Loss of miR-17~92 results in dysregulation of Cftr in nephron progenitors

Abstract

We have previously demonstrated that loss of miR-17~92 in nephron progenitors in a mouse model results in renal hypodysplasia and chronic kidney disease. Clinically, decreased congenital nephron endowment because of renal hypodysplasia is associated with an increased risk of hypertension and chronic kidney disease, and this is at least partly dependent on the self-renewal of nephron progenitors. Here, we present evidence for a novel molecular mechanism regulating the self-renewal of nephron progenitors and congenital nephron endowment by the highly conserved miR-17~92 cluster. Whole transcriptome sequencing revealed that nephron progenitors lacking this cluster demonstrated increased Cftr expression. We showed that one member of the cluster, miR-19b, is sufficient to repress Cftr expression in vitro and that perturbation of Cftr activity in nephron progenitors results in impaired proliferation. Together, these data suggest that miR-19b regulates Cftr expression in nephron progenitors, with this interaction playing a role in appropriate nephron progenitor self-renewal during kidney development to generate normal nephron endowment.

INTRODUCTION

Congenital anomalies of the kidney and urinary tract are the leading cause of childhood chronic kidney disease (61). The risk of chronic kidney disease is related to decreased renal reserve, resulting from the formation of fewer or abnormal nephrons during kidney development (3, 26). Multipotent Cbp/p300-interacting transactivator 1 (Cited1)⁺/Six2⁺ nephron progenitors give rise to multiple nephron segments, including podocytes, proximal tubules, distal tubules, and loops of Henle (37). The transition of nephron progenitors into epithelialized structures is dictated by a series of tightly orchestrated signaling events, including bone morphogenetic protein 7-phosphorylated Smad1/5/8 signaling, which induces the initial exit of Cited1⁺/Six2⁺ cells into a Cited1^-/Six2⁺ state followed by Wnt9b/β-catenin induction to undergo differentiation (4, 25). Nephron progenitors also undergo self-renewal to prevent premature depletion before the cessation of nephrogenesis (36). Indeed, ablation of nephron progenitors in utero has been shown to result in reduced nephron endowment (6). Emerging studies have now also implicated miRNAs in nephron progenitor biology (8, 22, 44).

miRNAs are 18- to 22-nt-long small noncoding RNAs that posttranscriptionally regulate mRNA levels and/or inhibit protein translation (19, 20). Biosynthesis of mature miRNAs is dependent on both Drosha and Dicer endoribonuclease activity (20, 28). Conditional deletions of Drosha or Dicer in various renal lineages including nephron progenitors, ureteric bud, renal stroma, and podocytes have been reported to result in diverse renal phenotypes (8, 18, 21, 22, 44, 46, 53, 54, 62). We have previously shown that conditional deletion of the miR-17~92 cluster in Six2⁺ nephron progenitors results in renal hypodysplasia, proteinuria, and chronic kidney disease in a mouse model (38). The miR-17~92 cluster includes the following miRNAs: miR-17, miR-18a, miR-19a, miR-20a, miR-19b-1, and miR-92a-1 (42). Interestingly, mutations in the miR-17-92a-1 cluster host gene represent the first miRNA mutations to be implicated in a human syndrome, Feingold syndrome type II (11, 17). However, the molecular targets of the miR-17~92 cluster in nephron progenitors remain unclear.

In the present study, comprehensive transcriptome and bioinformatics analyses of miR-17~92-null nephron progenitors revealed an association between dysregulated miR-17~92 downstream targets and the regulation of cellular proliferation/cell cycle, with Cftr predicted to be the most differentially expressed gene. In vitro assays demonstrated that miR-19b, a member of the miR-17~92 cluster, exhibits a specific repressive effect on Cftr expression and that miR-19b mimics were sufficient to inhibit Cl⁻ conductance by Cftr. Moreover, pharmacological inhibition of Cftr activity in nephron progenitor cultures impaired cellular proliferation, and miR-17~92-null progenitors exhibited disrupted cellular proliferation that resulted in low nephron endowment. Together, these results suggest that miR-17~92 acts to maintain normal nephron progenitor self-renewal during nephrogenesis, at least in part through the modulation of appropriate Cftr levels.

MATERIALS AND METHODS

Mouse strains.

Six2^TGC [Tg(Six2-EGFP/cre)1Amc/J, stock no. 009606, 129-Elite background, where GFP is green fluorescent protein] (31) and conditional miR-17~92^fl/fl (Mirc1^tm1.1Tyj/J, stock no. 008458, C57BL/6J background) (58) mice were obtained from The Jackson Laboratory (Bar Harbor, ME). Animals were genotyped using genomic DNA by PCR (29, 58). All experimental procedures were performed in accordance with University of Pittsburgh Institutional Animal Care and Use Committee guidelines, which adheres to the National Institutes of Health Guide for the Care and Use of Laboratory Animals.

FACS and RNA sequencing analysis.

Embryonic day 16.5 (E16.5) Six2^TGC;129-Elite (control) and Six2^TGC;miR-17~92^fl/fl (mutant) kidneys were dissociated into single cells using 0.05% Trypsin-EDTA (catalog no. 25300054, ThermoFisher) followed by FACS of Six2^GFPHi nephron progenitors directly into QIAzol Lysis Reagent (catalog no. 79306, Qiagen). FACS was performed at the Rangos Research Center Flow Cytometry Core Facility (UPMC Children’s Hospital of Pittsburgh, Pittsburgh, PA). Each replicate consisted of nephron progenitors pooled from 4 embryos, with a total of 12 samples/genotype. Total RNA extraction was done using a miRNeasy Micro Kit (catalog no. 217084, Qiagen). RNA quality was assessed on an Agilent 2100 Bioanalyzer using an Agilent RNA 6000 Pico Kit (catalog no. 5067-1513, Agilent Technologies). RNA samples with a RNA integrity number score of >9 were used for sequencing.

Whole transcriptome RNA sequencing (RNA-seq) was performed at the TUCF Genomics Core Facility (Tufts University, Boston, MA). Briefly, total RNA quantity and quality were first assessed by Fragment Analyzer (Advanced Analytical Technologies) followed by cDNA library construction from equivalent amounts of RNA using an Ovation RNA-Seq System V2 (catalog no. 7102-08, NuGEN), and 2 × 100-bp sequencing was performed using a dual flow cell on an Illumina HiSeq 2500 sequencing system, with all samples multiplexed across 2 lanes, yielding ~70 million reads/sample. Reads were aligned to the mm10 genome using Rsubread (35) and annotated using the built-in package in featureCounts (34) followed by statistical testing using the limma-voom package (50) (Bioconductor) in the R programming language. Genes with a B statistic score of >0 were considered statistically significant (46, 47, 48). Hierarchical distance clustering of the RNA-seq data was performed using Gene-E software (Broad Institute, Cambridge, MA), and pathway analysis was performed using ToppFun (7) and Ingenuity Pathway Analysis software (Qiagen). Reads per kilobase million (RPKM) output from the featureCounts package (34) were converted into transcripts per kilobase million (TPM) using the following formula (32):

$$\frac{\mathrm{R}\mathrm{P}\mathrm{K}\mathrm{M}}{\mathrm{\Sigma }\mathrm{R}\mathrm{P}\mathrm{K}\mathrm{M}\mathrm{o}\mathrm{f}\mathrm{a}\mathrm{l}\mathrm{}\mathrm{l}\mathrm{g}\mathrm{e}\mathrm{n}\mathrm{e}\mathrm{s}}\times {10}^6$$

The RNA-seq data are available at the National Center for Biotechnology Information Gene Expression Omnibus (GEO) under GEO Accession Number GSE98089).

Quantitative PCR was carried out as previously described on a Bio-Rad CFX96 Real-Time PCR Instrument, with threshold cycle (C_t) values normalized to Gapdh and the relative fold change computed using the $2^{-\Delta C_t}$ method (52). Statistical analysis was performed using the limma package (50) (Bioconductor) with batch correction in R, and genes with P values of <0.05 were considered statistically significant. The primers used for quantitative PCR are shown in Table 1.

Table 1.

Primer sequence used for quantitative PCR

Gene	Forward	Reverse
Gapdh	5′-AGGTCGGTGTGAACGGATTTG-3′	5′-TGTAGACCATGTAGTTGAGGTCA-3′
Cftr	5′-GGACTGTGTGGGTAGATGCTC-3′	5′-AGTGTCTCCCCAACAACTCTG-3′
Meox1	5′-GGACCTATCCTGCGTGTTCC-3′	5′-ACTGTTGCCTGAGGCTGTTT-3′
FgfR1	5′-GGGCATTCCTGTTGACCCAT-3′	5′-GCAGAGGTTGCCTTCGGTTA-3′
Eya2	5′-CTGCACCCGTTACTCCCATT-3′	5′-TCCACATGACATCGGGCTTC-3′
Rspo3	5′-GCAGTGTCCCTCAAAACCAAG-3′	5′-CCAGGCCCTAATTCTGAGCA-3′
Shisa3	5′-TTCACCCTCCCATACCCAGT-3′	5′-AACATCGGTGCGCCTAAAGA-3′
Mapk12	5′-TGGTTAACACCCCGACAGTG-3′	5′-GTCATCTGGGGCACTCGAAA-3′
Six4	5′-ACCCAGCTTGACAACAACCA-3′	5′-ACGACAGGCAACAGTCATGT-3′
Rabl2	5′-TCCCTTCACTTCTGCCGAGA-3′	5′-ACTGCAATCGAGTCCTGCTG-3′
Amotl1	5′-TCTCACGTCCACACTCTTGC-3′	5′-AGCCATCATGGGTTCCTGTG-3′
IL33	5′-CCGTAACGGCTCTCCTACTC-3′	5′-TCAGGGGACAACCTGGTAAC-3′
Sulf1	5′-CCGGCAAGGACACACAGTAT-3′	5′-ACCACCAACATTGTGAGCCA-3′
Osbpl8	5′-ACCTGCCACCACTGCATAAT-3′	5′-TGTCAAGGTTTTCCTGTGGCT-3′
Limd1	5′-TGGGGCTTTTGTGACTGGAC-3′	5′-CTGTGTGAGGGTTATCCGGG-3′
Six2	5′-TCTGCTCGGTATCCTTTGGG-3′	5′-TTAAAAATCGGGGTGGTGGTG-3′

Histology, 5-ethynyl-2′-deoxyuridine proliferation analysis,immunostaining, and section in situ hybridization.

Kidneys dissected from E16.5 embryos were fixed overnight in 4% paraformaldehyde and processed into paraffin, and immunofluorescence was carried out as previously described, with antigen retrieval in sodium citrate (47). Primary antibodies including Six2 (catalog no. 11562-1-AP, Proteintech), phosphohistone H3 (catalog no. 369A-1, Sigma), cleaved caspase-3 (catalog no. 9661, Cell Signaling), and Cftr (catalog no. ACL-006, Alomone Laboratories) were used at 1:100 dilution followed by detection with the respective secondary fluorophore antibodies (catalog nos. 711-545-152 and 111-585-003, Jackson ImmunoResearch) at 1:200 dilution. Immunohistochemistry for Cftr was performed using the VECTASTAIN Elite ABC Kit (rabbit IgG) and Diaminobenzidine Peroxidase Substrate Kit (catalog nos. PK-6101 and SK-4100, Vector Laboratories). For 5-ethynyl-2′-deoxyuridine (EdU) analysis, 30 μg/g body wt EdU was administered intraperitoneally to E16.5 Six2^TGC;129-Elite (control) and Six2^TGC;miR-17~92^fl/fl (mutant) pregnant dams 2 h before embryo collection. Kidney samples were processed into paraffin, and EdU staining was performed with the use of a Click-iT EdU Alexa Fluor 594 Imaging Kit (catalog no. C10339, ThermoFisher) in accordance with the manufacturer’s protocol.

Bright-field images were acquired using a Leica DM2500 microscope equipped with a Qimaging QICAM Fast 1394 camera, and immunofluorescence images were acquired using a Zeiss LSM710 confocal microscope. Section in situ hybridization was performed as previously described (47, 51), and the antisense probe was generated against Cftr exon 27 (NM_021050.2) using a PCR-based method with the following primers: forward 5′-GTCCCATAAAGTGGCCTGGA-3′ and reverse: 5′-CGATGTTAATACGACTCACTATAGGGCATCAGGGGGCCGTCTTAAC-3′.

miRNA/mRNA bioinformatics analysis and luciferase assay.

miR-17~92/Cftr binding interactions were analyzed using DIANA Tools microT-CDS (49) with a threshold filter of 0.1. For the luciferase assay, a portion of Cftr cDNA (NM_021050.2, nucleotides: 2103–4316) containing the miR-17~92 binding sites was cloned into the pmirGLO vector (catalog no. E1330, Promega). Fifty thousand human embryonic kidney (HEK)-293 cells were cotransfected with 50 ng pmirGLO-Cftr and 200 ng human argonaute-2 (to minimize RNA-induced silencing complex saturation effects, catalog no. 21981, Addgene) (2, 16, 27) using jetPRIME (catalog no. 114-07, Polyplus). The media were replaced with Opti-MEM (catalog no. 51985034, ThermoFisher) 5 h posttransfection and transfected with 10 nM miRNA mimics (catalog no. 47XXXX-001, Exiqon) using INTERFERin (catalog no. 409-10, Polyplus) for 48 h. The luciferase assay was carried out using the Dual-Luciferase Reporter Assay System per the manufacturer’s instructions (catalog no. E1910, Promega). Firefly ratios were normalized to the internal Renilla control. One-way ANOVA was used to determine statistical significance in GraphPad PRISM.

mIMCD3 cell transfection and short-circuit current recordings.

mIMCD3 cells were maintained in 75-cm² flasks in DMEM (catalog no. 10569010, ThermoFisher) supplemented with 10% FCS. For transfection with miR mimic or scrambled control RNA, cells were transferred to six-well culture dishes (catalog no. 140675, ThermoFisher) and transiently transfected (overnight) using Lipofectamine 2000 (catalog no. 11668027, ThermoFisher) according to the manufacturer’s instructions and as previously described for other epithelial cells (13, 14). Cells were then subcultured onto permeable filter supports (0.4-µm pore size, 0.33- or 4-cm² surface area, catalog nos. 3470 and 3450, Transwell, Corning) for use in Ussing chambers. After 48 h on filter supports, cells had formed a tight, transporting monolayer, and inserts were mounted in modified Ussing chambers (P2300, Physiologic Instruments) and continuously short circuited with an automatic voltage clamp (VCC MC8, Physiologic Instruments), as previously described (13, 14). The apical and basolateral chambers each contained 4 ml of Ringer solution (120 mM NaCl, 25 mM NaHCO₃, 3.3 mM KH₂PO₄, 0.8 mM K₂HPO₄, 1.2 mM MgCl₂, 1.2 mM CaCl₂, and 10 mM glucose). Chambers were constantly gassed with a mixture of 95% O₂ and 5% CO₂ at 37°C, which maintained pH at 7.4 and established a circulating perfusion bath within the Ussing chamber. Simultaneous transepithelial resistance was recorded by applying a 2-mV pulse/min via an automated pulse generator. Recordings were digitized and analyzed using PowerLab (AD Instruments). To examine CFTR-mediated transport in isolation, 10 μM amiloride (catalog no. A7410, Sigma) was added to the apical hemichamber to eliminate any contribution of epithelial Na⁺ channel transport to the measured short-circuit current (I_SC). Cells were then stimulated with the cAMP agonist forskolin (10 μM, catalog no. F3917, Sigma) to activate CFTR. Once a new steady-state I_SC had been reached, the CFTR-specific inhibitor 20 μM CFTR_INH-172 (catalog no. C2992, Sigma) was added to the apical surface to inhibit CFTR and determine the contribution of CFTR to the measured I_SC. To compare results from multiple transfection and recording days, I_SC was normalized to average CFTR-specific I_SC from control (scrambled) transfected mIMCD3 cells for each group.

Nephron progenitor culture and in vitro assays.

Nephron progenitors sorted from E16.5 Six2^TGC;129-Elite embryos were cultured and maintained in nephron progenitor expansion media (5) as previously described with the following modifications: 1) the CHIR concentration in the media was reduced to 1 μM, 2) media were supplemented with 1,000 U/ml mouse leukemia inhibitory factor (catalog no. ESG1106, Millipore), and 3) progenitors were cultured as three-dimensional aggregates (55, 56). For the Cftr activator assay, an equivalent number of live cells in each batch was pelleted and treated either with DMSO or 1 μM VX-770 (catalog no. A5047, Apex Bio) using the trypan blue exclusion assay for viable cells. At 48 h posttreatment, the pellet was dissociated into single cells, and viable and nonviable cell counts were performed using trypan blue exclusion to compare the cell counts of DMSO- or VX-770-treated pellets. Statistical analysis was conducted by repeated-measures one-way ANOVA followed by post hoc Tukey analysis for multiple comparisons.

Estimation of total glomerular number.

Control and mutant kidneys were harvested from mice at postnatal day 21, fixed in 4% paraformaldehyde, and serially sectioned at 4 μm. For the determination of the number of glomeruli, sections were stained with hematoxylin and eosin. A pair of consecutive sections out of every 26 sections was chosen and analyzed using a Walcom drawing tablet (Walcom) and Stereoinvestigator version 9.04 software using a physical fractionator probe (MBF Bioscience). The total glomerular number (N_glom) was calculated using the following equation:

$$N_{\text{glom}}=\frac{\text{Σ}{Q}^{-}}{2}\times \frac{A}{a}\times \frac{1}{\text{ASF}}$$

where N_glom is the total number of glomeruli in the entire kidney, ΣQ⁻ is the total number of glomeruli appearing and disappearing between two consecutive sections, A is the grid size, a is the counting frame size, and 1/ASF is the reciprocal of the section sampling fraction (the number of sections advanced between section pairs) (45).

Statistical analyses.

Unless indicated otherwise, statistical analyses were performed using a two-tailed t-test in GraphPad Prism. P values of <0.05 were considered as statistically significant. Values are represented as mean ± SE.

RESULTS

Loss of miR-17~92 results in an intrinsic nephron progenitor defect.

We have previously shown that deletion of the miR-17~92 cluster in nephron progenitors in mice results in chronic kidney disease (38). However, the molecular pathways regulated by miR-17~92 in nephron progenitors are unknown. Since the initial phenotypic changes occurred around E16.5 in our prior study, we isolated nephron progenitors from E16.5 Six2^TGC (control) and Six2^TGC;miR-17~92^fl/fl (mutant) kidneys based on GFP^Hi fluorescence driven by the Six2 locus (Fig. 1, A and B) for whole transcriptome RNA-seq. Initial analysis of the RNA-seq data revealed marked enrichment for nephron progenitors in both control- and mutant-sorted samples, as measured by Six2, Cited2, crystallin-μ (Crym), and eyes absent 1 (Eya1) TPM, and minimal cellular contamination from other nephron lineages based on markers associated with the renal stroma, ureteric bud, differentiating nephron, and vasculature in TPM (GEO Accession Number GSE98089; see Supplemental Table S1, available online at https://doi.org/10.6084/m9.figshare.7629263.v2). Furthermore, there were ~10,064 genes expressed in nephron progenitors, defined as having >5 TPM. Mutant nephron progenitors displayed reduced expression of several nephron progenitor-associated transcripts [Six2, mesenchyme homeobox 1 (Meox1), FGF receptor 1 (Fgfr1), eyes absent 2 (Eya2), R-spondin 3 (Rspo3), Shisa3, and MAPK12 (Mapk12)] when measured with quantitative PCR, supporting the idea that there is an intrinsic defect in mutant progenitors (Fig. 1C). As expected, miR-17~92 precursor transcripts were not evident in mutant nephron progenitors (GEO Accession Number GSE98089).

Fig. 1.

Whole transcriptome RNA sequencing of nephron progenitors isolated from Six2^TGC (control) and Six2^TGC; miR-17~92^fl/fl (mutant) embryonic day 16.5 (E16.5) kidneys is consistent with a nephron progenitor defect. A: immunofluorescence of a wild-type E16.5 kidney immunostained for Six2 to show Six2^Hi (self-renewing) and Six2^Lo (committed to differentiate) nephron progenitor populations. B: FACS plot demonstrating isolation of Six2^Hi nephron progenitors from control and mutant E16.5 kidneys. C: quantitative real-time PCR confirming the decreased expression of several nephron progenitor markers. n = 3 per group. *P < 0.05; **P < 0.01; ***P < 0.001. D and E: graphical representation of transcripts per kilobase million (TPM) for markers of self-renewing progenitors (left) and progenitors committed to differentiation (right). Data were curated from the RNA sequencing results with n = 3 per group.

Further analysis for molecular markers associated with the induction of nephron progenitors to differentiate into renal vesicles [family with sequence similarity 132 member A (Fam132a), cadherin 4 (Cdh4), FGF-9 (Fgf9), glutathione peroxidase 6 (Gpx6), lymphoid enhancer-binding factor 1 (Lef1), musashi RNA-binding protein 2 (Msi2), paired box 8 (Pax8), phospholipase C-η₁ (Plch1), coiled-coil domain containing 88C (Ccdc88c), Slc45a3, sorbin and SH3 domain containing 2 (Sorbs2), and Wnt4] or progenitor self-renewal [amphiphysin (Amph), UNC homeobox (Uncx), BTB (POZ) domain containing 11 (Btbd11), Cited1, chemokine (C-X-C motif) receptor 4 (Cxcr4), claudin 9 (Cldn9), Crym, ETS variant 5 (Etv5), family with sequence similarity 19 member A5 (Fam19a5), glial cell line-derived neurotrophic factor (Gdnf), integrin-α₈ (Itga8), phospholipase A₂ group VII (Pla2g7), Slc12a2, and transferrin receptor (Tfrc)] (25) did not reveal any significant changes between control and mutant nephron progenitors in the RNA-seq data (Fig. 1, D and E). Similarly, consistent with our previous findings (38), the RNA-seq data revealed no significant changes in Bim (Bcl2L11) (Supplemental Table S1, available online at https://doi.org/10.6084/m9.figshare.7629263.v2). The data, therefore, indicate that loss of miR-17~92 did not significantly impact signaling pathways that would otherwise promote ectopic nephron formation, rapid exhaustion of the progenitor pool, or cellular death.

Dysregulated transcripts in miR-17~92-null nephron progenitors are associated with proliferation and motility.

A total of 240 (B statistics > 0) differentially expressed genes, with 132 upregulated genes in mutant nephron progenitors, were identified in the RNA-seq data (Supplemental Table S1). Unbiased pathway analysis using ToppFun and Ingenuity Pathway Analysis revealed multiple biological processes predicted to be affected in mutant nephron progenitors, including cell-cell adhesion, motility, and cellular activation (Fig. 2 and Supplemental Table S2), all of which are known to be important for nephron progenitor behavior (9). Positive regulation of multicellular organismal process (Gene Ontology: 0051240), which includes cellular proliferation, was predicted to be highly associated with the mutant nephron progenitor phenotype (Fig. 2A and Supplemental Table S2). Similarly, Ingenuity Pathway Analysis also revealed an enrichment score for cell-cell signaling and proliferation (Table 2). Among the upregulated transcripts in mutant nephron progenitors, 15 transcripts have been predicted to be targets of miR-17~92 using the bioinformatics prediction algorithm from DIANA Tools (49), including Cftr, integrin-β₈ (Itgb8), stearoyl-CoA desaturase (Scd), very-low-density lipoprotein receptor (VldlR), sulfatase 1 (Sulf1), family with sequence similarity 160 member B2 (Fam160b2), phosphodiesterase 1C (Pde1c), tripartite motif containing 36 (Trim36), and TNF superfamily member 11 (Tnfsf11) (Table 3). Thus, these potential targets may act independently or in concert to regulate proliferation in nephron progenitors (Fig. 2B).

Fig. 2.

Bioinformatics analysis revealed dysregulated biological processes in embryonic day 16.5 Six2^TGC;miR-17~92^fl/fl nephron progenitors. A: ToppFun analysis of the RNA sequencing data demonstrated that the dysregulated transcripts are associated with multiple biological processes, including multicellular organisms process, cell-cell adhesion, and motility (marked with *). B: Ingenuity Pathway Analysis revealed an interlinked network of dysregulated transcripts in the mutant nephron progenitors that are involved in cellular proliferation. Upregulated transcripts are indicated in red. GO, Gene Ontology.

Table 2.

Ingenuity Pathway Analysis of dysregulated transcripts in embryonic day 16.5 Six2^TGC;miR-17~92^fl/fl nephron progenitors revealed candidate pathways and processes that are dysregulated

ID	Genes	Score	Focus Molecules	Top Diseases and Functions
1	Akt, ALK3-BMPR2, ANGPT4, ANKRD42, B3GNT9, BPI, CFTR, CLEC4C, CSF2, Cyp2j9, ERK1/2, FAM160B2, Fcna, GLP2R, IFNB1, Il15r, IL22R1-IL10R2, INSRR, Insulin, ITGB8, LRRC26, NfkB (complex), NPS, p38 MAPK, PDE1C, SCD, SIGLEC15, SULF1, Tnfrsf22/Tnfrsf23, TNFSF11, TRAF1-TRAF2-TRAF3, TRIM36, VLDLR, Ybx1-ps3, and ZNF25	27	10	Cell-to-cell signaling and interaction, cellular growth and proliferation, and connective tissue development and function
2	AHCY, ATF2, BTNL2, CD38, Cg, COMT, CRKL, CYLD, F2, FGL2, FL1HLA-B, IFN-α/β, IFNG, IL5, IL21, IRF1, LGALS3, Map3k7, MAPK9, MERTK, MET, miR-21, NFATC2, NR1I2, PTPRJ, RXRA, SAMSN1, SQSTM1, SYK, TBK1, TIGIT, TNFAIP2, TP73, and TRAF3IP1	4	2	Cellular development, cellular growth and proliferation, and hematological system development and function
3	ALX4, CDX1, CDX4, CEBPE, DLX1, DLX2, DLX5, EMX1, ERCC8, FOXA3, FOXE1, FOXH1, GTF3C1, HAND2, HOXA3, HOXA11, HOXB6, HOXB13, HOXC4, HOXD3, HOXD12, HOXD13, LDB1, LMX1B, MED16, MIXL1, NCAM1, POU3F4, POU4F2, PRDM5, PRRX1, RAX, SOX10, SOX15, and TLE6	2	1	Embryonic development, organismal development, and gene expression

Table 3.

Upregulated transcripts in embryonic day 16.5 Six2^TGC;miR-17~92^fl/fl nephron progenitors that are predicted argets of miR-17~92 based on DIANA TOOLS microT-CDS

miRNAs	mRNAs
miR-17-5p	Zfp9, Clec5a, Cftr, Sulf1, Tnfsf11, Trim36, Fam160b2,  and Itgb8
miR-18a-5p	Cftr and Ahcy
miR-19a-3p	Scd1, Cftr, Sulf1, Pde1c, and Itgb8
miR-19b-3p	Scd1, Cftr, Sulf1, Pde1c, and Itgb8
miR-20a-5p	Fgl2, Zfp9, Tnfsf11, Pde1c, Trim36, Fam160b2, Itgb8,  and Asb4
miR-92a-3p	Trim36 and Alx4

Indeed, several of the upregulated transcripts associated with cell motility were confirmed through quantitative PCR (Fig. 3, A and B). Notably, Cftr was found to be the most significantly upregulated transcript in mutant nephron progenitors. Based on the Genitourinary Development Molecular Anatomy Project (GUDMAP, https://www.gudmap.org/) database (40), expression of Cftr is present in multiple renal cell types, including nephron progenitors, developing renal vesicles, proximal tubules, and podocytes. We verified increased Cftr transcripts in the mutant nephron progenitors via quantitative PCR and confirmed the spatial expression pattern of Cftr using section in situ hybridization and immunohistochemistry, which is consistent with GUDMAP data (Fig. 3, C and D). Similarly, upregulation of Cftr expression in mutant nephron progenitors was also validated with immunohistochemistry (Fig. 3, E and F).

Fig. 3.

Cftr was the most differentially upregulated transcript in embryonic day 16.5 (E16.5) Six2^TGC;miR-17~92^fl/fl nephron progenitors compared with controls. A: heat map of the RNA sequencing results showing the most significantly differentially expressed transcripts in mutant versus control nephron progenitors, confirming that the biological replicates were consistent across the pooled samples. B: quantitative real-time PCR confirming the differential expression of Cftr, along with transcripts associated with cell motility. n = 3 per group. *P < 0.05; **P < 0.01; ***P < 0.001. C and D: section in situ hybridization demonstrating increased expression of Cftr in nephron progenitors in E16.5 mutant kidneys compared with control kidneys (left). C′ and D′: higher-magnification images of the dashed box in C and D, respectively. Yellow dashed lines denote the ureteric bud (UB). E and F: this was further validated using immunohistochemistry (IHC) for Cftr. E′ and F′: higher-magnification images of the dashed box in E and F, respectively. G, glomerulus; RV, renal vesicle; SISH, section in situ hybridization.

Cftr is regulated by miR-19b in vitro.

We next sought to verify whether Cftr is a bona fide downstream target of miR-17~92 in nephron progenitors. Using DIANA Bioinformatics Tools for miRNA-mRNA target prediction (49), we uncovered several candidate miR-17~92-binding sites in the Cftr transcript, including miR-17 (2 sites), miR-18a (1 site), and miR-19a and miR-19b (4 sites) (Fig. 4A and Table 3). A portion of the Cftr transcript containing the miR-17~92-binding sites was cloned downstream of the firefly gene in the pmirGLO vector, transfected into HEK-293 cells with the respective miRNA mimics, and assayed for downregulation of luciferase activity. Whereas miR-19a exhibited mild repression, miR-19b consistently exhibited a stronger repressive effect on Cftr-luciferase activity (Fig. 4B). Moreover, mutation of the miR-19b-binding site abolished this repression, suggesting that this interaction is specific (Fig. 4C). The results of the luciferase assay support a regulatory role for miR-19b on the Cftr transcript.

Fig. 4.

Cftr is regulated by miR-19b in vitro. A: miR-17~92 target sites on the Cftr transcript were predicted using DIANA Tools, and binding sites with the highest conservation scores are shown in the mouse and human CFTR genes. miR-17~92-binding sites on the Cftr exons and sequence conservation are indicated by the solid lines and enrichment peaks, respectively. B: luciferase reporter assays demonstrated a significant decrease in luciferase activity when pmirGLO-Cftr was transfected with miR-19b mimic but not miR-17, miR-18a, or miR-19a. n = 3 independent assays. *P < 0.05. C: repression of luciferase activity was specific to the miR-19b-binding site; when mutated, there was derepression of the luciferase activity. n = 5 independent assays. *P < 0.05.

To determine whether miR-19b could repress endogenous levels of Cftr, we used a mIMCD3 cell line that endogenously expresses Cftr and forms a polarized epithelium in Ussing chamber studies. Transfection of miR-19b was performed in the mIMCD3 cell line (Fig. 5A), and the relative amount of Cftr protein on the cell surface was quantified by measuring the levels of Cftr channel activity using an Ussing chamber (33). Under nonstimulated conditions, minimal Cftr current activity was recorded (Fig. 5B). The addition of forskolin, a cAMP agonist that phosphorylates and activates the Cftr channel, resulted in increased current activity in both scrambled and miR-19b mimic-treated cells (Fig. 5B). Cftr current was attenuated immediately upon the addition of CF_INH-172, a Cftr-specific inhibitor (Fig. 5B). Although Cftr was present in both scrambled and miR-19b mimic-treated cells, mIMCD3 cells treated with 25–100 nM miR-19b mimics consistently exhibited lower Cftr current activity in a dose-dependent manner (Fig. 5C). The results are consistent with the idea that miR-19b represses endogenous Cftr at the protein level, which is recorded as a decrease in functional Cftr activity.

Fig. 5.

Short-circuit current (I_SC) measurements from mIMCD3 cells transfected with miR-19b mimics have deceased Cftr activity compared with scrambled controls. A: RT-PCR confirmed the expression of Cftr in mIMCD3 cells but not in cells from the NIH-3T3 cell line. B: representative trace of mIMCD3 cells transfected with 50 nM miR-19b and control RNA. Cells were stimulated with forskolin to activate Cftr (white bar) and inhibited with the specific Cftr inhibitor (CF_INH-172; black bar) to determine the Cftr-specific I_SC. C: summarized data (mean ± SE) from experiments similar to B with I_SC expressed as a percentage of the average scrambled control I_SC for each transfection concentration. n = 6. *P < 0.05; ***P < 0.001.

Potentiation of Cftr channel activity results in decreased proliferation of nephron progenitors.

Cftr channel activity has been previously shown to play a functional role in cellular proliferation, with either overexpression or repression of Cftr resulting in decreased cellular proliferation (1, 30, 60). To determine if this is the case for nephron progenitors, we adopted an in vitro approach to assess whether perturbations to nephron progenitor Cftr channel activity would result in a negative impact on cellular proliferation. Nephron progenitors were first FACS sorted and cultured as three-dimensional aggregates in a chemically defined nephron progenitor expansion media (5) with several modifications made to improve previously reported culture conditions, including reduced CHIR concentration (mimic low levels of canonical Wnt signaling to promote self-renewal of nephron progenitors), inclusion of leukemia inhibitory factor (increase nuclear localization of Six2), and culturing as aggregates (maintain cell-cell contacts and improve overall survival) (5, 56). Using this modified protocol, we observed progressive growth of the pellet over a 6-day period with persistent Six2-driven GFP expression (Fig. 6, A–A”). To assess whether these nephron progenitors retain their multipotency in culture, pellets were treated with a high dose of CHIR (to mimic high canonical Wnt activation), which induced nephron differentiation into presumptive Wilms tumor⁺ glomerular podocytes and E cadherin⁺/LTL⁺ proximal tubules (Fig. 6, B–E).

Fig. 6.

Nephron progenitors cultured in modified nephron progenitor expansion media (NPEM) are multipotent and proliferate less when treated with VX-770. A, A′, and A′′: time-lapse imaging demonstrating nephron progenitor pellet growth in modified NPEM for 6 days with retained Six2-green fluorescent protein (GFP) signals (inset). B and B′: hematoxylin and eosin staining revealed that the nephron progenitor pellet differentiated readily into glomerular-like (arrowheads) and epithelial structures in response to high-dose CHIR at 3 μM. C–E: immunofluorescence staining demonstrated evidence for Wilms tumor (WT1)⁺ presumptive podocytes and E cadherin (Ecad)⁺/LTL⁺ proximal tubule formation in the differentiated pellet. F and G: cell counts revealed that treatment of nephron progenitor pellets with the Ctfr activator VX-770 resulted in reduced proliferation with no significant impact on cell death. n = 5 pellets from 3 independent batches. *P < 0.05.

After establishment of the nephron progenitor culture, we next determined the response of nephron progenitors to increased Cftr channel activity through the use of VX-770 (Ivacaftor), a United States Food and Drug Administration-approved drug for patients with cystic fibrosis that improves Cl⁻ transport by potentiating Cftr channel activity (23). Compared with DMSO-treated nephron progenitors at 48 h, VX-770-treated cells had reduced cell numbers after 48 h (Fig. 6, F and G). There was no significant difference in nonviable cells at 48 h, and both DMSO- and VX-770-treated nephron progenitor pellets demonstrated an increase in cell number over time, suggesting that cell death could not account for the reduced cell numbers in VX-770-treated cells. These data support the idea that increased Cftr channel activity in nephron progenitors has a negative impact on proliferation in this in vitro setting.

In addition, evaluation of the Six2^TGC;miR-17~92^fl/fl mutant nephron progenitors with increased Cftr demonstrated evidence for reduced proliferation at E16.5 on the basis of reduced EdU incorporation in Six2-positive cells in an in vivo setting (Fig. 7, A–C) and was consistent with reduced progenitors per ureteric tip at E16.5 (Fig. 7, D and E). This subsequently resulted in an ~38% reduction in final nephron number in postnatal day 21 mice (Fig. 7E) in control compared with mutant mice, a time point before the development of pathological changes associated with chronic kidney disease in this model (38). Together, these results demonstrate that misregulation of Cftr in the absence of miR-17~92 in mutant nephron progenitors causes impaired cellular proliferation and therefore reduced nephron endowment.

Fig. 7.

Reduced cellular proliferation and nephron endowment in Six2^TGC;miR-17~92^fl/fl kidneys. A–C: 5-ethynyl-2′-deoxyuridine (EdU) analysis and semiquantitative analysis demonstrated a significant reduction in EdU incorporation in Six2-positive mutant nephron progenitors at embryonic day 16.5. D and E: as a consequence of the reduced proliferation, mutant kidneys displayed reduced nephron progenitors per ureteric tip at embryonic day 16.5, with a 38% reduction in final nephron endowment in postnatal day 21 mutant mice. n = 3 per group with 10–15 optical images quantified. ***P < 0.001. For nephron counting, 4 Six2^TGC and 3 Six2^TGC;miR-17~92^fl/fl postnatal day 21 kidneys were quantified. *P < 0.05.

DISCUSSION

In our previous study (38), we demonstrated that loss of miR-17~92 resulted in renal hypodysplasia and chronic kidney disease. However, the molecular pathways regulated by miR-17~92 in nephron progenitors remained unknown. In the present study, we identified dysregulated transcripts in mutant nephron progenitors involved in cell-cell adhesion, cell motility, and proliferation. Of these, Cftr was noted to be the most significantly upregulated, was predicted to be an miR-17~92 target, and has been previously shown to be implicated in cell cycle progression (30, 31). In vitro experiments confirmed that miR-19b, one member of the miR-17~92 cluster, was sufficient to repress Cftr mRNA/protein. Furthermore, we demonstrated through our modified nephron progenitor cell culture assay that increased Cftr channel activity can impair cellular proliferation. In Six2^TGC;miR-17~92^fl/fl kidneys, there was an impairment of nephron progenitor proliferation with a subsequent deficit in nephron endowment. Together, these data support a model whereby miR-17~92 posttranscriptional regulation of Cftr is required for appropriate nephron progenitor proliferation, which ultimately impacts congenital nephron endowment.

Nephron progenitors represent a heterogeneous population of cells, which are molecularly and spatially distinct (4, 25, 43). Our RNA-seq data did not reveal evidence for a preponderance of nephron progenitors committed to differentiate (and thus likely to lead to ectopic nephron formation), and there was no difference in Bcl2L11 expression in mutant progenitors, a known target of the miR-17~92 cluster. This latter observation highlights the concept that the specificity of miRNA interactions with their mRNA targets is tissue or cell type context dependent (12). However, the expression of several key nephron progenitor markers, including Meox1, Fgfr1, and Eya2, were decreased, which raised the question of whether there was an alteration in cell cycle dynamics that allow progenitors to self-renew. In keeping with this, the analysis of dysregulated transcripts from the RNA-seq data identified several dysregulated pathways in the nephron progenitors, including cell-cell adhesion, cell motility, and proliferation.

Among several predicted miR-17~92 targets that were upregulated in mutant nephron progenitors, we identified Cftr for further study. Cftr is a transmembrane ion channel that regulates Cl⁻ transport in the cell (10). Although Cftr mutations in patients with cystic fibrosis are commonly associated with lung, pancreatic, and intestinal organ dysfunction (10), renal anomalies including defective proximal tubule endocytosis have been reported in a clinical cohort study (24). Despite the known expression of Ctfr in kidney development and described roles in cystogenesis in polycystic kidney disease (59), the functional role of Cftr during normal kidney development remains elusive. Nevertheless, Cftr-regulated Cl⁻ secretion has been shown to modulate G₂/M cell cycle progression in multiple contexts (15, 30, 31, 41). More importantly, an appropriate level of Cftr is known to be indispensable for supporting normal proliferation, with overexpression or underexpression of Cftr being detrimental (1, 60).

Interestingly, although four of six members of the miR-17~92 cluster were predicted to target Cftr, our data are consistent with this effect being primarily mediated by miR-19b in vitro. This may not be surprising given that members of this cluster are differentially involved in axial patterning, B cell development, and Myc-driven tumorigenesis (58). Cftr has also been proposed to be a target of miR-17 in macrophages, which suggests that miRNA-mediated repression of Cftr is indeed context specific (57). Finally, while miR-19a and miR-19b share a seed sequence, miR-19b exhibited stronger repression in the luciferase assay. We reasoned that differences in miRNA-centered binding sites (i.e., sequences outside of the seed sequence) could be responsible or that they may target a distinct set of not completely overlapping mRNAs, as has been previously reported for miR-10a and miR-10b (39). miRNAs are known to simultaneously target multiple transcripts within a single pathway (20), so we cannot exclude that the cell cycle defects observed in the Six2^TGC;miR-17~92^fl/fl nephron progenitors arise from multiple dysregulated transcripts acting in concert on the cell cycle. Nevertheless, our data suggest that modulation of Cftr activity in nephron progenitors is sufficient to impact cellular proliferation of nephron progenitors in culture. It is also possible that the upregulation of Cftr in mutant nephron progenitors is not wholly miR-19b dependent and that other as yet undefined regulators of Cftr expression are dysregulated.

In summary, we have evidence for a novel regulatory mechanism whereby miR-19b, a member of the miR-17~92 cluster, regulates Cftr expression in nephron progenitors during kidney development in the Six2^TGC;miR-17~92^fl/fl model. These findings have implications for understanding how nephron progenitors are regulated during nephrogenesis and understanding nephron endowment as it relates to chronic kidney disease. Furthermore, these data suggest that screening for renal anomalies would be beneficial in patients with Feingold syndrome type II.

GRANTS

J. Ho’s laboratory is supported by funding from National Institute of Diabetes and Digestive and Kidney Diseases (NIDDK) Grants R00-DK-087922 and R01-DK-103776, March of Dimes Basil O’Connor Starter Scholar Award, and NIDDK Diabetic Complications Consortium pilot Grant DK-076169. Y. L. Phua was a George B. Rathmann Research Fellow supported by the American Society of Nephrology Ben J. Lipps Research Fellowship Program. K. H. Chen and A. K. Marrone were supported by the Research Advisory Committee of Children’s Hospital of Pittsburgh. S. L. Hemker was supported by NIDDK Grant T32-DK-061296. D. Kostka’s laboratory is supported by National Institute of General Medical Sciences Grant GM-115836, and M. B. Butterworth’s laboratory is supported by NIDDK Grant DK-102843.

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the authors.

AUTHOR CONTRIBUTIONS

Y.L.P. and J.H. conceived and designed research; Y.L.P., K.H.C., S.L.H., A.K.M., A.J.B., and X.L. performed experiments; Y.L.P., K.H.C., S.L.H., A.K.M., A.J.B., X.L., A.C., D.K., M.B.B., and J.H. analyzed data; Y.L.P., A.C., D.K., M.B.B., and J.H. interpreted results of experiments; Y.L.P., A.C., and M.B.B. prepared figures; Y.L.P. drafted manuscript; Y.L.P., K.H.C., S.L.H., A.K.M., A.J.B., A.C., D.K., M.B.B., and J.H. edited and revised manuscript; Y.L.P., K.H.C., S.L.H., A.K.M., A.J.B., X.L., A.C., D.K., M.B.B., and J.H. approved final version of manuscript.