TGF-β signaling and its effect on glutaminase expression in LLC-PK1-FBPase+ cells
Abstract
During systemic acidosis, renal proximal tubular cells exhibit enhanced rates of bicarbonate and ammonium ion synthesis and undergo extensive hypertrophy. The former adaptations are accomplished, in part, by increased expression of glutaminase (GA). LLC-PK1-FBPase+ cells, a gluconeogenic line of porcine kidney cells, exhibit a rapid activation of the ERK1/2 and p38 MAPK pathways and a two- to threefold increase in GA mRNA when transferred to acidic medium (pH 6.9). Transforming growth factor-β (TGF-β), a potent activator of MAPK and Smad signaling cascades, also causes extensive renal hypertrophy. Thus the potential role of TGF-β in the renal response to metabolic acidosis was investigated. Western blot analyses established that in LLC-PK1-FBPase+ cells, TGF-β activated the ERK1/2, p38 MAPK, and Smad1/5/8 pathways, but not the JNK and Smad2/3 pathways. Addition of TGF-β to cells cultured in normal medium (pH 7.4) produced a steady increase in GA mRNA, resulting in a twofold induction after 18 h. Western blot analysis indicated that treatment with either TGF-β or acidic medium resulted in an increased level of fibronectin. However, the effects of the two treatments on both GA mRNA and fibronectin levels occurred with different time courses and were additive. In addition, the rates of ammonia production were decreased slightly by addition of TGF-β. Finally, a GA-luciferase reporter construct, which is activated 3.5-fold by treatment with acidic medium, is not affected by TGF-β. Therefore, TGF-β and metabolic acidosis activate some of the same signaling pathways in LLC-PK1-FBPase+ cells, but produce separate effects on GA expression.
the onset of a systemic metabolic acidosis initiates an array of adaptive responses in the rat renal proximal tubule that results in rapid and pronounced increases in the catabolism of glutamine and the synthesis of glucose (3, 34). This response also results in increased reabsorption and production of bicarbonate ions and an enhanced excretion of acid equivalents (11). This adaptation is sustained, in part, by increased expression of key enzymes of glutamine metabolism including mitochondrial glutaminase (GA) and glutamate dehydrogenase (GDH) and cytosolic phosphoenolpyruvate carboxykinase (PEPCK). The adaptive increases in GA results from an increased rate of synthesis that correlates with a comparable increase in the level of the GA mRNA (15). The latter increase results from selective stabilization of the GA mRNA (19, 20). During acidosis, the renal proximal tubule also undergoes extensive hypertrophy (29).
LLC-PK1-FBPase+ cells are a gluconeogenic and pH-responsive line of porcine renal epithelial cells that exhibit a number of metabolic features specific to the proximal tubule (9). These cells effectively model in vitro the adaptive responses in renal glutamine metabolism. On incubation with acidic medium (pH 6.9), LLC-PK1-FBPase+ cells respond with increased rates of ammonia production, accompanied by a two- to threefold increase in GA mRNA levels (10, 14). Furthermore, when incubated in low-potassium (0.7 mM)-containing medium for 24–72 h, to elicit an intracellular acidification at normal extracellular pH (7.4), LLC-PK1-FBPase+ cells again respond with increased levels of GA mRNA, suggesting that a fall in intracellular pH triggers this adaptive response (10). More recently, it was established that the pH-responsive induction of PEPCK mRNA in LLC-PK1-FBPase+ cells is mediated by the SB203580-sensitive p38α MAPK/ATF-2 signaling pathway (6, 28).
In addition to metabolic acidosis, a number of pathophysiological conditions, including compensatory renal growth, diabetes mellitus, high-protein diets, and chronic potassium deficiency cause renal hypertrophy (29, 38). In renal tubular epithelial cultures, cellular hypertrophy is produced in response to angiotensin II and insulin, to various growth factors and cytokines, and to increases in ammonia or glucose concentrations (7). Over the past several years, transforming growth factor-β (TGF-β) has been recognized as a central player in renal hypertrophy and fibrogenesis. TGF-β stimulates the production of extracellular matrix, including proteoglycans, fibronectin, collagens type I, II, and IV, and laminin in renal epithelial cells (2, 12, 26). In several animal models of diabetes, TGF-β mRNA and protein levels are significantly increased (7, 44, 45). The overexpression of this cytokine in the kidney may be a direct effect of hyperglycemia, as in vitro studies have established that high glucose levels enhance tubular production of TGF-β (13). Tubular cells, including LLC-PK1 cells, when cultured under high-glucose conditions (8, 13, 45), exhibit an increased phosphorylation, and thus activation, of ERK1/2 and p38 MAPK. Preincubation of the cells with the p38 MAPK inhibitor SB203580 or the MEK1/2 inhibitor PD098059 suppressed the high glucose-induced increases in protein content, [3H]leucine incorporation, and the protein-to-DNA ratio. Thus ERK1/2 and p38 MAPK are activated under high-glucose conditions and may mediate cellular hypertrophy (8).
It was recently hypothesized (37) that TGF-β may activate GA and apical Na+/H+ exchanger (NHE3) gene expression to enhance intracellular catabolism of glutamine. This hypothesis and the striking similarities between the hypertrophic action of TGF-β and the adaptive response of proximal tubular cells to metabolic acidosis prompted us to investigate the effects of TGF-β in LLC-PK1-FBPase+ cells. The resulting data suggest that TGF-β and metabolic acidosis activate some of the same signaling pathways. However, they produce additive effects on GA mRNA expression and cellular fibronectin levels and separate effects on glutamine metabolism.
MATERIALS AND METHODS
Cell culture.
Gluconeogenic LLC-PK1-FBPase+ cells (9) were cultured in DMEM with 5.5 mM d-glucose, 2 mM l-glutamine, and 26 mM NaHCO3 (pH 7.5 at 5% CO2), supplemented with 10% fetal bovine serum, 100 U/ml penicillin, and 100 mg/ml streptomycin. Culture media were prepared by supplementing bicarbonate- and glutamine-free DMEM-base (D-5030, Sigma-Aldrich) with 2.2 g/l (26.2 mM) NaHCO3, 1.0 g/l (17.8 mM) NaCl (to correct medium osmolarity), 1.0 g/l (5.5 mM) d-glucose, and l-glutamine from freshly prepared stock solutions to yield a final concentration of 2.0 mM (6, 9). Acidic culture medium (pH 6.9 at 5% CO2) was prepared similarly, except that 0.76 g/l (9.0 mM) NaHCO3 and 2.0 g/l NaCl (35 mM) were added to reduce pH and to maintain osmolarity, respectively (10). All media additives were tissue culture grade and were obtained from Sigma-Aldrich. Biochemicals and buffer chemicals were of the highest analytic grade available and were obtained from Roche Applied Science or Sigma-Aldrich.
Cultures were incubated at 37°C in a 5% CO2-95% air atmosphere. Cultures were maintained in 10-cm tissue culture dishes with 10 ml of culture medium, and the media were changed three times a week. Confluent monolayers were subcultured (split ratio 1:3) using 0.25% trypsin and 0.02% EDTA in Ca2+- and Mg2+-free buffered saline. Experimental cultures were grown for 10 days to produce confluent monolayers.
Incubation protocols and cell harvest.
To model a metabolic acidosis, cultures were incubated in acidic culture medium (pH 6.9) for 1.5–18 h. Control (pH 7.4) and acidic cultures were treated with porcine TGF-β1 (T-5050, Sigma-Aldrich). Ammonia was analyzed enzymatically in culture media samples (10).
For Western blotting, monolayer cultures were harvested in freshly prepared ice-cold lysis buffer (RIPA buffer) (6) and lysed on ice for 20 min. For Northern blot analysis, total RNA was isolated using acidic guanidinium isothyocyanate (TRI Reagent, MRC) (6, 10).
Western blotting and antibodies.
SDS-PAGE was performed under standard denaturing conditions. Equal amounts of protein were loaded onto each lane. Electrophoresis was performed overnight at a constant 80 V. Gels were blotted immediately following electrophoresis onto polyvinylidene difluoride membrane (Immobilon-P, Millipore), and blots were processed according to the instructions of the manufacturers of the antibodies. The following antibodies were used: PhosphoPlus Antibody Kits (Cell Signaling Technology) for the detection of phosphorylated forms of ERK1/2 (p44/42 MAPK Thr202/Tyr204 Antibody Kit 9100), SAPK/JNK (Thr183/Tyr185 Antibody Kit 9250), p38 (Thr180/Tyr182 Antibody Kit 9210), and MKK3/6 (Ser189/207 Antibody Kit 9230), respectively. The state of phosphorylation of transcription factors and MAPKAPK-2 was determined using phospho-ATF-2 (Thr71) antibody (Cell Signaling Technology 9221), phospho-Elk-1 (Ser383) antibody (Cell Signaling Technology 9181), and phospho-MAPKAPK-2 (Thr222) antibody (Cell Signaling Technology 3044). TGF-β-signaling via Smad activation was determined using phospho-Smad1 (Ser463/465)/Smad5 (Ser463/465)/Smad8 (Ser426/428) antibody (Cell Signaling Technology 9511), phospho-Smad2 (Ser465/467) antibody (Cell Signaling Technology 3101), and phospho-Smad2/3 antibody from Santa Cruz Biotechnology (sc-11769-R). Whole-protein antibodies were from either Cell Signaling Technology or from Santa Cruz Biotechnology, and anti-fibronectin antibody was from Sigma-Aldrich. The respective proteins were visualized using the LumiGLO System of enhanced chemiluminescence (Cell Signaling Technology). Blots were exposed to Hyperfilm ECL (Amersham).
p38 MAPK assay.
Confluent cultures of LLC-PK1-FBPase+ cells were serum starved for 24 h and then incubated with TGF-β1 for 0.5–16 h. The cells were then analyzed for p38 kinase activity using a nonradioactive p38 MAPK Assay Kit (model 9820, Cell Signaling Technology,). Cell lysates were first subjected to immunoprecipitation using phospho-p38 MAPK antibody. The resultant immunoprecipitate was subsequently incubated with ATF-2 fusion protein in the presence of ATP and kinase buffer. This allowed immunoprecipitated active p38 MAPK to phosphorylate the ATF-2 fusion protein. Kinase activity was assayed through detection of phosphorylated ATF-2 (Thr71) by Western blot analysis and quantified by densitometry.
Northern blot analysis and cDNA probes.
Formaldehyde-agarose gel electrophoresis of total RNA samples, transfer to GeneScreen Plus membranes (New England Nuclear), and hybridization and washings of blots were performed as described previously (6, 10). The GA cDNA probe (p930) is a KpnI and EcoRI fragment of a PCR-amplified 1.1-kb segment of the coding region of the pH-responsive 4.5-kb pig GA mRNA (10).
Luciferase assays.
A XhoI-HindIII fragment (−401 to +110) of the rat GA promoter was inserted at the XhoI and HindIII sites within the multicloning site of pGL3-Basic (Promega) to create p401GA-Luc3. A 1-kb SpeI-XbaI fragment encoding the 3′-untranslated region (UTR) of the GA mRNA was isolated from pCRScript-βG-GA (31) and ligated into the XbaI site of p401GA-Luc3 to generate p401GA-Luc3-GAUTR. A poly A signal is provided by the SV40 sequence within pGL3-Basic. LLC-PK1-FBPase+ cells were grown for 7 days in control medium in six-well plates. The cells were transiently cotransfected with 0.6 μg of p401GA-Luc3-GAUTR and 0.1 μg of pRLnull (Promega) per well by calcium phosphate transfection. Approximately 24 h later, the cells were washed twice with 2 ml of phosphate-buffered saline and then fed with 2 ml of control or acidic medium lacking serum. After 24 h, 2 μl of 4 mM HCl containing 1 mg/ml BSA (control) or 2 μl of 1 μg/ml TGF-β suspended in the HCl-BSA solution was added to the cells without replacing the medium. The cells were cultured for an additional 24 h and then washed with 2 ml of phosphate-buffered saline. Cell extracts were prepared and assayed using the reagents contained in the Dual Luciferase Reporter Assay. The firefly luciferase activity was divided by the corresponding Renilla luciferase activity to correct for differences in transfection efficiency. The mean of the ratio of the luciferase activities measured for the control samples was normalized to a value of one. The number of experiments was 4–6 independent repeats.
Statistical analysis of results was performed using an unpaired Student's t-test.
RESULTS
TGF-β activates the p38 MAPK signaling pathway in LLC-PK1-FBPase+ cells.
Initial experiments were performed to characterize the signaling pathways that are activated by porcine TGF-β1 in pig renal epithelial LLC-PK1-FBPase+ cells. Significant phosphorylation, and thus activation, of MKK3/6 and p38 MAPK was detectable within 1 h of TGF-β stimulation (Fig. 1A). Parallel blots probed with antibodies against total MKK3 and p38 MAPK established that TGF-β did not alter the basal expression of either kinase (Fig. 1A). The potential downstream targets of TGF-β1 activation of the p38 MAPK were investigated using antibodies that are specific for the phosphorylated forms of ATF-2, Elk-1, and MAPKAPK-2 (Fig. 1B). A pronounced phosphorylation of ATF-2 was also observed within 1 h of TGF-β treatment, consistent with the time course for activation of the upstream p38 MAPK. Phosphorylations of the transcription factor, Elk-1, and the MAPKAPK-2 were also evident, but both occurred with slower time courses. The TGF-β1-mediated activation of p38 was confirmed by a direct assay of p38 MAPK activity that quantified the site-specific phosphorylation of an ATF-2 fusion protein (Fig. 2). An initial increase in TGF-β1-induced p38 MAPK activity was again detectable within 1 h. By 3 h, the measured p38 MAPK activity was increased 2.5-fold. Thus the time course of activation closely paralleled the increase in phosphorylation of p38 MAPK (Fig. 1A).
Fig. 1.Transforming growth factor (TGF)-β activation of the p38 MAPK signaling in LLC-PK1-FBPase+ cells. Confluent cultures of LLC-PK1-FBPase+ cells were serum depleted for 24 h and then stimulated by the addition of 10 ng/ml porcine TGF-β1 for 0.5–16 h. Phosphorylation, and thus activation, of p38 MAPK, of the upstream MAP kinase kinases MKK3/MKK6 and of the downstream transcription factors ATF-2 and Elk-1, and of MAPKAP kinase-2, respectively, was characterized by Western blot analysis using phospho-specific antibodies. A: time course (0.5–16 h) of MKK3/6 and p38 MAPK phosphorylation by TGF-β1. Gels were also probed with antibodies against total MKK3 and p38 MAPK. B: time course of phosphorylation of transcription factors ATF-2 and Elk-1, and of MAPKAP kinase-2 in LLC-PK1-FBPase+ cells by TGF-β.
Fig. 2.Effect of TGF-β on p38 MAPK activity. LLC-PK1-FBPase+ cells were incubated with exogenous porcine TGF-β1 (2 ng/ml) for the indicated times. Cell lysates were prepared and assayed for p38 MAPK activity by immunoprecipitation with phospho-p38 MAPK antibody followed by detection of phosphorylated ATF-2 fusion protein through Western blotting with phospho-ATF-2(Thr71) antibody. Phosphorylation signals were quantified by densitometry. Data are means of 2 independent series of experiments.
TGF-β1 activates the ERK1/2 MAPK pathway in LLC-PK1-FBPase+ cells.
Phospho-p44/42 polyclonal antibodies were used to determine whether TGF-β1 also activates the ERK1/2 MAPK pathway. Stimulation with epidermal growth factor (EGF), a potent activator of the ERK1/2 cascade, served as a positive control (Fig. 3A). As shown in Fig. 3B, an increase in phosphorylation of ERK1/2 was observed as early as 1 h after stimulation with porcine TGF-β1. In comparison, EGF-induced activation of ERK1/2 occurred even more rapidly. However, EGF and TGF-β1 produced similar levels of phosphorylation of ERK1/2. Specific inhibitors were used to determine whether ERK1/2 phosphorylation occurs via TGF-β-mediated activation of the upstream kinases, MEK1/2. LLC-PK1-FBPase+ cells were preincubated with PD098059 or U0126 before stimulation with TGF-β1. U0126 strongly inhibited TGF-β-mediated ERK1/2 activation (Fig. 3C), while PD098059 produced a significant, but less pronounced inhibition (Fig. 3D).
Fig. 3.Western blot analysis of ERK1/2 phosphorylation in LLC-PK1-FBPase+ cells by TGF-β. Confluent LLC-PK1-FBPase+ monolayers were incubated for the indicated times with either 100 ng/ml of EGF (A) or 10 ng/ml of porcine TGF-β1 (B). Cell lysates were prepared, separated by SDS-PAGE, and subjected to Western blot analysis using antibodies specific for phospho-ERK1/2 or total ERK1/2. The experiments in B were repeated using cells that were preincubated with the MEK1/2 inhibitors U0126 (10 μM; C) and PD098059 (50 μM; D) for 30 min before addition of porcine TGF-β1.
Lack of SAPK/JNK activation by TGF-β1 in LLC-PK1-FBPase+ cells.
Phospho-SAPK/JNK polyclonal antibodies were used to also determine whether TGF-β1 activated the SAPK/JNK pathway in LLC-PK1-FBPase+ cells (Fig. 4). SAPK/JNK activation by anisomycin served as a positive control. In contrast to p38 MAPK and ERK1/2, no significant increase in phosphorylation of SAPK/JNK was observed even with 16 h of TGF-β1 treatment.
Fig. 4.Lack of SAPK/JNK activation by TGF-β in LLC-PK1-FBPase+ cells. Confluent LLC-PK1-FBPase+ cultures were incubated with porcine TGF-β1 (10 ng/ml) for the indicated times and subsequently subjected to Western blot analysis using antibodies specific for phospho-SAPK/JNK or total SAPK/JNK. As a positive control, LLC-PK1- FBPase+ cells were pretreated for 60 min with 5 μM anisomycin (ai; lane 1).
R-Smads Smad1/5/8 are transiently activated by TGF-β1 in LLC-PK1-FBPase+ cells.
The archetypal mode of intracellular TGF-β signaling is mediated through the receptor-regulated Smad proteins (R-Smads) (5, 24, 25). To determine which R-Smad subfamily is activated by TGF-β1 in LLC-PK1-FBPase+ cells, Western blot analyses were performed using anti-phospho-Smad2/3 and anti-phospho-Smad1/5/8 polyclonal antibodies. As shown in Fig. 5A, a strong Smad1/5/8 activation was observed. Increases in phosphorylation were evident as early as 0.5 h after stimulation with porcine TGF-β1. In contrast, no phosphorylation of Smad2/3 could be detected even after 16 h of TGF-β treatment (Fig. 5C).
Fig. 5.TGF-β activation of receptor-regulated Smad proteins (R-Smads) in LLC-PK1-FBPase+ cells. A: Western blot analysis of Smad1/5/8 activation. LLC-PK1-FBPase+ cells were stimulated with TGF-β1 (10 ng/ml) for 0.5–16 h and analyzed by Western blotting. B: to assess the potential role of p38 MAPK in the phosphorylation of Smad1/5/8, LLC-PK1-FBPase+ cultures were preincubated for 30 min with 10 μM SB203580 before addition of porcine TGF-β1. C: lack of TGF-β1 activation of Smad2/3 in LLC-PK1-FBPase+ cells. The phospho-specific antibodies detecting the phosphorylated forms and the antibodies against total Smad proteins were from either Cell Signaling Technology (*) or Santa Cruz Biotechnology (**).
Cross talk between MAPK and Smad pathways.
Previous studies demonstrated that SB203580 is a potent inhibitor of p38 MAPK in LLC-PK1-FBPase+ cells (6). However, preincubation of serum-starved cells with SB203580 did not block the TGF-β1-induced phosphorylation of Smad1/5/8 (Fig. 5B). Therefore, the prompt activation of Smad1/5/8 is not mediated via p38 MAPK. However, other kinases are also able to phosphorylate R-Smad proteins (25). For example, R-Smads contain phosphorylation sites for ERK1/2 (18). Therefore, the effect of MEK1/2 inhibition on TGF-β-induced Smad1/5/8 phosphorylation was also determined using the MEK1/2 inhibitors PD098059 or U0126 before stimulation with TGF-β1. Both MEK1/2 inhibitors blocked TGF-β-induced Smad1/5/8 phosphorylation (Fig. 6) but with different efficacies. Again, PD098059 only partially blocked Smad1/5/8 activation, whereas U0126 completely blocked Smad1/5/8 phosphorylation.
Fig. 6.Cross talk of Smad1/5/8 with ERK1/2 signaling. LLC-PK1-FBPase+ cultures were incubated for 0.5–3.0 h with 10 ng/ml TGF-β either alone or following preincubation for 30 min with the MEK1/2 inhibitors U0126 (10 μM) or PD098059 (50 μM). Cell lysates were separated by SDS-PAGE and probed for Smad1/5/8 phosphorylation using a phospho-specific antibody. A: densitometric quantification of three independent series of experiments. B: representative Western blots. Inset: representative Western chemiluminescence of anti-Smad1/5/8 total antibody.
TGF-β increases GA mRNA levels and stimulates fibronectin synthesis and deposition in LLC-PK1-FBPase+ cells.
The effect of porcine TGF-β1 on the cellular level of mitochondrial GA mRNA in LLC-PK1-FBPase+ cells was determined by Northern blot analysis. Previous studies have established that confluent cultures maintained in control (pH 7.4) medium express a constant basal level of GA mRNA, whereas the transfer of cells to acidic (pH 6.9) medium results in a gradual and sustained increase in GA mRNA (10, 14). The addition of TGF-β1 to cells cultured in normal medium produced a steady increase in GA mRNA levels that was initially evident after 3 h and that resulted in a twofold increase after 18 h (Fig. 7). However, the pH-responsive increase in GA mRNA and the effect of TGF-β were additive. By contrast, TGF-β decreased the release of ammonia into the culture medium. As displayed in Fig. 8, the acid-mediated increases in ammonia generation are comparable to earlier studies (10) and are statistically significant compared with control cells maintained at pH 7.4. However, the addition of TGF-β to cells treated with pH 6.9 medium produced a significant decrease in ammonium ion production compared with the acidic cultures without the cytokine.
Fig. 7.Northern blot analysis of the effects of TGF-β on cellular GA mRNA levels. LLC-PK1-FBPase+ cells were treated for 0–6 h using control medium (A) and for 0–18 h using control (pH 7.4, B, top) and acidic (pH 6.9) medium (B, middle). Confluent monolayers were stimulated with 10 ng/ml porcine TGF-β1 for the indicated times: 1.5–6 h in A or 3–18 h in B. Total cellular RNA was isolated, and samples (20 μg) were electrophoresed and blotted onto GeneScreen Plus membranes. Blots were hybridized with a GA cDNA probe (p930). Equal loading and RNA integrity were demonstrated by staining of 18S and 28S rRNA bands with ethidium bromide (B, bottom). C: GA mRNA levels were quantified by densitometric analysis and are reported as fold-changes compared with the untreated control (pH 7.4). Values are means ± SE of 3–9 independent sets of experiments. *P < 0.05 and **P < 0.01 compared with controls at either pH 7.4 or pH 6.9 before addition of TGF-β1.
Fig. 8.Rates of ammonia production by LLC-PK1-FBPase+ cells and response to TGF-β1. Confluent LLC-PK1-FBPase+ cultures were incubated in DMEM containing 2 mM freshly added glutamine at pH 7.4 or pH 6.9 in the absence and presence of TGF-β1 (10 ng/ml) for 72 h. At 24 h-intervals, culture media aliquots (50 μl) were taken and analyzed. Ammonia production rates are expressed as μmol·mg protein−1·24 h−1. Values are means ± SD of 6 independent series of experiments. *P < 0.01 compared with control cultures at pH 7.4. +P < 0.01 compared with acidic cultures in the absence of TGF-β1.
To compare the mechanisms that mediate the increases in GA mRNA, LLC-PK1-FBPase+ cells were transiently transfected with a luciferase construct that contained 401 bp of the promoter and 1 kb of the 3′-UTR of the rat KGA gene. Treatment of the transfected cells with acidic medium produced a 3.5-fold increase in luciferase activity (Fig. 9). However, addition of TGF-β1 had no effect on either control or acidic medium. Therefore, the observed effects of TGF-β on the levels of GA mRNA may be mediated through different elements than those used to produce the pH-responsive stabilization of the GA mRNA.
Fig. 9.Effect of TGF-β1 on luciferase activity of p401GA-Luc3-GAUTR in LLC-PK1-FBPase+ cells. Confluent cultures were transiently transfected with p401GA-Luc3-GAUTR and pRLnull and then treated for 24 h with control (pH 7.4, open bars) or acidic (pH 6.9, filled bars) medium in the absence or presence of TGF-β (2 ng/ml). Samples were assayed using a Dual Luciferase Reporter Assay kit, and the resulting firefly luciferase activities were divided by the corresponding Renilla luciferase activities. The mean of the ratios were normalized so that the control samples minus TGF-β were equal to one. Values are means ± SE of 4–6 independent samples. The increase in relative luciferase activities in control compared with acidic samples has a P < 0.001.
Western blots using an anti-fibronectin antibody were performed to assess the effects of TGF-β on the synthesis and deposition of extracellular matrix components by LLC-PK1-FBPase+ cells. A representative blot of five independent series of experiments is shown in Fig. 10. The transfer of cells to acidic (pH 6.9) culture conditions produced an increase in fibronectin deposition that peaked at 24 h. TGF-β also strongly stimulated cellular fibronectin synthesis. However, the increase in fibronectin deposition was maximal at 72 h after addition of TGF-β under both control and acidic culture conditions. Again, the pH-responsive increase in fibronectin synthesis and the effect of TGF-β were additive.
Fig. 10.Fibronectin biosynthesis by LLC-PK1-FBPase+ cells in response to metabolic acidosis and TGF-β. Confluent LLC-PK1 cultures were incubated under control (pH 7.4) or acidic conditions (pH 6.9) in the absence or presence of TGF-β1 (10 ng/ml) for 72 h. At 0, 24, and 72 h of incubation, confluent monolayers were harvested and subjected to Western blot analysis. Blots were stained with anti-fibronectin antibody.
DISCUSSION
The TGF-β family consists of >30 structurally related polypeptides including TGF-βs, activins, and bone morphogenetic proteins (BMP) (42). This family of secreted dimeric proteins regulates pivotal biological functions, including cell proliferation, differentiation, apoptosis, migration, and extracellular matrix production (22). Signaling by these cytokines occurs via ligand-induced heteromeric complex formation of distinct type I (TβRI) and type II (TβRII) receptors and the subsequent phosphorylation of TβRI by TβRII (Fig. 11). The activated TβRI propagates the signal within the cell through the phosphorylation of the COOH-terminal serine residue of specific R-Smad proteins. Whereas Smad2 and Smad3 act downstream of TGF-β and activin TβRI, Smad1, Smad5, and Smad8 are phosphorylated by BMP type I receptors (24, 25, 32). Activated R-Smads form heteromeric complexes with a common-partner (Co-) Smad (Smad4), which accumulate in the nucleus, where they control gene expression through interaction with other transcription factors, coactivators, and corepressors (1, 32).
Fig. 11.Pathways in TGF-β signal transduction. Signaling by TGF-β occurs via ligand-induced heteromeric complex formation of type I (TβRI) and type II (TβRII) receptors. Subsequently, TβRI activates the R-Smads Smad2 and Smad3, whereas the type II receptor subunit (TβRII) activates Smad1/5/8 and TGF-β-activated kinase 1 (TAK1), an upstream MAPKKK. Activated R-Smads (Smad2/3 and Smad1/5/8) form heteromeric complexes with Co-Smads (Smad4), which are transported to the nucleus to control specific gene expression. TAK1 mediates the TGF-β signal downstream via MKK3 and MKK6 to activate the p38 MAPK, and through MKK4 to activate SAPK/JNK, respectively. In addition, TGF-β can also activate ERK1/2. For further details see the text.
Besides the archetypal Smad signaling, TGF-β can also activate each of the three major MAPK cascades, the ERK1/2, SAPK/JNK, and p38 pathways (Fig. 11), independently of Smad activation (2, 26). The activation of JNK and p38 is initiated by an upstream MAPKKK, termed TAK1 (for TGF-β activated kinase 1) (39). TAK1, together with the upstream regulator TAB1 (33), mediates the TGF-β signal via MKK3 and MKK6 to activate the p38 MAPK (30), whereas the SAPK/JNK pathway is activated through MKK4 (21). TGF-β-mediated activation of ERK1/2 may be transduced through activation of Ras and MEK1/2 (5, 26). In addition, activated MEK1/2 and ERK1/2 can phosphorylate various R-Smads (5, 22).
To explore the hypothesis that TGF-β and metabolic acidosis may share a common mechanism, the effects of porcine TGF-β1 on the activation of various signaling pathways and the expression of GA mRNA in LLC-PK1-FBPase+ cells were delineated. The resulting data established that TGF-β1 enhanced phosphorylation of multiple proteins in p38 MAPK and ERK1/2 signaling cascades. In a number of cell lines, including HepG2, Chinese hamster ovary (CHO), and Madin-Darby canine kidney (MDCK) cells, TGF-β1 also activates the SAPK/JNK (12). However, in LLC-PK1-FBPase+ cells no activation of SAPK/JNK was observed following treatment with TGF-β1 (Fig. 4). The latter finding is consistent with our unpublished observation that treatment of LLC-PK1-FBPase+ cells with TGF-β1 produced no detectable signs of epithelial-mesenchymal transition (36) or apoptosis (4).
It is well established that TGF-β-induced activation of the ERK pathway can result in Smad phosphorylation and thus enhance Smad activation (5, 41). A rapid and pronounced activation of Smad1/5/8 was also observed in LLC-PK1-FBPase+ cells (Figs. 5A), while no phosphorylation of Smad2/3 was detectable (Fig. 5C). The former response was U0126 and PD098059 sensitive (Fig. 6). Thus, as shown previously for rat intestinal epithelial cells (41), TGF-β can signal through MEK1/2 to activate Smad1/5/8. In contrast, p38 MAPK does not activate Smad1/5/8, since TGF-β-induced Smad1/5/8 phosphorylation is insensitive to the p38 inhibitor SB203580 (6) (Fig. 5B). The significance and biological relevance of ERK1/2 and Smad1/5/8 activation in LLC-PK1-FBPase+ cells by TGF-β is unknown at present. Due to the lack of specific, discriminating antibodies, however, the identification of each of the Smad isoforms and their contribution in mediating the TGF-β signal cannot yet be accomplished. One alternative strategy would be the selective knock-down of Smad1, Smad5, and/or Smad8 by a small interfering RNA approach.
In a very recent study (27) it was shown that TGFβ-stimulated fibronectin secretion in a human proximal tubular epithelial cell line is solely mediated via the p38 MAPK pathway, independently of Smad activation. This is in line with the data shown in the present study. TGF-β activated p38 MAP kinase (Figs. 1 and 2) and stimulated fibronectin synthesis in LLC-PK1-FBPase+ cells (Fig. 10). The activation of Smad1/5/8 (Fig. 5), however, was insensitive to p38 inhibition. In contrast, in cultured rat mesangial cells, TGF-β-induced fibronectin expression was PD098059 sensitive, indicative of a role in MEK1/2-ERK1/2 signaling (16). There is still a controversy surrounding the effects of TGF-β in cultured renal cells and the signaling cascades involved. One explanation is the state of confluency of cultures under study (23).
TGF-β induces the expression of a whole array of target genes (18, 40, 43). Gene expression profiling by cDNA microarray analysis revealed a broad set of immediate-early TGF-β target genes, implicated in such diverse cellular actions as growth arrest, epithelial-to-mesenchymal transdifferentiation, extracellular matrix expansion, and apoptosis (1). In the present study, we established that stimulation of LLC-PK1-FBPase+ cells with TGF-β1 results in a gradual, but stable increase in GA mRNA levels (Fig. 7). By contrast, TGF-β1 produced only a slight but transient increase in PEPCK mRNA (data not shown). Treatment of LLC-PK1-FBPase+ cells with acidic medium or TGF-β1 activates p38 MAPK and the downstream transcription factor ATF-2. Binding of the phosphorylated ATF-2 to the CRE-1 element within the PEPCK promoter following onset of metabolic acidosis is likely to activate transcription of the PCK1 gene (6). Thus the transient increase in PEPCK mRNA levels that occurs when LLC-PK1-FBPase+ cells are stimulated with TGF-β1 may be due to the rapid and pronounced activation of p38 MAPK. However, a signaling pathway that is activated by TGF-β1, but not during metabolic acidosis, must produce a subsequent repression of PCK1 transcription or enhance the turnover of the PEPCK mRNA.
The pH-responsive increase in GA mRNA is mediated by binding of ζ-crystallin/NADPH:quinone reductase (35) to an eight-base AU sequence in the 3′-UTR that functions as a pH-response element (19). This adaptation occurs following a 8- to 12-h lag (15) and is not inhibited by SB203580 (6). Therefore, this response may require activation of a kinase that is upstream of p38 MAPK or, alternatively, it may be mediated by a different signaling pathway. Furthermore, when LLC-PK1-FBPase+ cells were treated with acidic medium (pH 6.9) and stimulated with TGF-β1, the observed effects on GA mRNA levels were additive. In contrast, acid-stimulated rates of ammonia production decreased in the presence of TGF-β1 (Fig. 8). This may be explained by a shift in glutamine metabolism to enhanced extracellular matrix synthesis (Fig. 10) at the expense of glutamine breakdown. Therefore, the observed effects of metabolic acidosis differ significantly from those mediated by TGF-β. This conclusion is also supported by the data obtained with the luciferase reporter construct that includes the 3′-UTR of the GA mRNA (Fig. 9). This construct exhibits a pronounced activation in LLC-PK1-FBPase+ cells that were treated with acidic medium. However, the activity of this construct was unaffected by treatment with TGF-β1. Therefore, the TGF-β1-mediated increase in GA mRNA is not likely to utilize the pH-response elements. The luciferase construct also contains the proximal region of the KGA promoter. Thus TGF-β1 stimulation of KGA gene expression may require elements that are upstream or downstream of this segment. Further studies are necessary to determine the molecular mechanism by which TGF-β1 causes a pronounced and sustained increase in GA mRNA levels.
In conclusion, TGF-β1 activates multiple signaling pathways in LLC-PK1-FBPase+ cells and enhances expression of GA mRNA, which is accompanied by increased cellular synthesis and deposition of fibronectin (Fig. 10). The latter is a well-known effect of TGF-β on renal glomerular and tubular cells in vivo and in vitro (13, 44, 45). However, the significant differences observed when the temporal effects of treatment with TGF-β or acidic medium and the response of the luciferase reporter construct are compared suggest that enhanced expression of the KGA gene during metabolic acidosis is not mediated by the autocrine factor, TGF-β.
GRANTS
This work was supported by Austrian Science Fund Grant P14981 (to G. Gstraunthaler) and by National Institute of Diabetes and Digestive and Kidney Diseases Grants DK-37124 and DK-43704 (to N. P. Curthoys).
FOOTNOTES
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
Parts of this study were presented at the Annual Meetings of the American Society of Nephrology and have appeared in abstract form: Philadelphia, PA, 2002 (J Am Soc Nephrol 13: 492A, 2002); San Diego, CA, 2003 (J Am Soc Nephrol 14: 73A, 2003); and St. Louis, MO, 2004 (J Am Soc Nephrol 15: 92A, 2004).
REFERENCES
- 1 Bottinger EP, Bitzer M. TGF-β signaling in renal disease. J Am Soc Nephrol 13: 2600–2610, 2002.
Crossref | PubMed | ISI | Google Scholar - 2 Choi ME. Mechanism of transforming growth factor-β1 signaling. Kidney Int Suppl 77: S53–S58, 2000.
PubMed | Google Scholar - 3 Curthoys NP, Gstraunthaler G. Mechanism of increased renal gene expression during metabolic acidosis. Am J Physiol Renal Physiol 281: F381–F390, 2001.
Link | ISI | Google Scholar - 4 Dai C, Yang J, Liu Y. Transforming growth factor-beta1 potentiates renal tubular epithelial cell death by a mechanism independent of Smad signaling. J Biol Chem 278: 12537–12545, 2003.
Crossref | PubMed | ISI | Google Scholar - 5 Derynck R, Zhang YE. Smad-dependent and Smad-independent pathways in TGF-β family signalling. Nature 425: 577–584, 2003.
Crossref | PubMed | ISI | Google Scholar - 6 Feifel E, Obexer P, Andratsch M, Euler S, Taylor L, Tang A, Wei Y, Schramek H, Curthoys NP, Gstraunthaler G. p38 MAPK mediates acid-induced transcription of PEPCK in LLC-PK1-FBPase+ cells. Am J Physiol Renal Physiol 283: F678–F688, 2002.
Link | ISI | Google Scholar - 7 Fine LG, Norman J. Cellular events in renal hypertrophy. Annu Rev Physiol 51: 19–32, 1989.
Crossref | PubMed | ISI | Google Scholar - 8 Fujita H, Omori S, Ishikura K, Hida M, Awazu M. ERK and p38 mediate high-glucose-induced hypertrophy and TGF-β expression in renal tubular cells. Am J Physiol Renal Physiol 286: F120–F126, 2004.
Link | ISI | Google Scholar - 9 Gstraunthaler G, Handler JS. Isolation, growth, and characterization of a gluconeogenic strain of renal cells. Am J Physiol Cell Physiol 252: C232–C238, 1987.
Abstract | ISI | Google Scholar - 10 Gstraunthaler G, Holcomb T, Feifel E, Liu W, Spitaler N, Curthoys NP. Differential expression and acid-base regulation of glutaminase mRNAs in gluconeogenic LLC-PK1-FBPase+ cells. Am J Physiol Renal Physiol 278: F227–F237, 2000.
Link | ISI | Google Scholar - 11 Halperin ML. Metabolic aspects of metabolic acidosis. Clin Invest Med 16: 294–305, 1993.
PubMed | ISI | Google Scholar - 12 Hanafusa H, Ninomiya-Tsuji J, Masuyama N, Nishita M, Fujisawa J, Shibuya H, Matsumoto K, Nishida E. Involvement of the p38 mitogen-activated protein kinase pathway in transforming growth factor-β-induced gene expression. J Biol Chem 274: 27161–27167, 1999.
Crossref | PubMed | ISI | Google Scholar - 13 Haneda M, Koya D, Isono M, Kikkawa R. Overview of glucose signaling in mesangial cells in diabetic nephropathy. J Am Soc Nephrol 14: 1374–1382, 2003.
Crossref | PubMed | ISI | Google Scholar - 14 Holcomb T, Curthoys NP, Gstraunthaler G. Subcellular localization of PEPCK and metabolism of gluconeogenic substrains of renal cell lines. Am J Physiol Cell Physiol 268: C449–C457, 1995.
Link | ISI | Google Scholar - 15 Hwang JJ, Curthoys NP. Effect of acute alterations in acid-base balance on rat renal glutaminase and phosphoenolpyruvate carboxykinase gene expression. J Biol Chem 266: 9392–9396, 1991.
PubMed | ISI | Google Scholar - 16 Inoki K, Haneda M, Ishida T, Mori H, Maeda S, Koya D, Sugimoto T, Kikkawa R. Role of mitogen-activated protein kinases as downstream effectors of transforming growth factor-β in mesangial cells. Kidney Int 58, Suppl 77: S-76–S-80, 2000.
Google Scholar - 17 Karlsson G, Liu Y, Larsson J, Goumans MJ, Lee JS, Thorgeirsson SS, Ringner M, Karlsson S. Gene expression profiling demonstrates that TGF-β1 signals exclusively through receptor complexes involving Alk5 and identifies targets of TGF-β signaling. Physiol Genomics 21: 396–403, 2005.
Link | ISI | Google Scholar - 18 Kretzschmar M, Doody J, Massague J. Opposing BMP and EGF signalling pathways converge on the TGF-β family mediator Smad1. Nature 389: 618–622, 1997.
Crossref | PubMed | ISI | Google Scholar - 19 Laterza OF, Curthoys NP. Specificity and functional analysis of the pH-responsive element within renal glutaminase mRNA. Am J Physiol Renal Physiol 278: F970–F977, 2000.
Link | ISI | Google Scholar - 20 Laterza OF, Hansen WR, Taylor L, Curthoys NP. Identification of an mRNA-binding protein and the specific elements that may mediate the pH-responsive induction of renal glutaminase mRNA. J Biol Chem 272: 22481–22488, 1997.
Crossref | PubMed | ISI | Google Scholar - 21 Lutz M, Knaus P. Integration of the TGF-β pathway into the cellular signalling network. Cell Signal 14: 977–988, 2002.
Crossref | PubMed | ISI | Google Scholar - 22 Massague J. How cells read TGF-β signals. Nat Rev Mol Cell Biol 1: 169–178, 2000.
Crossref | PubMed | ISI | Google Scholar - 23 Masszi A, Fan L, Rosivall L, McCulloch CA, Rotstein OD, Mucsi I, Kapus A. Integrity of cell-cell contacts is a critical regulator of TGF-β1-induced epithelial-to-myofibroblast transition. Am J Pathol 165: 1955–1967, 2004.
Crossref | PubMed | ISI | Google Scholar - 24 Miyazawa K, Shinozaki M, Hara T, Furuya T, Miyazono K. Two major Smad pathways in TGF-β superfamily signalling. Genes Cells 7: 1191–1204, 2002.
Crossref | PubMed | ISI | Google Scholar - 25 Moustakas A, Souchelnytskyi S, Heldin CH. Smad regulation in TGF-β signal transduction. J Cell Sci 114: 4359–4369, 2001.
PubMed | ISI | Google Scholar - 26 Mulder KM. Role of Ras and MAPKs in TGF-β signaling. Cytokine Growth Factor Rev 11: 23–35, 2000.
Crossref | PubMed | ISI | Google Scholar - 27 Niculescu-Duvaz I, Panish MK, Colville-Nash P, Dockrell MEC. The TGFβ1-induced fibronectin in human renal proximal tubular epithelial cells is p38 MAP kinase dependent but Smad independent. Nephron Exp Nephrol 105: e108–e116, 2007.
Crossref | PubMed | Google Scholar - 28 O'Hayre M, Taylor L, Andratsch M, Feifel E, Gstraunthaler G, Curthoys NP. Effects of constitutively active and dominant negative MKK3 and MKK6 on the pH-responsive increase in phosphoenolpyruvate carboxykinase mRNA. J Biol Chem 281: 2982–2988, 2006.
Crossref | PubMed | ISI | Google Scholar - 29 Preisig P. What makes cells grow larger and how do they do it? Renal hypertrophy revisited Exp Nephrol 7: 273–283, 1999.
Google Scholar - 30 Sano Y, Harada J, Tashiro S, Gotoh-Mandeville R, Maekawa T, Ishii S. ATF-2 is a common nuclear target of Smad and TAK1 pathways in transforming growth factor-β signaling. J Biol Chem 274: 8949–8957, 1999.
Crossref | PubMed | ISI | Google Scholar - 31 Schroeder JM, Ibrahim H, Taylor L, Curthoys NP. Role of deadenylation and AUF1 binding in the pH-responsive stabilization of glutaminase mRNA. Am J Physiol Renal Physiol 290: F733–F740, 2006.
PubMed | ISI | Google Scholar - 32 Shi Y, Massague J. Mechanisms of TGF-β signaling from cell membrane to the nucleus. Cell 113: 685–700, 2003.
Crossref | PubMed | ISI | Google Scholar - 33 Shibuya H, Yamaguchi K, Shirakabe K, Tonegawa A, Gotoh Y, Ueno N, Irie K, Nishida E, Matsumoto K. TAB1: an activator of the TAK1 MAPKKK in TGF-β signal transduction. Science 272: 1179–1182, 1996.
Crossref | PubMed | ISI | Google Scholar - 34 Stumvoll M, Perriello G, Meyer C, Gerich J. Role of glutamine in human carbohydrate metabolism in kidney and other tissues. Kidney Int 55: 778–792, 1999.
Crossref | PubMed | ISI | Google Scholar - 35 Tang A, Curthoys NP. Identification of zeta-crystallin/NADPH:quinone reductase as a renal glutaminase mRNA pH response element-binding protein. J Biol Chem 276: 21375–21380, 2001.
Crossref | PubMed | ISI | Google Scholar - 36 Tian YC, Fraser D, Attisano L, Phillips AO. TGF-β1-mediated alterations of renal proximal tubular epithelial cell phenotype. Am J Physiol Renal Physiol 285: F130–F142, 2003.
Link | ISI | Google Scholar - 37 Welbourne T, Routh R, Yudkoff M, Nissim I. The glutamine/glutamate couplet and cellular function. News Physiol Sci 16: 157–160, 2001.
Abstract | Google Scholar - 38 Wolf G, Ziyadeh FN. Molecular mechanisms of diabetic renal hypertrophy. Kidney Int 56: 393–405, 1999.
Crossref | PubMed | ISI | Google Scholar - 39 Yamaguchi K, Shirakabe K, Shibuya H, Irie K, Oishi I, Ueno N, Taniguchi T, Nishida E, Matsumoto K. Identification of a member of the MAPKKK family as a potential mediator of TGF-β signal transduction. Science 270: 2008–2011, 1995.
Crossref | PubMed | ISI | Google Scholar - 40 Yang YC, Piek E, Zavadil J, Liang D, Xie D, Heyer J, Pavlidis P, Kucherlapati R, Roberts AB, Bottinger EP. Hierarchical model of gene regulation by transforming growth factor β. Proc Natl Acad Sci USA 100: 10269–10274, 2003.
Crossref | PubMed | ISI | Google Scholar - 41 Yue J, Frey RS, Mulder KM. Cross-talk between the Smad1 and Ras/MEK signaling pathways for TGFβ. Oncogene 18: 2033–2037, 1999.
Crossref | PubMed | ISI | Google Scholar - 42 Yue J, Mulder KM. Transforming growth factor-β signal transduction in epithelial cells. Pharmacol Ther 91: 1–34, 2001.
Crossref | PubMed | ISI | Google Scholar - 43 Zavadil J, Bitzer M, Liang D, Yang YC, Massimi A, Kneitz S, Piek E, Bottinger EP. Genetic programs of epithelial cell plasticity directed by transforming growth factor-β. Proc Natl Acad Sci USA 98: 6686–6691, 2001.
Crossref | PubMed | ISI | Google Scholar - 44 Ziyadeh FN. Mediators of diabetic renal disease: the case for TGF-β as the major mediator. J Am Soc Nephrol 15, Suppl 1: S55–S57, 2004.
Crossref | PubMed | ISI | Google Scholar - 45 Ziyadeh FN, Sharma K. Overview: combating diabetic nephropathy. J Am Soc Nephrol 14: 1355–1357, 2003.
Crossref | PubMed | ISI | Google Scholar
