BASIC RESEARCH www.jasn.org

Transcription Factor 21 Is Required for Branching Morphogenesis and Regulates the Gdnf-Axis in Development

Shintaro Ide,1 Gal Finer,2,3 Yoshiro Maezawa,1 Tuncer Onay,3,4 Tomokazu Souma,3,4 Rizaldy Scott,3,4 Kana Ide,1 Yoshihiro Akimoto,5 Chengjin Li,6 Minghao Ye,3,4 Xiangmin Zhao,2,3 Yusuke Baba,1 Takuya Minamizuka,1 Jing Jin,3,4 Minoru Takemoto,1,7 Koutaro Yokote,1 and Susan E. Quaggin3,4

Due to the number of contributing authors, the affiliations are listed at the end of this article.

ABSTRACT Background The mammalian kidney develops through reciprocal inductive signals between the meta- nephric and ureteric bud. 21 (Tcf21) is highly expressed in the meta- nephric mesenchyme, including Six2-expressing cap mesenchyme and Foxd1-expressing stromal mesenchyme. Tcf21 knockout mice die in the perinatal period from severe renal hypodysplasia. In humans, Tcf21 mRNA levels are reduced in renal tissue from human fetuses with renal dysplasia. The molecular mechanisms underlying these renal defects are not yet known. Methods Using a variety of techniques to assess kidney development and expression, we compared the phenotypes of wild-type mice, mice with germline deletion of the Tcf21 gene,micewithstromal mesenchyme–specific Tcf21 deletion, and mice with cap mesenchyme–specific Tcf21 deletion. Results Germline deletion of Tcf21 leads to impaired ureteric bud branching and is accompanied by down- regulated expression of Gdnf-Ret-Wnt11, a key pathway required for branching morphogenesis. Selective removal of Tcf21 from the renal stroma is also associated with attenuation of the Gdnf signaling axis and leads to a defect in ureteric bud branching, a paucity of collecting ducts, and a defect in urine concentration capacity. In contrast, deletion of Tcf21 from the cap mesenchyme leads to abnormal glomerulogenesis and massive proteinuria, but no downregulation of Gdnf-Ret-Wnt11 or obvious defect in branching. Conclusions Our findings indicate that Tcf21 has distinct roles in the cap mesenchyme and stromal mes- enchyme compartments during kidney development and suggest that Tcf21 regulates key molecular pathways required for branching morphogenesis.

J Am Soc Nephrol 29: ccc–ccc, 2018. doi: https://doi.org/10.1681/ASN.2017121278

Received December 12, 2017. Accepted September 27, 2018.

Development of the mouse metanephros begins at S.I. and G.F. contributed equally to this work. approximately embryonic day (E) 10.5 with induc- Published online ahead of print. Publication date available at tion of the ureteric bud (UB) at the posterior end of www.jasn.org. the nephric duct (also known as the Wolffian duct) by signals from the adjacent metanephric mesen- Correspondence: Dr. Yoshiro Maezawa, Clinical Cell Biology and Medicine and Department of Diabetes, Metabolism and 1,2 chyme (MM). TheUBinvadestheMM,elon- Endocrinology, Chiba University Graduate School of Medicine, 1- gates, and branches to form the renal collecting 8-1 Inohana, Chuo-ku, Chiba 260-8670, Japan, or Dr. Susan E. system. As each branch forms, a subset of mesen- Quaggin, Division of Nephrology/Hypertension, Feinberg Car- diovascular and Renal Research Institute, Northwestern Univer- chymal cells aggregate and condense around the UB sity Feinberg School of Medicine, Suite 10-105, 303 E. Superior tips to form the Six2-expressing cap mesenchyme Street, Chicago, IL 60611. E-mail: [email protected] or (CM). This structure further differentiates into pre- [email protected] tubular aggregates, comma- and S-shaped bodies, Copyright © 2018 by the American Society of Nephrology

J Am Soc Nephrol 29: ccc–ccc, 2018 ISSN : 1046-6673/2912-ccc 1 BASIC RESEARCH www.jasn.org and subsequently into the Bowman’s capsule of the glomeru- Significance Statement lus, podocytes, and the nephron down to the distal tubule.3–5 The Six2-positive CM is surrounded by Foxd1-expressing Branching morphogenesis of the ureteric bud is central to forming a stromal cells that are also derived from the MM and give normal kidney, and the most severe forms of congenital anomalies of rise to interstitial cells, mesangial cells, and pericytes.6 Disrup- the kidney and urinary tract (CAKUT) arise from mutations in involved in branching. The authors report that deletion in mice of tion of any of these events results in congenital anomalies of transcription factor 21 (Tcf21) results in a spectrum of renal de- the kidney and the lower urinary tract (CAKUT); these rep- velopmental phenotypes that resemble human CAKUT. Germline resent 20%–30% of all anomalies identified in the prenatal deletion of Tcf21 or Tcf21 deleted specifically from the stromal period in humans.7–10 mesenchyme resulted in branching defects and reduced expression Gdnf-Ret-Wnt11 Cell type–specific basic helix-loop-helix transcription of (a key pathway required for branching mor- phogenesis), whereas Tcf21 deletion specifically from the cap 11–14 factors are key regulators of organ morphogenesis. mesenchyme resulted in glomerular rather than branching defects Transcription factor 21 (Tcf21) (Pod1/Capsulin/Epicardin) and no downregulation of Gdnf-Ret-Wnt11. These findings suggest is a class 2 basic helix-loop-helix transcription factor that that Tcf21 has a central role in regulating the Gdnf axis and renal showsrobustexpressioninmesenchymalcellsoftheuro- stromal factors crucial for branching. genital, respiratory, digestive, and cardiovascular systems – throughout embryogenesis and is involved in epithelial were kind gifts from Dr. Andy McMahon (University of mesenchymal interactions in the kidney, , and other Southern California), and are described elsewhere.5,26 15–18 organs. In disease, Tcf21 is reported to function as a Rosa26-mT/mG reporter mice (JAX stock no. 007676) were tumor suppressor and loss of Tcf21 expression and/or meth- obtained from The Jackson Laboratory. ylation of its have been identified in a variety of lung and urogenital tumors.19,20 In other studies, Tcf21 has been linked to ,21,22 podocyte differ- Immunostaining entiation, diabetic kidney disease,23 and changes in adipose For immunohistochemistry and lectin stainings, kidneys were fi phenotype.24 dissected and xed by 10% formalin or 4% paraformaldehyde fi In the kidney, previous work has shown that Tcf21 null overnight. Tissues were then embedded in paraf n and sliced m mutant mice are born with severely hypoplastic kidneys, and into 5- m-thick sections. Before immunostainings, sections fi that although Tcf21 is exclusively expressed in the mesen- were deparaf nized in xylene and ethanol and boiled in cit- fl chyme, in its absence, major defects in UB epithelial differen- rate buffer for antigen retrieval. For immuno uorescence, af- fi tiation and branching are observed.15,25 Currently, it remains ter xing by 4% paraformaldehyde overnight, tissues were unclear what signaling pathways are controlled by Tcf21. cryoprotected in 30% sucrose overnight and embedded in In this study, we report that Tcf21 regulates branching mor- Tissue-Tek O.C.T 4583 compound (Sakura Finetek Japan 2 m phogenesis by modifying the Gdnf-Ret-Wnt11 axis and pro- Co., Ltd., Tokyo, Japan) at 80°C and sliced into 10- m-thick vide data to support its pleiotropic functional roles that affect sections. UB-MM-stromal crosstalk. Using the Cre-LoxP system we These sections were blocked (0.3% Triton X-100, 3% show that Tcf21 has distinct roles in Foxd1-positive stromal albumin, 1 mM MgCl2, 1 mM CaCl2 in PBS) for 1 hour, cells and Six2-positive CM cells; where selective deletion of incubated with primary antibodies at 4°C overnight. After Tcf21 from the stromal cells results in branching defects, a washing several times by PBS, they were incubated with paucity of collecting ducts, and urine concentrating defects. secondary antibodies for 1 hour at room temperature. fl On the other hand, Tcf21 deletion in the CM leads to defects in Fluorescence-labeled Dolichos bi orus agglutinin (DBA) glomerulogenesis and massive proteinuria. or Lotus tetragonolobus lectin (LTL) was mixed with sec- ondary antibodies. For detailed methods and a full list of antibodies, see Supplemental Material and Supplemental METHODS Table 1.

Ethics Statement/Study Approval Real-Time PCR All mouse experiments were performed in accordance with Whole kidneys were dissected and preserved at 280°C after institutional guidelines for animal studies. All animal experi- freezing in liquid nitrogen. They were treated by the ho- ments were approved by the Animal Care Committee at the mogenizer and QIA shredder (79656; Qiagen) and total Center for Comparative Medicine of Northwestern University RNA was extracted by using Pure link RNA minikit (Evanston, IL) or by the Chiba University Ethics Committee (12183025; Life Technologies). Reverse transcription was (Chiba, Japan). performed using Super Script III Reverse transcription (18080044; Invitrogen) according to the manuals. Quanti- Mice and Genotyping tative real-time PCR was performed using Fast SYBR Green Tcf21 lox/lox and Tcf21-LacZ mice were created as previously Master Mix (Thermo Fisher Scientific). Data were analyzed described.23,25 The Six2-eGFPCre and FoxD1-eGFPCre mice as the ratio to GAPDH.

2 Journal of the American Society of Nephrology J Am Soc Nephrol 29: ccc–ccc,2018 www.jasn.org BASIC RESEARCH

A Control Tcf21 null E Control Tcf21 null Ret (E14.5) Calbindin / Jagged1

B 60 60

40 40 ** 20 *** 20 UB tips number number Wnt11 (E14.5)

0 Jagged1(+) spot 0

Tcf21 Tcf21 Control null Control null F Control Tcf21 null

C Gdnf Ret Wnt11 1.5 1.5 2.0

1.5 1.0 1.0

1.0 Gdnf (E14.5) ** 0.5 ** 0.5 0.5 Fold change *** 0.0 0.0 0.0

Tcf21 Tcf21 Tcf21 Control null Control null Control null

D Control Tcf21 null E-cadherin (E14.5) Gdnf (E12.5) Gdnf (E16.5) Gdnf (E14.5) E-cadherin (E16.5)

Figure 1. UB branching is impaired and Gdnf, Ret, and Wnt11 are downregulated in Tcf21 null kidneys. (A and B) Reduced number of UB tips (green, calbindin) and renal vesicles (red, Jagged1) in Tcf21 null kidney explants. Kidneys were dissected at E12.5, incubated for 48 hours and subjected to immunostainings. Scale bar, 50 mm. (C) Downregulation of Gdnf, Ret,andWnt11 mRNA expression in Tcf21 null kidneys at E14.5 by quantitative RT-PCR. (D) Low expression of Gdnf mRNA in the mesenchyme of Tcf21 null kidneys at E12.5 and E14.5 by in situ hybridization. Scale bar E12.5, 100 mm; E14.5, 50 mm. (E) Paucity of Ret and Wnt11 mRNA at the UB tips by in situ hybridization in Tcf21 null kidneys at E14.5. Scale bar, 50 mm. (F) Immunostaining of Gdnf and the UB marker E-cadherin of Tcf21 null kidneys and controls at E14.5 and E16.5 showing low expression of Gdnf and abnormal UB branching. Scale bar, 100 mm. *P,0.05; **P,0.01; ***P,0.001.

J Am Soc Nephrol 29: ccc–ccc, 2018 Transcription Factor 21 and Kidney Development 3 BASIC RESEARCH www.jasn.org

Explant Culture of Mouse Embryonic Kidneys A Embryos were dissected from pregnant mice on day 12.5 E18.5 Six2 Pax2 Wt1 postcoitum. Kidneys from embryos were cultured on Nucle- 1.5 n.s. 1.5 n.s. 4.0 n.s. opore Track-Etched Membranes (Whatman) floated on cul- 3.0 ture medium. After 48 hours, kidneys were fixed in 4% PFA 1.0 1.0 2.0 and stained for Calbindin and Jagged1. The number of 0.5 0.5 branching and Jagged1-positive spots were counted in a double- 1.0 Fold change blind manner. 0.0 0.0 0.0

Tcf21 Tcf21 Tcf21 mRNA In Situ Hybridization Control null Control null Control null mRNA in Situ hybridization was performed on formalin- B fixed, paraffin-embedded sections using the RNAscope 2.5 E14.5 Control Tcf21 null assay system (Advanced Cell Diagnostics) with RNAscope FFPE Reagent Kit, 2.5 HD-Reagent Kit-RED, 2.5 HD-Reagent Kit-BROWN. Recommended probes were used for this assay. SIX2

Statistical Analyses Six2 / E-cadherin Statistical analyses were carried out using GraphPad Prism 6.0 (GraphPad Software Inc.). Comparison of two groups was done by two-tailed t-test. A P-value ,0.05 was considered significant. Error bars showed 6SEM. PAX2 RESULTS Pax2 / E-cadherin

Tcf21 Regulates Branching Morphogenesis via Gdnf- Ret-Wnt11 Signaling Germline deletion of Tcf21 leads to hypodysplastic kidneys reminiscent of CAKUT in humans (Supplemental Figure 1).

At E12.5 +48 hours, Tcf21 null explant cultures show a very Wt1 abnormal UB tree (Figure 1, A and B). To determine how

Tcf21 regulates UB branching and collecting duct develop- Wt1 / E-cadherin ment, we first examined the expression of Gdnf, which is crit- ical for branching morphogenesis. Gdnf transcript level was , Figure 2. Absence of Tcf21 does not affect quantitative ex- markedly downregulated to 40% in Tcf21 null kidneys pression of Six2, Pax2, and Wt1 in Tcf21 null kidneys at E14.5 at E14.5 compared with wild-type by quantitative RT-PCR and E18.5. (A) Quantitative RT-PCR on whole kidney at E18.5 (Figure 1C). By in situ hybridization at E12.5 and E14.5 and showing no significant change in mRNA expression levels of by immunohistochemistry at E14.5 and E16.5, Tcf21 null kid- Six2, Pax2,andWt1 in Tcf21 null and control kidneys. (B) Im- neys also exhibit reduction of Gdnf (Figure 1, D and F). Next, munostaining of Six2, Pax2, and Wt1 at E14.5 shows largely in- we examined the expression of Ret, the cognate tyrosine kinase tact CM in Tcf21 null kidneys and controls. Scale bars, 200 or 100 for Gdnf, and that of Wnt11, a growth factor that is mm as indicated. necessary to maintain Gdnf expression. Similar to Gdnf, both Ret and Wnt11 transcript levels were decreased in Tcf21 null kidneys at E14.5, consistent with paucity of UB branch tips Further, the transcription factors that regulate Gdnf (Osr1, (Figure 1, C and E). The quantitative RT-PCR results were Eya1, Pax2, Six2, Hoxa11, Hoxd11, Wt1, Sall1,andGata3) normalized to Gapdh to account for size difference of the kid- were not decreased in Tcf21 null kidneys (Figure 2, Supple- neys. These results suggest that Tcf21 is required for normal mental Figure 2). We next examined the potential role of Tcf21 expression of Gdnf-Ret-Wnt11 and therefore is critical for UB in non-Gdnf–dependent pathways that regulate UB branch- branching. On the other hand, the expression pattern of mark- ing: fibroblast growth factor (Fgf), canonical Wnt, and reti- ers for CM (Six2, Pax2, and Wt1) was not decreased by quan- noic acid signaling. At E14.5 and E18.5, none of the titative RT-PCR and immunostaining in Tcf21 null kidneys transcripts of Aldehyde dehydrogenase family 1, subfamily a2 (Figure 2, A and B). This suggests that the decrease of Gdnf (Aldh1a2), Retinoic acid receptora (Rara), Fgf7, Fgf10, Reti- in Tcf21 null kidneys is not the result of loss of nephron pro- noic acid receptor b (Rarb), and Wnt4 were significantly genitor population, consistent with previous experiments.27 downregulated (Supplemental Figure 3). Taken together,

4 Journal of the American Society of Nephrology J Am Soc Nephrol 29: ccc–ccc,2018 www.jasn.org BASIC RESEARCH these results support a selective role for Tcf21 in regulating ROSA26-mTmG reporter mice. Immunostaining demon- the Gdnf axis. strated that Tcf21 is distributed both in a subset of the Six2- positive CM and in the derivatives of Foxd1-positive stromal Tcf21 Has Distinct Roles in the Stromal and CM during mesenchyme (Figure 3A). We hypothesized that the defect Kidney Development observed in UB branching and Mesenchymal to epithelial When MM cells begin to aggregate around the UB tips, the transformation in Tcf21 null kidneys represented convergence mesenchyme segregates to CM (expressing Cited1 and Six2) of signals from these distinct cell lineages that arise from and to stromal mesenchyme (expressing high levels of the MM. To dissociate these effects, we generated two Foxd1).28 We evaluated Tcf21 in each cell Cre/LoxP mouse models, StrTcf21 and CapTcf21, where compartment using Tcf21 LacZ/Wild, Foxd1-eGFPCre; Tcf21 is selectively deleted from the kidney stromal cells under

A Tcf21;beta-Galactosidase/FoxD1-Cre;(Rosa26)mTmG/Six2

LacZ=Tcf21 Stroma (FoxD1)

Cap Mesenchyme (Six2) Merge

B i ii iii iv

CapM CapM CapM CapM MM MM MM MM

UB

StrM UB StrM UB StrM UB StrM

Wild Type Tcf21 null CapTcf21 StrTcf21

Figure 3. Tcf21 is expressed in a subset of the CM and stromal mesenchyme. (A) Immunostaining of the kidney from Tcf21LacZ/+; FoxD1Cre; (ROSA26)mT/mG triple transgenic mice. Tcf21 LacZ expression by b-galactosidase staining (red) delineating normal expression of Tcf21 at E18.5. Derivatives of FoxD1-positive stromal mesenchyme was visualized by GFP using FoxD1-Cre;(ROSA26)mT/mG reporter mice (green). CM is represented by Six2 immunostaining (blue). Bar, 100 mm. (B) Schematic depiction of Tcf21 deletion in the mouse models utilized. White areas represent Tcf21-deleted cell lineages. (i) Wild-type; (ii) Tcf21 null kidney, Tcf21 is deleted in germline cells and hence from the entire MM and its derivatives; (iii) CapTcf21 kidney, Tcf21 is selectively deleted from the CM under Six2-eGFPCre driver; (iv) StrTcf21 kidney, Tcf21 is selectively deleted from the stromal mesenchyme under Foxd1-eGFPCre driver.

J Am Soc Nephrol 29: ccc–ccc, 2018 Transcription Factor 21 and Kidney Development 5 BASIC RESEARCH www.jasn.org

A D Control StrTcf21 Control StrTcf21 LTL / Uromodulin

B Control StrTcf21 DBA / Aquaporin2

E 15

Glomerulus 10

5 ** Aquaporin2 (+) area

/Total kidney area (%) 0 Control StrTcf21

Stroma F Control StrTcf21

C 15 O)

2 2000

**** Calbindin

10 1500

1000 **** 5 500 G 24h Urine volume(ml)

0 Osmolality (mosm/kgH 0 80 Control StrTcf21 Control StrTcf21 60 *

40

20 UB tips number 0 Control StrTcf21

6 Journal of the American Society of Nephrology J Am Soc Nephrol 29: ccc–ccc,2018 www.jasn.org BASIC RESEARCH

FoxD1-eGFPCre driver, and from the CM under Six2-eGFPCre stroma, whereas expression of members of the retinoic acid driver, respectively (Figure 3B, schematic illustration). and Fgf pathways do not appear to be Tcf21-dependent (Figure 5, D and E). Although the stroma of StrTcf21 Deletion of Tcf21 in Stromal Progenitors Results in mice appears disorganized, mRNA levels of stromal markers Abnormal UB Branching, Paucity of Collecting Ducts, were not depressed in whole-kidney lysate from mutants at and Disorganized Stroma E14.5 and P0 supporting normal stromal cell mass (Supple- Several lines of evidence support a role for Tcf21 in the devel- mental Figure 6). opment of the interstitial stroma.27 To study Tcf21’s function in the stroma, Tcf21 lox/lox mice were bred to Foxd1-eGFPCre Tcf21 in the CM Is Necessary for Proper Development mice to generate a stromal-specific Tcf21 knockout mice of the Glomerulus and Tubules but Not for the (StrTcf21). In this model, Tcf21 mRNA expression in the Collecting Ducts stroma is markedly decreased starting at E14.5 onwards, al- Deletion of Tcf21 from the CM (CapTcf21) results in smaller though it is still expressed at E12.5 (Supplemental Figure 4). kidneys (Figure 6A, Supplemental Figure 7A) and decreased WhentheStrTcf21micewereevaluated,theyappeared urine volume at P0 and proteinuria at 3 days of age (Figure healthy and survived for at least 10 weeks. On gross dissection, 6B). Histologically at P0, the glomeruli of CapTcf21 show the StrTcf21 kidneys are reduced in size at birth and their primitive and simplified structures. The tubules are cystic histology reveals normal glomeruli but disorganized stroma and atrophic, but the stromal structure does not show obvious (Figure 4, A and B, Supplemental Figure 5A). Notably, at 4 changes (Figure 6C). The simplified glomeruli show decreased weeks of age the StrTcf21 animals showed increased urine Podocin and Desmin, suggestive of a defect in mesangial in- volume and a severe defect in urine concentration capacity growth and/or mesangiolysis (Figure 6D), and ultrastructur- (Figure 4C). Corresponding with this finding is the severe re- ally, the podocytes have effaced foot processes (Supplemental duction of collecting ducts in StrTcf21 kidneys that markedly Figure 7B). The loss of tubular mass is also shown by reduced outnumber the mild reduction in proximal tubules (Figure 4, area of LTL-expressing proximal tubules and Uromodulin- D and E, Supplemental Figure 5, B and C). The paucity of positive thick ascending limb (Figure 6E). These results indi- collecting ducts is also evident by reduction of Aquaporin 2, cate that loss of Tcf21 in Six2-positive progenitors leads Calbindin, DBA, Epithelial sodium channel,andNa-H-ATPase to defects in glomerulogenesis, podocyte differentiation, and (Figure 4D, Supplemental Figure 5, C and D). To investigate tubular development. In contrast, absence of Tcf21 does not whether the loss of collecting ducts stemmed from abnormal obviously affect the distribution of the collecting ducts as UB development, we examined the UB tree in kidney explants shown by similar DBA-positive area between CapTcf21 and and noticed fewer UB tips in the StrTcf21 kidneys at E12.5 +48 controls (Figure 6F). In line with these findings are the results hours, suggesting abnormal branching although we cannot from kidney explants that show reduction in both renal vesi- rule out the possibility that artifacts from culturing for 48 cles (Jagged1) and proximal tubules (LTL), but only a mild hours may be different in mutants versus controls. Addi- decrease in UB tip number in CapTcf21 explants (Figure 7, A tionally, Wnt7b was decreased at E14.5 (Figures 4, F and G and B). The minimal effect of Tcf21 on branching in CapTcf21 and 5C). Wnt7b is a UB stalk maker important to cortico- kidneys is also noted by E-cadherin staining at E14.5 and sim- medullary organization and function.29 Further supportive ilar mRNA levels of UB markers Wnt7b and Sox9 (Figure 7C, of a primary branching defect is the finding of low expres- Supplemental Figure 7C). Importantly, loss of Tcf21 in the CM sion of Gdnf and Wnt11 mRNA in the StrTcf21 kidney at did not attenuate expression of Gdnf, Ret,orWnt11 transcripts E14.5 and P0 (Figure 5, A and B, Supplemental Figure 5, E at E14.5 (Figure 7D). Taken together, these results suggest that and F). Of note, although Ret mRNA level is not reduced although Tcf21 in the CM is required for the normal develop- here, its downstream effectors Etv4 and Etv5 are signifi- ment of the glomeruli and tubules, it is not necessary for nor- cantly downregulated at E14.5 and E18.5 in StrTcf21 kid- mal Gdnf axis and development of the collecting ducts. neys (Figure 5A, Supplemental Figure 5E). Taken together, these results suggest that Tcf21 expression in stromal cells is Bmp4 Is Upregulated in Tcf21 Null Kidneys required for normal expression of the Gdnf axis, develop- As described above, decreased Gdnf levels were noted only in ment of the collecting ducts, and normal organization of the Tcf21 null and StrTcf21 kidneys, whereas Gdnf had normal

Figure 4. Absence of Tcf21 from the renal stroma does not affect glomerulogenesis but results in urine concentration defect, abnormal collecting ducts, and branching morphogenesis. (A and B) StrTcf21 kidneys are reduced in size, have normal glomeruli, and disor- ganized stroma ([B] Scale bars: top, 500 mm; middle, 25 mm; bottom, 50 mm). (C) StrTc21 mice develop polyuria and urine concen- tration defect at 4 weeks. (D and E) StrTc21 kidneys show severe decrease of inner medulla. Thick ascending limbs that express Uromodulin and collecting ducts that express Aquaporin 2 are largely reduced. Scale bar, 500 mm. (F and G) Abnormal UB branching in kidney explants of StrTc21 dissected at E12.5. Scale bar, 50 mm. *P,0.05; **P,0.01; ***P,0.001; ****P,0.001.

J Am Soc Nephrol 29: ccc–ccc, 2018 Transcription Factor 21 and Kidney Development 7 BASIC RESEARCH www.jasn.org

A C Gdnf Ret Wnt11 Wnt7b Ecad 1.5 1.5 1.5 1.5 2.5 n.s. 2.0 n.s. 1.0 1.0 1.0 1.0 ** 1.5 * 1.0 0.5 0.5 0.5 0.5 Fold change Fold change ** 0.5 0.0 0.0 0.0 0.0 0.0

Control Control Control Control Control StrTcf21 StrTcf21 StrTcf21 StrTcf21 StrTcf21

Etv4 Etv5 D 2.0 2.0 Control StrTcf21 1.5 1.5

1.0 1.0 * *

Fold change 0.5 0.5 Aldh1a2 0.0 0.0

Control Control StrTcf21 StrTcf21

B E Fgf7 Fgf10 Control StrTcf21 1.5 2.0 n.s. n.s. 1.5 1.0 1.0 0.5 Gdnf

Fold change 0.5

0.0 0.0

Control Control StrTcf21 StrTcf21 Aldh1a2 Rara 1.5 1.5

1.0 n.s. 1.0 n.s.

0.5 0.5 Fold change

0.0 0.0

Control Control StrTcf21 StrTcf21 Wnt11 Ret

Figure 5. Absence of Tcf21 from the renal stroma leads to downregulation of the Gdnf axis at E14.5. (A and B) StrTcf21 kidneys have low mRNA of Gdnf and Wnt11 on quantitative real-time PCR and in situ hybridization. Ret transcript is not reduced but Etv4 and Etv5, downstream effectors of Ret, are downregulated in this model. Scale bar, 50 mm. (C) Reduced expression of the UB stalk marker Wnt7b, shown by quantitative real-time PCR. Expression of the pan-epithelial marker E-cadherin is not decreased. (D and E) mRNA expression of retinoic acid and Fgf signaling components that are critical for Gdnf axis are not significantly changed by in situ hybridization and quantitative real-time PCR. Scale bar, 100 mm. *P,0.05; **P,0.01.

8 Journal of the American Society of Nephrology J Am Soc Nephrol 29: ccc–ccc,2018 www.jasn.org BASIC RESEARCH

A D Control CapTcf21

Control CapTcf21 Podocin/ CD31

Control CapTcf21 B 20 **** 80 15 60

*** 10 Desmin/ CD31 40 ***

Urine / 5 20 Urine volume(µl) 0 creatinine ratio (mg/mg) 0 E Control CapTcf21 P3P0

C Control CapTcf21 LTL / Uromodulin

15 4

Glomerulus 3 10 ** 2 * 5 1 LTL (+) area

Uromodulin (+) area 0 0 /Total kidney area (%) /Total kidney area (%) Control CapTcf21 Control CapTcf21

F DBA Stroma Tubules

15 n.s. 10

5 DBA (+) area 0

/Total kidney area (%) Control CapTcf21

Figure 6. Absence of Tcf21 from the CM leads to abnormal glomeruli and tubules but does not affect the collecting ducts. (A) Absence of Tcf21 from the CM results in smaller kidneys at P0. (B) Urine volume is reduced and proteinuria is evident at P0 and P3 in CapTcf21 mice (volume measured by aspirating bladder content at the time of dissection). (C) Periodic acid–Schiff staining shows small and simplified glomeruli and cystic tubules in CapTcf21 mice. The histology of the renal stroma is not different from controls.

J Am Soc Nephrol 29: ccc–ccc, 2018 Transcription Factor 21 and Kidney Development 9 BASIC RESEARCH www.jasn.org expression in the CapTcf21 model, suggesting that branch- The expression of Tcf21 in the developing kidney is ing is dependent on Tcf21 expression in the stroma. Bmp4 complex.27 Although Tcf21 is not expressed in metaneph- is a member of the TGF-b superfamily that is expressed rogenic blast cells found at the periphery of the metaneph- throughout , initially in stromal ros before induction and condensation, it is expressed cells enveloping the nephric duct.30–35 BMP4 is one of the in both the Six2-expressing CM and in FoxD1-expressing key regulators of UB branching, inhibiting ectopic UB out- stromal cells. Somewhat unexpectedly, the branching de- growth by decreasing expression of Gdnf.34,36–38 Previous fects observed in conventional Tcf21 knockout mice were work showed that Tcf21-lacZ–expressing cells are present only observed in mice lacking Tcf21 from the stromal com- around the stalk of the UB and express Bmp4.27 Here, partment, whereas glomerular defects were observed in Bmp4 protein and mRNA levels were increased in Tcf21 CapTcf21 mutants. In addition, Gdnf-Ret-Wnt11 downre- null kidneys at E18.5 (Figure 8, A and B). Upregulation gulation was only observed in StrTcf21 and total knockout of Bmp4 was also seen in our previously conducted gene- mice. expression microarray analysis on total RNA extract from What can be learned from these distinct phenotypes? glomeruli of Tcf21 lacZ/lacZ null mice at E18.5 (Mouse Affy- The most likely explanation is that Tcf21 has distinct metrix GeneChips 430–2.0)39 (Figure 8C). In silico analysis functions in the stromal versus cap mesenchymal line- showed that the 3000 bp promoter and 59 UTR regions of ages. Interestingly, the phenotype of the Tcf21 null mice the Bmp4 gene contain the canonical binding sequences of was much more severe than that observed in either Tcf2140 (Figure 8D). Thus, Bmp4 expression increase in StrTcf21 or CapTcf21 kidneys, suggesting that the Tcf21 Tcf21 null kidneys may suggest its involvement in Tcf21- null phenotype is not a simple sum of each compartmen- mediated effects. talized phenotype. Although this may reflect timing or completeness of gene inactivation, other explana- tions are possible, including crosstalk between Tcf21- DISCUSSION expressing cell types or existence of additional subset(s) of mesenchymal cell types not represented by Six2 or In the mammalian kidney, interactions between the UB and the Foxd1 expression. MM promote growth and differentiation to generate a functional The role of the stromal cells as a key regulator of nephro- organ of appropriate size, shape, and cellular diversity.41 Per- genesis and branching morphogenesis has previously been turbations to these processes lead to CAKUT that, in many shown in Foxd1 null mice.50–52 Levinson et al.51 showed cases, stem from disruption of genes involved in renal organ- that Foxd1 null mice are born with hypodysplastic and ogenesis.42,43 The “budding theory” proposes that hypodys- fused midline kidneys, a similar but not identical renal plastic kidneys in CAKUT are the result of defects in UB phenotype observed in Tcf21 null mice. The importance branching.44,45 UB epithelium branching is principally con- of stromal Tcf21 for UB branching has been previously trolled by the Gdnf-Ret pathway and by the Wnt, Fgf, and suggested in Tcf21LacZ/GFP chimeras, alluding to a non- retinoic acid signaling pathways, but many other genes are cell-autonomous role of Tcf21 in the stroma.27 We now involved.2,46–48 Importantly, mutations in RET and GDNF, have shown that removal of Tcf21 in Foxd1-positive cells as well as in other genes that regulate branching of the UB, leads to downregulation of Gdnf-Ret-Wnt11,providinga have been associated with CAKUT.10,42 Although a functional potential mechanism to explain the phenotypes of both role of Tcf21 in human CAKUT has not yet been reported, Tcf21 and Foxd1 knockout mice. Whether Tcf21 expression Tcf21 has been shown to be downregulated in dysplastic fetal in the stroma is required for normal stromal development kidneys.49 is not completely clear. Here, we show that removal of Here, we report a spectrum of renal developmental phe- Tcf21 from stromal progenitors results in disorganized notypes that resemble human CAKUT that result from loss of interstitium that has normalexpressionofstromalcell Tcf21 from the germline, or from specificMMcompart- markers, suggesting that Tcf21 is not required for the de- ments. Kidneys from conventional Tcf21 knockout mice velopmentofnormaltotalstromalcellmassbutthatitmay show arrested Mesenchymal to epithelial transformation, have a role in stromal cell alignment. Lastly, identifying with malformed and very small kidneys. We also direct gene targets for Tcf21 in the kidney is crucial for identified a reduction in expression of the Gdnf-Ret-Wnt11 understanding its mechanistic action. In that regard, the signaling loop critically required for branching morphogen- observed increased levels of Bmp4, a stromal factor and a esis in null Tcf21 embryos. potent inhibitor of UB branching and Gdnf,33,34,53–56 is

Scale bar, 50 mm. (D) CapTcf21 kidneys demonstrates glomerular injury by reduced Podocin and Desmin. Scale bars, 25 mm. (E) Reduced area of proximal tubules (LTL) and thick ascending limb (Uromodulin) in CapTcf21 kidneys. Scale bar, 500 mm. (F) The collecting ducts (DBA) are thick but the area is similar in CapTcf21 compared with controls. Scale bar, 500 mm. *P,0.05; **P,0.01; ***P,0.001; ****P,0.001.

10 Journal of the American Society of Nephrology J Am Soc Nephrol 29: ccc–ccc,2018 www.jasn.org BASIC RESEARCH

A Control CapTcf21 / LTL / Jagged1 Calbindin

B 40 15 60 ** 30 10 *** 40 20 ***

number 5 20 10 UB tips number Jagged1(+) spot

0 LTL(+) spot number 0 0 Control CapTcf21 Control CapTcf21 Control CapTcf21

C Control CapTcf21 Ecad / Six2

D Gdnf Ret Wnt11 2.0 6.0 * 2.0 n.s. n.s. 1.5 1.5 4.0 1.0 1.0 2.0 0.5 0.5 Fold change

0.0 0.0 0.0 Control CapTcf21 Control CapTcf21 Control CapTcf21

Figure 7. Absence of Tcf21 from the CM leads to abnormal development of the glomeruli and tubules but does not affect the development of the collecting ducts or Gdnf axis expression. (A and B) Kidney explant assay at E12.5 for 48 hours of CapTcf21 and controls shows severely reduced Jagged1 (renal vesicle)- and LTL (proximal tubule)-positive spots, but counting of ureteric bud tips showed only mild decrease. Scale bar, 50 mm. (C) E-cadherin (green) immunostaining of CapTcf21 and control kidneys at E14.5 showing similar patterns of UB development. Scale bar, 100 mm. (D) Quantitative RT-PCR results of Gdnf, Wnt11, or Ret transcripts at E14.5 showing no reduction in the Gdnf axis in the absence of Tcf21 from the CM. *P,0.05; **P,0.01; ***P,0.001. intriguing; however, further study is required to verify di- that are critical for normal nephrogenesis. In addition, stro- rect regulatory effect. mal expression of Tcf21 is needed to allow normal In summary, the results support a model whereby UB branching morphogenesis via the Gdnf-Ret-Wnt11 Tcf21 has distinct roles in cap versus stromal MM autoregulatory loop (Supplemental Figure 8), underscoring

J Am Soc Nephrol 29: ccc–ccc, 2018 Transcription Factor 21 and Kidney Development 11 BASIC RESEARCH www.jasn.org

A C E18.5 Gene Probe ID FC Bmp4 Bmp4 * BMP4 1422912_at 2.66 10.0 3.0 8.0 2.0 D 6.0 * BMP4 4.0 3000bp 1.0 5’ UTR Exon1 2.0 Fold change Bmp4/beta-Actin 0.0 0.0 Control Tcf21 Control Tcf21 MA0832.1

null null 2.0

1.5

1.0

0.5 Control Tcf21 nullControl Tcf21 null 0.0 1234567891011121314

Relative Name Score Start End score Bmp4/ β -Actin Tcf21 9.83019 0.818755 822 835 B Tcf21 10.8315 0.834641 822 835 E18.5 Control Tcf21 null Tcf21 9.7289 0.817148 841 854 Tcf21 9.90098 0.819879 841 854 Tcf21 14.5714 0.893978 993 1006 Tcf21 14.5299 0.89332 1006993 Bmp4

Figure 8. Bmp4, a stromal factor and inhibitor of Gdnf, is upregulated in Tcf21 null mutants. (A and B) In Tcf21 null kidneys at E18.5, Bmp4 protein and mRNA are increased by Western blot, quantitative RT-PCR, and in situ hybridization, respectively. Scale bar, 50 mm. (C) Upregulation of Bmp4 by gene-expression microarray analysis on total RNA extract from glomeruli of Tcf21 null mice at E18.5. (D) Graphic depiction of Tcf21 binding site consensus sequences and their location relative to the Bmp4 gene promotor and 59 UTR (JASPAR database). *P,0.05. the importance of stromal-cap-MM-UB crosstalk in kidney drafted and revised the paper; all authors approved the final version of development. the manuscript. ThisworkwassupportedbyJapanSocietyforPromotion of Science grants 17K09687 (to Y.M.), 17K19643 (to K.Y.), and ACKNOWLEDGMENTS 17J03714 (to S.I.), by the National Kidney Foundation of Illinois Young Investigator Award (to G.F.), and by National Institutes We are grateful to Mrs. Aki Watanabe and Naoko Koizumi for of Health grants HL124120, EY025799, T32 DK108738, and technical support, and Dr. Akira Suto and Dr. Hiroshi Nakajima for P30DK114857 (to S.E.Q.). technical supports and fruitful discussion. S.E.Q., G.F., Y.M., and S.I. designed the study; S.I., Y.M., G.F., M.Y., X.Z., Y.B., T.M., R.S., C.L., K.I., T.O., and Y.A. carried out experiments; DISCLOSURES G.F., S.I., Y.M., T.S., R.S., J.J., S.E.Q., M.T., and K.Y., analyzed the data; S.E.Q. owns stock in and is a director of Mannin Research, and is on the G.F., Y.M., and S.I., made the figures; G.F., Y.M., S.I., and S.E.Q. external advisory board of Astra Zeneca.

12 Journal of the American Society of Nephrology J Am Soc Nephrol 29: ccc–ccc,2018 www.jasn.org BASIC RESEARCH

REFERENCES 21. Kim JB, Pjanic M, Nguyen T, Miller CL, Iyer D, Liu B, et al.: TCF21 and the environmental sensor aryl-hydrocarbon receptor cooperate to fl 1. Costantini F, Kopan R: Patterning a complex organ: Branching mor- activate a pro-in ammatory gene expression program in coronary ar- phogenesis and nephron segmentation in kidney development. Dev tery smooth muscle cells. PLoS Genet 13: e1006750, 2017 Cell 18: 698–712, 2010 22. Lu X, Wang L, Chen S, He L, Yang X, Shi Y, et al.: Coronary ARtery 2. Little MH, McMahon AP: Mammalian kidney development: Principles, DIsease Genome-Wide Replication And Meta-Analysis (CARDIo- progress, and projections. Cold Spring Harb Perspect Biol 4: a008300, 2012 GRAM) Consortium: Genome-wide association study in Han Chinese fi 3. Dressler GR: The cellular basis of kidney development. Annu Rev Cell identi es four new susceptibility loci for coronary artery disease. Nat – Dev Biol 22: 509–529, 2006 Genet 44: 890 894, 2012 4. Vainio S, Lin Y: Coordinating early kidney development: Lessons from 23. Maezawa Y, Onay T, Scott RP, Keir LS, Dimke H, Li C, et al.: Loss of the gene targeting. Nat Rev Genet 3: 533–543, 2002 podocyte-expressed transcription factor Tcf21/Pod1 results in podo- – 5. Kobayashi A, Valerius MT, Mugford JW, Carroll TJ, Self M, Oliver G, cyte differentiation defects and FSGS. J Am Soc Nephrol 25: 2459 et al.: Six2 defines and regulates a multipotent self-renewing nephron 2470, 2014 progenitor population throughout mammalian kidney development. 24. de Jong JM, Larsson O, Cannon B, Nedergaard J: A stringent validation Cell Stem Cell 3: 169–181, 2008 of mouse adipose tissue identity markers. Am J Physiol Endocrinol – 6. Kobayashi A, Mugford JW, Krautzberger AM, Naiman N, Liao J, Metab 308: E1085 E1105, 2015 McMahon AP: Identification of a multipotent self-renewing stromal 25. Quaggin SE, Schwartz L, Cui S, Igarashi P, Deimling J, Post M, et al.: The progenitor population during mammalian kidney organogenesis. Stem basic-helix-loop-helix protein pod1 is critically important for kidney and – Cell Reports 3: 650–662, 2014 lung organogenesis. Development 126: 5771 5783, 1999 7. Pope JC 4th, Brock JW 3rd, Adams MC, Stephens FD, Ichikawa I: How 26. Humphreys BD, Lin SL, Kobayashi A, Hudson TE, Nowlin BT, Bonventre they begin and how they end: Classic and new theories for the devel- JV, et al.: Fate tracing reveals the pericyte and not epithelial origin of fi fi – opment and deterioration of congenital anomalies of the kidney and myo broblasts in kidney brosis. Am J Pathol 176: 85 97, 2010 urinary tract, CAKUT. J Am Soc Nephrol 10: 2018–2028, 1999 27. Cui S, Schwartz L, Quaggin SE: Pod1 is required in stromal cells for – 8. Schedl A: Renal abnormalities and their developmental origin. Nat Rev glomerulogenesis. Dev Dyn 226: 512 522, 2003 Genet 8: 791–802, 2007 28. Little MH: Kidney Development, Disease, Repair, and Regeneration, 9. van der Ven AT, Vivante A, Hildebrandt F: Novel insights into the Amsterdam, Boston, Elsevier/AP, 2016 pathogenesis of monogenic congenital anomalies of the kidney and 29. Yu J, Carroll TJ, Rajagopal J, Kobayashi A, Ren Q, McMahon AP: A urinary tract. J Am Soc Nephrol 29: 36–50, 2018 Wnt7b-dependent pathway regulates the orientation of epithelial cell 10. Sanna-Cherchi S, Westland R, Ghiggeri GM, Gharavi AG: Genetic basis division and establishes the cortico-medullary axis of the mammalian of human congenital anomalies of the kidney and urinary tract. JClin kidney. Development 136: 161–171, 2009 Invest 128: 4–15, 2018 30. von Bubnoff A, Cho KW: Intracellular BMP signaling regulation in ver- 11. Weintraub H, Davis R, Tapscott S, Thayer M, Krause M, Benezra R, et al.: tebrates: Pathway or network? Dev Biol 239: 1–14, 2001 The myoD gene family: Nodal point during specification of the muscle 31. Hogan BL: Bone morphogenetic in development. Curr Opin cell lineage. Science 251: 761–766, 1991 Genet Dev 6: 432–438, 1996 12. Lee JE, Hollenberg SM, Snider L, Turner DL, Lipnick N, Weintraub H: 32. Wang RN, Green J, Wang Z, Deng Y, Qiao M, Peabody M, et al.: Bone Conversion of Xenopus ectoderm into neurons by NeuroD, a basic Morphogenetic Protein (BMP) signaling in development and human helix-loop-helix protein. Science 268: 836–844, 1995 diseases. Genes Dis 1: 87–105, 2014 13. Porcher C, Swat W, Rockwell K, Fujiwara Y, Alt FW, Orkin SH: The T cell 33. Dudley AT, Robertson EJ: Overlapping expression domains of bone leukemia oncoprotein SCL/tal-1 is essential for development of all morphogenetic protein family members potentially account for lim- hematopoietic lineages. Cell 86: 47–57, 1996 ited tissue defects in BMP7 deficient embryos. Dev Dyn 208: 349–362, 14. Srivastava D, Thomas T, Lin Q, Kirby ML, Brown D, Olson EN: Regula- 1997 tion of cardiac mesodermal and neural crest development by the bHLH 34. Miyazaki Y, Oshima K, Fogo A, Hogan BL, Ichikawa I: Bone morpho- transcription factor, dHAND. Nat Genet 16: 154–160, 1997 genetic protein 4 regulates the budding site and elongation of the 15. Quaggin SE, Vanden Heuvel GB, Igarashi P: Pod-1, a mesoderm-spe- mouse ureter. JClinInvest105: 863–873, 2000 cific basic-helix-loop-helix protein expressed in mesenchymal and 35. Davidson AJ: Mouse Kidney Development, Cambridge, MA, Stem- glomerular epithelial cells in the developing kidney. Mech Dev 71: 37– Book, 2008 48, 1998 36. Liu B, Chen Q, Tian D, Wu L, Dong H, Wang J, et al.: BMP4 reverses 16. Hidai H, Bardales R, Goodwin R, Quertermous T, Quertermous EE: multidrug resistance through modulation of BCL-2 and GDNF in glio- Cloning of capsulin, a basic helix-loop-helix factor expressed in pro- blastoma. Brain Res 1507: 115–124, 2013 genitor cells of the pericardium and the coronary arteries. Mech Dev 73: 37. Michos O, Gonçalves A, Lopez-Rios J, Tiecke E, Naillat F, Beier K, et al.: 33–43, 1998 Reduction of BMP4 activity by gremlin 1 enables ureteric bud out- 17. Lu J, Richardson JA, Olson EN: Capsulin: A novel bHLH transcription growth and GDNF/WNT11 feedback signalling during kidney factor expressed in epicardial progenitors and mesenchyme of visceral branching morphogenesis. Development 134: 2397–2405, 2007 organs. Mech Dev 73: 23–32, 1998 38. Majumdar A, Vainio S, Kispert A, McMahon J, McMahon AP: Wnt11 and 18. Robb L, Mifsud L, Hartley L, Biben C, Copeland NG, Gilbert DJ, et al.: Ret/Gdnf pathways cooperate in regulating ureteric branching during Epicardin: A novel basic helix-loop-helix transcription factor gene ex- metanephric kidney development. Development 130: 3175–3185, pressed in epicardium, branchial arch myoblasts, and mesenchyme of 2003 developing lung, gut, kidney, and gonads. Dev Dyn 213: 105–113, 1998 39. Cui S, Li C, Ema M, Weinstein J, Quaggin SE: Rapid isolation of glo- 19. Gooskens SL, Klasson TD, Gremmels H, Logister I, Pieters R, Perlman meruli coupled with gene expression profiling identifies downstream EJ, et al.: TCF21 hypermethylation regulates renal tumor cell clono- targets in Pod1 knockout mice. JAmSocNephrol16: 3247–3255, genic proliferation and migration. Mol Oncol 12: 166–179, 2018 2005 20. Richards KL, Zhang B, Sun M, Dong W, Churchill J, Bachinski LL, et al.: 40. Sazonova O, Zhao Y, Nürnberg S, Miller C, Pjanic M, Castano VG, et al.: of the candidate biomarker TCF21 is very frequent across a Characterization of TCF21 downstream target regions identifies a spectrum of early-stage nonsmall cell lung . 117: 606– transcriptional network linking multiple independent coronary artery 617, 2011 disease loci. PLoS Genet 11: e1005202, 2015

J Am Soc Nephrol 29: ccc–ccc, 2018 Transcription Factor 21 and Kidney Development 13 BASIC RESEARCH www.jasn.org

41. Rutledge EA, Benazet JD, McMahon AP: Cellular heterogeneity in the 50. Hatini V, Huh SO, Herzlinger D, Soares VC, Lai E: Essential role of ureteric progenitor niche and distinct profiles of branching morpho- stromal mesenchyme in kidney morphogenesis revealed by targeted genesis in organ development. Development 144: 3177–3188, 2017 disruption of Winged Helix transcription factor BF-2. Genes Dev 10: 42. Nicolaou N, Renkema KY, Bongers EM, Giles RH, Knoers NV: Genetic, 1467–1478, 1996 environmental, and epigenetic factors involved in CAKUT. Nat Rev 51. Levinson RS, Batourina E, Choi C, Vorontchikhina M, Kitajewski J, Nephrol 11: 720–731, 2015 Mendelsohn CL: Foxd1-dependent signals control cellularity in the 43. Vivante A, Hildebrandt F: Exploring the genetic basis of early-onset renal capsule, a structure required for normal renal development. De- chronic kidney disease. Nat Rev Nephrol 12: 133–146, 2016 velopment 132: 529–539, 2005 44. Piscione TD, Rosenblum ND: The molecular control of renal branching 52. Bard J: A new role for the stromal cells in kidney development. Bio- morphogenesis: Current knowledge and emerging insights. Differen- Essays 18: 705–707, 1996 tiation 70: 227–246, 2002 53. Basson MA, Akbulut S, Watson-Johnson J, Simon R, Carroll TJ, Shakya 45. Ichikawa I, Kuwayama F, Pope JC 4th, Stephens FD, Miyazaki Y: Para- R, et al.: Sprouty1 is a critical regulator of GDNF/RET-mediated kidney digm shift from classic anatomic theories to contemporary cell bi- induction. Dev Cell 8: 229–239, 2005 ological views of CAKUT. Kidney Int 61: 889–898, 2002 54. Grieshammer U, Le Ma, Plump AS, Wang F, Tessier-Lavigne M, Martin 46. Michos O, Cebrian C, Hyink D, Grieshammer U, Williams L, D’Agati V, GR: SLIT2-mediated ROBO2 signaling restricts kidney induction to a et al.: Kidney development in the absence of Gdnf and Spry1 requires single site. Dev Cell 6: 709–717, 2004 Fgf10. PLoS Genet 6: e1000809, 2010 55. Shakya R, Watanabe T, Costantini F: The role of GDNF/Ret signaling in 47. Ohuchi H, Hori Y, Yamasaki M, Harada H, Sekine K, Kato S, et al.: FGF10 ureteric bud cell fate and branching morphogenesis. Dev Cell 8: 65–74, acts as a major for FGF receptor 2 IIIb in mouse multi-organ 2005 development. Biochem Biophys Res Commun 277: 643–649, 2000 56. Short KM, Smyth IM: The contribution of branching morphogenesis to 48. Rosselot C, Spraggon L, Chia I, Batourina E, Riccio P, Lu B, et al.: Non- kidney development and disease. Nat Rev Nephrol 12: 754–767, 2016 cell-autonomous retinoid signaling is crucial for renal development. Development 137: 283–292, 2010 49. Jain S, Suarez AA, McGuire J, Liapis H: Expression profiles of congenital renal dysplasia reveal new insights into renal development and disease. This article contains supplemental material online at http://jasn.asnjournals. Pediatr Nephrol 22: 962–974, 2007 org/lookup/suppl/doi:10.1681/ASN.2017121278/-/DCSupplemental.

AFFILIATIONS

1Department of Clinical Cell Biology and Medicine, Graduate School of Medicine, Chiba University, Chiba, Japan; 2Division of Kidney Diseases, Ann and Robert H. Lurie Children’s Hospital of Chicago, Chicago, Illinois; 3Feinberg Cardiovascular and Renal Research Institute and 4Division of Nephrology/Hypertension, Northwestern University Feinberg School of Medicine, Chicago, Illinois; 5Department of Anatomy, Kyorin University School of Medicine, Tokyo, Japan; 6Lunenfeld-Tanenbaum Research Institute, Mount Sinai Hospital, Toronto, Ontario, Canada; and 7Division of Diabetes, Metabolism and Endocrinology, International University of Health and Welfare, Narita, Japan

14 Journal of the American Society of Nephrology J Am Soc Nephrol 29: ccc–ccc,2018