Involvement of Pax-2 in the Action of Activin A on Tubular Cell Regeneration
Akito Maeshima*,,
Kyoko Maeshima*,
Yoshihisa Nojima and
Itaru Kojima*
*Institute for Molecular and Cellular Regulation, Gunma University, Maebashi, Japan; Third Department of Internal Medicine, Gunma University School of Medicine, Maebashi, Japan.
Correspondence to Dr. Itaru Kojima, Institute for Molecular & Cellular Regulation, Gunma University, Maebashi 371-8512, Japan. Phone: 81-27-220-8835; Fax: 81-27-220-8893;E-mail: ikojima{at}showa.gunma-u.ac.jp
ABSTRACT. It has been recently shown that in ischemic rat kidneysactivin A is induced in tubular cells and inhibits their regeneration.The present study was conducted to further investigate the actionof activin A in tubular cells during regeneration. Among genesthought to be critical for kidney development, Pax-2 was upregulatedin tubular cells during regeneration after renal ischemia. Pax-2protein was localized in nuclei of tubular and interstitialcells, some of which co-expressed a mesenchymal cell marker,vimentin, suggesting that a population of Pax-2–positivecells have properties of immature progenitor-like tubular cells.The Pax-2–expressing cells co-expressed a cell proliferationmarker, BrdU, activin A, and the type II activin receptor. ActivinA modulated growth of BrdU/Pax-2 double-positive cells sincean administration of follistatin increased; conversely, exogenousactivin A decreased the number of BrdU/Pax-2 double-positivecells after renal ischemia. Activin A also reduced the expressionof Pax-2 in cultured metanephroi. A proximal tubular cell line,LLC-PK1 cells, was used to further study the mode of actionof activin A. The expression of Pax-2 was not detected in quiescentLLC-PK1 cells, but it was markedly increased when growth wasstimulated. Under this condition, activin A significantly inhibitedDNA synthesis and reduced the expression of Pax-2 in LLC-PK1cells. In contrast, blockade of the activin signaling by overexpressingdominantly negative mutant receptor enhanced the expressionlevel of Pax-2 in LLC-PK1 cells and induced an immature phenotype.These results suggest that activin A regulates tubular cellgrowth and differentiation by modulating the expression of Pax-2during regeneration.
In the kidney, regeneration, reconstruction, and maturationof tubular cells after renal injury have many parallels to thegrowth and differentiation that take place during kidney organogenesis(1). The adult tubular epithelium has a potential for regenerationafter damage. During acute tubular necrosis induced by renalischemia or renal toxins, normally quiescent cells undergo dedifferentiationand reobtain their potential to divide after enhancing theirDNA synthesis. After proliferation, the new cells then differentiateto restore the functional integrity of the nephron (2). Severalgrowth factors, including hepatocyte growth factor (HGF), epidermalgrowth factor (EGF), insulin-like growth factor-I (IGF-I), andbone morphogenetic protein-7 (BMP-7) have been shown to be involvedin tubular regeneration of the kidney (3). These factors arepotent regulators of kidney organogenesis (4,5). It has beenreported that administration of these growth factors promotestubular regeneration after a variety of insults (3). This suggeststhat regeneration processes may be at least partially controlledby the similar mechanism operating during development. However,little is known about the mechanism by which these factors modulatetubular regeneration, although it is considered that these factorsplay important roles in regeneration processes of the kidneyas mitogen, motogen, and morphogen (3).
Activin A, a member of the TGF- superfamily, modulates cellgrowth and differentiation in many types of cells (6,7). Activinsare dimeric proteins, and subunits of activin are expressedin various organs (8). An important modulator of activin isfollistatin (9). This protein specifically binds to activinsand related ligands with high affinity and blocks their actions(10,11). Follistatin is synthesized in the target cells of activinsand remains in the extracellular matrix. Furthermore, the productionof follistatin is regulated by activins. Hence, activin andfollistatin modulate cellular function in a complex manner.Activin A and follistatin are expressed in a developing kidney(12). Activin A disrupts ureteric bud branching in an embryonickidney in organ culture (13) and also inhibits branching tubulogenesisin an in vitro tubulogenesis model (14). The number of glomeruliis increased in the kidney of transgenic mice expressing dominantlynegative activin receptors (15). Collectively, activin A isa negative regulator of tubulogenesis during kidney development(16). We recently demonstrated the involvement of the activin-follistatinsystem in tubular regeneration after renal ischemia (17). ActivinA, which was not detected in normal kidney, was upregulatedin tubular cells after renal ischemia. Exogenous follistatinaccelerated renal regeneration by enhancing DNA synthesis andpreventing apoptosis in tubular cells. Presumably, exogenousfollistatin enhanced tubular regeneration by blocking the actionof endogenous activin A. However, the mechanism by which endogenousactivin A inhibits tubular regeneration still remains unknown.Considering that regeneration processes may recapitulate developmentalparadigms to restore organ or tissue function (1), it is quitepossible that activin modulates the expression of a set of developmentalgenes during tubular regeneration.
In the present study, we examined whether or not activin A modulatedthe expression of developmental genes during tubular regeneration.The results indicate that Pax-2, a transcription factor criticalfor kidney development (18), was re-expressed in tubular cellsafter renal ischemia. Exogenous follistatin increased; conversely,exogenous activin A decreased the number of BrdU/Pax-2 double-positivecells after renal ischemia in vivo. Activin A reduced the expressionof Pax-2 in embryonic kidney as well as in proximal tubularcell line in vitro. Furthermore, blockade of activin signalingenhanced the expression of Pax-2 and induced an immature phenotypein tubular cells. These results suggest that activin A regulatestubular regeneration by modulating the expression of Pax-2.
Materials
Recombinant human activin A, follistatin, and polyclonal rabbitanti-human activin A antibody were provided by Dr. Y. Eto ofthe central Research Laboratory, Ajinomoto Inc. (Kawasaki, Japan).Polyclonal antibody against Pax-2 was purchased from BerkeleyAntibody Company (Berkeley, CA), and polyclonal anti-activintype II receptor antibody was a generous gift from Dr. K. Miyazono(University of Tokyo, Japan). Polyclonal anti-E-cadherin antibodywas purchased from Santa Cruz Biotechnology, Inc. (Santa Cruz,CA). Polyclonal anti-vimentin antibody was obtained from NEOMARKERS (Fremont, CA). Cy3 (indocarbocyanine)-conjugated goatanti-rabbit IgG and FITC (fluorescence indocyanine)–conjugateddonkey anti-mouse IgG were purchased from Jackson ImmunoResearchLaboratories (West Grove, PA). [3H]Thymidine was from Dupont-NewEngland Nuclear (Boston, MA).
Experimental Protocol
Male Wistar rats weighing 200 to 300 g were obtained from theImai Animal Company (Saitama, Japan). Ischemia/reperfusion injurywas performed as described previously (17). Briefly, under anesthesiawith pentobarbital sodium (30 mg/kg body wt), renal ischemiawas induced by clamping both renal arteries for 45 min usinga nontraumatic vascular clamp. Core body temperature was maintained37°C by placing the animal on a homoiothermic table andwas monitored with a temperature-sensing rectal probe. Afterremoval of the clamp to allow reperfusion for the indicatedperiods, rats were sacrificed. Then, the kidneys were removedand fresh frozen for RNA extraction and histologic analyses.Sham-operations were performed in a similar manner, except forclamping the renal arteries. To analyze the efficacy of exogenousfollistatin or activin A after ischemic renal injury, the indicateddose of recombinant human (rh) follistatin or activin A dissolvedin 0.5 ml of physiologic saline was administrated via the tailvein at the time of reperfusion. Control animals received thesame volume of saline alone.
Reverse Transcription–PCR
Total RNA was isolated with the TRlzol Reagent (Life Technologies/BRL,Grand Island, NY) from whole kidneys or cultured metanephroi.First-strand cDNA was made from total RNA using a SuperscriptPreamplification System (Life Technologies/BRL) as describedpreviously (17) according to the manufacturer’s instructions.Contaminated genomic DNA was removed with RNase-free deoxyribonuclease(DNase). Five micrograms of DNase-treated RNA were incubatedwith 1 µl of oligodT at 70°C for 10 min. Two microliters10x PCR buffer, 1 µl of DTT (0.1 M), 2 µl of dNTPmix (10 mM), and 2 µl of MgCl2 (25 mM) were added to eachreaction. After incubation for 5 min at 42°C, 1 µlof reverse transcriptase was added. Samples were incubated at42°C for 50 min, then at 70°C for 15 min. RNase H (1µl) was added to each reaction, and samples were incubatedat 37°C for 20 min. PCR was performed as indicated by themanufacturer (Perkin-Elmer) with the primers shown in Table 1.Reactions contained 5 µl of a 10x PCR buffer, 2 µlof MgCl2 (50 mM), 1 µl of dNTP mix, 1 µl of 3'-primer,1 µl of 5'primer, 0.5 µl of Taq polymerase, and1 µl of cDNA. Samples were incubated at 95°C for 5min, followed by the indicated cycles of 30 s at 94°C, 30s at 58°C, 90 s at 72°C, and final extension at 72°Cfor 10 min in a Perkin-Elmer DNA Thermal Cycler. There were30 cycles of PCR for the Pax-2, WT-1, Wnt-4, Pax-8, BF-2, and18 cycles for GAPDH. Reactions without cDNA were used as a negativecontrol. Rat embryonic kidney (embryonic day 14) cDNA was usedas a positive control in each experiment. Reactions were repeatedat least twice.
Table 1. Sequences of PCR primers used in this study
Immunohistochemical Analyses
Kidneys were removed and embedded in a Tissue-Tek OCT compound(Miles, Inc., Elkhart, IN) and frozen in liquid nitrogen. Frozensections (4 µm) were cut with a Jung CM 3000 cryostat(Leica, Wien, Austria), mounted on poly-L-lysine–coatedslides, and fixed in 4% paraformaldehyde for 15 min at roomtemperature. Sections were then washed in PBS, pretreated with5% normal goat serum-PBS for 1 h, and covered with a primaryantibody at room temperature for 1 h. After washing in PBS,the sections were covered with a mixture of a Cy3-labeled goatanti-rabbit lgG antibody or FITC-labeled rabbit anti-mouse lgGantibody and 4'-diamidino-2-phenylindole (DAPI: Boehringer Mannheim).Immunofluorescence images were recorded with an Olympus AX70epifluorescence microscope (Olympus, Tokyo, Japan) equippedwith a PXL 1400 cooled-CCD camera system (Photometrics, Tucson,AR), which was operated with IP Lab Spectrum software (SignalAnalysis, Vienna, VA). For immunohistochemical controls, theprimary antibody was replaced with 5% normal goat serum-PBS,which did not show positive staining, confirming specificity.
In separate experiments, kidneys were removed and fixed with4% formaldehyde. An avidin-biotin coupling (ABC) immunoperoxidasetechnique using a Vectastain Elite ABC kit (Vector Laboratories,Burlingame, CA) was performed according to the manufacturer’sinstruction as described previously (17).
Bromodeoxyuridine Labeling
DNA synthesis in renal tubular cells was measured using bromodeoxyuridine(BrdU). At the indicated times after reperfusion, BrdU (100mg/kg), an analogue of thymidine, was injected intraperitoneallyinto rats. After 1 h, rats were sacrificed and the kidneys wereremoved and embedded in a Tissue-Tek OCT compound (Miles, Inc.,Elkhart, IN) and frozen in liquid nitrogen. The frozen sectionswere immunostained using a mouse anti-BrdU antibody (Amersham)as described above.
Quantification of BrdU, Pax-2, and BrdU/Pax-2 Double-Positive Cells
Quantification of BrdU-positive, Pax-2-positive, and BrdU/Pax-2double-positive cells in the kidney after renal ischemia wasperformed by counting the positive nuclei in tubular cells fromfive randomly selected fields of the outer medulla with epifluorescencemicroscope at x400 magnification. The results were expressedas a percentage of total tubular cells in five sections perrat kidney. The average of the five determinations was calculated.
Cell Culture
LLC-PK1 cells obtained from American Type Culture Collection(ATCC RL-1392) were cultured in complete medium consisting ofMedium 199 (ICN Biomedicals, Inc.) with 5% fetal bovine serum(FBS; Life Technologies/BRL), penicillin, and streptomycin inan atmosphere of 5% CO2 and 95% air at 37°C. The mediumwas changed every 3 to 4 d. To obtain quiescent cells, cellswere incubated in a serum-free medium for 48 h. LLC-PK1 cellsexpressing truncated activin type II receptor cDNA (LLC-PK1-tARII)and LLC-PK1 cells expressing PCXN2 plasmid vector (LLC-PK1-mock)were generated and cultured as described previously (19).
Measurement of DNA Synthesis
DNA synthesis was assessed by measuring [3H]thymidine incorporationinto TCA precipitable materials. Serum-starved cells culturedin a 24-well plate were incubated in complete medium containing5% FBS with or without 10 nM activin A for indicated times.Then, the cells were pulse-labeled with 1 µCi/ml [3H]thymidine for an additional 4 h. [3H]Thymidine incorporationwas measured as described previously (14).
Immunocytochemical Analyses
Cells grown on coverslips were washed, fixed with 4% paraformaldehyde,permeabilized with PBS containing 0.1% (vol/vol) Triton X-100,and incubated sequentially with 3% bovine serum albumin (BSA)in PBS. Cells were then incubated with a primary antibody atroom temperature for 1 h. After washing in PBS, cells were coveredwith a mixture of a Cy3-labeled goat anti-rabbit IgG antibodyand DAPI. Immunofluorescence images were recorded as describedabove.
Western Blot Analyses
Cells were washed three times with PBS, suspended in Laemmlibuffer, and heated to 100°C for 10 min. After centrifugation,supernatant was collected and the protein concentration wasdetermined by using a protein assay kit (Bio-Rad Laboratories).Twenty micrograms of protein from each sample were separatedby SDS-PAGE under reducing condition and transferred to a PVDFmembrane (Nihon Millipore Ltd., Yonezawa, Japan) by electroblotting.To reduce nonspecific antibody binding, the membrane was blockedwith 5% BSA and 0.1% NaN3 dissolved in Tris-saline (TS) for1 h at 37°C and then incubated overnight with a primaryantibody and washed with Tris-PBS (PBST). After incubation withperoxidase-labeled secondary antibody for 1 h at room temperature,the membrane was washed with PBST and analyzed by exposure tox-ray film using ECL Western blotting detection reagent (AmershamLife Science).
Organ Culture of Metanephros
Embryos were removed from anesthetized pregnant Wistar rats(Nihon SLC, Inc., Hamamatsu, Japan) on day 14 of the pregnancy.Metanephric rudiments were surgically removed from embryos.Metanephroi were explanted onto Transwell-clear (pore size,0.4 µm; Corning Incorporated, Corning, NY) at the interfacebetween air/5% CO2 atmosphere and medium and cultured at 37°Cin DMEM containing 5% FBS and antibiotics.
Statistical Analyses
The significance of differences between the means was comparedby t test; P values < 0.05 considered significant.
Changes in the mRNA Expression of Developmental Genes in the Kidney after Renal Ischemia
To identify the possible regulator of tubular regeneration,we analyzed the expression of developmental genes critical forkidney organogenesis (4), including Pax-2 (18), Wnt-4 (20,21),Pax-8 (22,23), WT-1 (24), and BF-2 (25), in the kidney afterrenal ischemia by RT-PCR. As shown in Figure 1, the expressionof Pax-2 was not detected in normal and sham-operated kidney.In contrast, a strong induction of Pax-2 expression was observedin the kidney after renal ischemia. The expression of Wnt-4was undetectable in either normal or ischemic kidney. The expressionof WT-1, Pax-8, and BF-2 was detected in normal and sham-operatedkidney, but the expression levels of these factors were notaltered after renal ischemia.
Figure 1. Time course of the expression of developmental genes in the kidney after renal ischemia. Whole kidney RNA was isolated at the indicated time points after ischemic renal injury and the expression of mRNA for Pax-2, Wnt-4, WT-1, Pax-8, and BF-2 was analyzed by reverse transcriptase-PCR (RT-PCR). Reactions without cDNA were used as negative controls (N). The reverse transcription product obtained from rat embryonic kidney was used as a positive control (P). M, molecular marker. Product sizes are indicated on the right. Representative results of three separate experiments are shown.
Colocalization of the Pax-2 Protein and Vimentin in the Kidney after Renal Ischemia
We next examined the localization of Pax-2 protein in the kidneyafter renal ischemia by immunohistochemistry. In normal kidney,Pax-2 protein was not detected in either tubular cells or glomeruliin the cortex (data not shown). In contrast, Pax-2 protein wasdetected in the nuclei of tubular cells (Figure 2A) and interstitialcells (Figure 2D) in the outer medulla after ischemic injury,where proliferating tubular cells are mainly localized. To characterizethe Pax-2–positive cells, we examined the expression ofvimentin, a mesenchymal cell marker, in the kidney after renalischemia. In normal kidney, vimentin was expressed in mesangialcells in glomeruli or interstitial fibroblasts, which are derivedfrom mesenchymal cells in the metanephros but not in tubularepithelial cells (data not shown). In contrast, vimentin wasobserved in the interstitial Pax-2–positive cells (Figure 2E)but was not in tubular cells (Figure 2B) in the outer medullaof ischemic kidney.
Figure 2. Colocalization of Pax-2 and vimentin in the kidney after renal ischemia. Localization of Pax-2 and vimentin in the tubular (A through C) and interstitial (D through F) areas of outer medulla in the kidney after renal ischemia was examined by indirect fluorescence immunostaining. Frozen sections from ischemic kidney 24 h after reperfusion were used for experiments. A and D: Pax-2 protein (red), nuclei (blue). B and E: vimentin (green), nuclei (blue). C and F: Nomarski images. Magnification, x1000.
Changes in the Number of BrdU-Positive Cells and Pax-2–Positive Cells in the Kidney after Renal Ischemia
To examine the relationship between Pax-2 expression and tubularcell growth, we first analyzed the change in the number of BrdU-positivecells in the kidney after renal ischemia. As shown in Figure 3A,BrdU-positive cells were slightly detected in ischemic kidneysat 18 h after reperfusion. The number of BrdU-positive cellspeaked maximum at 24 h after reperfusion and decreased thereafter.On the other hand, Pax-2–positive cells were significantlyobserved in ischemic kidneys at 12 h after reperfusion. Thenumber of Pax-2–positive cells peaked at 18 h after reperfusionand decreased thereafter, indicating that Pax-2 expression precedestubular cell growth in the kidney after renal ischemia.
Figure 3. Quantitative analysis of BrdU-positive cells and Pax-2–positive cells in the kidney after renal ischemia. The number of Pax-2–positive cells and BrdU-positive cells were analyzed in kidney sections obtained from ischemic kidney at the indicated times after reperfusion as described in Materials and Methods. Time courses of the number of BrdU-positive cells (A) and Pax-2–positive cells (B) in the kidney after renal ischemia are presented. Results are the mean ± SEM of three independent experiments.
Changes in the Number of BrdU/Pax-2 Double-Positive Cells in the Kidney after Renal Ischemia
To further examine whether Pax-2–positive cells were growing,we analyzed the localization of BrdU and Pax-2 in the kidneyafter renal ischemia. Indirect immunofluorescence staining demonstratedthat Pax-2–positive tubular cells (Figure 4A) were overlappedwith BrdU-positive cells (Figure 4B), suggesting that Pax-2–positivecells are actively engaged in cell proliferation (Figure 4D).Quantitative analysis showed that the number of BrdU/Pax-2 double-positivecells peaked at 24 h in ischemic kidneys after reperfusion (Figure 4E).
Figure 4. Quantitative analysis of BrdU/Pax-2 double-positive cells in the kidney after renal ischemia. Localization of BrdU and Pax-2 in the outer medulla of the kidney after renal ischemia was examined by indirect fluorescence immunostaining. Frozen sections from ischemic kidney 24 h after reperfusion were used for experiments. (A) Pax-2 protein (red). (B) BrdU (green). (C) Nuclei (blue). (D) merge. Magnification, x400. (E) Time course of the number of BrdU/Pax-2 double-positive cells in the kidney after renal ischemia. The number of BrdU/Pax-2 double-positive cells was analyzed in kidney sections obtained from ischemic kidney at the indicated times after reperfusion as described in Materials and Methods. Results are presented as the mean ± SEM of three independent experiments.
Co-localization of Pax-2, Activin A, and the Type II Activin Receptor in the Tubular Cells after Renal Ischemia
We demonstrated previously that activin A was upregulated intubular cells after renal ischemia and that activin receptorswere ubiquitously expressed in tubular cells (17). To examinewhether activin A produced in tubular cells was involved inthe expression of Pax-2, we analyzed the localization of Pax-2,activin A, and the type II activin receptor in the kidney afterrenal ischemia using serial sections. Type II activin receptorwas localized in tubular cells (Figure 5B). The distributionpattern was not altered by ischemia/reperfusion injury (datanot shown). As shown in Figure 5A and 5B, Pax-2–positivecells co-expressed the type II activin receptor. Furthermore,activin A was also expressed in these cells (Figure 5C).
Figure 5. Immunohistochemical localization of Pax-2-positive cells, activin A, and the type II activin receptor in the kidney after renal ischemia. Localization of Pax-2-positive cells (A), the type II activin receptor (B), and activin A (C) in the kidney after renal ischemia was examined as described in Materials and Methods. Consecutive sections from ischemic kidneys after 24 h of reperfusion were used. Note that Pax-2 (arrow in A) was co-localized with the type II activin receptor (brown staining around * in B), and activin A (arrow heads in C). Magnification, x1000. (D) Quantitative analysis of BrdU/Pax-2 double-positive cells. The number of BrdU/Pax-2 double-positive cells was analyzed in kidney sections obtained from normal kidney treated with rh-follistatin, sham-operated and ischemic kidney (I/R) treated with saline, rh-follistatin, and rh-activin A as described in Materials and Methods. Results are presented as the mean ± SEM of three independent experiments. * P < 0.05 compared with I/R+saline.
Effect of rh-Activin and rh-Follistatin on the Number of BrdU/Pax-2 Double-Positive Cells in the Kidney after Renal Ischemia
To examine whether or not endogenous activin A is involved inthe regulation of BrdU/Pax-2 double-positive cell growth duringregeneration process of the kidney, we analyzed the number ofBrdU/Pax-2 double-positive cells in an ischemic kidney treatedwith rh-activin A or rh-follistatin. As shown in Figure 5D,BrdU/Pax-2 double-positive cells were observed in ischemic kidneysbut not in sham-operated kidneys. Interestingly, rh-follistatinincreased; conversely, rh-activin A reduced the number of BrdU/Pax-2double-positive cells in the kidneys after renal ischemia. rh-Follistatindid not increase BrdU/Pax-2 double-positive cells in normalkidneys.
Expression of Pax-2 in Embryonic Kidney in Organ Culture
It was reported that activin A inhibited branching morphogenesisof the ureteric bud in organ culture (13). To examine whetheractivin A also modulated the expression of Pax-2 during development,we used metanephric organ culture system. As shown in Figure 6A,the expression level of Pax-2 was significantly decreasedin cultured metanephroi treated with activin A compared withthat in cultured metanephroi treated without activin A. Immunohistochemicalanalyses also demonstrated that Pax-2–positive cells wererarely observed in cultured metanephroi treated with activinA (Figure 6B).
Figure 6. Expression of the Pax-2 protein in embryonic kidney in organ culture. (A) Expression of the Pax-2 in cultured metanephroi treated with activin A. Total RNA was isolated from cultured metanephroi treated with (n = 5) or without activin A (n = 5) for 3 d. The expression of Pax-2 and GAPDH were analyzed by RT-PCR as described in Materials and Methods. The product sizes are indicated on the right. (B) Localization of the Pax-2 protein in cultured metanephroi was examined by indirect fluorescence immunostaining using an anti-human Pax-2 antibody. Frozen sections obtained from cultured metanephroi treated with (c and d) or without (a and b) 10 nM activin A for 3 d were used for experiments. (a and c) Pax-2 protein (red), nuclei (blue). (b and d) Nomarski images. Magnification, x400.
Expression of Pax-2 in Cultured Proximal Tubular Epithelial Cells
To further clarify the role of activin A in the regulation ofthe expression of Pax-2 in tubular cells, we also used a proximaltubular cell line, LLC-PK1 cells. First, we examined the relationshipbetween DNA synthesis and the expression of Pax-2 in LLC-PK1cells. Serum-starved LLC-PK1 cells were cultured in completemedium containing 5% FBS with or without 10 nM activin A forthe indicated times. As shown in Figure 7A, initiation of DNAsynthesis was observed 12 h after growth stimulation. The expressionlevel of Pax-2 was very low in quiescent cells, but it was significantlyincreased at 6 h and thereafter (Figure 7B), indicating thatupregulation of the Pax-2 expression preceded initiation ofDNA synthesis. Immunocytochemical staining also showed the presenceof Pax-2 protein in growing cells (Figure 7C-b) but not in quiescentcells (Figure 7C-a). These results suggest that induction ofPax-2 is associated with cell growth in LLC-PK1 cells, and itat least partly mimics in vivo events during tubular regeneration.
Figure 7. Effect of activin A on the expression of Pax-2 in LLC-PK1 cells. (A) Effect of activin A on DNA synthesis in LLC-PK1 cells. Serum-starved cells were cultured in complete medium containing 5% FBS with () or without (•) 10 nM activin A. DNA synthesis was analyzed by measuring [3H]thymidine incorporation. (B) Changes in the expression of mRNA for Pax-2 in LLC-PK1 cells. Serum-starved cells were cultured in complete medium containing 5% FBS with or without 10 nM activin A (Act), and total RNA was extracted at the indicated times. The expression of Pax-2 and GAPDH was analyzed by RT-PCR. Representative results of three experiments are shown. (C) The expression of Pax-2 protein in LLC-PK1 cells. Serum-starved cells (a and d) were cultured for 24 h in complete medium containing 5% FBS with (c and f) or without (b and e) activin A. The expression of Pax-2 was analyzed by indirect fluorescence immunostaining as described in Materials and Methods. (a through c) Pax-2 protein (red), nuclei (blue). (d throug f) Nomarski images. Magnification, x200.
We next examined the effect of activin A on the expression ofPax-2 in LLC-PK1 cells. Consistent with results obtained byin vivo experiments (Figure 5), activin A inhibited DNA synthesis(Figure 7A) and reduced the expression level of Pax-2 (Figure 7B).Activin A also decreased the number of Pax-2–positivenuclei in LLC-PK1 cells (Figure 7C-c).
Upregulated Expression of Pax-2 Protein in LLC-PK1 Cells Expressing Dominantly Negative Mutant Receptor
To further investigate the relationship between activin A andPax-2 expression in tubular cells, we used LLC-PK1-tARII cells(19). LLC-PK1-tARII is a stable cell line expressing truncatedtype II activin receptor that lacks intracellular kinase domain,in which activin signaling pathway was completely blocked (19).Serum-starved LLC-PK1-mock cells or LLC-PK1-tARII cells werecultured in complete medium containing 5% FBS, and the expressionof Pax-2 was examined at indicated times. As shown in Figure 8,the expression level of Pax-2 was transiently enhanced inLLC-PK1-mock cells, but it was decreased 48 h after growth stimulation.In LLC-PK1-tARII cells, the expression level of Pax-2 was alsoenhanced after growth stimulation. However, the increase continuedfor 48 h. Regarding that the expression of activin A was upregulatedin tubular cells after growth stimulation (19), this resultsuggest that the expression of Pax-2 in tubular cells was tonicallyinhibited by endogenous activin A.
Figure 8. Time course of Pax-2 expression in LLC-PK1-mock cells and LLC-PK1-tARII cells after growth stimulation. Serum-starved cells were cultured in complete medium containing 5% FBS for the indicated times. The expression of Pax-2 protein was analyzed by Western blotting as described in Materials and Methods.
Blockade of the Activin Signaling Induced an Immature Cell Phenotype in LLC-PK1 Cells
We further assessed cell phenotype in LLC-PK1-tARII cells. Weexamined the expression of E-cadherin, one of the cell adhesionmolecule expressed in differentiated epithelial tubular cells,in LLC-PK1-mock cells and LLC-PK1-tARII cells. Indirect immunofluorescencestaining showed that the expression of E-cadherin was observedand was localized to the lateral portion of the plasma membranein a linear staining pattern in LLC-PK1-mock cells (Figure 9A).In contrast, the expression of E-cadherin was almost absentin LLC-PK1-tARII cells (Figure 9B). Western blot analyses alsodemonstrated the decrease of E-cadherin protein in LLC-PK1-tARIIcells compared with that in LLC-PK1-mock cells (Figure 9C).
Figure 9. Expression of E-cadherin in LLC-PK1-mock cells and LLC-PK1-tARII cells. The expression of E-cadherin in serum-starved LLC-PK1-mock cells (A) and LLC-PK1-tARII cells (B) was examined by indirect fluorescence immunostaining as described in Materials and Methods. Magnification, x400. (C) Production of E-cadherin protein in serum-starved LLC-PK1-mock cells and LLC-PK1-tARII cells was examined by Western blotting as described in Materials and Methods.
In the present study, we examined the involvement of developmentalgenes critical for kidney organogenesis in tubular regeneration.We demonstrated that the expression of Pax-2 was upregulatedin tubular cells after ischemic injury (Figures 1 through 3).However, the expression of other transcription factors, suchas Pax-8, WT-1, Wnt-4, and BF-2, was not significantly changedin the kidneys after renal ischemia (Figure 1). These resultssuggest that, among transcription factors involved in renaldevelopment, developmental cascade controlled by Pax-2 is reactivatedand is potentially involved in tubular regeneration.
Pax-2 (18), a transcription factor belonging to the Pax family,which contains the DNA-binding paired domain, plays a key regulatoryrole during renal organogenesis (26). Proper temporal and spatialexpression of Pax-2 is tightly regulated during normal kidneydevelopment (27). Deregulated expression of Pax-2 was shownto be associated with the abnormality of the kidney in mice(28,29) and human (30,31). In renal tubular cell line, Pax-2acts as a transcription factor involved in tubular cell survival(32). These studies raise the possibility that Pax-2 is a strongcandidate of master gene controlling tubular cell proliferationand differentiation during kidney development
Pax-2 is downregulated when nephrogenesis is completed and isnot detected in mature kidneys (27). However, it was recentlyreported that Pax-2 was re-expressed in tubular cells damagedby renal toxins (33). Pax-2 was localized in proximal tubularcells (33), which are actively engaged in DNA synthesis duringregeneration (2). Furthermore, Pax-2–positive cells co-expresseda mesenchymal marker, vimentin, suggesting that these cellshave characteristics of immature progenitor-like tubular cells(33). Therefore, it is considered that Pax-2–positivecells mainly participate in cell proliferation and differentiationto reconstruct tubular structure during regeneration. Consistentwith the previous report (33), Pax-2–positive cells co-expressedvimentin in the ischemia/reperfusion model (Figure 2). Furthermore,Pax-2 protein was co-localized in BrdU-positive cells (Figures 4).Therefore, it is quite possible that Pax-2–positivecells are engaged in the regeneration process of the kidneyas renal epithelial progenitor-like tubular cells in this ischemia/reperfusionmodel as well as in the renal toxin injury model.
We previously demonstrated that the production of activin Ais upregulated in tubular cells after renal ischemia (17). Blockadeof endogenous activin A action by an administration of follistatinimproved renal dysfunction and protected the kidney from ischemicrenal injury by preventing apoptosis and promoting regeneration(17). However, the precise mechanism of endogenous activin Aaction still remains unknown. The present results demonstratethat activin A was localized in tubular cells expressing thetype II activin receptor in the kidney after renal ischemia(Figure 5, B and C). This result suggests that activin A producedin tubular cells acts as an autocrine factor in tubular cellssince activin A initiates its signal transduction pathway bybinding to the type II receptor. In addition, we also demonstratedthat the Pax-2–positive cells expressed the type II activinreceptor (Figure 5, A and B). Administration of activin A significantlyreduced the number of BrdU/Pax-2 double-positive tubular cellsin vivo (Figure 4). Inhibition of endogenous activin actionby exogenous follistatin increased the number of BrdU/Pax-2double-positive cells in the kidney after renal ischemia (Figure 4).We also demonstrated that activin A decreased the expressionof Pax-2 in embryonic kidney in organ culture system (Figure 6)as well as in proximal tubular cell line (Figure 7). Furthermore,blockade of the activin signaling by overexpressing dominant-negativemutant receptor promoted cell growth (19) and significantlyenhanced the expression of Pax-2 (Figure 8) and induced an immaturephenotype (Figure 9) in tubular cells. Collectively, it is likelythat the growth rate of the Pax-2–positive cells is controlledby an autocrine action of endogenous activin A during regeneration.
Although it is unclear at present by which mechanism activinA suppresses the expression of Pax-2 in tubular cells, it ispossible that activin A directly modulates the transcriptionof Pax-2. As an intracellular mediator of activin signaling,Smad proteins have been identified (34). Among the eight clonedSmad genes, Smad-2 and Smad-3 mediate the activin signals. Upondirect phosphorylation by the type I activin receptor, Smad-2or Smad-3 binds to its partner Smad-4 to form a heteromericcomplex and translocates into the nucleus, where it can potentiallyregulate the transcription of target genes. We found that theproposed Smad binding element (SBE) of CAGAC was present onthe upstream of the transcription start site in the sequenceof the human Pax-2 promoter (35). It is also known that bindingto the SBE is not sufficient for Smad-dependent transcriptionalactivation, and additional DNA contacts seem to be necessaryfor specific, high-affinity binding of a Smad complex to thetarget gene in many biologic systems (34). In this regard, wealso observed the presence of a binding site for transcriptionalco-factors of Smad protein, such as Sp1 (36) or NF-B (37). Therefore,it is quite possible that Smads bind to the Pax-2 promoter andrepress its transcription rates with transcriptional co-factors.Further study is necessary to address this issue.
In summary, we demonstrate here that Pax-2 was upregulated intubular cells after renal ischemia. Pax-2 protein was co-localizedwith activin A and the type II activin receptor. Furthermore,we show the inhibitory effect of activin A on the expressionof Pax-2 in tubular cells both in vivo and in vitro. ActivinA may be a critical regulator of tubular regeneration that modulatescell growth and differentiation of Pax-2–positive progenitor-liketubular cells.
Acknowledgments
This study was supported by a Grant-in-Aid for Scientific Researchfrom the Ministry of Science, Education, Sports and Cultureof Japan. The authors thank Mayumi Odagiri for her secretarialassistance.
Witzgall R, Brown D, Schwarz C, Bonventre JV: Localization of proliferating cell nuclear antigen, vimentin, c-fos, and clusterin in the post-ischemic kidney. J Clin Invest 93: 2175–2188, 1994
Toback FG: Regeneration after acute tubular necrosis. Kidney Int 41: 226–246, 1992[Medline]
Nigam SK, Lieberthal W: Acute renal failure. III. The role of growth factors in the process of renal regeneration and repair. Am J Physiol 279: F3–F11, 2000
Kuure S, Vuolteenaho R, Vainio S: Kidney morphogenesis: Cellular and molecular regulation. Mech Dev 92: 31–45, 2000[CrossRef][Medline]
Davies JA, Bard JBL: The development of the kidney. Curr Top Dev Biol 39: 245–301, 1998[Medline]
Mather JP, Moore A, Li RH: Activins inhibins and follistatins: Further thoughts on growing family of regulators. Proc Soc Exp Biol Med 215: 209–222, 1997[Abstract]
Ying SY, Zhang Z, Furst B, Batres Y, Huang G, Guangwu L: Activins and activin receptors in cell growth. Proc Soc Exp Biol Med 214: 114–122, 1997[Abstract]
Igawa T, Matsumoto K, Saito Y, Nakamura T: Hepatocyte growth factor may function as a renotropic factor for regeneration in rats with acute renal injury. Am J Physiol 265: F61–F69, 1993
Nakamura T, Takio K, Eto Y, Shibai H, Titani K, Sugino H: Activin-binding protein from rat ovary is follistatin. Science 247: 836–838, 1990[Abstract/Free Full Text]
Ueno N, Ling N, Ying SY, Esch F, Shimasaki S, Guillemin R: Isolation and partial characterization of follistatin: A novel Mr 35,000 monomeric protein that inhibits the release of follicle stimulating hormone. Proc Natl Acad Sci USA 84: 8282–8286, 1987[Abstract/Free Full Text]
Shimanaka M, Inoue S, Shimasaki S, Ling N: Follistatin binds to both activin and inhibin though the common beta-subunit. Endocrinology 128: 3313–3315, 1991[Abstract]
Tuuri T, Eramaa M, Hilden K, Ritvos O: The tissue distribution of activin A and B subunits and follistatin mRNA suggests multiple sites of action of the activin-follistatin system during human development. J Clin Endocrinol Metab 78: 1521–1524, 1994[Abstract]
Ritvos O, Tuuri T, Eramaa M, Sainio K, Hiden K, Saxen L, Gilbert SF: Activin disrupts epithelial branching morphogenesis in developing glandular organs of the mouse. Mech Dev 50: 229–245, 1995[CrossRef][Medline]
Maeshima A, Zhang YQ, Furukawa M, Naruse T, Kojima I: Hepatocyte growth factor induces branching tubulogenesis by modulating the activin-follistatin system. Kidney Int 58: 1511–1522, 2000[CrossRef][Medline]
Maeshima A, Shiozaki S, Tajima T, Nakazato Y, Naruse T, Kojima I: Number of glomeruli is increased in transgenic mice expressing truncated type II activin receptor. Biochem Biophys Res Commun 268: 445–449, 2000[CrossRef][Medline]
Maeshima A, Zhang YQ, Naruse T, Nojima Y, Kojima I: Involvement of the activin-follistatin system in tubular regeneration after renal ischemia in rats. J Am Soc Nephrol 12: 1685–1695, 2001[Abstract/Free Full Text]
Dressler GR, Douglass EC: Pax-2 is a DNA-binding protein expressed in embryonic kidney and Wilms tumor. Proc Natl Acad Sci USA 89: 1179–1183, 1992[Abstract/Free Full Text]
Maeshima A, Nojima Y, Kojima I: Activin A: An autocrine regulator of cell growth and differentiation in proximal tubular cells. Kidney Int 2002, in press
Vainio SJ, Uusitalo MS: A road to kidney tubules via Wnt pathway. Pediatr Nephrol 15: 151–156, 2000[CrossRef][Medline]
Stark K, Vainio S, Vassileva G, McMahon AP: Epithelial transformation of metanephric mesenchyme in the developing kidney regulated by Wnt-4. Nature 372: 679–683, 1994[CrossRef][Medline]
Mrowka C, Schedl A: Wilms tumor suppressor gene WT1: From structure to renal pathophysiologic features. J Am Soc Nephrol 11 [Suppl]: S106–S115, 2000[Abstract/Free Full Text]
Plachov D, Choedhury K, Walther C, Simon D, Guenet JL, Gruss P: Pax-8, a murine paired box gene expressed in the developing excretory system and thyroid gland. Development 110: 643–651, 1990[Abstract/Free Full Text]
Poleev A, Fickenscher H, Mudlos S, Winterpatcht A, Zabel B, Fidler A, Gruss P: PAX8, a human paired box gene: Isolation and expression in developing thyroid, kidney and Wilms tumor. Development 116: 611–623, 1992[Abstract]
Hatini V, Huh SO, Herzlinger D, Soares VC, Lai E: Essential role of stromal mesenchyme in kidney morphogenesis revealed by targeted disruption of Wingled Helix transcription factor BF-2. Genes Dev 10: 1467–1478, 1996[Abstract/Free Full Text]
Dressler Gr, Woolf AS: Pax-2 in development and renal disease. Int J Dev Biol 43: 463–468, 1999[Medline]
Rothenpieler UW, Dressler GR: Pax-2 is required for mesenchyme-to-epithelium conversion during kidney development. Development 119: 711–720, 1993[Abstract]
Dressler GR, Wilkinson JE, Rothenpieler UW, Patterson LT, Williams-Simons L, Westphal H: Deregulation of Pax-2 expression in transgenic mice generates severe kidney abnormalities. Nature 362: 65–67, 1993[CrossRef][Medline]
Torres M, Gomez-Pardo E, Dressler GR, Gruss P: Pax-2 controls multiple steps of urogenital development. Development 121: 4057–4065, 1995[Abstract]
Sanyanusin P, Schimmenti LA, McNoe LA, Ward TA, Pierpont ME, Sullivan MJ, Dobyns WB, Eccles MR: Mutation of the PAX2 gene in a family with optic nerve colobomas, renal anomalies and vesicoureteral reflux. Nat Genet 9: 358–364, 1995[CrossRef][Medline]
Winyard PJ, Risdon RA, Sams VR, Dressler GR, Woolf AS: The PAX2 transcription factor is expressed in cystic and hyperproliferative dysplastic epithelia in human kidney malformations. J Clin Invest 98: 451–459, 1996[Medline]
Torban E, Eccles MR, Favor J, Goodyer PR: PAX2 suppresses apoptosis in renal collecting duct cells. Am J Pathol 157: 833–842, 2000[Abstract/Free Full Text]
Imgrund M, Grone HJ, Kretzler M, Holzman L, Schlondorff D, Rothernpaler UW: Re-expression of the development gene Pax-2 during experimental acute tubular necrosis in mice. Kidney Int 56: 1423–1431, 1999[CrossRef][Medline]
Massague J, Wotton D: Transcriptional control by the TGF-beta/Smad signaling system. EMBO J 19: 1745–1754, 2000[CrossRef][Medline]
Stayner CK, Cunliffe HE, Ward TA, Eccles MR: Cloning and characterization of human PAX2 promoter. J Biol Chem 273: 25472–25479, 1998[Abstract/Free Full Text]
Feng XH, Lin X, Derynck R: Smad2, Smad3 and Smad4 cooperate with Sp1 to induce p15 (Ink4B) transcription in response to TGF-beta. EMBO J 19: 5178–5193, 2000[CrossRef][Medline]
Lopez-Rovira T, Chalaux E, Rosa JL, Batrons R, Ventura F: Interaction and functional cooperation of NF-kappa B with Smads. Transcriptional regulation of the junB promoter. J Biol Chem 275: 28937–28946, 2000[Abstract/Free Full Text]
Received for publication April 15, 2002.
Accepted for publication August 6, 2002.
This article has been cited by other articles:
V. Cantaluppi, L. Biancone, G. M. Romanazzi, F. Figliolini, S. Beltramo, F. Galimi, M. G. Camboni, E. Deriu, P. Conaldi, A. Bottelli, et al. Macrophage Stimulating Protein May Promote Tubular Regeneration after Acute Injury
J. Am. Soc. Nephrol.,
October 1, 2008;
19(10):
1904 - 1918.
[Abstract][Full Text][PDF]
S. Metsuyanim, N. Pode-Shakked, K. M. Schmidt-Ott, G. Keshet, G. Rechavi, D. Blumental, and B. Dekel Accumulation of Malignant Renal Stem Cells Is Associated with Epigenetic Changes in Normal Renal Progenitor Genes
Stem Cells,
July 1, 2008;
26(7):
1808 - 1817.
[Abstract][Full Text][PDF]
S.-L. Zhang, Y.-W. Chen, S. Tran, I. Chenier, M.-J. Hebert, and J. R. Ingelfinger Reactive Oxygen Species in the Presence of High Glucose Alter Ureteric Bud Morphogenesis
J. Am. Soc. Nephrol.,
July 1, 2007;
18(7):
2105 - 2115.
[Abstract][Full Text][PDF]
S. Gupta, C. Verfaillie, D. Chmielewski, S. Kren, K. Eidman, J. Connaire, Y. Heremans, T. Lund, M. Blackstad, Y. Jiang, et al. Isolation and Characterization of Kidney-Derived Stem Cells
J. Am. Soc. Nephrol.,
November 1, 2006;
17(11):
3028 - 3040.
[Abstract][Full Text][PDF]
V. Fonsato, S. Buttiglieri, M. C. Deregibus, V. Puntorieri, B. Bussolati, and G. Camussi Expression of Pax2 in Human Renal Tumor-Derived Endothelial Cells Sustains Apoptosis Resistance and Angiogenesis
Am. J. Pathol.,
February 1, 2006;
168(2):
706 - 713.
[Abstract][Full Text][PDF]
A. Maeshima, H. Sakurai, and S. K. Nigam Adult Kidney Tubular Cell Population Showing Phenotypic Plasticity, Tubulogenic Capacity, and Integration Capability into Developing Kidney
J. Am. Soc. Nephrol.,
January 1, 2006;
17(1):
188 - 198.
[Abstract][Full Text][PDF]
Q. Cai, N. I. Dmitrieva, J. D. Ferraris, H. L. Brooks, B. W. M. van Balkom, and M. Burg Pax2 expression occurs in renal medullary epithelial cells in vivo and in cell culture, is osmoregulated, and promotes osmotic tolerance
PNAS,
January 11, 2005;
102(2):
503 - 508.
[Abstract][Full Text][PDF]
S.-L. Zhang, B. Moini, and J. R. Ingelfinger Angiotensin II Increases Pax-2 Expression in Fetal Kidney Cells Via the AT2 Receptor
J. Am. Soc. Nephrol.,
June 1, 2004;
15(6):
1452 - 1465.
[Abstract][Full Text][PDF]
S. Yamashita, A. Maeshima, I. Kojima, and Y. Nojima Activin A Is a Potent Activator of Renal Interstitial Fibroblasts
J. Am. Soc. Nephrol.,
January 1, 2004;
15(1):
91 - 101.
[Abstract][Full Text][PDF]
J. Yoshino, T. Monkawa, M. Tsuji, M. Hayashi, and T. Saruta Leukemia Inhibitory Factor Is Involved in Tubular Regeneration after Experimental Acute Renal Failure
J. Am. Soc. Nephrol.,
December 1, 2003;
14(12):
3090 - 3101.
[Abstract][Full Text][PDF]
A. Maeshima, S. Yamashita, and Y. Nojima Identification of Renal Progenitor-Like Tubular Cells that Participate in the Regeneration Processes of the Kidney
J. Am. Soc. Nephrol.,
December 1, 2003;
14(12):
3138 - 3146.
[Abstract][Full Text][PDF]
Top
Abstract
Introduction
superfamily, modulates cell growth and differentiation in many types of cells (6,7). Activins are dimeric proteins, and subunits of activin are expressed in various organs (8). An important modulator of activin is follistatin (9). This protein specifically binds to activins and related ligands with high affinity and blocks their actions (10,11). Follistatin is synthesized in the target cells of activins and remains in the extracellular matrix. Furthermore, the production of follistatin is regulated by activins. Hence, activin and follistatin modulate cellular function in a complex manner. Activin A and follistatin are expressed in a developing kidney (12). Activin A disrupts ureteric bud branching in an embryonic kidney in organ culture (13) and also inhibits branching tubulogenesis in an in vitro tubulogenesis model (14). The number of glomeruli is increased in the kidney of transgenic mice expressing dominantly negative activin receptors (15). Collectively, activin A is a negative regulator of tubulogenesis during kidney development (16). We recently demonstrated the involvement of the activin-follistatin system in tubular regeneration after renal ischemia (17). Activin A, which was not detected in normal kidney, was upregulated in tubular cells after renal ischemia. Exogenous follistatin accelerated renal regeneration by enhancing DNA synthesis and preventing apoptosis in tubular cells. Presumably, exogenous follistatin enhanced tubular regeneration by blocking the action of endogenous activin A. However, the mechanism by which endogenous activin A inhibits tubular regeneration still remains unknown. Considering that regeneration processes may recapitulate developmental paradigms to restore organ or tissue function (1), it is quite possible that activin modulates the expression of a set of developmental genes during tubular regeneration. 
B (37). Therefore, it is quite possible that Smads bind to the Pax-2 promoter and repress its transcription rates with transcriptional co-factors. Further study is necessary to address this issue. 
) or without (•) 10 nM activin A. DNA synthesis was analyzed by measuring [3H]thymidine incorporation. (B) Changes in the expression of mRNA for Pax-2 in LLC-PK1 cells. Serum-starved cells were cultured in complete medium containing 5% FBS with or without 10 nM activin A (Act), and total RNA was extracted at the indicated times. The expression of Pax-2 and GAPDH was analyzed by RT-PCR. Representative results of three experiments are shown. (C) The expression of Pax-2 protein in LLC-PK1 cells. Serum-starved cells (a and d) were cultured for 24 h in complete medium containing 5% FBS with (c and f) or without (b and e) activin A. The expression of Pax-2 was analyzed by indirect fluorescence immunostaining as described in Materials and Methods. (a through c) Pax-2 protein (red), nuclei (blue). (d throug f) Nomarski images. Magnification, x200.
