PKB and megalin determine the survival or death of renal proximal tubule cells

Abstract

Renal proximal tubule cells have a remarkable ability to reabsorb large quantities of albumin through megalin-mediated endocytosis. This is an essential process for overall body homeostasis. Overstressing this endocytic system with a prolonged excess of albumin is injurious to proximal tubule cells. How these cells function and protect themselves from injury is unknown. Here, we show that megalin is the sensor that determines whether cells will be protected or injured by albumin. Megalin, through a novel mechanism, binds PKB in a D-3-phosphorylated phospholipid-insensitive manner, anchoring PKB in the luminal plasma membrane. Whereas low doses of albumin are protective, an overload of albumin decreases megalin expression followed by a reduction of plasma membrane PKB, PKB activity, and Bad phosphorylation induced by PKB. The result is albumin-induced apoptosis. These results reveal a model for PKB distribution in the plasma membrane and elucidate mechanisms involved in both the protective and toxic effects of albumin on proximal tubule cells. In addition, our findings suggest a mechanism for the progression of chronic kidney disease to end-stage renal disease.

Albumin, a major blood protein, is retained in the blood stream because it does not readily cross the glomerular filter into the kidney (1). The fraction that is filtered is reabsorbed by proximal tubule cells by clathrin- and receptor-mediated endocytosis, which efficiently removes albumin from the filtrate (1, 2). In many renal diseases, injury to the glomerulus breaks down the barrier function of the glomerulus, leading to excess filtration of albumin (2). Interestingly, long-term exposure of opossum kidney (OK) cells, a proximal tubule cell model, to high concentrations of albumin decreases albumin endocytosis (3), suggesting that high concentrations of albumin are actually toxic to the cells. Similarly, albuminuria is a well known marker for renal disease with a direct correlation between albuminuria and the progression of chronic kidney disease to end-stage renal disease (4). Over the past two decades, several reports have shown that albuminuria may cause tubular disease by overstressing the endocytic system with excess albumin (5–7).

Numerous studies support the importance of apoptosis in proximal tubule cells during tubular injury (8–10). However, a causative relationship between albumin and programmed cell death has not been described (11). In mouse proximal tubular epithelial cells, lower albumin concentrations act as a potent survival factor (12). Conversely, in the LLC-PK1 cell line, a model of the proximal tubular epithelium, high concentrations of albumin induce apoptosis (8). This apparent discrepancy may indicate that albumin could have a dual effect on the proximal tubule cells; lower concentration works as a survival factor, whereas higher concentration triggers toxic effects.

Neither the sensor nor the signal transduction cascade involved in albumin-induced toxic effects on proximal tubule cells has been identified or described. Albumin endocytosis occurs through the complex of at least two proteins, megalin and cubilin (1). An analysis of the megalin C-terminal cytoplasmic domains reveals the presence of motifs involved in protein–protein interaction, such as SH3 and PDZ domains, and protein kinase-phosphorylated sites (13, 14). These observations led some authors to propose that megalin-mediated signaling might be directly involved in the toxic effects of albumin overload (15).

One of the signal transduction pathways involved in cell survival and death is the phosphoinositide 3-kinase (PI-3K)/PKB pathway. Brunskill et al. (16) showed not only that albumin endocytosis in OK cells is modulated by PI-3K, but also that incubation with albumin increases PI-3K activity. PKB is a serine/threonine kinase and belongs to the AGC kinase family (17, 18). There are two important factors that regulate PKB activity: (i) translocation to plasma membrane and (ii) subsequent phosphorylation by plasma membrane-associated kinases (19, 20). Currently, it is widely accepted that localization of PKB at the plasma membrane depends on the level of D-3-phosphorylated phospholipids [PtdIns (3)P, PtdIns(3,4)P₂, and PtdIns(3,4,5)P₃], and that the inactive form of PKB is located in the cytosol (18, 19). PKB binding to phosphatidylinositol-3,4,5-triphosphate (PIP₃) is mediated by a pleckstrin homology domain located in the NH₂-terminal domain of PKB. Two phosphorylation sites in PKB are crucial for its activation: Ser-473 and Thr-308. Previously, we showed that Ang II, through activation of the resident pool of PKB in the plasma membrane, increases albumin endocytosis in LLC-PK1 in the presence of a high concentration of albumin (0.1–1.0 mg/ml; ref. 21). However, virtually nothing is known about the involvement of PKB, either in the protective or toxic effects of albumin on proximal tubule cells, or how this pathway modulates albumin endocytosis. Furthermore, the observation that PKB interacts with other proteins suggests that D-3-phosphorylated phospholipids are not solely responsible for the intracellular distribution of PKB (22).

We show here an interaction between PKB and megalin that is independent of D-3-phosphorylated phospholipids. We also demonstrate that megalin is the albumin sensor, mediating between cell protection and albumin-induced apoptosis. Low concentrations of albumin lead to activated PKB and phosphorylation of the Bad protein, known to inhibit apoptosis (23). On the other hand, overload of albumin leads to a decrease in megalin expression in the plasma membrane that is associated with PKB, PKB activity, and Bad phosphorylation by PKB. The result is an induction of albumin-induced apoptosis. These findings reveal a model of megalin signal transduction and plasma membrane PKB distribution. Our results provide a framework to understand how albumin can be both protective and harmful to proximal tubule cells.

Results

Interaction Between Megalin and PKB.

To access the location of PKB with respect to megalin at the plasma membrane, starved LLC-PK1 cells transiently expressing WT GFP-PKB, GFP-PKB (S473A), or GFP-PKB (T308A) were analyzed by using confocal microscopy. Zonula occludens 1 (ZO-1; blue) was used to mark the apical pole of the cells. Fig. 1 shows that WT GFP-PKB localizes at the apical membrane (Fig. 1a), along with megalin (b). The overlay of the GFP-PKB and megalin shows that PKB and megalin (yellow) colocalize at the apical membrane (Fig. 1c). The XZ crosssections of polarized LLC-PK1 cells confirm that PKB-GFP is present at the plasma membrane in association with megalin. The GFP-PKB distribution and its interaction with megalin did not change in cells transiently transfected with GFP-PKB (T308A) (Fig. 1 d–f) or GFP-PKB(S473A) (Fig. 1 g–i).

Fig. 1.

PKB interacts with megalin in the luminal membrane. LLC-PK1 cells were grown on Transwell cell culture inserts and transfected with either WT GFP-PKB (a–c) or its mutants GFP-PKB (T308A) (d–f) or GFP-PKB (S473A) (g–i). Blue (anti-ZO-1 antibody) represents tight junctions, green represents GFP, red represents megalin, and yellow (white arrows) represents the overlay of megalin and PKB imagines. ZO-1 staining is shown in all merged pictures, indicating that the pictures are captured at apical level. GFP-PKB and megalin display a punctuated pattern, appearing both near the plasma membrane and in the cytoplasm. [Scale bar, 10 μm (n = 3).]

To further confirm an association between PKB and megalin, endogenous PKB was immunoprecipitated, and megalin was detected by using antimegalin antibodies. The level of megalin was normalized to the total PKB coimmunoprecipitated. The band at 600 kDa that corresponds to the correct molecular mass for megalin was obtained (Fig. 6, which is published as supporting information on the PNAS web site).

It is well known that D-3-phosphorylated phospholipid-induced localization of PKB to the plasma membrane is followed by immediate phosphorylation of Ser-473 in the hydrophobic domain of PKB (17, 18). To explore the involvement of PKB phosphorylation on the megalin–PKB interaction, we used 0.5 μM wortmannin, a widely accepted selective inhibitor of PI-3K activity (24). Surprisingly, this inhibitor did not change the megalin–PKB interaction (Fig. 6). Importantly, wortmannin did inhibit basal PKB phosphorylation at both Thr-308 and Ser-473 residues (Fig. 6). This is consistent with published reports showing that this concentration of wortmannin completely inhibits all PI-3K isoforms (24). To study this further, LLC-PK1 cells were transiently transfected with WT GFP-PKB or the mutant, GFP-PKB (S473A; Fig. 6). PKB was immunoprecipitated by using a monoclonal GFP antibody and immunoblotted for megalin. The association between megalin and PKB was not changed by a single mutation at the PKB Ser-473 residue (S473A). Taken together, these results indicate that the association between PKB and megalin does not depend on PI-3K or the phosphorylation of Ser-473 and Thr-308 residues.

Physiological Concentrations of Albumin Modulate the Interaction Between PKB and Megalin.

Our first set of experiments demonstrated that, in the absence of albumin, localization of PKB at the plasma membrane depends upon an association with megalin in a D-3-phosphorylated phospholipid-insensitive manner. The next question is whether the association of megalin with PKB and PKB activity are modulated by albumin.

We observed that when cells are deprived of albumin and then rapidly exposed to lower physiological concentrations of albumin for 5 min (0.01–0.1 mg/ml), PKB disassociates from megalin without a change in overall amounts of either megalin or PKB (Fig. 2A). This is important, because it shows that the association between megalin and PKB is dynamically regulated by albumin and is not the result of spurious effects induced by experimental procedures.

Fig. 2.

Short-term effect of physiological albumin concentrations on PKB and megalin interactions and PKB activity. LLC-PK1 cells were grown on six-well plates, kept overnight in medium depleted of serum, and incubated with different concentrations of albumin for 5 min. Ctr, control; Wort, 0.5 μM wortmannin; and P-GSK-3, phosphorylated glycogen synthase kinase-3. Albumin concentrations are expressed in mg/ml. PKB was immunoprecipitated with mouse monoclonal PKB antibody where indicated. (A) Effect of albumin on PKB and megalin interactions (n = 4). (B) Effect of albumin on PKB phosphorylation (n = 6). (C) PKB activity measured by phosphorylation of GSK-3, as described in Materials and Methods. (Lower) Densitometry of phospho-GSK-3 (P-GSK-3) (n = 4). (D) Wortmannin and LY29004 reverse the stimulatory effect of albumin on PKB phosphorylation (n = 5). The megalin and phospho-residues bands were quantified and normalized by total PKB.

Next, we studied the effect of exposure to albumin for 5 min on PKB phosphorylation in the whole cell. Lower concentrations of albumin (0.01–0.1 mg/ml) increased phosphorylation of both Ser-473 and Thr-308 residues (Fig. 2B). The maximal stimulatory effect of 0.01 mg/ml occurs after 1 min of preincubation, but after 30 min, the increase noted in Fig. 2B returns to levels not significantly different from the control (data not shown). These data are consistent with the kinetics of PKB phosphorylation, which is well known (17, 18) to have rapid onset with transient duration. This effect is correlated with the activation of PKB activity measured by phosphorylation of glycogen synthase kinase (GSK); low concentrations increase PKB activity, whereas higher concentrations do not (Fig. 2C). The stimulatory effect of albumin 0.01 mg/ml on PKB phosphorylation is completely reversed by 0.5 μM wortmannin and LY29004 25 μM (Fig. 2D).

Finally, we determined whether PKB remains activated after it dissociates from megalin. To accomplish this, the cytosol was isolated, and PKB phosphorylation was measured (Fig. 7, which is published as supporting information on the PNAS web site). Albumin 0.01 mg/ml increases the phosphorylation of Ser-473 and Thr-308, and this effect is completely reversed by 0.5 μM wortmannin.

These data are consistent with a scenario whereby, in the absence of albumin, PKB associates with megalin at the plasma membrane independently of PI-3K. Physiological concentrations of albumin exposed for a short time to cells previously deprived of albumin enhance PI-3K activity (16) and lead, as shown here, to increases in PKB phosphorylation and activity and disassociation of activated PKB from megalin. Our previous work showed that the same physiological concentrations of albumin enhance endocytosis, and that this enhancement is also inhibited both by PI-3K and PKB inhibitors (21).

Pathophysiological Concentrations of Albumin Chronically Alter the Interaction Between PKB and Megalin.

Fig. 3A shows how pathophysiological concentrations of albumin modulate the interaction between PKB and megalin. Cells were preincubated with different concentrations of albumin for 24 h. In contrast to exposure of lower physiological concentrations of albumin for 5 min, the amount of PKB associated with megalin after 24-h exposure to 0.01–1.0 mg/ml albumin is similar to that in albumin-deprived cells. Interestingly, chronic exposure to high pathophysiological concentrations of albumin (10–20 mg/ml) decreases the association between PKB and megalin by 45%. This effect is concomitant with a decrease in megalin protein by 40% without any change in the total PKB protein (Fig. 3B). Again, this is in contrast to short-term exposure, which does not change the amount of either megalin or PKB.

Fig. 3.

Chronic effects of pathophysiological albumin concentrations on the PKB–megalin interaction. LLC-PK1 cells were grown on a six-well plate, kept overnight in medium depleted of serum in the absence or presence of albumin ranging from 0.01 to 20 mg/ml. Ctr, control. Albumin concentrations are expressed in mg/ml. (A) Effect of albumin on PKB–megalin interactions. PKB was immunoprecipitated with mouse monoclonal PKB antibody followed by immunoblotting for megalin and total PKB (n = 6). (B) Effect of albumin on PKB and megalin levels. Megalin and PKB bands were quantified and normalized by β-actin (n = 6). (C) Effect of albumin on megalin mRNA. Real-time PCR was used to quantify megalin. β-Actin was used as a control (n = 4).

To uncover the cause of the albumin-induced decrease in megalin, megalin mRNA was measured by using real-time PCR. The results show that albumin at 20 mg/ml decreased megalin mRNA by 40% (Fig. 3C), indicating that high concentrations of albumin decrease the transcription of the megalin gene. Furthermore, the possibility that the effect of albumin could be due to lysosomal degradation was ruled out, because 0.5 μM afilomycin A1 and 100 μM U64, inhibitors of lysosomal degradation (25, 26), did not change the albumin-induced decrease in megalin protein expression (data not shown).

To ascertain whether the decrease in the association between PKB and megalin could lead to a possible drop of PKB levels in the plasma membrane, we carried out experiments using confocal microscopy and isolated membrane fractions (Fig. 4A). The confocal images show that the expression of megalin at the luminal membrane decreased in cells incubated for 24 h with 20 mg/ml albumin. This effect is associated with decreased colocalization of PKB and megalin at the plasma membrane (Fig. 4A). Fig. 4B shows the presence of PKB in the membrane fraction in cells preincubated for 3 h with 0.5 μM wortmannin or for 24 h with 0.01 or 20 mg/ml albumin. Wortmannin and 0.01 mg/ml albumin did not change the presence of PKB in the membrane fraction, whereas 20 mg/ml albumin did decrease it by 40%.

Fig. 4.

Albumin modulates PKB in the plasma membrane. Albumin concentrations are expressed in mg/ml. (A) LL-CPK cells were grown on Transwell cell culture inserts and transiently transfected with GFP-PKB WT. Cells were kept overnight in medium depleted of serum without (a–c) or with (d–f) albumin (20 mg/ml). Blue (anti-ZO-1 antibody) represents tight junctions, green represents GFP, red represents megalin, and yellow (white arrows) represents the overlay of megalin and PKB images. [Scale bar, 10 μm (n = 3).] (B) LLC-PK1 cells were grown on a six-well plate, kept overnight in medium depleted of serum, and incubated for 3 h with 0.5 μM wortmannin (Wort) or for 24 h with 0.01 and 20 mg/ml albumin. The membrane fraction was isolated and immunoblotting for PKB and β-actin performed. Bands of PKB were quantified and normalized to β-actin (n = 4). (C) Chronic effect of albumin on PKB phosphorylation. PKB was immunoprecipitated with mouse monoclonal PKB antibody followed by immunoblotting with anti-phosphoThr-308, anti-phosphoSer-473, or total PKB antibodies. Phosphoresidues were quantified and normalized by total PKB (n = 6).

We next studied the effect of long-term exposure to albumin on PKB phosphorylation at both Ser-473 and Thr-308. Fig. 4C shows the effect of 24-h exposure to albumin. There is no detectable difference in PKB phosphorylation between the controls or cells exposed to low concentrations of albumin for long periods of time. This is consistent with low-dose exposures causing a transient increase reaching a maximum at 1 min and dissipating within 30 min. In contrast, the effect of long-term exposure to high pathophysiological albumin concentrations such as 10 and 20 mg/ml do indeed inhibit basal PKB phosphorylation on Ser-473 and Thr-308 (Fig. 4C).

These data suggest a pathophysiological scenario whereby chronic exposure to high concentrations of albumin decreases the expression of megalin and PKB in the plasma membrane and reduces PKB phosphorylation. High concentrations are also known to reduce both albumin binding and endocytosis (3).

PKB Inhibition Mediates Albumin-Induced Apoptosis.

It is well known that chronic exposure to high concentrations of albumin leads to tubular interstitial disease (1, 15), possibly by inducing apoptosis (9). PKB promotes cell survival through the phosphorylation of proteins involved in apoptosis such as Bad, which belongs in the Bcl-2 protein family (23). Phosphorylation of Bad by PKB at Ser-136 is important for the inhibition of the proapoptotic effect of Bad. This raises an important question regarding the impact of albumin-induced changes in PKB phosphorylation on cell survival shown in Fig. 4. Because we observed that high concentrations of albumin promote the inhibition of PKB activity (Fig. 2), we tested whether this effect could lead to a decrease in Bad phosphorylation at Ser-136 (Fig. 5A). We found that incubation with higher concentrations of albumin (10–20 mg/ml) for 24 h promoted dephosphorylation of Bad at Ser-136.

Fig. 5.

Albumin-induced apoptosis. LLC-PK1 cells were grown on six-well plates, kept for 24 h in medium depleted of serum in the absence or presence of albumin or in the presence of FBS, as indicated. Albumin concentrations are expressed in mg/ml. (A) Chronic effect of albumin on Bad phosphorylation. Phospho-Bad was detected by immunoblotting and normalized by total Bad (n = 4). (B and C) Cells were kept in medium with different concentrations of serum for 24 h in the absence or presence of albumin, as indicated in the figure legends. Apoptosis was measured by using annexin V-FITC as a marker of the early phase of apoptosis. Propidium iodide was used as a marker of cell viability. The stain was quantified by FACS (n = 4). (D) Cells were kept in medium without or with serum for 24 h. Protein extracted and immunoblotted with anti-phosphoThr-308, anti-phosphoSer-473, or total PKB antibodies was performed (n = 3).

Building on this observation, we next performed experiments measuring the effect of albumin on apoptosis in LLC-PK1 cells. The apoptotic cells were analyzed by FACS after staining with annexin V-FITC, a marker of early apoptosis (27, 28). The viability of the cells was measured by staining with propidium iodide, a marker of late apoptosis and necrosis (27, 28). The cells were washed with PBS and incubated with DMEM in the absence or presence of different concentrations of albumin (0.01–20 mg/ml) for 24, 48, and 72 h. We observed that higher concentrations of albumin (10–20 mg/ml) induced apoptosis after 24 h of incubation. In this condition, the number of annexin V-stained cells increased from 5.7 to either 11.5% or 14.7% in the presence of albumin at 10 and 20 mg/ml, respectively (Fig. 5B). LLC-PK1 cells maintained for 48 and 72 h showed a sharp decrease in viability even in the absence of albumin (data not shown). Furthermore, cell viability decreased from 87% to 37% and to 10% after 48 and 72 h of preincubation.

It is well known that FBS has several growth factors that increase PKB activity (19). In Fig. 5C, we show that incubation with different concentrations of FBS (2–10%) completely reversed albumin-induced apoptosis with a maximal effect observed at 10% FBS. In addition, we showed that 10% FBS induced PKB phosphorylation at Ser-473 and Thr-308 (Fig. 5D).

Discussion

Taken together, our data point to a model of how megalin, through its interaction with PKB, acts as a receptor/signal transduction mechanism that senses albumin in the lumen of the proximal tubule. Megalin can promote both the protection of normally functioning cells at low/physiological albumin concentrations and programmed cell death at high concentrations more typical of disease states.

PKB anchors to the plasma membrane through D-3-phosphorylated phospholipids formed by PI-3K and, immediately after binding, PKB is phosphorylated at the Ser-473 residue (17, 18). In this classic way, the phosphorylation of Ser-473 is indicative of D-3-phosphorylated phospholipid-mediated PKB binding to the plasma membrane. Watton and Downward (29) showed that, in polarized epithelial Madin–Darby canine kidney (MDCK) cells, a wortmannin-sensitive resident pool of PKB localizes to sites of cell–matrix and cell–cell contact even in the absence of activation by serum due to high basal PI-3K activity. Using confocal microscopy and coimmunoprecipitation experiments, we showed that LLC-PK1 cells kept overnight in medium depleted of serum have a resident pool of PKB in the plasma membrane because of a direct interaction between PKB and megalin. The interaction between PKB and megalin is not affected by wortmannin, demonstrating that it does not depend on PI-3K activation. One possible explanation for the differences observed between our results and those in MDCK cells could be that MDCK cells do not express megalin. Furthermore, the observation that the interaction of PKB and megalin is unaffected when cells are transiently transfected with GFP-PKB (S473A) or GFP-PKB (T308A) is further evidence that this association does not depend on the phosphorylation of Ser-473 and Thr-308 residues. These results represent an alternative mechanism for PKB association at the plasma membrane of proximal tubule cells in a D-3-phosphorylated phospholipid-independent manner through binding to megalin.

A longstanding open question is whether megalin could be the receptor for a signaling pathway and, if so, which one (1, 15)? Up to this point, no specific signaling mechanism has been linked to megalin. The observation that megalin has several motifs located in the cytosolic C-terminal domain that are able to interact with different proteins is suggestive of a mechanism that could trigger a cellular response (1, 15). For example, megalin interacts with Dab2, a protein involved in cellular growth and differentiation (30). But a link between the binding of Dab2 to megalin and a cellular response has not been established. Our work clearly shows that megalin binds to PKB, but what about signal transduction? Interestingly, when albumin is present for a short period at physiological concentrations, PKB phosphorylation and activity increase, whereas activated PKB dissociates from megalin. The initial interaction of megalin with PKB, which we noted in the absence of albumin, does not depend on PI-3K activity. In sharp contrast, the activation of PKB by physiological concentrations of albumin and the dissociation of PKB from megalin are both inhibited by wortmannin, strongly suggesting that both processes involve PI-3K. It is known that albumin increases PI-3K activity in OK cells (16), supporting our notion that stimulating PI-3K activity enhances PKB phosphorylation and induces the release of activated PKB from megalin. Does this pathway regulate endocytosis? Previous studies have shown that wortmannin abolishes albumin endocytosis in LLC-PK1 (21) and OK cells (16). Taken together, these results indicate that PKB is an element of the process leading to albumin endocytosis.

Albumin overload is usually associated with an abnormal increase in glomerular membrane permeability, leading to intensive megalin-mediated albumin reabsorption in proximal tubule cells (15). This albuminuria becomes a major risk factor for the progression of renal diseases (5, 15). A key element of our work is that the pathophysiological effects of albumin are triggered by a reduction in the expression of megalin. Megalin and cubulin form a complex that is critical for the reabsorption of low molecular-weight proteins from the glomerular filtrate (1). Given the importance of megalin in the process of protein reabsorption, it is probably the reduction in megalin expression that causes the reduced albumin binding and endocytosis that occur in kidney cells chronically exposed to high concentrations of albumin (3). It was shown that TGF-β1, which is involved in the progression of interstitial fibrosis, decreases megalin levels in OK cells (31). Furthermore, it was shown that albumin induces TGF-β1 production in human proximal tubular epithelial cells in a dose-dependent manner. These data indicate a possible connection between albumin-induced TGF-β1 release and lowered megalin expression, which may lead to tubule interstitial disease. One caveat concerning how these two processes are linked is that enhanced TGF-β1 production occurs at 0.1 mg/ml albumin, whereas we observed an albumin-induced decrease in megalin expression at 10 mg/ml. This makes it unlikely that the enhanced release of TGF-β1 is the cause of reduced megalin expression in our studies and leaves open the question of precisely how chronically saturated binding of albumin to megalin lowers megalin mRNA and protein expression.

Our results show an interesting switch in signaling that occurs between low and chronically high albumin concentrations. Low concentrations of albumin activate PKB, which promotes the phosphorylation of BAD, protecting against apoptosis. It has been shown that lower concentrations of albumin operate as a survival factor in primary cultures of mouse proximal tubule cells (12). The switch occurs at chronically high pathological concentrations of albumin, which decreases megalin mRNA and protein expression, promotes the decrease in PKB in the luminal membrane, and decreases PKB activity. The drop in PKB activity ultimately leads to a reduction in Bad phosphorylation and consequently induces apoptosis in proximal tubule cells. Erkan et al. (8) also observed that a high concentration of albumin (10–20 mg/ml) induced apoptosis in LLC-PK1 cells by assessing DNA cleavage and annexin V-FITC fluorescence. Thomas et al. (9), using an in vivo model, showed that an overload of albumin in proximal tubule induces both tubular apoptosis and cell proliferation, with apoptosis more marked than cell proliferation. Because our results show that megalin binds PKB at the plasma membrane, a pathological reduction in megalin expression subsequently causes a reduction in activated PKB. This ultimately leads to cell apoptosis and leaves one wondering whether inhibiting this process can be therapeutic. In a mouse model of glomerular injury, overexpression of AT₂ decreases urinary albumin excretion and protects against glomerular injury (32). We showed that albumin endocytosis in starved LLC-PK1 cells is increased by angiotensin II by the AT₂ receptor through activation of the plasma membrane resident pool of PKB (21), suggesting that therapeutically modulating PKB activity by either variation of megalin expression or direct modulation by a ligand binding to a receptor could be involved in progression of the renal disease destruction observed in some diseases.

Materials and Methods

Materials and Reagents.

Albumin, wortmannin, LY294002, genistein, and staurosporine were from Sigma (St. Louis, MO). Annexin V-FITC and propidium iodide were from BD PharMingen (San Diego, CA). Polyclonal phospho-PKB (Ser-473), monoclonal phospho-PKB (Thr-308), polyclonal Bad, polyclonal phospho-Bad (Ser-136) antibodies, and PKB kinase assay kits were from Cell Signaling Technology (Danvers, MA). Monoclonal GFP antibody was from Roche (Indianapolis, IN). The mouse monoclonal antibody for ZO-1 was from Zymed (San Francisco, CA). The rabbit anti-rat megalin antibody was kindly provided by Olivier Devuyst (Division of Nephrology, Université Catholique de Louvain, Louvain, Belgium) and GFP-tagged PKB constructs, by Jin Zhang (Department of Pharmacology, Johns Hopkins University School of Medicine).

Cell Culture and Transfection.

LLC-PK1 cells, a well characterized porcine proximal tubule cell line (American Type Culture Collection, Rockville, MD), were maintained in low-glucose DMEM with 10% FBS/1% penicillin/streptomycin (37°C and 5% CO₂). Cells were used 1 day postconfluence, typically 3 days after seeding. Before use, cells were kept overnight in medium depleted of serum and incubated with different compounds indicated in the figure legends. LipofectAMINE 2000 (Invitrogen, Carlsbad, CA) was used for cell transfection according to the manufacturer's instructions. Forty-eight hours after transfection, cells were used for immunostaining experiments.

Immunofluorescent Staining.

LL-CPK cells were grown in transwell inserts and transfected with GFP-tagged PKB WT and its mutants, GFP-PKB(T308A) or GFP-PKB(S473A). Twenty-four hours after transfection, they were placed in FBS-free medium in the conditions described in the figure legends. Forty-eight hours after transfection, the cells were used for immunostaining. After washing with PBS twice, the cells were fixed with 4% paraformaldehyde and blocked with 5% normal donkey serum in for 1 h. The cells were then incubated with the primary antibody for 1 h and washed with PBS three times, followed by the appropriate secondary antibody conjugated to Cy3 or Cy5 fluorescent dye (Jackson ImmunoResearch, West Grove, PA) for 1 h. After staining, the transwells were washed thoroughly in PBS, mounted with antiquenching medium (Vector Laboratories, Burlingame, CA), and then the slides were sealed.

Confocal Laser Microscopy.

The fluorescence label was examined with a confocal microscope (model LSM510, Zeiss, Oberkochen, Germany). Images were acquired by using the manufacturer's software. Contrast and brightness settings were chosen to ensure that all pixels were within the linear range. To obtain 3D images, each XY plane of the sample with a depth of 0.4 μm in the Z direction was scanned by confocal laser, and the picture serials along the z axis were combined, reconstructed, and presented as XZ and YZ cross-section images by using the Velocity software package (Improvision, Lexington, MA). Images were prepared for publication with Adobe Photoshop (Adobe Systems, San Jose, CA).

Immunoblotting and Immunoprecipitation.

The cells were washed with PBS⁺⁺, harvested, and incubated for 30 min in lysis buffer [20 mM Hepes. pH 7.4/2 mM EGTA/1% Triton X-100/400 μM PMSF/50 mM NaF/2 μM microcystin LR/2× complete protease inhibitor (Roche)/10 ng/μl leupeptin/10 ng/μl aprotinin/4 ng/μl elastatinal/2.5 mM 1,10 phenanthroline/100 μM l-1-tosylamido-2-phenylethyl chloromethyl ketone (tpck)] and cleared by centrifugation at 4°C for 10 min at 15,000 × g. The supernatant was retained, and the protein concentration was determined by using the Bio-Rad Protein Assay Kit (Rockford, Hercules, CA). When indicated, PKB or GFP were immunoprecipitated according to the manufacturer's instructions. Proteins were resolved on SDS/PAGE gels and transferred to PVDF or nitrocellulose membranes (Amersham Biosciences, Piscataway, NJ), according to the manufacturer's instructions. After antibody labeling, detection was performed with ECL (Amersham Biosciences). The images were acquired with FujiFilm (Edison, NJ) Image Reader (LAS-1000 Lite) and quantified with ImageGauge 4.0 (Fujifilm, Tokyo, Japan).

PKB Activity Assay.

PKB activity was measured by phosphorylation of GSK-3 according to the manufacturer's instructions (Cell Signaling Technology) and previously described (21). Protein were resolved on 10% SDS/PAGE gels and transferred to PVDF membranes. Western blot for phospho-GSK-3 was performed. The images were acquired with FujiFilm Image Reader and quantified with ImageGauge 4.0.

Cell Fractionation.

Cells were harvested and nuclei removed by centrifugation for 10 min at 1,000 × g at 4°C. The membrane and cytosol fractions were obtained by centrifugation at 100,000 × g for 1 h at 4°C. The pellet was resuspended in lysis buffer (composition as described above).

Apoptosis Measurements.

Annexin V-FITC in combination with PI was used to quantitatively determine the percentage of cells undergoing apoptosis, as described (27, 28). Briefly, cells were treated with different compounds, as indicated in figure legends, and the monolayer was released by a brief incubation with a Trypsin-EDTA solution. Cells (10⁵) were resuspended in 1× binding buffer (BD PharMingen) and incubated with annexin V-FITC for 15 min at room temperature, in the dark, followed by PI staining. Cells were analyzed within 1 h in a FACSCalibur flow cytometer, and Cellquest software (Becton Dickinson) was used to analyze the data. Early apoptotic cells were stained with annexin V alone, whereas late apoptotic or necrotic cells were stained with both annexin V and PI.

Megalin mRNA Quantitation by Real-Time PCR.

Total RNA was isolated by using TRIzol reagent (Invitrogen) by following the manufacturer's instructions. Total RNA (1 μg) was subjected to reverse transcription by using the iScript cDNA Synthesis Kit (Bio-Rad) and amplified by PCR. The primers pair used for megalin were: sense, 5′-ctgctcttgtagacctgggttc-3′; and antisense, 5′-tcggcacaggtacactcataac-3′ (GenBank accession no. CV869370). The primer pairs used for β-actin were: sense, 5′-ccagatcatgttcgagaccttc-3′; and antisense, 5′-tcttcatgaggtagtcggtcag-3′ (GenBank accession no. AY550069). The PCR product for megalin was separated by electrophoresis by using 1.5% agarose gel, and the single band of 138 bp was extracted from the gel by using a QIAquick Gel Extraction Kit (Qiagen, Valencia, CA) and sequenced to confirm identity. Megalin mRNA was quantified by real-time PCR by using SYBR Green Supermix (Applied Biosystems, Foster City, CA), according to the supplier's instructions.

Statistical Analysis.

Results are expressed as means ± SEM. Statistical significance was assessed by Student's unpaired t test. Significance was determined as P < 0.05.

Abbreviations

PI-3K

phosphoinositide 3-kinase

GSK

glycogen synthase kinase

ZO-1

zonula occludens 1

OK

opossum kidney.