Basic Research

Suppression of microRNA Activity in Kidney Collecting Ducts Induces Partial Loss of Epithelial Phenotype and Renal Fibrosis

Sachin Hajarnis, Matanel Yheskel, Darren Williams, Thomas Brefort, Bob Glaudemans, Huguette Debaix, Michel Baum, Olivier Devuyst and Vishal Patel
JASN February 2018, 29 (2) 518-531; DOI: https://doi.org/10.1681/ASN.2017030334

Abstract

microRNAs (miRNAs) are sequence-specific inhibitors of post-transcriptional gene expression. The physiologic function of these noncoding RNAs in postnatal renal tubules still remains unclear. Surprisingly, they appear to be dispensable for mammalian proximal tubule (PT) function. Here, we examined the effects of miRNA suppression in collecting ducts (CDs). To conclusively evaluate the role of miRNAs, we generated three mouse models with CD-specific inactivation of key miRNA pathway genes Dicer, Dgcr8, and the entire Argonaute gene family (Ago1, 2, 3, and 4). Characterization of these three mouse models revealed that inhibition of miRNAs in CDs spontaneously evokes a renal tubule injury–like response, which culminates in progressive tubulointerstitial fibrosis (TIF) and renal failure. Global miRNA profiling of microdissected renal tubules showed that miRNAs exhibit segmental distribution along the nephron and CDs. In particular, the expression of miR-200c is nearly 70-fold higher in CDs compared with PTs. Accordingly, miR-200s are downregulated in Dicer-KO CDs, its direct target genes Zeb1, Zeb2, and Snail2 are upregulated, and miRNA-depleted CDs undergo partial epithelial-to-mesenchymal transition (EMT). Thus, miRNAs are essential for CD homeostasis. Downregulation of CD-enriched miRNAs and the subsequent induction of partial EMT may be a new mechanism for TIF progression.

CKD affects millions of Americans. Irrespective of the primary cause, the final common pathway leading to kidney failure in all forms of progressive CKD is tubulointerstitial fibrosis (TIF). TIF is characterized by renal tubule atrophy associated with the accumulation of fibroblasts and inflammatory cells and the deposition of extracellular matrix. Downregulation of numerous renal tubule miRNAs is observed in models of TIF.1–5 Similarly, hypoxia signaling, a key pathogenic pathway that promotes TIF, globally downregulates miRNA biogenesis in kidneys. However, whether miRNA downregulation is sufficient to produce TIF, or if miRNAs are required for normal postnatal renal tubule function, is not entirely clear.

MicroRNAs (miRNAs) are short, noncoding RNAs that regulate the expression of genes at the post-transcriptional level. Biogenesis of miRNAs involves the transcription of the miRNA loci by RNA polymerase II and its associated machinery to produce primary miRNAs (pri-miRNAs).6,7 The pri-miRNAs are several kilobases in length, have secondary structures, and are capped, spliced, and polyadenylated.8 Within the nucleus, the pri-miRNAs are processed by the microprocessor complex, which comprises the RNAse III enzyme, DROSHA, and the double-stranded RNA (dsRNA)-binding protein, DiGeorge syndrome critical region 8 (DGCR8).9,10 The initial recognition of the pri-miRNA by DGCR8 is followed by DROSHA-mediated processing to generate the 60–70-nucleotide (nt) hairpin-shaped precursor miRNAs (pre-miRNAs).9–13 The pre-miRNAs are exported from the nucleus to the cytoplasm by exportin 5 (XPO5).14–16 Within the cytoplasm, the pre-miRNAs are processed by a second RNAse III enzyme, DICER1, and the dsRNA binding protein, TRBP (transactivation-responsive RNA-binding protein), to produce the mature, approximately 22-nt miRNA duplexes.17–20 These miRNA duplexes are further processed by members of the Argonaute (AGO) protein family (AGO1, AGO2, AGO3, or AGO4), which along with other proteins constitute the miRNA-induced silencing complex (miRISC). One strand of the miRNA is retained by the miRISC, where Watson–Crick base-pairing between the miRNA seed sequence and complementary sequences located primarily within the 3′-UTRs of mRNAs results in the inhibition of mRNA translation.21–23

The primary objective of our study was to elucidate the physiologic function of miRNAs in kidney tubules. Because deletion of Dicer from postnatal mammalian proximal tubules (PT) does not affect PT histology or function,24 we focused our studies on collecting ducts (CDs). Conditional Dicer inactivation is a commonly-used approach to investigate the effects of inhibiting miRNA activity. However, recent studies have uncovered several miRNA-independent functions of DICER1.25–31 Therefore, in addition to generating CD-specific Dicer mutant mice, we also produced mice with CD-specific deletion of Dgcr8, to disrupt the upstream miRNA microprocessor complex, and CD-specific deletion of the entire AGO gene family, to disrupt the downstream miRISC. Characterization of these various mouse models conclusively proved that suppression of miRNA activity in CDs spontaneously evokes a CKD-like, tubulointerstitial phenotype. Furthermore, we found that PTs and CDs exhibited distinct miRNA expression patterns and identified miR-200 family members as CD-enriched miRNAs. Additional studies revealed that miR-200 might aid in maintaining CD homeostasis by suppressing mRNA targets that activate epithelial-to-mesenchymal transition (EMT), profibrotic, and proinflammatory pathways.

Results

CD-Specific Inactivation of Dicer Produces Progressive Kidney Fibrosis

To study the role of miRNAs in CDs, we began by generating Pkhd1/Cre;DicerF/F (Dicer-KO) transgenic mice. Pkhd1 promoter drives Cre expression in CDs but not in any other parts of the kidney. Extremely sparse Pkhd1-Cre–mediated recombination is observed in CDs at postnatal day (P) 1. Pkhd1-Cre activity is observed in approximately 50% of CDs at P4 and approximately 100% of CDs at P7.32 To characterize Pkhd1-Cre activity in our mouse strains, we generated Pkhd1/Cre; R26R-EYFP reporter mice, which carry an enhanced yellow fluorescent protein (EYFP) reporter gene that is activated by Pkhd1-Cre–mediated loxP recombination. Consistent with previous studies, we observed that 100% of EYFP+ tubules were also positive for aquaporin-2 (AQP2), a CD marker (Supplemental Figure 1). Thus, this approach permitted targeted deletion of Dicer specifically from postnatal CDs. Quantitative PCR (Q-PCR) confirmed that the expression of Dicer was decreased in kidneys from Dicer-KO mice compared with control mice (Figure 1A). H&E staining revealed normal kidney histology at P14, including normal-appearing CDs, in Dicer-KO mice (Supplemental Figure 2). To assess CD number, kidney sections from P14 control and Dicer-KO mice were stained with AQP2. Quantification revealed that an equal number of AQP2-positive CDs were present in kidneys of 14-day-old control and Dicer-KO mice (Supplemental Figure 2). These observations suggested that Pkhd1/Cre-mediated inactivation of Dicer did not affect postnatal CD maturation.

Figure 1.
Figure 1.

Deletion of Dicer from CDs produces progressive kidney fibrosis. (A) Q-PCR established that the expression of Dicer was decreased by 50% in kidneys from 10-day-old Pkhd1/Cre;DicerF/F (Dicer-KO) mice compared with control mice (n=4). (B) Kaplan–Meier survival curves of Dicer-KO and control mice are shown. The median survival of Dicer-KO mice was reduced to approximately 6 months. (C) Gross images of kidneys from 280-day-old control and Dicer-KO mice are shown. Dicer-KO kidneys were smaller and atrophic compared with control kidneys. (D) H&E and PicroSirius Red staining from 35-day-old control mice demonstrated normal glomerular and tubular histology, whereas Dicer-KO mice exhibited progressive tubular atrophy and inflammatory infiltrates. (E) Serum creatinine levels of Dicer-KO mice compared with control mice at 3 weeks (P21, n=3–4), 5 weeks (P35, n=3), and 15 weeks of age (P105, n=5–6). (F and G) Q-PCR analysis showed that the expression of kidney injury markers Kim1 and Ngal is increased in kidneys from Dicer-KO mice compared with control mice (P10, n=4; P14, n=4; P21, n=6–7; P35, n=3). (H) Q-PCR analysis showed that the expression of the fibrotic markers Acta2, Col1a1, Vimentin, and Col4a6 was markedly increased in kidneys of 5-week-old Dicer-KO mice compared with control mice (n=3). (I) Kidney sections from 5-week-old control and Dicer-KO mice were stained with antibodies against FSP-1(green, a and d) and α-SMA (red, b and e), both markers of fibrosis, and F4/80+, a macrophage marker (green, c and f). Dicer-KO kidneys showed increased inflammation and interstitial fibrosis compared with control kidneys. Error bars indicate SEM. *P<0.05; ns, P>0.05.

Next, to determine whether Dicer is required for CD homeostasis in older mice, we monitored a cohort of control and Dicer-KO mice for approximately 1 year. Initially, Dicer-KO adults appeared healthy and were fertile. However, their median survival was reduced to only 6 months. In comparison, control mice lived well past 1 year (Figure 1B). Gross examination revealed that kidneys from aged Dicer-KO mice were small and atrophic (Figure 1C). To characterize this phenotype in more detail, we performed histologic and molecular analyses and measured the renal function of Dicer-KO mice at earlier ages. Hematoxylin and eosin (H&E) and PicroSirius Red staining revealed normal kidney histology at P18 (Figure 1D). However, at P21, Q-PCR analysis demonstrated a marked elevation in the expression of kidney tubule injury markers Kim1 and Ngal in Dicer-KO kidneys compared with control kidneys (Figure 1, F and G). By P35, patchy interstitial fibrosis surrounding atrophic cortical CDs was observed in Dicer-KO kidneys, whereas control kidneys exhibited normal histology (Figure 1D). Furthermore, the expression of several profibrotic genes (Acta2, Col1a1, Vim, and Col4a6) was increased in P35 Dicer-KO kidneys compared with control kidneys (Figure 1H). Histologic analysis of kidneys from older Dicer-KO mice (P105) revealed that the fibrotic lesions had markedly progressed with age. Accordingly, we observed a progressive rise in serum creatinine levels starting at P35, indicating that kidney fibrosis was associated with reduced renal function in Dicer-KO mice (Figure 1E). To further characterize the fibrotic lesions, kidney sections from P35 Dicer-KO and control mice were stained with antibodies against fibroblast-specific protein–1 (FSP-1), α–smooth muscle actin (α-SMA), and F4/80, a marker of inflammatory cells. The expression of FSP-1 and α-SMA was increased in the interstitium of Dicer-KO kidneys (Figure 1I). Moreover, in addition to fibrosis, immunostaining of Dicer-KO kidneys showed infiltration of F4/80+ inflammatory cells within the renal interstitium (Figure 1I). Costaining of FSP-1, α-SMA, and F4/80 with Dolichos Biflorus Agglutinin (DBA), a CD marker, showed that these markers were expressed in interstitial cells surrounding the CDs (Supplemental Figure 3). Interstitial expression of FSP-1, α-SMA, or F4/80 was not observed in control kidneys. Thus, CD-specific inactivation of Dicer spontaneously evoked a tubule injury–like response in adult mice, which culminated in renal failure due to progressive interstitial fibrosis and interstitial inflammation.

Deletion of Dgcr8 from CDs Produces Renal Fibrosis

The DGCR8-DROSHA microprocessor complex catalyzes the first step in the miRNA biogenesis pathway upstream of DICER1 by cleaving pri-miRNA transcripts into smaller pre-miRNAs. As an alternate approach to study the role of miRNAs, we next examined the effects of inhibiting miRNA biogenesis by disrupting the miRNA microprocessor complex in CDs. To address this question, we generated Pkhd1/Cre;Dgcr8F/F (Dgcr8-KO) mice. Western blot analysis confirmed that the expression of DGCR8 was reduced in kidneys of 35-day-old Dgcr8-KO mice compared with control mice (Figure 2A). Gross, histologic, and molecular studies of kidneys from Dgcr8-KO mice revealed a striking similarity to kidneys from adult Dicer-KO mice. H&E staining of kidney sections from Dgcr8-KO mice exhibited normal kidney histology at P14 (Supplemental Figure 2). Moreover, expression of AQP2 and the number of CDs was also similar between P14 control and Dgcr8-KO kidneys, indicating that Pkhd1/Cre-mediated deletion of Dgcr8 did not affect CD maturation (Supplemental Figure 2). However, at P21, Q-PCR analysis revealed a marked increase in the expression of kidney injury markers (Kim1 and Ngal) in Dgcr8-KO kidneys compared with control kidneys (Figure 2, F and G). By P35, H&E and PicroSirius Red staining demonstrated cortical CD atrophy and interstitial fibrosis (Figure 2C). Consistent with the histologic studies, Q-PCR analysis showed increased expression of several profibrotic genes in kidneys of 35-day-old Dgcr8-KO mice compared with control mice (Figure 2H). Kidneys from approximately 9-month-old Dgcr8-KO mice were small and atrophic, and histologic analysis revealed that the fibrotic lesions had significantly worsened (Figure 2, B and D). Serum creatinine levels progressively increased with age, indicating that kidney injury and fibrosis resulted in reduced renal function in Dgcr8-KO (Figure 2E).Immunofluorescence staining showed that the expression of FSP-1, α-SMA, and F4/80 was increased in kidneys from Dgcr8-KO mice compared with control mice, further indicating that Dgcr8 deletion resulted in interstitial fibrosis and inflammation (Figure 2I, Supplemental Figure 3). Therefore, deletion of Dgcr8 from CDs produces a phenotype which is similar to the one observed in Dicer-KO mice.

Figure 2.
Figure 2.

CD-specific deletion of Dgcr8 results in renal fibrosis. (A) Western blot analysis indicates reduced expression of DGCR8 in kidneys from 35-day-old Pkhd1/Cre;Dgcr8F/F (Dgcr8-KO) mice. GAPDH was used as a loading control. The protein bands were quantified using Quantity One imaging software from Bio-Rad. (B) Gross kidney images from 280-day-old control and Dgcr8-KO mice. Dgcr8-KO kidneys were small and atrophic compared with age-matched control kidneys. (C) H&E and (D) PicroSirius Red staining revealed progressive interstitial fibrosis and tubular atrophy in Dgcr8-KO mice compared with control mice. (E) Serum creatinine levels were elevated in Dgcr8-KO mice compared with control mice (n=4–5). Q-PCR analysis revealed upregulation of kidney injury markers (F) Kim1 and (G) Ngal, indicating progressive renal injury in Dgcr8-KO mice compared with controls (P14, n=3; P21, n=3; and P35, n=5). (H) Q-PCR analysis showed that expression of profibrotic genes Acta2, Col1a1, Vimentin, and Col4a6 was increased in Dgcr8-KO mice compared with control mice (n=5–6). (I) Immunostaining of kidney sections from 35-day-old mice showed increased expression of FSP-1 (red, a and d), α-SMA (red, b and e), and F4/80 (green, c and f) in Dgcr8-KO mice compared with control mice. Error bars indicate SEM. *P<0.05; ns, P>0.05.

CD-Specific Disruption of the miRISC Results in Kidney Fibrosis

In addition to miRNA biogenesis, both DICER1 and DGCR8 regulate other biologic processes such as DNA damage repair and the generation of snoRNAs and lncRNAs. Thus, the kidney fibrosis phenotype could arise due to miRNA-independent functions of DICER1 and DGCR8. To exclude that possibility, we studied the effects of disrupting the multiprotein miRISC that is essential for the normal miRNA function downstream of DICER1 and DGCR8. Because the AGO protein family members are integral components of the miRISC, we generated mice lacking all AGO proteins in the CDs. The murine AGO proteins (AGO1, AGO2, AGO3, and AGO4) are encoded by four genes located on two distinct genomic loci. The Ago3, 1, and 4 gene cluster is located on chromosome 4, whereas Ago2 is located on chromosome 15. To study the role of AGO proteins in kidney, we characterized three different mouse models. First, we generated Pkhd1/Cre;Ago2F/F (Ago2-KO) mice by breeding Pkhd1/Cre mice with Ago2F/F mice that harbor loxP sites flanking exons 9–11 of Ago2. Q-PCR analysis confirmed that the expression of Ago2 was decreased in kidneys from Ago2-KO mice compared with control mice (Figure 3A). Second, we characterized the Ago1/3/4-KO mice, which harbor a germline deletion of the entire Ago1/3/4 gene cluster. Accordingly, by Q-PCR we observed no expression of Ago1, Ago3, or Ago4 in kidneys of Ago1/3/4-KO mice compared with control mice (Figure 3A). Finally, to avoid functional redundancies by the AGO proteins, we generated mice that lacked activity of all four AGO proteins. Ago1/3/4-KO mice were crossed with Pkhd1/Cre;Ago2F/F (Ago2-KO) mice to produce Ago1/3/4-KO; Pkhd1/Cre;Ago2F/F (Ago1–4-KO) mice. Ago1–4-KO mice lacked expression of Ago1, Ago3, and Ago4 from all tissues, whereas Ago2 expression was inhibited only in the renal CDs. Both Ago2-KO and Ago1/3/4-KO mice had normal renal histology and function and exhibited no signs of kidney injury, renal failure, interstitial fibrosis, or the presence of inflammatory cells (Figure 3, B and D–H). In contrast, adult Ago1–4-KO mice recapitulated the Dicer-KO and Dgcr8-KO renal phenotypes. The Ago1–4-KO mice exhibited normal kidney histology at P14 (Supplemental Figure 2). However, by P35, atrophic renal tubules and surrounding interstitial fibrosis were observed. Q-PCR analyses revealed that the expression of markers of kidney injury and fibrosis was increased in Ago1–4-KO kidneys compared with control, Ago2-KO, and Ago1/3/4-KO kidneys (Figure 3, E–G). Gross examination revealed that, whereas kidneys from aged Ago2-KO and Ago1/3/4-KO mice appeared healthy, kidneys from aged (5-month-old) Ago1–4-KO mice were small and atrophic, indicating that fibrosis had progressed with age (Figure 3C). Finally, immunofluorescence staining showed increased interstitial expression of FSP-1, α-SMA, and F4/80 in Ago1–4-KO kidneys (Figure 3H, Supplemental Figure 3). Interstitial FSP-1, α-SMA, and F4/80 expression was not seen in kidneys from control, Ago2-KO, and Ago1/3/4-KO mice. Collectively, these results demonstrate that independent deletion of Ago2 or Ago1/3/4-KO gene cluster does not affect CD homeostasis. However, similar to Dicer or Dgcr8 inactivation, combined deletion of all Ago genes in CDs produces renal failure due to progressive interstitial fibrosis and interstitial inflammation.

Figure 3.
Figure 3.

CD-specific disruption of the miRISC recapitulates the fibrotic phenotype of Dicer-KO and Dgcr8-KO mice. (A) Q-PCR revealed that the expression of Ago2 was decreased by 30% in kidneys from Pkhd1/Cre;Ago2F/F (Ago2-KO) mice compared with control mice (n=3). Expression of Ago1, 3, and 4 was ablated in Ago1/3/4−/− (n=5–6). (B) H&E and PicroSirius Red staining of kidney sections from 5-week old Ago2-KO (Pkhd1/Cre;Ago2F/F) and Ago1/3/4-KO (Ago1–3-4−/−) mice revealed normal glomerular and kidney histology. In contrast, interstitial fibrosis, inflammation, and tubular atrophy was observed in combined Ago1–4-KO mice (Ago1–3-4−/−; Pkhd1/Cre;Ago2F/F). (C) Kidneys from 150-day-old Ago1–4-KO mice were smaller and atrophic compared with Ago1/3/4-KO kidneys. (D) Serum creatinine levels progressively increased in Ago1–4-KO mice. Similar progressive increase was not observed in Ago1/3/4-KO mice (P35, n=3–6; P150, n=3–8). (E–G) Q-PCR analyses revealed increased expression of Kim1 and Ngal, and fibrosis markers Acta2, Col1a1, Vimentin, and Col4a6, in Ago1–4-KO kidneys compared with control, Ago2-KO, and Ago1–3-4−/− kidneys from 5-week-old mice (n=3–6). (H) Kidneys of 5-week-old mice from the indicated models were stained with antibodies against fibrosis marker FSP-1 (green, a–d), α-SMA (red, e–h), and a mouse macrophage marker F4/80+ (green, i–l). Ago1–4-KO kidneys showed increased inflammation and interstitial fibrosis compared with control, Ago2-KO, and Ago1–3-4−/− kidneys. Error bars indicate SEM. *P<0.05; ns, P>0.05.

Segmental miRNA Expression Pattern Is Observed along the Nephron and CDs

Although our results have shown that miRNAs play an essential role in CD homeostasis, a previous study found that DICER1 and, by extension, miRNAs are dispensable for PT homeostasis.24 We reasoned that one potential explanation for these divergent observations may be that CDs and PTs exhibit distinct miRNA expression patterns. Global miRNA expression using RNA extracted from whole kidneys has been previously reported.33 However, the segmental distribution of miRNAs along the length of the nephron and CD is not known. Therefore, we began by first identifying CD-enriched miRNAs in wild-type kidneys. miRNA microarray analysis was performed using RNA isolated from various nephron segments and CDs. Kidneys from adult wild-type C57 BL/6J mice were sectioned through the midsagittal plane. The cortex and inner medulla were removed, treated with collagenase, and microdissected into various nephron segments, viz., glomerulus (GL), proximal convoluted tubule (PCT), proximal straight tubule (PST), thick ascending limbs (TAL), distal convoluted tubule (DCT), and CDs. Q-PCR analysis of segment-specific marker genes confirmed the purity of each fraction (Supplemental Figure 4). Correlation matrix analysis of the microarray data showed that different biologic samples obtained from the same nephron/CD fractions showed near-identical miRNA expression profiles, further indicating the purity of each fraction (Figure 4A). The complete miRNA expression profile for each nephron segment is shown in Supplemental Table 1. Unsupervised hierarchic clustering of the microarray data revealed that the global miRNA expression profiles segregated into four distinct clusters: GL, PCT/PST, TAL/DCT, and CD (Figure 4B). Expression of ten miRNAs was validated by Q-PCR (miR-143–3p, miR-195a-5p, miR-107–3p, miR-34a-5p, miR-193–3p, miR-378a-5p, miR-874–3p, miR-155–5p, miR-200c-3p, and miR-96–5p). Consistent with the microarray data, Q-PCR analysis also revealed that miRNAs displayed a segmental expression pattern along the nephron and CD (Figure 3C). Next, we identified sixteen high-abundance miRNAs that were expressed at 50% or greater levels in CD compared PCT/PST. Two of these sixteen CD-enriched miRNAs, miR-200c and miR-200a, belong to the miR-200 miRNA family. miR-200c expression was approximately 70-fold enriched in the CD compared with the PTs (Figure 4D). Next, we identified CD-enriched miRNAs that may have contributed to the renal fibrosis phenotype of Dicer-KO mice. CD tubules from P18 Dicer-KO and control mice were microdissected, and the isolated RNA was used for miRNA microarray analysis (Figure 4E, Supplemental Table 3). Of the 19 miRNAs that showed reduced expression in Dicer-KO CDs, three (miR-200c, miR-200a, and miR-429) belonged to the miR-200 family. Intriguingly, miR-200c, which exhibits the highest enrichment in CDs compared with other fractions, was the top downregulated (reduced by approximately 80%) miRNA in Dicer-KO CDs compared with control CDs. Furthermore, the expression of other miR-200 family members, miR-200a and miR-429, were reduced by 50% and 33%, respectively, in Dicer-KO compared with control CDs (Figure 4E). Thus, these results indicate that the miR-200 members are enriched in the CD, and their expression is significantly reduced when Dicer is ablated from the CD.

Figure 4.
Figure 4.

CDs exhibit a distinct miRNA expression profile and show high expression of miR-200 family members. Kidneys from three male and three female wild-type C57 BL/6J mice were microdissected into the following nephron segments: GL, PCT, PST, TAL, DCT, and CD. miRNA microarray analysis was performed using RNA from these segments. (A) Correlation matrix shows the pairwise correlations between all samples. Different biologic samples obtained from the same renal fractions are shown in red dashed boxes. Correlations closer to 1.0 within each of the nephron segment groups indicate high data quality and reproducibility. (B) AGglomerative NESting was used to compute unsupervised agglomerative hierarchic clustering of the nephron segments dataset. The complete dataset was employed, without any additional prior filtering step. Clusters with the shortest average Euclidean distance are combined stepwise. AGglomerative NESting analysis of the miRNA profiles indicates that GL samples form a distinct cluster, whereas PCT and PST samples show no clear separation from each other. Similarly, TAL and DCT samples are not separated from each other but they cluster separately from other groups. The CD group appears as a distinct cluster with some degree of similarity to the TAL/DCT cluster. (C) Consistent with the miRNA microarray data, Q-PCR validation showed that miRNAs exhibit segmental expression pattern. Values are shown as relative expression of the miRNA in various fractions compared with the fraction in which the miRNA is enriched. (D) Sixteen high-abundance miRNAs that were expressed at 50% or greater levels in normal CDs compared with PCTs are shown. miR-200c expression is enriched approximately 68-fold, whereas miR-200a is enriched approximately two-fold in CDs compared with PCT. (E) miRNA microarrays were performed on microdissected CDs from 18-day-old control or Dicer-KO mice. Of the 19 miRNAs that showed reduced expression in Dicer-KO CDs, three (miR-200c, miR-200a, and miR-429) belonged to the miR-200 family. miR-200c was the top downregulated miRNA with an 80% reduction in expression. F, female; M, male; vs, versus; WT, wild-type.

Suppression of miRNA Activity Induces Partial EMT in CDs

The miR-200 miRNA family members are well known inhibitors of EMT. Recent studies have shown that the induction of partial EMT, a process whereby renal tubule epithelial cells undergo dedifferentiation and express mesenchymal markers while still retaining the expression of epithelia-related genes, contributes to the progression of kidney fibrosis by promoting inflammation. Therefore, we tested whether downregulation of miR-200 family members was associated with the induction of EMT or partial EMT in CD cells in vitro and in vivo. Expression of miR-200 was inhibited in a cell line that was derived from mouse inner medullary collecting duct cells (mIMCD3), using locked nucleic acid (LNA)–modified anti-miRs against miR-200 family members. Q-PCR analyses revealed that expression of several direct miR-200 targets that promote EMT (Zeb1, Zeb2, Snai1, and Tgfb2) was upregulated in anti–miR-200–treated compared with scramble-treated mIMCD3 cells (Figure 5A). In addition, expression of key inflammatory signaling cytokines Mcp1, Il6, and Cxcl2 was also elevated in anti–miR-200–treated compared with scramble-treated mIMCD3 cells. However, the expression of two key epithelial-specific genes, Hnf-1β and E-cadherin, did not change in anti–miR-200–treated compared with scramble-treated cells. To determine whether a similar gene expression pattern was also observed in vivo, we analyzed Dicer-KO, Dgcr8-KO, and Ago1–4-KO kidneys. Q-PCR analysis revealed that, similar to miR-200 knockdown cells, expression of miR-200 targets Zeb1, Zeb2, Snai1, Snai2, and Tgfb2 and inflammatory cytokines Mcp1, Il6, and Cxcl2 was increased, whereas the expression of Hnf1-β and E-cadherin remained unchanged in Dicer-KO, Dgcr8-KO, and Ago1–4-KO kidneys, compared with their respective control kidneys (Figure 5, B–D).

Figure 5.
Figure 5.

Inhibition of miR-200 activity evokes a profibrotic and inflammatory response. (A) mIMCD3 cells were treated with 10 nM each of LNA-modified anti–miR-200a, anti–miR-200b, and anti–miR-200c to inhibit activity of miR-200 members. Scrambled anti-miR (30 nM)–treated cells were used as controls. Q-PCR analysis indicated increased expression of miR-200-targets and mesenchymal genes Zeb1, Zeb2, Snai1, Tgfb2, and Tgfbr2, in anti–miR-200–treated cells compared with scramble-treated cells. Expression of inflammatory cytokines Mcp1, Il-6, and Cxcl2 was increased, whereas levels of epithelial markers Hnf1b and Cdh1 (E-cadherin) were unchanged in mIMCD3 cells with reduced miR-200 activity. Q-PCR analysis indicated that expression of miR-200 targets, mesenchymal markers, inflammatory genes, and epithelial markers in kidneys from (B) Dicer-KO (n=3–4), (C) Dgcr8-KO (n=5–6), and (D) Ago1–4-KO (n=5–6) mice were similar to that observed in miR-200 knockdown cells. Error bars indicate SEM. *P<0.05; ns, P>0.05.

To test whether EMT was observed in vivo, we first examined whether mesenchymal markers are expressed in mutant CDs. Kidney sections from control, Dicer-KO, Dgcr8-KO, and Ago1–4-KO mice were costained with mesenchymal markers Vimentin or SNAIL2 and a CD marker (DBA) or a PT marker Lotus Tetragonolobus Agglutinin (LTA). Expression of both Vimentin and SNAIL2 was observed in DBA-positive CDs in kidneys from Dicer-KO, Dgcr8-KO, and Ago1–4-KO mice but not in kidneys from control mice (Supplemental Figure 5). Similar colocalization was not observed in PTs (Supplemental Figure 6). On the basis of the Q-PCR data, we next tested whether the miRNA-depleted CDs expressed both epithelial and mesenchymal markers. Tubules in kidneys of 35-day-old control mice revealed expression of HNF-1β and E-cadherin (Supplemental Figure 7) but no coexpression of E-cadherin and Vimentin or SNAIL2 (Figure 6). In contrast, tubules in kidneys of 35-day-old Dicer-KO, Dgcr8-KO, and Ago1–4-KO kidneys retained HNF-1β and E-cadherin expression (Supplemental Figure 7) and exhibited coexpression of E-cadherin both with Vimentin and SNAIL2 (Figure 6). The number of Vimentin- and SNAIL2-positive CDs increased in aged (150-day-old) Dicer-KO, Dgcr8-KO, and Ago1–4-KO mice (Figure 6, B and D). As a second approach to examine EMT, we introduced R26R-EYFP reporter in control and Dicer-KO mice to perform lineage tracing. We observed that EYFP+ cells in both control and Dicer-KO kidney were surrounded by Entactin, a tubular basement membrane marker (Figure 7A). Moreover, we observed no colocalization of EYFP with FSP-1, indicating that Dicer-KO CDs remain with the tubular compartment and do not directly contribute to the FSP-1+ cells in the interstitium (Figure 7B). Finally, consistent with our earlier findings, we observed persistent expression of HNF-1β and de novo expression of SNAI2 in a subset of EYFP+ tubules in Dicer-KO kidneys (Figure 7, C and D). Collectively, these results indicate that suppression of miRNA activity in renal CDs is associated with the induction of partial EMT, which underlies the kidney fibrosis phenotype observed in Dicer-KO, Dgcr8-KO, and Ago1–4-KO mice (Figure 7E).

Figure 6.
Figure 6.

Suppression of miRNA activity induces partial EMT in CDs. Kidney sections from the indicated mouse models at 5 weeks and 5 months of age were costained with antibodies against E-cadherin (green), an epithelial cell junction marker, and mesenchymal markers Vimentin (red) (A and B) or SNAI2 (red) (C and D). The merged panels show colocalization of both the epithelial and mesenchymal markers within kidney tubules.

Figure 7.
Figure 7.

Lineage tracing revealed that Dicer-KO CDs undergo partial EMT. Kidney sections from 35-day-old Pkhd1/Cre; R26R-EYFP and Dicer-KO; R26R-EYFP mice were costained with antibodies against GFP and (A) Entactin, (B) FSP1, (C) HNF-1β, or (D) SNAI2. EYFP+ tubules in both control and Dicer-KO kidneys were surrounded by the basement membrane marker Entactin and did not colocalize with FSP-1. Thus, Dicer-KO CDs remained within the tubular compartment and do not directly contribute to FSP-1+ interstitial cells. HNF-1β expression was observed in EYFP+ tubules from control and Dicer-KO kidney. De novo expression of SNAI2 was observed in a subset of EYFP+ tubules in Dicer-KO kidneys, indicating that the mutant CDs exhibit both epithelial and mesenchymal markers suggesting partial EMT. (E) The proposed mechanism is shown.

Discussion

Our study has provided significant new insights into the physiologic function of miRNAs in renal tubules. First, we show that miRNAs are essential for CD homeostasis. miRNAs regulate multiple aspects of embryonic renal tubule development33 and they have emerged as drug targets for genetic kidney diseases such as autosomal dominant polycystic kidney disease.34–38 However, the physiologic function of miRNAs in postnatal renal tubules is still unknown. A recent study found that inactivation of Dicer in postnatal PTs does not affect kidney histology or function, suggesting that miRNAs are dispensable for PT homeostasis.24 In contrast, we found that CD-specific inactivation of Dicer spontaneously evoked a renal tubule injury–like response marked by increased levels of injury markers Ngal and Kim1. Over time, this process culminated into kidney failure due to TIF and inflammation. Although conditional Dicer inactivation is a commonly-used approach to investigate the effects of inhibiting miRNA activity, recent studies have shown that DICER1 is a multifaceted enzyme with several other miRNA-independent functions.25–31 Therefore, in addition to Dicer inactivation, we also disrupted the upstream microprocessor and the downstream miRISC. CD-specific disruption of both the microprocessor complex and miRISC recapitulated the kidney fibrosis phenotype of Dicer-KO mice, thereby conclusively proving that miRNAs are necessary for CD homeostasis.

The second major insight of our work is that miRNAs aid in CD homeostasis by inhibiting the expression of mRNAs that activate EMT, profibrotic, and proinflammatory pathways. To identify miRNAs that underlie the Dicer-KO phenotype, we began by profiling global miRNA expression in microdissected renal tubules. Solely on the basis of miRNA expression profiles, renal tubules can be faithfully divided into three distinct groups (PCT/PST, TAL/DCT, and CD). This observation suggests that kidney miRNAs have unique, segment-specific functions.

We identified miR-200 family members, in particular miR-200c, as CD-enriched miRNAs that may underlie the fibrotic phenotype of our mouse models. A very well described function of the miR-200 family is to repress genes that induce EMT and fibrosis. In keeping with these observations, we found that expression of key activators of EMT and fibrosis (Zeb1, Zeb2, Snai1, Tgfb2, Tgfbr2) was upregulated in Dicer, Dgcr8, and Ago1–4-KO kidneys as well as miR-200 knockdown cultured CD cells. Moreover, we observed de novo expression of mesenchymal markers in CDs of Dicer-KO, Dgcr8-KO, and Ago1–4-KO mice. However, considering that the mutant CD cells remained within the tubular compartment and continued to express epithelial markers, suppression of miRNA activity induced only partial EMT in CDs. Rather than directly generating myofibroblasts, as would be expected if the CDs underwent EMT, the partial EMT CD cells could still promote fibrosis by secreting proinflammatory cytokines. Indeed, expression of proinflammatory cytokines was increased in Dicer, Dgcr8, and Ago1–4-KO kidneys and miR-200 knockdown CD cells. Whether EMT contributes to kidney fibrosis is controversial. However, two recent studies showed that partial EMT is not only observed in renal tubules but is necessary and sufficient to promote inflammation and TIF in mouse models of kidney injury.39–41

Finally, our work suggests that dysregulated miRNA expression in CDs might be a new mechanism for the initiation and progression of TIF. Several previous studies have highlighted the fibrogenic potential of CD injury. For example, when exposed to fibrogenic stimuli, CD cells readily undergo EMT.42,43 Similarly, CDs exhibit partial EMT in the nonhuman primate model of fetal urinary tract obstruction.44 In murine unilateral urinary tract obstruction models, CDs have been implicated in promoting interstitial inflammation and inhibiting the profibrotic TGF-β signaling.45,46 However, it is far less clear if CDs can contribute to TIF when the initial site of injury is not within CDs. miR-200c and miR-200a are downregulated in patients with IgA and diabetic nephropathy, respectively47—both diseases that initially affect the GL. miR-200c is also downregulated in mouse models of ischemic kidney injury, which primarily affect the PTs.48,49 Considering that these miRNAs are enriched in CDs and are known to be antifibrotic, the pathogenic effect of their downregulation is likely to be mediated through CDs. Hypoxia signaling, a major pathogenic pathway in TIF, represses Dicer in the kidney.48 Because Dicer inactivation in CDs, but not in PTs, recapitulates the phenotype of kidney hypoxia, the pathogenic consequence of global miRNA downregulation in hypoxia models is also likely to be mediated via CDs.

In summary, we have used independent but complementary approaches to examine the physiologic function of miRNAs in CDs. We found that suppression of miRNA activity through CD-specific deletion of key miRNA pathway enzymes results in TIF and renal failure resembling CKD. Global expression profiling revealed marked differences in miRNA expression patterns between the PT and CDs. We identified miR-200 family members as CD-enriched miRNAs and show that downregulation of these miRNAs is associated with partial EMT in the CDs. Thus, our results indicate that miRNAs are essential for CD function. Downregulation of miRNAs and subsequent induction of partial EMT in CDs may be a new general mechanism for TIF progression. Moreover, our work shines a new light into the crucial function of CDs in initiation and progression of TIF.

Concise Methods

Mice

Kidney-specific Dicer mutant mice were produced by breeding Pkhd1/Cre mice50 with DicerF/F mice.51 The Dgcr8F/F (MMRRC stock number: 32051), Ago2F/F (JAX stock number: 022558), and Ago3/1/4def (referred to as Ago1/3/4-KO mice in the manuscript; JAX stock number: 014152) mice were obtained from Jackson Laboratories. The R26R-EYFP mice were provided by Dr. Frank Costantini (Columbia University). The generation of the following mouse lines is discussed in the Results section: Pkhd1/Cre;Dgcr8F/F, Pkhd1/Cre;Ago2F/F, and Ago1/3/4-KO;Pkhd1/Cre;Ago2F/F (Ago1–4-KO). All mice were maintained on a B6 genetic background. All experiments involving animals were conducted under the auspices of the University of Texas Southwestern Medical Center at Dallas and University of Zurich Animal Care and Use Committee.

Tissue Harvesting and Analysis

Mice were anesthetized under approved protocols, blood was obtained by cardiac puncture, and the right kidney was flash frozen for molecular analysis. The left kidney was perfused with cold PBS and 4% (wt/vol) paraformaldehyde and then harvested. Kidneys were fixed with 4% paraformaldehyde for 2 hours, and then embedded in paraffin for sectioning. Sagittal sections of kidneys were stained with H&E or PicroSirius Red for additional analysis.

Renal Function Tests

Serum creatinine was measured using capillary electrophoresis and BUN was measured using the Vitros 250 Analyzer.

Immunofluorescence Staining

The following antibodies and dilutions were used on paraffin-embedded or frozen sections for immunofluorescence staining: anti-GFP (1:400, GFP-1020; Aves), anti-F4/80 (1:50, ab6640; Abcam), anti–α-smooth muscle (1:400, C6198; Sigma-Aldrich), anti-S100A4 (FSP-1) (1:200, ab41532; Abcam), anti-Vimentin (1:100, D21H3; Cell Signaling), anti-SNAIL2 (1:400, C19G7; Cell Signaling), anti-AQP2 (1:400, A7310; Sigma-Aldrich), anti–E-cadherin (1:200, 13–1900; ThermoFisher Scientific), anti–HNF1-β (1:400, sc-22840; SantaCruz Biotechnology). Secondary antibodies were conjugated to Alexa Fluor 488 or Alexa Fluor 594 (1:400; Molecular Probes). Lectins used were Lotus Tetragonolobus Agglutinin (1:300, FL-1321; Vector Laboratories) and DBA (1:200, FL-1031; Vector Laboratories).

Western Blot Analysis

Total protein was extracted from Dgcr8-KO and control kidneys. Twenty micrograms of protein was loaded on a 4%–15% SDS-polyacrylamide gel, and the proteins were transferred to a nitrocellulose membrane. The membrane was blocked with 5% milk solution and probed overnight at 4°C with a primary rabbit anti-DGCR8 antibody (1:1000, Abcam 36865). Goat–anti-rabbit HRP-conjugated IgG was used as a secondary antibody, and the blot was developed using the SuperSignal West Dura Extended Duration substrate from Pierce. The protein bands were quantified using Quantity One imaging software from Bio-Rad.

Cell Culture

Mouse inner medullary (mIMCD3) cells were grown in DMEM low-glucose medium supplemented with 10% FBS and maintained in 5% CO2 atmosphere at 37°C. mIMCD3 cells were plated in six-well plates (2×105 cells/well) and transfected with 10 nM each of anti–miR-200a, anti–miR-200b, and anti–miR-200c (Exiqon) with Lipofectamine 2000 (Life Technologies). Thirty nanomolar of scrambled anti-miR was used as controls. The cells were harvested 48 hours after transfection and lysed in Qiazol. The sequences of the miR-200 anti-miR and the scramble anti-miRs have been previously published.33

RNA Isolation and Quantitative RT-PCR

Total RNA was isolated from cultured cells or mouse kidneys using the miRNeasy Mini kit (Qiagen). First-strand cDNA was synthesized using the SuperScript III Reverse transcription from Life Technologies. Quantitative real-time PCR was performed by loading samples in triplicates on a CFX Connect Real-Time PCR Detection System. We used 18S RNA to normalize gene expression for mRNA, and the data were analyzed using the Bio-Rad CFX software.

Microdissection of Kidney Tubules

Kidney tubules were isolated as previously described.52 Kidney slices were placed into a prewarmed collagenase type I (1.5 mg/ml dissolved in DMEM/F12; Worthington, Lakewood, NJ) solution in a 15-ml tube, with vigorous shaking at 37°C for 10–15 minutes. After digestion, the individual nephron segments were dissected in 4°C Hank’s solution and transferred by adhering the tubules to small glass beads (0.5-mm diameter; Thomas Scientific, Swedesboro, NJ) and then transferring the beads to 1.5-ml tubes containing Qiazol to isolate total RNA.

miRNA Microarrays

Total RNA was isolated from microdissected nephron segments from wild-type kidneys and analyzed using the SurePrint mouse miRNA microarrays (Agilent Technologies). To examine the miRNA expression patterns, total RNA was isolated from microdissected CDs from Dicer-KO and control mice, with miRNeasy kits, and miRNA microarrays analysis was performed by LC Sciences (Houston, TX) as previously described.33

miRNA Isolation and Quantitative Real-Time PCR

The validity of the miRNA microarray results was confirmed by reverse transcription quantitative real-time PCR analysis for two distinct miRNAs (see Supplemental Table 4 for primers) for each of the five nephron segments microdissected from wild-type kidneys. The analysis was obtained in five independent samples, each pooled from six wild-type kidneys. Total RNA was reverse transcribed using miScript II RT kit (QIAGEN). The cDNA (diluted 10×) was used for a 40-cycle Q-PCR reaction with miScript specific Primers (forward primer) and miScript Universal Primers ([reverse primer] sequences proprietary of QIAGEN) with miScript SYBR green PCR KIT (QIAGEN). All reactions were performed on CFX96 Real-Time PCR Detection System (Bio-Rad Laboratories, Hercules, CA). Two endogenous small RNA control targets (RNU6–6P and SNORD61: sequences proprietary of QIAGEN) were used to normalize and the expression values of the respective miRNA were compared using the ∆∆CT method.

Statistical Analyses

Data are shown as mean±SEM. Statistical analysis was performed using t test. ANOVA followed by Dunnett’s post hoc test was used for multiple comparisons. Log-rank (Mantel–Cox) test was used to compare differences in the survival analyses. P<0.05 was considered statistically significant.

Disclosures

None.

Acknowledgments

We thank the University of Texas Southwestern O’Brien Kidney Research Core Center for providing critical reagents.

The work from the authors’ laboratory is supported by National Institutes of Health grants R01DK102572 (to V.P.). These studies were also supported in part by the European Community’s Seventh Framework Programme (FP7/2007-2013) under grant agreement no. 246539 (Marie Curie) and grant no. 305608 (EURenOmics), the Swiss National Science Foundation (31003A_169850), and the Rare Disease Initiative Zurich (radiz), a clinical research priority program of the University of Zurich, Switzerland.

Footnotes

References

  1. ↵
    1. Zarjou A,
    2. Yang S,
    3. Abraham E,
    4. Agarwal A,
    5. Liu G
    : Identification of a microRNA signature in renal fibrosis: Role of miR-21. Am J Physiol Renal Physiol 301: F793–F801, 2011pmid:21775484
    OpenUrlCrossRefPubMed
    1. Wang B,
    2. Koh P,
    3. Winbanks C,
    4. Coughlan MT,
    5. McClelland A,
    6. Watson A,
    7. Jandeleit-Dahm K,
    8. Burns WC,
    9. Thomas MC,
    10. Cooper ME,
    11. Kantharidis P
    : miR-200a prevents renal fibrogenesis through repression of TGF-β2 expression. Diabetes 60: 280–287, 2011pmid:20952520
    OpenUrlAbstract/FREE Full Text
    1. Xiong M,
    2. Jiang L,
    3. Zhou Y,
    4. Qiu W,
    5. Fang L,
    6. Tan R,
    7. Wen P,
    8. Yang J
    : The miR-200 family regulates TGF-β1-induced renal tubular epithelial to mesenchymal transition through Smad pathway by targeting ZEB1 and ZEB2 expression. Am J Physiol Renal Physiol 302: F369–F379, 2012pmid:22012804
    OpenUrlCrossRefPubMed
    1. Hajarnis SS,
    2. Patel V,
    3. Aboudehen K,
    4. Attanasio M,
    5. Cobo-Stark P,
    6. Pontoglio M,
    7. Igarashi P
    : Transcription factor Hepatocyte Nuclear Factor-1β (HNF-1β) regulates microRNA-200 expression through a long noncoding RNA. J Biol Chem 290: 24793–24805, 2015pmid:26292219
    OpenUrlAbstract/FREE Full Text
  5. ↵
    1. Chung AC,
    2. Huang XR,
    3. Meng X,
    4. Lan HY
    : miR-192 mediates TGF-beta/Smad3-driven renal fibrosis. J Am Soc Nephrol 21: 1317–1325, 2010pmid:20488955
    OpenUrlAbstract/FREE Full Text
  6. ↵
    1. Lee Y,
    2. Jeon K,
    3. Lee JT,
    4. Kim S,
    5. Kim VN
    : MicroRNA maturation: Stepwise processing and subcellular localization. EMBO J 21: 4663–4670, 2002pmid:12198168
    OpenUrlAbstract/FREE Full Text
  7. ↵
    1. Ha M,
    2. Kim VN
    : Regulation of microRNA biogenesis. Nat Rev Mol Cell Biol 15: 509–524, 2014pmid:25027649
    OpenUrlCrossRefPubMed
  8. ↵
    1. Lee Y,
    2. Kim M,
    3. Han J,
    4. Yeom KH,
    5. Lee S,
    6. Baek SH,
    7. Kim VN
    : MicroRNA genes are transcribed by RNA polymerase II. EMBO J 23: 4051–4060, 2004pmid:15372072
    OpenUrlAbstract/FREE Full Text
  9. ↵
    1. Gregory RI,
    2. Yan KP,
    3. Amuthan G,
    4. Chendrimada T,
    5. Doratotaj B,
    6. Cooch N,
    7. Shiekhattar R
    : The microprocessor complex mediates the genesis of microRNAs. Nature 432: 235–240, 2004pmid:15531877
    OpenUrlCrossRefPubMed
  10. ↵
    1. Denli AM,
    2. Tops BB,
    3. Plasterk RH,
    4. Ketting RF,
    5. Hannon GJ
    : Processing of primary microRNAs by the microprocessor complex. Nature 432: 231–235, 2004pmid:15531879
    OpenUrlCrossRefPubMed
    1. Han J,
    2. Lee Y,
    3. Yeom KH,
    4. Kim YK,
    5. Jin H,
    6. Kim VN
    : The Drosha-DGCR8 complex in primary microRNA processing. Genes Dev 18: 3016–3027, 2004pmid:15574589
    OpenUrlAbstract/FREE Full Text
    1. Lee Y,
    2. Ahn C,
    3. Han J,
    4. Choi H,
    5. Kim J,
    6. Yim J,
    7. Lee J,
    8. Provost P,
    9. Rådmark O,
    10. Kim S,
    11. Kim VN
    : The nuclear RNase III Drosha initiates microRNA processing. Nature 425: 415–419, 2003pmid:14508493
    OpenUrlCrossRefPubMed
  13. ↵
    1. Landthaler M,
    2. Yalcin A,
    3. Tuschl T
    : The human DiGeorge syndrome critical region gene 8 and Its D. melanogaster homolog are required for miRNA biogenesis. Curr Biol 14: 2162–2167, 2004pmid:15589161
    OpenUrlCrossRefPubMed
  14. ↵
    1. Bohnsack MT,
    2. Czaplinski K,
    3. Gorlich D
    : Exportin 5 is a RanGTP-dependent dsRNA-binding protein that mediates nuclear export of pre-miRNAs. RNA 10: 185–191, 2004pmid:14730017
    OpenUrlAbstract/FREE Full Text
    1. Lund E,
    2. Güttinger S,
    3. Calado A,
    4. Dahlberg JE,
    5. Kutay U
    : Nuclear export of microRNA precursors. Science 303: 95–98, 2004pmid:14631048
    OpenUrlAbstract/FREE Full Text
  16. ↵
    1. Yi R,
    2. Qin Y,
    3. Macara IG,
    4. Cullen BR
    : Exportin-5 mediates the nuclear export of pre-microRNAs and short hairpin RNAs. Genes Dev 17: 3011–3016, 2003pmid:14681208
    OpenUrlAbstract/FREE Full Text
  17. ↵
    1. Lee HY,
    2. Doudna JA
    : TRBP alters human precursor microRNA processing in vitro. RNA 18: 2012–2019, 2012pmid:23006623
    OpenUrlAbstract/FREE Full Text
    1. Chendrimada TP,
    2. Gregory RI,
    3. Kumaraswamy E,
    4. Norman J,
    5. Cooch N,
    6. Nishikura K,
    7. Shiekhattar R
    : TRBP recruits the Dicer complex to Ago2 for microRNA processing and gene silencing. Nature 436: 740–744, 2005pmid:15973356
    OpenUrlCrossRefPubMed
    1. Tomari Y,
    2. Zamore PD
    : Perspective: Machines for RNAi. Genes Dev 19: 517–529, 2005pmid:15741316
    OpenUrlAbstract/FREE Full Text
  20. ↵
    1. Meister G,
    2. Tuschl T
    : Mechanisms of gene silencing by double-stranded RNA. Nature 431: 343–349, 2004pmid:15372041
    OpenUrlCrossRefPubMed
  21. ↵
    1. Liu J,
    2. Valencia-Sanchez MA,
    3. Hannon GJ,
    4. Parker R
    : MicroRNA-dependent localization of targeted mRNAs to mammalian P-bodies. Nat Cell Biol 7: 719–723, 2005pmid:15937477
    OpenUrlCrossRefPubMed
    1. Gregory RI,
    2. Chendrimada TP,
    3. Cooch N,
    4. Shiekhattar R
    : Human RISC couples microRNA biogenesis and posttranscriptional gene silencing. Cell 123: 631–640, 2005pmid:16271387
    OpenUrlCrossRefPubMed
  23. ↵
    1. Eichhorn SW,
    2. Guo H,
    3. McGeary SE,
    4. Rodriguez-Mias RA,
    5. Shin C,
    6. Baek D,
    7. Hsu SH,
    8. Ghoshal K,
    9. Villén J,
    10. Bartel DP
    : mRNA destabilization is the dominant effect of mammalian microRNAs by the time substantial repression ensues. Mol Cell 56: 104–115, 2014pmid:25263593
    OpenUrlCrossRefPubMed
  24. ↵
    1. Wei Q,
    2. Bhatt K,
    3. He HZ,
    4. Mi QS,
    5. Haase VH,
    6. Dong Z
    : Targeted deletion of Dicer from proximal tubules protects against renal ischemia-reperfusion injury. J Am Soc Nephrol 21: 756–761, 2010pmid:20360310
    OpenUrlAbstract/FREE Full Text
  25. ↵
    1. Nakagawa A,
    2. Shi Y,
    3. Kage-Nakadai E,
    4. Mitani S,
    5. Xue D
    : Caspase-dependent conversion of Dicer ribonuclease into a death-promoting deoxyribonuclease. Science 328: 327–334, 2010pmid:20223951
    OpenUrlAbstract/FREE Full Text
    1. Kaneko H,
    2. Dridi S,
    3. Tarallo V,
    4. Gelfand BD,
    5. Fowler BJ,
    6. Cho WG,
    7. Kleinman ME,
    8. Ponicsan SL,
    9. Hauswirth WW,
    10. Chiodo VA,
    11. Karikó K,
    12. Yoo JW,
    13. Lee DK,
    14. Hadziahmetovic M,
    15. Song Y,
    16. Misra S,
    17. Chaudhuri G,
    18. Buaas FW,
    19. Braun RE,
    20. Hinton DR,
    21. Zhang Q,
    22. Grossniklaus HE,
    23. Provis JM,
    24. Madigan MC,
    25. Milam AH,
    26. Justice NL,
    27. Albuquerque RJ,
    28. Blandford AD,
    29. Bogdanovich S,
    30. Hirano Y,
    31. Witta J,
    32. Fuchs E,
    33. Littman DR,
    34. Ambati BK,
    35. Rudin CM,
    36. Chong MM,
    37. Provost P,
    38. Kugel JF,
    39. Goodrich JA,
    40. Dunaief JL,
    41. Baffi JZ,
    42. Ambati J
    : DICER1 deficit induces Alu RNA toxicity in age-related macular degeneration. Nature 471: 325–330, 2011pmid:21297615
    OpenUrlCrossRefPubMed
    1. Wei W,
    2. Ba Z,
    3. Gao M,
    4. Wu Y,
    5. Ma Y,
    6. Amiard S,
    7. White CI,
    8. Rendtlew Danielsen JM,
    9. Yang YG,
    10. Qi Y
    : A role for small RNAs in DNA double-strand break repair. Cell 149: 101–112, 2012pmid:22445173
    OpenUrlCrossRefPubMed
    1. Francia S,
    2. Michelini F,
    3. Saxena A,
    4. Tang D,
    5. de Hoon M,
    6. Anelli V,
    7. Mione M,
    8. Carninci P,
    9. d’Adda di Fagagna F
    : Site-specific DICER and DROSHA RNA products control the DNA-damage response. Nature 488: 231–235, 2012pmid:22722852
    OpenUrlCrossRefPubMed
    1. Knight SW,
    2. Bass BL
    : A role for the RNase III enzyme DCR-1 in RNA interference and germ line development in Caenorhabditis elegans. Science 293: 2269–2271, 2001pmid:11486053
    OpenUrlAbstract/FREE Full Text
    1. Ketting RF,
    2. Fischer SE,
    3. Bernstein E,
    4. Sijen T,
    5. Hannon GJ,
    6. Plasterk RH
    : Dicer functions in RNA interference and in synthesis of small RNA involved in developmental timing in C. elegans. Genes Dev 15: 2654–2659, 2001pmid:11641272
    OpenUrlAbstract/FREE Full Text
  31. ↵
    1. Grishok A,
    2. Pasquinelli AE,
    3. Conte D,
    4. Li N,
    5. Parrish S,
    6. Ha I,
    7. Baillie DL,
    8. Fire A,
    9. Ruvkun G,
    10. Mello CC
    : Genes and mechanisms related to RNA interference regulate expression of the small temporal RNAs that control C. elegans developmental timing. Cell 106: 23–34, 2001pmid:11461699
    OpenUrlCrossRefPubMed
  32. ↵
    1. Ma M,
    2. Tian X,
    3. Igarashi P,
    4. Pazour GJ,
    5. Somlo S
    : Loss of cilia suppresses cyst growth in genetic models of autosomal dominant polycystic kidney disease. Nat Genet 45: 1004–1012, 2013pmid:23892607
    OpenUrlCrossRefPubMed
  33. ↵
    1. Patel V,
    2. Hajarnis S,
    3. Williams D,
    4. Hunter R,
    5. Huynh D,
    6. Igarashi P
    : MicroRNAs regulate renal tubule maturation through modulation of Pkd1. J Am Soc Nephrol 23: 1941–1948, 2012pmid:23138483
    OpenUrlAbstract/FREE Full Text
  34. ↵
    1. Hajarnis S,
    2. Lakhia R,
    3. Yheskel M,
    4. Williams D,
    5. Sorourian M,
    6. Liu X,
    7. Aboudehen K,
    8. Zhang S,
    9. Kersjes K,
    10. Galasso R,
    11. Li J,
    12. Kaimal V,
    13. Lockton S,
    14. Davis S,
    15. Flaten A,
    16. Johnson JA,
    17. Holland WL,
    18. Kusminski CM,
    19. Scherer PE,
    20. Harris PC,
    21. Trudel M,
    22. Wallace DP,
    23. Igarashi P,
    24. Lee EC,
    25. Androsavich JR,
    26. Patel V
    : microRNA-17 family promotes polycystic kidney disease progression through modulation of mitochondrial metabolism. Nat Commun 8: 14395, 2017pmid:28205547
    OpenUrlCrossRefPubMed
    1. Li X
    1. Hajarnis S,
    2. Lakhia R,
    3. Patel V
    : MicroRNAs and polycystic kidney disease. In: Polycystic Kidney Disease, edited by Li X, Brisbane (AU), Codon Publications, 2015
    1. Lakhia R,
    2. Hajarnis S,
    3. Williams D,
    4. Aboudehen K,
    5. Yheskel M,
    6. Xing C,
    7. Hatley ME,
    8. Torres VE,
    9. Wallace DP,
    10. Patel V
    : MicroRNA-21 aggravates cyst growth in a model of polycystic kidney disease. J Am Soc Nephrol 27: 2319–2330, 2016pmid:26677864
    OpenUrlAbstract/FREE Full Text
    1. Patel V,
    2. Williams D,
    3. Hajarnis S,
    4. Hunter R,
    5. Pontoglio M,
    6. Somlo S,
    7. Igarashi P
    : miR-17~92 miRNA cluster promotes kidney cyst growth in polycystic kidney disease. Proc Natl Acad Sci U S A 110: 10765–10770, 2013pmid:23759744
    OpenUrlAbstract/FREE Full Text
  38. ↵
    1. Yheskel M,
    2. Patel V
    : Therapeutic microRNAs in polycystic kidney disease. Curr Opin Nephrol Hypertens 26: 282–289, 2017pmid:28399020
    OpenUrlPubMed
  39. ↵
    1. Grande MT,
    2. Sánchez-Laorden B,
    3. López-Blau C,
    4. De Frutos CA,
    5. Boutet A,
    6. Arévalo M,
    7. Rowe RG,
    8. Weiss SJ,
    9. López-Novoa JM,
    10. Nieto MA
    : Snail1-induced partial epithelial-to-mesenchymal transition drives renal fibrosis in mice and can be targeted to reverse established disease. Nat Med 21: 989–997, 2015pmid:26236989
    OpenUrlCrossRefPubMed
    1. Nieto MA,
    2. Huang RY,
    3. Jackson RA,
    4. Thiery JP
    : Emt: 2016. Cell 166: 21–45, 2016pmid:27368099
    OpenUrlCrossRefPubMed
  41. ↵
    1. Lovisa S,
    2. LeBleu VS,
    3. Tampe B,
    4. Sugimoto H,
    5. Vadnagara K,
    6. Carstens JL,
    7. Wu CC,
    8. Hagos Y,
    9. Burckhardt BC,
    10. Pentcheva-Hoang T,
    11. Nischal H,
    12. Allison JP,
    13. Zeisberg M,
    14. Kalluri R
    : Epithelial-to-mesenchymal transition induces cell cycle arrest and parenchymal damage in renal fibrosis. Nat Med 21: 998–1009, 2015pmid:26236991
    OpenUrlCrossRefPubMed
  42. ↵
    1. Ivanova L,
    2. Butt MJ,
    3. Matsell DG
    : Mesenchymal transition in kidney collecting duct epithelial cells. Am J Physiol Renal Physiol 294: F1238–F1248, 2008pmid:18322023
    OpenUrlCrossRefPubMed
  43. ↵
    1. Smith JP,
    2. Pozzi A,
    3. Dhawan P,
    4. Singh AB,
    5. Harris RC
    : Soluble HB-EGF induces epithelial-to-mesenchymal transition in inner medullary collecting duct cells by upregulating Snail-2. Am J Physiol Renal Physiol 296: F957–F965, 2009pmid:19244405
    OpenUrlCrossRefPubMed
  44. ↵
    1. Butt MJ,
    2. Tarantal AF,
    3. Jimenez DF,
    4. Matsell DG
    : Collecting duct epithelial-mesenchymal transition in fetal urinary tract obstruction. Kidney Int 72: 936–944, 2007pmid:17667982
    OpenUrlCrossRefPubMed
  45. ↵
    1. Gewin L,
    2. Bulus N,
    3. Mernaugh G,
    4. Moeckel G,
    5. Harris RC,
    6. Moses HL,
    7. Pozzi A,
    8. Zent R
    : TGF-beta receptor deletion in the renal collecting system exacerbates fibrosis. J Am Soc Nephrol 21: 1334–1343, 2010pmid:20576806
    OpenUrlAbstract/FREE Full Text
  46. ↵
    1. Fujiu K,
    2. Manabe I,
    3. Nagai R
    : Renal collecting duct epithelial cells regulate inflammation in tubulointerstitial damage in mice. J Clin Invest 121: 3425–3441, 2011pmid:21821915
    OpenUrlCrossRefPubMed
  47. ↵
    1. Wang G,
    2. Kwan BC,
    3. Lai FM,
    4. Choi PC,
    5. Chow KM,
    6. Li PK,
    7. Szeto CC
    : Intrarenal expression of microRNAs in patients with IgA nephropathy. Lab Invest 90: 98–103, 2010pmid:19901913
    OpenUrlCrossRefPubMed
  48. ↵
    1. Ho JJ,
    2. Metcalf JL,
    3. Yan MS,
    4. Turgeon PJ,
    5. Wang JJ,
    6. Chalsev M,
    7. Petruzziello-Pellegrini TN,
    8. Tsui AK,
    9. He JZ,
    10. Dhamko H,
    11. Man HS,
    12. Robb GB,
    13. Teh BT,
    14. Ohh M,
    15. Marsden PA
    : Functional importance of Dicer protein in the adaptive cellular response to hypoxia. J Biol Chem 287: 29003–29020, 2012pmid:22745131
    OpenUrlAbstract/FREE Full Text
  49. ↵
    1. Godwin JG,
    2. Ge X,
    3. Stephan K,
    4. Jurisch A,
    5. Tullius SG,
    6. Iacomini J
    : Identification of a microRNA signature of renal ischemia reperfusion injury. Proc Natl Acad Sci U S A 107: 14339–14344, 2010pmid:20651252
    OpenUrlAbstract/FREE Full Text
  50. ↵
    1. Williams SS,
    2. Cobo-Stark P,
    3. Hajarnis S,
    4. Aboudehen K,
    5. Shao X,
    6. Richardson JA,
    7. Patel V,
    8. Igarashi P
    : Tissue-specific regulation of the mouse Pkhd1 (ARPKD) gene promoter. Am J Physiol Renal Physiol 307: F356–F368, 2014pmid:24899057
    OpenUrlCrossRefPubMed
  51. ↵
    1. Harfe BD,
    2. McManus MT,
    3. Mansfield JH,
    4. Hornstein E,
    5. Tabin CJ
    : The RNaseIII enzyme Dicer is required for morphogenesis but not patterning of the vertebrate limb. Proc Natl Acad Sci U S A 102: 10898–10903, 2005pmid:16040801
    OpenUrlAbstract/FREE Full Text
  52. ↵
    1. Cheng CJ,
    2. Yoon J,
    3. Baum M,
    4. Huang CL
    : STE20/SPS1-related proline/alanine-rich kinase (SPAK) is critical for sodium reabsorption in isolated, perfused thick ascending limb. Am J Physiol Renal Physiol 308: F437–F443, 2015pmid:25477470
    OpenUrlCrossRefPubMed
View Abstract
Back to top

In this issue

View Selected Citations (0)
Print
Download PDF
Email Article
Citation Tools
Request Permissions
Suppression of microRNA Activity in Kidney Collecting Ducts Induces Partial Loss of Epithelial Phenotype and Renal Fibrosis
Sachin Hajarnis, Matanel Yheskel, Darren Williams, Thomas Brefort, Bob Glaudemans, Huguette Debaix, Michel Baum, Olivier Devuyst, Vishal Patel
JASN Feb 2018, 29 (2) 518-531; DOI: 10.1681/ASN.2017030334
del.icio.us logo Digg logo Reddit logo Twitter logo CiteULike logo Facebook logo Google logo Mendeley logo

Jump to section

Keywords