BASIC RESEARCH www.jasn.org

Increased Expression of Secreted Frizzled-Related 4 in Polycystic Kidneys

Daniel Romaker,* Michael Puetz,* Sven Teschner,* Johannes Donauer,* Marcel Geyer,* Peter Gerke,* Brigitta Rumberger,* Bernd Dworniczak,† Petra Pennekamp,† Bjo¨rn Buchholz,‡ H.P.H. Neumann,* Rajiv Kumar,§ Joachim Gloy,* Kai-Uwe Eckardt,‡ and Gerd Walz*

*Renal Division, University Hospital Freiburg, Freiburg, Germany; †Department of Human Genetics, Mu¨nster, and ‡Department of Nephrology and Hypertension, University of Erlangen-Nuremberg, Erlangen, Germany; and §Departments of Medicine, Biochemistry and Molecular Biology, Nephrology Research Unit, Mayo Clinic and Foundation, Rochester, Minnesota

ABSTRACT Autosomal dominant polycystic kidney disease (ADPKD) is a common hereditary disease associated with progressive renal failure. Although cyst growth and compression of surrounding tissue may account for some loss of renal tissue, the other factors contributing to the progressive renal failure in patients with ADPKD are incompletely understood. Here, we report that secreted frizzled-related protein 4 (sFRP4) is upregulated in human ADPKD and in four different animal models of PKD, suggesting that sFRP4 expression is triggered by a common mechanism that underlies cyst formation. Cyst fluid from ADPKD kidneys activated the sFRP4 promoter and induced production of sFRP4 protein in renal tubular epithelial cell lines. Antagonism of the vasopressin 2 receptor blocked both promoter activity and tubular sFRP4 expression. In addition, sFRP4 selectively influenced members of the canonical Wnt signaling cascade and promoted cystogenesis of the zebrafish pronephros. sFRP4 was detected in the urine of both patients and animals with PKD, suggesting that sFRP4 may be a potential biomarker for monitoring the progression of ADPKD. Taken together, these observations suggest a potential role for SFRP4 in the pathogenesis of ADPKD.

J Am Soc Nephrol 20: 48–56, 2009. doi: 10.1681/ASN.2008040345

Autosomal dominant polycystic kidney disease brane , and active fluid secretion,4 but it has (ADPKD) occurs in approximately one of 1000 hu- remained unclear how these alterations explain the mans and causes ESRD in Ͼ50% of all affected pa- accelerated tissue loss in vivo. tients.1 Cyst formation starts during embryogenesis To understand better molecular mechanisms as- but typically does not compromise renal function sociated with the progression of renal failure in AD- until later in life. Mutations of either PKD1 or PKD, we generated profiles of ADPKD kid- PKD2 cause the disease, but why cysts, present in neys and identified secreted Frizzled-related Ͻ1% of all nephrons, cause renal failure remains elusive. Persistent proliferation, secretion, and cyst Received April 1, 2008. Accepted July 24, 2008. expansion seem to damage the surrounding tissue by reactive changes of the extracellular matrix.2 In Published online ahead of print. Publication date available at www.jasn.org. addition, increased of healthy paren- chyma seems to contribute to progressive renal fail- Correspondence: Dr. Gerd Walz, Renal Division, University Hos- pital Freiburg, Hugstetter Strasse 55, D-79106 Freiburg, Ger- 3 ure in human disease. Epithelial cells isolated from many. Phone ϩ49-761-2703250; Fax: ϩ49-761-2703245; E-mail: cystic kidneys show increased levels of proto-onco- [email protected] gene expression, a mislocalization of integral mem- Copyright ᮊ 2009 by the American Society of Nephrology

48 ISSN : 1046-6673/2001-48 J Am Soc Nephrol 20: 48–56, 2009 www.jasn.org BASIC RESEARCH protein 4 (sFRP4), a member of an evolving family of secreted Pkd2-deficient mice to address the question of whether sFRP4 molecules that antagonize the Wnt signaling cascade,5,6 as a is upregulated in polycystic kidney disease. Pkd2 (Ϫ/Ϫ) ani- differentially regulated gene. Members of the sFRP family can mals develop renal cysts but die between embryonic day 15 directly sequester Wnt,7 secreted glycoproteins that bind and (E15) and E20 of embryogenesis.13 Western blot analysis of activate Frizzled receptors to stabilize ␤-catenin, and initiate T kidneys from Pkd2 (Ϫ/Ϫ) mice at E16 revealed increased cell factor/lymphocyte enhancer factor (TCF/LEF)-dependent sFRP4 expression (Figure 1C). To determine whether sFRP4 gene transcription.8 The sFRP share a cysteine-rich domain, upregulation was also detectable in animal models of nephron- which mediates homodimerization and interaction with Friz- ophthisis, an autosomal recessive form of PKD, we also per- zled receptors to block Wnt binding.9 Since abnormal Wnt formed Western blot analysis in Invs (Ϫ/Ϫ) mice. These mice signaling has been implicated in the pathogenesis of polycystic lack functional NPHP2 (Inversin); the corresponding human kidney disease (reviewed in8), and a dysregulation of ␤-catenin gene is mutated in the infantile form of nephronophthisis degradation results in rapid-onset of polycystic kidney dis- (type II). As shown in Figure 1D, Invs (Ϫ/Ϫ) mice revealed a ease,10,11 we decided to analyze the regulation of sFRP4 expres- similar upregulation of sFRP4, suggesting that excessive sFRP4 sion, the effects of sFRP4 on various components of the Wnt production accompanies cyst formation independent of the signaling cascade and furthermore examined the effects of underlying gene mutation. Cysts typically lose their connec- sFRP4 overexpression on pronephric kidney development in tion to the draining tubules. Nevertheless, sFRP4 was detect- vivo. able in the urine of several patients with ADPKD but not in the urine of healthy volunteers (Figure 1E). To investigate urinary sFRP4 further, we examined the excretion of sFRP4 in Han: RESULTS SPRD rats, a slowly progressive PKD model. The urine of Han: SPRD rats was collected from two different animals at three Increased Expression of sFRP4 in Cystic Kidneys of different time points during a 3-wk interval; the urine was Patients with ADPKD and Animal Models of PKD normalized for creatinine and urea concentrations. As shown Microarray analysis of ADPKD kidneys identified sFRP4 as a in Figure 1F, the excretion of sFRP4 increased during the de- differentially regulated gene12; reverse transcription–PCR picted time interval. Because the urine of mice is notoriously (RT-PCR) confirmed that most ADPKD kidneys express in- difficult to standardize, we determined sFRP4 expression in creased amounts of this gene (Figure 1, A and B). We used kidney lysates of pcy mice, a mouse model of type III (NPHP3)

A D Figure 1. Increased sFRP4 expression in ADPKD kidneys. (A) Microarray analysis of to- tal RNA extracted from ADPKD kidneys re- vealed that sFRP4 expression normalized for glyceraldehyde-3-phosphate dehydrogenase (GAPDH) expression was increased in com- parison with tissue from normal kidneys. Two ADPKD kidneys could not be analyzed be- cause of spotting errors. (B) The microarray data were confirmed by semiquantitative RT- B E PCR. (C and D) Western blot analysis of whole-kidney lysates from Pkd2 (Ϫ/Ϫ) at E16 (C) and INVS-deficient mice at E19 (D) dem- F onstrated that sFRP4 was upregulated in kid- C neys of homozygote knockout embryos, com- pared with heterozygotes or wild-type mice. ␥-Tubulin was used as a loading control. PCR primers directed against the neomycin cas- G sette (Neo) or the thymidin kinase cassette (TK) as well as gene-specific primers were used to distinguish from among wild-type, heterozygote, and homozygote mice. (E) The urine of six patients with ADPKD (lanes 1 through 6) but not the urine of healthy indi- viduals (lanes A through C) contained hsFRP4. (F) Urine samples from two different heterozygote, male Han:SPRD rats (Cy/ϩ) were collected at postnatal weeks 6, 9, and 12 (corresponding to lanes 1 through 3). The samples were standardized to contain the same amount of either creatinine or urea, demonstrating that sFRP4 excretion in the urine increases with the progression of the disease. (G) Equal amounts of protein (1 ␮g) were obtained from pcy mice kidney lysates, ranging from 10 to 56 wk of age. Analysis by Western blot revealed increased sFRP4 protein level over time; actin was used to control for equal loading.

J Am Soc Nephrol 20: 48–56, 2009 FRP4 in Polycystic Kidneys 49 BASIC RESEARCH www.jasn.org nephronophthisis.14 Figure 1G confirmed that sFRP4 concen- cells (a cell line derived from the inner medulla cortical collect- tration increased over time in this animal model of cystic kid- ing duct) and PT1 cells (a cell line derived from the proximal ney disease as well. Thus, sFRP4 expression is elevated in AD- tubule) to cyst fluid. As shown in Figure 2, B and C, cyst fluid PKD and four different animal models of PKD suggesting that triggered the expression of sFRP4 protein after approximately a common final pathway of renal cystogenesis and/or cyst pro- 2 h of exposure. Both IMCD and PT1 cells responded to cyst gression induces sFRP4 expression. fluid within the first 5 d after seeding (Figure 2, B and C, left) but became unresponsive by day 10 (Figure 2, B and C, right). Cyst Fluid Stimulates sFRP4 Expression Instead, the 10-d-old cells expressed low basal levels of sFRP4, Cyst fluid contains several growth factors and hormones (e.g., suggesting that senescent and/or highly differentiated cells be- EGF, vasopressin) as well as a lipophilic substance that stimu- come increasingly unresponsive to the stimulating effect of lates the generation of cAMP.15,16 Because the sFRP4 promoter cyst fluid. contains multiple cis-acting elements, including a LEF1-, STAT3-, and cAMP-responsive element binding site,17,18 we Vasopressin 2 Receptor Antagonist SR121463 Inhibits analyzed whether cyst fluid activates the sFRP4 promoter. As sFRP4 Expression shown in Figure 2A, cyst fluid significantly augmented the ac- Vasopressin 2 receptor (V2R) antagonists (V2RA) have been tivity of the sFRP4 promoter fragment Ϫ1417/ϩ83. To test shown to ameliorate the progression of renal disease in mouse whether the cyst fluid–mediated promoter activation is associ- models of PKD by blocking the production of cAMP in re- ated with the production of sFRP4 protein, we exposed IMCD sponse to vasopressin.19,20 Vasopressin (antidiuretic hormone) triggered the expression of sFRP4 in IMCD cells (Figure 3, A and B), albeit only at high concentrations (10 ␮M). Although IMCD cells are derived from the collecting duct, V2R expres- sion was barely detectable by RT-PCR in the IMCD clone used in this experiments (data not shown). Thus, the low V2R ex- pression may explain why high concentrations of antidiuretic hormone were needed to elicit sFRP4 expression in IMCD cells. Because several hormones and growth factors have been detected in cyst fluid, we examined whether the V2R antago- nist SR121463 can block sFRP4 expression triggered by cyst fluid. SR121463, at concentrations as low as 1 nM, almost com- pletely suppressed the cyst fluid–induced sFRP4 expression in IMCD cells (Figure 3C). V2RA ameliorate PKD progression in animal models and are now used in clinical trials to investigate their therapeutic benefit. Because our results indicate that sFRP4 expression can be blocked by V2RA in vitro, we moni- tored the expression of sFRP4 in response to systemic applica- tion of the V2R antagonist in vivo. SR121463 was fed during a 15-wk period to pcy mice. As demonstrated in Figure 3D, V2RA-treated pcy mice expressed less sFRP4 than vehicle- treated pcy mice. Thus, sFRP4 expression seems to correlate with the progression of PKD and responds to V2RA aimed to ameliorate this disease.

sFRP4 Blocks Select Wnt Pathway Components To understand the consequences of increased sFRP4 expres- Figure 2. The sFRP4 promoter is activated by cyst fluid in a sion in PKD, we decided to address the effects of sFRP4 on Wnt concentration-dependent manner. (A) HEK 293T cells were trans- signaling. Previous findings demonstrated that sFRP4 abro- fected with the sFRP4 promoter fragment Ϫ1417 to ϩ83, driving gates Xenopus Wnt8 (XWnt8)-mediated double-axis forma- ␤ the expression of a secreted alkaline phosphatase and -galac- tion,21 an established system to monitor canonical Wnt activa- tosidase, to normalize for transfection efficiency. Twelve hours tion during Xenopus embryogenesis. We expressed sFRP4 in after transfection, the cells were incubated with diluted cyst fluid HEK 293T cells and confirmed that sFRP4 blocks the XWnt8- (CF) as indicated; the promoter activity was assayed 4 h later. All ␤ experiments were performed in triplicate and repeated at least mediated stabilization of -catenin, a hallmark of the activated three times. (B and C) IMCD (B) or PT1 (C) cells were stimulated canonical Wnt branch (Figure 4A). sFRP are thought to block with CF after5dofculture (left) or after 10 d of culture (right) and canonical Wnt signaling through direct interaction with solu- assayed for sFRP4 expression by Western blot analysis; ␥-tubulin ble Wnt molecules.7 Consistent with this hypothesis, sFRP4 was used as a loading control. blocked XWnt8 but had no effect on the stabilization of ␤-cate-

50 Journal of the American Society of Nephrology J Am Soc Nephrol 20: 48–56, 2009 www.jasn.org BASIC RESEARCH nin mediated by Dishevelled (Dsh), a cyto- plasmic adaptor molecule that acts down- stream of Wnt and Frizzled receptors. Interestingly, sFRP4 also had no effect on the stabilization of ␤-catenin mediated by XWnt3a, suggesting that sFRP4 selectively in- terferes with some but not all Wnt molecules (Figure 4A). When we analyzed the TCF/LEF- dependent activation of the TOPFLASH lucif- erase reporter, we confirmed that sFRP4-con- ditioned medium (sFRP4 CM) blocked XWnt8 but had no effect on XWnt3-mediated activation of the reporter (Figure 4B). Figure 3. Vasopressin and the V2RA SR121463 affect sFRP4 expression levels in To corroborate our observation further, vitro and in vivo. (A) Expression of sFRP4 in response to increasing concentrations we monitored the formation of a secondary of vasopressin (VP; 0.1 to 10.0 ␮M) in IMCD cells; ␥-tubulin served as a loading body axis after ectopic stimulation of the ca- control. (B) Time course of sFRP4 expression in response to 10 ␮M vasopressin; nonical during early ␥-tubulin served as a loading control. (C) The V2RA SR121463 blocked sFRP4 Xenopus embryogenesis. Both XWnt3a and expression, triggered by CF in IMCD cells. (D) Kidneys of two pcy mice (lanes 1 and XWnt8 induced a secondary body axis (Figure 2), treated with SR121463 over a period of 15 wk, and two vehicle treated pcy mice 5); however, whereas sFRP4 almost com- (lanes 3 and 4) were explanted and homogenized. Equal amounts of protein were pletely rescued the phenotypic changes caused analyzed on Western blot and compared for sFRP4 concentration. Results showed by XWnt8, it had no effect on either Dsh- or reduced amounts of sFRP4 expression in treated animals in comparison with vehi- cle-treated mice. Actin was used as a loading control. XWnt3a-mediated axis duplication. This analysis confirms that sFRP4 selectively blocks distinct Wnt molecules. To investigate further mecha- nisms that mediate the effects of sFRP4 on the canonical Wnt pathway, we analyzed the influence of sFRP4 on rat Frizzled 1 (RFz1)- and Xenopus Frizzled 8 (XFz8)-mediated activation of the canonical Wnt signaling cascade. The stabilization of ␤-catenin by either RFz1 or XFz8 was blocked by sFRP4 CM, suggesting that sFRP4 inhibits canonical Wnt signaling by binding to both soluble Wnt molecules and Frizzled receptors (Figure 6A). Consistent with this hypothesis, sFRP4 CM blocked the activation of the TOPFLASH reporter, induced by either RFz1 or XFz8 (Figure 6B). sFRP4 Promotes Cystogenesis of the Zebrafish Pronephros The zebrafish pronephros is a well-established model sys- tem to monitor renal development in vivo.22 To understand the consequences of ectopic sFRP4 expression, we analyzed Figure 4. sFRP4 blocks XWnt8- but not XWnt3a-mediated acti- the phenotypic changes caused by human sFRP4 mRNA vation of the canonical Wnt pathway. (A) HEK 293T cells were transfected with equal amounts (2 ␮g) of pcDNA3 vector, XWnt8, during zebrafish embryogenesis. Microinjection of sFRP4 Dsh, and XWnt3a and incubated with or without sFRP4-CM for was well tolerated with little mortality (Figure 7A). The 12 h. Accumulation of cytoplasmic ␤-catenin was used to monitor Wt1b:GFP transgenic zebrafish line was used to visualize the activation of canonical Wnt signaling; ␥-tubulin served as a changes of the glomerulus and proximal tubuli by fluores- loading control. (B) HEK 293T cells were co-transfected with the cence microscopy.23 No pronephric cysts were detectable in TOPFLASH reporter construct and ␤-galactosidase, to normalize control embryos or embryos injected with 50 pg of sFRP4; for transfection efficiency, in combination with a control (CTL) however, higher amounts (100 and 200 pg) caused clearly plasmid, XWnt8, Dsh, or XWnt3a and incubated with or without detectable pronephric cysts in 6 to 8% of embryos (Figure 7, sFRP4-CM for 12 h. B and C). Surprising, ectopic expression of sFRP4 also in- terfered with the development of a normal left–right asym- in zebrafish expressing sFRP4 (Figure 8A). A dorsal body metry. Heterotaxia of the zebrafish pancreas, observed in curvature is the hallmark of ciliary defects and characteristic Ͻ10% of control animals, increased to approximately 30% for TRPP2-deficient zebrafish.24,25 Whereas none of the

J Am Soc Nephrol 20: 48–56, 2009 FRP4 in Polycystic Kidneys 51 BASIC RESEARCH www.jasn.org

DISCUSSION

Several sFRP molecules are expressed during renal develop- ment in the metanephric kidney. In embryonic kidney ex- plants, sFRP1 binds to Wnt4, an essential component of mes- enchymal-to-epithelial conversion in the developing kidney, and blocks kidney tubule formation and bud branching.26 In addition to their role in canonical Wnt signaling, sFRP mem- bers modulate bone morphogenetic protein signaling during early Xenopus laevis embryogenesis and mediate a cross-talk between hedgehog and Wnt signaling pathways.27,28 Increased apoptosis rates observed in tissues that express sFRP family members suggest that these molecules promote apoptosis.29–37 Although the role of apoptosis in ADPKD remains controver- sial,38 sFRP4 could not only affect the cells lining the cysts but also diffuse to the surrounding tissue and perhaps contribute to the increased apoptotic rates that have been observed in ADPKD kidneys.3 Furthermore, our data indicate that sFRP4 inhibits select members of the canonical Wnt signaling path- way. This selectivity might explain why ␤-catenin can still ac- cumulate during PKD progression even in the presence of ele- vated sFRP4 concentrations. Because the sFRP4 promoter region contains a TCF/LEF-binding site and can be stimulated by constitutively active ␤-catenin,39 it is conceivable that ca- nonical Wnt, not blocked by sFRP4, stimulate sFRP4 produc- tion.

Figure 5. sFRP4 blocks double-axis formation caused by XWnt8 but has no effect on XWnt3a-mediated activation of the canonical Wnt signaling cascade in Xenopus embryogenesis. Xenopus lae- vis eggs were injected dorsolaterally at the four-cell stage with Dsh, XWnt3a, XWnt8, and sFRP4 mRNA as indicated and scored at tadpole stage 37/38. Whereas sFRP4 blocked the formation of a secondary axis mediated by XWnt8, the XWnt3a-induced sec- ondary axis was not rescued by co-injection of sFRP4 mRNA. The Figure 6. sFRP4 inhibits Frizzled-mediated activation of the ca- percentages of no axis duplication and partial or complete axis nonical wnt signaling branch. (A) HEK 293T cells were transfected duplication are shown in the bars; numbers of scored tadpoles are with pCDNA6 (CTL), RFz1, or XFz8 and incubated with sFRP4-CM ␤ given on top of each bar. for 12 h as indicated. Cytoplasmic -catenin levels were used to monitor the activation of the canonical Wnt pathway; ␥-tubulin served as a loading control. The Frizzled-mediated accumulation of cytosolic ␤-catenin was suppressed by sFRP4. (B) HEK 293T control animals displayed this abnormality, 30 to 60% of Ն cells were co-transfected with the TOPFLASH reporter construct zebrafish embryos injected with 50 pg sFRP4 showed an and pcDNA6 (CTL), RFz1, or XFz8; ␤-galactosidase was used to abnormal body curvature (Figure 8B). Taken together, our normalize for transfection efficiency. sFRP4-CM inhibited the results suggest that sFRP4 can promote cystogenesis in the RFz1- and XFz8-mediated activation of the TOPFLASH reporter zebrafish embryo. construct.

52 Journal of the American Society of Nephrology J Am Soc Nephrol 20: 48–56, 2009 www.jasn.org BASIC RESEARCH

Hypermethylation and epigenetic inactivation of sFRP pro- moter regions has been observed in breast, gastric, renal, and other cancers,40,41 suggesting that sFRP molecules act as tumor suppressors to curtail Wnt signaling. Their increased expres- sion in some cancers indicates that they are expressed to con-

Figure 8. Ectopic expression of sFRP4 causes heterotaxia and an abnormal body curvature in zebrafish embryos. (A) The Wt1b:GFP transgene labels the pancreas of zebrafish embryo (dorsal view, anterior to the top). Microinjection of sFRP4 increased the fre- Figure 7. sFRP4 promotes cyst formation in the zebrafish pro- quency of pancreatic heterotaxia (arrow) from Յ10% in control nephros. (A) Zebrafish embryos were injected with 50, 100, and Wt1b:GFP transgenic zebrafish (CTL) to 30% in zebrafish injected 200 pg of human sFRP4, and survival was monitored at 5 to 7, 24, with 50 pg of sFRP4; zebrafish were analyzed at 72 hpf. (B) Dorsal and 55 h postfertilization (hpf). Human sFRP4 was well tolerated body curvature, not observed in control animals, was present in and not associated with substantial mortality (CTL, uninjected 30 to 60% of animals injected with Ն50 pg of sFRP4. To quantify control embryos). Depicted is the number of surviving embryos. the changes, microinjected zebrafish embryos were grouped into (B) Cysts, as indicated by the asterisks, were detectable within the three dysmorph classes (Dys I through III) according to the degree proximal tubuli adjacent to the single zebrafish glomerulus. Ar- of body curvature abnormalities. rows point to pronephric ducts. The Wt1b:GFP transgenic ze- brafish line was used to visualize the proximal pronephros by 42,43 fluorescence microscopy (CTL, control embryos). (C) Injection of strict aberrant Wnt signaling. Ciliary defects result in PKD sFRP4 (100 to 200 pg) resulted in 6 to 8% pronephric cysts (CTL, (reviewed in reference44) and are associated with a deregula- control embryos). tion of Wnt signaling.45,46 It is conceivable that sFRP4 expres-

J Am Soc Nephrol 20: 48–56, 2009 FRP4 in Polycystic Kidneys 53 BASIC RESEARCH www.jasn.org sion in PKD is caused by an abnormal activation of the Wnt 94°C for 2 min, followed by multiple cycles of denaturation at 94°C at cascade to balance the detrimental effects of uncontrolled Wnt 1 min, annealing at 58°C for 1 min, and elongation at 72°C for 1 min. signaling. Alternatively, the damaging effects of progressively Cycle number was adjusted to mRNA expression level of the different expanding cysts may trigger sFRP4 expression. For example, . The PCR products were separated on a 2% agarose gel and sFRP1 is upregulated after myocardial infarction to limit the analyzed in relation to the corresponding control band. consequences of ischemic damage.47 Thus, increased sFRP4 expression could represent an adaptive mechanism to protect Cell Culture the kidney from the detrimental effects of progressively ex- Conditionally immortalized proximal tubular epithelial (PT) cells panding cysts; however, the developmental changes caused by were maintained at 33°C in DMEM-F12 supplemented with 2% calf sFRP4 expression in zebrafish embryos argue that sFRP4 might serum. For differentiation of the PT cells, temperature was raised to represent a disease-promoting factor. Although sFRP4 micro- 37°C. Immortalized mIMCD-3 cells (referred to as IMCD cells) were injection is well tolerated, it is associated with pronephric cyst grown in DMEM - F12 medium supplemented with 10% calf serum. formation, heterotaxia, and an abnormal body curvature. Human embryonic kidney cells (HEK-293T cells) were grown in These phenotypic changes are characteristic of zebrafish em- DMEM supplemented with 10% calf serum. bryos with ciliary defects22 and suggest that sFRP4 can interfere with cellular programs that maintain normal tubular geome- Western Blot Analysis Cells were harvested after extensive washing with cold PBS in PBS and try. The defective left-right patterning in sFRP4-expressing ze- lysed in a buffer containing 20 mM Tris-HCL (pH 7.5), 1% Triton brafish may result from an early interference with noncanoni- ϩ X-100, 25 mM NaF, 12.5 mM Na4P2O7, 0.1 mM EDTA, 50 mM cal Ca2 -dependent Wnt signaling at the dorsal forerunner NaCl, 2 mM Na VO , and protease inhibitors. After centrifugation cells, the precursors that establish the ciliated epithelial cell 3 4 (15,000 g for 15 min, 4°C) and ultra centrifugation steps of the lysates layer lining the zebrafish Kuppfer’s vesicle.48 The upregulation (45,000 rpm for 30 min, 4°C) equal amounts of protein were sepa- of sFRP4 in several animal models of PKD as well as human rated on SDS gels and further processed with ␤-catenin (Transduc- ADPKD suggests that increased sFRP4 expression is not linked tion Laboratories, San Jose, CA), ␥-tubulin (Sigma, Hamburg, Ger- to a particular gene defect. Although sFRP4 is therefore not many), and ␤-actin (Cell Signaling, Danvers, MA) antibodies for useful to diagnose a specific form of PKD, monitoring urine Westen blot analysis. The sFRP-4 monoclonal antibody was recently and/or blood levels may be helpful to assess the progression of described.50 disease in patients with PKD; however, sFRP4 is also detectable in the urine of healthy individuals by mass spectrometry,49 and Reporter Assays its value as a marker to monitor cystic kidney disease progres- Reporter assays were performed using Great EscAPe SEAP chemilu- sion needs to be evaluated in large patient collectives with PKD minescence detection kit (BD Bioscience, Heidelberg, Germany). The as well as other renal diseases. results are presented as relative luciferase or alkaline phosphatase ac- tivities after normalization for ␤-galactosidase activity.

CONCISE METHODS Genotyping and Organ Preparation Kidneys were explanted from Pkd2-, Invs-orpcy -mice at E16 (Pkd2), Materials E20 (Invs) or and processed for Western blot analysis immediately Patient material was obtained after informed consent. Tissue samples after organ removal. Additionally, material from the head was used were removed and immediately either snap-frozen in liquid nitrogen for genotyping. For Western blot analysis, the kidneys were homoge- for RNA preparation or fixed in 4% PFA. Cyst fluid was collected by nized and lysed for 30 minutes on ice in a buffer containing 1% Triton aspiration and stored at either 4°C (short-term) or Ϫ20°C (long- and 0.1 mM EDTA. Equal amounts were used for Western blot anal- term). Urine samples were centrifuged at 14,000 rpm for 5 min. The ysis. Pkd2 and Invs E16 and E20 embryos were genotyped by PCR. The supernatant was stored at either 4°C (short-term) or Ϫ20°C (long- following primers were used to genotype Pkd2-mice: 5Ј-CCCATGGC- term) for further analysis. For RNA isolation, frozen tissue was ho- GATGCCTGCTTGCCG-3Ј,5Ј-GGCGATAGAAGGCGATGCGC- mogenized in 4 M guanidinium isothiocyanate/0.72% ␤-mercapto- TGCG-3Ј (neomycin cassette), 5Ј-CGTCCAATGAATTTGCACC- ethanol and subsequently purified on a caesium chloride gradient. AACAAGAACGC-3Ј,5Ј-CTTTCGTCCTGCTCCAGGCAAGCG- GAGC-3Ј (to amplify a Pkd2 specific region to distinguish between wild- Gene Expression Profiling and Semiquantitative RT- type, heterozygote, and homozygote animals) and Invs mice: 5Ј- PCR GATTACGTAATAGTGGTCCCTCAGG-3Ј,5Ј-CTGTCCAGTGC- Microarray analysis and validation by RT-PCR were performed as ACCATGTGGACCT-3Ј (thymidine kinase cassette), and 5Ј-GTATT- described. 51 For semiquantitative RT-PCR, total RNA (1 ␮g) was TACTCAGTGGCCTCAG-3Ј,5Ј-CCGATCACAGGATTGCTAG-3Ј (to purified with the RNase free DNase set (Qiagen, Hilden, Germany) amplify exon 3). and transcribed into cDNAs using an oligo (dT) primer and Super- script II reverse transcriptase (Invitrogen, Carlsbad, CA). For each Zebrafish Microinjections PCR reaction, ␤-actin or GAPDH expression was used as an internal Microinjections and analysis were performed as recently described.25 standard control. The amplification cycle consisted of a hot start at Briefly: zebrafish embryos were microinjected at the 1 to 2 cell stage

54 Journal of the American Society of Nephrology J Am Soc Nephrol 20: 48–56, 2009 www.jasn.org BASIC RESEARCH with various amounts of synthesized capped human sFRP4 mRNA 12. Schieren G, Rumberger B, Klein M, Kreutz C, Wilpert J, Geyer M, diluted in a solution containing 200 mM KCL, 0.1% Phenol Red, and Faller D, Timmer J, Quack I, Rump LC, Walz G, Donauer J: Gene 10 mM HEPES (pH 7.5). Embryos were dechorionated manually at profiling of polycystic kidneys. Nephrol Dial Transplant 21: 1816– 1824, 2006 24–30 hours post fertilization (hpf), and were kept at 28.5°C in 13. Pennekamp P, Karcher C, Fischer A, Schweickert A, Skryabin B, Horst Danieau’s solution with 0.003% 1-Phenyl-2-thiourea (Sigma) to sup- J, Blum M, Dworniczak B: The ion channel polycystin-2 is required for press pigmentation. Staging was done according to hpf. Zebrafish left-right axis determination in mice. Curr Biol 12: 938–943, 2002 embryos were analyzed under a Leica MZ16 stereo-microscope 14. Olbrich H, Fliegauf M, Hoefele J, Kispert A, Otto E, Volz A, Wolf MT, (Leica, Solms, Germany). Pictures were taken with a SPOT Insight Sasmaz G, Trauer U, Reinhardt R, Sudbrak R, Antignac C, Gretz N, Walz G, Schermer B, Benzing T, Hildebrandt F, Omran H: Mutations in Fire Wire System and processed with the SPOT Imaging Software a novel gene, NPHP3, cause adolescent nephronophthisis, tapeto- (Diagnostic Instruments, Sterling Heights, MI). retinal degeneration and hepatic fibrosis. Nat Genet 34: 455–459, 2003 15. Grantham JJ, Ye M, Gattone VH 2nd, Sullivan LP: In vitro fluid secre- tion by epithelium from polycystic kidneys. J Clin Invest 95: 195–202, ACKNOWLEDGMENTS 1995 16. Yamaguchi T, Nagao S, Takahashi H, Ye M, Grantham JJ: Cyst fluid This study was supported by Deutsche Forschungsgesellschaft grant from a murine model of polycystic kidney disease stimulates fluid secretion, cyclic adenosine monophosphate accumulation, and cell WA 597/10 to G.W. proliferation by Madin-Darby canine kidney cells in vitro. Am J Kidney We thank the members of the Renal Division for helpful discus- Dis 25: 471–477, 1995 sion. We gratefully acknowledge the gift of the vasopressin 2 receptor 17. Yam JW, Chan KW, Wong VK, Hsiao WL: Transcriptional activity of the antagonist SR121463, provided by Sanofi-Aventis. Han:SPRD rats promoter region of rat frizzled-related protein gene. Biochem Biophys were kindly provided by N. Gretz, Mannheim, and the sFRP4 pro- Res Commun 286: 94–100, 2001 18. Wong VK, Yam JW, Hsiao WL: Cloning and characterization of the moter constructs by W.L. Wendy Hsiao, Hong Kong. promoter region of the mouse frizzled-related protein 4 gene. Biol Chem 384: 1147–1154, 2003 19. Gattone VH 2nd, Wang X, Harris PC, Torres VE: Inhibition of renal cystic disease development and progression by a vasopressin V2 DISCLOSURES receptor antagonist. Nat Med 9: 1323–1326, 2003 None. 20. Torres VE, Wang X, Qian Q, Somlo S, Harris PC, Gattone VH 2nd: Effective treatment of an orthologous model of autosomal dominant polycystic kidney disease. Nat Med 10: 363–364, 2004 21. Abu-Jawdeh G, Comella N, Tomita Y, Brown LF, Tognazzi K, Sokol SY, Kocher O: Differential expression of frpHE: a novel human stromal REFERENCES protein of the secreted frizzled gene family, during the endometrial cycle and malignancy. Lab Invest 79: 439–447, 1999 1. Grantham JJ: The etiology, pathogenesis, and treatment of autosomal 22. Drummond IA: Kidney development and disease in the zebrafish. dominant polycystic kidney disease: Recent advances. Am J Kidney J Am Soc Nephrol 16: 299–304, 2005 Dis 28: 788–803, 1996 23. Perner B, Englert C, Bollig F: The Wilms tumor genes wt1a and wt1b 2. Wilson PD: Polycystic kidney disease. N Engl J Med 350: 151–164, control different steps during formation of the zebrafish pronephros. 2004 Dev Biol 309: 87–96, 2007 3. Woo D: Apoptosis and loss of renal tissue in polycystic kidney dis- 24. Sun Z, Amsterdam A, Pazour GJ, Cole DG, Miller MS, Hopkins N: A eases. N Engl J Med 333: 18–25, 1995 genetic screen in zebrafish identifies cilia genes as a principal cause of 4. Wilson PD: Epithelial cell polarity and disease. Am J Physiol 272: cystic kidney. Development 131: 4085–4093, 2004 F434–F442, 1997 25. Fu X, Wang Y, Schetle N, Gao H, Putz M, von Gersdorff G, Walz G, 5. Jones SE, Jomary C: Secreted Frizzled-related proteins: Searching for Kramer-Zucker AG: The subcellular localization of TRPP2 modulates its relationships and patterns. Bioessays 24: 811–820, 2002 function. J Am Soc Nephrol 19: 1342–1351, 2008 6. Kawano Y, Kypta R: Secreted antagonists of the Wnt signalling path- 26. Yoshino K, Rubin JS, Higinbotham KG, Uren A, Anest V, Plisov SY, way. J Cell Sci 116: 2627–2634, 2003 Perantoni AO: Secreted Frizzled-related proteins can regulate meta- 7. Dennis S, Aikawa M, Szeto W, d’Amore PA, Papkoff J: A secreted nephric development. Mech Dev 102: 45–55, 2001 frizzled related protein, FrzA, selectively associates with Wnt-1 protein 27. Lee HX, Ambrosio AL, Reversade B, De Robertis EM: Embryonic and regulates wnt-1 signaling. J Cell Sci 112: 3815–3820, 1999 dorsal-ventral signaling: Secreted frizzled-related proteins as inhibi- 8. Benzing T, Simons M, Walz G: Wnt signaling in polycystic kidney tors of tolloid proteinases. Cell 124: 147–159, 2006 disease. J Am Soc Nephrol 18: 1389–1398, 2007 28. He J, Sheng T, Stelter AA, Li C, Zhang X, Sinha M, Luxon BA, Xie J: 9. Bafico A, Gazit A, Pramila T, Finch PW, Yaniv A, Aaronson SA: Inter- Suppressing Wnt signaling by the hedgehog pathway through sFRP-1. action of frizzled related protein (FRP) with Wnt ligands and the frizzled J Biol Chem 281: 35598–35602, 2006 receptor suggests alternative mechanisms for FRP inhibition of Wnt 29. Zhou Z, Wang J, Han X, Zhou J, Linder S: Up-regulation of human signaling. J Biol Chem 274: 16180–16187, 1999 secreted frizzled homolog in apoptosis and its down-regulation in 10. Saadi-Kheddouci S, Berrebi D, Romagnolo B, Cluzeaud F, Peuchmaur breast tumors. Int J Cancer 78: 95–99, 1998 M, Kahn A, Vandewalle A, Perret C: Early development of polycystic 30. Ellies DL, Church V, Francis-West P, Lumsden A: The WNT antagonist kidney disease in transgenic mice expressing an activated mutant of cSFRP2 modulates programmed cell death in the developing hind- the beta-catenin gene. Oncogene 20: 5972–5981, 2001 brain. Development 127: 5285–5295, 2000 11. Qian CN, Knol J, Igarashi P, Lin F, Zylstra U, Teh BT, Williams BO: 31. Jones SE, Jomary C, Grist J, Stewart HJ, Neal MJ: Modulated expres- Cystic renal neoplasia following conditional inactivation of apc in sion of secreted frizzled-related proteins in human retinal degenera- mouse renal tubular epithelium. J Biol Chem 280: 3938–3945, 2005 tion. Neuroreport 11: 3963–3967, 2000

J Am Soc Nephrol 20: 48–56, 2009 FRP4 in Polycystic Kidneys 55 BASIC RESEARCH www.jasn.org

32. Schumann H, Holtz J, Zerkowski HR, Hatzfeld M: Expression of se- related protein 2 in a primary canine mammary tumor cell line: A creted frizzled related proteins 3 and 4 in human ventricular myocar- candidate tumor marker for mammary tumor cells. In Vitro Cell Dev dium correlates with apoptosis related gene expression. Cardiovasc Biol Anim 39, 221–227, 2003 Res 45: 720–728, 2000 43. Lee JL, Chang CJ, Wu SY, Sargan DR, Lin CT: Secreted frizzled-related 33. Dufourcq P, Couffinhal T, Ezan J, Barandon L, Moreau C, Daret D, protein 2 (SFRP2) is highly expressed in canine mammary gland tu- Duplaa C: FrzA, a secreted frizzled related protein, induced angio- mors but not in normal mammary glands. Breast Cancer Res Treat 84: genic response. Circulation 106: 3097–3103, 2002 139–149, 2004 34. Ko J, Ryu KS, Lee YH, Na DS, Kim YS, Oh YM, Kim IS, Kim JW: Human 44. Fliegauf M, Benzing T, Omran H: When cilia go bad: Cilia defects and secreted frizzled-related protein is down-regulated and induces apo- ciliopathies. Nat Rev Mol Cell Biol 8: 880–893, 2007 ptosis in human cervical cancer. Exp Cell Res 280: 280–287, 2002 45. Gerdes JM, Liu Y, Zaghloul NA, Leitch CC, Lawson SS, Kato M, Beachy 35. Drake JM, Friis RR, Dharmarajan AM: The role of sFRP4, a secreted PA, Beales PL, Demartino GN, Fisher S, Badano JL, Katsanis N: Dis- frizzled-related protein, in ovulation. Apoptosis 8: 389–397, 2003 ruption of the basal body compromises proteasomal function and 36. Lacher MD, Siegenthaler A, Jager R, Yan X, Hett S, Xuan L, Saurer S, perturbs intracellular Wnt response. Nat Genet 39: 1350–1360, 2007 Lareu RR, Dharmarajan AM, Friis R: Role of DDC-4/sFRP-4, a secreted 46. Corbit KC, Shyer AE, Dowdle WE, Gaulden J, Singla V, Reiter JF: Kif3a frizzled-related protein, at the onset of apoptosis in mammary involu- constrains beta-catenin-dependent Wnt signalling through dual ciliary tion. Cell Death Differ 10: 528–538, 2003 and non-ciliary mechanisms. Nat Cell Biol 10: 70–76, 2008 37. Bodine PV, Billiard J, Moran RA, Ponce-de-Leon H, McLarney S, 47. Barandon L, Couffinhal T, Ezan J, Dufourcq P, Costet P, Alzieu P, Mangine A, Scrimo MJ, Bhat RA, Stauffer B, Green J, Stein GS, Lian Leroux L, Moreau C, Dare D, Duplaa C: Reduction of infarct size and JB, Komm BS: The Wnt antagonist secreted frizzled-related protein-1 prevention of cardiac rupture in transgenic mice overexpressing FrzA. controls osteoblast and osteocyte apoptosis. J Cell Biochem 96, Circulation 108: 2282–2289, 2003 1212–1230, 2005 48. Schneider I, Houston DW, Rebagliati MR, Slusarski DC: Calcium fluxes 38. Piontek K, Menezes LF, Garcia-Gonzalez MA, Huso DL, Germino GG: in dorsal forerunner cells antagonize beta-catenin and alter left-right A critical developmental switch defines the kinetics of kidney cyst patterning. Development 135: 75–84, 2008 formation after loss of Pkd1. Nat Med 13: 1490–1495, 2007 49. Adachi J, Kumar C, Zhang Y, Olsen JV, Mann M: The human urinary 39. Lee AY, He B, You L, Dadfarmay S, Xu Z, Mazieres J, Mikami I, proteome contains more than 1500 proteins, including a large pro- McCormick F, Jablons DM: Expression of the secreted frizzled-related portion of membrane proteins. Genome Biol 7: R80, 2006 protein gene family is downregulated in human mesothelioma. On- 50. Berndt T, Craig TA, Bowe AE, Vassiliadis J, Reczek D, Finnegan R, Jan cogene 23: 6672–6676, 2004 De Beur SM, Schiavi SC, Kumar R: Secreted frizzled-related protein 4 40. Suzuki H, Watkins DN, Jair KW, Schuebel KE, Markowitz SD, Chen is a potent tumor-derived phosphaturic agent. J Clin Invest 112: WD, Pretlow TP, Yang B, Akiyama Y, Van Engeland M, Toyota M, 785–794, 2003 Tokino T, Hinoda Y, Imai K, Herman JG, Baylin SB: Epigenetic inacti- 51. Donauer J, Rumberger B, Klein M, Faller D, Wilpert J, Sparna T, vation of SFRP genes allows constitutive WNT signaling in colorectal Schieren G, Rohrbach R, Dern P, Timmer J, Pisarski P, Kirste G, Walz cancer. Nat Genet 36: 417–422, 2004 G: Expression profiling on chronically rejected transplant kidneys. 41. Urakami S, Shiina H, Enokida H, Hirata H, Kawamoto K, Kawakami T, Transplantation 76: 539–547, 2003 Kikuno N, Tanaka Y, Majid S, Nakagawa M, Igawa M, Dahiya R: Wnt 52. Simons M, Gloy J, Ganner A, Bullerkotte A, Bashkurov M, Kronig C, antagonist family genes as biomarkers for diagnosis, staging, and Schermer B, Benzing T, Cabello OA, Jenny A, Mlodzik M, Polok B, prognosis of renal cell carcinoma using tumor and serum DNA. Clin Driever W, Obara T, Walz G: Inversin, the gene product mutated in Cancer Res 12: 6989–6997, 2006 nephronophthisis type II, functions as a molecular switch between Wnt 42. Lee JL, Chang CJ, Chueh LL, Lin CT: Expression of secreted frizzled- signaling pathways. Nat Genet 37: 537–543, 2005

56 Journal of the American Society of Nephrology J Am Soc Nephrol 20: 48–56, 2009