BASIC RESEARCH www.jasn.org

Von Hippel-Lindau Acts as a Metabolic Switch Controlling Nephron Progenitor Differentiation

Kasey Cargill,1,2,3 Shelby L. Hemker,1,2,3 Andrew Clugston,1,2,3,4 Anjana Murali,1,2 Elina Mukherjee,1,2 Jiao Liu,5,6 Daniel Bushnell,1,2 Andrew J. Bodnar,1,2 Zubaida Saifudeen,5,6 Jacqueline Ho,1,2,3 Carlton M. Bates,1,2,3 Dennis Kostka,1,3,4 Eric S. Goetzman,1,3,7 and Sunder Sims-Lucas1,2,3

1Rangos Research Center, University of Pittsburgh Medical Center Children’s Hospital of Pittsburgh, Pittsburgh, Pennsylvania; 2Division of Nephrology, Department of Pediatrics, University of Pittsburgh School of Medicine, Pittsburgh, Pennsylvania; 3Department of Pediatrics, University of Pittsburgh School of Medicine, Pittsburgh, Pennsylvania; 4Department of Developmental Biology and 7Division of Medical Genetics, Department of Pediatrics, University of Pittsburgh School of Medicine, Pittsburgh, Pennsylvania; 5Section of Pediatric Nephrology, Department of Pediatrics and 6The Hypertension and Renal Centers of Excellence, Tulane University Health Sciences Center, New Orleans, Louisiana

ABSTRACT Background Nephron progenitors, the cell population that give rise to the functional unit of the , are metabolically active and self-renew under glycolytic conditions. A switch from to mitochon- drial respiration drives these cells toward differentiation, but the mechanisms that control this switch are poorly defined. Studies have demonstrated that kidney formation is highly dependent on oxygen concen- tration, which is largely regulated by von Hippel-Lindau (VHL; a protein component of a ubiquitin ligase complex) and hypoxia-inducible factors (a family of transcription factors activated by hypoxia). Methods To explore VHL as a regulator defining nephron progenitor self-renewal versus differentiation, we bred Six2-TGCtg mice with VHLlox/lox mice to generate mice with a conditional deletion of VHL from Six2+ nephron progenitors. We used histologic, immunofluorescence, RNA sequencing, and metabolic as- says to characterize kidneys from these mice and controls during development and up to postnatal day 21. Results By embryonic day 15.5, kidneys of nephron progenitor cell–specific VHL knockout mice begin to exhibit reduced maturation of nephron progenitors. Compared with controls, VHL knockout kidneys are smaller and developmentally delayed by postnatal day 1, and have about half the number of glomeruli at postnatal day 21. VHL knockout nephron progenitors also exhibit persistent Six2 and Wt1 expression, as well as decreased mitochondrial respiration and prolonged reliance on glycolysis. Conclusions Our findings identify a novel role for VHL in mediating nephron progenitor differentiation through metabolic regulation, and suggest that VHL is required for normal kidney development.

JASN 30: ccc–ccc, 2019. doi: https://doi.org/10.1681/ASN.2018111170

Nephron development initially involves reciprocal signaling between the ureteric epithelium (UE) and the cap mesenchyme (CM), a region comprising re- Received November 28, 2018. Accepted April 1, 2019. 1 nal cells known as nephron progenitors. These Published online ahead of print. Publication date available at nephron progenitors are under strict regulation ei- www.jasn.org. 2 ther to maintain a progenitor state or differentiate. Correspondence: Dr. Sunder Sims-Lucas, Division of Pediatric The transcription factor Six2 plays a major role in Nephrology, Children’s Hospital of Pittsburgh, 4401 Penn Avenue, the maintenance of the proliferating progenitor Pittsburgh, PA 15224. Email: [email protected] – population.3 5 In self-renewing nephron progenitors, Copyright © 2019 by the American Society of Nephrology

JASN 30: ccc–ccc,2019 ISSN : 1046-6673/3007-ccc 1 BASIC RESEARCH www.jasn.org both Six2 and the cofactor Cited1 are initially expressed; Significance Statement however, Cited1 expression becomes downregulated at the beginning of differentiation.2,3,6,7 As nephron progenitors Nephron progenitors, the self-renewing cells that give rise to mature, Six2 also becomes downregulated and the cells be- nephrons, are particularly metabolically active, relying primarily on gin to express differentiation markers indicating commit- glycolysis for energy generation early in development. As devel- opment progresses, nephron progenitors switch from glycolysis to 4,8 ment to a mesenchyme-to-epithelial transition (MET). mitochondrial respiration for energy—mediated by an unknown This transition, marked by Wnt4 expression, occurs as CM cells mechanism—and undergo differentiation. To study the potential make mature cell-cell contacts and ultimately form functional role in this metabolic shift of von Hippel-Lindau (VHL), a protein nephrons.9 The maturing nephron progenitors give rise to component of a ubiquitin ligase complex, the authors generated – fi VHL the nephron: a complex structure comprising a glomerulus nephron progenitor cell speci c knockout mice. In addition to identifying VHL as a critical regulator of nephron progenitors’ and epithelialized tubules that are connected to the collecting metabolic switching, the authors’ findings demonstrate that this duct system.10 switch also plays a large role in the differentiation process, and Kidney formation is highly dependent on oxygen concen- suggest that VHL is required for normal kidney development. tration11 and previous studies show that low levels of oxygen (hypoxia)canresultinrenalmalformationsincludingneph- Herein, we report an interrogation into the role of VHL/HIF ron deficit.12,13 The kidney initially develops in a relatively signaling in mediating nephron progenitor differentiation. hypoxic environment and as the vasculature perfuses, oxygen Global VHL knockout mice are embryonic lethal,16 suggest- levels increase.14 This tightly regulated change in oxygenation ing that VHL plays a critical role in embryonic development. coordinates with the inductive signals from the UE to drive To circumvent embryonic lethality, we generated nephron proliferation and subsequent differentiation of the nephron 2 2 progenitor-specific VHL knockout mice (VHLNP / )tointer- progenitors. Hypoxia-inducible factors (HIFs) play a large role rogate the role of VHL in nephron development. Our investiga- in mediating cellular responses to oxygen fluctuations. Al- tion reveals that loss of VHL results in metabolic dysregulation, though there are three HIF isoforms (HIF-1a,HIF-2a,and which inhibits nephron progenitor differentiation. HIF-3a), HIF-1a is considered the master regulator of devel- opmental responses to hypoxia.15 HIF-1a is known to stimu- late angiogenesis and maturation of the vasculature, which METHODS drives blood flow into hypoxic areas, thereby increasing oxygen concentration.14,16 Increases in oxygen facilitate HIF-1a’sdeg- Mouse Model radation via recruitment of von Hippel-Lindau (VHL),14,16 a We bred transgenic Six2-TGCtg mice [JAX009606; Tg(Six2-EGFP/ protein component of a ubiquitin ligase complex. If prenatal cre)1Amc/J]26 with VHL-floxed mice [JAX12933; B6.129S4(C)- exposure to pathologic hypoxia occurs in the nephron pro- Vhltm1Jae/J]27 to generate a deletion of VHL in Six2cre+ cells. genitors, VHL is not recruited to degrade HIF-1a,allowing The University of Pittsburgh Institutional Animal Care and Use for its sustained expression and transcriptional activation Committee approved all experiments (approval no. 16088935). of containing hypoxia-response elements (HREs).11–13 The genotyping primers for the Six2-TGCtg and Vhl alleles are Many confirmed HREs are found in genes that encode gly- listed in Supplemental Table 1. colytic .11,12,14 High rates of glycolysis allow progen- itor cells to rapidly undergo expansion through proliferation Western Blotting and generation.15–17 However, as differentiating Kidneys were lysed in radioimmunoprecipitation assay buffer cells mature, they utilize mitochondrial respiration to maintain (ThermoFisher Scientific) and protein was quantified using a energy demands of the cell.17–19 HIF-1a is largely responsible Bradford assay (Bio-Rad, Hercules, CA). Samples were electro- for the metabolic profile of progenitor cells and functions in a phoresed on Bolt 4%–12% Bis-Tris Plus Gels and transferred to dual role by promoting glycolysis as well as suppressing mi- 0.2 mm nitrocellulose membranes (ThermoFisher Scientific). tochondrial respiration.18,20,21 It has also been shown that Antibodies used include anti-VHL (1:250; Santa Cruz Biotech- nephron progenitors exhibit a gradual, decreased reliance nology), anti–HIF-1a (1:250; Novus), and anti-a/b-tubulin on glycolysis between embryonic day (E) 13.5 and 19.5.22 (1:1000; Cell Signaling Technology). Blots were imaged using a Although early-stage nephron progenitors rely heavily on ProteinSimple FluoroChem System (ProteinSimple) and quan- glycolysis,22 terminally differentiated nephrons have a unique tified using ImageJ (ImageJ/National Institutes of Health metabolic profile depending on the particular nephron seg- Image). ment.23 The proximal tubule segment, which is responsible for a majority of the reabsorption performed by the kidney, Histology and Immunostaining utilizes fatty acids, oxidative , and gluco- Renal histology was assessed at E13.5, E15.5, postnatal day 1 neogenesis to support these critical functions.23–25 For these (P1), and P21. Tissue was fixed in 4% paraformaldehyde and reasons, we hypothesize that the switch from glycolysis to mi- processed in paraffin.28 Samples were sectioned at 4 mm tochondrial respiration is regulated via VHL-mediated HIF-1a and subjected to hematoxylin and eosin or immunostaining. degradation in nephron progenitors during development. The following primary antibodies or lectins were used at a 1:100

2 JASN JASN 30: ccc–ccc,2019 www.jasn.org BASIC RESEARCH dilution unless indicated: anti-Six2 (Proteintech), anti-Ncam software package (version 37.41).34 Reads were aligned to (Sigma-Aldrich), anti-Glut1 (Abcam), anti–phospho-histone transcripts assembled from the mm10 genome (GRCm38, H3 (Cell Signaling), terminal deoxynucleotidyl – GENCODE M17)35 using the Spliced Transcripts Alignments mediated digoxigenin-deoxyuridine nick-end labeling (EMD to a Reference software package (version 2.5.3a).36 Data anal- Millipore), anti-Lef1 (Cell Signaling), anti-Sall1 (Abcam), anti- ysis was performed using R programming (version 3.4.3).37 Wt1 (Santa Cruz Biotechnology), anti-Jag1 (1:20; Santa Cruz Reads aligned to known transcripts were counted using the Biotechnology), anti-Pck1 (Proteintech), dolichos biflorus Bioconductor GenomicAlignments package (version 1.10.1).38 agglutinin (Vector Laboratories), and lotus tetragonolobus Differential expression between kidney samples was calcu- lectin (Vector Laboratories). Samples were imaged using a lated using the DESeq2 R package (version 1.18.1).39 Func- Leica microscope and LAS X software (Leica Microsystems). tional annotation analysis was performed using the DAVID database (release 6.8; https://david.ncifcrf.gov/).40,41 Data Three-Dimensional Reconstruction were deposited into Expression Omnibus (accession E13.5 embryos were collected for three-dimensional recon- number: GSE122900). structions of kidneys as previously described.29 Three- dimensional reconstruction was performed as described,29 Real-Time Quantitative PCR using a Zeiss Model Axio Imager M1 (Zeiss) microscope with RNA was extracted using a miRNeasy Mini Kit (Qiagen) from Stereo Investigator Stereology image tracing software (MBF E17.5 kidneys. The Superscript First Strand cDNA kit (Invi- Bioscience). trogen) was used for cDNA synthesis and samples were ana- lyzed using a C1000 Thermal Cycler (Bio-Rad) to determine Physical Dissector/Fractionator Combination Method mRNA expression levels (normalized to Rn18s). The following for Glomeruli Counting genes were analyzed: Pgk1, Gapdh, Tpi1, Bnip3, Ddit4, Pdk1, P21 kidneys were fixed, processed, and serially sectioned at Eno1, Slc16a3, Cited1, Six2, Lef1, Sall1, Osr1, Eya1, Jag1, Sox9, 4 mm. Samples were subjected to hematoxylin and eosin stain- Foxd1,andWnt4 (Supplemental Table 1). ing.30 The nth (reference section) and nth+2 (lookup section) sections were counted for glomeruli as described,30,31 using a Magnetic-Activated Cell Sorting Isolation and Zeiss Model Axio Imager M1 (Zeiss) microscope with Stereo Seahorse Extracellular Flux Investigator Stereology image tracing software (MBF Nephron progenitors were isolated from P1 kidneys using Bioscience). magnetic-activated cell sorting (MACS; Miltenyi Biotec) and cultured as previously described.3,22 Nephron progenitors Quantification of P1 Nephron Number were subjected to a mitochondrial stress test by flux analysis P1 kidneys were fixed, processed, and serially sectioned at on a XF24 Extracellular Flux Analyzer (Agilent Seahorse 4 mm. Samples were immunostained with anti-Wt1 (1:100; Technologies).22 XF sensor cartridges and assay medium Cell Signaling Technologies), and visualized using a Vectastain were prepared as previously described.22 Metabolic profiling ABC Kit (Vector Laboratories).32 Every 20 sections, Wt1+ was performed by measuring basal respiration for 25 minutes structures were identified from consecutive sections using a followed by the sequential injection of the substrates oligomy- Zeiss Model Axio Imager M1 (Zeiss) microscope with Stereo cin (1 mg/ml), cyanide p-trifluoromethoxy-phenylhydrazone Investigator Stereology image tracing software (MBF Bioscience). (0.3 mM), and rotenone (0.1 mM) with antimycin A (1 mM).22 Glomerular number was quantified using the equation by Data were normalized to cell number. Cullen-McEwen et al.30 Pyruvate Oxidation Serum Analysis P1 kidneys were homogenized in DMEM without supplemen- P21 mice were anesthetized and subject to cardiac puncture tation. Homogenates were incubated at 37°C with uniformly for blood collection. Serum was isolated by centrifugation labeled 14C-pyruvate in glass tubes sealed with rubber stoppers and samples were analyzed for urea (Kansas State fitted with hanging baskets. The baskets held filter paper with University). potassium hydroxide (1M). After 30 minutes, perchloric acid (0.5M) was introduced and the reactions were incubated at 14 42,43 14 RNA-Sequencing 37°C for 1 hour to collect C-CO2. C-pyruvate incor- E17.5 kidneys were dissociated and sorted by FACS for GFP+ poration was measured by scintillation counting and normal- cells (Flow Cytometry Core). RNA was extracted using a ized to cellular protein content. miRNeasy Mini Kit (Qiagen). The Health Sciences Sequencing Core at University of Pittsburgh Medical Center Children’s Mitochondrial Density Hospital of Pittsburgh performed library construction and Kidneyswere digestedusing 0.3% collagenase to create a single- RNA-sequencing (single-end reads, 75 bp). Bioinformatics cell suspension. Then, 13106 cells were stained using 200 nM quality control was performed using FastQC (version 0.11.5),33 MitoTracker Red FM (ThermoFisher Scientific). Cells were an- and adapters were trimmed using BBDuk from the BBMap alyzed on a BD LSRFORTESSA cell analyzer (BD Biosciences).

JASN 30: ccc–ccc,2019 Von Hippel-Lindau Acts as a Metabolic Switch 3 BASIC RESEARCH www.jasn.org

Mean fluorescence intensity was determined using FlowJo differences in kidney volume, UE volume, or glomerular software (version 10.1). volume (Figure 2, A–H). Next, we evaluated the expression of nephron progenitor Kidney Explants and renal vesicle markers through immunofluorescence stain- 2 2 2 E13.5 kidneys were cultured on 0.4 mm polyethylene tere- ing of E15.5 and P1 VHLNP+/ and VHLNP / kidneys as E15.5 phthalate membrane inserts as previously described.32 One is when we first observe a phenotype and P1 is largely an older kidney from each embryo was cultured in improved minimal progenitor population. Staining against the nephron progen- essential media (containing 50 mg/ml transferrin) alone and itor marker Six2 with Ncam to mark the nascent nephrons at the other was cultured in media containing 10 mMYN1 E15.5 (Figure 3, A and B) and P1 (Figure 3, C and D) revealed 2 2 inhibitor.22,32 After 72 hours kidneys were fixed in 4% para- that VHLNP / kidneys have persistent Six2 expression in formaldehyde for 1 hour. Kidneys were immunostained Ncam+ structures. Further, Ncam expression was also de- with antibodies against Six2 and Jag1, following previously creased at both E15.5 and P1, suggesting there may be alter- described methods.32 ations in cell adhesion between nephron progenitors (Figure 3, A–D). For these reasons, we quantified the number of Six2+ Statistical Analyses cells per cap at P1 and found an increase (P=0.006; Figure 3I). At least three biologic replicates were used for each experiment. Next, we performed immunostaining against Jag1 (differenti- Statistical significance was determined by a two-tailed t test ating structures) and quantified the number of Jag1+ structures (a=0.05), using GraphPad Prism 7. Where appropriate, data per P1 kidney. We saw a reduction in the number of Jag1+ are presented as mean6SEM. structures (P=0.008; Figure 3, E–H and J), which further sug- 2 2 gests that VHLNP / kidneys exhibit decreased differentiation. Additionally, we performed immunostaining for Wt1, Sall1, and RESULTS Lef1, which support these findings (Supplemental Figure 2). In 2 2 particular, persistent Wt1 expression in VHLNP / nephron pro- VHL Expression Is Required for Differentiation of genitors was consistent with maintenance of a progenitor-like Nephron Progenitors state (Supplemental Figure 2, A and B). To evaluate differentia- To determine the functional significance of VHL in nephron tion capacity, we performed unbiased, stereological counting of progenitors, we conditionally deleted VHL from Six2+ neph- Wt1+ glomeruli at P1 and found a 30% decrease (P=0.01; Figure ron progenitors by crossing transgenic Six2-TGCtg mice26 to 4A). Although there is a significant decrease in the number of VHLlox/lox mice27 (Figure 1A). This produced VHL wild-type nephrons, the kidney-to-body weight ratio at P1 was un- 2 2 2 [Six2-TGC+/+; VHLlox/lox (VHLNP+/+)], heterozygous knock- changed between VHLNP+/ and VHLNP / mice (Figure 4B). 2 out [Six2-TGC tg/+; VHL+/lox (VHLNP+/ )], and homozygous analysis of whole E17.5 kidneys showed no 2 2 knockout [Six2-TGC tg/+; VHLlox/lox (VHLNP / )] genotypes. differences in several nephron progenitor markers, including 2 Six2-TGCtg/+ embryos enabled GFP visualization in VHLNP+/ Cited1, Sall1, Eya1, Six2,andOsr1 (Figure 4C). Gene expres- 2 2 and VHLNP / animals. Because of a slight hypoplastic phe- sion analysis of the differentiation markers Lef1, Wnt4,and notype associated with the Six2GFPcre transgene,44 we used Jag1 revealed significant decreases in expression reflecting 2 VHLNP+/ littermate controls for all experiments. VHL pro- an inability to properly undergo differentiation (Figure 4C). tein expression (Figure 1B) and HIF-1a protein stabilization There were no changes in Foxd1 (stromal progenitors) or Sox9 2 2 2 (Figure 1C) in VHLNP+/ and VHLNP / kidneys confirmed (ureteric tip identity) gene expression (Figure 4C). Together, sufficient knockdown and stabilization, respectively. Moreover, this suggests that VHL does not induce profound alterations subsequent RNA-sequencing analysis (see below) validated during the self-renewal phase and likely plays a larger role in VHL knockout specifically in the nephron progenitors. the regulation of nephron progenitors that are poised to exit the 2 2 Histologic examination of VHLNP / kidneys did not progenitor niche. reveal any significant alterations in development at E13.5 To explore alterations in proliferation and cell death as a 2 2 (Figure 1, D and E). However, by E15.5 the VHLNP / kidneys possible causes of decreased differentiation, we performed im- begin to exhibit reduced maturation of the nephron progeni- munofluorescence staining with anti–phospho-histone H3 2 2 tors (Figure 1, F and G). By P1 VHLNP / kidneys are smaller and terminal deoxynucleotidyl transferase–mediated digoxi- and remain developmentally delayed (Figure 1, H and I, Sup- genin-deoxyuridine nick-end labeling, respectively. There plemental Figure 1) with fewer nephron progenitor-derived were no differences in proliferation or apoptosis (Supplemen- structures (see below). To determine the onset of the devel- tal Figure 3, A–P). We also performed cell cycle analysis on opmental phenotype, we performed three-dimensional re- P1 nephron progenitors using propidium iodide to investigate constructions of E13.5 kidneys to evaluate kidney size and potential irregularities in the cell cycle or senescence. Cell- maturation, which supported the histologic examination. cycle analysis showed no differences in G1/G0, S, or G2/M The number of glomerular structures, renal vesicles, and phases (Supplemental Figure 3, Q and R). Collectively, these immature glomeruli were not significantly altered be- data indicate that reductions in nephron number do not likely 2 2 2 tween VHLNP+/ and VHLNP / kidneys nor were there any stem from decreased proliferation, increased cell death, cell

4 JASN JASN 30: ccc–ccc,2019 www.jasn.org BASIC RESEARCH

A VHL Six2-cre EGFP loxP Exon 1 loxP Exon 2

VHLNP-/- loxP Exon 2

B C NP+/- NP-/- NP+/- NP-/-

VHL VHL VHL VHL

VHL HIF-1α

α/β tubulin α/β tubulin

*** * 0.3 0.6

0.2 0.4

0.1 0.2 Relative Quantification Relative Quantification

0.0 0.0 VHLNP+/- VHLNP-/- VHLNP+/- VHLNP-/-

E13.5 E15.5 P1 D F H NP +/- VHL

E G I NP -/- VHL

Figure 1. Conditional deletion of VHL from Six2+ nephron progenitors leads to HIF-1a stabilization and reduced nephron structures. (A) Experimental breeding strategy to obtain successful VHL deletion in nephron progenitors using Six2-TGCtg and VHLlox/lox mice. (B) 2 2 2 Whole kidney homogenates from P1 VHLNP+/ and VHLNP / mice subjected to Western blotting with antibodies against VHL and 2 2 2 a/b-tubulin to confirm VHL deletion; ***P,0.001; VHLNP+/ n=3; VHLNP / n=5. Error bars indicate SEM. (C) Whole kidney homogenates 2 2 2 from P1 VHLNP+/ and VHLNP / mice subjected to Western blotting with antibodies against HIF-1a and a/b-tubulin to confirm HIF-1a

JASN 30: ccc–ccc,2019 Von Hippel-Lindau Acts as a Metabolic Switch 5 BASIC RESEARCH www.jasn.org

A Glomerular Structures at E13.5 B Tissue Volume at E13.5 80 8×107 VHLNP+/- 60 6×107 VHLNP-/- ) 3

40 4×107 Volume (um Structure Counts 20 2×107

0 0 All Glomerular Renal Immature Kidney Ureteric All Glomerular Structures Vesicles Glomeruli Epithelium Structures

Glomeruli UB + Glomeruli Semi-reconstruction CEG NP +/- VHL z z z

x y x y x y

DF H NP -/- VHL z z z y x y x y x

2 2 2 Figure 2. VHLNP / mice exhibit altered kidney development after E13.5. (A) Three-dimensional reconstruction of E13.5 VHLNP+/ and 2 2 VHLNP / kidneys showing glomerular structure formation. No significant differences in glomerular structure, renal vesicle formation, or 2 2 2 immature glomeruli number. (B) Three-dimensional reconstruction of E13.5 VHLNP+/ and VHLNP / kidneys showing tissue volume. No significant differences in kidney volume, UE volume, or glomerular volume. (C–H) Three-dimensional images depicting all glo- merular structures (green), ureteric branching patterns (pink), and a reconstruction of the kidneys. n=4. Error bars indicate SEM. UB, ureteric bud. cycle dysregulation, or cellular senescence before cessation of Supplemental Figure 4, A–C). Variance and expression of dis- 2 2 nephrogenesis in VHLNP / kidneys. tribution were estimated using DESeq2 and minus-average plotting (Supplemental Figure 4, D and E). We observed a NP2/2 VHL Nephron Progenitors Display Dysregulation downregulation of VHL,confirming efficient deletion of of Metabolic Gene Expression this loxP-flanked fragment by Cre (Supplemental To better understand the mechanisms underlying how VHL Figure 5A). Controlling the false discovery rate at 5% and focus- contributes to development, we performed RNA-sequencing ing on genes with at least two-fold up- or downregulation, we on E17.5 GFP+ nephron progenitors using FACS. Hierarchical found that 245 known genes were significantly altered: 222 of distance clustering and principal component analysis showed these genes were upregulated and only 23 were downregulated 2 2 2 2 2 clear grouping of VHLNP+/ and VHLNP / samples (n=4; in VHLNP / nephron progenitors (Figure 5A). The top stabilization; *P,0.05; n=4. Error bars indicate SEM. (D and E) Hematoxylin and eosin staining of kidneys sectioned at E13.5 appear developmentally normal. (F–I) Hematoxylin and eosin staining of kidneys sectioned at E15.5 and P1 exhibit hypoplasia and reduced developing glomeruli (arrows).

6 JASN JASN 30: ccc–ccc,2019 www.jasn.org BASIC RESEARCH

E15.5 P1

VHLNP+/- VHLNP-/- VHLNP+/- VHLNP-/- ABCD Six2 / Ncam

E FGH Six2 / Jagged1

I J 24 ** 50 **

22 40 20 30 18 20 16 Six2+ Cells/Cap 10 14 Jagged1+ Structures

12 0 VHLNP+/- VHLNP-/- VHLNP+/- VHLNP-/-

2 2 Figure 3. VHLNP / kidneys exhibit Six2 misexpression, decreased cell adhesion, and fewer Jagged1+ structures. (A–D) E15.5 and P1 immunofluorescence staining with antibodies against Six2 (red) and Ncam (green) show prolonged expression of Six2 in Ncam+ 2 2 structures (arrowheads). Reduced Ncam staining was also shown in VHLNP / kidneys (*). (E–H) Six2 (red) and Jag1 (green) staining reveals expansion of Six2+ cells (+) and fewer Jag1+ structures (arrows). (I) The number of Six2+ cells per cap was significantly increased 2 2 in P1 VHLNP / kidneys. **P,0.01; n=3. Error bars indicate SEM. (J) The number of Jag1+ structures is significantly decreased in P1 2 2 VHLNP / kidneys. **P,0.01; n=3. Error bars indicate SEM. most-dysregulated genes included Bnip3l, Bnip3, Pgm2, can be found in Supplemental Table 2. Additionally, we vali- Fam162a, Pfkp, Slc2a3, Kbtbd11, Stc2, Adamts13,andMgarp dated several of the RNA-sequencing data results by real-time 2 2 2 (Supplemental Figure 5B). Functional enrichment analysis quantitative PCR using cDNA from VHLNP+/ and VHLNP / using DAVID (http://david.ncifcrf.gov) generated the top whole kidneys (Figure 5D). most affected term pathways, which included NP2/2 glycolysis (16 genes; 45.7% of KEGG glycolysis/gluconeogen- VHL Nephron Progenitors Exhibit Decreased 2 esis genes; P=1.13310 21), cellular response to hypoxia Mitochondrial Respiration 2 (13 genes; 12.6% of KEGG HIF-1 genes; P=6.65310 10), RNA-sequencing data analysis implicated metabolic dysregu- and HIF-1a signaling pathway (12 genes; 11.4% of KEGG lation as a potential driver of defective nephron progenitor 2 2 2 HIF-1a genes; P=3.78310 8), among others (Figure 5B). differentiation in VHLNP / kidneys. To compare overall mi- 2 2 2 HIFs, the main targets of VHL, are known to transcriptionally tochondrial function, P1 VHLNP+/ and VHLNP / nephron activate genes containing HREs, several of which were found to progenitors were MACS-isolated and subjected to a mito- be altered, including those associated with glycolysis (Pgk1, chondrial stress test using a Seahorse XF24 extracellular flux Gapdh,andTpi1), hypoxia response (Bnip3 and Ddit4), analyzer to measure oxygen consumption.3,22 Oxygen con- and HIF-1a signaling (Pdk1 and Eno1) (Figure 5C, Supple- sumption is an indicator of mitochondrial respiration in the 2 2 mental Figure 5, C and D) in VHLNP / nephron progenitors. presence of a series of metabolic inhibitors and uncoupling A complete list of dysregulated genes in the HIF-1a pathway agents.42,45 Overall oxygen consumption, both before and

JASN 30: ccc–ccc,2019 Von Hippel-Lindau Acts as a Metabolic Switch 7 BASIC RESEARCH www.jasn.org

A B 3000 * 1.0

0.8 2000 0.6

Nglom 0.4 1000 Mass (g) 0.2

0 0.0 VHLNP+/- VHLNP-/- VHLNP+/- VHLNP-/- C 200 VHLNP+/- VHLNP-/-

150 ***

100

50 % Normalized to Control

0 Cited 1 Sall1 Eya1 Six2 Osr1 Lef1 Wnt4 Jag1 Foxd1 Sox 9

2 2 2 2 Figure 4. VHLNP / kidneys display decreased differentiation. (A) VHLNP / kidneys have significantly fewer glomeruli (30% reduction) compared with littermate controls. *P,0.05; n=4. Error bars indicate SEM. (B) No significant differences in the kidney-to-body weight ratio at P1; n=3. (C) No significant differences in nephron progenitor gene expression (Cited1, Sall1, Eya1, Six2, Osr1). Renal vesicle genes (Lef1, Wnt4, Jag1) were all significantly increased. *P,0.05; n=4. Error bars indicate SEM. after substrate addition, resulted in decreased oxygen consump- both before and after substrate administration (Figure 7A). Basal tion rate (OCR) (Figure 6A). Basal oxygen consumption was extracellular acidification rate was increased (Figure 7B). 2 2 significantly decreased in VHLNP / nephron progenitors com- Moreover, the glycolytic response to oligomycin (mitochondrial 2 2 2 pared with VHLNP+/ littermate controls (Figure 6B). Moreover, respiration inhibitor) was significantly higher in VHLNP / 2 mitochondrial ATP-linked respiration was significantly decreased nephron progenitors compared with VHLNP+/ nephron pro- (Figure 6C). Nonmitochondrial respiration was also calculated genitors demonstrating increased glycolytic capacity (Figure 7C). by averaging the oxygen consumption after addition of rotenone/ We evaluated glycolytic dependence by culturing E13.5 kidney 2 2 antimycin A and was increased in VHLNP / nephron progenitors explants in media alone or media containing 10 mM Y1N (gly- (P=0.02). This indicates decreased reliance on the mitochondria colysis inhibitor) for 72 hours. Explants were then subjected to 2 2 for energy production. P1 VHLNP / kidneys also exhibited de- immunostaining against Six2 and Jag1 to mark the nephron creased mitochondrial pyruvate oxidation to CO2 (Figure 6D), progenitors and developing structures, respectively. Visual further confirming decreased mitochondrial . To identification of the number of differentiating structures re- 2 2 better understand the decrease in mitochondrial respiration, vealed (1) VHLNP / kidneys cultured in YN1 were capable of 2 2 we analyzed mitochondrial density by flow cytometry using differentiating whereas VHLNP / kidneys cultured in media 2 MitoTracker Red. There were no significant differences at P1 alone underwent minimal differentiation, and (2) VHLNP+/ 2 2 in mitochondrial density in the GFP+ cell population (Figure 6E). and VHLNP / kidneys cultured in YN1 differentiated similarly 2 2 This data indicates that P1 VHLNP / nephron progenitors ex- (Figure 7, D–G). This suggests that the differentiation pheno- 2 2 hibit decreased dependency on mitochondrial ATP for energy type arises from an inability of VHLNP / nephron progenitors production, which is not due to mitochondrial insufficiency. to metabolically switch from glycolysis to mitochondrial respi- ration during nephrogenesis, further validating the importance NP2/2 VHL Nephron Progenitors Exhibit of VHL during kidney development. Increased Glycolysis In addition to oxygen consumption, the Seahorse XF24 extra- VHL Deletion Results in Abnormalities of Juvenile cellular flux analyzer monitors the extracellular acidification Mouse Kidneys 2 2 2 2 rate, as an indicator of anaerobic glycolysis. VHLNP / MACS- VHLNP / mice experience premature death around P30; isolated nephron progenitors demonstrated increased glycolysis therefore we characterized the renal phenotype at P21.

8 JASN JASN 30: ccc–ccc,2019 www.jasn.org BASIC RESEARCH

A Expression Change (LFC > 2, padj < 0.05) B Functional Annotation Enrichment VHLNP+/- VHLNP-/-

2 Glycolysis 16 genes Glycolytic process 16 genes 1 Glycolysis/ 17 genes

0 Carbon metabolism 16 genes

Biosynthesis of amino acids 14 genes -1 Biosynthesis of antibiotics 19 genes -2 Cellular response to hypoxia 13 genes

HIF-1α signaling pathway 12 genes

MHC class II protein complex 6 genes

Fructose & mannose metabolism 8 genes

051015 -Log10(FDR) Glycolysis Gene Expression C D 500 VHLNP+/- 3,000 * * VHLNP-/- 400

2,000 * 300

TPM * * * 200 * 1,000 *

100 % Normalized to Control

0 0

Pkm Tpi1 Pfkl Hk2 Pfkp Gck Pgk1 Gpi1 AldocEno1 Tpi1 Eno1b Pgam1Gapdh Pgk1 Bnip3 Ddit4 Pdk1 Eno1 Aldoart1 Aldoart2 Gapdh Slc16a3 VHLNP+/- VHLNP-/-

2 2 2 Figure 5. RNA-sequencing reveals genetic dysregulation in isolated VHLNP+/ and VHLNP / nephron progenitors. (A) Heat map 2 2 2 showing differential gene expression in VHLNP+/ and VHLNP / nephron progenitors. Cutoff for significance was set to LFC .2and P,0.05. (B) DAVID functional annotation enrichment analysis showing the pathways associated with the gene expression data. (C) Annotation-associated gene expression showing glycolysis gene dysregulation using DAVID analysis (Gene Ontology terms). (D) Real- time quantitative PCR validation of significantly dysregulated genes in the glycolysis, cellular response to hypoxia, and HIF-1a signaling pathways. *P,0.05; RNA-sequencing n=4; real-time quantitative PCR validation n=3. Error bars indicate SEM. FDR, false discovery rate; LFC, log fold change; TPM, transcripts per million.

To assess renal structure and function of the juvenile (P21) hypothesis, we performed immunostaining against Glut1 mice, we performed gross and histologic examinations, which (glycolysis) and Pck1 (gluconeogenesis), which did not reveal 2 revealed smaller kidneys with fewer glomerular struc- differences in protein expression between the VHLNP+/ and 2 2 tures (Figure 8, A–D, Supplemental Figure 6, A and B). The VHLNP / kidneys (Figure 8, E–H), suggesting that the de- 2 2 VHLNP / kidneys appeared to have decreased glomeruli and veloped tubules are functional. slight dilation of the proximal tubules (Figure 8, A–D), which To determine whether the defect we observed was specific is likely owing to a compensatory mechanism to overcome the to nephron progenitors or due to a defect in their differenti- reduction in nephron number.46 In fact, we found that the ation into renal vesicles, we generated a mouse model with a 2 2 VHLNP / kidneys have a 50% reduction in nephron number conditional deletion of VHL specifically in the renal vesicles 2 2 (P,0.001; Figure 8I). Although VHLNP / kidneys were utilizing a Wnt4creGFP47 using the same breeding strategy as 2 smaller than VHLNP+/ kidneys, the kidney-to-body weight described for our VHLNP model above. Interestingly, we did ratio was not different (Figure 6). Additionally, BUN concentra- not see significant alterations in body weight, kidney weight, 2 2 tion was significantly increased in VHLNP / mice (Figure 8J), or kidney histology (Supplemental Figure 7 and data not 2 whichwehypothesizedwasduetoglomerularinsufficiency shown) between heterozygous (VHLRV+/ ) and homozygous 2 2 and not a result of tubular malfunction. To evaluate this (VHLRV / ) mice. This suggests that VHL is dispensable in

JASN 30: ccc–ccc,2019 Von Hippel-Lindau Acts as a Metabolic Switch 9 BASIC RESEARCH www.jasn.org

A 80 O F R+A B Basal Oxygen Consumption C ATP-linked Respiration * 60 60 * 40 VHLNP+/- VHLNP-/- 30 40 40 20 20 20 OCR (pMoles/min) 10 OCR (pMoles/min) OCR (pMoles/min) 0 0 0 0 15 30 45 60 75 90 105 120 135 VHLNP+/- VHLNP-/- VHLNP+/- VHLNP-/- Time [min]

DEPyruvate Oxidation Mitochondrial density 30 * 8

6 20 4 10 Relative MFI 2

pMoles/min/mg protein 0 0 VHLNP+/- VHLNP-/- VHLNP+/- VHLNP-/-

Figure 6. VHL deletion suppresses cellular oxygen consumption. (A) P1 MACS-isolated nephron progenitors were subjected to oxygen consumption (OCR) measurements in a Seahorse XF24 extracellular flux analyzer. Measurements of OCR were taken before and after sequential administration of the metabolic inhibitors oligomycin (ATP synthase inhibitor), cyanide p-trifluoromethoxy-phenylhydrazone (mitochondrial oxidative phosphorylation uncoupler), and rotenone/antimycin A (respiratory chain inhibitors). (B) The average oxygen 2 2 consumption over the first 30 minutes produced the basal oxygen consumption, which was significantly decreased in VHLNP / nephron progenitors. (C) ATP-linked respiration was calculated by subtracting basal respiration by the average OCR after oligomycin NP2/2 14 addition. ATP-linked respiration was significantly decreased in VHL nephron progenitors. (D) Pyruvate oxidation to CO2 was 2 2 significantly decreased in VHLNP / whole tissue homogenates. *P,0.05; n=3. Error bars indicate SEM. (E) Mitochondrial density was not significantly altered P1 nephron progenitors. Error bars indicate SEM.

2 2 2 the differentiated renal vesicle, and thus also dispensable for VHLNP+/ and VHLNP / nephron progenitors. We conclude 2 2 nephron function. Taken together, we show that VHLNP / that the mitochondria are healthy and retain the ability to mice have malformations resulting in fewer glomeruli and re- function normally; however, they lack substrate (i.e.,pyru- duced renal function, which is not attributed to a postnatal vate) because of the HIF-mediated upregulation of PDK, defect and therefore results from the abnormal development which effectively blocks pyruvate utilization by the mitochon- of the nephron progenitors. dria. Instead of pyruvate shuttling into the mitochondria, it is preferentially converted into lactate and recycled back into the glycolytic pathway. The profound metabolic alteration DISCUSSION leads to decreased differentiation indicated by reduced nephron numbers at both P1 and P21. This model simulates From our study it is clear that the VHL/HIF pathway plays a pathologic hypoxia, which has previously shown to cause a major role in differentiation decisions during nephron pro- reduction in nephrons13; however, the underlying mechanisms genitor maturation. Our data reveals a previously unknown have not been previously identified. role for VHL in mediating metabolic switching during kidney It is well known that kidney formation is highly dependent development that is broadly applicable to other developing on oxygen concentration, which is largely regulated by VHL 2 2 systems. Mechanistically, our data suggests that VHL is re- and HIFs. Of particular interest is the fact that VHLNP / quired for nephron progenitors to adapt to increasing oxygen- nephron progenitors have access to oxygen, but remain unable 2 2 ation and, in response, alter their primary metabolism from to respond to the oxygen. We hypothesize that the VHLNP / glycolysis to oxidative phosphorylation. Moreover, we propose differentiation phenotype is due to prolonged maintenance this metabolic switching is an essential process for differenti- of a progenitor niche, as demonstrated by persistent Six2 ating nephron progenitors. Specifically, nephron progenitors and Wt1 expression and decreased nephron number. Six2- lacking VHL exhibit upregulated glycolysis and downregulated expressing cells represent a multipotent nephron progenitor mitochondrial respiration. Although mitochondrial respiration population expressed throughout kidney development,48,49 and was suppressed, mitochondrial density did not differ between Six2 must be downregulated to initiate MET.49,50 Furthermore,

10 JASN JASN 30: ccc–ccc,2019 www.jasn.org BASIC RESEARCH

Basal ECAR Oligomycin-stimulated Glycolysis A 40 O F R+A B C 30 15 ** 30 20 10 20

10 5 10 ECAR (mpH/min) ECAR (mpH/min) ECAR (mpH/min)

0 0 0 0 1530456075 90 105 120 135 VHLNP+/- VHLNP-/- VHLNP+/- VHLNP-/- Time [min] VHLNP+/- VHLNP-/- Media Control YN1 Inhibitor D E NP +/- VHL

FG NP -/- VHL

Figure 7. VHL deletion leads to increased glycolysis. (A) P1 MACS-isolated nephron progenitors were subjected to extracellular acidification (ECAR) measurements in a Seahorse XF24 extracellular flux analyzer as done in Figure 6. (B) Basal glycolysis was de- 2 2 termined by averaging the extracellular acidification over the first 30 minutes and exhibited an upregulated trend (P=0.09) in VHLNP / nephron progenitors. (C) Oligomycin-stimulated glycolysis was calculated by subtracting the basal ECAR from the average of the values 2 2 obtained after oligomycin addition. Oligomycin-stimulated glycolysis was significantly increased in VHLNP / nephron progenitors. 2 2 **P,0.01; n=3. Error bars indicate SEM. (D–G) Immunofluorescence staining of kidney explants revealed that in media alone VHLNP / 2 kidneys differentiated less than VHLNP+/ kidneys in media alone. When cultured in the presence of the glycolysis inhibitor YN1 both 2 2 2 VHLNP / and VHLNP+/ kidneys underwent similar levels of differentiation. F, FCCP (Carbonyl cyanide-p-trifluoromethoxyphenylhydrazone; O, Oligomycin; R+A, Rotenone & Antimycin A. another study has shown that HIF-1a blocks MET,51 anec- of normal differentiation; however, after this time they fail to essary process for differentiating nephron progenitors, undergo sufficient differentiation. A previous study by Wang which serves as another potential mechanism leading to de- et al.53 also investigated renal structure and function in Six2- creased nephron number. Although our model has a signif- creEGPF-VHL null mice for the purpose of studying renal cell icant reduction in nephron number, our data suggests that carcinoma. Although our model exhibits profound abnor- either a subset of nephron progenitors remain capable of malities, their mice are histologically normal and do not maturing into nephrons or that the differentiation process exhibit any malformations or dysfunction until approxi- 2 2 is not as efficient in VHLNP / mice. Similarly, when VHL is mately 4 months of age.53 We see an accelerated phenotype, deleted from Foxd1-expressing stromal progenitors, the re- whichweattributetodifferencesinmousebackground sulting offspring exhibit nephron deficit and postnatal death strains.54–56 by 14 days of age, which was attributed to increased HIF Another significant function of VHL-mediated HIF deg- expression.52 Our data shows that the nephrons that do radation is the regulation of metabolism. Previous interro- form are either derived from nephron progenitors that un- gations of HIF implicate it in the regulation of glycolysis and derwent differentiation early (before E15.5) or due to inef- oxidative phosphorylation.17,18 Although VHL has not been 2 2 ficient nephron progenitor differentiation. The VHLNP / interrogated as a regulator of metabolism in nephron progen- mice do not exhibit a profound phenotype until E15.5, itors, HIF-1a is a known mediator in other cell types, suggest- which we hypothesize is because of the physiologically hyp- ing that VHL must also play an important role in regulating oxic environment the kidneys are subjected to in early de- the metabolic profile of these progenitors. Moreover, VHL- velopment, before VHL is critical. Therefore in the absence related cancers exhibit similarly decreased mitochondrial res- of VHL before E15.5, the nephron progenitors are capable piration in favor of glycolic metabolism even in the presence

JASN 30: ccc–ccc,2019 Von Hippel-Lindau Acts as a Metabolic Switch 11 BASIC RESEARCH www.jasn.org

A B EG NP +/- NP +/- VHL VHL

C D FH NP -/- NP -/- VHL VHL

I 15000 **** J 100 **** 80 10000 60

Nglom 40 5000 BUN mg/dL 20

0 0 VHLNP+/- VHLNP-/- VHLNP+/- VHLNP-/-

2 2 Figure 8. VHLNP / mice exhibit a glomerular deficit at P21. (A–D) Hematoxylin and eosin staining of kidneys sectioned at P21 showing reduced nephron number (black arrows). (E–H) There are no differences in Glut1 or Pck1 expression in P21 kidneys. Lotus 2 2 2 2 2 tetragonolobus lectin (LTL) staining reveals proximal tubule dilation in VHLNP / + kidneys. (I) VHLNP+/ and VHLNP / glomerular counts at P21 revealed approximately 50% reduction in glomerular number (Nglom). ****P,0.001; n=4. Error bars indicate SEM. (J) 2 2 2 2 2 BUN levels are significantly higher in VHLNP / mice compared with littermate controls. ****P,0.001; VHLNP+/ n=8; VHLNP / n=6. Error bars indicate SEM.

of oxygen, a phenomenon known as the Warburg effect.57–59 glycolysis, mitochondrial respiration, and cell fate are 2 2 Although our VHLNP / model is not a cancer model, these ongoing. mice do exhibit severe alterations in the nephron progenitor metabolic profile consistent with the Warburg effect. In ad- dition to enhanced glycolysis, a previous study investigated ACKNOWLEDGMENTS the metabolic profiles of E13.5 and P0 nephron progenitors and showed that younger progenitors demonstrate a higher Cargill designed and performed experiments, provided intellec- dependence on glycolysis whereas older progenitors are tual input, performed data analysis, and wrote the entirety of the more dependent on oxidative phosphorylation.22 Our study manuscript.Dr.Hemker,Dr.Murali,Dr.Mukherjee,Dr.Bushnell, 2 2 suggests that the phenotype associated with the VHLNP / Dr. Bodnar, and Dr. Liu conducted experiments and provided mice can be attributed to metabolic alterations leading to intellectual input. Dr. Clugston and Dr. Kostka performed RNA- an inability of the cells to transition from glycolysis to oxi- sequencing data analysis and wrote RNA-sequencing methods. dative phosphorylation. Dr.Hemker,Dr.Bates,Dr.Ho,Dr.Saifudeen,Dr.Goetzman,and Insummary,ourstudysuggeststhatVHLiscrucialfor the Dr. Sims-Lucas edited the manuscript and provided intellectual metabolic switching required during progenitor cell differ- input to the project. All authors approved the final version of this entiation, and validates that strict regulation of oxygenation manuscript. is vital for cellular differentiation. Thus, VHL is also an The authors would like to thank the flow cytometry core at the essential mediator dictating proliferative, poised, and dif- University of Pittsburgh Medical Center, Children’s Hospital of Pitts- ferentiating nephron progenitors and likely plays a critical burgh Rangos Research Center for assistance throughout this project. role in cell fate decisions. Additional mechanistic interro- This publication was made possible by our funding sources: gation to describe the roles of VHL and HIFs in anaerobic three-dimensional reconstruction experiments are supported by

12 JASN JASN 30: ccc–ccc,2019 www.jasn.org BASIC RESEARCH

The O’Brien Center National Institute of Diabetes and Digestive and 7. Boyle S, Shioda T, Perantoni AO, de Caestecker M: Cited1 and Cited2 Kidney Diseases (NIDDK) P30 (DK079307), Cargill is supported by are differentially expressed in the developing kidney but are not re- – the University of Pittsburgh Medical Center, Children’sHospitalof quired for nephrogenesis. Dev Dyn 236: 2321 2330, 2007 8. Nagalakshmi VK, Yu J: The ureteric bud epithelium: Morphogenesis Pittsburgh and by NIDDK F31 (DK116370), and Sims-Lucas is sup- and roles in metanephric kidney patterning. Mol Reprod Dev 82: 151– ported by NIDDK K01 (DK096996) and R03 (DK110503). 166, 2015 9. Tanigawa S, Wang H, Yang Y, Sharma N, Tarasova N, Ajima R, et al.: Wnt4 induces nephronic tubules in metanephric mesenchyme by a DISCLOSURES non-canonical mechanism. Dev Biol 352: 58–69, 2011 Ms. Cargill and Mrs. Hemker as well as Drs. Saifudeen, Ho, Bates, Goetzman 10. Little MH, McMahon AP: Mammalian kidney development: Princi- and Sims-Lucas report grants from National Institute of Diabetes and Digestive ples, progress, and projections. Cold Spring Harb Perspect Biol 4: and Kidney Diseases (NIDDK), during the conduct of this study. Dr. Kostka, a008300, 2012 report grants from National Institute of General Medical Sciences (NIGMS), 11. Hemker SL, Sims-Lucas S, Ho J: Role of hypoxia during nephrogenesis. – during the conduct of this study. Pediatr Nephrol 31: 1571 1577, 2016 12. Rymer C, Paredes J, Halt K, Schaefer C, Wiersch J, Zhang G, et al.: Renal blood flow and oxygenation drive nephron progenitor differentiation. Am J Physiol Renal Physiol 307: F337–F345, 2014 SUPPLEMENTAL MATERIAL 13. Gonzalez-Rodriguez P Jr., Tong W, Xue Q, Li Y, Hu S, Zhang L: Fetal hypoxia results in programming of aberrant angiotensin ii receptor This article contains the following supplemental material online at expression patterns and kidney development. Int J Med Sci 10: 532– http://jasn.asnjournals.org/lookup/suppl/doi:10.1681/ASN.2018111170/-/ 538, 2013 14. Bernhardt WM, Schmitt R, Rosenberger C, Münchenhagen PM, DCSupplemental. Gröne HJ, Frei U, et al.: Expression of hypoxia-inducible transcrip- Supplemental Figure 1. P1 renal histology reveals smaller kidneys tion factors in developing human and rat kidneys. Kidney Int 69: 114– with fewer glomeruli. 122, 2006 2 2 Supplemental Figure 2. VHLNP / kidneys remain in an early 15. Majmundar AJ, Wong WJ, Simon MC: Hypoxia-inducible factors and progenitor state with decreased differentiation. the response to hypoxic stress. Mol Cell 40: 294–309, 2010 2 2 Supplemental Figure 3. VHLNP / kidneys do not exhibit alter- 16. Iyer NV, Kotch LE, Agani F, Leung SW, Laughner E, Wenger RH, et al.: Cellular and developmental control of O2 homeostasis by hypoxia-in- ations in proliferation or apoptosis across development. ducible factor 1 alpha. Genes Dev 12: 149–162, 1998 Supplemental Figure 4. RNA-sequencing gene expression clustering. 17. Formenti F, Constantin-Teodosiu D, Emmanuel Y, Cheeseman J, Supplemental Figure 5. RNA-sequencing gene expression. Dorrington KL, Edwards LM, et al.: Regulation of human metabolism by 2 2 Supplemental Figure 6. VHLNP / kidneys are significantly smaller hypoxia-inducible factor. Proc Natl Acad Sci U S A 107: 12722–12727, 2 than VHLNP+/ kidneys. 2010 18. Papandreou I, Cairns RA, Fontana L, Lim AL, Denko NC: HIF-1 mediates Supplemental Figure 7. Wnt4VHL kidneys do not exhibit any adaptation to hypoxia by actively downregulating mitochondrial oxy- histological differences by P21. gen consumption. Cell Metab 3: 187–197, 2006 Supplemental Table 1. List of primers for real-time quantitative 19. Shyh-Chang N, Daley GQ, Cantley LC: Stem cell metabolism in tissue PCR and genotyping. development and aging. Development 140: 2535–2547, 2013 Supplemental Table 2. List of genes dysregulated in the HIF-1a 20. Baldewijns MM, van Vlodrop IJ, Vermeulen PB, Soetekouw PM, van pathway. Engeland M, de Bruïne AP: VHL and HIF signalling in renal cell carci- nogenesis. J Pathol 221: 125–138, 2010 21. LaGory EL, Wu C, Taniguchi CM, Ding CC, Chi JT, von Eyben R, et al.: Suppression of PGC-1a is critical for reprogramming oxidative me- REFERENCES tabolism in renal cell carcinoma. Cell Reports 12: 116–127, 2015 22. Liu J, Edgington-Giordano F, Dugas C, Abrams A, Katakam P, Satou R, 1. Hatini V, Huh SO, Herzlinger D, Soares VC, Lai E: Essential role of et al.: Regulation of nephron progenitor cell self-renewal by in- – stromal mesenchyme in kidney morphogenesis revealed by targeted termediary metabolism. JAmSocNephrol28: 3323 3335, 2017 disruption of Winged Helix transcription factor BF-2. Genes Dev 10: 23. Cargill K, Sims-Lucas S: Metabolic requirements of the nephron [pub- 1467–1478, 1996 lished online ahead of print December 15, 2018]. Pediatr Nephrol 2. CebrianC,AsaiN,D’Agati V, Costantini F: The number of fetal nephron 10.1007/s00467-018-4157-2 progenitor cells limits ureteric branching and adult nephron endow- 24. Forbes JM: Mitochondria-power players in kidney function? Trends ment. Cell Reports 7: 127–137, 2014 Endocrinol Metab 27: 441–442, 2016 3. Brown AC, Muthukrishnan SD, Oxburgh L: A synthetic niche for neph- 25. Rognstad R, Katz J: Gluconeogenesis in the kidney cortex. Quantitative ron progenitor cells. Dev Cell 34: 229–241, 2015 estimation of carbon flow. JBiolChem247: 6047–6054, 1972 4. Li Y, Liu J, Li W, Brown A, Baddoo M, Li M, et al.: p53 Enables metabolic 26. Kobayashi A, Valerius MT, Mugford JW, Carroll TJ, Self M, Oliver G, fitness and self-renewal of nephron progenitor cells. Development 142: et al.: Six2 defines and regulates a multipotent self-renewing nephron 1228–1241, 2015 progenitor population throughout mammalian kidney development. 5. Chen S, Brunskill EW, Potter SS, Dexheimer PJ, Salomonis N, Cell Stem Cell 3: 169–181, 2008 Aronow BJ, et al.: Intrinsic age-dependent changes and cell-cell 27. Haase VH, Glickman JN, Socolovsky M, Jaenisch R: Vascular tumors in contacts regulate nephron progenitor lifespan. Dev Cell 35: 49–62, with targeted inactivation of the von Hippel-Lindau tumor sup- 2015 pressor. Proc Natl Acad Sci U S A 98: 1583–1588, 2001 6. Tanigawa S, Taguchi A, Sharma N, Perantoni AO, Nishinakamura R: 28. Sims-Lucas S: Analysis of 3D branching pattern: Hematoxylin and Eosin Selective In Vitro propagation of nephron progenitors derived from method. In: Kidney Development: Methods and Protocols,editedby embryos and pluripotent stem cells. Cell Rep 15: 801–813, 2016 Michos O, Totowa, NJ, Humana Press, 2012, pp 73–86

JASN 30: ccc–ccc,2019 Von Hippel-Lindau Acts as a Metabolic Switch 13 BASIC RESEARCH www.jasn.org

29. Sims-LucasS,ArgyropoulosC,KishK,McHughK,BertramJF,QuigleyR, 46. Wintour EM, Moritz KM, Johnson K, Ricardo S, Samuel CS, Dodic M: et al.: Three-dimensional imaging reveals ureteric and mesenchymal de- Reduced nephron number in adult sheep, hypertensive as a result of fects in Fgfr2-mutant kidneys. J Am Soc Nephrol 20: 2525–2533, 2009 prenatal glucocorticoid treatment. J Physiol 549: 929–935, 2003 30. Cullen-McEwen LA, Armitage JA, Nyengaard JR, Bertram JF: Estimating 47. Shan J, Jokela T, Peltoketo H, Vainio S: Generation of an allele to in- nephron number in the developing kidney using the physical disector/ activate Wnt4 gene function conditionally in the mouse. Genesis 47: fractionator combination. Methods Mol Biol 886: 109–119, 2012 782 –788, 2009 31. Cullen-McEwen LA, Douglas-Denton RN, Bertram JF: Estimating total 48. O’Brien LL, Guo Q, Lee Y, Tran T, Benazet JD, Whitney PH, et al.: nephron number in the adult kidney using the physical disector/ Differential regulation of mouse and human nephron progenitors by fractionator combination. Methods Mol Biol 886: 333–350, 2012 the Six family of transcriptional regulators. Development 143: 595– 32. Cerqueira DM, Bodnar AJ, Phua YL, Freer R, Hemker SL, Walensky LD, 608, 2016 et al.: Bim gene dosage is critical in modulating nephron progenitor 49. Self M, Lagutin OV, Bowling B, Hendrix J, Cai Y, Dressler GR, et al.: Six2 survival in the absence of microRNAs during kidney development. is required for suppression of nephrogenesis and progenitor renewal in FASEB J 31: 3540–3554, 2017 the developing kidney. EMBO J 25: 5214–5228, 2006 33. Andrews, S: FastQC: A Quality Control Tool For High Throughput Sequence 50. Chung E, Deacon P, Marable S, Shin J, Park JS: Notch signaling pro- Data, 2010. Available at: https://www.bioinformatics.babraham.ac.uk/ motes nephrogenesis by downregulating Six2. Development 143: projects/fastqc/. Accessed October 6, 2011 3907–3913, 2016 34. Bushnell B: BBMap (Version 35.14), 2015. Available at: https:// 51. Evans AJ, Russell RC, Roche O, Burry TN, Fish JE, Chow VW, et al.: sourceforge.net/projects/bbmap/. Accessed February 19, 2014 VHL promotes E2 box-dependent E-cadherin transcription by 35. Church DM, Schneider VA, Graves T, Auger K, Cunningham F, Bouk N, HIF-mediated regulation of SIP1 and snail. MolCellBiol27: 157– et al.: Modernizing reference genome assemblies. PLoS Biol 9: 169, 2007 e1001091, 2011 52. Gerl K, Steppan D, Fuchs M, Wagner C, Willam C, Kurtz A, et al.: 36. Dobin A, Davis CA, Schlesinger F, Drenkow J, Zaleski C, Jha S, et al.: STAR: Activation of hypoxia signaling in stromal progenitors impairs Ultrafast universal RNA-seq aligner. Bioinformatics 29: 15–21, 2013 kidney development. Am J Pathol 187: 1496–1511, 2017 37. R Core Team: R: The R Project for Statistical Computing, 2016. Avail- 53. Wang SS, Gu YF, Wolff N, Stefanius K, Christie A, Dey A, et al.: Bap1 is able at: https://www.r-project.org. Accessed May 30, 2016 essential for kidney function and cooperates with Vhl in renal tumori- 38. Lawrence M, Huber W, Pagès H, Aboyoun P, Carlson M, Gentleman R, genesis. Proc Natl Acad Sci U S A 111: 16538–16543, 2014 et al.: Software for computing and annotating genomic ranges. PLOS 54. Puri TS, Shakaib MI, Chang A, Mathew L, Olayinka O, Minto AW, et al.: Comput Biol 9: e1003118, 2013 Chronic kidney disease induced in mice by reversible unilateral ureteral 39. Love MI, Huber W, Anders S: Moderated estimation of fold change and obstruction is dependent on genetic background. Am J Physiol Renal dispersion for RNA-seq data with DESeq2. Genome Biol 15: 550, 2014 Physiol 298: F1024–F1032, 2010 40. Huang W, Sherman BT, Lempicki RA: Bioinformatics enrichment tools: 55. Walkin L, Herrick SE, Summers A, Brenchley PE, Hoff CM, Korstanje R, Paths toward the comprehensive functional analysis of large gene lists. et al.: The role of mouse strain differences in the susceptibility to Nucleic Acids Res 37: 1–13, 2009 fibrosis: A systematic review. Fibrogenesis Tissue Repair 6: 18, 2013 41. Huang W, Sherman BT, Lempicki RA: Systematic and integrative anal- 56. Hartner A, Cordasic N, Klanke B, Veelken R, Hilgers KF: Strain differ- ysis of large gene lists using DAVID bioinformatics resources. Nat ences in the development of hypertension and glomerular lesions Protoc 4: 44–57, 2009 induced by deoxycorticosterone acetate salt in mice. Nephrol Dial 42. de Moura MB, Uppala R, Zhang Y, Van Houten B, Goetzman ES: Transplant 18: 1999–2004, 2003 Overexpression of mitochondrial sirtuins alters glycolysis and mito- 57. Hervouet E, Demont J, Pecina P, Vojtísková A, Houstek J, Simonnet H, chondrial function in HEK293 cells. PLoS One 9: e106028, 2014 et al.: A new role for the von Hippel-Lindau tumor suppressor protein: 43. Collins CL, Bode BP, Souba WW, Abcouwer SF: Multiwell 14CO2- Stimulation of mitochondrial oxidative phosphorylation complex bio- capture assay for evaluation of substrate oxidation rates of cells in genesis. Carcinogenesis 26: 531–539, 2005 culture. Biotechniques 24: 803–808, 1998 58. Lu H, Forbes RA, Verma A: Hypoxia-inducible factor 1 activation by 44. Park JS, Valerius MT, McMahon AP: Wnt/beta-catenin signaling aerobic glycolysis implicates the Warburg effect in carcinogenesis. regulates nephron induction during mouse kidney development. J Biol Chem 277: 23111–23115, 2002 Development 134: 2533–2539, 2007 59. Chan DA, Sutphin PD, Nguyen P, Turcotte S, Lai EW, Banh A, et al.: 45. Brand MD, Nicholls DG: Assessing mitochondrial dysfunction in cells. Targeting GLUT1 and the Warburg effect in renal cell carcinoma by Biochem J 435: 297–312, 2011 chemical synthetic lethality. Sci Transl Med 3: 94ra70, 2011

14 JASN JASN 30: ccc–ccc,2019