The signaling protein Wnt5a promotes TGFβ1-mediated macrophage polarization and kidney fibrosis by inducing the transcriptional regulators Yap/Taz

M2 macrophage polarization is known to underlie kidney fibrosis. We previously reported that most of the members of the Wnt family of signaling proteins are induced in fibrotic kidneys. Dysregulation of the signaling protein Wnt5a is associated with fibrosis, but little is known about the role of Wnt5a in regulating M2 macrophage activation that results in kidney fibrosis. Here, using murine Raw 264.7 cells and bone marrow–derived macrophages, we found that Wnt5a enhanced transforming growth factor β1 (TGFβ1)-induced macrophage M2 polarization as well as expression of the transcriptional regulators Yes-associated protein (Yap)/transcriptional coactivator with PDZ-binding motif (Taz). Verteporfin blockade of Yap/Taz inhibited both Wnt5a- and TGFβ1-induced macrophage M2 polarization. In mouse models of kidney fibrosis, shRNA-mediated knockdown of Wnt5a expression diminished kidney fibrosis, macrophage Yap/Taz expression, and M2 polarization. Moreover, genetic ablation of Taz in macrophages attenuated kidney fibrosis and macrophage M2 polarization in mice. Collectively, these results indicate that Wnt5a promotes kidney fibrosis by stimulating Yap/Taz-mediated macrophage M2 polarization.

Monocyte-macrophage lineage may be activated and categorized as either classically activated (M1) or alternatively activated (M2) phenotype in response to various stimuli (1). Proinflammatory M1 macrophages causes acute tissue injury, whereas persistent accumulation of profibrotic M2 macrophages drives the fibrotic response (2,3). During the past decade, great achievement has been made in defining the molecular networks involved in modulating the polarized activation of macrophages (4). However, the mechanisms for regulating macrophage polarization during kidney fibrosis remain to be further understood.
The Wnt family, containing at least 19 Wnt ligands in humans, is classified into canonical and noncanonical pathways based on whether or not ␤-catenin is activated (5,6). Our published studies found that most of the Wnt family members are induced in fibrotic kidney tissue. We also demonstrated that the activation of canonical Wnt signaling can promote podocyte injury and macrophage M2 polarization through distinct molecular mechanisms (7,8). Wnt5a, a ligand that binds Fizzled5 and triggers noncanonical Wnt signaling, is up-regulated in various fibrotic diseases (9 -13). Furthermore, activation of Wnt5a is associated with epithelial-to-mesenchymal transition of tubular epithelial cells during renal fibrosis (13). In addition, many studies have found that Wnt5a can regulate macrophage function through distinct mechanisms (14 -16). However, the role and mechanisms for Wnt5a in regulating macrophage M2 polarization as well as its contribution to kidney fibrosis are obscure.
Like Wnt signaling, Hippo signaling is evolutionarily conserved and pivotal in regulating embryonic development and disease progression (17). Yes-associated protein (Yap) 2 /transcriptional coactivator with PDZ-binding motif (Taz) acts as the target and downstream effector of this pathway. When dephosphorylated, Yap/Taz accumulate and are relocated to the nucleus where it interacts with a number of transcription factors that may promote cell growth, differentiation, and survival (18). Yap and Taz have divergent but critical roles in nephrogenesis (19). Ablation of Taz displays kidney cyst formation accompanied by enhanced Wnt/␤-catenin signaling, suggesting that a cross-talk exists between Hippo and canonical Wnt signaling. Additionally, TGF␤1, a well-known profibrotic cytokine, induces robust Taz but not Yap protein expression in both mesenchymal and epithelial cells. Sustained Taz activation promotes epithelial maladaptive repair, suggesting a potential role of Hippo signaling in promoting kidney fibrosis (20 -23). In addition, it has been reported that Taz regulates reciprocal differentiation of TH17 cells and Treg cells (24). Wnt5a may also induce Yap/Taz activation through stimulating Yap dephosphorylation in bone marrow stromal cells (25). Moreover, blocking Yap/Taz prevents tumor-associated macrophage M2 polarization (26). Therefore, it is highly possible that Wnt5a may promote macrophage M2 polarization and contribute to kidney fibrosis by modulating Yap/Taz activity.
In this study, we found that Wnt5a could exacerbate TGF␤1induced macrophage M2 polarization through Yap/Taz up-regulation. In mouse models with kidney fibrosis, down-regulation of Wnt5a was able to reduce kidney fibrosis, macrophage M2 polarization, and Yap/Taz expression. Specific ablation of Taz gene in macrophages could markedly inhibit macrophage M2 polarization and kidney fibrosis and in mice.

Wnt5a exacerbates TGF␤1-stimulated macrophage alternative activation
Wnt5a is one of the major components of the Wnt family of proteins. It is up-regulated in various fibrotic diseases. Activation of Wnt5a is associated with epithelial-to-mesenchymal transition of tubular epithelial cells during renal fibrosis (13). In addition to tubular epithelial cells, macrophage also plays a very important role in renal fibrosis. To explore the role of Wnt5a in regulating macrophage M2 polarization, we treated Raw 264.7 cells and bone marrow-derived macrophages (BMMs) with TGF␤1 with or without Wnt5a for 24 h. In Raw 264.7 cells, TGF␤1 could largely up-regulate arginase-1 (Arg-1), found in inflammatory zone 1 (Fizz1), and chitinase-like lectin (Ym1) mRNAs but not mannose receptor (MR) expression. Wnt5a plus TGF␤1 could significantly enhance Arg-1, MR, Fizz1, and Ym1 mRNA expression compared with treatment with TGF␤1 alone in both Raw 264.7 cells and BMMs (Fig. 1, A and B). In parallel, Wnt5a could further up-regulate TGF␤1-induced Arg-1 protein expression in both Raw 264.7 cells and BMMs in a time-and dose-dependent manner ( Fig. 1, C, D, F, and G). In addition, we found that Wnt5a could further up-regulate TGF␤1-induced ␣-SMA expression in both Raw 264.7 cells and BMMs (Fig. 1, E and H). Therefore, Wnt5a treatment could exacerbate TGF␤1-stimulated macrophage M2 polarization.

Yap/Taz mediate Wnt5a-exacerbated macrophage M2 polarization
We then wanted to elucidate the mechanisms for Wnt5a in promoting TGF␤1-stimulated macrophage M2 polarization. We first examined ␤-catenin signaling in Raw 264.7 cells. The results showed that Wnt5a could decrease ␤-catenin abundance and inhibit TOPFlash luciferase activity, suggesting the inactivation of ␤-catenin signaling in macrophages (Fig. S1, A and B). We then investigated the phosphorylation status for Stat3 or Stat6 in Wnt5a-treated Raw 264.7 cells. Western blot results showed that TGF␤1 could induce Stat3 phosphorylation at Tyr-705 in a time-dependent manner, whereas Wnt5a could not enhance it (Fig. S1C). Stat6 phosphorylation at Tyr-641 was reported to be involved in macrophage M2 polarization (27). However, it was not induced by treatment with either TGF␤1 alone or TGF␤1 plus Wnt5a (Fig. S1C). and BMMs (B) that were stimulated with TGF␤1 (2 ng/ml) with or without Wnt5a (100 ng/ml) for 24 h. The qRT-PCR data were normalized to a reference gene, Gapdh. *, p Ͻ 0.05 versus cells treated with vehicle (Veh.) alone, n ϭ 3; #, p Ͻ 0.05 versus cells treated with TGF␤1 alone, n ϭ 3. Error bars represent S.E. C-H, Western blot assay showing the abundance for Arg-1 and ␣-SMA in Raw 264.7 cells (C-E) and BMMs (F-H) that were treated with TGF␤1 with or without Wnt5a for 12 or 24 h.

Wnt5a promotes kidney fibrosis
Recently, it was reported that ovatodiolide prevents polarization of M2 tumor-associated macrophages through the Yap oncogenic pathway (26). Park et al. (25) also reported that alternative Wnt signaling could activate Yap/Taz. To investigate whether Wnt5a exacerbates TGF␤1-induced macrophage M2 polarization through Yap/Taz activation, we treated Raw 264.7 cells and BMMs with TGF␤1 with or without Wnt5a. Western blot results showed that TGF␤1 or Wnt5a treatment alone could not obviously increase Yap/Taz abundance, but TGF␤1 plus Wnt5a could largely up-regulate Yap/Taz expression (Fig.  2, A and B). In addition, TGF␤1 plus Wnt5a could promote Yap/Taz to undergo nuclear translocation in BMMs (Fig. 2C). To investigate the role of Yap/Taz induction in Wnt5a-exacerbated macrophage M2 polarization, we treated Raw 264.7 cells with verteporfin, a Yap inhibitor (28), followed by TGF␤1 or TGF␤1 plus Wnt5a treatment to induce macrophage M2 polarization. The results showed that verteporfin could largely inhibit Yap/Taz expression as well as macrophage M2 polarization, suggesting a pivotal role for Yap/Taz activation in modulating TGF␤1 plus Wnt5a-promoted macrophage M2 polarization (Fig. 2, D and E). Moreover, we generated BMMs with Taz ablation and examined macrophage M2 polarization. BMMs isolated from Csf1r-Cre ϩ , Taz fl/fl mice were treated with 4-hydroxytamoxifen for 5 days to induce Taz gene deletion and stimulated with TGF␤1 with or without Wnt5a to induce macrophage M2 polarization (Fig. 2F). Ablation of Taz could markedly inhibit both protein and mRNA expression of Arg-1 stimulated by TGF␤1 plus Wnt5a in BMMs (Fig. 2, F and G). To further explore whether Taz induction is sufficient to enhance TGF␤1-stimulated macrophage M2 polarization, we transfected the BMMs with Taz(S89A) expression plasmid followed by TGF␤1 treatment for 24 h (Fig. 2H). The results showed that expression of exogenous Taz could enhance TGF␤1-stimulated Arg-1 expression, which mimicked the effect of Wnt5a in exacerbating TGF␤1-induced macrophage M2 polarization (Fig.  2I). Together, these results demonstrated that Wnt5a may exacerbate TGF␤1-induced macrophage M2 polarization through up-regulating Yap/Taz expression.

Induction of Wnt5a/b and Yap/Taz in the fibrotic kidneys
To explore the role of Wnt5a in kidney fibrosis, we generated mouse models of kidney fibrosis by unilateral ureter obstruc- Wnt5a promotes kidney fibrosis tion (UUO) and ischemic/reperfusion injury (IRI), respectively. Western blot results showed that Wnt3a protein abundance in UUO kidneys was slightly decreased, whereas it was increased in IRI kidneys. Wnt5a/b protein abundance was largely increased in the fibrotic kidneys after UUO (Fig. 3A) or IRI (Fig.  3B). Immunohistochemical staining showed that Wnt5a/b abundance was increased in the fibrotic area of UUO or IRI kidneys (Fig. 3C). Of note, the abundance of Yap/Taz was also largely elevated in kidneys at different time points as indicated after UUO (Fig. 3A) or IRI (Fig. 3B). In addition, coimmunostaining results showed that, in the UUO or IRI kidneys, a large amount of Yap/Taz was detected to undergo nuclear translocation in F4/80-positive macrophages (Fig. 3C). Thus, these data suggest that Wnt5a/b and Yap/Taz are concurrently up-regulated in the fibrotic kidneys after UUO or IRI in mice.

Down-regulation of Wnt5a attenuates kidney fibrosis in mice
The above data showed that Wnt5a was up-regulated in the fibrotic kidneys. To further explore whether Wnt5a induction promotes kidney fibrosis, we generated a mouse model with short hairpin RNA (shRNA)-mediated knockdown of Wnt5a gene. Western blot and real-time qRT-PCR analyses demonstrated the down-regulation of Wnt5a, whereas Wnt3a protein abundance was not changed as measured by Western blot assay in livers (Fig. 4, A and B) and kidneys (Fig. 4, C and D) from mice injected with Wnt5a shRNA compared with those injected with pcDNA6.2.
We then investigated whether down-regulation of Wnt5a could attenuate kidney fibrosis. A mouse model of kidney fibrosis induced by UUO or IRI was used (Fig. 4, E and F). At day 14 after UUO, marked tubular atrophy and interstitial fibrosis were detected in mice injected with pcDNA6.2, whereas in mice injected with Wnt5a shRNA plasmid, tubular damage and interstitial fibrosis were remarkably attenuated (Fig. 4, G and H). Immunofluorescence staining and Western blotting showed that FN and ␣-SMA expression was remarkably increased in kidneys from mice injected with pcDNA6.2 after UUO, whereas FN and ␣-SMA expression was much less in mice injected with Wnt5a shRNA (Fig. 4, K and L). Similarly, at day 28 after IRI, kidney injury, fibrotic area, total collagen content, and FN and ␣-SMA expression were largely attenuated in mice injected with Wnt5a shRNA compared with those injected with pcDNA6.2 ( Fig. 4, I, J, M, and N). Collectively, we concluded that down-regulation of Wnt5a ameliorates kidney fibrosis after UUO or IRI in mice.

Wnt5a promotes kidney fibrosis Down-regulation of Wnt5a inhibits macrophage accumulation and M2 polarization in the UUO or IRI kidneys
Macrophage accumulation and activation have been reported to play a crucial role in kidney fibrosis. Previous studies also reported that Wnt5a can promote cell migration (29). The above data showed that Wnt5a could stimulate macrophage activation and M2 polarization. We then wanted to know whether Wnt5a down-regulation can decrease macrophage accumulation in the fibrotic kidneys after UUO or IRI. We stained kidney tissues with antibody against F4/80 to identify macrophages. At day 14 after UUO or day 28 after IRI, macrophage accumulation was remarkably increased in the UUO or IRI kidneys from mice injected with pcDNA6.2, whereas it was much less in those from mice injected with Wnt5a shRNA (Fig. 5, A and B). We then sorted macrophages from the UUO or IRI kidneys and examined the abundance for Arg-1 and MR, the markers for M2 polarized macrophages. Western blot results showed that Arg-1 and MR were largely decreased in macrophages from mice injected with Wnt5a shRNA compared with those injected with pcDNA6.2 (Fig. 5, C and E). Moreover, immunofluorescence staining demonstrated that MR was largely decreased in macrophages from mice injected with Wnt5a shRNA after UUO (Fig.  5D) or IRI (Fig. 5F) compared with those injected with pcDNA6.2. Consistently, mRNA abundance for M2 macrophage markers, including Arg-1, MR, Fizz1, and Ym1, was markedly reduced in macrophages from mice injected with Wnt5a shRNA plasmid compared with those injected with pcDNA6.2 after UUO or IRI (Fig. 5, G and H). Flow cytometry analysis for CD11b and CD206 further demonstrated that down-regulation of Wnt5a could inhibit macrophage M2 polarization within the UUO or IRI kidneys (Fig. 5, I and J). Therefore, we conclude that knockdown of Wnt5a diminishes macrophage M2 polarization in the kidneys after UUO or IRI.

Down-regulation of Wnt5a reduces Yap/Taz expression in macrophages from the fibrotic kidneys
In cultured macrophages, we found that Wnt5a plus TGF␤1 could up-regulate Yap/Taz expression, which mediates Wnt5a-exacerbated macrophage M2 polarization. We then detected Yap/Taz expression in macrophages from UUO or IRI kidneys. A Western blot assay showed that Yap/ Taz expression was induced in macrophages from mice injected with pcDNA6.2, whereas it was much less in mice injected with Wnt5a shRNA after UUO or IRI (Fig. 6, A and C). In addition, the mRNA abundance for Yap/Taz target genes, including Ankrd1 and Ctgf, in macrophages was largely elevated from mice injected with pcDNA6.2 but was much less in those from mice injected with Wnt5a shRNA after UUO or IRI (Fig. 6, B and D). Immunofluorescence staining for Yap/Taz and F4/80 further confirmed the reduction of Yap/Taz in macrophages from mice injected with Wnt5a shRNA compared with those injected with pcDNA6.2 after UUO (Fig. 6E). Therefore, these results suggest that knockdown of Wnt5a may down-regulate macrophage Yap/Taz expression and diminish macrophage M2 polarization in the kidneys after UUO or IRI.

Ablation of Taz in macrophages attenuates macrophage M2 polarization and kidney fibrosis in mice with UUO nephropathy
To further decipher the role and mechanisms for Taz induction in macrophage M2 polarization and kidney fibrosis, we generated a mouse model with inducible macrophage Taz deletion with a Cre-LoxP system (Fig. 7, A and B). Mice with macrophage ablation of Taz were generated by intraperitoneal injection of tamoxifen for 5 consecutive days in Csf1r-Cre ϩ , Taz fl/fl mice and named as Mac-Taz Ϫ/Ϫ . The same gender with genotype Csf1r-Cre Ϫ , Taz fl/fl littermates were injected with tamoxifen and named as Mac-Taz ϩ/ϩ (Fig. 7C). Western blot analysis demonstrated the ablation of Taz in macrophages from Mac-Taz Ϫ/Ϫ mice (Fig. 7, D and E). Kidney histology was comparable between the knockouts and control littermates, whereas in the UUO kidneys from the knockouts, interstitial fibrotic area and total collagen content were much less compared with their control littermates (Fig. 7, F and G). Moreover, FN or ␣-SMA abundance was largely decreased in the UUO kidneys from Mac-Taz Ϫ/Ϫ kidneys compared with that from Mac-Taz ϩ/ϩ kidneys (Fig. 7, H and I).
We then wanted to know whether ablation of Taz inhibits macrophage accumulation and M2 polarization in the UUO kidneys. The number of F4/80-positive macrophages was largely decreased in the UUO kidneys from mice with macrophage Taz deletion compared with their control littermates (Fig. 8A). The mRNA abundance for Arg-1, Fizz1, and chitinase 3-like 3/Ym1 was largely increased in macrophages from Mac-Taz ϩ/ϩ kidneys after UUO but was much less in macrophages from Mac-Taz Ϫ/Ϫ kidneys (Fig. 8B).

Wnt5a promotes kidney fibrosis
Macrophages from the fibrotic kidneys were sorted with CD115 microbeads, and Western blot analysis showed that Arg-1 abundance in macrophages enriched from Mac-Taz Ϫ/Ϫ UUO kidneys was much less compared with those from Mac-Taz ϩ/ϩ mice (Fig. 8C). Additionally, immuno-staining results showed that deletion of Taz in macrophages could inhibit macrophage M2 polarization (Fig. 8D). Therefore, it can be concluded that ablation of Taz in macrophages attenuates macrophage accumulation, M2 polarization, and kidney fibrosis after UUO in mice.

Discussion
In this study, we report that Wnt5a could exacerbate TGF␤1induced macrophage M2 polarization and contribute to kidney fibrosis. Additionally, we demonstrate that Yap/Taz induction mediated Wnt5a plus TGF␤1-promoted macrophage M2 polarization.
The Wnt family members are crucial players in regulating cellular and organ function. Previous studies demonstrated that Wnt/␤-catenin signaling activation promotes kidney fibrosis through stimulating fibroblast activation and macrophage M2 polarization (8,30). Among all the Wnt family members, Wnt5a is a pleiotropic cytokine that regulates the development of various organs and postnatal cellular function. Wnt5a may promote inflammation and fibrosis in multiple organs and tissues (31). In this study, we report that Wnt5a is crucial for promoting kidney fibrosis. Wnt5a has been reported to be involved in cell migration, invasion, and polarity (32). Under proinflammatory conditions, Wnt5a induces immunosuppressive macrophage activation (33). In this study, we found that down-regulation of Wnt5a could attenuate macrophage accumulation and M2 polarization. In cultured macrophages, Wnt5a could enhance TGF␤1-induced macrophage M2 polarization. Therefore, we conclude that Wnt5a may promote kidney fibrosis through stimulating macrophage activation and M2 polarization.
Alternative (M2) macrophage activation is known to be driven by multiple cytokines and transcriptional factors (26, 34 -36). Our published studies reported that Wnt/␤-catenin signaling modulates macrophage M2 polarization through Stat3 activation (8). Although Wnt5a is considered as a noncanonical Wnt family member, studies have shown that it can activate ␤-catenin in dendritic cells as well as several other cell types (37). Here, we found that Wnt5a could down-regulate ␤-catenin expression and inhibit ␤-catenin signaling activation in cultured macrophages, which suggests that ␤-catenin signaling may not mediate Wnt5a-exacerbated macrophage M2 polarization. Stat3 and Stat6 signaling are two major pathways for mediating macrophage M2 polarization. In this study, we found that TGF␤1 alone could stimulate Stat3 but not Stat6 phosphorylation, whereas Wnt5a plus TGF␤1 could not enhance Stat3 or Stat6 phosphorylation. Thus, we can conclude that Wnt5a exacerbates TGF␤1-stimulated macrophage M2 polarization through Yap/Taz induction based on the following reasons. First, Wnt5a plus TGF␤1 could largely up-regulate Yap/Taz expression in both Raw 264.7 cells and BMMs. Second, suppression of Yap/Taz activation could markedly inhibit Wnt5a plus TGF␤1-stimulated macrophage M2 polarization. Third, overexpression of exogenous Taz could mimic the effect of Wnt5a on exacerbating TGF␤1-stimulated macrophage M2 polarization. Fourth, in animal models, down-regulating Wnt5a expression resulted in less macrophage Yap/Taz induction or macrophage M2 polarization.
Yap and Taz are transcriptional coregulators that bind primarily to enhancer elements by using Tead factors as DNA-

Wnt5a promotes kidney fibrosis
binding platforms (38). The Hippo pathway and its essential effectors Yap and Taz are necessary for fibroblast activation and tissue fibrosis (39 -42). In the UUO model, Yap/Taz display nuclear translocation, and verteporfin, a chemical that inhibits Yap-Tead interaction and Yap transcriptional activity, reduces renal fibrosis (23). In this study, we found that ablation of Taz in macrophages could inhibit Wnt5a plus TGF␤1-induced macrophage M2 polarization. Moreover, inducible ablation of

Wnt5a promotes kidney fibrosis
Taz in macrophages could largely inhibit macrophage M2 polarization and kidney fibrosis after UUO or IRI. Thus, it can be concluded that Wnt5a and TGF␤1 may promote macrophage M2 polarization and kidney fibrosis through converging into Yap/Taz induction. It is of note that the previous studies reported that Yap and Taz bind Smad2/3 through the coiledcoil region, and this interaction may dictate the subcellular localization of Smad2/3 (43,44). However, more investigation is still needed to decipher the mechanisms for Yap/Taz activation in regulating TGF␤1-stimulated macrophage M2 polarization.
In summary, our study demonstrated that Wnt5a promotes macrophage M2 polarization via regulating Yap/Taz activity and contributes to kidney fibrosis after UUO or IRI in mice. Targeting Wnt5a/Yap/Taz in macrophages may represent a new therapeutic strategy for protecting against kidney fibrosis in patients with chronic kidney disease.

Mice and animal models
Male C57BL/6 mice weighing ϳ18 -20 g were acquired from the specific pathogen-free laboratory animal center of Nanjing Medical University. Mice were sacrificed, and kidneys were harvested at different time points after UUO. For the kidney IRI model, the left renal pedicle of the mouse was clamped for 35 min. The right kidneys were not removed. Mice were sacrificed, and kidneys were harvested at different time points after IRI.
Mice expressing tamoxifen-inducible MerCreMer fusion protein under the control of macrophage-specific mouse Csf1r promoter (019098; FVB-Tg(Csf1r-cre/Esr1*)) were ordered from The Jackson Laboratory (Bar Harbor, ME). FVB-Tg(Csf1rcre/Esr1*) mice were crossed with C57BL/6J mice for eight generations to obtain Csf1r-Cre transgenic mice on a C57BL/6J background. Homozygous Taz floxed mice were kindly pro-vided by Dr. Randy L. Johnson from MD Anderson Cancer Center. By mating Taz floxed mice with Csf1r-Cre/Esr1* transgenic mice, mice that were heterozygous for the Taz floxed allele were generated (genotype, Csf1r-Cre ϩ/Ϫ , Taz fl/wt ). These mice were cross-bred with homozygous Taz floxed mice (genotype, Taz fl/fl ) to generate offspring with different littermates (Csf1r-Cre ϩ/Ϫ , Taz fl/fl ; Csf1r-Cre ϩ/Ϫ , Taz fl/wt ; Csf1r-Cre Ϫ/Ϫ , Taz fl/wt ; and Csf1r-Cre Ϫ/Ϫ , Taz fl/fl ). Mice with genotype Csf1r-Cre ϩ/Ϫ , Taz fl/fl and the same gender littermates with genotype Csf1r-Cre Ϫ/Ϫ , Taz fl/fl were used in the study. Genotyping was performed by PCR assay using DNA extracted from the mouse tail. Csf1r-Cre ϩ/Ϫ , Taz fl/fl mice and Csf1r-Cre Ϫ/Ϫ , Taz fl/fl control littermates were intraperitoneally injected with tamoxifen (T5648, Sigma-Aldrich) at 25 mg/kg for 5 consecutive days, and 2 days after the last injection, the mice were subjected to UUO operation. All animals were housed in the specific pathogenfree laboratory animal center of Nanjing Medical University according to the guidelines of the Institutional Animal Care and Use Committee at Nanjing Medical University.

Plasmids
shRNA specific for mouse Wnt5a gene was ordered from Ruizhen, Nanjing, China. BLOCK-iT Pol II miR RNAi Expression Vector kits were used (catalog number K4935, Invitrogen). The target sequence of the murine Wnt5a is GAGTTCGTG-GACGCTAGAGAA. The mouse Wnt5a-specific shRNA was cloned into pcDNA6.2. Mice on the C57BL/6J background were injected with Wnt5a shRNA in the tail vein at 1 mg/kg to induce endogenous Wnt5a down-regulation. For the UUO model, Wnt5a shRNA plasmid was injected at 1 day before the surgery and 7 days after the surgery. For the IRI model, Wnt5a shRNA plasmid was injected at 7 and 14 days after the surgery. 3XFlag pCMV5-TOPO TAZ(S89A) was a gift from Jeff Wrana (Addgene plasmid 24815) (43). BMMs were seeded on Wnt5a promotes kidney fibrosis 12-well culture plates to 90 -95% confluence in complete medium containing 10% fetal bovine serum and 10 ng/ml macrophage colony-stimulating factor. BMMs were transfected with pTAZ or empty vector using Lipofectamine 2000 (Invitrogen) according to the manufacturer's instruction.

Real-time qRT-PCR assay
Total RNA was extracted using TRIzol reagent (catalog number 15596018, Invitrogen) according to the manufacturer's instructions. cDNA was synthesized using 1 g of total RNA, ReverTra Ace (catalog number R111-02, Vazyme, Nanjing, China), and oligo(dT) [12][13][14][15][16][17][18] primers according to the manufacturer's protocol. Gene expression was measured by real-time qRT-PCR (catalog number Q141-02, Vazyme) and a 7300 realtime PCR system (Applied Biosystems, Foster City, CA). Gapdh was detected as an internal control. The primers are listed in Table 1. The relative amount of mRNA or gene to internal control was calculated using the equation 2⌬CT in which ⌬CT ϭ CT gene Ϫ CT control .

Histology and immunohistochemical staining
Mouse kidney sample were fixed in 10% neutral formalin and embedded in paraffin. 3-m-thick sections were stained with periodic acid-Schiff, Masson's trichrome, and Sirius Red. Paraffin-embedded kidney sections were deparaffinized, hydrated, and subjected to antigen retrieval, and endogenous peroxidase activity was quenched by 3% H 2 O 2 . Sections were then blocked with 10% normal donkey serum followed by incubating with anti-Wnt5a/b (catalog number 2530, Cell Signaling Technol-ogy) at 4°C overnight. After incubation with secondary antibody for 1 h at room temperature, sections were then incubated with ABC reagents for 1 h at room temperature before being subjected to the substrate 3-amino-9-ethylcarbazole for staining (Vector Laboratories, Burlingame, CA). Slides were viewed under a Nikon Eclipse 80i microscope equipped with a digital camera (DS-Ri1, Nikon, Shanghai, China).

Immunofluorescence staining
Kidney cryosections at 3-m thickness were fixed for 15 min with 4% paraformaldehyde followed by permeabilization with 0.2% Triton X-100 in 1ϫ PBS for 5 min at room temperature. After blocking with 2% donkey serum for 60 min, the slides were immunostained with the following antibodies: anti-FN (catalog number F3648, Sigma-Aldrich), anti-␣-SMA (catalog number A5228, Sigma-Aldrich), anti-F4/80 (catalog number 14-4801, eBioscience, San Diego, CA), anti-MR (catalog number ab64693, Abcam), and anti-Yap/Taz (catalog number 8418, Cell Signaling Technology). BMMs cultured on coverslips were washed with cold 1ϫ PBS and fixed with cold methanol:acetone (1:1) for 10 min at Ϫ20°C. After three extensive washings with 1ϫ PBS, the cells were treated with 0.1% Triton X-100 for 5 min, blocked with 2% normal donkey serum in 1ϫ PBS buffer for 40 min at room temperature, and incubated with anti-Yap/ Taz (catalog number 8418, Cell Signaling Technology) followed by staining with FITC-conjugated secondary antibodies. Cells were also stained with 4Ј,6-diamidino-2-phenylindole (DAPI) to visualize the nuclei. Slides were viewed with a Nikon Eclipse 80i epifluorescence microscope equipped with a digital camera. The F4/80-positive macrophage number was counted from 10 randomly selected fields in the cortical area for each sample under 400ϫ magnification, and an average number of positive cells for each section was calculated.

Kidney monocyte/macrophage enrichment
After perfusion with cold 1ϫ PBS, kidneys were removed, minced into fragments, and then digested in Dulbecco's modified Eagle's medium containing 1 mg/ml collagenase (catalog number 17018-029, Gibco) and 0.1 mg/ml DNase (catalog number 10104159001, Roche Applied Science) for 1 h at 37°C with intermittent agitation. The fragments were filtered through 40-m mesh (Falcon, BD Biosciences) to obtain a single-cell suspension. Macrophages were enriched from the single-cell suspension with CD115 microbeads and a MACS

Statistical analyses
All data examined are presented as mean Ϯ S.E. Statistical analyses of the data were performed using SigmaStat software (Jandel Scientific Software, San Rafael, CA). Comparison between groups was made using one-way analysis of variance followed by the Student-Newman-Keuls test. Paired or unpaired t test was used to compare two groups. A p value of 0.05 or lower was considered statistically significant.