Aquaporin 3 Expression Is Up-Regulated by TGF-β1 in Rat Peritoneal Mesothelial Cells and Plays a Role in Wound Healing
2012; Elsevier BV; Volume: 181; Issue: 6 Linguagem: Inglês
10.1016/j.ajpath.2012.08.018
ISSN1525-2191
AutoresHye-Myung Ryu, Eun-Joo Oh, Sun-Hee Park, Chan‐Duck Kim, Ji‐Young Choi, Jang‐Hee Cho, In‐San Kim, Tae‐Hwan Kwon, Ho-Young Chung, Min Yoo, Yong-Lim Kim,
Tópico(s)Parathyroid Disorders and Treatments
ResumoAquaporin 3 (AQP3) is expressed in many tissues including the peritoneum and kidney. In cultured mesothelial cells, glucose up-regulates AQP3, which may be important for water transport through the peritoneal membrane. However, there has been no research into the role of AQP3 in human peritoneal mesothelial cell (HPMC) migration or peritoneal fibrosis. We investigated the effects of transforming growth factor-β1 (TGF-β1) on AQP3 expression in HPMCs. We also investigated the role of AQP3 in the peritoneal wound healing process in rats. Chronic exposure to glucose-containing solution increased peritoneal myofibroblasts, with TGF-β1 and AQP3 expression in a model of long-term peritoneal dialysis. In vitro, TGF-β1 induced AQP3 expression in HPMCs. AQP3 knockdown by small-interfering RNA inhibited TGF-β1–induced AQP3 and α-smooth muscle actin expression and also slowed HPMC migration. AQP3 overexpression induced faster migration of HPMCs. Treatment with an extracellular signal-regulated kinase inhibitor and p38 kinase inhibitor attenuated TGF-β1–induced AQP3 expression in HPMCs. These data suggest that TGF-β1 induces AQP3 and that AQP3 has a critical role in TGF-β–induced HPMC migration. These findings provide evidence of a novel role for AQP3 in peritoneal fibrosis and wound healing. The effect of TGF-β1 on AQP3 expression in HPMCs is mediated, at least in part, by ERK and p38 signaling. Aquaporin 3 (AQP3) is expressed in many tissues including the peritoneum and kidney. In cultured mesothelial cells, glucose up-regulates AQP3, which may be important for water transport through the peritoneal membrane. However, there has been no research into the role of AQP3 in human peritoneal mesothelial cell (HPMC) migration or peritoneal fibrosis. We investigated the effects of transforming growth factor-β1 (TGF-β1) on AQP3 expression in HPMCs. We also investigated the role of AQP3 in the peritoneal wound healing process in rats. Chronic exposure to glucose-containing solution increased peritoneal myofibroblasts, with TGF-β1 and AQP3 expression in a model of long-term peritoneal dialysis. In vitro, TGF-β1 induced AQP3 expression in HPMCs. AQP3 knockdown by small-interfering RNA inhibited TGF-β1–induced AQP3 and α-smooth muscle actin expression and also slowed HPMC migration. AQP3 overexpression induced faster migration of HPMCs. Treatment with an extracellular signal-regulated kinase inhibitor and p38 kinase inhibitor attenuated TGF-β1–induced AQP3 expression in HPMCs. These data suggest that TGF-β1 induces AQP3 and that AQP3 has a critical role in TGF-β–induced HPMC migration. These findings provide evidence of a novel role for AQP3 in peritoneal fibrosis and wound healing. The effect of TGF-β1 on AQP3 expression in HPMCs is mediated, at least in part, by ERK and p38 signaling. The aquaporins (AQPs) are a family of small integral membrane proteins that transport water alone or water plus small solutes such as glycerol.1Agre P. King L.S. Yasui M. Guggino W.B. Ottersen O.P. Fujiyoshi Y. Engel A. Nielsen S. Aquaporin water channels: from atomic structure to clinical medicine.J Physiol. 2002; 542: 3-16Crossref PubMed Scopus (915) Google Scholar To date, 13 subtypes of AQPs have been identified in mammals.2Morishita Y. Matsuzaki T. Hara-chikuma M. Andoo A. Shimono M. Matsuki A. Kobayashi K. Ikeda M. Yamamoto T. Verkman A. Kusano E. Ookawara S. Takata K. Sasaki S. Ishibashi K. Disruption of aquaporin-11 produces polycystic kidneys following vacuolization of the proximal tubule.Mol Cell Biol. 2005; 25: 7770-7779Crossref PubMed Scopus (213) Google Scholar, 3Itoh T. Rai T. Kuwahara M. Ko S.B. Uchida S. Sasaki S. Ishibashi K. Identification of a novel aquaporin, AQP12, expressed in pancreatic acinar cells.Biochem Biophys Res Commun. 2005; 330: 832-838Crossref PubMed Scopus (160) Google Scholar AQP3 is permeable not only to water but also to glycerol and is thus sometimes called "aquaglyceroporin."1Agre P. King L.S. Yasui M. Guggino W.B. Ottersen O.P. Fujiyoshi Y. Engel A. Nielsen S. Aquaporin water channels: from atomic structure to clinical medicine.J Physiol. 2002; 542: 3-16Crossref PubMed Scopus (915) Google Scholar AQP3 is expressed specifically in skin keratinocytes, in epithelial cells in the airway, and in the collecting ducts of the kidneys.4Mobasheri A. Wray S. Marples D. Distribution of AQP2 and AQP3 water channels in human tissue microarrays.J Mol Histol. 2005; 36: 1-14Crossref PubMed Scopus (153) Google Scholar Recent research has found that AQPs contribute to cell migration in endothelial cells,5Saadoun S. Papadopoulos M.C. Hara-Chikuma M. Verkman A.S. Impairment of angiogenesis and cell migration by targeted aquaporin-1 gene disruption.Nature. 2005; 434: 786-792Crossref PubMed Scopus (646) Google Scholar brain astroglial cells,6Saadoun S. Papadopoulos M.C. Watanabe H. Yan D. Manley G.T. Verkman A.S. Involvement of aquaporin-4 in astroglial cell migration and glial scar formation.J Cell Sci. 2005; 118: 5691-5698Crossref PubMed Scopus (376) Google Scholar and kidney proximal tubule cells.7Hara-Chikuma M. Verkman A.S. Aquaporin-1 facilitates epithelial cell migration in kidney proximal tubule.J Am Soc Nephrol. 2006; 17: 39-45Crossref PubMed Scopus (150) Google Scholar In cultured mesothelial cells, glucose up-regulates both mRNA and protein expression of AQP1 and AQP3, which may be important for water transport through the peritoneal membrane.8Lai K.N. Li F.K. Lan H.Y. Tang S. Tsang A.W. Chan D.T. Leung J.C. Expression of aquaporin-1 in human peritoneal mesothelial cells and its upregulation by glucose in vitro.J Am Soc Nephrol. 2001; 12: 1036-1045PubMed Google Scholar, 9Lai K.N. Leung J.C. Chan L.Y. Tang S. Li F.K. Lui S.L. Chan T.M. Expression of aquaporin-3 in human peritoneal mesothelial cells and its up-regulation by glucose in vitro.Kidney Int. 2002; 62: 1431-1439Crossref PubMed Google Scholar However, there has been no research into the role of AQP3 in human peritoneal mesothelial cell (HPMC) migration or peritoneal fibrosis. Peritoneal injury, whether induced by infection, ischemia, inflammation, or surgery, can lead to peritoneal adhesions and fibrosis. Transforming growth factor-β (TGF-β) is a primary regulator of this process.10Chegini N. TGF-beta system: the principal profibrotic mediator of peritoneal adhesion formation.Semin Reprod Med. 2008; 26: 298-312Crossref PubMed Scopus (79) Google Scholar To investigate the role of AQP3 in a model of animal peritoneal injury, we used the long-term peritoneal dialysis (PD) model. Long-term exposure to PD fluids results in peritoneal injury, with concurrent structural changes and functional decline such as ultrafiltration loss.11Selgas R. Bajo A. Jiménez-Heffernan J.A. Sánchez-Tomero J.A. Del Peso G. Aguilera A. López-Cabrera M. Epithelial-to-mesenchymal transition of the mesothelial cell: its role in the response of the peritoneum to dialysis.Nephrol Dial Transplant. 2006; 21: ii2-ii7Crossref PubMed Scopus (90) Google Scholar Peritoneal dialysis fluids are hyperosmotic solutions that are usually glucose based. High glucose concentrations and glucose degradation products from PD solutions lead to production of TGF-β and vascular endothelial growth factor by mesothelial cells.12Hirahara I. Kusano E. Yanagiba S. Miyata Y. Ando Y. Muto S. Asano Y. Peritoneal injury by methylglyoxal in peritoneal dialysis.Perit Dial Int. 2006; 26: 380-392PubMed Google Scholar The peritoneum is the largest extensive serous membrane in the body, covering the visceral organs and lining the abdominal cavity. The peritoneum comprises a monolayer of mesothelial cells that provide the first line of defense in peritoneal protection during chemical, surgical, or bacterial insult.13Yung S. Li F.K. Chan T.M. Peritoneal mesothelial cell culture and biology.Perit Dial Int. 2006; 26: 162-173PubMed Google Scholar Pathologic changes in the peritoneal membrane due to long-term PD are characterized by loss of mesothelial cells and enlargement of the submesothelial compact zone due to interstitial fibrosis.14Williams J.D. Craig K.J. Topley N. Von Ruhland C. Fallon M. Newman G.R. Mackenzie R.K. Peritoneal Biopsy Study GroupMorphologic changes in the peritoneal membrane of patients with renal disease.J Am Soc Nephrol. 2002; 13: 470-479PubMed Google Scholar Myofibroblasts in peritoneal tissue are α-smooth muscle actin (α-SMA)–positive interstitial cells.15Arora P.D. McCulloch C.A. Dependence of collagen remodelling on alpha-smooth muscle actin expression by fibroblasts.J Cell Physiol. 1994; 159: 161-175Crossref PubMed Scopus (246) Google Scholar Fibrosis is a disease process common to the kidney and lungs in the context of reiterative injury or infection,16Desmoulière A. Chaponnier C. Gabbiani G. Tissue repair, contraction, and the myofibroblast.Wound Repair Regen. 2005; 13: 7-12Crossref PubMed Scopus (710) Google Scholar and cell transdifferentiation is a generic process by which myofibroblasts are generated in injured tissues such as the kidney and lung.17Desmoulière A. Darby I.A. Gabbiani G. Normal and pathologic soft tissue remodeling: role of the myofibroblast, with specific emphasis on liver and kidney fibrosis.Lab Invest. 2003; 83: 1689-1707Crossref PubMed Scopus (296) Google Scholar, 18Thannickal V.J. Toews G.B. White E.S. Lynch III, J.P. Martinez F.J. Mechanisms of pulmonary fibrosis.Annu Rev Med. 2004; 55: 395-417Crossref PubMed Scopus (530) Google Scholar Myofibroblasts are key cells in the wound-healing response and are responsible for wound contraction, recruitment of inflammatory cells, and remodeling of the extracellular matrix.16Desmoulière A. Chaponnier C. Gabbiani G. Tissue repair, contraction, and the myofibroblast.Wound Repair Regen. 2005; 13: 7-12Crossref PubMed Scopus (710) Google Scholar TGF-β, a member of a family of growth factors that are important in wound healing and response to injury,19Bottinger E.P. Bitzer M. TGF-β signaling in renal disease.J Am Soc Nephrol. 2002; 13: 2600-2610Crossref PubMed Scopus (661) Google Scholar, 20Margetts P.J. Oh K.H. Kolb M. Transforming growth factor-beta: importance in long-term peritoneal membrane changes.Perit Dial Int. 2005; 25: S15-S17PubMed Google Scholar is also a key fibrogenic growth factor in the peritoneum and mediates peritoneal fibrosis. TGF-β stimulates fibroblast proliferation and increases the synthesis of a number of extracellular matrix components. Although transient TGF-β1 activity is involved in the repair and regeneration of tissues, persistent TGF-β1 function leads to excessive fibrosis.21Cutroneo K.R. TGF-beta-induced fibrosis and SMAD signaling: oligo decoys as natural therapeutics for inhibition of tissue fibrosis and scarring.Wound Repair Regen. 2007; 15: S54-S60Crossref PubMed Scopus (147) Google Scholar However, the relationship between TGF-β1 and AQP3 is not known. The objectives of the present study were to investigate the effects of TGF-β1 on AQP3 expression in HPMCs and the role of AQP3 both in vitro and in vivo in the wound healing process and fibrosis. The rat PD model, which is a long-term infusion model, was used to investigate the role of AQP3 in peritoneal injury. In brief, permanent PD catheters were inserted in 18 male 8-week-old Sprague-Dawley rats (Hyochang Science, Daegu, Korea) that were divided into 2 groups: the control group (C), in which each rat had a catheter but received no dialysis solution infusion, and the dialysis group (D), in which each rat had a catheter and received dialysis solution infusion for 8 weeks. Group D rats received infusions of 25 mL 4.25% glucose dialysis solution (Dianeal; Baxter Healthcare Ltd., Singapore) b.i.d. for 8 weeks.22Yu M.A. Shin K.S. Kim J.H. Kim Y.I. Chung S.S. Park S.H. Kim Y.L. Kang D.H. HGF and BMP-7 ameliorate high glucose-induced epithelial-to-mesenchymal transition of peritoneal mesothelium.J Am Soc Nephrol. 2009; 20: 567-581Crossref PubMed Scopus (107) Google Scholar At the end of the 8 weeks, all rats were sacrificed, and peritoneal tissue was collected. AQP3 and α-SMA expression in peritoneal tissue was analyzed using Western blot analysis. All experiments were performed according to animal experimental procedures approved by the Animal Care and Use Committee of Kyungpook National University. Immunofluorescence was performed as described previously.22Yu M.A. Shin K.S. Kim J.H. Kim Y.I. Chung S.S. Park S.H. Kim Y.L. Kang D.H. HGF and BMP-7 ameliorate high glucose-induced epithelial-to-mesenchymal transition of peritoneal mesothelium.J Am Soc Nephrol. 2009; 20: 567-581Crossref PubMed Scopus (107) Google Scholar Primary antibodies against pan-cytokeratin (Thermo Fisher Scientific, Inc., Fremont, CA), α-SMA (Abcam PLC, Cambridge, UK), TGF-β1 (Santa Cruz Biotechnology, Inc., Santa Cruz, CA), and AQP3 (Alomone Laboratories, Ltd., Jerusalem, Israel) were used. HPMCs were collected from omental tissue from patients undergoing abdominal surgery. Informed consent was provided by each patient. HPMCs were isolated using the trypsin-EDTA method and cultured in M199 medium containing 20% fetal bovine serum (FBS). To confirm the HPMC population before starting experiments, we usually tested the HPMC population via immunofluorescence staining of cytokeratin and α-SMA. To minimize the fibroblast contamination, we used a cobblestone-shaped HPMC, and if the fibroblastoid-shaped cell was contained above 5%, we discarded the cell. The cells were plated at confluence and allowed to rest in 1% FBS-containing M199 medium for 24 hours to allow growth synchronization. The cells were then exposed to recombinant human TGF-β1 (R&D Systems, Inc., Minneapolis, MN) at concentrations of 2 ng/mL for 48 hours. For dose-dependent studies, TGF-β1 was used at 1, 2, 3, 4, and 5 ng/mL. For time-dependent studies, HPMCs were treated for 24, 48, and 72 hours with 2 ng/mL TGF-β1. For phosphorylation studies, HPMCs were treated for 1 hour with 2 ng/mL TGF-β1. Western blot analysis was performed as described previously,23Oh E.J. Ryu H.M. Choi S.Y. Yook J.M. Kim C.D. Park S.H. Chung H.Y. Kim I.S. Yu M.A. Kang D.H. Kim Y.L. Impact of low glucose degradation product bicarbonate/lactate-buffered dialysis solution on the epithelial-mesenchymal transition of peritoneum.Am J Nephrol. 2010; 31: 58-67Crossref PubMed Scopus (24) Google Scholar using primary antibodies against α-SMA (Sigma-Aldrich, St. Louis, MO), AQP3 (Alomone Laboratories), β-actin (Sigma-Aldrich), ERK (Santa Cruz Biotechnology), phosphorylated (p) ERK, p38, p-p38, c-Jun N-terminal kinase (JNK), and p-JNK (Cell Signaling Technology, Inc., Beverly, MA). A horseradish-peroxidase–conjugated polyclonal goat anti-rabbit immunoglobulin or goat anti-mouse immunoglobulin (Dako, Glostrup, Denmark) were used as secondary antibodies for Western blot analysis. Positive immunoreactive bands were quantified via densitometry and compared with human β-actin expression. Real-time RT-PCR was performed as described previously.23Oh E.J. Ryu H.M. Choi S.Y. Yook J.M. Kim C.D. Park S.H. Chung H.Y. Kim I.S. Yu M.A. Kang D.H. Kim Y.L. Impact of low glucose degradation product bicarbonate/lactate-buffered dialysis solution on the epithelial-mesenchymal transition of peritoneum.Am J Nephrol. 2010; 31: 58-67Crossref PubMed Scopus (24) Google Scholar The sequences of the primer pairs were as follows: AQP3, forward 5′-GACCTTTGCCATGTGCTTCCT-3′ and reverse 5′-CCAAAAACTATTCCAGCACCCA-3′; β-actin, forward 5′-CTGTCCACCTTCCAGCAGATGT-3′ and reverse 5′-CGCAACTAAGTCATAGTCCGCC-3′; and α-SMA, forward 5′-TCCGGAGCGCAAATACTCTGT-3′ and reverse 5′-CCGGCTTCATCGTATTCCTGT-3′. Human AQP3 small-interfering RNA (siRNA) and nontargeting siRNA as a negative control were purchased from Dharmacon, Inc. (Chicago, IL) and used at 20 nmol/L. Opti-MEM transfection media and lipofectamine (both from Invitrogen, Paisley, UK) were used for transfection. HPMCs were seeded at 1 day before transfection and cultured so that they were 40% to 50% confluent on the following day. RNA interference (RNAi) duplexes for AQP3 were mixed with lipofectamine to form a transfection complex that was added to cells in 6-well plates. At 24 hours after knockdown of AQP3, cells were incubated in M199 medium containing 1% FBS for 24 hours and then with or without 2 ng/mL TGF-β1 for 48 hours. Real-time RT-PCR, Western blot analysis, and immunofluorescence were performed to analyze AQP3 and α-SMA expression. Proliferation was measured via 5-bromo-2-deoxyuridine (BrdU) incorporation. Cells were plated in 96-well plates at 1 × 104 cells per well. After 24 hours of siRNA transfection, cells were incubated in M199 medium containing 1% FBS for 24 hours and then were treated with 2 ng/mL TGF-β1 for 24 hours. BrdU labeling solution was added, and cells were reincubated for an additional 24 hours. Incorporated BrdU was detected by using a BrdU ELISA kit (Calbiochem Corp., La Jolla, CA) according to the manufacturer's instructions. Adenoviral transfection of the human AQP3 gene was used to investigate the effect of AQP3 on HPMCs. In brief, human AQP3 cDNA was purchased from Origene Technologies, Inc. (Rockville, MD), and AQP3 cDNA was cloned into the pGEM-T Easy Vector (Promega Corp., Madison, WI). Restriction enzyme analyses and DNA sequencing were used to confirm the accuracy of the pGEM-T–AQP3 sequence. AQP3 cDNA from pGEM-T–AQP3 was inserted into the pShuttle-CMV vector. pShuttle/AQP3 plasmids were linearized with restriction enzymes and co-transformed into Escherichia coli by using pAdEasy-1 vector. Recombinant plasmids were transfected into the adenovirus packing cell line HEK293. Green fluorescent protein adenovirus (Ad-GFP) (Seoulin Corp., Seoul, Korea) was used as a negative control. HPMCs were infected with AQP3 (Ad-AQP3) and Ad-GFP at a concentration of 100 multiplicity of infection (MOI) for 24 hours. After transfection, the cells were incubated in M199 medium containing 1% FBS for 24 hours, then further incubated with or without 2 ng/mL TGF-β1 for 48 hours. Real-time RT-PCR, Western blot analysis, and immunofluorescence were performed to analyze AQP3 and α-SMA expression. Immunofluorescence was performed as described previously.23Oh E.J. Ryu H.M. Choi S.Y. Yook J.M. Kim C.D. Park S.H. Chung H.Y. Kim I.S. Yu M.A. Kang D.H. Kim Y.L. Impact of low glucose degradation product bicarbonate/lactate-buffered dialysis solution on the epithelial-mesenchymal transition of peritoneum.Am J Nephrol. 2010; 31: 58-67Crossref PubMed Scopus (24) Google Scholar For immunofluorescence staining, primary antibodies against α-SMA (Sigma-Aldrich), pan-cytokeratin (Thermo Fisher Scientific), and AQP3 (Alomone Laboratories) were used. Fluorescein-conjugated secondary antibodies (AlexaFluor 488 and AlexaFluor 594; Invitrogen-Molecular Probes, Inc., Eugene, OR) were used for immunofluorescence staining. The nuclei were counterstained with DAPI, and the slides were mounted with anti-fade mounting reagent (Invitrogen-Molecular Probes). The slides were viewed using a Zeiss confocal scanning laser microscope using the LSM 5 EXCITER (Carl Zeiss AG, Oberkochen, Germany). We used the mouse anti–α-SMA (Sigma-Aldrich) for Western blot analysis and immunofluorescence staining in in vitro experiments. In in vivo experiments, for double immunofluorescence staining with mouse anti–pan-cytokeratin (Thermo Fisher Scientific), we used rabbit anti–α-SMA (Abcam). Cells were plated onto Labtek slides in M199 medium plus 20% FBS. Cells were washed with PBS and fixed with 3.7% paraformaldehyde for 30 minutes at room temperature. The cells were incubated with Alexa594-phalloidin (Invitrogen-Molecular Probes) diluted in 1% bovine serum albumin–PBS at a final concentration of 5 U/mL for 30 minutes. Cells were counterstained with DAPI. HPMCs in 6-well plates were wounded by means of manual scraping with a yellow pipette tip. Plates were washed twice to remove nonadherent cells and were incubated with serum-free M199 medium for 24 hours. The cells were then transfected with siAQP3 or adenovirus-mediated AQP3 for 24 hours. After transfection, the cells were treated with 10 μg/mL mitomycin C for 1 hour before the scratch to block proliferation. Then the cells were incubated with recombinant human TGF-β1 at 2 ng/mL, and digital images were obtained at 0, 12, and 24 hours. Healing was assessed on 5 parts of the scratch at each time point using the Internet-based software ImageJ to compare wound width with that at time 0. At least 5 regions of wound closure on each scratch were measured in 3 separate experiments, and the results were analyzed statistically. The matrigel invasion assay was performed to assess the effects of AQP3 on the migration of TGF-β1–treated HPMCs. Transwell inserts (12-wells, polycarbonate, 8-μm pore size) from Corning, Inc. (Corning, NY) were coated with 200 μL matrigel (final concentration, 1.0 mg/mL in ice-cold serum-free medium) (BD Biosciences, San Jose, CA) and allowed to dry at 37°C for 3 hours. Growth medium (10% FBS containing 5 μg/mL fibronectin) was added to the lower wells of the chambers. At 24 hours after transfection, cells were washed twice with serum-free medium and trypsinized, and 200 μL cell suspension (1 × 105 cells) from each sample was added to each well. After 24 hours of incubation at 37°C, the cells on the top surface of the chamber were gently removed using cotton swabs. Migrated cells remaining on the bottom surface were stained with Diff-Quik (Fisher Scientific, Pittsburgh, PA) or 0.1% crystal violet (Sigma-Aldrich). The number of cells that had migrated to the lower side of the filter was counted under a light microscope at ×200 magnification in 5 randomly selected fields. HPMCs were treated for 1 hour with 20 μmol/L of either PD98059, SB203580, or SP600125, which are specific chemical inhibitors of ERK, p38, or JNK, respectively. Subsequently, 2 ng/mL TGF-β1 was added, and the cells were incubated for 48 hours. The level of AQP3 protein expression after treatment with mitogen-activated protein kinase (MAPK) inhibitor was assessed using Western blotting. The t-test was used for comparisons where indicated. The results are expressed as mean ± SEM. P < 0.05 was considered significant. Statistical analyses were performed using SigmaPlot software (version 11.0; Systat Software, Inc., San Jose, CA). We first used the rat model of chronic PD to investigate the role of AQP3 in peritoneal injury and fibrosis. Group C rats had a catheter without infusion of dialysis solution, and group D rats had a catheter with infusion of dialysis solution for 8 weeks. After 8 weeks of experimental PD, the peritoneal thickness of the abdominal wall in group D rats was thicker than in group C rats, and increased vascularity was noted in group D rats (data not shown). Also at 8 weeks, the number of dual-stained cytokeratin- and α-SMA–positive myofibroblasts in the submesothelial layer in group D rats had increased, compared with those in group C rats, which was associated with decreased cytokeratin and increased α-SMA expression (Figure 1A). There was strong TGF-β1 expression in the thickened submesothelial matrix of the parietal peritoneum in group D rats (Figure 1B). AQP3 was detected in the peritoneal mesothelial layer in group C rats and in the thickened submesothelial matrix of the parietal peritoneum in group D rats (Figure 1B). Western blot analysis revealed that AQP3 and α-SMA protein expression was significantly higher in the abdominal walls of group D rats than in group C rats (P < 0.05) (Figure 1C). To confirm the HPMC population, we tested it by using immunofluorescence staining of cytokeratin and α-SMA. At confocal microscopy, cytokeratin was expressed in the great majority of confluent cells, whereas α-SMA–positive cells were rare (see Supplemental Figure S1 at http://ajp.amjpathol.org). To investigate the relationship of AQP3 and TGF-β1 in HPMCs, HPMCs were incubated in medium containing 1 to 5 ng/mL TGF-β1 for 48 hours or were treated with 2 ng/mL TGF-β1 for 24, 48, and 72 hours. Protein and mRNA expression levels of α-SMA, a mesenchymal marker, and AQP3 were significantly up-regulated after TGF-β1 treatment. Specifically, TGF-β1 induced α-SMA and AQP3 expression in a dose- and time-dependent manner (Figure 2). RNAi experiments were performed to further examine the role of AQP3 in TGF-β1-treated HPMCs. Human AQP3 siRNA (siAQP3; 20 nmol/L) or nontargeting siRNA (siN; 20 nmol/L) were transfected into HPMCs. At 24 hours after siAQP3 or siN transfection, HPMCs were treated with 2 ng/mL TGF-β1 or were left untreated for 48 hours, and AQP3 mRNA and protein expression were determined by means of real-time RT-PCR and Western blot analysis, respectively. Transfection with siAQP3 decreased AQP3 mRNA expression by 80% compared with control (P < 0.05) (Figure 3A) and also inhibited AQP3 protein expression compared with control (P < 0.05) (Figure 3B). Furthermore, TGF-β1–induced AQP3 mRNA and protein expression levels in HPMCs were attenuated by siAQP3 transfection (Figure 3, A and B). Immunofluorescence staining demonstrated that the AQP3 signals were lower in siAQP3-transfected cells and were attenuated in siAQP3- or TGF-β1–treated cells compared with the appropriate control cells. The morphologic changes induced by TGF-β1 treatment were substantially reduced after siAQP3 transfection. In particular, the cobblestone-like appearance of TGF-β1– or siAQP3-treated cells remained unchanged (Figure 3C). We also performed real-time RT-PCR, Western blot analysis, and indirect immunofluorescence staining to investigate changes in α-SMA expression after siAQP3 transfection. Expression of α-SMA protein and mRNA was significantly decreased compared with the control after siAQP3 transfection (P < 0.05) (Figure 3, A and B). Expression of α-SMA was undetectable at immunofluorescence in untreated HPMCs but was strongly expressed in TGF-β1–treated HPMCs. Furthermore, expression of α-SMA was diminished in HPMCs transfected with siAQP3 and treated with TGF-β1 compared with cells treated only with TGF-β1 (Figure 3C). To investigate the role of AQP3 in HPMC migration, we used a wound-healing assay and a transwell migration assay to investigate HPMCs transfected with siAQP3. HPMCs were treated with mitomycin C for 1 hour before the scratch to prevent proliferation. In the in vitro wound-healing assay, wound closure was analyzed after cells were scraped from a confluent cell monolayer. TGF-β1 accelerated wound closure, whereas siAQP3 inhibited TGF-β1–induced HPMC motility, and as a result, the wound remained open (Figure 4A). Matrigel invasion assays demonstrated remarkable reductions in the invasive properties of HPMCs after treatment with siAQP3. The staining of cells that had migrated through the polycarbonate membrane demonstrated that the number of invasive cells was significantly reduced in HPMCs treated with siAQP3 or TGF-β1 compared with TGF-β1 alone (Figure 4B). To verify whether wound closure in HPMCs is due to TGF-β1–induced proliferation or TGF-β1–induced migration, we performed BrdU incorporation assays (Figure 4C). BrdU incorporation in these cells was inhibited by TGF-β1 and was not changed by siAQP3 (Figure 4C). To determine and optimize the adenoviral infection efficiency in mesothelial cells, we infected HPMCs with Ad-AQP3 at various MOIs for 24 hours. AQP3 mRNA and α-SMA mRNA were significantly increased with increasing Ad-AQP3 in a dose-dependent manner (Figure 5A). Ad-AQP3–transfected HPMCs exhibited more prominent immunofluorescence staining of AQP3 than did control cells (Figure 5B). To investigate the effects of AQP3 overexpression, HPMCs were infected with 100 MOIs of either Ad-GFP or Ad-AQP3 for 24 hours, after which the cells were incubated with 2 ng/mL TGF-β1 or were left untreated for 48 hours. AQP3 mRNA and protein were significantly increased after transfection of Ad-AQP3 compared with control (P < 0.05) (Figure 5). TGF-β1–treated cells exhibited a greater increase in AQP3 protein expression; however, the AQP3 mRNA level was not significantly different in cells transfected with Ad-AQP3 compared with cells subjected to Ad-AQP3 transfection plus TGF-β1 treatment. AQP3 overexpression increased α-SMA protein and mRNA expression (Figure 5, C and D). To examine the consequences of AQP3 expression on stress fiber formation, cells infected with siAQP3 or Ad-AQP3 were stained with phalloidin-Alexa594 conjugate. At phalloidin staining, TGF-β1 stimulated reorganization of F-actin filaments from an apical ringlike structure to stress fiber projections. F-actin stress fiber formation was attenuated in cells treated with TGF-β1 or siAQP3, and increased after transfection of Ad-AQP3 (Figure 5E). To investigate whether AQP3 gene transfer influenced HPMC migration, HPMCs were infected with 100 MOIs of either a control adenovirus expressing Ad-GFP or an adenovirus expressing human AQP3. After Ad-AQP3 transfection, we observed faster migration in AQP3-overexpressing cells than in control cells but not in Ad-GFP–transfected cells (Figure 6, A and B). We could not find a significant difference between TGF-β1– or Ad-AQP3–treated groups and the TGF-β1–treated group on cell migration. Activation of the MAPK pathways after TGF-β1 treatment of HPMCs for 1 hour was confirmed using Western blot analysis of p-ERK1/2, p-JNK, and p-p38, respectively, using p-protein–specific antibodies. Increased levels of p-ERK1/2, p-JNK, and p-p38 were detected after 1 hour of TGF-β1 treatment (Figure 7). TGF-β1 induces a motile phenotype in HPMCs. The motile phenotype may be mediated, at least in part, by the MAPK pathway (Figure 7A). HPMCs were pretreated with specific MAPK inhibitors (each at 20 μmol/L) before treatment with TGF-β1. The basal AQP3 levels in the presence of inhibitor were not significantly different from control. The treatment with an ERK inhibitor (PD98059) and a p38 kinase inhibitor (SB203580) significantly attenuated TGF-β1–induced AQP3 expression in HPMCs. In contrast,
Referência(s)