Rapamycin inhibits protein kinase C activity and stimulates Na+ transport in A6 cells.

Rapamycin and FK506 have unique cellular effects despite the fact that they bind to the same set of immunophilins, the FK506 binding proteins (FKBP). We have previously reported that rapamycin (RAP) stimulates sodium transport in A6 cells. FK506 did not stimulate sodium transport but did inhibit the stimulation seen in RAP-treated cells. Since FKBP12 has been shown to have sequence homology with an endogenous inhibitor of protein kinase C (PKC) and PKC inhibition has been shown to increase Na+ channel activity in A6 cells, studies to determine the effect of RAP on PKC activity and its relationship to sodium transport were performed. Here we report that RAP stimulates sodium transport, and the effect is not additive to that seen with a cell-permeant inhibitor of PKCα and -β subtypes. RAP significantly inhibits endogenous PKC activity in A6 cells both in membrane and cytosolic preparations. There is a strong correlation between the degree of inhibition of PKC activity and the stimulation of sodium transport by RAP. RAP has no effect on Na+/K+-ATPase activity over this time course. Purified recombinant FKBP12 with or without FK506 has no effect on PKC activity when incubated with a rat brain-derived PKC preparation of known activity. By contrast, RAP plus FKBP12 significantly inhibits PKC activity. RAP plus FKBP12 inhibits the PKCα and not the -β subtype. The results demonstrate inhibition of PKC activity by RAP and not FK506 through its binding to FKBP12. The inhibition of PKC activity by RAP stimulates sodium transport in A6. The results therefore imply the existence of an endogenous RAP-like ligand which when bound to FKBP12 could regulate Na+ channel activity through this mechanism.

FKBP25). The cyclophilins bind the structurally distinct macrocyclic peptide cyclosporin. The immunosuppressive effects of these agents are not mediated solely by the drug themselves or by the immunophilin to which they bind. Rather the binding of drug to the immunophilin exposes an effector region on the drug which then inhibits downstream pathways (1,2). At present it is know that two distinct drugs that bind to the same immunophilin can have drastically different effects. For example, although FK506 and RAP both bind with high affinity to FKBP12 only FK506 inhibits calcineurin (1,2). In contrast, the RAP⅐FKBP complex has been found to inhibit cell proliferation by inhibiting p70 S6 kinase activation (22).
A6 cells, derived from Xenopus laevis are a well characterized, high resistance epithelia with transport characteristics similar to the mammalian cortical collecting duct (3). Na ϩ channel activity is known to be increased in response to aldosterone (4,23), insulin (5), and vasopressin (4, 6 -9), each acting through unique pathways. In addition, inhibition of protein kinase C (PKC) activity has been shown to increase sodium channel activity in cell-attached patches of A6 (3). We have recently reported that RAP stimulates transepithelial sodium transport in A6 (10). The effect is seen within 15 min of addition and persists for 4 h. An excess of FK506 had no effect on basal transport but completely inhibited the RAP-induced effect. This suggested that the effect was mediated through a RAP-immunophilin complex (10), since both drugs bind to the same set of immunophilins.
In this regard, FKBP12 is known to have sequence homology to an endogenous PKC inhibitor I-2 (11). This suggested that RAP might inhibit PKC activity through its interaction with this immunophilin. RAP has been found to block a classical PKC (cPKC)-dependent and phosphatidylinositol 3-kinase-dependent stimulation of p70 S6 kinase activity (12,22). While RAP has no effect on phosphatidylinositol 3-kinase activity (12), the effect of RAP on cPKC activity has not been reported. The effect might be peculiar to the RAP⅐FKBP12 complex since it has been clearly demonstrated that FK506 does not have an effect on PKC activity (13). Studies were therefore performed to determine the effect of RAP on PKC activity and its relationship to sodium transport in A6.

EXPERIMENTAL PROCEDURES
Cell Culture-A6 cells were grown as described previously in amphibian media in 10% fetal calf serum on millicell inserts (Millipore Corp.) and studied when they exhibited stable electrical resistance (10 -14 days) (10,23). Amiloride-sensitive short circuit current (I sc ) was measured in a sterile in-hood modified Ussing chamber as described previously (10). Nystatin was added to the apical solution (100 units/ml) to permeabilize the apical membrane while the cells were in the shortcircuited state. The peak of the subsequent increase in I sc was recorded (10) and is a measure of maximal Na ϩ /K ϩ -ATPase activity.
PKC Assay-A6 cells were exposed to drug or diluent for the appropriate time and then washed three times with ice-cold calcium-free phosphate-buffered saline. Cells were then scraped from the filters using a rubber policeman into ice-cold phosphate-buffered saline and centrifuged at 1500 ϫ g for 5 min to pellet the cells. The cell pellet was resuspended in homogenization buffer containing 100 mM Tris-HCl, 1 mM EDTA, 1 mM DTT, leupeptin (10 g/ml), chymostatin (10 g/ml), phenylmethylsulfonyl fluoride (1 mM), sodium vanadate (0.1 mM), okadaic acid (0.5 M), and 10% glycerol. The resulting suspension was centrifuged at 100,000 ϫ g for 1 h at 4°C. The supernatant represents the crude cytosolic fraction, and the pellet represents the crude membrane fraction. Alternatively, crude cytosolic and membrane fractions were made as described above but from cells not exposed to drug in vivo. Samples were protein-matched and then exposed in homogenization buffer to a final concentration of RAP (1 M), N-myristoylated PKC inhibitor (50 M), phorbol myristic acid (PMA) or RAP (10 nM) and FK506 (1 M), and PKC activity was measured. PKC activity was measured using a Spinzyme assay kit (Pierce, catalog number 29510) using fluorescent dye labeled n-acetylated peptide sequence (4 -14) of myelin basic protein (Pierce catalog number 29527) as a substrate. This is a specific substrate of ␣, ␤, and ␥ isoforms of PKC and a poor substrate for PKA (14). Briefly, crude cytosolic or membrane fractions containing 10 g of protein were mixed with reaction buffer to a final concentration of 2 mM ATP, 10 mM MgCl 2 , 0.1 mM CaCl 2 , 0.002% Triton X-100, 20 mM Tris (hydroxymethyl)aminomethane, 0.2 mM phosphatidyl L-serine, pH 7.4, and allowed to incubate for 30 min at 30°C. Bound phosphorylated substrate was eluted using 0.1 M ammonium HCO 3 Ϫ and 0.02% sodium azide, pH 8, and detected by measuring absorbance at 570 nM. To generate a standard curve, absorbance was measured in samples containing a range of known activity (0 -0.020 units per reaction) of protein kinase C from rat brain. 1 unit of specific activity is defined as the amount that will transfer 1 nmol of phosphate to histone H1 per min at 30°C.
To determine if RAP or FK506 affected exogenous PKC activity in the presence of exogenous FKBP12, PKC activity from rat brain of known activity (10 microunits) was incubated in the presence of 7 g of purified human recombinant FKBP12 from Escherichia coli (15) with or without 10 M FK506 or 10 M RAP or FK506 and RAP for 10 min at room temperature. Reaction buffer to a final concentration of 2 mM ATP, 10 mM MgCl 2 , 0.1 mM CaCl 2 , 0.002% Triton X-100, 20 mM Tris(hydroxymethyl)aminomethane, 0.2 mM phosphatidyl L-serine, pH 7.4, was then added and allowed to incubate for 30 min at 30°C, and PKC activity was determined. In parallel experiments, FKBP12 was incubated with PKC␣ or PKC␤ subtypes in the presence or absence of RAP, and PKC activity was determined as described above.
Na ϩ /K ϩ -ATPase Assay-This was performed as described previously (10). Na ϩ /K ϩ -ATPase activity is determined from the difference between P i released in control tubes and that released in identical samples containing 1 mM ouabain (10).
Western Blotting-This was performed as described previously (23). Crude cytosolic and crude membrane fractions were protein-matched and solubilized in sample buffer containing 5% SDS, 8% 2-mercaptoethanol, 62.5 mM Tris-HCl, 10% glycerol, 0.025% bromphenol blue, pH 6.8. Samples were heated at 95°C for 5 min and subjected to SDSpolyacrylamide gel electrophoresis using 5% stacking gels and 15% separating gels. Rainbow colored molecular weight standards (Sigma) were run in adjacent lanes. For FKBP12 detection, samples were solubilized in sample buffer containing 10 mM DL-dithiothreitol (DTT) and were heated overnight at 37°C, and then for 20 min at 95°C. Resolved proteins were transferred to nitrocellulose membranes as described previously (23), and transfers were overlaid with 1:1000 dilutions of specific primary antibody in 1% non-fat dried milk in phosphate-buffered saline (1% Blotto). Reactive proteins were detected with a 1:2000 dilution of horseradish peroxidase-conjugated second antibody in 1% Blotto and signal-detected by an enhanced chemiluminescence system (Amersham Corp.) and exposed to Kodak X-Omat AR film (Eastman Kodak Co.).
Materials-N-myristoylated Phe-Ala-Arg-Lys-Gly-Ser-Leu-Arg-Gln, representing the pseudo-substrate sequence from cPKC␣ and -␤, was obtained from BIOMOL (Plymouth Meeting, PA). This is a highly specific inhibitor which is N-terminally myristoylated to allow membrane permeability (16,17). RAP was obtained from BIOMOL and stored as a 2 mM solution in ethanol at Ϫ70°C and diluted into serum-free media on the day of the experiment. Human recombinant FKBP12 expressed in E. coli and rat brain PKC and PKC␣ and -␤ subtypes from rat brain were obtained from Sigma. Protein kinase C assay materials were from Pierce. FKBP12 antibody directed against amino acids 3-16 of the FKBP12 was the kind gift of Dr. Andrew Marks and Dr. Thottala Jayaraman of the Mt. Sinai School of Medicine (27). Monoclonal antibody against the PKC␣ subtype (catolog number P16520) was purchased from Transduction Laboratories (Lexington, KY). Horseradishperoxidase-conjugated second antibodies were purchased from Jackson Immunoresearch (West Grove, PA). All other reagents were from Sigma.
Statistics-Data were analyzed using one-way analysis of variance on Number Cruncher statistical software (NCSS, Hintze, Kaysville, UT). Data are expressed as means Ϯ S.E. A p value of less than 0.05 was considered significant.

RESULTS
We have previously shown that 1 and 10 nM RAP stimulates transepithelial sodium transport in A6 cells and that this effect is blocked by a 100-fold excess of FK506 (10). To determine the dose-response relationship of this effect, cells were exposed to dilutions of RAP, and the short circuit current at 1 h was measured and expressed as the % increase relative to control I sc . These data are shown in Fig. 1. The K d is approximately 10 nM. Sodium channel activity in A6 cells is known to be modulated by several hormones each acting through unique pathways (4). Vasopressin activates sodium channels through a PKA-associated phosphorylation (4, 6 -9), and insulin stimulates Na ϩ transport through a mechanism dependent upon tyrosine kinase activity (5). Since the stimulation of sodium transport by insulin or vasopressin was similar to the time course seen with RAP, we considered the possibility that RAP might be acting through a common pathway. As shown in Fig.  2, the simultaneous addition of insulin and RAP resulted in an additive effect on sodium transport compared with when RAP was added alone. Cells exposed to vasopressin had a significant increase in sodium transport, and when RAP was added after 60 min to these cells, there was a further stimulation of transport. These data suggest that RAP stimulates sodium transport via an independent mechanism.
Since RAP binds to FKBP12 which has been reported to have nearly identical sequence homology to an endogenous inhibitor of PKC (11,13), the effect of RAP on stimulating sodium transport might be through the inhibition of PKC. To determine the effect of PKC inhibition on sodium transport, we exposed A6 cells to an N-myristoylated cell-permeant peptide pseudosubstrate sequence of the PKC␣ and -␤ isoforms and measured the effect of I sc . As shown in Fig. 3, the PKC inhibitor stimulated sodium transport. When the PKC inhibitor and RAP were added simultaneously to A6 cells, no additional stimulation of I sc was noted. This suggested that the mechanism of action of RAP on stimulation of sodium transport was via inhibition FIG. 1. Dose-response curve of rapamycin stimulation of sodium transport in A6. We have previously shown that 1 and 10 nM RAP stimulate transepithelial sodium transport in A6 cells and that this effect is blocked by a 100-fold excess of FK506 (10). To determine the dose-response relationship of this effect, cells were exposed to dilutions of RAP, and the short circuit current (I sc ) at 1 h was measured and expressed as the % increase relative to control I sc . The K d is approximately 10 nM. n ϭ 4 for each point.
of PKC activity. PKC activity was measured in crude cytosolic and membrane fractions that were exposed to either PKC inhibitor, RAP, or phorbol 12-myristate 13-acetate (PMA) in vitro. PMA is known to stimulate PKC activity and inhibit sodium transport in these cells (18,19). As shown in Table I, PKC activity was significantly inhibited by both the PKC inhibitor and RAP in the membrane fractions. PKC activity in the cytosol was significantly inhibited by RAP, although not by the PKC inhibitor. PMA stimulated PKC activity in both the membrane and cytosolic fractions. We next measured PKC activity in the crude cytosolic and crude membrane fractions from cells that were exposed for 1 h to either 50 nM PKC inhibitor, 1.2 M RAP, 1.2 M FK506, or 0.1 M PMA. As shown in Table II, PKC activity is greatest in the membrane fractions of control cells. As expected, PKC activity was stimulated in the membrane fractions of PMA-treated cells and decreased in the cytosol (25). The PKC activity was significantly inhibited in both the cytosolic and membrane fractions of cell treated with the N-myristoylated PKC inhibitor. RAP inhibited PKC activity to the same degree as the PKC inhibitor in both the cytosolic and membrane fractions. FK506 had no effect on PKC activity in either the cytosol or the membrane fraction. To determine if the effect of the agents on activity was due to changes in the association of the PKC␣ isoenzyme with the membrane and cytosolic fractions, the experiment shown in Fig. 4 was performed. PMA treatment resulted in a decrease in the amount of PKC␣ isoenzyme in the cytosolic fraction and an increase in the membrane fraction relative to control as has been described previously (35). Although RAP inhibits PKC activity in the membrane fraction (Table II), RAP did not inhibit the association of the PKC␣ isoenzyme with the membrane fraction (Fig. 4).
Since FK506 inhibits the stimulation of sodium transport by PKC, we measured the effect of a 100-fold excess of FK506 on the RAP-associated inhibition of PKC activity. FK506 prevented the RAP-associated inhibition of PKC activity (Table II).
This implied the presence of FKBP12 in both the cytosolic and membrane fractions. Immunoblots of membrane and cytosolic fractions with an antibody directed against FKBP12 are shown in Fig. 5. FKBP12 was detected in both fractions (Fig. 5, lanes  A and C). Detection of the protein in the membrane fractions required treatment with sample buffer containing 10 mM DTT and overnight incubation at 37°C and then heating for 20 min at 100°C before SDS-PAGE (Fig. 5, lane C).
To determine the relationship of the inhibition of PKC activity with the stimulation of sodium transport by RAP, the doseresponse curves for the inhibition of PKC activity and the stimulation of sodium transport by RAP were compared as follows. A6 cells were exposed to serial dilutions of RAP or diluent (1 ϫ 10 Ϫ5 -1 ϫ 10 Ϫ10 M), I sc was measured at 1 h and recorded as relative to control I sc at 1 h. Cells were then scraped and PKC activity measured in the membrane fraction. As shown in Fig. 6, there is a strong linear relationship between the inhibition of PKC activity and the stimulation of sodium transport by RAP.
Since it has been shown that inhibition of PKC is associated FIG. 3. PKC inhibitor stimulates sodium transport in A6, and the effect is not additive to that seen with rapamycin alone.
Since RAP binds to FKBP12 which has been reported to have nearly identical sequence homology to an endogenous inhibitor of PKC (11,13), the effect of RAP on stimulating sodium transport might be through the inhibition of PKC. To determine the effect of PKC inhibition on sodium transport, we exposed A6 cells to an N-myristoylated cell-permeant peptide pseudosubstrate sequence of the PKC␣ and -␤ isoforms and measured the effect of I sc . PKC inhibitor stimulated sodium transport. When the PKC inhibitor and RAP were added simultaneously to A6 cells, no additional stimulation of I sc was noted. This suggested that the mechanism of action of RAP on stimulation of sodium transport was via inhibition of PKC activity. n ϭ 4.

TABLE I
Rapamycin inhibits PKC activity in crude cell preparations of A6 exposed to drug in vitro Crude membrane and cytosolic preparations were prepared, exposed to drugs as indicated, and PKC activity measured as described in the text (n ϭ 6). Results are shown in microunits, where 1 unit of specific activity will transfer 1 nmol of phosphate to histone H1 per min at 30°C. CON, control; RAP, rapamycin; PKC inhibitor, N-terminal myristoylated pseudosubstate inhibitor of PKC ␣ and ␤ subtypes.  2. The stimulation of sodium transport by rapamycin is additive to stimulatory effects of insulin or vasopressin. The simultaneous addition of insulin and RAP resulted in an additive effect on sodium transport compared when RAP or insulin was added alone. Cells exposed to vasopressin had a significant increase in sodium transport, and when RAP was added after 60 min to these cells, there was a further stimulation of transport. These data suggest that RAP stimulates sodium transport via an independent mechanism. n ϭ 6.
with activation of Na ϩ /K ϩ -ATPase in certain kidney cell lines (20, 21), we measured the effect of this agent on Na ϩ /K ϩ -ATPase activity by two methods. When the apical membrane is permeabilized to cations with nystatin while in the short-circuited state, the peak of the increase in I sc reflects maximal Na ϩ /K ϩ -ATPase activity (10). As can be seen in Fig. 7, cells exposed to 1 M RAP for 1 h demonstrate a peak nystatinassociated current that is not different than control. Next, A6 cells were exposed to 1 M RAP or diluent for 1 h, and enzymatic Na ϩ /K ϩ -ATPase activity was measured in whole cell lysates. RAP had no effect on Na ϩ /K ϩ -ATPase activity in A6.
To determine if RAP inhibits PKC activity through its interaction with FKBP12, the following experiment was performed. Purified FKBP12 was incubated in the presence of either diluent, 10 M RAP, 10 M RAP, and 100 M FK506 at room temperature for 10 min in the presence of 10 microunits of PKC activity from rat brain. As shown in Table III, FKBP12 had no effect on PKC activity. By contrast, RAP, when incubated in the presence of FKBP12, significantly inhibited PKC activity. When RAP was incubated with a 10-fold excess of FK506 in the presence of FKBP12, PKC activity was increased toward control values. By contrast, FK506 ϩ FKBP12 had no effect on PKC activity (control: 9.98 Ϯ 0.33; FKBP12: 10.25 Ϯ 0.33 microunits, n ϭ 4). We next measured the effect of drugs and FKBP12 on the activity of PKC␣ and -␤ isoenzymes. Purified FKBP12 was incubated in the presence of drugs and 8 microunits of PKC␣ or PKC ␤ isoenzyme (one unit of PKC␣ or -␤ will transfer 1 nmol of phosphate to histone I per min at pH 7.4 at 30°C). RAP significantly inhibited the PKC␣ isoenzyme but not the PKC ␤ isoenzyme (PKC␣ activity: FKBP12 alone 7.77 Ϯ 0.7; FKBP ϩ RAP: 4.76 Ϯ 0.8 microunits, p Ͻ 0.02, n ϭ 4. PKC␤ activity: FKBP12 alone 7.3 Ϯ 0.6; FKBP12 ϩ RAP 6.9 Ϯ 0.6 microunits, p ϭ not significantly different than FKBP12 alone; n ϭ 4). DISCUSSION RAP is a fungal macrolide which has recently been introduced as an immunosuppressant for use in solid organ transplantation (1, 2). Although RAP and FK506 are structurally similar and bind to a common set of immunophilins, the FK506 binding proteins, their mechanism of action and effects on T-cell function are remarkably dissimilar. It has been shown that it is the ligand-immunophilin complex and not the immunophilin or the ligand itself that determines the specific downstream effect peculiar to each drug (1,2). Whereas the FK506⅐FKBP12 complex inhibits calcineurin-dependent lymphokine gene transcription and T-cell activation (1), the RAPimmunophilin complex inhibits cytokine-stimulated T-cell proliferation in part through the inhibition of both PKC-dependent and PKC-independent p70 S6 kinase serine phosphorylations (22). A6 cells conduct sodium in a vectorial fashion through apically localized Na ϩ channels and basolaterally localized Na ϩ / K ϩ -ATPase pumps. Sodium channel activity is known to be affected by several hormones including vasopressin (4, 6 -9), insulin (5), and aldosterone (4,23), each acting through different mechanisms. It has previously been shown that PKC stimulation inhibits Na ϩ transport (18,19), and inhibition stimulates Na ϩ channel activity (5,24). Since RAP stimulates sodium transport in A6 and FKBP12 has sequence homology to an endogenous PKC inhibitor (11), it seemed likely that the effect of RAP on sodium transport was occurring through inhibition of PKC by its interaction with FKBP12. Previous studies have failed to demonstrate an effect of FK506 on PKC activity (13); however, the effect of RAP on PKC activity has not been reported.
Here we demonstrate that RAP stimulates sodium transport FIG. 5. FKBP12 is found in both the cytosolic and membrane preparation. Crude membrane and cytosolic proteins were prepared from cells exposed to diluent and solubilized in sample buffer containing containing 10 mM DTT. Samples were heated at 95°C for 3 min before resolving on SDS-PAGE. Proteins were transferred and probed with FKBP12 antibody and detected with an enhanced chemiluminescence system and autoradiography. Migration of rainbow colored molecular weight markers are noted at the left. A, proteins from cytosolic fraction; B, proteins from membrane fraction. The FKBP12 is present in the cytosolic fraction. There are reactive proteins at 24 kDa in the membrane fraction. Lane C, membrane fractions were prepared as above but solubilized in sample buffer containing 10 mM DTT, heated overnight at 37°C, and then heated at 95°C for 20 min before resolving on SDS-PAGE. Proteins were transferred to nitrocellulose and probed with the FKBP12 antibody and detected as in A. Under stringent reducing conditions, FKBP12 is found in the membrane fractions. n ϭ 2 for each sample.

TABLE II
RAP but not FK506 inhibits PKC activity in whole cells exposed to drug in vivo A6 cells were exposed to drug or diluent as described; crude cytosol and membrane preparations were made, and PKC activity was measured as described. n ϭ 6. RAP/FK, 10 nM RAP, 1 M FK506. PMA, phorbol myristic acid. See Table I 4. Effect of rapamycin and PMA on the association of PKC␣ subtype with crude membrane or cytosolic fractions as quantified by Western blotting. Crude membrane and cytosolic proteins were prepared from cells exposed to 1 M RAP, 0.1 M PMA, or diluent for 1 h, and and protein-matched samples were resolved by SDS-PAGE, transferred to nitrocellulose, and probed using an antibody specific for the PKC␣ subtype. Reactive proteins were detected with a horseradish peroxidase-conjugated second antibody, enhanced chemiluminescence, and autoradiography. The PKC␣ protein is 80 kDa. CYTO, crude cytosolic preparation; MEM, crude membrane preparation. C, control; P, PMA; R, rapamycin. RAP did not block association of the PKC␣ subtype with the membrane fraction. n ϭ 4 for each sample. in A6 with a K d of 10 nM (Fig. 1). This is similar to the dissociation constant for RAP binding to FKBP12 in human T-cells (1,2). The effect is blocked by a 100-fold excess of FK506 (10). These results imply that the effect is mediated through a common FK binding protein. The effect was additive to that induced by either insulin or vasopressin. By contrast, a cellpermeant inhibitor of cPKC␣ and -␤ subtypes stimulates sodium transport in this cell line, and the simultaneous addition of RAP plus this inhibitor resulted in no additional increase in Na ϩ transport (Fig. 3). These results suggested that RAP was stimulating sodium transport via inhibition of PKC activity. PKC activity was measured in crude cytosolic and membrane fractions. As shown in Tables I and II, phorbol stimulates PKC activity in the membrane fractions. RAP and the cell-permeant PKC inhibitor inhibit PKC activity in membranes to the same extent. The PKC inhibitor inhibited PKC activity in the membrane fractions only. When intact A6 cells were exposed to these agents (Table II), the bulk of PKC activity was membrane-associated, and as expected PMA treatment resulted in the translocation of PKC activity from the cytosol to the membrane (32). This can be attributed in part to translocation of the PKC␣ isoenzyme to the membrane fraction as shown in Fig. 4. RAP resulted in inhibition in PKC activity in both the cytosol and membrane fractions, similar to that seen with the PKC inhibitor (Table II). This inhibition cannot be accounted for by a decrease in the amount of PKC␣ isoenzyme in either the membrane or cytosolic fraction in RAP-treated cells (Fig. 4). FK506 prevented the inhibition of PKC activity by RAP in both the membrane and cytosolic fractions (Table II). FK506 itself has no effect on PKC activity, just as it had no effect on basal Na ϩ transport (10). Thus FK506 prevents the inhibition of PKC activity by RAP just as it was found to inhibit the stimulation of Na ϩ transport by RAP (10). The presence of FKBP12 in the membrane fractions (Fig. 5) provides further support for this being an immunophilin-mediated effect. Previous studies have shown a typical distribution of the FKBP's to the nucleus and cytoplasm or attached to the cell cytoskeleton (33). The finding of FKBP12 in the crude membrane described here is not inconsistent with its association with the cytoskeletal elements. The requirement for stringent reducing conditions to resolve this protein in the membrane fractions supports the notion that it might be associated with other cellular proteins. There is evidence that the regulation of the amiloride-sensitive sodium channel is in part dependent upon its association with the actin cytoskeleton (34). Whether FKBP12 is associated with and has actions on the cytoskeleton will be the subject of future study. The finding that the degree of PKC inhibition is strongly correlated with the degree of stimulation of transepithelial sodium FIG. 6. The stimulation of sodium transport and inhibition of PKC activity by rapamycin are strongly related. To determine the relationship of the inhibition of PKC activity with the stimulation of sodium transport by RAP, the dose-response curves for the inhibition of PKC activity and the stimulation of sodium transport by RAP were compared as follows. A6 cells were exposed to serial dilutions of RAP or diluent (1 ϫ 10 Ϫ5 -1 ϫ 10 Ϫ10 M), and I sc was measured at 1 h and recorded as relative to control I sc at 1 h. Cells were then scraped, and PKC activity was measured in the membrane fraction. There is a strong linear relationship between the inhibition of PKC activity and the stimulation of sodium transport by RAP. The numbers above each point represent the concentration of RAP used. 95% confidence intervals are shown above and below the regression line. n ϭ 4 for each point.

FIG. 7.
Rapamycin has no effect on enzymatic Na ؉ /K ؉ -ATPase activity or nystatin-stimulated short circuit current. Since it has been shown that inhibition of PKC is associated with activation of Na ϩ /K ϩ -ATPase in certain kidney cell lines (20,21), we measured the effect of this agent on Na ϩ /K ϩ -ATPase activity by two methods. When the apical membrane is permeabilized to cations with nystatin while in the short-circuited state, the peak of the increase in I sc reflects maximal Na ϩ /K ϩ -ATPase activity (10). As demonstrated by the two bars on the left side of this figure, cells exposed to 1 M RAP for 1 h demonstrate a peak nystatin-associated current that is not different than control. RAP, rapamycin; CON, control; n ϭ 4. Next, cells were scraped from filters and Dounce-homogenized, and Na ϩ /K ϩ -ATPase activity is determined from the difference between P i released in control tubes and that released in identical samples containing 1 mM ouabain (10) as described in the text. Activity is measured in M P i /mg protein/h as shown on the right axis. CON, control (2.0 Ϯ 0.01); RAP, rapamycin (1.8 Ϯ 0.08), n ϭ 4.

RAP-FKBP12 inhibits PKC activity
To determine if RAP or FK506 in the presence of exogenous FKBP12 affected exogenous PKC activity, PKC activity from rat brain of known activity (10 microunits) was incubated in the presence of 7 g of purified human recombinant FKBP12 from E. coli (15) (Sigma) with or without 10 M FK506 or 10 M RAP, or FK506 and RAP for 10 min at room temperature. Then reaction buffer to a final concentration of 2 mM ATP, 10 mM MgCl 2 , 0.1 mM CaCl 2 , 0.002% Triton X-100, 20 mM Tris(hydroxymethyl)aminomethane, 0.2 mM phosphatidyl-L-serine, pH 7.4, was added and allowed to incubate for 30 min at 30°C, and PKC activity was determined. In parallel experiments, FKBP12 was incubated with PKC␣ or PKC␤ subtypes in the presence or absence of RAP, and PKC activity was determined as described above. n ϭ 4. transport by RAP (Fig. 6) strongly suggests that RAP stimulates sodium transport via inhibition of PKC.
The time course of stimulation of sodium transport by RAP strongly suggests that this effect is a channel phenomenon rather than an effect on basolateral Na ϩ /K ϩ -ATPase activity. Nevertheless, since PKC stimulation has been shown to inhibit Na ϩ /K ϩ -ATPase activity in different renal cell lines (20), we measured Na ϩ /K ϩ -ATPase activity by two methods in cells exposed to RAP at the peak of the sodium transport response. There was no effect of RAP on either nystatin-stimulated sodium transport or enzymatic Na ϩ /K ϩ -ATPase activity. This provides strong support for the hypothesis that the effect of the PKC inhibition induced by RAP is through a sodium channel effect.
Finally, to determine if RAP inhibits PKC through interaction with FKBP12, we examined the effect of this protein on PKC activity (Table III). FKBP12 alone has no effect on PKC activity, but in the presence of RAP, PKC activity is significantly inhibited. This effect is partially reversed in the presence of a 10-fold excess of FK506 (Table III). The effect seems to be specific for the ␣ isoenzyme of PKC since there is no effect on the activity of the ␤ isoenzyme of PKC.
Taken together, these data demonstrate that RAP inhibits PKC activity in A6 cells through its interaction with FKBP12 and that inhibition of PKC is the mechanism by which RAP stimulates sodium transport. It thus seems likely that PKC exerts a tonic inhibitory effect on sodium channel activity as has been previously suggested (3,26). The finding that FK506 has no effect on sodium transport or PKC activity, although it binds tightly to FKBP12, represents another example of the peculiar effects determined by the immunophilin-immunophilin ligand complex. The ability of a single immunophilin to present two immunosuppressive ligands to effectors associated with two distinct pathways raises the possibility that the immunophilins may function as general presenting molecules, by analogy to the way that major histocompatibility complex molecules present a large number of peptides to the polymorphic T-cell receptors (1). This concept implies the existence of endogenous immunophilin ligands which, through their binding to the same immunophilin, would either activate or inhibit some downstream pathway depending on the cellular need (1,2).
PMA has been shown to stimulate phosphorylation of the 150-and 50-kDa subunits of the amiloride-sensitive sodium channel (26), and inhibition of PKC activity results in an increase in the open probability of the sodium channel in cellattached patches (3,24). It is likely that dephosphorylation of the channel itself or some channel regulatory protein results in an increase in channel opening. It is therefore interesting to postulate the existence of an endogenous RAP-like ligand that might ultimately regulate basal sodium channel activity through inhibition of PKC-associated phosphorylations. Identification of the relevant proteins that are regulated in this fashion will be the subject of future studies. Moreover, FKBP12 is associated with the ryanodine calcium channel (27)(28)(29) and is associated with calcium transients in human neutrophils (30). Since transepithelial sodium transport is dependent upon Na ϩ / Ca 2ϩ exchange (3,31) and cPKC is dependent on intracellular Ca 2ϩ concentration (32), this suggests that RAP and perhaps an endogenous Rap-like ligand might affect intracellular calcium as another means of altering sodium channel activity.