Transactivation of the IGF-1R by aldosterone

【摘要】  Activation of epithelial sodium channels (ENaC) by aldosterone, insulin, or insulin-like growth factor-1 (IGF-1) in renal epithelial cells (including the Xenopus laevis renal cell line A6) appears to share some common signaling elements subsequent to the initial insulin or IGF-1 receptor activation. Previously, the convergence point for insulin or IGF-1 and aldosterone signaling was assumed to be downstream of the receptor at the level of phosphatidylinositol 3-kinase (PI3-K); however, this study shows aldosterone directly transactivates the IGF-1 receptor (IGF-1R). In A6 cells, 10-min exposure to aldosterone increased the phosphorylation of the IGF-1 receptor, insulin receptor substrate-1 (IRS-1), and Akt (PKB). Furthermore, aldosterone activated PI3-K and phosphorylation of the most downstream element, Akt, was blocked by the specific PI3-K inhibitor LY-294002. Transactivation requires aldosterone binding to the mineralocorticoid/glucocorticoid receptor and does not require transcription.

【关键词】  PIkinase insulinlike growth factor Akt A cells

IN THE TRADITIONAL ALDOSTERONE mechanism, aldosterone, like other steroid hormones, enters cells and binds to cytosolic steroid (mineralocorticoid) receptors. The steroid-bound receptor translocates to the nucleus, where it interacts with steroid response elements to regulate gene expression and, ultimately, protein expression.

Aldosterone is the primary hormone that regulates sodium transport in sodium-transporting epithelia by increasing the activity of epithelial sodium channels (ENaC). The effect of aldosterone on ENaC has been described in several different renal epithelial cells including the amphibian A6 renal cell line ( 55 ). The effect of aldosterone on sodium transport occurs in three phases ( 32 ). During the first (latent) phase, there is no increase in transepithelial sodium transport (measured as transepithelial sodium current, I te ). During the second (early) phase, which occurs between 30 min and 4 h, there is a rapid four- to sixfold increase in I te. Finally, the late phase of aldosterone action (after 3 h) involves an additional two- to fourfold increase in current and requires gene transcription, including the synthesis of new sodium channel mRNA ( 39 ).

Ultimately, aldosterone-mediated changes in ENaC activity control systemic blood pressure by regulating total body sodium content, which is directly correlated with blood volume and blood pressure.

In addition to the well-documented three phases of aldosterone's action, there are also rapid effects of this hormone, which are difficult to reconcile with the traditional view of the mechanism of aldosterone's action. In this traditional view, aldosterone binds mineralocorticoid or glucocorticoid receptors (MR/GR) with subsequent changes in gene expression that are necessary for aldosterone's effects. While aldosterone certainly does act in this manner, the rapid responses to aldosterone (and other steroids) do not appear to require gene transcription and may not even require activation of cytosolic steroid receptors ( 9, 12, 17, 19, 32, 41, 42, 54 ). While the long-term effects of aldosterone on ENaC subunit synthesis have been well documented, aldosterone also stimulates ENaC after 30 min to 1 h of aldosterone treatment, and this effect does not require the synthesis of new ENaC subunits ( 32 ). The question remains whether this short-term effect requires gene transcription and binding of aldosterone to cytosolic steroid receptors.

Some reports suggest that the rapid effects of aldosterone may be mediated by an as yet unidentified plasma membrane receptor ( 56, 59 ) rather than the MR/GR. Since the rapid action of aldosterone is not blocked by actinomycin D (ActD), the events appear not to require new gene transcription or protein translation ( 40 ). For example, in isolated rat heart, aldosterone has an immediate, positive, inotropic effect even in the presence of the mineralocorticoid receptor inhibitor spironolactone ( 1 ). Nonetheless, in many studies, once the effect of aldosterone is shown to be transcriptionally independent, the requirement for mineralocorticoid receptors is often not investigated.

Renal Na + reabsorption and ENaC activity are increased by aldosterone but are also increased by the growth factors insulin and IGF-1 ( 6, 44 ). Like some other steroid and growth factor signaling pathways, aldosterone and insulin/IGF-1 appear to share some common signaling elements. It has been suggested that the two signaling cascades, insulin/IGF-1 and aldosterone, converge at the level of phosphatidylinositol 3-kinase (PI3-K) because PI3-K is required for both aldosterone- and insulin-stimulated/regulated sodium reabsorption ( 5, 43 ). In the presence of the PI3-K inhibitor LY-294002, both insulin and aldosterone are incapable of increasing Na + transport through ENaC.

Blazer-Yost and other investigators ( 6, 23, 57 ) have shown that MDCK-C7 cells respond to IGF-1, but not insulin, due to a lack of insulin receptors (IRs); however, insulin and IGF-1 are each capable of activating each other's receptor, which can confuse the issue of which receptor is actually responsible for activating the signaling elements in common with aldosterone signaling. Since some IGF-1 signaling also requires PI3-K activity ( 34 ), it is possible that the effects of insulin on renal epithelial cells are actually due to insulin activation of IGF-1 receptors.

While several investigators have suggested that PI3-K is an integration point of the aldosterone and insulin/IGF-1 signaling pathways, the possibility that the pathways converge at a point before PI3-K activation has not been carefully investigated. IR substrate-1 (IRS-1) protein as well as the IR and the IGF-1 receptor (IGF-1R) are upstream of PI3-K in the peptide signaling pathways. Therefore, it is possible that the actual convergence point for insulin/IGF-1 and aldosterone could occur at a level higher than that of PI3-K.

Recent work has suggested that the interaction between steroid and peptide hormone signaling may occur through transactivation of growth factor receptors. Normally, growth factors bind to growth factor receptors leading to autophosphorylation of the receptors and subsequent phosphorylation of downstream signaling elements: in the case of insulin/IGF-1 receptors the first downstream element is IRS-1. However, under some circumstances, growth factor receptors can be transactivated, that is, activated by phosphorylation in the absence of ligand binding; the phosphorylation activates all the same downstream signaling elements that normal ligand binding will activate.

Aldosterone has been shown to transactivate the epidermal growth factor receptor (EGF-R) in Madin-Darby canine kidney (MDCK) cells ( 18 ). This effect is seen in the absence of EGF, although the effect is enhanced when EGF is present in addition to aldosterone. Not only is the EGF-R activated, as measured by changes in tyrosine phosphorylation, but also ERK1/2 phosphorylation is increased in a manner similar to EGF alone. These data show that aldosterone is capable of using the EGF-R pathway to elicit a rapid response in renal epithelium ( 19 ). Additional experiments performed in Chinese hamster ovary cells, which do not normally express EGF-R, have confirmed these results when the cells were transfected with the receptor; however, transactivation of EGF-R and activation of ERK1/2 by aldosterone does require low concentrations of EGF (that would normally not activate the receptor). The mechanism of EGF-R transactivation may involve activation by aldosterone of the cytosolic tyrosine kinase Src, and subsequent phosphorylation and activation of EGF-R. Furthermore, aldosterone enhanced the phosphorylation of c-Src, and its effect was abolished with the Src inhibitor 4-amino-5-(4-chloro-phenyl)-7-(t-butyl)pyrazolo[3,4-d]pyriociaine ( 28 ).

In this study, we show that the IGF-1R is transactivated in the presence of aldosterone in A6 renal epithelial cells from the kidney of Xenopus laevis. This means the two signaling pathways converge at the level of the receptor rather than farther downstream at the level of PI3-K. We have also shown that transactivation requires the MR/GR but does not require gene transcription.

METHODS

Cell culture. All experiments were performed using A6 cells (American Tissue Type Culture), subclone 2F3. Cells were cultured as described previously ( 35, 50 ), in a mixture of three parts Coon's F-12 medium and seven parts Leibovitz's L-15 medium, modified for amphibian cells with 104 mM NaCl, 25 mM NaHCO 3, pH 7.4, and supplemented with 10% fetal bovine serum, 1.5 µM aldosterone, 0.6% penicillin, and 1% streptomycin. 2F3 cell monolayers were plated on permeable 3.8-cm 2 inserts [either 0.02-µm Anopore membrane (Nalge NUNC International, Naperville, IL) or 0.4-µm polyester membrane (Costar)]. Cells were grown to confluence for all experiments and were washed twice with PBS buffer before serum and aldosterone deprivation for 48 h. Cells were then replenished with 1.5 µM aldosterone only for 10 min, 30 min, or 4 h, or with 100 nM IGF-1 for 10 min. While a wide range of aldosterone concentrations has been used to study A6 cells, we have chosen to use 1.5 µM aldosterone, since this is a concentration that has been used by other groups to produce maximum activation of Na + transport in A6 cells ( 22, 31 ). Figure 1 shows a dose-response curve for the effect of increasing concentrations of aldosterone on ENaC activity, serving as a further rationale for our concentration selection.

Fig. 1. Aldosterone activates the amiloride-sensitive epithelial Na + channel (ENaC) in a dose-dependent manner. ENaC activity was measured using single-channel methods as previously described ( 21, 38 ) as the number of channels ( N ) x channel open probability ( P o ) = NP o. Cells were deprived of serum and steroid for 48 h, after which they were exposed to aldosterone at varying concentrations (0.001-100 µM) for 30-45 min and activity was recorded. The K 1/2 is 0.021 ± 0.0046 µM. In subsequent experiments, we used 1.5 µM aldosterone to achieve 98-99% activation of signaling pathways leading to ENaC.

I te measurements. The transepithelial voltage ( V te ) and transepithelial resistance ( R te ) across cell monolayers were measured using an epithelial volt-ohmmeter equipped with chop-stick electrodes (World Precision Instruments, Sarasota, FL). The equivalent short-circuit current ( I sc ) was calculated according to Ohm's law ( I sc = V te / R te ) and then corrected for the surface area of the insert.

Western blotting. On the basis of space considerations, the blots using insulin are not shown. Furthermore, insulin was not used in any other experiments. A6 cells were washed twice with PBS buffer and then harvested on ice with RIPA buffer: 10 mM NaPO 4, 150 mM NaCl, 1% sodium deoxycholate, 1% NP-40, 0.1% SDS, containing protease inhibitors (phenylmethylsulfonyl fluoride, leupeptin, antipain, PLCK, and TPCK). Phosphatase inhibitors were not required to observe increased phosphorylation of IGF-1R signaling proteins in A6 cells. In the presence of protease inhibitors, phosphorylation of Akt at serine 473 increases in the presence of both aldosterone (1.5 µM) and IGF-1 (100 nM). The addition of phosphatase inhibitors does not change the amount of phosphorylation in the presence of aldosterone or IGF-1.

After standardization of total protein by the DC-protein assay (Bio-Rad) and addition of sample buffer (glycerol, SDS, and DTT), lysates were boiled at 95°C for 10 min. Proteins were separated by SDS-PAGE (7.5% gels) and subsequently transferred to nitrocellulose. Western blot analysis was performed by blocking the nitrocellulose in either 5% BSA, 5% milk, or 3% milk, each with 0.1% Tween 20, and then probing with antibodies to detect signaling proteins in the IGF-1 pathway. Phospho-Y1131IGF-1R, phospho-S473 Akt, and Akt antibodies were from Cell Signaling Technology (Beverly, MA); PI3-K p85 -subunit and IRS-1 antibodies were from Upstate (Lake Placid, NY), and phospho-Y632IRS-1, IR, IR, IGF-1R, IGF-1R, and pan-p110 PI3-K subunit were from Santa Cruz Biotechnology (Santa Cruz, CA). GAPDH antibodies were from Santa Cruz Biotechnology or Chemicon (Temecula, CA). Immunoblots were incubated with the primary antibodies overnight at 4°C and then with a species-appropriate horseradish peroxidase (KPL, Gaithersburg, MD)- or alkaline phosphatase (Tropix/Applied Biosystems, Foster City, CA)-conjugated secondary antibody for 1 h.

Western blots were developed using either the ECL detection system (LumiGlo, Cell Signaling Technology) or CDP-Star (Applied Biosystems, Foster City, CA). Blots were quantified using Kodak 1D Image Analysis.

Immunoprecipitation/coimmunoprecipitation. Immunoprecipitation of p110 of PI3-K was performed on cellular lysates treated with or without aldosterone for 30 min. The primary (pan-p110, Santa Cruz Biotechnology) antibody was pulled out of the lysates using ImmunoPure Plus Immobilized Protein A beads at room temperature, followed by four washes of the beads with lysis buffer. Samples were then boiled in sample buffer with DTT for 10 min, followed by rocking for 30 min at room temperature, after which supernatants were collected and the boiling/rocking step was repeated. The supernatants from each step were combined and concentrated at a high temperature in a SpeedVac for 5 min, and then they were subsequently loaded on a 7.5% SDS-PAGE gel. Western blotting with p85 PI3-K (Upstate) was performed following the procedure outlined above.

Quantative RT-PCR. Quantitative real-time PCR (Q-RT-PCR) of cDNA reverse transcribed from RNA from A6 cells exposed to aldosterone for 24 h was performed with receptor-specific primers for the IGF-1R and IR subunits, designed according to published sequences for X. laevis (GenBank). Selected primers were tested for primer-dimer formation using Qiagen Oligo analysis and were BLAST searched against the sequence database to ensure further specificity. PCR was performed using these primers to ensure that a single fragment of expected size was generated for each set. Additionally, the PCR products were not seen in control reactions with RNA only, indicating that no genomic DNA was present in the RNA used to generate the cDNA. Melting curves were run for all primer pairs to ensure primer specificity and appropriate binding. Q-RT-PCR was performed using SYBR Green and iCycler software (Bio-Rad). PCR efficiencies were calculated using the equation [10 (1/-s) ] - 1, where s = slope, and the relative copy number difference was calculated by using the equation n = f ( Ct), where f = the PCR efficiency and Ct = the difference in cycle numbers between the groups ( 20 ).

Chemicals and solutions. Chemicals were obtained from Sigma, Fisher, Bio-Rad, and Calbiochem. RU-486 was a generous gift from Dr. S. Russ Price, Emory University.

Statistical analysis. Data are reported as means ± SE. Statistical analysis was performed using SigmaStat software (Jandel Scientific). ANOVA or t -tests were used to compare the data when appropriate using the Holm-Sidak posttest. P < 0.05 was considered significant.

RESULTS

A6 cells have much more IGF-1R than IR. While previous studies have identified the presence of both IGF-1R and IR in toad urinary bladder using electron microscopy ( 7 ), there appear to be no studies addressing the presence of IGF-1R in A6 cells. While some tissues express more IR (liver, fat, muscle cells), most cells express more IGF-1R than IR ( 4 ). We used Western blots to determine the relative amount of IR and IGF-1R. Both IR and IGF-1R are composed of two subunits, and, both derived from the same propeptide and both necessary for function. We were unable to detect IR by Western blotting in A6 cells, although the antibody did weakly detect some IR (possibly cross-reactivity with IGF-1R ); however, both subunits of the IGF-1R ( and ) were visible (data not shown). To further address the issue of receptor subunit presence, we performed Q-PCR for both the IR and IGF-1R and found a 4.8-cycle difference in the average threshold crossings for 24 separate experiments with primers for each subunit ( P < 0.001 between groups). Because the PCR reaction efficiencies were both 94%, we were able to calculate that that there is at least 20 times more mRNA present for IGF-1R in A6 cells than mRNA for IR [fold-difference = f ( Ct), where f = 1.88 and Ct = 4.8 cycles]. This is the minimum difference since the threshold crossing for the IR is near the threshold crossing limit in the absence of DNA. This serves as further evidence that there is much less IR present in A6 cells, and therefore it is most likely the IGF-1R and its signaling pathway that are being activated in our system.

A6 cells are more responsive to IGF-1 than insulin. The ability of IGF-1 and insulin to increase Na + transport through ENaC has been documented in the literature ( 6 ). To confirm that our cells were able to respond to IGF-1 and insulin, cells were treated with insulin (20 nM) and IGF-1 (10 and 100 nM) in addition to the control. As seen in Fig. 2, there is a dramatic increase in I sc in the presence of either the low or high concentration of IGF-1; however, the response to 20 nM insulin is significantly lower. This suggests, in combination with the Q-RT-PCR data, that A6 cells respond better to IGF-1 than insulin due to the presence of more IGF-1Rs in the cell membrane.

Fig. 2. IGF-1 increases ENaC-mediated Na + transport more than insulin. A6 cells respond to IGF-1 more than insulin. Insulin (20 nM) and IGF-1 (10 or 100 nM) were applied to cells on the basolateral side of the monolayer, and transepithelial voltage and resistance measurements were taken, from which transepithelial current was calculated. The response of A6 cells to 10 nM IGF-1 was nearly identical to 100 nM IGF-1 from 60 min onward. That 10 nM IGF-1 increases Na + transport significantly more than 20 nM IGF-1 again suggests that the IGF-1 receptor (IGF-1R) is present in a higher quantity in A6 cells. I sc, short-circuit current. Values are means ± SE; n = 6 for each data point. ** P < 0.05 for insulin vs. control. * P < 0.05 between both IGF-1 conditions and insulin, and both IGF-1 conditions and control.

Aldosterone increases the phosphorylation of IGF-1R. Phosphorylation of the IGF-1R is necessary for receptor activation. Typically, this is an autophosphorylation event in response to ligand that takes place within the tyrosine kinase domain (residues 973-1229) on the cytoplasmic side of the -subunit ( 34 ). Furthermore, three of these tyrosine residues are initially phosphorylated and are essential for activation of the IGF-1R: Y1131, Y1135, and Y1136 ( 25 ). Therefore, we examined the phosphorylation of the IGF-1R in A6 cells that were exposed to either no hormone, aldosterone (1.5 µM), or IGF-1 (100 nM). All exposures were for 10 min. Lysates were probed with a phospho-specific antibody against tyrosine 1131 (Y1131) of the IGF-1R since it is one of the residues required for receptor activation. Blots were then stripped and reprobed with an antibody against IGF-1R as well as GAPDH to quantify and confirm equal loading, respectively. As seen in Fig. 3, A and B, phosphorylation of the IGF-1R at Y1131 increases significantly in the presence of aldosterone. Simultaneously, we used IGF-1 as a positive control to show activation of the IGF-1R. Previous reports have shown that the IGF-1R binds IGF-1 with a high affinity, although insulin can also bind to the IGF-1R, but with a much lower affinity ( 33 ). In fact, in another experiment (data not shown) 100 nM insulin produces less phosphorylation of Y1131 than 100 nM IGF-1 or 1.5 µM aldosterone. This result also serves as confirmation that the IGF-1R is being activated as opposed to the IR, since insulin is incapable of producing as large an effect as either aldosterone or IGF-1.

Fig. 3. Aldosterone increases the phosphorylation of IGF-1R at Y1131. A : representative Western blots of phosphorylated IGF-1R and total IGF-1R from A6 cells exposed to aldosterone (Aldo; 1.5 µM) or IGF-1 (100 nM, positive control) for 10 min each. All conditions were deprived of serum and hormones for 48 h, followed by the 10-min treatment; control lanes were replenished with serum- and hormone-free media containing vehicle only (negative control). pY1131 IGF-1R blots were stripped and reprobed with a non-phospho-IGF-1R antibody and with GAPDH. Phosphorylation of IGF-1R appears to be comparable in the presence of either aldosterone or IGF-1. B : summary graph of relative phosphorylation of the IGF-1R at Y1131. On the basis of space considerations, the blots using insulin are not shown. Furthermore, insulin was not used in any other experiments. * P < 0.05 between aldosterone and control; n = 3 for each bar.

Aldosterone increases the phosphorylation of IRS-1. IRS-1 is the first downstream element from the IGF-1R (and IR). The IRS protein family contains six isoforms, two of which, IRS-1 and IRS-2, have been extensively studied for their role in metabolic signaling pathways. In human IRS-1, there are several critical tyrosine residues that must be phosphorylated for the protein to become fully activated and continue the IGF-1 signaling cascade ( 14 ). Therefore, we examined one of these, Y632, with a commercially available phospho-antibody to confirm that IRS-1 is being phosphorylated after aldosterone had activated IGF-1R ( Fig. 4 A ). Blots were stripped and reprobed with IRS-1 Ab for quantification ( Fig. 4 B ). The increase in phosphorylation is seen not only in the experimental group treated with 1.5 µM aldosterone but also in the positive control (cells exposed to 100 nM IGF-1).

Fig. 4. Phosphorylation of IRS-1 occurs in the presence of aldosterone. A : representative Western blot showing changes in phosphorylation of Y632, a critical residue for activation of IRS-1. Reported molecular mass of IRS-1 is 165-185 kDa. Cells were exposed to aldosterone (1.5 µM) or IGF-1 (100 nM) for 10 min, after 48 h of serum and hormone deprivation. Control cells were serum/hormone deprived and were then replenished with the same media during the experiment. Middle : blot reprobed from top with a non-phospho-IRS-1 antibody; bottom : blot reprobed with GAPDH. B : summary graph of relative phosphorylation between all 3 conditions. * P < 0.05 between aldosterone/IGF-1 and control for 3 separate experiments. C : representative Western blot showing the absence of IRS-2 in A6 cells. IRS-2 has a reported molecular mass of 175 kDa and is present in muscle lysates from Xenopus laevis hindlimb. However, it is absent in hormone-deprived, aldosterone-, and IGF-1-treated A6 cells.

Although, as demonstrated above, aldosterone promotes phosphorylation of IRS-1, in some cells insulin (or IGF-1) also activates IRS-2. IRS-2 is more strongly activated by the IR than by the IGF-1R ( 27, 45 ). Therefore, to obtain further evidence that we are primarily observing effects of an IGF-1R-mediated pathway, we probed for the presence of IRS-2 in our A6 model renal epithelial cell line. As seen in Fig. 4 C, there is no IRS-2 present in A6 cells either in the absence of hormone or when exposed to either 1.5 µM aldosterone or 100 nM insulin. However, the antibody is capable of detecting IRS-2 in X. laevis muscle lysates, showing that the antibody is functional in X. laevis.

Aldosterone causes a change in the association of PI3-K subunits. Previous reports have claimed that PI3-K is required for both aldosterone- and insulin-regulated sodium transport in A6 cells ( 5, 43 ), and many others have shown that IGF-1 acts through PI3-K by promoting association of the p85 regulatory subunit with the p110 catalytic subunit of PI3-K ( 2, 8, 24, 26, 30, 37, 47, 51, 53, 58 ). We wanted first to reconfirm the requirement for PI3-K by measuring I te in the presence of a classic inhibitor of PI3-K activity, LY-294002. As seen in Fig. 5 A, cells exposed to aldosterone showed an increase in current, while cells that were pretreated with 50 µM LY-294002 for 30 min before aldosterone exposure have less current than the steroid-free control. Besides Na + transport, activation of PI3-K also activates the signaling molecule Akt, by phosphorylation, and Akt phosphorylation is considered a marker of PI3-K activation. Therefore, we performed an additional experiment using cells pretreated with LY-294002 for 30 min, followed by aldosterone treatment for 10 min. Cells were then probed with pSer473-Akt, a downstream marker of IGF-1R pathway activation. In the presence of aldosterone, there is significant phosphorylation of Akt; in the presence of aldosterone and LY-294002, there is little Akt phosphorylation ( Fig. 5 B ).

Fig. 5. A : transepithelial current measurements were taken from A6 cells grown on permeable supports. After serum and aldosterone deprivation for 48 h, cells were pretreated with LY-294002 (50 µM), both apically and basolaterally, for 30 min. LY-294002 was then replenished in either the absence or presence of 1.5 µM aldosterone, and currents were measured over a 4-h time course. At the end of the experiment, 10 µM amiloride (AMIL) was added to the cells for 10 min to confirm that current measurements were a result of activation of the ENaC. ** P < 0.05 between aldosterone and all other conditions. * P < 0.05 between no aldosterone and LY conditions with and without aldosterone; n = 6 for each data point. Values are means ± SE. B : representative Western blot for pSer473 Akt, a downstream signaling element of phosphatidylinositol 3-kinase (PI3-K). Cells were pretreated with LY-294002 (50 µM) for 30 min to block PI3-K, followed by a 10-min treatment with or without aldosterone (1.5 µM). Blots were then probed with pSer473 Akt to show that in the presence of LY-294002, downstream signaling by PI3-K is blocked. The blots were subsequently stripped and reprobed with Akt antibody to demonstrate equal initial starting amounts of total Akt.

To further measure activation of PI3-K, we measured aldosterone-induced association of the two PI3-K subunits, p85 and p110, which is necessary for PI3-K activation. Cells were treated with or without aldosterone for 30 min (lysates were generated using Triton X-100 buffer instead of RIPA to maintain protein association). Immunoprecipitation was performed with p85, followed by Western blotting with p110. As shown in Fig. 6, there is an increase in the association of the two subunits in the presence of aldosterone.

Fig. 6. Aldosterone causes an increase in the association of PI3-K subunits. A : representative immunoprecipitation (IP) of the p110 subunit of PI3-K, using a pan-p110 antibody, followed by Western blotting for the p85 subunit. IB, immunoblotting. Cells were exposed to aldosterone (1.5 µM) for 30 min, or remained serum and hormone deprived for the entire experiment. An increase in association between the 2 subunits can be seen in the aldosterone-treated cells over the control. B : summary graph showing the increase in density of the p85/p110 PI3-K complex between control and aldosterone-treated cells. * P < 0.05; n = 3 separate experiments.

Aldosterone increases the phosphorylation of Akt. The protein signal that serves as the hallmark of the activation of the IGF-1R cascade is Akt. After activation of IRS-1 through binding with activated IGF-1R, first PI3-K, and finally Akt, are activated ( 11, 29 ); phosphorylation of Akt leads to its activation ( 49 ). In the presence of both aldosterone and IGF-1, phosphorylation of Akt at serine 473 (pSer473Akt) increases significantly over the control ( Fig. 7, A and B ). The activation of this downstream element in the presence of only aldosterone (no serum, no other hormones, no growth factors) is final confirmation of the activation of the IGF-1R signaling pathway by aldosterone.

Fig. 7. Aldosterone increases the phosphorylation of Akt. A : representative Western blots showing the phosphorylation of Akt by aldosterone at serine 473 (S473). Bottom : blot stripped and reprobed from top with a nonphosphorylated pan-Akt antibody. Cells were treated with aldosterone (1.5 µM), IGF-1 (100 nM), or no hormone for 10 min. The increase in Akt phosphorylation in the presence of IGF-1 is expected to be greater than aldosterone because IGF-1 binds its receptor and has been well documented to robustly phosphorylate Akt in an antiapoptotic manner. The blots were subsequently stripped and reprobed with Akt antibody to demonstrate equal initial starting amounts of total Akt ( bottom ). B : summary graph of relative protein densities. * P < 0.05 for aldosterone/IGF-1 vs. control; n = 4 separate experiments.

Rapid activation of Akt by aldosterone is not dependent on transcription, while long-term Na + transport is. Aldosterone typically exerts its effect by binding to MR/GR, after which the steroid-bound receptor translocates to the nucleus where the steroid-bound receptor acts as a transcription factor, controlling the synthesis of new proteins. However, aldosterone-mediated transactivation of IGF-1R does not necessarily require gene expression or even aldosterone binding to the MR/GR. To test whether the phosphorylation of Akt by aldosterone involves transcription, we treated cells with ActD at a concentration of 1 µg/ml for 30 min, followed by 1.5 µM aldosterone for 10 min. ActD did not affect aldosterone-induced Akt phosphorylation or total Akt protein levels ( Fig. 8 ).

Fig. 8. Actinomycin D (ActD) does not inhibit Akt phosphorylation in the presence of aldosterone. A : representative Western blots showing phosphorylated Akt ( top ) and Akt ( bottom ). Cells were pretreated for 30 min with ActD (1 µg/ml), followed by 10 min of treatment with 1.5 µM aldosterone, or no hormone. ActD was replenished with the hormone treatment. B : summary graph of relative band densities taken as the ratio of pSerAkt to Akt; n = 3 separate experiments. * P < 0.05 vs. control.

Since aldosterone typically exerts its effect on ENaC by increasing the levels of new ENaC subunits at the membrane ( 3 ), there is a component of the effect of aldosterone that requires transcription. Previous reports have described the inhibition of Na + transport in renal cells by ActD ( 13, 16 ). To confirm that the long-term effect of aldosterone on Na + transport requires protein synthesis in A6 cells, cells were treated with no hormone, aldosterone, or IGF-1, in either the presence or absence of 1 µg/ml ActD. As seen in Fig. 9, Na + transport is rapidly and robustly increased by IGF-1 within 15 min of treatment, while aldosterone requires a much longer time period to have a large effect. In the presence of ActD, however, Na + transport is not immediately inhibited. Instead, ActD does not begin to affect Na + transport until 1 h after treatment. Of interest is that the effect of IGF-1 is also inhibited in the presence of ActD, but not to the degree that aldosterone is inhibited.

Fig. 9. ActD prevents aldosterone-induced, and decreases IGF-1-induced, Na + transport through ENaC. Cells were serum and hormone deprived, followed by either a pretreatment with ActD (1 µg/ml) for 30 min, or continued serum/hormone deprivation. Transepithelial voltage and resistance measurements were taken, and transepithelial current was calculated. At the end of the experiment, cells were treated with 10 µM amiloride for 10 min to ensure that the currents being observed were due to Na + transport through ENaC. Values are means ± SE; n = 6 for each data point. There is no significant difference between any of the 6 conditions at time 0 when hormone is applied and after amiloride treatment at the end of the experiment. * P < 0.05 between IGF-1 compared with all other conditions. # P < 0.05 between Aldo+ActD and No aldo+ActD at 15 min. P < 0.05 between IGF-1+ActD and all other conditions at all time points except No aldo and aldosterone at 8 h, and No aldo+ActD at 24 h. P < 0.05 between aldosterone and control, and aldosterone vs. aldosterone+ActD. P < 0.05 between aldosterone and all other conditions.

Rapid activation of IGF-1R signaling elements is dependent on the MR/GR. In A6 cells, aldosterone is a ligand for both the MR and the GR: activation of either receptor by an appropriate ligand will increase Na + transport ( 10, 48 ). We have shown above that aldosterone transactivates the IGF-1R. The transactivation could either be due to a direct effect of aldosterone on the IGF-1R, an effect of aldosterone binding to some hitherto unidentified membrane receptor, or it could involve binding of aldosterone to its steroid receptor even though receptor-mediated gene expression is not necessary for transactivation. The reorganization of the steroid receptor complex when it binds steroid (either MR or GR) itself represents a signaling event since several signaling molecules are released from the receptor complex after steroid binding. In fact, Moura and Worcel ( 40 ) demonstrated a rapid Na + efflux that is blockable by the MR antagonists spironolactone and RU-28318, yet is unaffected by ActD, showing that the rapid effect of aldosterone can be dependent on the MR without requiring transcription. Figure 10 shows that, in the presence of aldosterone, phosphorylation of Akt at serine 473 occurs after 10 min. This effect is eliminated in the presence of either 10 µM spironolactone (MR inhibitor), 10 µM RU-486 (GR inhibitor), or a combination of the two inhibitors, demonstrating that aldosterone-mediated transactivation requires aldosterone binding to the MR/GR but does not require gene transcription.

Fig. 10. Spironolactone and RU-486 inhibit the rapid actions of aldosterone on the IGF-1R signaling cascade. A : representative Western blots showing changes in phosphorylation of Akt by aldosterone at serine 473 in the presence or absence of 10 µM spironolactone and 10 µM RU-486. Cells were pretreated with the inhibitors for 30 min, followed by treatment with aldosterone (1.5 µM) for 10 min. The expected increase in Akt phosphorylation after 10 min of aldosterone treatment occurs; however, this phosphorylation is diminished in the presence of either spironolactone, RU-486, or both inhibitors. Blots were stripped and reprobed for total Akt ( bottom ). B : summary graph of relative protein densities. The addition of spironolactone or RU-486 prevents the phosphorylation of Akt in the presence of aldosterone. * P < 0.05 between column 2 (aldosterone only) and all other conditions; n = 3 separate experiments.

DISCUSSION

In this study, we examined the transactivation of the IGF-1R by aldosterone. We have demonstrated that not only does aldosterone transactivate the IGF-1R by phosphorylation in the absence of IGF-1, the normal ligand for the receptor, but also that key downstream components of this signaling pathway are activated in the presence of aldosterone. Furthermore, this effect is specific for the IGF-1R, as it does not appear to be present in sufficient quantities to be important in our model system, the A6 X. laevis renal epithelial cell line. Finally, this effect is still dependent on the interaction of aldosterone with its classic receptor; however, steroid receptor-induced transcriptional activation is not required.

Transactivation of growth factor receptors is a mechanism by which alternative signaling molecules (like steroids) can use the same signal transduction pathways as growth factors despite not being natural extracellular ligands for the receptors. Thus such transactivation can provide greater signaling flexibility with the same complement of signaling elements already within the cell. Steroids and steroid receptors typically produce their effects by promoting gene expression. Steroid-induced transactivation not only allows steroids to activate an alternative pathway but also does not require immediate changes in gene expression (although Akt may, in the long term, lead to changes in gene expression, albeit, different genes from those activated by steroid receptors) ( 46 ).

Others have shown that there are membrane receptors for some steroids (like estrogen) ( 15 ) and have suggested that there also might be a similar, but as yet unidentified, receptor for aldosterone ( 36 ). Despite the rapidity of transactivation, our data show ( Fig. 10 ) that it is unnecessary to invoke such a receptor to explain transactivation since all of the IGF-1R phosphorylation and Akt activation can be blocked by inhibiting the classic, cytosolic mineralocorticoid or glucocorticoid receptors. Nonetheless, our data also show that even though MR/GR are required for transactivation, the transactivation of IGF-1R does not require transcription ( Fig. 8 ). Previous research has shown that calcineurin activity is increased within 30 min of aldosterone treatment and that this effect is inhibited by spironolactone, but not by ActD ( 52 ). This evidence suggests a possible role for heat shock proteins and the reorganization of the steroid receptor complex proteins in the mediation of the IGF-1R transactivation signaling cascade as well, since we have observed that the effect is blocked by spironolactone and RU-486, but not ActD.

Aldosterone can transactivate other receptor tyrosine kinases (growth factor receptors). Specifically, EGF-R can be rapidly (10 min) transactivated by aldosterone in both transfected and native systems ( 18, 19, 28 ). However, aldosterone-mediated transactivation of EGF-R appears to require the addition of EGF at low concentrations (below that which would normally activate the receptor). By eliminating all serum-derived growth factors and other peptide hormones and treating A6 cells only with aldosterone, we have shown that aldosterone alone is sufficient to activate the IGF-1 signaling pathway. We have not tested, but speculate that low levels of IGF-1 would potentiate aldosterone-induced transactivation of IGF-1R.

The present work has shown that exposure to aldosterone for 10 min activates the IGF-1R signaling cascade in A6 cells. Phosphorylation of the IGF-1R pathway signaling components, IRS-1 and Akt, as well as the changes in the association of the PI3-K subunits suggest that transactivation of the IGF-1R by aldosterone may be a mechanism for early-phase aldosterone signaling in A6 cells. The maintenance of ENaC current until 1 h after ActD treatment suggests that this earliest phase of aldosterone's effect on ENaC current is the result of this rapid, nontranscriptional/translational effect ( Fig. 9 ). Whether the signaling through IGF-1R transactivation is sustained or whether it only contributes for a short period during the early phase is unclear and will require more experiments to determine the time course of transactivation. If the receptor were activated by IGF-1, it would desensitize by being internalized. Whether a transactivated receptor is also internalized and, thereby, inactivated is unknown. If a transactivated receptor were not internalized, then transactivation would produce synergy because desensitization would be decreased. Therefore, it is possible that desensitization is impacted by the presence of aldosterone, allowing for the potentiation of the IGF-1 effect. Additionally, the temporal responses to aldosterone and IGF-1 may influence the synergistic effect. While the response to IGF-1 is rapid, the response to aldosterone is longer and more sustained. Therefore, the total activation of Na + transport through ENaC would appear to be larger because the duration of the combined IGF-1 and aldosterone response would be longer.

The mechanism by which aldosterone is capable of transactivating IGF-1R is not completely clear. Gekle et al. ( 28 ) suggest that aldosterone activates c-Src, which promotes EGF-R transactivation. Experiments investigating the role of Src or other small cytosolic tyrosine kinases will be performed to determine whether these signaling components are involved in IGF-1R transactivation in A6 cells.

In conclusion, our data show that aldosterone is capable of inducing the phosphorylation of the IGF-1R, leading to transactivation of the receptor. Transactivation leads to the subsequent activation of the classic IGF-1R signaling cascade, wherein IRS-1, PI3-K, and Akt are all activated. The changes in phosphorylation of the IGF-1R, IRS-1, and Akt, as well as the increased association of the PI3-K subunits in the presence of aldosterone, imply that transactivation should increase ENaC activity since PI3-K activates both Akt and the homologous kinase, i.e., serum- and glucocorticoid-dependent kinase ( Fig. 11 ). Although it is widely known and accepted that aldosterone, insulin, and IGF-1 all cause an increase in Na + transport and that the two pathways converge no later than at the level of PI3-K ( 5, 43 ), our current study shows that the convergence point for aldosterone and insulin/IGF-1 is farther upstream, at the level of IGF-1R transactivation. We feel that the activation of the IGF-1R is a rapid, transcription/translation-independent response to aldosterone and that this receptor transactivation event that may help to explain some of the early responses to aldosterone that have previously been somewhat perplexing.

Fig. 11. Flow chart of cellular signaling elements. Aldosterone enters the cell and is capable of transactivating the IGF-1R and its pathway signaling components in A6 renal epithelial cells. The effect on Akt is blocked by the PI3-K inhibitor LY-294002. While the steps between aldosterone and activation of the IGF-1R are still unclear, the mineralocorticoid/glucocorticoid receptor (MR/GR) complex is required for the rapid action of aldosterone.

GRANTS

This research was supported by National Institute of Diabetes and Digestive and Kidney Diseases Grants P01-DK-064399 and R37-DK-037963 to D. C. Eaton.

ACKNOWLEDGMENTS

We thank Otor Al-Khalili for assistance with Q-RT-PCR.

【参考文献】
  Barbato J, Mulrow P, Shapiro J, Franco-Saenz R. Rapid effects of aldosterone and spironolactone in the isolated working rat heart. Hypertension 40: 130-135, 2002.

Benito M, Valverde A, Lorenzo M. IGF-I: a mitogen also involved in differentiation processes in mammalian cells. Int J Biochem Cell Biol 28: 499-510, 1996.

Bhalla V, Soundararajan R, Pao A, Li H, Pearce D. Disinhibitory pathways for control of sodium transport: regulation of ENaC by SGK1 and GILZ. Am J Physiol Renal Physiol 291: F714-F721, 2006.

Blakesley V, Scrmgeour A, Esposito D, LeRoith D. Signaling via the insulin-like growth factor-I receptor: does it differ from insulin receptor signaling? Cytokine Growth Factor Rev 7: 153-159, 1996.

Blazer-Yost B, Paunesca T, Helman S, Lee K, Vlahos C. Phosphoinositide 3-kinase is required for aldosterone-regulated sodium reabsorption. Am J Physiol Cell Physiol 277: C531-C536, 1999.

Blazer-Yost B, Record R, Oberleithner H. Characterization of hormone-stimulated Na + transport in a high-resistance clone of the MDCK cell line. Pflügers Arch 432: 685-691, 1996.

Blazer-Yost B, Shah N, Jarett L, Cox M, Smith R. Insulin and IGF1 receptors in a model renal epithelium: receptor localization and characterization. Biochem Int 28: 143-153, 1992.

Castrillo A, Bodelon O, Bosca L. Inhibitory effect of IGF-I on type 2 nitric oxide synthase expression in Ins-1 cells and protection against activation-dependent apoptosis: involvement of phosphatidylinositol 3-kinase. Diabetes 49: 209-217, 2000.

Christ M, Eisen C, Aktas J, Theisen K, Wehling M. The inositol-1,4,5-triphosphate system is involved in rapid effects of aldosterone in human mononuclear leukocytes. J Clin Endocrinol Metab 77: 1452-1457, 1993.

Claire M, Machard B, Lombes M, Oblin M, Bonvalet J, Farman N. Aldosterone receptors in A6 cells: physicochemical characterization and autoradiographic study. Am J Physiol Cell Physiol 257: C665-C677, 1989.

Delafontaine P, Song YH, Li Y. Expression, regulation, and function of IGF-1, IGF-1R, and IGF-1 binding proteins in blood vessels. Arterioscler Thromb Vasc Biol 24: 435-444, 2004.

Doolan C, Harvey B. Modulation of cytosolic protein kinase C and calcium ion activity by steroid hormones in rat distal colon. J Biol Chem 271: 8763-8767, 1996.

Edelman I, Bogoroch R, Porter G. On the mechanism of action of aldosterone on sodium transport: the role of protein synthesis. Proc Natl Acad Sci USA 50: 1169-1177, 1963.

Esposito D, Li Y, Cama A, Quon M. Tyr612 and Tyr632 in human insulin receptor substrate-1 are important for full activation of insulin-stimulated phosphatidylinositol 3-kinase activity and translocation of GLUT4 in adipose cells. Endocrinology 142: 2833-2840, 2001.

Falkenstein E, Tillmann H, Christ M, Feuring M, Wehling M. Multiple actions of steroid hormones-a focus on rapid, nongenomic effects. Pharmacol Rev 52: 513-555, 2000.

Gaeggeler NP, Gonzalez-Rodriguez E, Jaeger N, Logging-Cueni D, Horregaard R, Loffing J, Horisberger JD, Rossier B. Mineralocorticoid versus glucocorticoid receptor occupancy mediating aldosterone-stimulated sodium transport in a novel renal cell line. J Am Soc Nephrol 16: 878-891, 2005.

Gekle M, Freudinger R, Mildenberger S, Schenk K, Marschitz I, Schramek H. Rapid activation of Na + /H + exchange in MDCK-cells by aldosterone involves MAP kinases ERK1/2. Pflügers Arch 441: 781-786, 2001.

Gekle M, Freudinger R, Mildenberger S, Silbernagl S. Aldosterone interaction with epidermal growth factor receptor signaling in MDCK cells. Am J Physiol Renal Physiol 282: F669-F679, 2002.

Gekle M, Freudinger R, Mildenberger S, Silbernagl S. Rapid actions of aldosterone on cells from renal epithelium: the possible role of EGF-receptor signaling. Steroids 67: 499-504, 2002.

Ginzinger D. Gene quantification using real-time quantitative PCR: an emerging technology hits the mainstream. Exp Hematol 20: 503-512, 2002.

Helms M, Chen X, Ramosevac S, Eaton D, Jain L. Dopamine regulation of amiloride-sensitive sodium channels in lung cells. Am J Physiol Lung Cell Mol Physiol 290: L710-L722, 2006.

Hendron E, Stockand J. Activation of mitogen-activated protein kinase (mitogen-activated protein kinase/extracellular signal-regulated kinase) cascade by aldosterone. Mol Biol Cell 13: 3042-3054, 2002.

Hoffman C, Crettaz M, Bruns P, Hessel P, Hadawi G. Cellular responses elicited by insulin mimickers in cells lacking detectable plasma membrane insulin receptors. J Cell Biochem 27: 401-414, 1985.

Isenovic E, Muniyappa R, Milivojevic N, Rao Y, Sowers J. Role of PI3-kinase in isoproterenol and IGF-1 induced ecNOS activity. Biochem Biophys Res Commun 285: 954-958, 2001.

Kato H, Faria T, Stannard B, Roberts C, LeRoith D. Essential role of tyrosine residues 1131, 1135, and 1136 of the insulin-like growth factor-l (IGF-I) receptor in IGF-I action. Mol Endocr 8: 40-50, 1994.

Kim E, Shin H, Park J. Genistein inhibits insulin-like growth factor-I receptor signaling in HT-29 human colon cancer cells: a possible mechanism of the growth inhibitory effect of genistein. J Med Food 8: 431-438, 2005.

Kim J, Park BC, Kido Y, Accili D. Mitogenic and metabolic effects of type I IGF receptor overexpression in insulin receptor-deficient hepatocytes. Endocrinology 142: 3354-3360, 2001.

Krug A, Schuster C, Cassner B, Freudinger R, Mildenberger S, Troppmair J, Gekle M. Human epidermal growth factor receptor-1 expression renders Chinese hamster ovary cells sensitive to alternative aldosterone signaling. J Biol Chem 277: 45892-45897, 2002.

Kulik G, Weber M. Akt - dependent and -independent survival signaling pathways utilized by insulin-like growth factor I. Mol Cell Biol 18: 6711-6718, 1998.

Kwon M, Ling Y, Maile L, Badley-Clark J, Clemmons D. Recruitment of the tyrosine phosphatase Src homology 2 domain tyrosine phosphatase-2 to the p85 subunit of phosphatidylinositol-3 kinase (PI3-K) is required for insulin-like growth factor-I-dependent PI-3 kinase activation in smooth muscle cells. Endocrinology 147: 1458-1465, 2006.

Lebowitz K, An B, Edinger R, Zeidel M, Johnson J. Effect of altered Na + entry on expression of apical and basolateral transport proteins in A6 epithelia. Am J Physiol Renal Physiol 285: F524-F531, 2003.

LeMoellic C, Ouvrard-Pascaud A, Capurro C, Cluzeaud F, Pay M, Jaisser F, Farman N, Blot-Chabaud M. Early nongenomic events in aldosterone action in renal collecting duct cells: PKC activation, mineralocorticoid receptor phosphorylation, and cross-talk with the genomic response. J Am Soc Nephrol 15: 1145-1160, 2004.

LeRoith D. Insulin-like growth factor 1 receptor signaling-overlapping or redundant pathways? Endocrinology 141: 1287-1288, 2000.

LeRoith D, Werner H, Beitner-Johnson D, Roberts C. Molecular and cellular aspects of the insulin-like growth factor 1 receptor. Endocr Rev 16: 143-162, 1995.

Ling B, Eaton D. Effects of luminal Na + on single Na + channels in A6 cells, a regulatory role for protein kinase C. Am J Physiol Renal Fluid Electrolyte Physiol 256: F1094-F1103, 1989.

Manegold J, Falkenstein E, Wehling M, Christ M. Rapid aldosterone effects on tyrosine phosphorylation in vascular smooth muscle cells. Cell Mol Biol 45: 805-813, 1999.

Martelli A, Borgatti P, Bortul R, Manfredini M, Massari L, Capitani S, Neri L. Phosphatidylinositol 3-kinase translocates to the nucleus of osteoblast-like MC3T3-E1 cells in response to insulin-like growth factor I and platelet-derived growth factor but not to the proapoptotic cytokine tumor necrosis factor alpha. J Bone Miner Res 15: 1716-1730, 2000.

Marunaka Y, Eaton D. Effects of vasopressin and cAMP on single amiloride-blockable Na channels. Am J Physiol Cell Physiol 260: C1071-C1084, 1991.

May A, Puoti A, Gaeggeler H, Horisberger J, Rossier B. Early effect of aldosterone on the rate of synthesis of the epithelial sodium channel alpha subunit in A6 renal cells. J Am Soc Nephrol 8: 1813-1822, 1997.

Moura A, Worcel M. Direct action of aldosterone on transmembrane Na + efflux from arterial smooth muscle. Hypertension 6: 425-430, 1984.

Oberleithner H, Kersting U, Silbernagl S, Steigner W, Vogel U. Fusion of cultured dog kidney (MDCK) cells. II. Relationship between cell pH and K + conductance in response to aldosterone. J Membr Biol 111: 49-56, 1989.

Oberleithner H, Weigt M, Westphale H, Wang W. Aldosterone activates Na + /H + exchange and raises cytoplasmic pH in target cells of the amphibian kidney. Proc Natl Acad Sci USA 84: 1464-1468, 1987.

Record R, Froelich L, Vlahos C, Blazer-Yost B. Phosphatidylinositol 3-kinase activation is required for insulin-stimulated sodium transport in A6 cells. Am J Physiol Endocrinol Metab 274: E611-E617, 1998.

Record R, Johnson M, Lee S, Blazer-Yost B. Aldosterone and insulin stimulate amiloride-sensitive sodium transport in A6 cells by additive mechanisms. Am J Physiol Cell Physiol 271: C1079-C1084, 1996.

Rother K, Imai Y, Caruso M, Beguinit F, Formisano P, Accili D. Evidence that IRS-2 phosphorylation is required for insulin action in hepatocytes. J Biol Chem 273: 17491-17497, 1998.

Saito Y, Berk B. Transactivation: a novel signaling pathway from angiotensin II to tyrosine kinase receptors. J Mol Cell Cardiol 33: 3-7, 2001.

Sanchez-Margalet V, Zoratti R, Sung C. Insulin-like growth factor-1 stimulation of cells induces formation of complexes containing phosphatidylinositol-3-kinase, guanosine triphosphatase-activating protein (GAP), and p62 GAP-associated protein. Endocrinology 136: 316-321, 1995.

Schmidt T, Husted R, Stokes J. Steroid hormone stimulation of Na + transport in A6 cells is mediated via glucocorticoid receptors. Am J Physiol Cell Physiol 264: C875-C884, 1993.

Song G, Ouyang G, Bao S. The activation of Akt/PKB signaling pathway and cell survival. J Cell Mol Med 1: 59-71, 2005.

Stockand J, Spier B, Worrell R, Yue G, Al-Baldawi N, Eaton D. Regulation of Na + reabsorption by the aldosterone-induced small G protein K-Ras2A. J Biol Chem 274: 35449-35454, 1999.

Tack I, Elliot S, Potier M, Rivera A, Striker G, Striker L. Autocrine activation of the IGF-I signaling pathway in mesangial cells isolated from diabetic NOD mice. Diabetes 51: 182-188, 2002.

Tumlin J, Lea J, Swanson C, Smith C, Edge S, Someren J. Aldosterone and dexamethasone stimulate calcineurin activity through a transcription-independent mechanism involving steroid receptor-associated heat shock proteins. J Clin Invest 99: 1217-1223, 1997.

Ueki K, Fruman D, Brachmann S, Tseng Y, Cantley L, Kahn C. Molecular balance between the regulatory and catalytic subunits of phosphoinositide 3-kinase regulates cell signaling and survival. Mol Cell Biol 22: 965-977, 2002.

Urbach V, VanKerkhove E, Maguire D, Harvey B. Rapid activation of KATP channels by aldosterone in principal cells of frog skin. J Physiol 491: 111-120, 1996.

Watlington C, Perkins F, Munson P, Handler J. Aldosterone and corticosterone binding and effects on Na + transport in cultured kidney cells. Am J Physiol Renal Fluid Electrolyte Physiol 242: F610-F619, 1982.

Wehling M, Christ M, Theisen K. Membrane receptors for aldosterone: a novel pathway for mineralocorticoid action. Am J Physiol Endocrinol Metab 263: E974-E979, 1992.

Yeh T, Roth R. Insulin receptor signaling in Madin-Darby canine kidney cells overexpressing the human insulin receptor. Diabetes 13: 1297-1303, 1994.

Zhang S, Tsiaras W, Araki T, Wen G, Minichiello L, Klein R, Neel B. Receptor-specific regulation of phosphatidylinositol 3'-kinase activation by the protein tyrosine phosphatase Shp2. Mol Cell Biol 22: 4062-4072, 2002.

Zhou Z, Bubien J. Nongenomic regulation of ENaC by aldosterone. Am J Physiol Cell Physiol 281: C1118-C1130, 2001.

作者单位：1 Center for Cell and Molecular Signaling, Department of Physiology, Emory University School of Medicine, and 2 Graduate Program in Biochemistry, Cell, and Developmental Biology, Emory University Graduate School of Arts and Sciences, Atlanta, Georgia