Akt and Mammalian Target of Rapamycin Regulate Separate Systems of Proteolysis in Renal Tubular Cells
Wen Shen*,
Nikia S. Brown*,
Patrick F. Finn,
J. Fred Dice and
Harold A. Franch*
* Research Service, Atlanta Veterans Affairs Medical Center, Decatur, and Renal Division, Department of Medicine, Emory University School of Medicine, Atlanta, Georgia; and Department of Physiology, Tufts University School of Medicine, Boston, Massachusetts
Address correspondence to: Dr. Harold A. Franch, Renal Division, Emory University School of Medicine, W.M.B., Room #338, 1639 Pierce Drive, N.E., Atlanta, GA 30322. Phone: 404-727-3959; Fax: 404-727-3425; E-mail: hfranch{at}emory.edu
Received for publication November 4, 2005.
Accepted for publication June 14, 2006.
EGF suppresses proteolysis via class 1 phosphatidylinositol3-kinase (PI3K) in renal tubular cells. EGF also increases theabundance of glycolytic enzymes (e.g., glyceraldehyde-3-phosphatedehydrogenase [GAPDH]) and transcription factors (e.g., pax2)that are degraded by the lysosomal pathway of chaperone-mediatedautophagy. To determine if EGF regulates chaperone-mediatedautophagy through PI3K signaling, this study examined the effectof inhibiting PI3K and its downstream mediators Akt and themammalian target of rapamycin (mTOR). Inhibition of PI3K withLY294002 prevented EGF-induced increases in GAPDH and pax2 abundancein NRK-52E renal tubular cells. Similar results were seen withan adenovirus encoding a dominant negative Akt (DN Akt). Expressionof a constitutively active Akt increased GAPDH and pax2 abundance.An mTOR inhibitor, rapamycin, did not prevent EGF-induced increasesin these proteins. Neither DN Akt nor rapamycin alone had aneffect on total cell protein degradation, but both partiallyreversed EGF-induced suppression of proteolysis. DN Akt no longeraffected proteolysis after treatment with a lysosomal inhibitor,methylamine. In contrast, methylamine or the inhibitor of macroautophagy,3-methyladenine, did not prevent rapamycin from partially reversingthe effect of EGF on proteolysis. Notably, rapamycin did notincrease autophagasomes detected by monodansylcadaverine staining.Blocking the proteasomal pathway with either MG132 or lactacystinprevented rapamycin from partially reversing the effect of EGFon proteolysis. It is concluded that EGF regulates pax2 andGAPDH abundance and proteolysis through a PI3K/Akt-sensitivepathway that does not involve mTOR. Rapamycin has a novel effectof regulating proteasomal proteolysis in cells that are stimulatedwith EGF.
EGF and its associated receptor signaling represent a conservedpathway for a variety of epithelial responses, including cellgrowth. EGF signaling is a key component of renal tubular growthduring renal development, recovery from acute renal failure,renal hypertrophy, cyst development, and renal cell carcinoma(1). These signaling pathways regulate cell growth by modulatingthe synthesis or degradation of specific proteins. Whereas thepathways that regulate protein synthesis have been well studied,until recently, little progress was made in our understandingof the regulation of intracellular protein degradation (2,3).
In mammalian tissues examined, lysosomal macroautophagy and/orproteasomal proteolysis often are regulated during cell growth(2,3). Autophagy refers to processes of importing intracellularproteins across the lysosomal membrane for degradation (4).In macroautophagy, a double-membrane vesicle forms around targetedproteins or organelles followed by fusion of that vesicle withlysosomes. Transformed cells have decreased macroautophagy,whereas increased macroautophagy has been linked to apoptosis(5). In the ATP-ubiquitin-proteasome system, a heat-shock protein,ubiquitin, is covalently coupled to lysines in target proteinsby an ATP-dependent process (6). Ubiquitin-conjugated proteinsare recognized by the 26S proteasome cap and degraded in theproteasome. This system accounts for the majority of proteindegradation in many cell types (6). There is increasing evidencethat insulin and IGF-1 suppress the ubiquitin/proteasome systemin skeletal muscle growth (3).
We have shown that a novel system of proteolysis, chaperone-mediatedautophagy (CMA), is suppressed in kidney cells that grow inculture and in diabetic renal hypertrophy (7,8). In CMA, a chaperoneprotein, hsc73, binds to a five–amino acid sequence (similarto Lys-Phe-Glu-Arg-Gln [KFERQ]) in proteins that are destinedfor degradation (9,10). Proteins that are bound to hsc73 subsequentlybind to the lysosome-associated membrane protein 2a on the lysosomalsurface (11); lysosome-associated membrane protein 2a acts asthe receptor, allowing import of the target protein into thelysosome (12). Approximately 30% of all kidney proteins containa KFERQ motif (i.e., are potential substrates of CMA), includingglycolytic enzymes, the enzymes of the Kennedy pathway of phospholipidssynthesis, and transcription factors such as pax2 and c-fos(10). Growth factors suppress CMA in kidney cells, but starvationor oxidative stress activate CMA, thereby reducing the abundanceof KFERQ-containing proteins. Because protein or calorie starvationinhibits renal hypertrophy (13), stimulation of CMA could providean explanation for the effects of diet on renal growth.
Class 1 phosphatidylinositol 3-kinase (PI3K) signaling has beenimplicated as regulating proteolysis in a number of tissues.This enzyme consists of an 85-kD regulatory subunit (p85) anda 110-kD effector subunit (p110) (14). The p110 subunit placesphosphates on the 3 hydroxyl group of inositol (PtdIns). PtdIns3,4 or PtdIns 3,4,5 mediate downstream signaling by activating3-phosphoinositide-dependent protein kinase 1 which phosphorylatesAkt (15). Blocking class 1 PI3K or Akt activity inhibits cellgrowth and protein accumulation and makes cells susceptibleto apoptosis (15). 3-Phosphoinositide-dependent protein kinase1 and Akt mediate these responses by phosphorylating other downstreameffectors, such as glycogen synthetase kinase (GSK), which regulatesglucose metabolism (16), and p70S6 kinase, which stimulatesprotein synthesis (14). The mammalian target of rapamycin (mTOR)modulates PI3K signaling in two ways. When bound to rictor,it phosphorylates Akt in a rapamycin-independent manner, butwhen complexed with raptor, it phosphorylates p70S6 kinase,an activity that is inhibited by rapamycin (17). This rapamycin-sensitivepathway has been lined to regulation of macroautophagy by nutrients:in isolated hepatocytes and hepatocellular carcinoma cell lines,rapamycin prevents PI3K/Akt signaling from suppressing macroautophagy(18,19). This finding seems to be tissue specific, because macroautophagycan be regulated by rapamycin-independent pathways in fibroblastsand skeletal muscle (20).
We have shown that signaling through PI3K is required for EGFto suppress proteolysis (21). We also determined that EGF suppressedlysosomal proteolysis, including CMA (7), but we did not establishwhether CMA or another system of proteolysis was regulated byPI3K. We now investigate the EGF-induced signaling intermediatesthat regulate proteolysis and the substrates of CMA in NRK-52Erenal tubular epithelial cells. Our results indicate that PI3Kand Akt but not the rapamycin-sensitive activity of mTOR arerequired for EGF to increase pax2 and glyceraldehyde-3-phosphatedehydrogenase (GAPDH) and suppress lysosomal proteolysis. Surprising,rapamycin stimulates a proteasomal system of proteolysis withoutaffecting lysosomal proteolysis in EGF-treated cells.
Rapamycin and other chemicals were from Sigma Chemical (St.Louis, MO) unless specified. Recombinant human EGF was fromR&D Systems (Minneapolis, MN); methylamine, 3-methyladenine,MG132, and lactacystin were from Calbiochem (La Jolla, CA).L-[U-14C] phenylalanine and chemiluminescence (ECL) reagentswere from Amersham Bioscience (Piscataway, NJ). DMEM, newborncalf serum, and penicillin-streptomycin were from Life Technologies(Grand Island, NY). Anti–phospho-Akt (Ser473) and anti–phospho-p70S6kinase (Thr389) antibodies were from Cell Signaling (Beverly,MA). Pax2 and GAPDH antibodies were from Santa Cruz Biotechnology(Santa Cruz, CA). Akt1 dominant negative (DN Akt) and constitutivelyactive (CA Akt) adenoviruses were gifts of Dr. Kenneth Walsh(Boston, MA) (22). Adenoviruses that expressed green fluorescenceprotein (GFP) alone (ad.EGFP) were gifts from Dr. Bert Vogelstein(Baltimore, MD). We have previously published construction ofadenoviruses that express a dominant negative p85 (DN p85) subunitof class 1 PI3K (21).
Cell Culture and Adenoviral Infection
NRK-52E cells (a rat kidney epithelial cell line) that wereobtained from the American Type Culture Collection (Bethesda,MD) at passage 15 were grown in DMEM supplemented with 5% newborncalf serum (7). Studies were performed on cells from passages19 to 29. Toxicity in the protein degradation assay was accessedby cell counts recovered and in other experiments by trypanblue exclusion (21). Concentrations of inhibitors or adenovirusesthat were used in experiments were usually <25% of the lowestconcentration that produces detectable toxicity. For most experiments,cells were grown to confluence and rendered quiescent by serumremoval 48 h before experimental treatment, but 5% calf serumwas required for the Akt1 CA adenovirus experiments.
Viruses were propagated in HEK293 cells and purified by gradientdensity centrifugation in CsCl; final yields generally were1010 to 1011 plaque-forming units. Ad.EGFP (a virus that containsGFP alone) was used as a transfection control in all experimentsthat involved adenoviruses (23). For adenoviral infection, cellswere grown to 70 to 80% confluence and incubated with adenovirusesin 10% FBS DMEM for 48 h. The multiplicity of infection (MOI)is indicated in the figures. Efficiency of transfection wasevaluated by fluorescence microscopy for the Ad.EGFP and DNp85 viruses. An MOI of 10 pfu/cell of the Ad.EGFP typicallyproduced 50 to 60% transfection efficiency with no evidenceof toxicity.
Protein Degradation
Protein degradation was measured after cellular proteins werelabeled for 3 d with 0.5 µCi of L-[14C]phenylalanine perwell (specific activity 1.25 mCi/M phenylalanine) (24). Afterlabeling, a 2-h chase was performed to remove L-[14C]phenylalaninethat was released from short-lived proteins. The chase mediumwas replaced with 3 ml of experimental medium supplemented withgrowth factors and enzyme inhibitors as indicated. Unlabeledphenylalanine (5 mM) was added to the medium to minimize reuseof the phenylalanine that was released by protein breakdown.Four aliquots of the medium were removed at various times upto 48 h and [14C]phenylalanine that was released from labeledproteins was measured after precipitation of proteins with trichloroaceticacid (10% vol/vol). At the end of the sampling period, cellmonolayers were solubilized in 1% SDS (1 ml/well) to determinethe radioactivity that remained in the cells, and the rate ofprotein degradation was determined by calculating the slopeof the logarithm of [14C]phenylalanine that remained in cellprotein at time 0 and at four other time points. The mean ofthe slope of the degradation curves of the six samples of asingle experiment was used in the figures, but all results wereconfirmed by three repeated experiments on separate days. Onlystudies with first-degree kinetics were used, and this resultedin the exclusion of a very few studies, because this relationshipremained linear with all treatments (Figures 6A and 7A, references[24,25]; data not shown). Total radioactivity that was recoveredfrom cells (calculated from the amount released into the mediaplus what remained in the cell monolayer) was an indicator ofcell viability and did not change with any experimental treatment.
Figure 6. EGF but not rapamycin alters lysosomal proteolysis. Cells that were grown and treated as in Figure 1 were labeled with C14 phenylalanine, and the release of radioactivity was measured as described in the Materials and Methods section. The rate of protein degradation is measured as the slope of the log of the percentage of the counts that remained in cells at different times (A). To facilitate viewing of multiple comparisons, we show most data as the mean value of this slope (B). The figures are the representative experiments of three repeats (n = 6/group), *P < 0.05 versus control; +P < 0.05 versus EGF alone. (A) Cells were treated with EGF (E; 10 nM) alone, rapamycin (R; 50 nM) alone, or both (ER), or vehicle (C). (B) Cells were treated with lysosomal inhibitor methylamine (10 mM) in the presence of EGF, rapamycin, or both, or vehicle.
Figure 7. Akt regulates lysosomal proteolysis. NRK-52E cells were infected with GFP, DN Akt at MOI of approximately 10 pfu/cell, or CA Akt with increasing MOI (approximately 1 pfu/cell, approximately 6 pfu/cell, approximately 25 pfu/cell, and approximately 50 pfu/cell) for 48 h. (A) GFP- and DN Akt–infected cells were treated with EGF (E; 10 nM) or vehicle (C). The result of protein degradation assay is expressed as indicated in Figure 5A. The slope of the EGF line is significantly different from all other lines (P < 0.05). (B) GFP-infected cells were treated with EGF (E; 10 nM) or vehicle (C) as the positive and negative control. The result of protein degradation assay is expressed as indicated in Figure 5B (n = 6 per group). (C) GFP- and DN Akt–infected cells were treated with EGF alone (E; 10 nM), methylamine alone (M; 10 mM), or both (EM), or vehicle (C). The result of protein degradation assay is expressed as indicated in Figure 5B. *P < 0.05 versus control; +P < 0.05 versus EGF alone.
Immunoblotting
Cells were lysed in RIPA buffer (20 mM Tris-HCl [pH 7.4], 0.1%SDS, 2.5 mM EDTA, 50 mM NaF, 1% Triton, 10% glycerol, 1% deoxycholate,10 mM Na4P2O7, 1 mM Na3VO4, 1 mM PMSF, and 10 µg/ml aprotinin).Proteins were separated by SDS-PAGE and transferred to nitrocellulosemembranes. After immunoblotting with specific antibodies, blotswere developed using ECL. Developed images were scanned andquantified using BioRad Quantity One software.
Monodansylcadaverine Staining
NRK-52E cells were treated with EGF, rapamycin, or EGF plusrapamycin for 16 h with or without 10 mM 3-methyladenine. Cellsthen were washed twice with Hanks balanced salt buffer and stainedwith monodansylcadaverine (MDC) as previously reported (26).Briefly, cells were incubated with 0.05 mM MDC for 10 min, thenwashed four times with PBS. The fluorescence was visualizedusing a Zeiss Axiovert fluorescence microscope (Thornwood, NY),and were images analyzed using Adobe Photoshop v7.0. Using areference grid, the number of MDC-positive vesicles per areawas calculated for each treatment group.
Statistical Analyses
Data are expressed as mean ± SE. Differences betweenmultiple groups were evaluated by ANOVA using the Student-Newman-Keulstest for multiple comparisons. P < 0.05 was considered statisticallysignificant.
PI3K Regulates pax2 and GAPDH Abundance
CMA degrades specific proteins that are important for renaltubular cell growth, including the glycolytic enzyme GAPDH andthe paired-box transcription factor, pax2 (7). Because EGF suppressesdegradation of pax2 and GAPDH by CMA in renal epithelial cellswithout significantly increasing their mRNA (27,28), we usedthe abundance of these proteins as markers of CMA. As expected,EGF significantly increased both pax2 and GAPDH abundance (P< 0.001) but not actin, a protein that lacks a KFERQ sequence(Figure 1). The mTOR inhibitor rapamycin (50 nM) did not significantlyaffect basal levels of GAPDH or pax2 or affect the EGF-stimulatedincreases in these KFERQ-containing proteins. This concentrationof rapamycin was sufficient to prevent the phosphorylation ofp70 S6 kinase (Figure 2B) and reduce EGF-induced protein synthesisby approximately 25% (data not shown). In contrast, the PI3Kinhibitor LY294002 (25 µM) significantly reversed EGF-inducedaccumulation of both pax2 and GAPDH. To confirm involvementof class 1 PI3K, we used adenovirus that express a mutant DNp85 with the inner SH2 domain deleted (21,29). Expression ofthe DN p85 suppressed EGF-induced Akt phosphorylation by 37%and blunted the EGF-induced rise in GAPDH and pax2 (Figure 3).These results suggest that EGF-induced upregulation of pax2and GAPDH requires class 1 PI3K but not mTOR.
Figure 1. LY294002 but not rapamycin regulates pax2 and glyceraldehyde-3-phosphate dehydrogenase (GAPDH) abundance. NRK-52E cells were treated with vehicle (C), EGF alone (E; 10 nM) with or without rapamycin (50 nM), or with or without LY294002 (25 µM). Cells were grown and made quiescent as described in Materials and Methods. All reagents were added freshly in serum-free media every 24 h. Western analysis was carried out by using the same amount of protein extracts (10 µg). Representative blots are shown along with densitometry results normalized to actin for all samples. Lines indicate statistical comparisons between groups. (A) Western analysis was performed by using an antibody against pax2. Cells were treated for 96 h (n = 4). (B) Western analysis was performed by using antibody against GAPDH. Cells were treated for 48 h (n = 4). (C) Western analysis with antibody against actin. Cells were treated for 48 h, but similar results were seen at 96 h (n = 3). No significant difference was found between groups (densitometry data not shown).
Figure 2. The effect of proteolytic inhibitors on Akt/p70S6 kinase phosphorylation. NRK-52E cells were treated with 10 mM methylamine, 10 mM 3-methyladenine, or 0.5 µM MG132 for 1 h, followed by 50 nM rapamycin (R) treatment for 30 min and/or 10 nM EGF treatment (E) for 15 min. Western analysis was carried out by using the same amount of protein extracts (15 µg). A representative blot of three total is shown for each treatment. (A) Western analysis with an antibody against phospho-Akt (Ser473; n = 3). (B) Western analysis with an antibody against phospho-p70S6 kinase (Thr389; n = 3). Significant decreases in phosphorylation were only seen with 3-methyladenine (P < 0.05).
Figure 3. Class 1 phosphatidylinositol 3-kinase (PI3K) p85 subunit dominant negative adenovirus regulates Pax2 and GAPDH abundance in NRK-52E cells. NRK-52E cells were infected with either green fluorescence protein (GFP) or p85 dominant negative adenoviruses (p85 DN) at a series of dosages for 48 h. The multiplicities of infection (MOI) of p85 DN were approximately 10 plaque-forming units (pfu)/cell, approximately 30 pfu/cell, and approximately 90 pfu/cell. Western analysis was carried out by using the same amount of protein extracts (10 µg). The statistical analysis results are shown. (A) Cells were treated with EGF (E; 10 nM) or vehicle (C) for 15 min. Immunoblot analysis was performed by using antibody against phospho-Akt (Ser473; n = 3). (B) Cells were treated as indicated for 96 h. Immunoblot analysis was performed by using antibody against Pax2 (n = 4). (C) Cells were treated as indicated for 48 h. Immunoblot analysis was performed by using antibody against GAPDH (n = 4). (D) Immunoblotting with antibody against actin (n = 4).
Akt Regulates Pax2 and GAPDH Abundance
In NRK-52E cells, type 1 PI3K activity is required for phosphorylationof the downstream mediator Akt (21). Next, we used a kinase-dead,Akt1 dominant negative protein (DN Akt; Akt1 with T308A, S473Amutations; a gift from Dr. K. Walsh [22]) that was deliveredby an adenoviral vector as a tool to examine whether Akt regulatespax2 and GAPDH abundance. Expression of DN Akt suppressed EGF-inducedphosphorylation of the Akt downstream effector GSK3 by approximately42% (Figure 4A). It also suppressed EGF-induced accumulationof pax2 and GAPDH in a dose-dependent manner without changingthe level of actin (Figure 4, B and C). These results suggestthat Akt is the downstream effector of PI3K that regulates CMA.
Figure 4. DN Akt adenovirus regulates pax2 and GAPDH abundance. NRK-52E cells were infected and treated as indicated. The MOI of DN Akt were approximately 5 pfu/cell, approximately 25 pfu/cell, and approximately 75 pfu/cell. Western analysis was performed by using the same amount of protein extracts (10 µg). Lines indicate statistical comparisons between groups. (A) NRK-52E cells were infected with either GFP or DN Akt for 48 h and then treated with EGF (E; 10 nM) or vehicle (C) for 15 min. Western analysis was performed by using antibody against phospho-GSK3 (Ser9; n = 3). (B) Cells were treated for 96 h. Western analysis was performed by using antibody against pax2 (n = 4). (C) Cells were treated for 48 h. Western analysis was performed by using antibody against GAPDH (n = 4). (D) Western analysis was performed on the same samples as in C with antibody against actin (n = 3). There were no statistically significant differences between treatments (densitometry data not shown).
To confirm the role of Akt in regulating the abundance of KFERQproteins, we expressed a constitutively active form of Akt (CAAkt: myristoylated Akt1 [AxCAMyr-Akt], a gift from Dr. K. Walsh[22]) using an adenoviral vector. This virus, unlike the DNAkt, caused detectable toxicity at an MOI of >10 in serum-starvedcells, so experiments shown were performed in the presence of5% calf serum. Similar but more variable results were obtainedin serum-starved cells with an MOI of 5 or 10 (data not shown).With CA Akt, GSK3 phosphorylation increased with increasingMOI of virus (Figure 5A). CA Akt increased pax2 and GAPDH proteinabundance without affecting actin abundance (Figure 5, B throughD). These results indicate that Akt activation is sufficientto regulate pax2 and GAPDH abundance. In summary, our resultsstrongly suggest that PI3K acting through Akt regulates thelysosomal pathway of CMA, whereas mTOR does not.
Figure 5. Constitutively active Akt (CA Akt) adenovirus regulates pax2 and GAPDH abundance. NRK-52E cells were infected with either GFP or CA Akt for 48 h with increasing MOI (approximately 1 pfu/cell, approximately 6 pfu/cell, approximately 25 pfu/cell, and approximately 50 pfu/cell). Lines indicate statistical comparisons between groups. (A) Western analysis was performed by using antibody against phospho-GSK3 (Ser9). GFP-infected cells were treated with EGF (E; 10 nM) or vehicle (C) as the positive and negative control. (B) Cells were treated as indicated in A for 96 h. Western analysis was performed by using antibody against pax2 (n = 3). (C) Cells were treated for 48 h as in B. Western analysis was performed by using antibody against GAPDH (n = 3). (D) Western analysis was performed on the same samples as in C with antibody against actin (n = 3). There were no statistically significant differences between treatments (densitometry data not shown).
Effect of Proteolytic Inhibitors on PI3K Downstream Signaling
To study the effect of these signaling molecules on systemsof proteolysis, we used a lysosomal inhibitor that alkalinizesacidic compartments (10 mM methylamine), a proteasome inhibitorthat blocks its trypsin-like site (0.5 µM MG132), anda class III PI3K inhibitor that blocks vesicle fusion in macroautophagy(10 mM 3-methyladenine). We previously showed that these concentrationsof methylamine and MG132 provide maximal proteolytic inhibitionand are specific for their proteolytic system in NRK-52E cells(24). Moreover, this concentration of 3-methyladenine is thecurrently accepted standard for inhibiting macroautophagy inepithelial cells (30). In initial experiments, we examined theeffects of proteolytic inhibitors on phosphorylation of severalkey signaling molecules that are involved in PI3K signaling.Neither methylamine nor MG132 altered the phosphorylation ofAkt (Ser473) and p70S6 kinase (Thr389; Figure 2). As expected,3-methyladenine partially inhibited cell signaling through Aktand p70S6 kinase. This last result is consistent with the observationof Proud and colleagues (31), who reported that this class IIIPI3K inhibitor also weakly inhibits class I PI3K in fibroblasts.
EGF and Akt Suppress a Lysosomal System, whereas Rapamycin Affects a Different System
In prelabeled NRK-52E cells, the release of C14-labeled phenylalanineinto the medium is an established method to measure the overalldegradation rate of long-lived proteins (7,21,24,32). Phenylalanineis an appropriate amino acid to label because it is not significantlysynthesized or degraded in renal epithelial cells (33). To determinewhether EGF, proteolytic inhibitors, or cell signaling manipulationsalter the kinetics of protein degradation, we plotted the meanvalues of the counts that remained at each time point (an exampleis shown in Figure 6A). All treatments resulted in first-orderkinetics (Figures 6A and 7A) (21,24). Because statistical comparisonsare difficult to visualize when plotting the degradation curves,we expressed most data as the mean slope of the plot of countsthat remained versus time (Figure 6B).
Treatment with EGF decreased protein degradation by 24.8 ±3.9% (P < 0.05; n = 6), and rapamycin significantly attenuatedthis effect (P < 0.05; n = 6). It is interesting that rapamycinalone had no significant effect on proteolysis (Figure 6A).Consistent with our previous study (24), EGF did not furthersuppress protein degradation in the presence of the lysosomalinhibitor 10 mM methylamine (Figure 6B). These results indicatethat EGF-induced suppression of proteolysis in renal tubularcells depends on a lysosomal system. When rapamycin is present,EGF no longer suppresses proteolysis to the same extent. However,this effect of rapamycin still is seen when methylamine is present.We interpret these data to mean that although rapamycin increasesproteolysis when EGF is present, it does so by activating anonlysosomal proteolytic pathway. These results are consistentwith the Western blotting data showing no effect on KFERQ-containingproteins (Figure 1). It is interesting that rapamycin increasesproteolysis when only methylamine is present, suggesting thatinhibition of a lysosomal system of proteolysis is sufficientfor rapamycin to stimulate proteolysis.
To confirm that Akt is regulating a lysosomal pathway, we measuredproteolysis in cells that were treated with the DN Akt1 adenovirus.DN Akt alone had no significant effect on proteolysis, but itreversed EGF-induced suppression of proteolysis by >50% (P< 0.01; Figure 7A). Similarly, the use of the CA Akt1–expressingadenovirus caused a dose-dependent decrease in proteolysis (Figure 7B).In the presence of the lysosomal inhibitor methylamine, we foundthat DN Akt no longer affected proteolysis in EGF-treated cells(Figure 7C). These results are consistent with Akt influencinga lysosomal system of proteolysis.
Macroautophagy Is Not Regulated by EGF or Rapamycin
Because it was reported that growth factor signaling throughTOR proteins acts as a negative regulator of macroautophagy(34), we examined proteolysis through this system. We treatedcells with 3-methyladenine, which blocks the class III PI3Kthat is required for macroautophagy. Consistent with our reportthat EGF regulates a lysosomal pathway that is independent ofmacroautophagy (namely, CMA), we found that EGF still suppressedproteolysis despite 3-methyladenine treatment (Figure 8) (7).There may be a small change in the degree of the EGF responsewhen 3-methyladenine is present, consistent with the slightdecrease in Akt phosphorylation with 3-methyladenine treatment.Rapamycin still reversed the suppression of proteolysis thatwas produced by EGF when 3-methyladenine was present. Unlikemethylamine, the presence of 3-methyladenine was not sufficientfor rapamycin to stimulate proteolysis. These results suggestthat both EGF and rapamycin affect protein breakdown independentof macroautophagy.
Figure 8. An inhibitor of macroautophagy does not influence the effect of EGF or rapamycin. NRK-52E cells were grown and treated as in Figure 1, and proteolysis was measured as in Figure 5. Cells were treated with macroautophagy inhibitor 3-methyladenine (10 mM) in the presence of EGF, rapamycin, or both, or vehicle. Data are presented as mean slope. The figure is a representative experiments of three repeats (n = 6/group). *P < 0.05 versus control; +P < 0.05 versus EGF alone.
To confirm this conclusion, we further examined macroautophagyin NRK-52E cells by using the autofluorescence chemical MDC.MDC stains double-membrane vesicles, allowing identificationof autophagasomes in the cell cytoplasm (26). Quiescent NRK-52Ecells show ample granular, cytoplasmic MDC staining, and thisstaining is decreased by 3-methyladenine (10 mM; Figure 9).In contrast, MDC staining was not affected by EGF and/or rapamycin,indicating that neither treatment alters the number of autophagasomes(Figure 9). Thus, we found no evidence of regulation of macroautophagyin this model.
Figure 9. Rapamycin does not increase autophagic vacuoles. After treatment as in Figure 1, NRK-52E cells were incubated with monodansylcadaverine (MDC) as described in the Materials and Methods section. Intracellular MDC was measured by fluorescence photometry. A representative photomicrograph is shown of four repeats. (A) Control. (B) 3-Methyladenine (10 mM). (C) EGF (10 nM). (D) Rapamycin (50 nM) (E) EGF (10 nM) and rapamycin (50 nM). (F) Quantitative values for each treatment group are plotted with and without 3-methyladenine present (n = 4). The decrease with 3-methyladenine is statistically significant in all groups (P < 0.05), but there is no significant difference between treatments. Magnification, x100.
Rapamycin Stimulates a Proteasomal System
Because rapamycin did not significantly affect either CMA ormacroautophagy in EGF-treated cells, we tested whether it increasesproteasomal proteolysis using MG132 (0.5 µM). As we haveshown previously (24), MG132 strongly reduced proteolysis, butit did not prevent EGF from further suppressing proteolysis(Figure 10A). It is interesting that when MG132 was added, rapamycinno longer reversed the suppression of proteolysis by EGF. Asa second test, we treated the cells with the more specific proteasomalinhibitor lactacystin and again found that the proteasomal inhibitoreliminated the effect of rapamycin (Figure 10B). These resultsstrongly indicate that rapamycin activates a proteasomal pathwayin the presence of EGF.
Figure 10. Proteasome inhibitors eliminate the effect of rapamycin on proteolysis. NRK-52E cells were grown and treated as in Figure 1, and proteolysis was measured as in Figure 5. The figures are the representative experiments of three repeats (n = 6/group). *P < 0.05 versus control; +P < 0.05 versus EGF alone. (A) Cells were treated with proteasomal inhibitor MG132 (0.5 µM) in the presence of EGF, rapamycin, or both, or vehicle. (B) Cells were treated with proteasomal inhibitor lactacystin (10 µM) in the presence of EGF, rapamycin, or both, or vehicle.
Eukaryotic cells have evolved a complex set of mechanisms tomodulate protein synthesis and degradation to allow for homeostasis,growth, or atrophy. Type I PI3K and its downstream mediatorsAkt and mTOR are critical enzymes that control and integrateprotein synthesis and degradation (3,35,36). Recent reportshave identified type I PI3K as a key regulator of protein degradation,but both the signaling elements that are downstream of typeI PI 3 kinase and the proteolytic systems vary widely in differentcell types. For example, PI3K regulates ubiquitin/proteasome-dependentproteolysis in skeletal muscle by a mechanism that involvesAkt and the FoxO transcription factors (37,38), whereas in hepatocytesand in human colon cancer HT-29 cells, regulation of macroautophagythrough PI3K involves Akt and mTOR (19,39,40). In NIH 3T3 fibroblasts,regulation of macroautophagy does not involve mTOR (20). Thus,it is clear that the PI3K signaling pathways regulate differentsystems of proteolysis through different downstream effectorsin different cell types.
In NRK-52E cells, the suppression of proteolysis by EGF is preventedby inhibiting Akt or mTOR, but the accumulation of the substratesof CMA is blocked only by Akt inhibition. Importantly, PI3Kand Akt activity is increased in the renal cortex during diabeticrenal hypertrophy (41), a condition whereby CMA is reduced (8).The proto-oncogene Akt seems to be situated uniquely to regulateprotein and energy metabolism. Akt activates mTOR, which influences4EBP-1 and p70 S6 kinase to stimulate protein synthesis, andinactivates glycogen synthetase kinase GSK3 and the FoxO transcriptionfactors to influence glucose metabolism (42). However, it isnot immediately clear how Akt regulates CMA. Our data demonstratethat mTOR is not involved in the regulation of CMA. Becauserapamycin suppressed the phosphorylation of p70 S6 kinase (35)(Figure 2) but not lysosomal protein degradation, it is unlikelythat p70 S6 kinase is involved in the regulation of lysosomalproteolysis. We also found that inhibiting the mitogen-activatedprotein kinase pathway decreases p70 S6 kinase phosphorylationand suppresses both protein synthesis and degradation (21) (W.S.and H.A.F., manuscript in preparation), further suggesting thatp70 S6 kinase phosphorylation in renal tubular cells does notcorrelate with changes in proteolysis.
Akt also phosphorylates and inactivates subclass O of the FBox (Forkhead) family of transcription factors, FoxO. When dephosphorylated,FoxO proteins increase the transcription of many genes thatare important for surviving the unfed state; phosphorylationby Akt prevents this response (43). In skeletal muscle, insulinspecifically downregulates proteasomal proteolysis through aprocess that involves PI3K signaling (25). One way that insulinregulates the ubiquitin-proteasome system occurs through muscle-specificubiquitin E3 ligases, whose transcription is driven by PI3K,Akt, and FoxO (37,38). Because these effectors are muscle specific,there is at least the possibility that similar signaling mightregulate a lysosomal system.
Quiescent NRK-52E cells have considerable activity of theirlysosomal proteolytic pathways that can be suppressed by EGF(24), but we have never previously examined macroautophagy.MDC staining shows ample macroautophagic vesicles in control,quiescent NRK-52E cells (Figure 9), indicating that macroautophagyis active. EGF treatment did not alter MDC staining in NRK-52Ecells, a result that is consistent with our observation thatEGF regulates CMA in these cells (7). Furthermore, rapamycindid not stimulate staining over the baseline levels of activity,even when EGF was present. Furthermore, 3-methyladenine didnot greatly alter the effects of EGF or rapamycin on proteolysis(Figure 8). The small inhibition of EGF-induced proteolysiswith 3-methyladenine, seen in Figure 8, most likely resultsfrom the partial inhibition of Akt phosphorylation, seen inFigure 2. Therefore, unlike reports in other epithelial cells(e.g., hepatocytes and intestinal carcinoma cells [18,34,44]),we found no evidence indicating that macroautophagy is significantlyregulated in renal epithelial cells.
That rapamycin acts by stimulating protein degradation in EGF-treatedcells via a proteasomal pathway is a novel and unexpected finding.Rapamycin had no effect unless growth factors or lysosomal inhibitorsalso were present (Figure 6). Some studies have suggested thatby degrading common substrates, a form of communication betweenlysosomal and proteasomal pathways exists. Lysosomes may sequesterand degrade certain ubiquitinated proteins (45), and monoubiquitinationmay target plasma membrane protein for rapid entry into thelysosomal pathway (46). This suggests a model that mTOR signalingprevents the proteasome from increasing degradation of proteinsthat ordinarily would be destroyed by CMA. Rapamycin has beendescribed to accelerate the degradation of many proteasomalsubstrates, presumably by upregulation of ubiquitin conjugation(47). However, bulk proteolytic flux through the ubiquitin-proteasomepathway can be regulated by multiple factors, including theabundance of ubiquitin, the abundance of 26S proteasomes, therate of translocation of the 19S cap, and the proteolytic rateof the 20S core particle (48).
Growth factors, including EGF, play a significant role in increasingprotein synthesis and suppressing protein degradation in renalhypertrophy (49). PI3K and Akt suppress the lysosomal CMA systemin EGF-treated renal tubular cells, leading to accumulationof KFERQ-containing proteins, whereas mTOR seems to regulatea separate proteasomal system in these cells. As inhibitorsof growth factor–mediated signaling (e.g., rapamycin andAkt inhibitors) become more widespread in clinical medicine,working out the distinct signaling pathways that regulate proteolyticsystems in different tissues may lead to greater understandingof the mechanism of action and adverse effects of the pharmacologicagents.
Acknowledgments
This work was supported by National Institutes of Health grantK08 DK02496, a Veterans Administration Merit Review Award (toH.A.F.), and a fellowship grant from the National Kidney Foundationof Georgia (to W.S.).
We thank Drs. S. Russ Price and James L. Bailey for help, support,and careful reading of the manuscript.
Footnotes
Published online ahead of print. Publication date availableat www.jasn.org.
Franch HA, Price SR: Molecular signaling pathways regulating muscle proteolysis during atrophy. Curr Opin Clin Nutr Metab Care 8
: 271–275, 2005[Medline]
Kim J, Klionsky DJ: Autophagy, cytoplasm-to-vacuole targeting pathway, and pexophagy in yeast and mammalian cells. Annu Rev Biochem 69
: 303–342, 2000[CrossRef][Medline]
Shintani T, Klionsky DJ: Autophagy in health and disease: A double-edged sword. Science 306
: 990–995, 2004[Abstract/Free Full Text]
Mitch WE, Goldberg AL: Mechanisms of muscle wasting: The role of the ubiquitin-proteasome pathway. N Engl J Med 335
: 1897–1905, 1996[Free Full Text]
Franch HA, Sooparb S, Du J: A mechanism regulating proteolysis of specific proteins during renal tubular cell hypertrophy. J Biol Chem 276
: 19126–19131, 2001[Abstract/Free Full Text]
Sooparb S, Price SR, Shaoguang J, Franch HA: Suppression of chaperone-mediated autophagy in the renal cortex during acute diabetes mellitus. Kidney Int 65
: 2135–2144, 2004[CrossRef][Medline]
Cuervo AM: Autophagy: Many paths to the same end. Mol Cell Biochem 263
: 55–72, 2004[CrossRef][Medline]
Majeski AE, Dice JF: Mechanisms of chaperone-mediated autophagy. Int J Biochem Cell Biol 36
: 2435–2444, 2004[CrossRef][Medline]
Terlecky SR, Dice JF: Polypeptide import and degradation by isolated lysosomes. J Biol Chem 268
: 23490–23495, 1993[Abstract/Free Full Text]
Cuervo AM, Dice JF: A receptor for the selective uptake and degradation of proteins by the lysosome. Science 273
: 501–503, 1996[Abstract]
Kobayashi S, Venkatashalam MA: Differential effects of calorie restriction on glomeruli and tubules of the remnant kidney. Kidney Int 42
: 710–717, 1992[Medline]
Vanhaesebroeck B, Waterfield MD: Signaling by distinct classes of phosphoinositide 3-kinases. Exp Cell Res 253
: 239–254, 1999[CrossRef][Medline]
Rameh LE, Cantley LC: The role of phosphoinositide 3-kinase lipid products in cell function. J Biol Chem 274
: 8347–8350, 1999[Free Full Text]
Jope RS, Johnson GV: The glamour and gloom of glycogen synthase kinase-3. Trends Biochem Sci 29
: 95–102, 2004[CrossRef][Medline]
Sarbassov DD, Guertin DA, Ali SM, Sabatini DM: Phosphorylation and regulation of Akt/PKB by the rictor-mTOR complex. Science 307
: 1098–1101, 2005[Abstract/Free Full Text]
Petiot A, Ogier-Denis E, Blommaart EFC, Meijer AJ, Codogno P: Distinct classes of phosphatidylinositol 3'-kinases are involved in signaling pathways that control macroautophagy in HT-29 cells. J Biol Chem 275
: 992–998, 2000[Abstract/Free Full Text]
Arico S, Petiot A, Bauvy C, Dubbelhuis PF, Meijer AJ, Codogno P, Ogier-Denis E: The tumor suppressor PTEN positively regulates macroautophagy by inhibiting the phosphatidylinositol 3-kinase/protein kinase B pathway. J Biol Chem 276
: 35243–35246, 2001[Abstract/Free Full Text]
Furuta S, Hidaka E, Ogata A, Yokota S, Kamata T: Ras is involved in the negative control of autophagy through the class I PI3-kinase. Oncogene 23
: 3898–3905, 2004[CrossRef][Medline]
Franch HA, Wang X, Sooparb S, Brown NS, Du J: Phosphatidylinositol 3-kinase activity is required for epidermal growth factor to suppress proteolysis. J Am Soc Nephrol 13
: 903–908, 2002[Abstract/Free Full Text]
Skurk C, Maatz H, Kim HS, Yang J, Abid MR, Aird WC, Walsh K: The Akt-regulated forkhead transcription factor FOXO3a controls endothelial cell viability through modulation of the caspase-8 inhibitor FLIP. J Biol Chem 279
: 1513–1525, 2004[Abstract/Free Full Text]
He TC, Zhou S, da Costa LT, Yu J, Kinzler KW, Vogelstein B: A simplified system for generating recombinant adenoviruses. Proc Natl Acad Sci U S A 95
: 2509–2514, 1998[Abstract/Free Full Text]
Franch HA, Curtis PV, Mitch WE: Mechanisms of renal tubular cell hypertrophy: mitogen-induced suppression of proteolysis. Am J Physiol 273
: C843–C851, 1997
Franch HA, Raissi S, Wang X, Zheng B, Bailey JL, Price SR: Acidosis impairs insulin receptor substrate-1-associated phosphoinositide 3-kinase signaling in muscle cells: Consequences on proteolysis. Am J Physiol Renal Physiol 287
: F700–F706, 2004[Abstract/Free Full Text]
Munafo DB, Colombo MI: A novel assay to study autophagy: Regulation of autophagosome vacuole size by amino acid deprivation. J Cell Sci 114
: 3619–3629, 2001
Franch HA, Shay JW, Alpern RJ, Preisig PA: Involvement of pRB family in TGFbeta-dependent epithelial cell hypertrophy. J Cell Biol 129
: 245–254, 1995[Abstract/Free Full Text]
Liu S, Cieslinski DA, Funke AJ, Humes HD: Transforming growth factor-beta 1 regulates the expression of Pax-2, a developmental control gene, in renal tubule cells. Exp Nephrol 5
: 295–300, 1997[Medline]
Rodrigeuz-Viciana P, Warne PH, Khwaja A, Marte BM, Pappin D, Das P, Waterfield MD, Ridley A, Downward J: Role of phosphoinositide 3-OH kinase in cell transformation and control of the actin cytoskeleton by Ras. Cell 89
: 401–410, 1997
Mizushima N: Methods for monitoring autophagy. Int J Biochem Cell Biol 36
: 2491–2502, 2004[CrossRef][Medline]
Beugnet A, Tee AR, Taylor PM, Proud CG: Regulation of targets of mTOR (mammalian target of rapamycin) signaling by intracellular amino acid availability. Biochem J 372
: 555–566, 2003[CrossRef][Medline]
Franch HA: Modification of the epidermal growth factor response by ammonia in renal cell hypertrophy. J Am Soc Nephrol 11
: 1631–1638, 2000[Abstract/Free Full Text]
Jurkovitz CT, England BK, Ebb RG, Mitch WE: Influence of ammonia and pH on protein and amino acid metabolism in LLC-PK1 cells. Kidney Int 42
: 595–601, 1992[Medline]
Blommaart EFC, Luiken JJFP, Meijer AJ: Autophagic proteolysis: Control and specificity. Histochem. J 29
: 365–385, 1997[CrossRef][Medline]
Harris TE, Lawrence JC Jr: TOR signaling. Science STKE 2003
: re 15, 2003
Proud CG: Turned on by insulin. Nature 371
: 747–748, 1995
Sandri M, Sandri C, Gilbert A, Skurk C, Calabria E, Picard A, Walsh K, Schiaffino S, Lecker SH, Goldberg AL: Foxo transcription factors induce the atrophy-related ubiquitin ligase atrogin-1 and cause skeletal muscle atrophy. Cell 117
: 399–412, 2004[CrossRef][Medline]
Stitt TN, Drujan D, Clarke BA, Panaro F, Timofeyva Y, Kline WO, Gonzalez M, Yancopoulos GD, Glass DJ: The IGF-1/PI3K/Akt pathway prevents expression of muscle atrophy-induced ubiquitin ligases by inhibiting FOXO transcription factors. Mol Cell 14
: 395–403, 2004[CrossRef][Medline]
Blommaart EF, Luiken JJ, Blommaart PJ, van Woerkom GM, Meijer AJ: Phosphorylation of ribosomal protein S6 is inhibitory for autophagy in isolated rat hepatocytes. J Biol Chem 270
: 2320–2326, 1995[Abstract/Free Full Text]
Shigemitsu K, Tsujishita Y, Hara K, Nanahoshi M, Avruch J, Yonezawa K: Regulation of translational effectors by amino acid and mammalian target of rapamycin signaling pathways. Possible involvement of autophagy in cultured hepatoma cells. J Biol Chem 274
: 1058–1065, 1999[Abstract/Free Full Text]
Feliers D, Duraisamy S, Faulkner JL, Duch J, Lee AV, Abboud HE, Choudhury GG, Kasinath BS: Activation of renal signaling pathways in db/db mice with type 2 diabetes. Kidney Int 60
: 495–504, 2001[CrossRef][Medline]
Yang ZZ, Tschopp O, Baudry A, Dummler B, Hynx D, Hemmings BA: Physiological functions of protein kinase B/Akt. Biochem Soc Trans 32
: 350–354, 2004[CrossRef][Medline]
Boon L, Blommaart PJE, Meijer AJ, Lamers WH, Schoolwerth AC: Acute acidosis inhibits liver amino acid transport: No primary role for the urea cycle in acid-base balance. Am J Physiol 267
: F1015–F1020, 1994
Lenk SE, Susan PP, Hickson I, Jasionowski T, Dunn WA Jr: Ubiquitinated aldolase B accumulates during starvation-induced lysosomal proteolysis. J Cell Physiol 178
: 17–27, 1999[CrossRef][Medline]
De Wit R, Makkinje M, Boonstra J, Verkleij AJ, Post JA: Hydrogen peroxide reversibly inhibits epidermal growth factor (EGF) receptor internalization and coincident ubiquitination of the EGF receptor and Eps15. FASEB J 15
: 306–308, 2001[Free Full Text]
Riesterer O, Zingg D, Hummerjohann J, Bodis S, Pruschy M: Degradation of PKB/Akt protein by inhibition of the VEGF receptor/mTOR pathway in endothelial cells. Oncogene 23
: 4624–4635, 2004[CrossRef][Medline]
Price SR: Increased transcription of ubiquitin-proteasome system components: Molecular responses associated with muscle atrophy. Int J Biochem Cell Biol 35
: 617–628, 2003[CrossRef][Medline]
Hammerman MR, O’Shea M, Miller SB: Role of growth factors in regulation of renal growth. Annu Rev Physiol 55
: 305–321, 1993[CrossRef][Medline]
This article has been cited by other articles:
X. Wang, J. Hu, and S. R. Price Inhibition of PI3-kinase signaling by glucocorticoids results in increased branched-chain amino acid degradation in renal epithelial cells
Am J Physiol Cell Physiol,
May 1, 2007;
292(5):
C1874 - C1879.
[Abstract][Full Text][PDF]
Top
Abstract
Introduction
by approximately 42% (Figure 4A). It also suppressed EGF-induced accumulation of pax2 and GAPDH in a dose-dependent manner without changing the level of actin (Figure 4, B and C). These results suggest that Akt is the downstream effector of PI3K that regulates CMA. 
