The inhibition of mTOR kinase after renal transplantation has been associated with podocyte injury and proteinuria; however, the signaling pathways regulating these effects are not well understood. We found that prolonged rapamycin treatment in podocytes leads to an increase in glycogen synthase kinase 3β (GSK3β) phosphorylation, resulting in inactivation of total GSK3β kinase activity. To investigate the cellular consequences of the inactivation of GSK3β, we used two inhibitors reducing kinase activity and studied the cross talk between GSK3 function and the Akt/mammalian target of rapamycin (mTOR) pathway. Both GSK3 inhibitors reduced the phosphorylation of the mTOR downstream target, p70S6K, indicating that GSK3 inhibition in podocytes is able to cause similar effects as treatment with rapamycin. Moreover, GSK3 inhibition was accompanied by the reduced expression of slit diaphragm-associated proteins and resulted in an altered cytoskeletal structure and reduced motility of podocytes, suggesting that GSK3 kinase can modulate Akt/mTOR-dependent signaling in podocytes.
- slit diaphragm
after kidney transplantation, calcineurin inhibitor (CNI) toxicity is a major factor for the development of chronic allograft dysfunction. Conversion to rapamycin from CNI has been shown to improve renal function in transplant recipients. However, there is a great body of evidence that many patients develop proteinuria and some develop nephrotic syndrome when converted to sirolimus, indicating that rapamycin disturbs podocyte function (2, 5, 27, 38). Podocytes are highly specialized cells forming the epithelium in the glomerula of the kidney (32, 35). Their complex three-dimensional shape and unique cell-cell contact, the slit diaphragm, constitutes an essential part of the blood-urine filtration barrier. Defects in slit diaphragm proteins, such as nephrin or specific cytoplasmic adaptor proteins such as CD2-associated protein (CD2AP), Nck1/Nck2, or α-actinin 4, lead to breakdown of the filtration barrier and consequently to proteinuric glomerular disease (23, 24, 43). Previously, we and others showed that mammalian target of rapamycin (mTOR) inhibition by rapamycin resulted in reduced expression of mTOR pathway-related proteins and slit diaphragm components in cultured podocytes. This was accompanied by reduced podocyte motility and adhesion. These effects might contribute to podocyte damage by rapamycin, which can ultimately lead to proteinuria (28, 29, 46). Interestingly, rapamycin treatment also correlates with reduced phosphorylation of Akt Ser473 in this system, indicating that prolonged administration also alters the mTORC2 function, which is acting upstream of the mTORC1 (46).
In many cells, activation of the Akt/mTOR pathway correlates with increased phosphorylation and thereby inactivation of glycogen synthase kinase 3 (GSK3), suggesting that Akt/PKB acts upstream of GSK3 (18, 26). However, originally identified as a regulator of glycogen metabolism, ubiquitously expressed GSK3 protein has been established as a crucial regulatory enzyme contributing to many diverse cellular processes in eukaryotic cells (21, 22). GSK3 is essential for embryonic development, regulation of cellular structure, gene expression, polarity, and cell motility. In addition, GSK3 has been implicated in many diseases such as diabetes, cancer, Alzheimer's, and bipolar disorder (14, 39). Thus GSK3 has multiple cell- and tissue-specific biological functions and seems to be embedded in a complex network of various signaling pathways (21, 22).
GSK3 kinases consist of two isoforms, GSK3α and GSK3β, but little is known about their isoform-specific roles. GSK3 activity is significantly reduced by phosphorylation of an N-terminal serine, Ser9 in GSK3β and Ser21 in GSK3α (8, 11). These inhibitory phosphorylations of GSK3 are mediated by several kinases, among them PKA, PKB (identical to Akt), or PKC (21). Other mechanisms contributing to the regulation of GSK3 activity include the phosphorylation state of its substrates, its changes in subcellular localization, and regulation by binding proteins (21). Recently, Matsui and colleagues (30) found that the inhibition of GSK3 exacerbates proteinuria in rats that were treated with puromycin aminonucleoside (PAN). PAN administration causes transient proteinuria in rats and therefore serves as a well-established animal model for analyzing molecular and cellular aspects of glomerular diseases (36). The authors found that PAN treatment was accompanied by a suppression of nephrin mediated by the transcriptional regulator Snail (30).
Interestingly, Boini et al. (3) used a mouse model that carries a mutation in which the serine of GSK3β is replaced by alanine, leading to a constitutively activated GSK3. In this model, constitutive GSK3 activation induced the suppression of nephrin and podocin expression, correlating with proteinuria and glomerular injury (3). Both studies indicate a correlation between imbalanced GSK3β activity and reduced expression of nephrin. The intracellular part of nephrin associates with CD2AP and Nck and is involved in the regulation of both, local actin rearrangements in the secondary foot processes (1, 19, 20, 41, 42, 45) and the stimulation of the PI3K/Akt pathway (16, 49). However, so far, little is known about cross talk between the Akt/mTOR pathway and GSK3 signaling in podocytes. The aim of the present study was to investigate the functional role of GSK3 in the podocyte and to identify potential links between GSK3 and Akt/mTOR signaling.
Here, we show for the first time that prolonged rapamycin treatment of podocytes leads to increased phosphorylation and thereby inhibition of GSK3β. Furthermore, we were able to show that the inhibition of GSK3β causes similar effects on p70S6K phosphorylation, slit diaphragm protein expression, and podocyte motility as the inhibition of mTOR by rapamycin, suggesting that GSK3β kinase activity is able to modulate mTOR-dependent signaling in podocytes.
Cell culture and treatment.
Human immortalized podocytes were cultivated as described previously (40). Briefly, cells were grown in RPMI 1640 medium supplemented with 10% FCS and supplements at the permissive temperature of 33°C in 5% CO2 to cell confluence of 60–80%, then shifted to the nonpermissive temperature of 37°C (5% CO2) to allow terminal differentiation for 10–14 days. The mTOR inhibitor rapamycin was applied as before (10 or 100 ng/ml for 120 h, Sigma-Aldrich) (46). Rapamycin was dissolved in ethanol at a stock concentration of 1 mg/ml and further diluted in distilled water or medium. SB-216763 is a widely used GSK3 inhibitor which specifically inhibits GSK3. Podocytes were treated with SB-216763 in the concentration of 10 μM (Sigma-Aldrich) for 1, 6, 24, 36, or 48 h. SB-216763 was dissolved in DMSO at a concentration of 10 mM and further diluted in culture medium. In addition, we used LiCl in a final concentration of 20 mM. Respective solvent controls (DMSO or 20 mM NaCl) were performed for each treatment.
Cellular extract preparation and Western blotting.
Cellular lysates were prepared in lysis buffer [1% Triton X-100, 20 mM Tris·HCl (pH 7.5), 25 mM NaCl, 50 mM NaF, 15 mM Na4P2O7, and 1.5 mM EDTA] containing protease inhibitor (Complete, Roche). Cells were disrupted by shearing with a 0.4-mm needle. Lysates were centrifuged at 10,000 g for 30 min at 4°C. The supernatant was removed and stored at −20°C until further use.
For Western blot analysis, samples were diluted with Laemmli buffer and boiled. Proteins were separated by 8–15% SDS-PAGE (Bio-Rad). After transferring of proteins to a polyvinylidene difluoride membrane (Millipore), reactive sites were blocked at room temperature in blocking buffer (5% skim milk powder dissolved in TBS-T containing 0.1 M Tris, pH 7.4, 1 M NaCl, and 0.05% Tween 20). Primary antibodies were diluted in blocking buffer and incubated at 4°C overnight. Antibodies from rabbit against CD2AP, Nck1, GSK3β, phospho-GSK3β, p-β-catenin, mTOR, p-mTOR (Ser2448), p70S6K, p-p70S6K, Raptor, Rictor, Akt, p-Akt (Ser473, -Thr308, -Thr450), and p-PDK1 were diluted 1:1,000 (Cell Signaling Technology). Mouse antibody against β-tubulin (Sigma-Aldrich), rabbit antibodies against β-catenin (Santa Cruz Biotechnology), α-actinin-4 (Alexis Biochemicals), SGK, and p-SGK (Upstate) were diluted 1:1,000 [monoclonal antibody against WT1 (Dako), 1:500]. The polyclonal rabbit antibody against nephrin was a kind gift of Dr. Lawrence Holzman (University of Pennsylvania, Philadelphia, PA). After washing with TBS-T, membranes were incubated with the respective secondary horseradish peroxidase-coupled antibodies (Dianova). After continued washing in TBS-T, membranes were incubated with a chemiluminescence detection reagent (Roche).
Light microscopy of human podocytes.
Podocytes were shifted to a nonpermissive temperature to allow differentiation for 10–14 days. Cells were then treated with the GSK inhibitor SB-216763 (10 μM) or a respective solvent control (DMSO) for 24 or 48 h, and images were taken with an Axiovert microscope (×10).
Immunofluorescence analysis of glomerular podocytes.
Podocytes were cultured on collagen-coated coverslips for indirect immunofluorescence analysis. Coverslips were incubated with collagen A (Biochrom) at a final concentration of 100 μg/ml in PBS for 30 min at 37°C and then washed twice with PBS. Cells were fixed with 4% paraformaldehyde supplemented with 4% sucrose in PBS for 20 min. All steps were performed at room temperature. Samples were washed with PBS and incubated with 50 mM NH4Cl in PBS to quench reactive amino groups. After continued washing, coverslips were permeabilized with PBS containing 0.2% gelatin and TX-100 0.2% (PBS-TG). Then, samples were blocked with 10% goat serum diluted in PBS-TG for 20 min. Immunofluorescence staining was performed by incubating the coverslips for 1 h with respective primary antibodies diluted in PBS-TG containing 2% goat serum. Mouse antibodies against β-tubulin (Sigma-Aldrich) and WT1 (Dako) and rabbit polyclonal antibodies against α-actinin-4 (Alexis Biochemicals) and cortactin (clone 4F11, Upstate) were diluted 1:100, and a mouse antibody against vimentin (Dako; clone V9) was diluted 1:20. Afterward, coverslips were washed in PBS-TG and incubated with fluorochrome-conjugated secondary antibodies diluted 1:1,000 (Molecular Probes), sometimes containing Alexa Fluor 595 phalloidin (1:100, Molecular Probes). After washing with PBS, coverslips were rinsed in distilled water and cells were mounted in Mowiol containing DAPKO. Samples were examined with an Axio Observer Z1 microscope and ApoTome (objective: EC Plan Neofluar ×40/1.30*Oil DIC M27, Zeiss) using Axio Vision 4.7.
To evaluate the change in shape of the podocyte phenotype after treatment with the specific GSK3 inhibitor SB-216763 (10 μM), the length and width of podocyte images were measured using Adobe Photoshop 9.0 software. The ratio of longitudinal to diagonal cell diameter was calculated on the basis of >100 evaluated cells/group in three independent experiments.
Confluent differentiated AB 8 podocytes were grown on six-well cell culture dishes. Cells were pretreated with SB-216763 (10 μM) or respective solvent controls for 12 h and wounded with two strokes at a 90° angle with a sterile 0.4-mm 200-μl Gilson style extension length tip. Images of the same area were acquired after 0, 10, and 14 h using an Axiovert microscope (Zeiss). The number of cells which had migrated into a 2-mm2 wound area was quantified (n = 18 independent experiments) (9, 46).
Experiments were performed 3–24 times, and data are expressed as means ± SE. Controls and experiments were compared using an unpaired t-test. Statistical significance was deemed as P < 0.05, P < 0.01, and P < 0.001.
Prolonged mTOR inhibition results in increased phosphorylation of GSK3β in podocytes.
We and others demonstrated that rapamycin treatment, especially for >24 h, significantly reduces the expression of slit diaphragm proteins. Moreover, we observed effects on the cytoskeletal structure of podocytes that were accompanied by reduced cell adhesion and motility (28, 46). Surprisingly, we observed that prolonged rapamycin administration additionally results in increased phosphorylation of GSK3β while total GSK3β remained unchanged (Fig. 1A).
GSK3 functions in a well-known protein complex with axin and casein kinase I. If β-catenin is recruited to this complex, it is first phosphorylated by casein kinase I followed by GSK3 phosphorylation and thereby targeted for proteasomal degradation (21). Thus, in the case of inactivation of GSK3 enzymatic activity, the level of phosphorylated β-catenin and the p-β-catenin vs. total-β-catenin ratio decreases. This phenomenon can be utilized to assay GSK3 inhibition and GSK3 enzyme activity, respectively (7).
To prove whether the increase in GSK3β phosphorylation correlates with a decrease in GSK3 kinase activity, we analyzed the total β-catenin and p-β-catenin levels in the rapamycin-treated podocyte lysates by quantitative Western blotting and observed that the increased phospho-GSK3β level correlates with significantly decreased β-catenin phosphorylation (Fig. 1B), whereas both total expression levels as well as the expression level of the β-tubulin loading control (Fig. 1C) remained unchanged.
Treatment of podocytes with SB-216763 and LiCl results in fast inhibition of GSK3 activity.
Our results (summarized in Fig. 1) show that prolonged inhibition of mTOR reduces the kinase activity of GSK3, due to increased levels of the phosphorylated inactive form of GSK3 (or the p-GSK3/total GSK3 ratio).
Next, we used two established GSK3 inhibitors to analyze the consequences of reduced GSK3β activity in podocytes in more detail. The first inhibitor, SB-216763, is highly specific for GSK3 because it binds in the enzymatic pocket of this enzyme (6). As a second inhibitor, we used LiCl. LiCl has been described to inhibit GSK3 activity directly by competing with magnesium as well as indirectly due to its stimulating effect on GSK3-phosphorylating upstream kinases and/or its inhibiting effect on phosphatases which dephosphorylate GSK3 (33, 37, 48).
We administered these specific GSK3 inhibitors to our cells, first to keep the natural expression level of GSK in podocytes constant and second to avoid the nonphysiological overexpression of dominant-negative or -positive forms of GSK3 mutants or GSK3 knockout (e.g., RNAi) strategies.
The addition of SB-216763 and LiCl to podocytes led to a decreased level of β-catenin phosphorylation after 1, 6, and 24 h, demonstrating the fast inhibitory effect on GSK3 enzymatic activity by both inhibitors (Fig. 2). Thus both inhibitors are potent drugs for analyzing the consequences of reduced GSK3 activity in podocytes. The total expression levels of GSK3 and catenin remained unchanged, showing that only enzymatic activity but not the basal expression level is altered by these inhibitors (Fig. 2). Both inhibitors inhibit total GSK3 activity in podocytes. However, they have different effects on the intracellular presence of the phopshorylated form of GSK3β. LiCl administration results in an increased level of the phosphorylated form of GSK3. By contrast, SB-216763 treatment of podocytes results in a reduced level of the GSK3β phosphor-form.
GSK3 inhibition results in fast reduction of p70S6K phosphorylation.
The effects shown in Fig. 1 suggest cross talk between mTOR signaling and GSK3β function. Therefore, we analyzed this aspect in more detail and tested the effect of GSK3β inhibition on the phosphorylation of the well-established mTOR downstream target p70S6K. Surprisingly, we found that GSK3β inhibition by both SB-216763 as well as by LiCl resulted in dramatically reduced p70S6K phosphorylation (Fig. 3). The effects became significant after an incubation of 6 h and remained stable for 24 h (Fig. 3) and also during a longer incubation (see Fig. 5A). The total p70S6K level remained unchanged.
GSK3 inhibition results in a delayed dephosphorylation of Akt.
We analyzed the influence of GSK3 inhibition on the mTOR signaling pathway in more detail and observed decreased phosphorylation of Akt (Ser473 and Thr308) after 24 h. This effect is significant for LiCl as well as for SB-216763 treatment (Fig. 4). After 6 h, we observed a similar trend in all experiments, but only Akt Ser473 phosphorylation was significantly reduced in SB-216763 inhibitor-treated podocytes. The total expression of Akt, mTOR, and tubulin, which served as a loading control, remained constant (Fig. 4). Compared with the effect observed on p70S6K phosphorylation after 6 h (Fig. 3), the reduced phosphorylation of Akt (and mTOR in SB-216763-treated cells; see Supplementary Fig. S1) became significant much later.
Prolonged GSK3 inhibition results in reduced expression of nephrin, CD2AP, Nck1, and Wilms' tumor antigen 1.
Previously, we found that especially the prolonged incubation of podocytes with rapamycin not only influenced the phosphorylation of mTOR and Akt, it also reduced the expression levels of the slit diaphragm proteins (e.g., nephrin), diminished cell motility, and altered the cytoskeleton (46). To analyze the potential cross talk between the GSK3 and mTOR signaling pathway, we therefore also investigated the influence of prolonged GSK3 inhibition on podocytes (similar to the prolonged mTOR inhibition by rapamycin). Here, we focused on the SB-216763 inhibitor, because in contrast to LiCl it only targets GSK3 enzymatic activity directly.
Prolonged treatment of podocytes with this agent (>24 h) resulted in not only significant downregulation of the intracellular levels of p-β-catenin and p-p70S6K (Fig. 5A) but also reduced intracellular level of the slit diaphragm protein nephrin, the nephrin-interacting adaptors CD2AP and Nck1, and the zinc-finger transcription factor Wilms' tumor antigen 1 (WT1). WT1 is exclusively expressed in glomerular podocytes in the adult kidney and is known to activate the transcription of nephrin, which in turn is able to activate Akt/mTOR signaling (12, 13, 31) (Fig. 5B). The expression level of α-actinin-4 and β-tubulin (Fig. 5C) was not altered by GSK3 inhibition.
Interestingly, SB-216763 treatment not only influenced the expression of WT1 but also changed its cellular localization (Supplementary Fig. S2). While under control conditions a major portion of cellular WT1 was localized in the nucleus, it was mostly translocated into the cytosol upon treatment with SB-216763. In this context, it is interesting that Letavernier and colleagues (28) recently observed that rapamycin treatment of primary podocytes also causes WT1 translocation from the nucleus into the cytosol.
GSK3 inhibition leads to rearrangements of actin and β-tubulin fibers in a parallel fashion and reduces migratory abilities of podocytes.
The podocyte is uniquely located as part of the filtration barrier. It has to withstand local blood pressure. This requires a highly dynamic adaptation of the cell shape to local conditions for which rearrangements of actin fibers are essential. We investigated whether GSK3 inhibition has similar effects on the cytoskeleton and cell motility and cell shape as rapamycin treatment. Treatment of podocytes with SB-216763 led to a change from a polygonal cell shape to an elongated “fibroblast-like” phenotype. GSK3 inhibition significantly changed the proportion of the longitudinal-to-diagonal diameter of the podocyte in favor of the longitudinal diameter (Fig. 6). This was more pronounced after longer treatment. As this effect implies, that GSK inhibition in podocytes leads to cytoskeletal alterations, we stained the cytoskeleton proteins actin and β-tubulin and made costainings with the actin-associated proteins α-actinin-4, cortactin, and vimentin. When GSK3 was inhibited, actin and β-tubulin fibers were rearranged in a parallel fashion, indicating a disturbed cytoskeleton formation (Fig. 6). The staining of α-actinin-4, cortactin, and vimentin showed that their typical distribution along or at the end of actin stress fibers was not influenced by GSK3 inhibition.
As GSK3 inhibition altered the morphology and cytoskeleton of podocytes fundamentally, we analyzed whether this also resulted in a change in motility of the cells. Wound healing assays showed a significantly reduced ability of podocytes to migrate into the wound after GSK3 inhibition compared with control solvent-treated cells (see Fig. 7). The effect was highly significant after 10 h of treatment with SB-216763 (P < 0.001). After 14 h of GSK inhibition, the reduced migratory capability of podocytes was even more pronounced (P < 0.001).
The conversion from calcineurin inhibitors to rapamycin in transplant patients can be accompanied by proteinuria and occasionally by nephrotic syndrome. To identify molecular mechanisms and to elucidate the podocyte's participation in the onset of rapamycin-associated proteinuria, we investigated the influence of rapamycin on these specialized cells. We and others have recently shown that an inhibition of the mTOR kinase by rapamycin correlates with reduced expression of slit diaphragm proteins (e.g., nephrin, Nck, WT1) and redistribution of the actin cytoskeleton, resulting in dotlike structures (5, 28, 29, 46).
Here, we show that prolonged rapamycin treatment of podocytes leads to enhanced phosphorylation of GSK3β without changing the total expression level, strongly suggesting that prolonged rapamycin administration also reduces the total enzymatic activity of GSK3. To investigate this point in more detail, we took advantage of two GSK3-inhibiting agents, the highly specific inhibitor SB-216763 (6) and LiCl, which has been shown to be a potent inhibitor of GSK3β kinase (48). Similar to prolonged mTOR inhibition, GSK3β inhibition reduces phosphorylation of p70S6K and the expression of slit diaphragm proteins in podocytes. Moreover, we found reduced motility of treated cells that was accompanied by cytoskeletal rearrangements. An intact cytoskeletal architecture and the ability of a fast dynamic adaptation of cell shape to local conditions is essential for the three-dimensional structure of the podocyte, which has to its withstanding pressure changes at the filtration barrier (10, 35). When podocytes were treated with a GSK3 inhibitor, their cellular morphology and cytoskeletal architecture were severely altered (Fig. 6). This implies a disruption of physiological actin dynamics. Furthermore, GSK3 inhibition strongly decreased migratory abilities of podocytes (Fig. 7). Both effects suggest a diminished cell dynamic and might contribute to the development of proteinuria.
Together, these data suggest that there is cross talk between these two pathways in podocytes and that GSK3 inhibition might modulate rapamycin-dependent effects (see scheme in Fig. 8). In addition, we observed that both prolonged inhibition of mTOR (28, 46) as well as of GSK3 (this study) has effects on Akt, mTOR, and GSK3 phosphorylation, suggesting close homeostasis between these three kinases and their signaling pathways.
GSK inhibition does not only result in downregulation of the expression of the membrane components nephrin, CD2AP, and Nck1, but additionally seems to influence the phosphorylation of Akt and mTOR (Fig. 4, Supplementary Fig. S1). This is surprising as Akt has been described as an upstream activator of GSK3 in other cells (26). Moreover, usually, GSK3β inhibits mTORC1 activity by phosphorylation of TSC2/tuberin, which acts in concert with TSC1/harmartin (and the GTPase Rheb) as a regulator upstream of mTOR in many cells. Here, we show not only that in podocytes the effect on p70S6K becomes significant after 6 h but also that the decrease in mTOR and Akt phosphorylation predominantly occurs after 24 h. This suggests that GSK3β directly influences p70S6K activity via a so far unknown mechanism or that GSK3 affects mTOR kinase activity independently of the phosphorylation status of Akt and mTOR (Fig. 8).
It is also possible that additional regulating components of both signaling pathways act as modulating factors in these pathways, as in the case of the TSC1/TSC2 complex, for example. TSC1/2 has several phosphorylation sites playing a role in its activation and inactivation, suggesting that besides Akt, additional kinases regulate this complex (15, 34). A differentiated regulation of all these enzymes in diverse cells might explain some of the discrepancies observed in different cell lines and tissues concerning the effects of Akt, mTOR, and GSK3 (5, 17).
In the adult kidney, WT1 is expressed specifically in the podocyte and is known to be essential for filtration barrier function (31). Inhibition of both mTOR (28) as well as GSK3 (this study) led to a reduced expression level of WT1 and a shift from a nuclear to a more cytoplasmic intracellular localization. WT1 functions as a transcriptional activator of the nephrin gene (12) and directly regulates the expression level of nephrin, which in turn activates Akt and organizes the actin cytoskeletal rearrangements (47). Thus our data suggest a link between GSK3 activity and a WT1-dependent gene in podocytes. In addition, Srichai et al. (44) identified the shuttling factor WT1-interacting protein (WTIP), which seems to be part of a multiprotein complex in the foot processes of healthy podocytes and may link the CD2AP/nephrin complex to the actin cytoskeleton. They also found that WTIP translocates into the nucleus following podocyte damage, where it represses WT1-dependent gene expression, resulting in redistribution of slit diaphragm proteins and actin rearrangement characteristic of foot process effacement (25, 44).
Our observations are in line with the in vivo studies of Matsui et al. (30), who describe that GSK3 inhibition aggravated proteinuria in PAN-treated rats and that a GSK3 inhibitor enhances the PAN-dependent decrease in nephrin expression. Boini and colleagues (3) observed a reduced expression of slit diaphragm proteins accompanied by increased proteinuria in knockin mice expressing GSK3 variants (GSK3α S21A and GSK3β S9A, respectively). These variants cannot be inactivated by phosphorylation, so consequently must be permanently active. Also, these GSK3 knockin mice show multiple phenotypes that are not limited to kidney or glomerular podocytes (e.g., effects on blood pressure, food and water intake, glomerular filtration rate), indicating multiple GSK3 functions depending on tissue- and cell-specific signaling networks (3, 4).
However, in summary, all mentioned studies clearly argue that an imbalanced GSK3 kinase function associates with a decreased expression of slit diaphragm proteins, indicating tight signaling homeostasis among GSK3, Akt, and mTOR kinase activity in glomerular podocytes. Due to the fact that (prolonged) mTOR and/or GSK3β inhibition also caused various effects on “upstream acting” Akt, we hypothesize that Akt functions rather as an additional and/or parallel acting kinase in a complex signaling network than simply as an upstream regulator of GSK3 and mTOR in podocytes (see Fig. 8). This study supports the hypothesis that (see diagram in Fig. 8) GSK3- and Akt/mTOR-dependent signaling pathways could be linked. This suggests that rapamycin-induced or -associated proteinuria in kidney transplant patients (5) might thus be mediated or modulated by an accompanied imbalance of GSK3 activity.
This study was supported by Wyeth Pharma (Germany) and a grant from Innovative Medizinische Forschung Münster (I-WE 110719).
No conflicts of interest, financial or otherwise, are declared by the authors.
Present address of B. George: Univ. of Pennsylvania School of Medicine, Renal Electrolyte and Hypertension Division, Philadelphia, PA.
We thank Katja Brinkmann for excellent technical assistance, all laboratory members for helpful comments and discussions, and Dr. Lawrence Holzman (Univ. of Pennsylvania, School of Medicine, Philadelphia, PA) for the anti-nephrin antibody.
- Copyright © 2011 the American Physiological Society