaKβ₁, and NaKβ₂ subunits. Hydropathy analysis (not shown) by the Kyte and Doolittle (19) algorithm predicts that HKcβ contains a charged cytoplasmic amino terminus, a single transmembrane domain, and a large extracellular carboxy-terminal domain. The predicted secondary structure of HKcβ is similar to those of rat HKgβ, NaKβ₁, and NaKβ₂ subunits. The hydrophobic residue-rich region of HKcβ is between amino acids Trp-36 and Leu-62. These observations establish substantial structural similarities between HKcβ and other P-type ATPase β-subunits.

Expression of HKcβ subunit mRNA in rat tissues.

Expression of HKcβ mRNA was analyzed in several tissues by Northern blot analysis (Fig. 2). HKcβ cDNA probe hybridized with a transcript of ∼1.9 kb in distal colon, proximal colon, ileum, jejunum, stomach, liver, lung, kidney, brain, testes, spleen, and heart. It should be noted that HKcβ cDNA that was cloned from colon is 1.73 kb. HKcβ message was highly expressed in testis and lung; the message was slightly smaller in size in testis than those identified in other tissues.

Regulation of HKcβ expression.

The identification of a colon-derived putative β-subunit isoform permitted studies to evaluate the role of HKcβ in the transport function of the rat distal colon. Because HKcβ is expressed in rat distal colon, it could be associated with the regulation of K transport and/or Na transport, which are closely linked to colonic H-K-ATPase and Na-K-ATPase, respectively (2). Expression of HKcβ was analyzed by Northern blot with mRNA isolated from distal colon of normal, dietary Na-depleted, and dietary K-depleted rats (Fig. 3). Na depletion stimulates Na and K absorption and K secretion (2, 37), whereas K depletion enhances only K absorption in distal colon (12). Figure3A demonstrates that HKcβ mRNA expression is increased in distal colon of K-depleted rats compared with normal rats. HKcβ message expression was normalized to glyceraldehyde-3-phosphate dehydrogenase expression and quantitated by densitometric analysis. K depletion resulted in a 3.5-fold increase in HKcβ message expression compared with controls (Fig.3B). In contrast, HKcβ mRNA abundance in Na-depleted rats was not significantly altered (Fig. 3), although both dietary K depletion and Na depletion induce comparable increases in active K absorption in the rat distal colon (11, 12). In parallel studies with mRNA isolated from proximal colon, an organ in which active K absorption is not present (11), dietary K depletion did not increase HKcβ mRNA abundance (Fig.4).

As shown in Fig. 2, HKcβ mRNA expression was highest in testis and lung. Therefore, additional Northern blot analyses were performed to determine whether dietary K depletion increased HKcβ mRNA abundance in these two organs. Figure 5 demonstrates that dietary K depletion did not alter HKcβ mRNA expression in either lung or testis. In both tissues, two distinct bands with equal intensity were identified; the smaller bands may represent alternatively spliced products of HKcβ mRNA. It should be noted that HKcβ cDNA probe hybridizes with slightly smaller mRNA species in testis than in lung. Therefore, the selective increase in HKcβ mRNA expression in distal colon of K-depleted rats (Fig. 3) suggests that this β-subunit may be associated with the colonic H-K-ATPase α-subunit.

To assess the significance of the increase in HKcβ mRNA by K depletion, the expression of NaKβ₁ mRNA abundance in distal and proximal colon of normal, Na-depleted, and K-depleted rats was also analyzed by Northern blot; these results are presented in Fig.6. Expression of NaKβ₁ mRNA in distal colon of K-depleted rats did not change compared with control animals. In contrast, NaKβ₁ mRNA abundance was modestly increased in Na-depleted rats, an observation consistent with the stimulation of active Na absorption in the distal colon by Na depletion (37). Northern blot analyses were performed to determine the presence of other β-subunit mRNAs in rat distal colon. Neither HKgβ nor NaKβ₂ cDNA probes hybridized with mRNA prepared from rat distal colon (data not shown).

Expression of HKcβ in apical and basolateral membranes.

The results of the Northern blot analyses (Fig. 3) are consistent with the thesis that HKcβ is the β-subunit required for colonic H-K-ATPase function. HKcβ has significant (80%) homology to a recently identified human β-subunit that was designated NaKβ₃ (18). This assignment of NaKβ₃ was based on chromosomal localization and sequence relatedness. As a consequence, additional studies were performed with HKcβ to establish whether HKcβ is present in apical and/or basolateral membranes and whether HKcβ is associated with HKcα.

Several studies were performed that established the specificity of HKcβ antibody to HKcβ protein without evidence of cross-reactivity to NaKβ₁ protein.1) The specificity of HKcβ antibody was initially confirmed by Western blot analysis against the GST-HKcβ fusion protein (data not shown).2) Immunodepletion experiments were performed in which GST-HKcβ fusion protein and HKcβ antibody complex mixture was used as an antibody source for Western blot analysis. Figure7Ademonstrates that no protein was identified in apical (lane 1) or in basolateral (lane 2) membranes, but the immunoglobulin band (lane 3) that was present in the original blot (Fig.8B,lane 3) was identified.3) Expression studies of HKcβ cDNA in COS-7 cells demonstrated that HKcβ protein was identified by the HKcβ antibody only in cells transfected with HKcβ cDNA (Fig.7B,lane 3). HKcβ antibody did not cross-react with NaKβ₁ protein in cells transfected with NaKβ₁cDNA (Fig. 7B,lane 4).4) Highly purified Na-K-ATPase protein from rabbit renal medulla that consists of NaKα₁ and NaKβ₁ proteins was not identified by HKcβ antibody (Fig.7C) but was identified by NaKβ₁ antibody (Fig.7D). These several observations establish the specificity of HKcβ antibody to HKcβ protein.

Figure 8 presents a Western blot analysis of apical and basolateral membranes using antibodies to HKcα, HKcβ, and NaKβ₁ and preimmune serum. HKcα was predominantly expressed in apical membranes (Fig.8A,lane 1), confirming previous studies of the localization of HKcα protein to the apical membrane of surface cells of rat distal colon (20). The very low expression of HKcα protein in basolateral membrane (Fig.8A,lane 2) probably reflects minimal contamination of the basolateral membrane preparation by apical membranes. Figure 8B demonstrates that HKcβ protein was expressed in both apical (lane 1) and basolateral (lane 2) membranes in approximately equal amounts. In contrast, NaKβ₁protein was expressed only in basolateral membranes (Fig.8C,lane 2). The presence of HKcβ and not NaKβ₁ in apical membrane supports a possible role of HKcβ as the β-subunit for HKcα. The specificity of HKcβ protein expression in rat colonic membranes was provided by the immunodepletion experiment shown in Fig.7A. This observation confirms that the protein band at 45 kDa in apical and basolateral membranes is specific to HKcβ antibody.

Additional Western blot analyses were performed on apical and basolateral membranes from normal and K-depleted rats using HKcβ antibody. Figure 9 presents the results of two of the three separate preparations of apical and basolateral membranes. Each preparation was prepared from at least six normal or K-depleted rats. HKcβ protein expression was significantly increased by 216% in apical membranes but was reduced by 30% in basolateral membranes (Fig. 9). The results presented in Figs. 3 and 9 suggest that K depletion regulates HKcβ mRNA and protein.

Identification of physical interaction of HKcβ and HKcα.

Experiments were designed to establish whether HKcβ protein was associated with HKcα protein in the apical membrane of distal colon. Therefore, apical membrane proteins from rat distal colon were immunoprecipitated with HKcα antibody (seeCoimmunoprecipitation). The resulting antigen-antibody complex was analyzed by Western blot using HKcβ or NaKβ₁ antibodies. Figure 8B(lane 3) demonstrates the presence of HKcβ protein in the immunoprecipitate, demonstrating the physical interaction of HKcα and HKcβ proteins. In contrast, NaKβ₁ was not identified in the immunoprecipitate (Fig. 8C,lane 3). The specificity of the bands of the immunoprecipitate was confirmed by performing a Western blot analysis of the protein blot in Fig.8C using preimmune serum following stripping. The preimmune serum identified only the immunoglobulin, as shown in Fig. 8D,lane 3. In addition, the immunodepletion experiments (Fig. 7A) established the specificity of HKcβ antibody to HKcβ protein. These observations are consistent with HKcβ functioning as the β-subunit for the apical membrane localized α-subunit of colonic H-K-ATPase.

DISCUSSION

Active K absorption and an apical membrane H-K-ATPase have been studied extensively in the rat large intestine. Active K absorption is restricted to the distal colon and is upregulated both by dietary K depletion and by aldosterone and Na depletion (11, 12). It is generally accepted that active K absorption is energized by apical membrane H-K-ATPase, the α-subunit of which (HKcα) was recently cloned (7,14) and has been studied extensively (5, 6, 15, 20, 33). HKcα message is restricted to surface (and the upper 20% of the crypt) cells of the distal colon, whereas its protein is localized to the apical membrane of surface cells of the rat distal colon (14, 20). Although several β-subunit isoforms of both Na-K-ATPase and noncolonic H-K-ATPase have been isolated, a colonic H-K-ATPase β-subunit had not previously been identified.

Recent studies have established that dietary Na depletion upregulates HKcα message and protein abundance and apical membrane H-K-ATPase activity in the rat distal colon but does not alter HKcα message and protein expression in the kidney (15, 33). In contrast to the effect of Na depletion on HKcα expression in the colon, dietary K depletion did not affect HKcα message and protein expression in the rat distal colon (33). Of potential importance, HKcα protein expression was enhanced in the principal cell in the kidney of dietary K-depleted rats compared with control rats. Thus the mechanism by which dietary K depletion increases active K absorption in the rat distal colon is not known. Lescale-Matys et al. (21) demonstrated in LLC-PK₁ cells that a decrease in extracellular K concentration was associated with an increase in NaKβ₁ message and protein but not in NaKα₁ message and protein. Their study concluded that regulation of Na-K-ATPase activity by K concentration was pretranslational and that β-subunit synthesis was rate limiting. These observations with LLC-PK₁ cells parallel the present results with rat distal colon. Dietary K depletion enhances active K absorption but is associated with an increase in HKcβ (Fig. 3) but not HKcα mRNA abundance (33). As a result, we speculate that the mechanism of regulation of H-K-ATPase function by dietary K depletion is mediated via HKcβ and not via HKcα.

The data presented in Fig. 3 are consistent with the role of HKcβ mRNA in the regulation of H-K-ATPase by dietary K depletion. These Northern blot analyses using the newly cloned colon-derived β-subunit establish that HKcβ mRNA abundance was enhanced in dietary K-depleted rats. Specificity of this observation is provided by four important experiments. First, dietary K depletion did not increase the mRNA abundance of the NaKβ₁ subunit (Fig. 6), indicating that dietary K depletion does not result in a nonspecific increase in message abundance of all β-subunits. Second, HKcβ mRNA abundance was not increased in Na-depleted rats (Fig. 3), indicating that the increase in HKcβ message was specific and was not merely secondary to an increase in active K absorption. Third, although HKcβ message is present in the proximal colon (Fig. 4), a tissue in which active K absorption is not present (11), HKcβ mRNA abundance was not increased in dietary K-depleted animals in proximal colon. Fourth, expression of HKcβ mRNA abundance was not altered in lung and testis of K-depleted rats compared with normal rats (Fig. 5). Thus the effect of K depletion on HKcβ mRNA is tissue specific, as HKcβ mRNA is upregulated in the distal but not in the proximal colon, testis, or lung of K-depleted rats. These observations provide compelling evidence that HKcβ is closely linked to the stimulation of active K absorption by dietary K depletion in the rat distal colon.

Controversy exists regarding the role of a β-subunit in the functional expression of HKcα. Although HKcα was expressed as ouabain-insensitive H-K-ATPase in Sf9 cells without the apparent presence of a β-subunit (20), the expression of HKcα in oocytes requires the coinjection of cRNAs of aBufo marinus urinary bladder β-subunit (6) or of HKgβ or NaKβ₁ (5). It is of interest that the amphibian β-subunit (17) that induced⁸⁶Rb uptake in oocytes when coinjected with HKcα (6) has significant homology (52.3%) to HKcβ. The demonstration that dietary K depletion resulted in an increase in HKcβ mRNA abundance but not an enhancement of HKcα mRNA abundance strongly suggests that the regulation of active K absorption by dietary K depletion is mediated by this β-subunit and not by HKcα. Because Na depletion stimulates HKcα but not HKcβ message abundance in the rat distal colon (Refs. 15, 33 and Fig. 3), differential regulation of HKcα and HKcβ subunits plays a significant role in the stimulation of active K absorption in rat distal colon by K depletion and Na depletion. In addition, because dietary K depletion increased the abundance of HKcβ message but not that of NaKβ₁, these observations suggest that HKcβ may be the specific β-subunit required for optimal H-K-ATPase function in the distal colon.

The close identity between HKcβ and the recently reported human putative NaKβ₃ (2, 22) requires comment. The latter cDNA sequence was identified from the human-expressed sequence tag data bank and was designated NaKβ₃ based solely on chromosomal localization and sequence relatedness, but without demonstration of membrane localization or physical association with an α-subunit. In addition, this human putative NaKβ₃ has an amino acid sequence with a 55% identity to a recently cloned β-subunit from amphibian bladder (17), which, when coexpressed in Xenopusoocytes with an amphibian bladder H-K-ATPase α-subunit, stimulated H/K and not Na/K function (16). Because HKcβ protein has 52% identity to the amphibian bladder β-subunit, HKcβ may be the rat homologue of this amphibian bladder β-subunit. Thus it is possible that the human putative NaKβ₃ may also manifest H-K-ATPase function.

The identification of HKcβ protein in basolateral membranes (see Fig.8B,lane 2) raised the possibility that HKcβ might also function with α-subunits of Na-K-ATPase and that therefore the upregulation of HKcβ mRNA by dietary K depletion (Fig.3) was causally related to an increase in Na-K-ATPase activity by dietary K depletion. Such a possibility would be consistent with the previous observations that dietary K depletion and incubation of LLC-PK₁ cells in a low-K medium are associated with increases in NaKβ₁ subunit and/or Na-K-ATPase activity (21). Therefore, additional experiments were performed to determine whether dietary K depletion was associated with an increase in Na-K-ATPase activity in rat distal colon. These studies demonstrated that Na-K-ATPase activity in basolateral membranes isolated from rat distal colon was not altered by dietary K depletion (unpublished observations). Thus the observed increase in HKcβ mRNA abundance in dietary K depletion cannot be responsible for an increase in Na-K-ATPase. Therefore, HKcβ is uniquely regulated by dietary K depletion in the distal colon and likely is the β-subunit required for H-K-ATPase in the rat distal colon.

H-K-ATPase is localized to the apical membrane in the distal colon (8), in contrast to the localization of Na-K-ATPase to the basolateral membrane. The Western blot studies presented in Fig. 8 present three important observations that indicate HKcβ is the β-subunit for the colonic H-K-ATPase. First, HKcβ protein was expressed in apical membrane (Fig. 8B,lane 1), the site at which HKcα protein is selectively expressed (see Fig. 8A,lane 1). Second, although HKcβ protein was identified in both apical and basolateral membranes (Fig.8B,lanes 1 and2), K depletion resulted in an increase in HKcβ protein expression only in apical membranes (Fig.9). The selective increase in HKcβ protein in the apical membrane requires comment. It is possible that the protein recognized by the HKcβ antibody in the basolateral membrane is not HKcβ protein but a closely related β-subunit that was recognized by the HKcβ antibody but not regulated by K depletion. The mechanism for the 30% decrease of HKcβ protein in basolateral membrane of K-depleted rats is not known. Third, coimmunoprecipitation experiments demonstrated the physical association between HKcβ and HKcα proteins in the apical membrane (Fig. 8B,lane 3). In addition, the preimmune serum (Fig. 8D,lane 3) and immunodepletion experiment (Fig. 7A,lane 3) did not identify any protein bands except immunoglobulins. These observations confirm that HKcβ protein bands are specific, and the two additional low-molecular weight protein bands in the immunoprecipitate may be a HKcβ degradation product. In contrast, an antibody to NaKβ₁ identified protein in basolateral but not in apical membranes (Fig.8C,lane 2). As a result, it is unlikely that NaKβ₁ protein is associated with HKcα protein or is the β-subunit for H-K-ATPase. The demonstration of the selective increase in HKcβ protein expression in apical membrane of rat distal colon and the association of HKcβ protein with HKcα protein in apical membrane (Fig.8B,lane 3) strongly support the possibility that HKcβ is most likely the β-subunit for colonic H-K-ATPase α-subunit.

Although dietary K depletion unequivocally stimulates active K absorption in the large intestine of both rat (12) and mouse (25), enhancement of H-K-ATPase in dietary K depletion has been inconstant. It is generally believed that active K absorption in the distal colon is a result of an H/K exchange energized by an apical membrane H-K-ATPase (2), but the effect of dietary K depletion on colonic active K absorption and H-K-ATPase activity is complex. Confounding variables of the presumed stimulation of H-K-ATPase by dietary K depletion include its duration and the presence of two H-K-ATPase isoforms and two components of active K absorption. The two H-K-ATPases have different sensitivities to ouabain and spatial distributions (8, 31), and the two active K absorptive processes also have different sensitivities to ouabain, as well as different responsiveness to aldosterone (28). Additionally, direct demonstration of an H/K exchange in colonic apical membranes has not as yet been established, and Feldman and Ickes (10) presented evidence of a dissociation between active K absorption and protein secretion.

In conclusion, we have identified a β-subunit from rat distal colon, and its mRNA is upregulated in a tissue-specific manner in the distal colon of K-depleted but not of Na-depleted rats. This β-subunit protein localized in the apical membrane, physically associated with HKcα protein. Its protein expression is selectively increased in the apical membrane of K-depleted rats. These observations are all consistent with the probability that HKcβ is the putative β-subunit for colonic H-K-ATPase.