Ouabain-sensitive H,K-ATPase Functions as Na,K-ATPase in Apical Membranes of Rat Distal Colon*

Na,K-ATPase activity has been identified in the apical membrane of rat distal colon, whereas ouabain-sensitive and ouabain-insensitive H,K-ATPase activities are localized solely to apical membranes. This study was designed to determine whether apical membrane Na,K-ATPase represented contamination of basolateral membranes or an alternate mode of H,K-ATPase expression. An antibody directed against the H,K-ATPase a subunit (HKc a ) inhibited apical Na,K-ATPase activity by 92% but did not alter basolateral membrane Na,K-ATPase activity. Two distinct H,K-ATPase isoforms exist; one of which, the ouabain-insensitive HKc a , has been cloned. Because dietary sodium depletion mark-edly increases ouabain-insensitive active potassium absorption and HKc a mRNA and protein expression, Na,K-ATPase and H,K-ATPase activities and protein expression were determined in apical membranes from control and sodium-depleted rats. Sodium depletion substantially increased ouabain-insensitive H,K-ATPase activity and HKc a protein expression by 109–250% but increased ouabain-sensitive Na,K-ATPase and H,K-ATPase activities by only 30% and 42%, respectively. These studies suggest that apical membrane Na,K-ATPase activity is an alternate mode of ouabain-sensi-tive H,K-ATPase and does not solely represent basolateral membrane contamination. H,K-ATPase is present in the apical membranes of the mammalian distal colon and is closely linked to ATPase activity measured in the presence of 1 m M ouabain represents ouabain-insensitive activity, whereas ouabain-sensitive activity was calculated by subtracting ouabain-insensitive activity from total activ- ity. Preliminary studies indicated that maximal inhibition was achieved at 1 and 3 m M ouabain. Thus, the present study used 1 m M ouabain to distinguish the ouabain-sensitive and ouabain-insensitive components of H,K-ATPase. The specific activities were expressed as nanomoles of P i liberated per milligram of protein per minute. Results presented represent mean 6 S.E. of triplicate assays from at least three different membrane preparations. en- hanced chemiluminescence procedure (Amersham Pharmacia Biotech). The resultant autoradiographs were used to quantitate HKc a and NaK a 1 protein abundance in colonic membranes from normal and sodium-depleted animals using a personal densitometer SI with Image- Quant software (Molecular Dynamics, Sunnyvale, CA). activities ATPase activities were measured either in presence of Mg 2 1 alone or in the presence of Mg 2 1 plus K 1 , or Mg 2 1 plus Na 1 and K 1 . Total H,K-ATPase activity represents ATPase activity in the presence of Mg 2 1 plus K 1 , minus that in the presence of Mg 2 1 alone. Total Na,K-ATPase activity the difference ATPase activity in Mg 2 1 plus Na and K , and H,K-ATPase activity. ATPase activities were also measured in the presence of 1 m M ouabain. Activity in the presence of ouabain represents ouabain-insensitive ATPase activity. Ouabain-sensitive H,K-ATPase was determined by subtracting ouabain-insensitive activity from total activity. Mg 2 1 -ATPase activity was 221.7 13.8 and 202.6 6 9.6 nmol of P i liberated/mg protein z min in apical and basolateral membranes, respectively. Results represent S.E.

H,K-ATPase is present in the apical membranes of the mammalian distal colon and is closely linked to active potassium absorption that most likely is the result of an apical membrane H-K exchange (1). The colonic H,K-ATPase is a member of the gene family of P-type ATPases, which include ouabain-sensitive Na,K-ATPase, gastric parietal cell H,K-ATPase that is ouabain-insensitive, and ATP1AL1 (2)(3)(4)(5)(6). Colonic H,K-ATPase consists of ␣ (HKc␣) and ␤ (HKc␤) 1 subunits, which have been identified and characterized, and is up-regulated by dietary sodium depletion and aldosterone (7)(8)(9).
Previous studies established that, although colonic H,K-ATPase is exclusively present in apical membranes of the rat distal colon (10), there are two distinct H,K-ATPases. One is ouabain-sensitive; the other is ouabain-insensitive (11). Subsequent investigations revealed that both ouabain-sensitive and ouabain-insensitive H,K-ATPase activities were present in ap-ical membranes isolated from surface epithelial cells, whereas only ouabain-sensitive H,K-ATPase activity and potassium-dependent intracellular alkalinization were identified in apical membranes of crypt epithelial cells (12). Because HKc␣ mRNA and protein are selectively present only in surface (and upper 20% of crypt) epithelial cells of the rat distal colon (10,13), it is likely that HKc␣ only encodes the ouabain-insensitive H,K-ATPase and not the ouabain-sensitive H,K-ATPase. In contrast, HKc␣ cRNA induces ouabain-sensitive 86 Rb uptake and intracellular alkalinization when expressed in Xenopus oocytes (14 -16).
The studies that established the presence of H,K-ATPase in apical membranes of rat distal colon also demonstrated significant Na,K-ATPase activity in apical membranes, which was assumed to represent partial contamination of the apical membranes by basolateral membranes (11). Recently, Cougnon et al. reported that HKc␣ cRNA could express both H-K and Na-K transport functions in Xenopus oocytes (15). As a result, it is not unlikely that our prior identification of Na,K-ATPase activity in apical membranes represented an expression of one or both H,K-ATPases and not basolateral membrane contamination. The present study was designed to explore this possibility by assessing the expression and activity of both H,K-ATPase and Na,K-ATPase and HKc␣ and NaK␣1 proteins in apical and basolateral membranes of normal and dietary sodium-depleted rats. These results have been presented in a preliminary communication (17).

EXPERIMENTAL PROCEDURES
Apical Membrane Preparation-Apical membranes were prepared from the distal colon of both normal and sodium-depleted rats (Harlan Sprague-Dawley, 200 -250 g) by the method of Stieger et al. (18) as described previously (19). Sodium depletion was produced by feeding a sodium-free diet for 6 to 7 days, as described previously (20) 2 . Crude membranes were also prepared from normal rat distal colon. For crude membranes, homogenates of colonocytes were centrifuged at 30,000 ϫ g for 30 min.
Basolateral Membrane Preparation-Basolateral membranes were prepared using the sucrose density gradient centrifugation method of Biber et al. (21), also as described previously (22). All membranes were resuspended in 20 mM Tris-HCl buffer (pH 7.4) containing 250 mM sucrose and 1 mM phenylmethylsulfonyl fluoride. Protein was assayed using the method of Lowry et al. (23).
Enzyme Assays-H,K-ATPase and Na,K-ATPase activities were determined by the method of Forbush et al. (24), as described previously (11). H,K-ATPase activity represents the difference in activity between that in the presence and absence of 20 mM K. Na,K-ATPase activity represents the difference in activity between that determined in the presence of both sodium and potassium and H,K-ATPase activity. * This study was supported in part by a research grant (DK 18777-19) from the NIDDK, National Institutes of Health. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
ATPase activity measured in the presence of 1 mM ouabain represents ouabain-insensitive activity, whereas ouabain-sensitive activity was calculated by subtracting ouabain-insensitive activity from total activity. Preliminary studies indicated that maximal inhibition was achieved at 1 and 3 mM ouabain. Thus, the present study used 1 mM ouabain to distinguish the ouabain-sensitive and ouabain-insensitive components of H,K-ATPase. The specific activities were expressed as nanomoles of P i liberated per milligram of protein per minute. Results presented represent mean Ϯ S.E. of triplicate assays from at least three different membrane preparations.
Western Blot Analysis-SDS-polyacrylamide gel electrophoresis was performed using the standard protocol as described previously (7,8). In brief, to avoid aggregation, the protein samples of colonic apical and basolateral membranes were incubated at 37°C for 20 min before loading. Proteins were electrophoretically transferred from SDS-polyacrylamide gel electrophoresis to nitrocellulose membrane (Biotrace; Gelman Science Inc., Ann Arbor, MI) in 192 mM glycine, 25 mM Tris (pH 8.5), 20% methanol for 5 h at 60 V. After blocking nonspecific sites with 20 mM Tris, 137 mM NaCl, 0.1% Tween-20 buffer (pH 7.5) containing 5% nonfat dry-milk, immunostaining was performed with polyclonal antibodies to the ␣-subunits of the colonic HKc␣ (M1) at 1:1000 dilution (10) and Na,K-ATPase (NaK␣1) at 1:2000 dilution (25). Anti-rabbit IgG horseradish peroxidase conjugate (1:5000 dilution) was used as the secondary antibody for M1 antibody-stained blots, whereas anti-mouse IgG horseradish peroxidase conjugate (1:5000 dilution) was used as the secondary antibody for NaK␣1 antibody-stained blots. HKc␣ and NaK␣1 antibody-specific protein bands were visualized using the enhanced chemiluminescence procedure (Amersham Pharmacia Biotech). The resultant autoradiographs were used to quantitate HKc␣ and NaK␣1 protein abundance in colonic membranes from normal and sodium-depleted animals using a personal densitometer SI with Image-Quant software (Molecular Dynamics, Sunnyvale, CA).

RESULTS
The results of both potassium-activated Mg-ATPase (H,K-ATPase) and Na/K-activated Mg-ATPase (Na,K-ATPase) activities in apical and basolateral membranes prepared from the distal colon of normal rat are presented in Table I. Similar to previous studies (11), H,K-ATPase activity is exclusively present in apical membranes; no H,K-ATPase activity was identified in basolateral membranes. Forty-four percent of total H,K-ATPase activity was ouabain-sensitive, while the remaining activity (56%) was ouabain-insensitive. Although Na,K-ATPase is a well-established basolateral membrane marker (26), Na,K-ATPase activity was present in apical and basolateral membranes in rat distal colon (Table I). As expected, Na,K-ATPase activity in basolateral membranes was completely ouabain-sensitive. Similar to a previous study (11), Na,K-ATPase activity was also identified in apical membranes and was also ouabain-sensitive. There are at least two possible explanations for the presence of ouabain-sensitive Na,K-ATPase activity in apical membranes: contamination of apical membranes by basolateral membranes or an alternate mode of H,K-ATPase expression. The latter would be consistent with the recent observation in Xenopus oocytes of the expression of H,K-ATPase ␣ cRNA (HKc␣) as sodium-potassium dependent 86 Rb uptake (15).
To address these two possibilities, two different experimental approaches were employed: 1) the effect of a polyclonal antibody to HKc␣ (M1) on H,K-ATPase and Na,K-ATPase activities in apical and basolateral membranes; and 2) Western blot analyses of NaK␣1 expression, an established basolateral membrane marker.
Previous studies established that M1, a polyclonal antibody raised against a HKc␣ fusion protein, inhibited H,K-ATPase activity in apical membranes in rat distal colon but did not affect the activity of rabbit renal Na,K-ATPase activity (10). Therefore, the effect of M1 antibody on H,K-ATPase and Na,K-ATPase activities in both apical and basolateral membranes was examined (Figs. 1 and 2). As shown in Fig. 1, H,K-ATPase activity in apical membranes was completely inhibited by M1 antibody, whereas Na,K-ATPase activity was reduced by 92%. In contrast, Na,K-ATPase activity in basolateral membranes was not altered by M1 antibody but, as expected, was completely inhibited by ouabain (Fig. 2). These results suggest that Na,K-ATPase activity in apical membranes is an ATPase encoded by HKc␣ (or a closely related) cDNA and not by NaK␣1 cDNA. This result also suggests that the M1 antibody-insensitive fraction of Na,K-ATPase activity in the apical membrane might represent contamination from basolateral Na,K-ATPase but at most would represent no more than 8% of the total apical membrane Na,K-ATPase.
Western blot analyses of apical and basolateral membranes were also performed with M1 antibody (Fig. 3) and with an antibody to NaK␣1 (25) (Fig. 4). Similar to the earlier observations, both HKc␣ and NaK␣1 antibodies identify 100-kDa proteins (7,25). HKc␣ protein was predominantly expressed in apical membrane with modest expression in basolateral membranes (Fig. 3). In addition to a 100-kDa protein, HKc␣ antibody also recognizes two other proteins (approximately 70-and 27-kDa proteins) with less intensity that might represent another HKc␣-related protein or nonspecific binding. As expected, NaK␣1 protein was primarily present in basolateral membrane (Fig. 4). NaK␣1 protein expression (20%) was noted in the apical membrane, but there was minimal enrichment of NaK␣1 expression in the apical membrane compared with that in the crude membranes (Fig. 5B). Similar studies with HKc␣ protein revealed substantial enrichment in apical membranes compared with that in the crude membranes (Fig. 5A).
These observations are not consistent with the possibility that the Na,K-ATPase activity that was identified in apical membranes solely represents contamination with basolateral membranes containing Na,K-ATPase. Because recent studies in Xenopus oocytes have proposed that H,K-ATPase could also be expressed as a Na-K exchange (15), the presence of Na,K-TABLE I H,K-ATPase and Na,K-ATPase activities in apical and basolateral membranes from normal rat distal colon ATPase activities were measured either in presence of Mg 2ϩ alone or in the presence of Mg 2ϩ plus K ϩ , or Mg 2ϩ plus Na ϩ and K ϩ . Total H,K-ATPase activity represents ATPase activity in the presence of Mg 2ϩ plus K ϩ , minus that in the presence of Mg 2ϩ alone. Total Na,K-ATPase activity represents the difference between ATPase activity in the presence of Mg 2ϩ plus Na ϩ and K ϩ , and H,K-ATPase activity. ATPase activities were also measured in the presence of 1 mM ouabain. Activity in the presence of ouabain represents ouabain-insensitive ATPase activity. Ouabain-sensitive H,K-ATPase was determined by subtracting ouabain-insensitive activity from total activity. Mg 2ϩ -ATPase activity was 221.7 Ϯ 13.8 and 202.6 Ϯ 9.6 nmol of P i liberated/mg protein ⅐ min in apical and basolateral membranes, respectively. Results represent mean Ϯ S.E. of triplicate assays from three different membrane preparations. ATPase activity in apical membranes of rat distal colon might represent an alternate mode of H,K-ATPase expression. There is substantial evidence for the presence of two distinct H,K-ATPases in the apical membranes of the rat distal colon; one isoform is ouabain-insensitive, is present primarily in surface cells and has been cloned and identified (5,9,10,12,13), whereas the other isoform is ouabain-sensitive, is present in both surface and crypt cells, and has not, as yet, been cloned (12). Because dietary sodium depletion and aldosterone differentially enhance the expression of these two isoforms in the rat distal colon (7,27), the effect of dietary sodium depletion on H,K-ATPase and Na,K-ATPase activities and HKc␣ and NaK␣1 protein expression was, therefore, determined. Fig. 6 presents the results of H,K-ATPase activity in apical and basolateral membranes. Dietary sodium depletion increased total H,K-ATPase activity by 80%, which largely reflected a 109% stimulation of the ouabain-insensitive fraction. Dietary sodium depletion also substantially increased HKc␣ protein expression in apical membranes by 3.5-fold (Fig. 3, A   FIG. 1. Effect of M1 antibody on H,K-ATPase (p) and Na,K-ATPase (o) activities in apical membranes from normal rat distal colon. H,K-ATPase and Na,K-ATPase activities were measured, as described in the legend to Table I. H,K-ATPase and Na,K-ATPase activities were measured in normal and in M1 antibody-treated apical membranes. Antibody-treated apical membranes were mixed with M1 antibody (30 g/mL) and incubated at 37°C for 60 min. Control membranes were mixed with PBS and were similarly incubated. ATPase activities in M1 antibody-treated membranes were also measured in presence of 1 mM ouabain. Results represent mean Ϯ S.E. of triplicate assays from three different membrane preparations.
FIG. 2. Effect of M1 antibody (10) on Na,K-ATPase activity in basolateral membranes from normal rat distal colon. Na,K-ATPase activity was measured, as described in the legend to Table I. Control and antibody treatment of basolateral membranes was performed as described for apical membrane in the legend to Fig. 1. Na,K-ATPase activity in M1 antibody-treated membrane was also measured in the presence of 1 mM ouabain. Results represent mean Ϯ S.E. of triplicate assays from three different membrane preparations. HKc␣) antibody (10). A, blot was stained with M1 antibody (1:1000), followed by anti-rabbit IgG horseradish peroxidase conjugate as secondary antibody, as described under "Experimental Procedures." B, relative abundance of HKc␣ protein in apical and basolateral membranes from normal (open bars) and sodium-depleted (cross-hatched bars) rats was quantitated using densitometry. Values represent mean Ϯ S.E. of three experiments. and B). In contrast, the ouabain-sensitive component of H,K-ATPase activity was enhanced by only 42%. H,K-ATPase activity was not identified in the basolateral membranes of either normal or dietary sodium-depleted rats. In contrast, HKc␣ protein was also found to be expressed in basolateral membranes of both normal and sodium-depleted animals (Fig. 3, A  and B).

FIG. 3. Western blot analysis of apical and basolateral membranes prepared from normal and sodium-depleted rat distal colon by M1 (
The effect of dietary sodium depletion on Na,K-ATPase activity in both apical and basolateral membranes was also ex-amined and is presented in Fig. 7. Only ouabain-sensitive Na,K-ATPase activity was identified in basolateral membranes from either normal or dietary sodium depleted rats (Fig. 7B). Dietary sodium depletion resulted in only a minimal (10%) increase in ouabain-sensitive Na,K-ATPase activity in basolateral membranes (Fig. 7B). Ouabain-sensitive Na,K-ATPase activity in apical membranes was increased by 30% (Fig. 7A). In contrast to HKc␣ protein, sodium depletion did not significantly alter the NaK␣1 protein expression either in apical or in basolateral membranes (Fig. 4, A and B). DISCUSSION This study was designed to establish the nature of Na,K-ATPase activity that has been found in the apical membrane of rat distal colon (Ref. 11 and Table I). Na,K-ATPase, or the sodium pump, is a basolateral enzyme that has been frequently used as the standard basolateral membrane enzyme marker (24,26). At the time that Na,K-ATPase activity was initially reported in the apical membrane (11), there was no acceptable apical membrane marker (18). As a consequence, the presence of Na,K-ATPase in apical membranes was considered secondary to basolateral membrane contamination (11,17). Since  antibodies (10, 25). A, blot was stained with M1 antibody (1:1000) followed by anti-rabbit IgG horseradish peroxidase conjugate as secondary antibody, as described under "Experimental Procedures." B, blot was stained with NaK␣1 antibody (1:2000), followed by anti-mouse IgG horseradish peroxidase conjugate as secondary antibody, as described under "Experimental Procedures."  Table I. In parallel studies, H,K-ATPase activity was not identified in basolateral membranes prepared from the distal colon of either normal or sodium-depleted rats (data not shown). Results represent mean Ϯ S.E. of triplicate assays from three different membrane preparations. then, H,K-ATPase has been identified in apical membranes and is now considered an excellent enzyme marker of the apical membrane of colonic epithelial cells.
Critical assessment of the present results depends on the specificity of the M1 antibody to identify H,K-ATPase and not Na,K-ATPase. Several observations provide substantial evidence of the specificity of the M1 antibody. First, M1 antibody in immunohistochemical studies localized proteins to plasma membranes and cytoplasm only in Sf-9 cells that had been transfected with HKc␣ cDNA (10). Second, additional immunohistochemical studies revealed that M1 antibody only localized proteins specific to the apical membrane of surface and upper crypt cells of distal (and not proximal) colon of the rat (10). Third, in HEK293 transfected with HKc␣ cDNA, but not in untransfected cells, M1 antibody identified a membrane-bound protein in both immunohistochemical and Western blot (100 kDa) studies (9). Fourth, M1 antibody also inhibited potassium-activated, but not Na/K-activated ATPase activities in apical membranes of rat distal colon and in Sf-9 cells transfected with HKc␣ cDNA (10). In contrast, M1 antibody did not inhibit either highly purified Na/K-activated ATPase isolated from rabbit renal medulla (10) or Na/K-activated ATPase activity in basolateral membranes of rat distal colon (Fig. 2); nor did M1 antibody recognize any proteins in rat parietal cells (i.e. gastric H,K-ATPase) (7).
The immunoblot in Fig. 4 provides evidence of NaK␣1 protein in apical membranes from normal and sodium-depleted animals. Therefore, Na,K-ATPase activity in apical membranes represents, at least in part, contamination of apical membranes by basolateral membranes. However, these present studies with polyclonal antibody (M1) against the ␣ subunit of H,K-ATPase provide compelling evidence that apical membrane Na,K-ATPase activity is not solely a result of basolateral membrane contamination of the apical membrane preparation (Figs. 1 and 2). First, Na,K-ATPase activity in apical membranes was virtually (92%) abolished by the M1 antibody (Fig.  1). Second, Na,K-ATPase activity in basolateral membrane was not altered by M1 antibody (Fig. 2).
HKc␣ protein, but not H,K-ATPase activity, was identified in basolateral membranes (Fig. 3 and Table I). The presence of HKc␣ protein in basolateral membranes was much less than that in apical membranes. Because previous immunofluorescent studies have demonstrated that HKc␣ protein is localized only in apical membranes and not in basolateral membranes (10), it is likely that the presence of HKc␣ protein identified in basolateral membranes represents contamination from apical membranes. The failure to establish the presence of H,K-ATPase activity in basolateral membranes is probably due to the different techniques used to prepare apical and basolateral membranes (19,22). This possibility is based on the recent observations that anion exchange isoform 2 (AE2)-specific protein (28) and anion exchange (Cl-HCO 3 ) activities (29) were not detected in basolateral membranes prepared from kidney and colon, respectively, in the absence of protease inhibitors. In contrast, AE2 protein and Cl-HCO 3 exchange activity were readily identified in basolateral membranes prepared in the presence of protease inhibitors. Because the method to prepare colonic basolateral membranes does not usually include protease inhibitors, we suspect that, similar to AE2 protein and Cl-HCO 3 exchange, H,K-ATPase was also denatured with loss of activity during the basolateral membrane preparation.
The demonstration that apical membrane Na,K-ATPase activity does not solely represent basolateral membrane contamination required studies to explain the function and origin of Na,K-ATPase in the apical membrane. Recent experiments with 86 Rb uptake in Xenopus oocytes using HKc␣ cRNA provided evidence that H,K-ATPase can be expressed both as an H-K exchange and as an Na-K exchange (15). Thus, there was a possibility that Na,K-ATPase activity in apical membrane of rat distal colon represented an alternate mode of expression of H,K-ATPase, especially as the function of H,K-ATPase in native tissue has not been well-delineated.
H,K-ATPase has been extensively studied during the past decade since the report of the successful cloning of its ␣ subunit by Crowson and Shull in 1992 (5). Nonetheless, considerable controversy exists especially regarding its ␤ subunit and its sensitivity to ouabain (8 -10, 12, 14 -16). Studies with HKc␣ cRNA in Xenopus oocytes and 86 Rb uptake have yielded evidence of ouabain-sensitive function (14 -16). In contrast, experiments in HEK293, a mammalian cell line, and in Sf9 insect cells have demonstrated the expression of HKc␣ cDNA as ouabain-insensitive H,K-ATPase (9,10). Because HKc␣ mRNA and protein are primarily expressed in surface and not in crypt epithelial cells (10,13) and because HK function in crypt cells is solely ouabain-sensitive (12), it would be necessary to postu- FIG. 7. Effect of dietary sodium depletion on Na,K-ATPase activity. Na,K-ATPase activity was measured in apical (A) and basolateral (B) membranes from normal (open bars) and sodium-depleted (cross-hatched bars) rat distal colon. ATPase activities were measured either in the presence of Mg 2ϩ plus K ϩ , or Mg 2ϩ plus Na ϩ and K ϩ . Total Na,K-ATPase activity represents ATPase activity in the presence of Mg 2ϩ plus Na ϩ and K ϩ , minus that in the presence of Mg 2ϩ plus K ϩ . ATPase activities were also measured in the presence of 1 mM ouabain. Activity in the presence of ouabain represents ouabain-insensitive ATPase activity. Ouabain-sensitive H,K-ATPase was calculated by subtracting ouabain-insensitive activity from the total activity. Results presented represent mean Ϯ S.E. of triplicate assays from three different membrane preparations. late the existence of three H,K-ATPase to account for ouabainsensitive expression of HKc␣ cDNA.
There is, however, excellent physiological evidence for the existence of two distinct H,K-ATPases in the rat distal colon. First, ouabain-sensitive and ouabain-insensitive components of H,K-ATPases have been identified in the apical membrane of the rat distal colon (Table I and Ref. 11). Second, active potassium absorption, energized by apical membrane H,K-ATPase, has two components: a fraction that is sensitive to mucosal sodium and is ouabain-insensitive and a fraction that is insensitive to mucosal sodium and is ouabain-sensitive (27). Third, aldosterone secondary to dietary sodium depletion stimulates only one of these two components of active potassium absorption, the so-called mucosal sodium-sensitive, ouabain-insensitive component (27). Fourth, although HKc␣ mRNA and protein are localized solely to surface cells of the rat distal colon, H,K-ATPase activity and potassium-dependent recovery of intracellular pH (pH i ) to an acid load that is sensitive to ouabain are localized primarily in crypt cells (12). Fifth, aldosterone up-regulates HKc␣ mRNA and protein, but dietary potassium depletion, which also enhances active potassium absorption, does not alter HKc␣ message and protein suggesting that the regulation of active potassium absorption by dietary potassium depletion is linked to another H,K-ATPase isoform (7). Sixth, although a defect in active colonic potassium absorption is evident in potassium balance studies in HKc␣ knockout mice (30), this is not a lethal gene deletion, suggesting there is a second colonic potassium-absorptive process present in these transgenic mice. Finally, aldosterone differentially enhances the ouabain-sensitive and the ouabain-insensitive components of H,K-ATPase (Fig. 6).
The results of the effect of dietary sodium depletion on Na,K-ATPase and H,K-ATPase function in apical and basolateral membranes (Figs. 3, 4, 6, and 7) provide compelling evidence that Na,K-ATPase represents an alternate mode of expression of a ouabain-sensitive H,K-ATPase isoform that has not as yet been identified and cloned. First, as expected, Na,K-ATPase activity in the apical membrane is almost exclusively ouabainsensitive with no ouabain-insensitive activity observed in apical (or basolateral) membranes in sodium-depleted rats. Second, the ouabain-sensitive components of Na,K-ATPase and H,K-ATPase were similarly increased by dietary sodium depletion (30% and 42%, respectively). Third, in contrast, aldosterone enhanced the ouabain-insensitive component of H,K-ATPase activity and HKc␣ protein expression in apical membranes by 2.1-and 3.5-fold, respectively (Figs. 3 and 6). Fourth, transfection of HEK293 cells by HKc␣ cDNA and a colonic H,K-ATPase ␤ subunit resulted in the expression of ouabain-insensitive H,K-ATPase activity without evidence of any Na,K-ATPase expression (9). As a result, it is likely that the cDNA that encodes the ouabain-sensitive component of H,K-ATPase has not as yet been cloned, and we suspect that this ATPase may also function as a Na-K exchange in apical membranes of rat distal colon. Studies performed in the rat outer medullary collecting duct are also consistent with many of these present observations (31,32). Hayashi and Katz (31,32) provided evidence that dietary potassium depletion increased a luminal membrane ouabain-sensitive Na,K-ATPase that regulated potassium reabsorption.
Codina et al. (33) also recently reported studies designed to identify the significance of Na,K-ATPase activity identified in apical membranes prepared from rat distal colon. Significant differences exist in the methods and experimental observations between the present study and that of Codina et al. (33). These differences include the HKc␣ antibody and the methods employed to prepare colonic apical membranes and to assay ATPase activity. Furthermore, Codina et al. (33) found that sodium-dependent K-ATPase activity was much greater than the sodium-independent activity that was completely ouabaininsensitive. In contrast, the present results established that Na,K-ATPase activity was approximately equal to H,K-ATPase and that H,K-ATPase activity was 56% ouabain-insensitive (Table I). Despite these differences, both studies concluded that apical membrane Na,K-ATPase activity represents an alternate mode of H,K-ATPase expression.