INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

PHOSPHATIDYLINOSITIDE 3-kinase (PI3K) is a member of the family of phosphoinositide kinases that acts to catalyze the phosphorylation of inositol phospholipids on the 3 position of the inositol ring (2, 7, 25, 31). The resulting 3-phosphoinositides are biologically active lipids, which appear to play a broad and complex role in eukaryotic cell regulation at a number of sites. These lipids have been observed to transduce both tyrosine kinase receptor signals (2, 19) and G protein-coupled receptor signals (27, 28). They have been implicated in the regulation of protein secretion (8, 33) and vesicle trafficking (11, 14). These lipids have also been observed to participate in regulation of the actin cytoskeleton (10, 12). Many of these observations have been made with the use of the fungal metabolite wortmannin, which specifically inhibits PI3K activity and has provided a powerful and selective tool with which to study the role of PI3K in diverse cellular functions (1).

Transepithelial sodium transport mediated by the amiloride-sensitive Na⁺ channel is regulated by a number of hormones, including aldosterone, insulin, and vasopressin (4, 9). Cellular mechanisms involved with regulating sodium transport include methylation of the channel (23), activation of either protein kinase A or C (PKA and PKC) (4, 21, 24, 32), activation of G proteins (6, 18, 22), and activation of tyrosine kinases (9, 16, 17). Activation of the Na⁺ channel may also be dependent on the actin cytoskeleton (4, 5). Since PI3K potentially interacts with many of these pathways, the role of this enzyme in hormonally regulated sodium transport was investigated by use of the PI3K inhibitor, wortmannin.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

A6 cells. All studies were performed on A6 cells grown on semipermeable supports. Cells were grown as described (22) in amphibian media (BioWhittaker, Walkersville, MD) with 10% fetal bovine serum (Sigma, St. Louis, MO) in an atmosphere of humidified air-4% CO₂ at 28°C. Cells were grown on Millicell-HA inserts (Millipore, Bedford, MA). Transepithelial potential difference and short-circuit current (I_sc) were measured using a sterile in-hood short-circuiting chamber as previously described (22). During incubations, wortmannin or LY-294002 was added to both sides of the semipermeable supports. Hormones (aldosterone, insulin, vasopressin) were added to the basolateral medium.

Chemicals. Aldosterone, arginine vasopressin, insulin, rapamycin, 8-bromo-cAMP, wortmannin, and forskolin were purchased from Sigma, St. Louis, MO. [-³²P]ATP was purchased from ICN. Anti-PI3K p85 antibody was obtained from Upstate Biotechnology. LY-294002 was obtained from Calbiochem, San Diego, CA. All other reagents were purchased from Sigma.

Protein kinase A activity. PKA activity was measured using a colorimetric assay for kemptide phosphorylation (Pierce). A6 cells were disrupted by passage through a small-gauge needle in a buffer containing 100 mM Tris · HCl, 1 mM EDTA, 1 mM dithiothreitol, 0.1 mM Na₃VO₄, 0.5 µM okadaic acid, and protease inhibitors (pepstatin A, leupeptin, antipain, and phenylmethylsulfonyl fluoride, each at 5 µg/ml) and 10% glycerol. The resulting suspension was subjected to centrifugation for 1 h at 100,000 g at 4°C. The resulting supernatant (crude cytosolic fraction) was protein matched, and PKA activity was measured.

PI3K activity. PI3K activity was measured on the immunoprecipitated enzyme as previously described (1). A6 cells on filters were exposed to 100 mU/ml arginine vasopressin or diluent for 10 min, and the cells were scraped from the filters. Cells were suspended in 10 mM Tris · HCl, 1% Triton X-100, 150 mM NaCl, 0.5% Nonidet P-40, 1 mM EDTA, 0.2 mM Na₃VO₄, 100 mM NaF, and protease inhibitors, pH 7.4. Cells were incubated in Eppendorf tubes at 4°C for 30 min, vortexing every 10 min. The samples were centrifuged at 14,000 g, and the supernatant containing PI3K was transferred to a fresh tube. PI3K was immunoprecipitated using the anti-PI3K antibody and GammaBind beads for 4 h at 4°C with rotation. The beads containing the PI3K subunit were then washed once in PBS, twice in 0.5 M LiCl, 0.1 M Tris-Cl, pH 7.4, and once in phosphatidylinositide kinase buffer (10 mM MgCl₂, 20 mM HEPES, pH 7.4). Beads were collected by centrifugation and resuspended in 40 µl of phosphatidylinositide kinase buffer containing 200 µg/ml phosphatidylinositol/phosphatidylserine each. The reaction was initiated with the addition of 10 µl of [-³²P]ATP (4,000 Ci/mmol). The samples were mixed every 2 min for 10 min at room temperature, and the reaction was terminated by the addition of 40 µl of 1 M HCl. Lipids were extracted from the organic phase after the addition of 80 µl chloroform/methanol (1:1). Samples were spotted on thin-layer chromatography plates and developed in chloroform/methanol/water/NH₃ (90:70:17:3). Radioactivity was quantified using Bio-Rad Molecular Analysis software.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

We first examined the effect of wortmannin on basal I_sc. When cells were incubated with 50 nM wortmannin for 3 h, there was no effect on I_sc (not shown). The effect of wortmannin on hormonally stimulated Na⁺ transport was then examined. Preincubation with wortmannin for 1 h and subsequent coincubation with aldosterone had no effect on the I_sc response to aldosterone (Fig. 1). Similarly, there was no effect of wortmannin on the I_sc response to insulin (Fig. 2). In contrast, preincubation with 50 nM wortmannin completely inhibited the I_sc response to a maximal concentration of vasopressin in these cells (Fig. 3). Similar results were obtained with simultaneous addition of wortmannin and vasopressin (not shown). To ensure that wortmannin was not acting nonspecifically, we employed a second inhibitor of PI3K, LY-294002 (29, 30). When LY-294002 was added to A6 cells at a concentration of 5 µM (three times its K_i of PI3K), there was complete inhibition of vasopressin-stimulated I_sc (Fig. 4).

View larger version (13K): [in this window] [in a new window]   Fig. 1.   Wortmannin has no effect on aldosterone-stimulated short-circuit current (Isc). A6 cells were preincubated in wortmannin (50 nM) or diluent for 1 h. Aldosterone (1 µM) was added at time 0, and Isc was measured sequentially; , aldosterone alone; , aldosterone + wortmannin. Results are means ± SE; n = 6.

View larger version (12K): [in this window] [in a new window]   Fig. 2.   Wortmannin has no effect on insulin-stimulated Isc. Cells were preincubated with wortmannin (50 nM) or diluent for 1 h, and then 100 mU/ml insulin was added to serosal solution. Isc is shown from time of addition of insulin (time 0); , insulin alone; , insulin + wortmannin; n = 6.

View larger version (13K): [in this window] [in a new window]   Fig. 3.   Wortmannin inhibits vasopressin-stimulated Isc. A6 cells were preincubated with wortmannin or diluent for 1 h. Arginine vasopressin (100 mU/ml) was added to serosal solution; , vasopressin alone; , vasopressin + wortmannin. Inhibition of vasopressin current is significant at each time point; n = 6.

View larger version (12K): [in this window] [in a new window]   Fig. 4.   LY-294002 inhibits vasopressin-stimulated Isc. A6 cells were incubated with vasopressin (100 mU/ml) or vasopressin and 5 µM LY-294002, added simultaneously; , vasopressin alone; , vasopressin + LY-294002. Inhibition of vasopressin current is significant at each time point; n = 6.

Inhibition of vasopressin action by wortmannin suggested that this hormone might act, in part, through stimulation of PI3K. To examine this possibility, we measured the effect of vasopressin on PI3K activity in A6 cells. PI3K was immunoprecipitated from A6 cells following exposure to vasopressin (100 mU/ml for 10 min) and compared with activity from control cells as described in METHODS. Phosphorylation of phosphatidylinositol was quantified by autoradiography and densitometry. Activity from control cells was 240 ± 10.9 U and from cells exposed to vasopressin was 408 ± 10.8 U (n = 3, P < 0.01). These results demonstrate that vasopressin stimulated PI3K activity in A6 cells. The next experiments sought to determine the possible site of wortmannin inhibition of vasopressin-stimulated Na⁺ transport. Vasopressin acts to increase apical Na⁺ channel number and activity through activation of adenylate cyclase and subsequent activation of PKA (16). Wortmannin (50 nM) has no effect on unstimulated PKA activity in A6 cells (wortmannin/control PKA activity, 1.04 ± 0.08; n = 4, P = not significant). Figure 5 demonstrates that wortmannin inhibits vasopressin activation of PKA activity in A6 cells. If PI3K is acting to couple receptor activation by vasopressin to enzyme stimulation, then agents that bypass these steps, such as cAMP or forskolin, might still stimulate transport even in the presence of PI3K inhibition. We therefore examined the effect of wortmannin on the transport response to cAMP and forskolin in A6 cells. Figure 6 shows the effects of a permeable cAMP analog, 8-bromo-cAMP on I_sc in A6 cells. Wortmannin does not inhibit the I_sc response to this agent. Figure 7 demonstrates that wortmannin does not inhibit the stimulation of I_sc mediated by forskolin, an agent known to stimulate adenylate cyclase independent of vasopressin receptor activation. There is an early (5 min) apparent enhancement of forskolin-stimulated I_sc by wortmannin, which is not sustained at 15 min.

View larger version (11K): [in this window] [in a new window]   Fig. 5.   Wortmannin inhibits vasopressin-induced increase in protein kinase A (PKA) activity. A6 cells were preincubated with wortmannin or diluent for 1 h and exposed to vasopressin (100 mU/ml) or diluent for 1 h. Cells were then scraped from filters, and PKA activity was measured as described in METHODS. Units for PKA activity are determined from a standard curve. Column 1, control PKA activity; column 2, PKA activity after vasopressin addition; column 3, PKA activity following vasopressin in presence of wortmannin; n = 4.

View larger version (12K): [in this window] [in a new window]   Fig. 6.   Wortmannin has no effect on 8-bromo-cAMP stimulation of Isc. Following 1-h preincubation with () or without () wortmannin, 2 mM 8-bromo-cAMP was added to basolateral surface at time 60 min; n = 12.

View larger version (12K): [in this window] [in a new window]   Fig. 7.   Wortmannin has no effect on forskolin-stimulated Isc. Effect of 2 µM forskolin on Isc in presence () or absence () of wortmannin (50 nM). Wortmannin or diluent was present for 1 h prior to addition of forskolin; n = 12.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

PI3K has been implicated in signal transduction of a number of cellular processes including G protein-coupled receptor signaling, activation of the actin cytoskeleton, tyrosine kinase-activated receptor signals, and secretion or targeting of proteins. Many of these processes have also been implicated in activation of the amiloride-sensitive epithelial Na⁺ channel by hormones. Aldosterone stimulation of Na⁺ channel activity is dependent on transcription and translation of new proteins and occurs in several phases. The early phase, from 1-4 h involves activation of preexisting channels without significant synthesis of new channels, whereas the late phase, from 4-12 h, may involve transcription of new channel subunits as well as basolateral membrane increase in Na⁺-K⁺-ATPase activity (4, 9, 13). The activation of channels occurs by one of several proposed mechanisms including G protein activation (9, 13, 22), channel subunit carboxylmethylation (23), or possibly phosphorylation (26). Insulin activation of channel activity is dependent on tyrosine kinase activity and appears to be primarily due to an increase in open probability of existing channels (9, 16, 17), although some studies have suggested an increase in total number of active channels (3). Finally, vasopressin activation of channel activity appears to be dependent on PKA activity, which appears to induce the insertion of new channels into the apical membrane of responsive epithelia (4, 9, 15). This study sought to examine the possible participation of PI3K in Na⁺ transport activation by use of the specific inhibitor of PI3K, wortmannin.

The results demonstrate that wortmannin has no effect on basal transport or transport stimulated by either aldosterone or insulin. The lack of an effect on insulin-stimulated Na⁺ transport is somewhat surprising, since insulin has been described as stimulating PI3K in other systems and this has been linked to membrane insertion of glucose transporters and stimulation of enzyme activity (8, 19). Although insulin activity in A6 cells is dependent on tyrosine kinase activity (17), subsequent PI3K activation does not appear to be involved in Na⁺ channel activation by insulin as judged by sensitivity to wortmannin. In contrast to the lack of effect of wortmannin on basal transport or transport stimulated by either aldosterone or insulin, transport stimulation by vasopressin is completely inhibited by wortmannin and by a second inhibitor of PI3K, LY-294002. This suggests that vasopressin might stimulate PI3K activity in A6 cells, and current results indicate that it does. Moreover, the PI3K inhibitor, wortmannin, blocks vasopressin stimulation of PKA. This result suggests that vasopressin stimulation of PKA activity is dependent on activation of PI3K as a signal transduction pathway linking receptor activation to enzyme stimulation as has been seen in other systems (27). Stimulation of transport by forskolin, which activates adenylate cyclase, and exogenous cAMP are both unaffected by wortmannin. It appears that channel activation or insertion into the apical membrane is not dependent on wortmannin-sensitive PI3K activity. These results indicate that wortmannin inhibits vasopressin action in A6 cells at a site prior to activation of adenylate cyclase and suggest that PI3K activity is required for vasopressin receptor activation of that enzyme in these cells.