Hedgehog-Gli Pathway Activation during Kidney Fibrosis - Core

The American Journal of Pathology, Vol. 180, No. 4, April 2012 Copyright © 2012 American Society for Investigative Pathology. Published by Elsevier Inc. All rights reserved. DOI: 10.1016/j.ajpath.2011.12.039

Cardiovascular, Pulmonary, and Renal Pathology

Hedgehog-Gli Pathway Activation during Kidney Fibrosis

Steven L. Fabian,* Radostin R. Penchev,* Benoit St-Jacques,† Anjali N. Rao,* Petra Sipilä,‡ Kip A. West,§ Andrew P. McMahon,¶储** and Benjamin D. Humphreys*¶ From the Renal Division,* Brigham and Women’s Hospital, Department of Medicine, Harvard Medical School, Boston, Massachusetts; the Institute for Research in Immunology and Cancer,† University of Montreal, Montreal, Quebec, Canada; the Department of Physiology,‡ Institute of Biomedicine, University of Turku, Finland; Infinity Pharmaceuticals,§ Cambridge, Massachusetts; the Harvard Stem Cell Institute,¶ Holyoke Center, Cambridge, Massachusetts; and the Departments of Stem Cell and Regenerative Biology,储 and Molecular and Cellular Biology,ⴱⴱ Harvard University, Cambridge, Massachusetts

The Hedgehog (Hh) signaling pathway regulates tissue patterning during development, including patterning and growth of limbs and face, but whether Hh signaling plays a role in adult kidney remains undefined. In this study, using a panel of hedgehog-reporter mice, we show that the two Hh ligands (Indian hedgehog and sonic hedgehog ligands) are expressed in tubular epithelial cells. We report that the Hh effectors (Gli1 and Gli2) are expressed exclusively in adjacent platelet-derived growth factor receptor-␤positive interstitial pericytes and perivascular fibroblasts, suggesting a paracrine signaling loop. In two models of renal fibrosis, Indian Hh ligand was upregulated with a dramatic activation of downstream Gli effector expression. Hh-responsive Gli1-positive interstitial cells underwent 11-fold proliferative expansion during fibrosis, and both Gli1- and Gli2-positive cells differentiated into ␣-smooth muscle actinpositive myofibroblasts. In the pericyte-like cell line 10T1/2, hedgehog ligand triggered cell proliferation, suggesting a possible role for this pathway in the regulation of cell cycle progression of myofibroblast progenitors during the development of renal fibrosis. The hedgehog antagonist IPI-926 abolished Gli1 induction in vivo but did not decrease kidney fibrosis. However, the transcriptional induction of Gli2 was unaffected by IPI-926, suggesting the existence of smoothened-independent Gli activation in this

model. This study is the first detailed description of paracrine hedgehog signaling in adult kidney, which indicates a possible role for hedgehog-Gli signaling in fibrotic chronic kidney disease. (Am J Pathol 2012, 180: 1441–1453; DOI: 10.1016/j.ajpath.2011.12.039)

The Hedgehog (Hh) signaling pathway plays a crucial role in regulating a diverse range of developmental processes in the mammalian embryo, including ventralization of the neural tube, patterning and growth of limbs and face, the formation of organs (such as the lung and gut), development of hair follicles, and decisions of leftright asymmetry.1,2 In the kidney, sonic hedgehog (Shh) expression in papillary collecting duct and ureteric epithelium regulates adjacent mesenchymal cell proliferation and differentiation, and either germline Shh deletion or deletion of Shh from collecting duct leads to severe renal developmental abnormalities, including renal aplasia or hypoplasia.3–5 Mutations affecting the Hh signaling member Gli3 in humans with Pallister-Hall syndrome are associated with renal malformation, further implicating Hh in human renal morphogenesis.6,7 Three Hh ligands are found in mice and humans: 1) Shh, 2) desert hedgehog (Dhh), and 3) Indian hedgehog (Ihh) ligands.1 These secreted, lipid-modified proteins can act at short or long distances by binding to the membrane receptor Patched1 (Ptch1) on target cells, thereby releasing tonic inhibition by Ptch1 on the transmembrane protein smoothened (Smo). Derepressed Smo translocates to the primary cilium, inhibiting produc-

Supported by the NIH (RO1 DK088923 to B.D.H. and T32 training grant DK007527 to S.L.F.) and the Harvard Stem Cell Institute (B.D.H). Accepted for publication December 29, 2011. Disclosures: K.A.W. holds stock options or bond holdings in a for-profit corporation or self-directed pension plan (Infinity Pharmaceuticals) and is also a full-time employee of Infinity Pharmaceuticals. None of the other authors disclosed any conflict of interest. Supplemental material for this article can be found on http://ajp. amjpathol.org or at doi: 10.1016/j.ajpath.2011.12.039. Address reprint requests to Benjamin D. Humphreys, M.D., Ph.D., Harvard Institutes of Medicine Room 554, 4 Blackfan Circle, Boston, MA 02115. E-mail: [email protected].

1441

1442 Fabian et al AJP April 2012, Vol. 180, No. 4

tion of the truncated repressor forms of the Gli2 and Gli3 transcription factors and promoting preservation of their full-length activator forms, which induce transcription of Hh target genes, including Gli1 and Ptch1, both of which serve as readouts of Hh pathway activation.8 Hh signaling has multiple, context-dependent downstream effects, such as controlling expression of patterning genes (ie, Pax2 and Sall1) or regulating cell cycle by activating Cyclin D1 and N-Myc.5 Little is known about a role for the Hh pathway in the adult kidney. In cancer and solid organ injury models, recent evidence suggests that epithelial-derived Hh ligands can be reactivated in pathological states to transmit signals to surrounding mesenchymal cells. For example, in carcinogenesis, Hh ligands from the epithelial tumor act on adjacent stroma to promote a favorable tumor microenvironment.9 –11 In murine bladder injury, epithelial Shh induces Wnt expression in surrounding stromal cells, which in turn stimulates stromal and epithelial proliferation in a paracrine signaling loop.12 Hh pathway reactivation has also been implicated in organ fibrosis. Both chronic cholestasis and nonalcoholic steatohepatitis are characterized by increased Hh signaling during fibrosis,13,14 and Hh signaling promotes activation of hepatic stellate cells to the myofibroblastic phenotype.15 In lung fibrosis, Shh is upregulated in airway epithelial cells, and Ptch1 expression is increased in the pulmonary interstitium.16 Collectively, these observations suggest that mesenchymal cells may be targets of Hh signaling in pathological states, just as they are in development. Myofibroblasts (the scar forming cells in fibrosis) derive from mesenchymal progenitors in the kidney,17,18 and because of this, we hypothesized that the Hh pathway would be activated in these cells during renal fibrogenesis. Using complementary techniques, including a variety of genetic reporter mice, we demonstrate that the Hh ligands (Ihh and Shh) are expressed in tubular epithelial cells of the kidney, whereas the Hh effectors (Gli1 and Gli2) are expressed in platelet-derived growth factor receptor-␤ (PDGFR-␤)-expressing interstitial pericytes and perivascular fibroblasts. Both Ihh expression and downstream Hh signaling were substantially activated during renal fibrosis, as Hh-responsive pericytes and perivascular fibroblasts proliferated and differentiated into myofibroblasts. Hh ligand drove cell proliferation in a pericyte-like cell line, suggesting that epithelial-derived Hh ligands might direct mesenchymal cell proliferation during renal fibrosis. Pharmacological inhibition of Smo completely suppressed Gli1 induction, but it did not inhibit fibrosis, suggesting that Gli2, whose induction was not inhibited, may be the more important Gli effector in renal fibrosis.

Materials and Methods Animal Experiments All mouse studies were performed according to the animal experimental guidelines issued by the Animal Care and Use Committee at Harvard University. Wild-type

mice were from Charles River Laboratories (Wilmington, MA); FVB/N mice were used for unilateral ureteral obstruction (UUO) and C57BL/6 mice were used for unilateral ischemia reperfusion injury (UIRI) time course experiments and quantitative PCR studies. Ptch1-nLacZ (JAX Stock 003081), Gli1-nLacZ (JAX Stock 008211), Gli2nLacZ (JAX stock 007922), Shh-GFPCre (JAX Stock 005622), and R26-LacZ knock-in mice were purchased from Jackson Laboratories (Bar Harbor, ME). To create Ihh-nLacZ reporter mice, an Ihh-nLacZ reporter allele was constructed and the Ihh locus targeted in the embryonic stem cells, replacing most of the first exon of Ihh with an NLS-LacZ-pA cassette (see Supplemental Figure S1at http://ajp.amjpathol.org). Mice of 8 to 12 weeks were anesthetized with pentobarbital sodium (60 mg/kg body weight) before surgery, and body temperatures were controlled at 36.5 to 37.5°C throughout all procedures. Each time point represents three to five mice as indicated. For UUO, the left kidney was exposed through a flank incision and the left ureter tied off at the level of the lower pole with two 4.0 silk ties. Mice were sacrificed 3 to 14 days after obstruction. For UIRI, the left kidney was exposed through a flank incision, and the renal pedicle was clamped with nontraumatic microaneurysm clamps (Roboz Surgical Instrument Co., Gaithersburg, MD), which were removed after 28 minutes. Reperfusion was visually verified. Two hours after surgery, 1 mL of 0.9% NaCl intraperitoneally was administered.

Tissue Preparation and Histology Mice were anesthetized, euthanized, and immediately perfused via the left ventricle with ice-cold PBS for 1 minute. Kidneys were either snap frozen or fixed in 4% paraformaldehyde on ice for 2 hours, then incubated in 30% sucrose in PBS at 4°C overnight. OCT-embedded (Sakura Finetek, Torrance, CA) kidneys were cryosectioned into 7-␮m sections. LacZ activity was measured on paraformaldehydefixed frozen sections by standard 5-bromo-4-chloro-3-indolyl-␤-D-galactopyranoside (X-gal) staining for 1 to 6 days at 37°C, and counterstained with nuclear fast red and mounted. To quantify nLacZ cell number, 100⫻ images were taken of the entire cortex (Gli1 and Ptch1), the inner cortex (Ihh) or cortex and medulla (Gli2) of the midsagittal kidney sections containing papilla from at least four different animals; the number of positive cells were then counted in each 100⫻ image using a manual cell counter from ImageJ (http://rsbweb.nih.gov/ij). Primary antibodies included rabbit anti-␤-galactosidase (MP Biomedicals, Solon, OH, Cat. #55976, 1:2000), chicken anti-green fluorescent protein (Aves Labs, Tigard, OR, Cat #GFP-1020, 1: 500), rat anti-PDGFR-␤ (eBioscience, San Diego, CA, Cat. #14–1402, 1:500), Cy3conjugated ␣-smooth muscle actin (␣-SMA) (Sigma-Aldrich, St. Louis, MO, Cat. #C6198, 1:500), rat anti-F4/80 (Abcam, Cambridge, MA, Cat. #ab6640, 1:500), fluorescein isothiocyanate (FITC)-Lotus tetragonolobus lectin (Vector Labs, Burlingame, CA, Cat. #FL-1321, 1:500), rabbit anti-CD31 (Abcam, Cat. #ab28364, 1:500), rabbit anti-aquaporin 2 (Abcam, Cat. #ab15116, 1:500), fluorescein isothiocyanate (FITC)-conjugated Dolichos biflorus agglutinin (Vector Labs, Cat. # FL1031, 1:500), and rabbit anti-NKCC2 (Alpha Diagnostic,

Hedgehog Pathway in Renal Fibrosis 1443 AJP April 2012, Vol. 180, No. 4

San Antonio, TX, Cat. #NKCC21-A, 1:100). Secondary antibodies were either FITC or Cy3-conjugated (Jackson ImmunoResearch, West Grove, PA) incubated for 30 minutes, with DAPI nuclear counterstain followed by mounting in ProlongGold (Invitrogen, Carlsbad, CA). Images were obtained by confocal (Nikon C1 eclipse, Nikon, Melville, NY) or standard microscopy (Nikon eclipse 90i, Nikon). Anti-LacZ antibodies reliably labeled LacZ-expressing interstitial cells, although the autofluorescence in tubular epithelia blunted their sensitivity in tubular epithelial cells. Therefore, in certain situations X-gal staining followed by indirect immunofluorescence was performed with pseudocoloring of the X-gal stain.

quantified by Bradford Assay and 25 ␮g separated by 10% polyacrylamide gel electrophoresis. Proteins were transferred to polyvinylidene difluoride membrane, blocked in 5% milk in phosphate buffered saline tween 20, probed overnight at 4°C with goat anti-Shh-N antibody (Santa Cruz, Cat. #sc-1194, 1:200) or mouse anti-␣-SMA (Sigma-Aldrich, Cat. #A2547, 1:5000), or for 1 hour at room temperature with mouse anti-glyceraldehydes-3-phosphate dehydrogenase (Abcam, Cat. #ab9484, 1:5000), washed, probed with antigoat- or mouse-horseradish peroxidase (Dako, Carpinteria, CA, 1:5000) for 1 hour at room temperature, and the antigen antibody complex was visualized using the ECL detection system (PerkinElmer, Waltham, MA).

Cell Culture Experiments 10T1/2 cells (ATCC) were grown in Basal Media Eagle (Gibco, Billings, MT) with 10% fetal bovine serum supplemented with penicillin and streptomycin and 2 mmol/L glutamine. Shh conditioned media was generated from supernatants of Cos7 cells stably transfected with pcDNA3-N-Shh or pcDNA3 control plasmid. For propidium iodide cell cycle analysis and Bromodeoxyuridine (BrdU) uptake cell proliferation assays, cells were grown on 6 well plates, serum starved by incubating in 0.5% fetal bovine serum for 12 hours, and then stimulated for 24 hours with either Shh conditioned media, Cos7 control media, 500 nmol/L smoothened agonist (SAG) (Santa Cruz, Santa Cruz, CA, Cat. #sc-202814) or water control in 0.5% or 10% fetal bovine serum. For the BrdU uptake assay, the cells were incubated in 10 ␮m BrdU for 2 hours before harvesting and then stained using the BrdUFITC flow kit (BD Pharmingen, San Diego, CA). For the cell cycle analysis, cells were fixed in ice-cold 100% ethanol, incubated with propidium iodide (400 ␮g/mL propidium iodide, 2 mmol/L MgCl2, 100 mg/mL RNase), and subject to fluorescence-activated cell sorting (FACS) analysis.

IPI-926 Experiments IPI-926 (5 mg/mL) stock solution (Infinity Pharmaceuticals, Cambridge, MA) was prepared fresh for each experiment by dissolving in hydroxyproplyl-␤-cyclodextrin (vehicle) and sonicating. Mice were given 40 mg/kg body weight IPI-926 or vehicle by gastric lavage daily until the day before sacrifice, with Gli1-LacZ mice receiving their first dose the day before UUO surgery and being sacrificed on day 7 of UUO, and BALB/c and C57BL/6 mice receiving their first dose 2 days before UUO surgery and being sacrificed on day 10 of UUO.

Western Blot To confirm the presence of Shh in conditioned media by Western blot, 5 ␮L of conditioned media was first separated by 10% polyacrylamide gel electrophoresis. To determine the relative amount of ␣-SMA protein in kidneys from IPI926- versus vehicle-treated mice, the lower kidney pole from UUO and contralateral (CLK) kidneys were homogenized in radioimmunoprecipitation assay buffer with protease inhibitors using a handheld rotor, the total protein

Real-Time PCR RNA was extracted from snap-frozen tissue stored at ⫺80°C or cells using standard techniques (RNeasy, Qiagen, Germantown, MD). Reverse transcription was performed with the iScript cDNA synthesis kit (Bio-Rad, Hercules, CA) producing cDNA. Real-time PCR was performed using iQ-SYBR Green supermix (BioRad) and the iQ5 Multicolor Real-Time PCR Detection system (BioRad) for detection of mRNA levels. Glyceraldehydes-3phosphate dehydrogenase was used as the internal control.

Statistical Analysis Statistical analyses were performed using Graph Pad Prism software (version 5.0) (GraphPad Software, Inc., San Diego, CA). Analysis of variance was used to compare data among groups followed by a Tukey’s post test to compare all groups to each other or a Dunnett’s post test to compare all groups to the control group. A twotailed Student’s t-test was used when only two groups were being compared. All results were repeated at least twice. A P value of less than 0.05 was considered significant. The results are presented as mean ⫾ SEM.

Results Hh Ligands, Shh, and Ihh are Expressed in Tubular Epithelial Cells To define the expression pattern of Hh pathway members in renal fibrosis, we used available Ptch1-nLacZ, Gli1nLacZ, and Gli2-nLacZ reporter mice, and generated Ihh-nLacZ knockin reporter mice. Because Shh-GFPCre reporter mice exhibited unexpectedly low green fluorescent protein fluorescence, historical Shh expression was assessed in Shh-GFPCre; R26-LacZ bigenic mice, in which cytoplasmic LacZ expression marks cells that either actively express Shh or expressed Shh at one time in development (see Supplemental Figure S2 at http:// ajp.amjpathol.org). LacZ expression in kidney sections from Shh-GFPCre; R26-LacZ adult mice was present exclusively in the papilla, corresponding to in situ hybridization staining of Shh mRNA in P1 kidney (Figure 1A), as well as ureteral urothelium (see Supplemental Figure S3


Figure 1. Expression pattern of the Hedgehog (Hh) pathway in the mouse kidney. Reporter mice sonic hedgehog (Shh)-GFPCre; R26-LacZ Ihh-nLacZ, Ptch1-nLacZ, Gli1-nLacZ, and Gli2-nLacZ adult mice), in situ hybridization, and quantitative PCR were used to measure and localize the Hh pathway expression. In situ results are from P1 kidney; reporter mouse and quantitative PCR are from the adult. A: Shh is observed only in the papilla (p) and ureter (arrow). Ihh is expressed in inner cortex and medulla (m), and Ptch1 is expressed in cortex (c), medulla, and papilla. Both Indian hedgehog (Ihh) and Patched1 (Ptch1) exhibit prominent staining at the corticomedullary junction. B: Gli1 and Gli2 are expressed in cortex, medulla, and papilla though Gli1, similar to Ihh and Ptch1, was highest at the corticomedullary junction, whereas Gli2 was highest in the inner medulla and papilla. C–E: Quantitative PCR confirms the differences in regional expression for all Hh pathway members: Shh, Ihh, desert hedgehog (Dhh) (C); Ptch1 (D); Gli1, GLi2, Gli3 (E). N ⫽ 3 mice throughout. *P ⬍ 0.05, **P ⬍ 0.005, and ***P ⬍ 0.001 by analysis of variance followed by Tukey’s post test. Dotted lines mark corticomedullary and medullopapillary junctions. Scale bar ⫽ 50 ␮m.

at http://ajp.amjpathol.org) as expected.3 LacZ expression in Shh-GFPCre; R26-LacZ kidneys was localized to aquaporin 2 positive collecting ducts (Figure 2A; see also Supplemental Figure S4 at http://ajp.amjpathol.org). We generated Ihh-nLacZ knockin mice to report Ihh expression (see Supplemental Figure S1 at http://ajp. amjpathol.org). Ihh was expressed predominantly in the inner cortex and outer medulla at the corticomedullary junction, with reduced expression seen throughout the rest of the medulla (Figure 1A). In situ hybridization in P1 mouse kidneys confirmed staining in the outer medulla (Figure 1A), consistent with previous findings during mouse development.3 As with Shh, Ihh was exclusively expressed in tubular epithelial cells (Figure 2B). Most Ihh-nLacZ tubular cells in the inner cortex and outer medulla co-stained with the proximal tubular marker Lotus tetragonolobus lectin (Figure 2B; see also Supplemental Figure S4 at http://ajp.amjpathol. org), consistent with a previous report of Ihh expression in dissected proximal tubules by real-time PCR.19 In addition, occasional Ihh-nLacZ was observed in thin limbs of Henle (see Supplemental Figure S5A at http://ajp.amjpathol.org), demonstrating Ihh expression of tubular epithelial cells with squamous morphology lacking brush borders. These cells did not costain with collecting duct markers aquaporin 2 or

Dilochus biflorus agglutinin, the thick ascending limb marker Na-K-2Cl cotransporter or the endothelial marker CD31 (see Supplemental Figure S5B at http://ajp.amjpathol. org). Relative mRNA expression as determined by quantitative PCR from dissected kidney cortex, medulla, and papilla confirmed that Shh is the most highly expressed Hh ligand in the papilla, and Ihh is the most highly expressed ligand in the medulla and cortex. Dhh expression was minimal (Figure 1C).

The Hh Receptor Ptch1 and Downstream Effectors Gli1 and Gli2 are Expressed in Interstitial Pericytes and Perivascular Fibroblasts To define the cell types that respond to Hh ligand, we examined the expression patterns of Ptch1 and Gli effectors in the adult kidney. Ptch1 and Gli1 are readouts of Hh pathway activity, and their expression defines Hh-responsive cells. Gli2 lies directly upstream of Gli1 and other Hh transcriptional targets.1 Ptch1 (Figure 1A) and Gli1 (Figure 1B) were expressed strongly at the corticomedullary junction, suggesting that these cells may be responding to Ihh in that region, whereas Gli2 (Figure 1B)


Figure 2. Hedgehog (Hh) ligands are expressed in tubular epithelial cells, and intersititial cells respond to Hh ligands. High-powered images of LacZ expression and in situ hybridization reveal that sonic hedgehog (Shh) and Indian hedgehog (Ihh) expression is restricted to tubular epithelial cells. A: 5-bromo-4-chloro-3-indolyl-␤-D-galactopyranoside (X-gal) staining of ShhGFPCre; R26-LacZ kidney with subsequent immunofluorescent detection of aquaporin 2 (AQP2) demonstrates that Shh is expressed in the collecting duct. B: In the outer medulla, Ihh-positive cells have a cuboidal morphology (arrows) and co-stain with the proximal tubular marker Lotus tetragonolobus lectin (LTL). C: Patched1 (Ptch1)-nLacZ expression is present in the glomerulus (g), tubules (t), endothelium (arrows), and interstitium (arrowheads), whereas Gli1-nLacZ and Gli2-nLacZ expression was restricted to interstitial cells. Ptch1, Gli1 and Gli2-positive cells were enriched in the perivascular space around arteries and arterioles (lower three panels). Scale bar ⫽ 50 ␮m. nLacZ, nuclear lacZ.

was expressed most prominently in the inner medulla and papilla. Ptch1 and a lesser amount of Gli1 expression was observed in the inner medulla and papilla as well, likely in response to Ihh in the inner medulla and Shh in the papilla. In situ studies of Ptch1 in P1 kidney sections were consistent with Ptch1-nLacZ expression in adult mice (Figure 1A) and embryonic kidney.20 Ptch1 was also expressed in occasional tubular epithelial cells, glomerular cells, and endothelial cells, in addition to interstitial cells (Figure 2C; see also Supplemental Figure S6 at http://ajp.amjpathol.org). In contrast, Gli1 and Gli2 were exclusively expressed in interstitial cells in the adult kidney (Figure 2C). Though there has been a prior report of Gli1

expression in tubules, especially in the setting of decreased transcriptional repressor Glis2,21 we did not observe X-gal staining of tubular epithelial cells using our Gli1-nLacZ reporter mouse, even in kidneys from newborn and 7-day-old mice (Figure 3A). We did, however, observe X-gal staining of epithelial cells in the ureteric bud in the nephrogenic zone in kidneys from Gli2-nLacZ newborn mice that was decreased in kidneys from 7-day-old mice and almost completely absent in kidneys from 14-day-old mice (Figure 3B). A higher density of Ptch1, Gli1, and Gli2-positive interstitial cells were observed closely associated with vessels (Figure 2C, bottom three panels). Quantitative mRNA comparisons confirmed that Ptch1 and Gli2 were most prominently ex-

Figure 3. Expression of Gli2, but not Gli1, occurs in the epithelial cells of the ureteric bud in the nephrogenic zone of developing kidneys. A: 5-bromo-4-chloro-3-indolyl-␤D-galactopyranoside (X-gal) staining of kidneys from Gli1-nLacZ mice is exclusively interstitial and does not reveal staining of ureteric bud epithelium (arrow) in P0 and P7 mice. B: X-gal staining of kidneys from Gli2-nLacZ mice reveals staining of ureteric bud epithelium (arrows) in the outer cortex in newborn mice (P0). X-gal staining in epithelial cells is decreased (arrows) in kidneys from 7-day-old mice (P7) and is almost completely absent in kidneys from 14-day-old mice (P14). Scale bar ⫽ 50 ␮m. nLacz, nuclear LacZ.


Figure 4. The Hedgehog (Hh) pathway is activated in two different models of kidney fibrosis. A–D: Corticomedullary lysates were obtained on days 3, 7, and 14 after unilateral ureteral obstruction (UUO) or sham surgery, and mRNA expression was normalized to glyceraldehydes-3-phosphate dehydrogenase (GADPH). Collagen 1␣1 (Col1␣1) and ␣-smooth muscle actin (␣-SMA) (A) increased during UUO, confirming fibrosis. Patched1 (Ptch1) (B) increased slightly by day 14 of UUO. Gli1 and Gli3 levels (C) progressively rose throughout UUO, and Gli2 levels increased on days 7 and 14. Indian hedgehog (Ihh) (D) significantly increased after UUO, peaking at day 3, whereas sonic hedgehog (Shh) and desert hedgehog (Dhh) were minimally expressed. N ⫽ 5 UUO and 5 sham mice per time point. E–H: Medullary lysates for fibrotic markers, Gli1, Gli2, Gli3, and Ptch1, and cortical lysates for Ihh were obtained on days 3, 7, 14, and 28 after ischemia reperfusion injury (UIRI), or day 3 after sham surgery. mRNA expression was normalized to GAPDH. Similar to the peak increase of Col1␣1 and ␣-SMA at day 7 (E), Gli1, Gli2, Gli3 (G), Ihh (H), and to a lesser degree Ptch1 (F), increase after UIRI with a peak at day 7. Gli1 also exhibits a second peak on day 28. N ⫽ 4 mice for UIRI days 1, 3, 14, and 28, and N ⫽ 3 mice for UIRI day 7 and sham. *P ⬍ 0.05, **P ⬍ 0.005, and ***P ⬍ 0.001 by analysis of variance followed by Dunnett’s post test.

pressed in the medulla and papilla, and Gli1 mRNA was highest in the medulla (Figure 1, D and E). Gli3 was also highest in the medulla and papilla (Figure 1E), and was expressed the highest overall when comparing the three Gli effectors in kidney.

Hh Pathway Upregulation in Renal Fibrosis To investigate Hh pathway modulation during renal fibrosis, we measured mRNA expression of Hh pathway members in corticomedullary kidney lysates from adult mice after 3, 7, and 14 days of chronic injury by UUO compared to sham controls. Expression of the fibrotic marker Collagen 1␣1 (Col1␣1) and the myofibroblast marker ␣-SMA progressively increased relative to sham, confirming fibrosis (Figure 4A). A progressive increase in Gli1 and Gli3 mRNA expression occurred on days 3, 7, and 14 of UUO, and a progressive increase in Gli2 mRNA expression occurred on days 7 and 14 (Figure 4C). Gli1 and Gli3 demonstrated a more

robust induction relative to sham with a 13.6 ⫾ 4.3-fold increase in Gli1 and a 15.2 ⫾ 5.7-fold increase in Gli3 on day 14 versus a 3.5 ⫾ 1.9-fold increase in Gli2. Gli1 transcription reflects active Hh signaling, and because of this, the results indicate that the Hh pathway is activated during renal fibrosis. Ptch1 expression also increased (Figure 4B), although only by 2.1 ⫾ 0.4-fold, perhaps reflecting its stronger baseline expression compared to Gli1. Next, we asked which Hh ligand might account for increased Hh signaling. Shh expression did not change during UUO, although Ihh was induced transcriptionally, peaking at day 3 with a 4.5 ⫾ 0.5 fold increase and remained elevated thereafter (Figure 4D). A similar 3.4 ⫾ 0.8 increase in Ihh mRNA at UUO day 3 was observed in a second independent experiment (N ⫽ 5). Dhh was also increased relative to sham at all time points, although the absolute amount of Dhh was extremely low -35.5 ⫾ 3.7fold lower than Ihh mRNA levels, indicating that Ihh is the primary Hh ligand induced by chronic renal injury. To


Figure 5. During unilateral ureteral obstruction (UUO) the number of Gli1, Gli2, and Patched1 (Ptch1)-expressing cells increases. A: Low-power images of cortex from 5-bromo-4-chloro-3-indolyl-␤-D-galactopyranoside (X-gal) staining of kidney sections from Gli1-nLacZ mice show a progressive increase in X-gal staining with increasing duration of UUO. B: Higher-power images of boxed areas from A demonstrate that Gli-nLacZ-positive cells remain exclusively interstitial. C: Quantification of the number of Gli1-nLacZ-positive cells in the kidney cortex shows a substantial increase during UUO. D: X-gal staining for Gli2 and Ptch1 shows an increase in the number of X-gal-positive interstitial cells in 7-day UUO kidneys compared to uninjured kidneys. X-gal staining for Indian hedgehog (Ihh) expression remains restricted to tubules in both uninjured and UUO kidneys. E: The number of Gli2-expressing cells per 200⫻ field of view increases in the cortex and medulla, although the increase in the cortex is not as dramatic as that seen for Gli1. Ptch1-expressing cells per 100⫻ field of view of the cortex increase in the interstitium, but not the tubules. Ihh-nLacZ mice do not show an increase in the number of X-gal-positive cells per 100⫻ field of view in the cortex during UUO. N ⫽ 4 mice per time point for Gli1 and Ptch1, and N ⫽ 3 mice per time point for Gli2 and Ihh. Scale bar ⫽ 50 ␮m. *P ⬍ 0.05, **P ⬍ 0.005, and ***P ⬍ 0.001 by analysis of variance followed by Dunnett’s post test. CLK, contralateral kidney; nLacZ, nuclear LacZ.

address the generalizability of these findings, we investigated a second model of renal fibrosis, unilateral ischemia reperfusion injury. UIRI has been validated as a model of renal fibrosis in previous reports22 and a dramatic increase in ␣-SMA immunofluorescent staining in UIRI day 14 kidneys compared to CLK provided further confirmation that a robust fibrotic response was achieved (see Supplemental Figure S7 at http://ajp.amjpathol.org). In this model, Ptch1, Gli1, Gli2, and Gli3 mRNA were all significantly increased in medullary kidney lysates relative to sham, with peak levels observed on day 7 in parallel with the peak increase in expression of Col1␣1 and ␣-SMA (Figure 4, E–G). Medullary lysates did not show an increase in Ihh (data not shown), although Ihh was increased at all time points in cortical lysates with a

peak increase of 3.9 ⫾ 0.3 at day 7 (Figure 4H). Hh pathway, therefore, is activated in two separate mouse models of kidney fibrosis.

Myofibroblasts Respond to Hh Signals during Fibrosis To further define the cells that respond to Hh ligands, we quantitated tubular versus interstitial expression of Gli1, Gli2, and Ptch1 during UUO. Gli1 and Gli2 remained exclusively expressed in the interstitium in UUO kidneys without detectable tubular expression (Figure 5, B and D). Compared to uninjured kidneys, cortical Gli1-nLacZ cells increased by 4.1 ⫾ 1.1-fold at 3 days, 10.5 ⫾


1.8-fold at 7 days, and 10.7 ⫾ 0.8-fold at 14 days after UUO (Figure 5, A–C). The number of LacZ-expressing cells in Gli2-nLacZ mice increased as well, but to a lesser degree, with only a 1.7 ⫾ 0.3 increase in the cortex and

3.9 ⫾ 0.5 increase in the medulla (Figure 5, D and E; see also Supplemental Figure S8A at http://ajp.amjpathol.org). There was a 1.9 ⫾ 0.5-fold decrease in the number of Ptch1-nLacZ tubular epithelial cells, but there was a 4.1

Figure 6. Gli1, Gli2, and Patched1 (Ptch1)-positive interstitial cells co-express the pericyte marker platelet-derived growth factor receptor-␤ (PDGFR-␤) and acquire expression of the myofibroblast marker ␣-smooth muscle actin (␣-SMA) in unilateral ureteral obstruction (UUO) kidneys. To determine which interstitial cell-types express Gli1, Gli2, and Ptch1 in uninjured and injured kidneys, day 7 UUO and contralateral (CLK) kidney sections from Gli1-nLacZ, Gli2-nLacZ, and Ptch1-nLacZ mice were immunofluorescently co-labeled for nLacZ and various cell-type specific markers and pictures were obtained from the inner renal cortex. nLacZ in interstitial cells from all three reporter mice colocalized with the pericyte marker PDGFR-␤ in both CLK (arrows) and UUO kidneys. nLacZ in interstitial cells for all three reporter mice colocalized with the myofibroblast marker ␣-SMA in the UUO kidneys. Gli1, Gli2, and Ptch1-expressing cells often come into close contact with cells expressing the macrophage marker F4/80 and endothelial marker CD31, although the vast majority of these cells do not co-express these markers (insets) for CLK and UUO. nLacZ, nuclear LacZ.


⫾ 0.6-fold increase in the number of Ptch1-nLacZ interstitial cells (Figure 5, D and E; see also Supplemental Figure S8B at http://ajp.amjpathol.org). In contrast with the transcriptional induction of Ihh observed during renal fibrosis, there was no increase in the number of IhhnLacZ cells in UUO. Ihh-nLacZ expression remained localized to tubular epithelial cells in the inner cortex and outer medulla after UUO (Figure 5, D and E; see also Supplemental Figure S8C at http://ajp.amjpathol.org). Thus, the increase in Ihh mRNA expression was not due to an increase in the number of Ihh expressing cells at the level of sensitivity of the Ihh-nLacZ reporter. During development, epithelial-derived Hh regulates mesenchymal proliferation and differentiation; we therefore sought to more precisely define the interstitial cell type that was responding to Hh signals and asked whether these cells were proliferating during renal fibrosis. A protocol for detection of nuclear LacZ by immunofluorescence was developed for this purpose. Gli1nLacZ-positive cells uniformly co-expressed the pericyte and perivascular fibroblast marker PDGFR-␤ in both uninjured and injured kidneys (Figure 6). In the fibrotic but not uninjured kidney, Gli1-nLacZ-positive cells also ac-

quire the myofibroblast marker ␣-SMA (Figure 6). Macrophages and endothelial cells were often closely opposed to Gli1-nLacZ-positive cells; there was, however, no overlap in the Gli1 expression domain among either of these cell types (Figure 6). The close association between Gli1nLacZ-positive cells and endothelial cells is consistent with the possibility that some or all of these cells are pericytes. Gli2-nLacZ and Ptch1-nLacZ also colocalized with PDGFR-␤ in uninjured and injured kidneys, with the majority of them co-expressing the myofibroblast marker ␣-SMA during injury, but not the macrophage marker F4/80 or endothelial marker CD31 (Figure 6). To investigate the correlation between Gli1 expression and cell proliferation in UUO, Gli1-nLacZ expressing cells were costained with the cell cycle marker Ki-67. Ki-67positive cells were observed in both tubules and in the interstitium on day 3 of UUO. The percentage of Gli1nLacZ positive cells that were co-stained for Ki-67 was 12.6 ⫾ 1.2% compared to only 1.3 ⫾ 0.4% in uninjured kidneys (Figure 7, A and B). These results indicate that many Hh-responsive cells are proliferating in the early stages of renal fibrosis.

Figure 7. Gli1-expressing cells undergo increased proliferation early in unilateral ureteral obstruction (UUO), and stimulation of a pericyte-like cell line with Hedgehog (Hh) agonists induces cell proliferation in vitro. A: To determine whether proliferation of Gli1-expressing cells increases during UUO, sections from days 0, 3, and 7 UUO kidneys from Gli1-nLacZ mice were co-labeled for nLacZ and Ki-67. B: The percentage of nLacZ-positive cells co-staining for Ki-67 was determined by dividing the number of nLacZ⫹; Ki-67⫹ double positive cells (arrows) by the total number of nLacZpositive cells in the outer medulla. Scale bar ⫽ 25 ␮m. N ⫽ 3 mice for days 0 and 7; N ⫽ 4 mice for day 3. A greater percentage of Gli1-nLacZ cells costained with Ki-67 at day 3 compared to days 0 and 7. C: Stimulation of 10T1/2 cells with sonic hedgehog (Shh) ligand or smoothened agonist (SAG) induces a large upregulation of Gli1 as assessed by quantitative PCR. Platelet-derived growth factor (PDGF) and transforming growth factor (TGF)-␤ had no effect. D, E: 10T1/2 cells labeled with propidium iodine and analyzed by fluorescence-activated cell sorting after stimulation with either Shh (D), SAG (E), or 10% fetal bovine serum (FBS) show an increase in the percentage of cells in S⫹G2M and a decrease in the percentage of cells in G1 compared to Cos7 control media for Shh and water vehicle control for SAG. F, G: The percentage of 10T1/2 cells positive for 5-bromo-2-deoxyuridine (BrdU) uptake increases with stimulation with Shh (F), SAG (G), or 10% FBS compared to Cos7 control media or vehicle control. N ⫽ 3 per condition. *P ⬍ 0.05, ** P ⬍ 0.005, and ***P ⬍ 0.001 by analysis of variance followed by Dunnett’s post test (B and D–G) and by two-tailed Student’s t-test (C). nLacZ, nuclear LacZ.


Exogenous Hh Ligand Drives Cell Cycle Progression in Pericyte-Like Cells Next we asked whether Hh ligand could directly induce proliferation of pericyte-like cells in vitro. The mouse mesenchymal cell line 10T1/2 is hedgehog-responsive and multipotent,23 and it can be induced to differentiate into ␣-SMA⫹ mature pericytes by transforming growth factor␤.24 Kidney pericytes are ␣-SMA negative but gain ␣-SMA expression as they differentiate into myofibroblasts during fibrosis,25 so we reasoned that 10T1/2 cells might be a good model for uninjured kidney pericytes. The presence of Shh in conditioned media from Cos7 cells stably transfected with pcDNA-N-Shh was confirmed by Western blot (see Supplemental Figure S9A at http://ajp.amjpathol.org). Then we confirmed that the media containing Shh activates Gli1 expression in these cells26,27 by 153.9 ⫾ 8.2-fold under our conditions. Consistent with this, the Smo agonist SAG induced a 107.5 ⫾ 6.2-fold increase in Gli1 gene expression (Figure 7C). Gli2 and Gli3 were only minimally affected (see Supplemental Figure S9B at http://ajp.amjpathol.org). Neither platelet-derived growth factor nor transforming growth factor-␤, both increased in UUO, induced Gli1 expression (Figure 7C). Although 10T1/2 cells have been used to model Hh-induced differentiation, the effect of Hh agonists on cell proliferation in these cells has not been reported. Hh pathway activation either with Shh or SAG induced proliferation of serum-starved 10T1/2 pericytes, as assessed by cell cycle analysis (Figure 7, D and E). In confirmation of these results, Shh and SAG also stimulated BrdU uptake as quantitated by FACS analysis (Figure 7, F and G). These in vitro results suggested that Hh could drive pericyte proliferation during fibrotic injury and are consistent with prior reports that Hh signaling can regulate proliferation of mouse and human mesenchymal cells in vitro.2,28

Pharmacological Inhibition of Hh Signaling Does Not Attenuate Fibrosis We next investigated the functional role of kidney Hh signaling in vivo by pharmacological inhibition. Cyclopamine is a well-characterized Smo inhibitor, buts its use in vivo is limited by its short half life29 and off-target effects at higher doses.30,31 We, therefore, used the cyclopamine derivative IPI-926, which has the advantages of a long half-life, increased potency, and oral bioavailability.32 IPI-926 nearly completely abolished Gli1 induction after 7 days of UUO, as reflected by the expression of Gli1-nLacZ (Figure 8, A and B). The efficacy of IPI-926 in inhibiting Hh signaling was further confirmed by quantitative PCR from day 10 UUO corticomedullary kidney extracts from BALB/c mice; the increase in Gli1 mRNA expression seen in UUO kidneys from the vehicle-treated mice was completely suppressed, and a decrease in the CLK controls was also seen. Importantly, the increase in Gli2 mRNA seen in UUO was not suppressed by IPI-926, suggesting that the increase in Gli2 in this setting is not smoothened dependent. Despite complete inhibition of

Figure 8. IPI-926 inhibits Hedgehog (Hh) signaling in unilateral ureteral obstruction (UUO) kidneys. A: To determine whether IPI-926 inhibits Hh signaling, IPI-926 (40 mg/kg) (N ⫽ 5) or vehicle (N ⫽ 5) was given daily by gastric lavage for 8 days to Gli1-nLacZ mice starting the day before UUO. Mice were sacrificed on UUO day 7. B: The increase in cortical 5-bromo-4chloro-3-indolyl-␤-D-galactopyranoside (X-gal) staining seen in UUO kidney sections from vehicle-treated Gli1-nLacZ mice compared to contralateral kidney (CLK) is completely abolished in IPI-926-treated mice. C: mRNA expression by quantitative PCR of Gli1, Gli2, Gli3, and Patched1 (Ptch1) from UUO and CLK kidney lysates from BALB/c mice treated in a blinded fashion with IPI-926 (N ⫽ 6) or vehicle (N ⫽ 8) shows inhibition of Gli1 expression in CLK and UUO. The increase in Gli2 and Gli3 expression in UUO compared to CLK was not inhibited by IPI-926. BALB/c mice were sacrificed on UUO day 10 and treated daily with IPI-926 40 mg/kg or vehicle for 12 days starting 2 days before surgery. *P ⬍ 0.05, **P ⬍ 0.005, and ***P ⬍ 0.001 by two-tailed Student’s t-test.

Gli1 by IPI-926, there was no decrease in renal fibrosis, as assessed by change in Col1␣1, fibronectin, or ␣-SMA gene transcription, or ␣-SMA protein levels by Western blot at UUO day 10 (Figure 9, A and B). In a blinded assessment


Figure 9. IPI-926 does not decrease renal fibrosis in unilateral ureteral obstruction (UUO). A: mRNA expression by quantitative PCR of fibrotic markers in day 10 UUO corticomedullary kidney lysates from BALB/c mice did not decrease with 12 days of IPI-926 treatment (N ⫽ 6) initiated 2 days before surgery compared to vehicle treatment (N ⫽ 8). B: The increase in ␣-smooth muscle actin (␣-SMA) protein expression in UUO compared to contralateral kidney (CLK) as determined by Western blot and quantified by integrated optical density (IOD) ␣-SMA/IOD glyceraldehydes-3-phosphate dehydrogenase (GAPDH) was not inhibited by IPI-926-treatment. C and D: Trichome stains (C) of UUO day 10 kidney sections from C57BL/6 mice and blindly scored (D) for interstitial fibrosis/tublar atrophy percentage (IF/TA%) by a pathologist did not show a difference between IPI-926-treated (N ⫽ 6) and vehicle-treated groups (N ⫽ 7). C57BL/6 mice also received daily IPI-926 (40 mg/kg) or vehicle for 12 days beginning 2 days before UUO. ns, nonsignificant by two-tailed Student’s t-test.

of interstitial fibrosis/tubular atrophy percentage by trichome stain at UUO day 10, there also showed no difference between IPI-926 and vehicle-treated groups. These experiments establish that Gli1 induction in this model is mediated by Hh ligand, but Gli1 does not mediate renal fibrosis in this model.

Discussion Activation of canonical Hh signaling in mesenchymal cells during tissue injury has been recently observed in the bladder, liver, and lung.12,13,16 That scar-forming myofibroblasts derive from mesenchymal progenitors in the kidney,17,25 supporting the hypothesis tested here that Hh-Gli signaling is reactivated in renal fibrosis and that myofibroblasts and their progenitors responds to Hh ligands. These findings also support the general concept that kidney injury responses often reactivate developmental signaling pathways,33 such as the Wnt,34 Notch,35 and fibroblast growth factor pathways.36

Our results confirm that in the uninjured kidney, Ihhproducing cells are localized to outer medullary tubular epithelia and Shh expression is restricted to papillary collecting duct.3,19 Most Ihh-producing cells were in proximal tubule, with some expression in thin limbs of Henle. Expression of Ptch1 and Gli1 is strongest in medullary stroma during development4 and consistent with this, their expression was strongest in the outer medulla of the adult kidney. Ihh induction drives Ptch1 and Gli1 expression in cortex and medulla during fibrosis, because it is expressed in adjacent tubular epithelium, and because Gli1 induction was completely inhibited by the Smo inhibitor IPI-926. The epithelial localization of both Ihh and Shh in the kidney, combined with our demonstration of stromal expression of Gli1 and Gli2 in renal interstitium, indicates that Hh is acting in a paracrine fashion in kidney fibrosis, as it does during renal development.3,20 We observed transcriptional induction of Ihh in renal fibrosis but nontranscriptional mechanisms may also contribute to Hh pathway activation in target cells. Re-


lease of pre-formed Hh ligand has been recently reported to occur from peripheral nerves in skin,37 and whether such a mechanism operates in the kidney remains to be tested. Smo inhibition did not decrease fibrosis, although redundant pathways for myofibroblast proliferation may exist in this model. Equally important, although Smo inhibition inhibited Gli1 induction, it did not suppress Gli2 induction. Gli1 and Gli2 can have redundant roles themselves,38 and Gli1 is dispensable for many Hh effector functions.39 Our results, therefore, indicate that Gli2 might be the more important Gli effector in renal fibrosis. Recently, evidence indicates that other signaling pathways may sensitize target cells to Hh ligand40 or induce ligand-independent, noncanonical Hh pathway activation. Both the RAS-RAF-MEK and PI3K/AKT pathways can potentiate Gli1 function or activate Gli signaling independent of Smo,40 – 42 and both of these pathways are implicated in renal myofibroblast activation.43– 45 Transforming growth factor-␤, whose critical role in renal fibrosis is well described,46 can also activate Gli2 expression independent of Ptch1/Smo in human fibroblasts47 and in cancer.48 Whether noncanonical, Smo-independent Gli activation occurs in kidney fibrosis, and defining the extent to which other more established pro-fibrotic pathways might modulate Hh-Gli signaling in the adult kidney are critical questions that require further investigation. The functional role of Hh-Gli signaling in renal pericytes, perivascular fibroblasts, and myofibroblasts in vivo remains to be defined. Our in vitro evidence suggested the hypothesis that Hh signaling might contribute to mesenchymal cell proliferation during injury, consistent with its known role in regulating ureteral stromal cell proliferation during development. Our in vivo data, however, do not support this model. Other roles for Hh signaling in renal injury responses are also possible. Hh can drive pro-angiogenic signaling in mesenchymal cells after injury49 or during carcinogenesis.50 Whether Hh-mediated pro-angiogenic signaling might occur in either acute or chronic injury is an interesting possibility because angiogenic signals are important in both diseases.51,52 Another question raised by these studies is why Gli1, Gli2, and Ptch1 are expressed in only some myofibroblasts. Are the Hh-responsive pericytes and perivascular fibroblasts different from their neighboring stromal cells? A growing literature documents Hh pathway activation in mesenchymal stem cell biology,28 and Hh is classically known as a stem cell-promoting factor.37,53 In the future it will be important to define possible functional differences between Gli1-positive and Gli1-negative interstitial cells. Finally, strong evidence implicates cortical Gli3 repressor function in regulating ureteric tip gene expression and patterning during renal development.20 The activation of Hh signaling in cortex that we report here suggests that the balance of Gli activator and repressor forms may be altered during kidney injury. In summary, we demonstrate, for the first time, strong activation of the Hh-Gli pathway during renal fibrosis. Chronic injury induces Ihh expression, which acts in a paracrine fashion on interstitial pericytes and perivascular fibroblasts to activate Gli effector expression. These findings define the Hh-Gli pathway as a novel

developmental signaling pathway that is strongly upregulated in renal fibrosis. Future studies are required to define the functional roles of Gli effector proteins in kidney fibrosis.

Acknowledgments We thank Derek DiRocco for help with the unilateral ischemia reperfusion experiments and Vanesa Bijol for help scoring fibrosis severity.

References 1. Ingham PW, McMahon AP: Hedgehog signaling in animal development: paradigms and principles. Genes Dev 2001, 15:3059 –3087 2. Mao J, Kim BM, Rajurkar M, Shivdasani RA, McMahon AP: Hedgehog signaling controls mesenchymal growth in the developing mammalian digestive tract. Development 2010, 137:1721–1729 3. Yu J, Carroll TJ, McMahon AP: Sonic hedgehog regulates proliferation and differentiation of mesenchymal cells in the mouse metanephric kidney. Development 2002, 129:5301–5312 4. Cain JE, Rosenblum ND: Control of mammalian kidney development by the Hedgehog signaling pathway. Pediatr Nephrol 2011, 26:1365– 1371 5. Hu MC, Mo R, Bhella S, Wilson CW, Chuang PT, Hui CC, Rosenblum ND: GLI3-dependent transcriptional repression of Gli1. Gli2 and kidney patterning genes disrupts renal morphogenesis. Development 2006, 133:569 –578 6. Kang S, Graham JM Jr, Olney AH, Biesecker LG: GLI3 frameshift mutations cause autosomal dominant Pallister-Hall syndrome. Nat Genet 1997, 15:266 –268 7. Bose J, Grotewold L, Ruther U: Pallister-Hall syndrome phenotype in mice mutant for Gli3. Hum Mol Genet 2002, 11:1129 –1135 8. Sasaki H, Nishizaki Y, Hui C, Nakafuku M, Kondoh H: Regulation of Gli2 and Gli3 activities by an amino-terminal repression domain: implication of Gli2 and Gli3 as primary mediators of Shh signaling. Development 1999, 126:3915–3924 9. Tian H, Callahan CA, DuPree KJ, Darbonne WC, Ahn CP, Scales SJ, de Sauvage FJ: Hedgehog signaling is restricted to the stromal compartment during pancreatic carcinogenesis. Proc Natl Acad Sci U S A 2009, 106:4254 – 4259 10. Yauch RL, Gould SE, Scales SJ, Tang T, Tian H, Ahn CP, Marshall D, Fu L, Januario T, Kallop D, Nannini-Pepe M, Kotkow K, Marsters JC, Rubin LL, de Sauvage FJ: A paracrine requirement for hedgehog signalling in cancer. Nature 2008, 455:406 – 410 11. Olive KP, Jacobetz MA, Davidson CJ, Gopinathan A, McIntyre D, Honess D, Madhu B, Goldgraben MA, Caldwell ME, Allard D, Frese KK, Denicola G, Feig C, Combs C, Winter SP, Ireland-Zecchini H, Reichelt S, Howat WJ, Chang A, Dhara M, Wang L, Ruckert F, Grutzmann R, Pilarsky C, Izeradjene K, Hingorani SR, Huang P, Davies SE, Plunkett W, Egorin M, Hruban RH, Whitebread N, McGovern K, Adams J, Iacobuzio-Donahue C, Griffiths J, Tuveson DA: Inhibition of Hedgehog signaling enhances delivery of chemotherapy in a mouse model of pancreatic cancer. Science 2009, 324:1457–1461 12. Shin K, Lee J, Guo N, Kim J, Lim A, Qu L, Mysorekar IU, Beachy PA: Hedgehog/Wnt feedback supports regenerative proliferation of epithelial stem cells in bladder. Nature 2011, 472:110 –114 13. Omenetti A, Porrello A, Jung Y, Yang L, Popov Y, Choi SS, Witek RP, Alpini G, Venter J, Vandongen HM, Syn WK, Baroni GS, Benedetti A, Schuppan D, Diehl AM: Hedgehog signaling regulates epithelialmesenchymal transition during biliary fibrosis in rodents and humans. J Clin Invest 2008, 118:3331–3342 14. Syn WK, Choi SS, Liaskou E, Karaca GF, Agboola KM, Oo YH, Mi Z, Pereira TA, Zdanowicz M, Malladi P, Chen Y, Moylan C, Jung Y, Bhattacharya SD, Teaberry V, Omenetti A, Abdelmalek MF, Guy CD, Adams DH, Kuo PC, Michelotti GA, Whitington PF, Diehl AM: Osteopontin is induced by hedgehog pathway activation and promotes fibrosis progression in nonalcoholic steatohepatitis. Hepatology 2011, 53:106 –115


15. Choi SS, Omenetti A, Witek RP, Moylan CA, Syn WK, Jung Y, Yang L, Sudan DL, Sicklick JK, Michelotti GA, Rojkind M, Diehl AM: Hedgehog pathway activation and epithelial-to-mesenchymal transitions during myofibroblastic transformation of rat hepatic cells in culture and cirrhosis. Am J Physiol Gastrointest Liver Physiol 2009, 297: G1093–1106 16. Stewart GA, Hoyne GF, Ahmad SA, Jarman E, Wallace WA, Harrison DJ, Haslett C, Lamb JR, Howie SE: Expression of the developmental Sonic hedgehog (Shh) signalling pathway is up-regulated in chronic lung fibrosis and the Shh receptor patched 1 is present in circulating T lymphocytes. J Pathol 2003, 199:488 – 495 17. Humphreys BD, Lin SL, Kobayashi A, Hudson TE, Nowlin BT, Bonventre JV, Valerius MT, McMahon AP, Duffield JS: Fate tracing reveals the pericyte and not epithelial origin of myofibroblasts in kidney fibrosis. Am J Pathol 2010, 176:85–97 18. Kriz W, Kaissling B, Le Hir M: Epithelial-mesenchymal transition (EMT) in kidney fibrosis: fact or fantasy? J Clin Invest 2011, 121:468 – 474 19. Valentini RP, Brookhiser WT, Park J, Yang T, Briggs J, Dressler G, Holzman LB: Post-translational processing and renal expression of mouse Indian hedgehog. J Biol Chem 1997, 272:8466 – 8473 20. Cain JE, Islam E, Haxho F, Chen L, Bridgewater D, Nieuwenhuis E, Hui CC, Rosenblum ND: GLI3 repressor controls nephron number via regulation of Wnt11 and Ret in ureteric tip cells. PLoS ONE 2009, 4:e7313 21. Li B, Rauhauser AA, Dai J, Sakthivel R, Igarashi P, Jetten AM, Attanasio M: Increased hedgehog signaling in postnatal kidney results in aberrant activation of nephron developmental programs. Hum Mol Genet 2011, 20:4155– 4166 22. Yang L, Besschetnova TY, Brooks CR, Shah JV, Bonventre JV: Epithelial cell cycle arrest in G2/M mediates kidney fibrosis after injury. Nat Med 2010, 16:535–543 23. Taylor SM, Jones PA: Multiple new phenotypes induced in 10T1/2 and 3T3 cells treated with 5-azacytidine. Cell 1979, 17:771–779 24. Hirschi KK, Rohovsky SA, D’Amore PA: PDGF. TGF-beta, and heterotypic cell-cell interactions mediate endothelial cell-induced recruitment of 10T1/2 cells and their differentiation to a smooth muscle fate. J Cell Biol 1998, 141:805– 814 25. Grgic I, Duffield JS, Humphreys BD: The origin of interstitial myofibroblasts in chronic kidney disease. Pediatr Nephrol 2012, 27:183– 193 26. Williams KP, Rayhorn P, Chi-Rosso G, Garber EA, Strauch KL, Horan GS, Reilly JO, Baker DP, Taylor FR, Koteliansky V, Pepinsky RB: Functional antagonists of sonic hedgehog reveal the importance of the N terminus for activity. J Cell Sci 1999, 112(pt 23):4405– 4414 27. Ingram WJ, Wicking CA, Grimmond SM, Forrest AR, Wainwright BJ: Novel genes regulated by Sonic Hedgehog in pluripotent mesenchymal cells. Oncogene 2002, 21:8196 – 8205 28. Plaisant M, Giorgetti-Peraldi S, Gabrielson M, Loubat A, Dani C, Peraldi P: Inhibition of hedgehog signaling decreases proliferation and clonogenicity of human mesenchymal stem cells. PLoS ONE 2011, 6:e16798 29. Lipinski RJ, Hutson PR, Hannam PW, Nydza RJ, Washington IM, Moore RW, Girdaukas GG, Peterson RE, Bushman W: Dose- and route-dependent teratogenicity, toxicity, and pharmacokinetic profiles of the hedgehog signaling antagonist cyclopamine in the mouse. Toxicol Sci 2008, 104:189 –197 30. Zhao C, Chen A, Jamieson CH, Fereshteh M, Abrahamsson A, Blum J, Kwon HY, Kim J, Chute JP, Rizzieri D, Munchhof M, VanArsdale T, Beachy PA, Reya T: Hedgehog signalling is essential for maintenance of cancer stem cells in myeloid leukaemia. Nature 2009, 458: 776 –779 31. Zhang X, Harrington N, Moraes RC, Wu MF, Hilsenbeck SG, Lewis MT: Cyclopamine inhibition of human breast cancer cell growth independent of Smoothened (Smo). Breast Cancer Res Treat 2009, 115:505–521 32. Tremblay MR, Lescarbeau A, Grogan MJ, Tan E, Lin G, Austad BC, Yu LC, Behnke ML, Nair SJ, Hagel M, White K, Conley J, Manna JD, Alvarez-Diez TM, Hoyt J, Woodward CN, Sydor JR, Pink M, MacDougall J, Campbell MJ, Cushing J, Ferguson J, Curtis MS, McGovern K, Read MA, Palombella VJ, Adams J, Castro AC: Discovery of a potent and orally active hedgehog pathway antagonist (IPI-926). J Med Chem 2009, 52:4400 – 4418

33. Safirstein R: Renal regeneration: reiterating a developmental paradigm. Kidney Int 1999, 56:1599 –1600 34. Lin SL, Li B, Rao S, Yeo EJ, Hudson TE, Nowlin BT, Pei H, Chen L, Zheng JJ, Carroll TJ, Pollard JW, McMahon AP, Lang RA, Duffield JS: Macrophage Wnt7b is critical for kidney repair and regeneration. Proc Natl Acad Sci U S A 2010, 107:4194 – 4199 35. Bielesz B, Sirin Y, Si H, Niranjan T, Gruenwald A, Ahn S, Kato H, Pullman J, Gessler M, Haase VH, Susztak K: Epithelial Notch signaling regulates interstitial fibrosis development in the kidneys of mice and humans. J Clin Invest 2010, 120:4040 – 4054 36. Villanueva S, Cespedes C, Vio CP: Ischemic acute renal failure induces the expression of a wide range of nephrogenic proteins. Am J Physiol Regul Integr Comp Physiol 2006, 290:R861–R870 37. Brownell I, Guevara E, Bai CB, Loomis CA, Joyner AL: Nerve-derived sonic hedgehog defines a niche for hair follicle stem cells capable of becoming epidermal stem cells. Cell Stem Cell 2011, 8:552–565 38. Barsoum I, Yao HH: Redundant and differential roles of transcription factors Gli1 and Gli2 in the development of mouse fetal Leydig cells. Biol Reprod 2011, 84:894 – 899 39. Park HL, Bai C, Platt KA, Matise MP, Beeghly A, Hui CC, Nakashima M, Joyner AL: Mouse Gli1 mutants are viable but have defects in SHH signaling in combination with a Gli2 mutation. Development 2000, 127:1593–1605 40. Stecca B, Ruiz IAA: Context-dependent regulation of the GLI code in cancer by HEDGEHOG and non-HEDGEHOG signals. J Mol Cell Biol 2010, 2:84 –95 41. Pasca di Magliano M, Sekine S, Ermilov A, Ferris J, Dlugosz AA, Hebrok M: Hedgehog/Ras interactions regulate early stages of pancreatic cancer. Genes Dev 2006, 20:3161–3173 42. Ji Z, Mei FC, Xie J, Cheng X: Oncogenic KRAS activates hedgehog signaling pathway in pancreatic cancer cells. J Biol Chem 2007, 282:14048 –14055 43. Grande MT, Fuentes-Calvo I, Arevalo M, Heredia F, Santos E, MartinezSalgado C, Rodriguez-Puyol D, Nieto MA, Lopez-Novoa JM: Deletion of H-Ras decreases renal fibrosis and myofibroblast activation following ureteral obstruction in mice. Kidney Int 2010, 77:509 –518 44. Bechtel W, McGoohan S, Zeisberg EM, Muller GA, Kalbacher H, Salant DJ, Muller CA, Kalluri R, Zeisberg M: Methylation determines fibroblast activation and fibrogenesis in the kidney. Nat Med 2010, 16:544 –550 45. Rodriguez-Pena AB, Grande MT, Eleno N, Arevalo M, Guerrero C, Santos E, Lopez-Novoa JM: Activation of Erk1/2 and Akt following unilateral ureteral obstruction. Kidney Int 2008, 74:196 –209 46. Bottinger EP, Bitzer M: TGF-beta signaling in renal disease. J Am Soc Nephrol 2002, 13:2600 –2610 47. Dennler S, Andre J, Alexaki I, Li A, Magnaldo T, ten Dijke P, Wang XJ, Verrecchia F, Mauviel A: Induction of sonic hedgehog mediators by transforming growth factor-beta: smad3-dependent activation of Gli2 and Gli1 expression in vitro and in vivo. Cancer Res 2007, 67:6981– 6986 48. Javelaud D, Alexaki VI, Dennler S, Mohammad KS, Guise TA, Mauviel A: TGF-beta/SMAD/GLI2 signaling axis in cancer progression and metastasis. Cancer Res 2011, 71:5606 –5610 49. Pola R, Ling LE, Silver M, Corbley MJ, Kearney M, Blake Pepinsky R, Shapiro R, Taylor FR, Baker DP, Asahara T, Isner JM: The morphogen Sonic hedgehog is an indirect angiogenic agent upregulating two families of angiogenic growth factors. Nat Med 2001, 7:706 –711 50. Chen W, Tang T, Eastham-Anderson J, Dunlap D, Alicke B, Nannini M, Gould S, Yauch R, Modrusan Z, Dupree KJ, Darbonne WC, Plowman G, de Sauvage FJ, Callahan CA: Canonical hedgehog signaling augments tumor angiogenesis by induction of VEGF-A in stromal perivascular cells, Proc Natl Acad Sci U S A 2011, 108:9589 –9594 51. Chade AR: Renovascular disease, microcirculation, and the progression of renal injury: role of angiogenesis. Am J Physiol Regul Integr Comp Physiol 2011, 300:R783–R790 52. Basile DP, Fredrich K, Chelladurai B, Leonard EC, Parrish AR: Renal ischemia reperfusion inhibits VEGF expression and induces ADAMTS-1, a novel VEGF inhibitor. Am J Physiol Renal Physiol 2008, 294:F928 –F936 53. Palma V, Lim DA, Dahmane N, Sanchez P, Brionne TC, Herzberg CD, Gitton Y, Carleton A, Alvarez-Buylla A, Ruiz i Altaba A: Sonic hedgehog controls stem cell behavior in the postnatal and adult brain. Development 2005, 132:335–344