MicroRNAs in kidney physiology and disease

 Abstract

MicroRNAs (miRNAs) are small non-coding RNA molecules that regulate gene expression. They have important roles during kidney development, homeostasis and disease. In particular, miRNAs participate in the onset and progression of tubulointerstitial sclerosis and end-stage glomerular lesions that occur in various forms of chronic kidney disease (CKD). Therefore, miRNAs represent potential new therapeutic targets for a debilitating disease that continues to increase in prevalence worldwide and for which fully effective therapies are lacking. Several lines of research aimed at improving common CKD diagnostic tools and avoiding invasive kidney biopsies have also identified circulating miRNAs as possible diagnostic and even prognostic biomarkers of kidney disease. This Review discusses current understanding of the function of miRNAs in CKD, focusing on functions specifically involved in the transforming growth factor β1 pathway, which is activated in CKD. miRNAs that, according to available evidence, seem to be involved in diabetic nephropathy, IgA nephropathy, lupus nephritis, polycystic kidney disease and graft rejection, are also discussed.

 Key points

• MicroRNAs (miRNAs) are key players in kidney development and physiology
• Transforming growth factor β1 is a major regulator of kidney fibrosis; its signalling is finely regulated by miRNAs
• miRNAs contribute to both the induction and progression of chronic kidney disease (CKD)
• Current translational research on miRNAs in kidney disease is mainly focused on developing reliable biomarkers for diagnosis and prognosis of CKD and renal transplantation
• miRNAs represent novel therapeutic targets for CKD, but delivery and safety issues must be taken into account before translation into clinical practice

Introduction

Chronic kidney disease (CKD) is an important public health problem that is closely linked to major non-communicable diseases such as diabetes mellitus and hypertension.¹ Less common causes of CKD are hereditary diseases such as polycystic kidney disease (PKD), IgA nephropathy (IgAN) and lupus nephritis, or herbal and environmental toxins. Regardless of the disease aetiology, progression of CKD results in tubulointerstitial and glomerular fibrosis owing to excessive deposition of extracellular matrix (ECM). Other key features of CKD are inflammatory cell infiltration, tubular cell atrophy, mesangial cell hypertrophy and podocyte apoptosis.² All of these pathological events are mainly instigated by the cytokine transforming growth factor β1 (TGF-β1).³

 

Current therapies that target the renin–angiotensin–aldosterone system⁴ are not always effective in halting progression to end-stage renal disease, a condition that requires renal replacement therapy by dialysis or kidney transplantation. Blockade of TGF-β1 signalling using TGF-β1 neutralizing antibodies or specific inhibitors of TGF-β1 receptors might be a promising therapeutic approach for CKD as these strategies have been shown to attenuate renal fibrosis in various animal models.⁵ The success of these approaches in preclinical studies has prompted the search for therapies that interfere with the epigenetic regulation of TGF-β and genes involved in the TGF-β signalling pathway. In this context, non-coding RNA species, including microRNAs (miRNAs), have been widely implicated in the pathogenesis and progression of CKD, particularly in the regulation of TGF-β1-mediated fibrosis. Correction of miRNA expression by in vivo delivery of miRNA mimics or inhibitors has therefore emerged as a promising novel therapeutic strategy for the treatment of CKD. In addition, several lines of current research aimed at improving common CKD diagnostic tools and avoiding invasive kidney biopsies have identified circulating miRNAs as possible diagnostic and even prognostic biomarkers.^{6, 7} This Review focuses on current research investigating the roles of miRNAs in normal kidney physiology and diseases; particular attention is given to the TGF-β1 pathway and its regulation by miRNAs.

Biogenesis and function of miRNAs

miRNAs are epigenetic regulators of gene expression that are able to modulate several cellular processes, from development to disease conditions. The human miRNAome is composed of 1,881 precursors and 2,588 mature miRNAs, which regulate at least 60% of protein-coding genes.⁸ Since 2010, the number of miRNAs included in miRBase has grown by approximately two-thirds owing to the advent of small RNA deep-sequencing techniques.⁹

 

Here, we briefly mention the key points of miRNA biogenesis and function, as these processes are described thoroughly elsewhere.^{10, 11} Transcription of miRNAs occurs from individual or clustered genes (that is, the miR-194–192 cluster), although some miRNAs can be encoded from distinct genomic loci.¹² Genes encoding miRNAs are located in non-coding sequences or in introns of either protein-coding genes (miR-trons) or non-coding RNA.¹³ Intronic miRNAs are usually coordinately expressed with their host gene and most of the time they both affect the same signalling pathway.^{14, 15, 16}

 

RNA polymerase II transcribes miRNAs in the nucleus as long capped and polyadenylated hairpin transcripts, called primary miRNAs (pri-miRNAs). These are processed into smaller ~70 nucleotide stem–looped structures, called precursor miRNAs (pre-miRNAs) by the ribonuclease III-like enzyme Drosha together with the microprocessor complex subunit DGCR8. Pre-miRNAs are then exported to the cytoplasm by exportin-5/GTP-binding nuclear protein Ran, where the ribonuclease Dicer yields 22 nucleotide miRNA duplexes consisting of the guide and passenger strands (miRNA:miRNA*). The guide strands are finally assembled into the RNA-induced silencing complex (RISC) and bind through their 'seed sequence' (nucleotides 2–8) to fully or partially complementary sites within the 3′ untranslated region of target mRNAs. miRNAs rarely bind to the coding regions of mRNA or genomic DNA, including promoter regions.¹⁷ Target recognition by miRNAs leads to mRNA translational repression and/or mRNA deadenylation and decay,^{18, 19} although positive regulation has been described in a few cases (Figure 1).^{20, 21} Beyond the simplistic concept that miRNAs act as repressors of a single transcript, emerging evidence indicates that they are modulators of many hundreds of proteins often involved in related signalling pathways.^{22, 23}

Figure 1: Interplay between TGF-β1/Smad signalling and the miRNA machinery. 

TGF-β1/Smad signalling can modulate miRNA expression by regulating transcription and/or facilitating the processing of pri-miRNA into pre-miRNA in the nucleus. Pre-miRNA is exported to the cytoplasm where it is processed into the final mature miRNA, which is loaded into the RISC. Mature miRNA guides the RISC to silence target mRNAs through mRNA cleavage, translational repression or deadenylation. Rarely, transcriptional activation can occur. Abbreviations: Co-Smad, common mediator Smad; miRNA, microRNA; P, phosphorylation; pre-miRNA, precursor miRNA; pri-miRNA, primary miRNA; RISC, RNA-induced silencing complex; R-Smad, receptor-regulated Smad; SBS, Smad binding site; TGF-β1, transforming growth factor β1; TGFBR1, TGF-β receptor type-1; TGFBR2, TGF-β receptor type-2; UTR, untranslated region.

miRNAs are usually stable with a long half-life, but individual miRNAs can undergo rapid decay in specific cellular contexts in the presence of particular environmental stimuli or cellular factors. Of note, the presence of certain sequences in miRNAs that determine decay rate, together with several ribonucleases that degrade miRNAs, have been identified.²⁴

miRNAs also have a key role as regulators of cellular crosstalk.⁶ They can be actively secreted into the extracellular microenvironment or into body fluids and captured by other cells, thus altering their transcriptional programme. In addition, circulating miRNAs can be derived from apoptotic cells, with the likely purpose of carrying alarm signals from apoptotic cells to other cells.⁶ miRNAs circulate in blood, urine and other body fluids either packaged into microvesicles and/or exosomes, or transported by RNA-binding proteins (Argonaute2 complexes) and lipoproteins, which protect them from degradation by ribonucleases.⁶ The consequent stability of miRNAs in biological fluids, together with the fact that miRNAs can be repeatedly collected by non-invasive means and detected with high accuracy and specificity by amplification methods, renders miRNAs potentially better biomarkers than proteins and mRNAs. However, isolation and quantification of miRNAs is time-consuming and expensive, and further optimization of these procedures is required before they can be used in routine clinical practice.

miRNAs in development and physiology

Analyses of miRNA expression profiles have identified a set of miRNAs expressed mainly in the adult human kidney (including miR-215, miR-146a and miR-886); other miRNAs, such as miR-192, miR-194, miR-21, miR-200a, miR-204 and let-7a–g, are enriched in the kidney as well as in other organs.^{14, 25, 26} However, some discordant findings regarding miRNA expression profiles have been reported, probably owing to use of different detection methods and to differences in the intrinsic nature of the analysed samples. The picture is further complicated by the fact that expression levels of selected miRNAs differ between fetal and adult kidneys, suggesting that besides being tissue-specific, miRNA expression is specific to developmental stage.²⁷

The generation of mouse models with conditional knockdown of miRNA-processing enzymes or knockdown of specific miRNAs in various cell lineages has enabled the roles of miRNAs during kidney development and homeostasis to be studied. In nephron progenitor cells, conditional deletion of Dicer results in their apoptosis—mediated by the proapoptotic protein Bim—and in the premature termination of nephrogenesis.^{28, 29} Ablation of the miR-17~92 cluster, composed of miRNAs normally expressed in the developing kidney with well-known roles in development, leads to defective proliferation of progenitor cells and reduced numbers of developing nephrons.³⁰ Mice deficient in miR-17~92 are characterized by renal hypodysplasia and develop glomerular dysfunction and proteinuria.³⁰

In ureteric buds, impaired miRNA processing leads to excessive cell proliferation and apoptosis accompanied by disrupted ciliogenesis within the epithelium of ureteric buds and consequent development of renal cysts.^{29, 31} In addition, one study suggested that miRNAs modulate later stages of renal tubule maturation as removal of Dicer from maturing renal tubules and collecting ducts leads to the formation of tubular and glomerular cysts in mice through a mechanism that probably involves modulation of Pkd1 gene dosage by miRNAs.³²

miRNAs also have a key role in podocyte homeostasis. Indeed, podocyte-specific deletion of Dicer or Drosha in the developing kidney causes disruption of the glomerular filtration barrier, leading to proteinuria and collapsing glomerulopathy with glomerular and tubulointerstitial fibrosis.^{33, 34, 35, 36, 37} Moreover, miRNA processing is important for maintenance of adult podocyte function and differentiation as selective deletion of Drosha in adult mice results in collapsing glomerulopathy,³⁷ whereas inactivation of Dicer in postnatal proximal tubules does not affect their histology or function and protects mice from renal ischaemia–reperfusion injury.³⁸

miRNAs are also involved in regulating renal physiology, from the control of blood pressure to the maintenance of whole-body fluid and electrolyte homeostasis. The role of miRNAs in blood pressure control was ascertained by generation of mice with specific inactivation of Dicer in renin-expressing juxtaglomerular cells, which leads to loss of these cells accompanied by reduced plasma renin concentration, hypotension and kidney fibrosis.^{39, 40} Several miRNAs are modulated by hypertonic conditions and have a key role in the control of osmolarity balance by regulating Na⁺ and K⁺ levels in the different portions of the nephron. miR-192 suppresses the Na⁺/K⁺-ATPase β1 subunit gene in human renal epithelial cells,⁴¹ and in the distal nephron it inhibits the serine/threonine kinase WNK1, a protein that is essential for the coordinated regulation of electrolyte transport in the kidney.⁴² Regulation of Na⁺ transport by miRNAs has been ascertained in the mouse cortical collecting duct where aldosterone reduces the expression of a subset of miRNAs that target ankyrin-3. In turn, overexpression of ankyrin-3 increases sodium transport mediated by the epithelial Na channel.⁴³ Moreover, renal medullary cortical collecting duct cells exposed to high Na⁺ concentration are characterized by reduced levels of miR-200 and miR-717 in association with increased levels of their common target NFAT5,⁴⁴ which controls the cellular response to osmotic stress. miRNAs are also able to regulate K⁺ secretion. Indeed, in the cortical collecting duct, high dietary K⁺ intake stimulates the transcription of miR-802 and miR-194, which regulate ROMK channel activity by targeting caveolin-1 (a negative regulator of the ROMK channel)⁴⁵ and intersectin-1 (involved in mediating the WNK-induced endocytosis of the ROMK channel),⁴⁶ respectively. Finally, in the thick ascending limb of Henle, miR-9 and miR-374 regulate the expression of claudin-14, which is critical for Ca²⁺ reabsorption in the kidney.⁴⁷ Interestingly, the transcriptional levels of both miRNAs are directly regulated by the Ca²⁺-sensing receptor CASR.⁴⁸

Figure 2: miRNA-regulatory networks in proximal tubular epithelial cells and mesangial cells in response to TGF-β1, and in podocytes in response to TGF-β1 and high glucose.

Up and down black arrows represent upregulated and downregulated, respectively. miRNA-dependent signalling loops through which TGF-β1 amplifies its signal are shown in green. Abbreviations: ECM, extracellular matrix; HDAC, histone deacetylase; miRNA, microRNA; TGF-β1, transforming growth factor β1; TGFBR1, TGF-β receptor type-1.

miRNAs in immune diseases

IgA nephropathy (IgAN) is characterized by aberrant glycosylation of IgA1, a reaction catalysed in part by C1GALT1, which has been proposed to be a target of miR-148b. miR-148b is upregulated in peripheral blood mononuclear cells (PBMCs) isolated from patients with IgAN, and patients carrying the 1365G/G genotype in the miR-148b-binding site of C1GALT1 have increased C1GALT1 binding affinity for miR-148b and consequently lower enzyme expression (Table 1).¹⁰⁵

Global miRNA expression analysis in kidney biopsy samples from patients with IgAN found that dysregulated levels of miRNAs related to fibrosis (downregulation of miR-200c and upregulation of miR-192, miR-141 and miR-205)¹⁰⁶ and to the immune response (upregulation of miR-155 and miR-146a)¹⁰⁷ correlated with disease severity and progression. Moreover, glomerular endothelial cells of patients with IgAN are characterized by reduced expression of miR-223, which causes glomerular endothelial proliferation, a pathological hallmark of IgAN.¹⁰⁸ Low levels of miR-223 were also found in circulating endothelial cells, providing a possible non-invasive method for evaluating the severity of IgAN (Table 1). Other potential biomarkers of IgAN are miR-155,¹⁰⁹ miR-146a,¹⁰⁹ miR-17¹⁰⁴ and miR-93—which are found at increased levels—and miR-29b and miR-29c—which are found at decreased levels—in the urine of patients with IgAN.¹⁰⁹

miRNA dysregulation also has a key role in lupus nephritis, an autoimmune disorder with a complex pathophysiology. miRNA expression analysis in kidney biopsy samples of patients with class II lupus nephritis (characterized by pure mesangial involvement), identified 36 upregulated and 30 downregulated miRNAs compared with healthy controls.¹¹⁰ miR-150 has also been demonstrated to promote renal fibrogenesis in lupus nephritis (Table 1).^{111, 112} miR-150 targets SOCS1, a negative regulator of the JAK/STAT signalling pathway, which regulates a wide range of genes involved in cell proliferation, inflammation and fibrosis. In human mesangial cells and pTECs, TGF-β1 upregulates miR-150, resulting in decreased expression of SOCS1 and increased production of profibrotic proteins. SOCS1-knockout mice spontaneously develop a lupus-like disease; likewise, the kidneys and in particular the tubular cells of patients with fibrosing lupus nephritis are characterized by increased expression of miR-150 in association with lower levels of SOCS1.¹¹¹

In patients with lupus nephritis of differing severities, a correlation was identified between clinical disease severity and glomerular and tubulointerstitial expression of miR-638, miR-198 and miR-146.¹¹³ Analysis of miRNA expression profiles in plasma,¹¹⁴ urinary sediment¹¹⁵ and PBMCs¹¹⁶ revealed a subset of miRNAs associated with lupus nephritis (miR-342-3p, miR-223 and miR-20a in plasma, miR-221 and miR-222 in urinary sediment, and miR-371-5p, miR-1224-3p and miR-423-5p in PBMCs), which are potential promising disease biomarkers.¹¹⁷

miRNAs in kidney transplantation

miRNA profiling could be a promising tool for monitoring the status of transplanted kidneys, which despite advances in immunosuppressive therapy can undergo acute rejection (an event that is becoming increasingly less common) or chronic rejection. Two research groups have identified miRNAs that were differentially expressed in biopsy samples from the kidneys of patients with acute rejection, although no overlapping miRNAs were found.^{127, 128} In particular, Anglicheau and colleagues demonstrated that increased intragraft levels of miR-142-5p, miR-155 and miR-223, probably derived by graft-invading immune cells, and decreased levels of miR-10b, miR-30a-3p and let-7c, expressed by resident renal cells, were predictive of renal graft function.¹²⁸ One study identified miRNA signatures in kidney biopsy samples that were able to distinguish between acute cellular and humoral rejection and delayed graft function.¹²⁹ Of note, Lorenzen et al.¹³⁰ analysed miRNA expression in the urine of patients with acute rejection and proposed miR-210 as a novel biomarker of acute rejection and predictor of long-term graft function.

 

Interestingly, analysis of a blood-derived miRNA signature of drug-free tolerant patients after kidney transplantation identified eight modulated miRNAs. Among these miRNAs, overexpression of miR-142-3p in PBMCs was correlated with drug-free tolerance, probably owing to the negative regulation of TGF-β signalling by miR-142-3p. Thus, miR-142-3p expression represents a promising predictor of patients who could reduce or even avoid immunosuppressive therapies.¹³¹

 

Three reports studied miRNA changes during chronic rejection characterized by interstitial fibrosis and tubular atrophy (IF/TA). miR-142-3p, miR-204 and miR-211 were differentially expressed in kidney biopsy samples and urine of patients with chronic rejection characterized by IF/TA compared with patients with normal histology and a functioning allograft.¹³² A subsequent study confirmed miR-142-3p overexpression in biopsy samples with IF/TA, and also reported dysregulation of miR-21, miR-142-5p, the cluster comprising miR-506 on chromosome X, miR-30b and miR-30c in these samples.¹³³ Investigation of miRNA expression in urinary cell pellets from patients diagnosed with chronic rejection with IF/TA compared with those with normal renal function identified 22 differentially expressed miRNAs, mainly associated with inflammation.¹³⁴ More importantly, these researchers identified a subset of miRNAs (miR-200b, miR-375, miR-423-5p, miR-193b and miR-345), which are promising biomarkers for monitoring graft function and anticipating progression to chronic rejection, as they were differentially expressed between the two groups soon after kidney transplantation but before histological injury was evident.¹³⁴

Potential uses of miRNAs for therapy

Mature miRNAs possess distinct features that make them potentially suitable as therapeutic agents, including their short sequence and their high homology across multiple vertebrate species. Manipulation of the activity of specific miRNAs in the kidney can be achieved by in vivo delivery of mimics to restore miRNA levels or inhibitors to block miRNA function. miRNA mimics are double-stranded synthetic oligonucleotides that accomplish the endogenous functions of the miRNA of interest, but following chemical modifications possess increased stability and are efficiently taken up by cells. The most widely adopted strategy so far to block miRNA function is with chemically modified oligonucleotides (anti-miRs) designed against the mature miRNA sequence that are stable in circulation and are cell permeable (2′-O-methyl-group-modified oligonucleotides or locked nucleic acid anti-miRs). In addition to anti-miRs, miRNA inhibition can be achieved by expression of miRNA-target sequences able to capture pathogenic miRNAs (miRNA sponge), short hairpin RNA plasmids to abrogate miRNA expression via RNA interference, or using oligonucleotides complementary either to the 3′ untranslated region of the target mRNA binding site sequence (masking approach) or to the sequence of the miRNA (erasers).

 

Many studies in experimental animal models have focused on the therapeutic potential of miRNAs in CKD and promising results in halting renal fibrosis have been obtained by knocking down miR-21,^{22, 65, 92} miR-29c,⁹⁹ miR-214,²³ miR-433,⁹⁴ and miR-192^{69, 76} or overexpressing miR-29b (Table 2).^{66, 80} Successful kidney transfection was achieved by intraperitoneal, intravenous or subcutaneous injection of either mimics or inhibitors or, more frequently, by intravenous injection of plasmids expressing miRNAs or short-hairpin RNAs.

 

However, many obstacles must be overcome before miRNA-based therapies for CKD can be translated into clinical practice, including delivery methods and safety concerns. Indeed, the target miRNA should be kidney-specific in order to avoid any potential adverse effects in other tissues and organs, and should affect only one target (or targets acting in the same pathway) to avoid effects on unintended templates, as in the case of miR-21, the knockdown of which induces cell death in addition to halting renal fibrosis.⁶⁵ So far, these limitations have been partially overcome either by local administration of miRNA-based drugs or by using vectors containing kidney-specific and inducible promoters.

Conclusions

 

The importance of miRNAs in the kidney field is increasingly recognized as they enable researchers to understand in-depth the pathways that have a role in kidney physiology and disease. They can also provide an explanation for divergent transcriptomic and proteomic data. Moreover, an miRNA-based therapy that either restores or blocks miRNA expression and activity is very attractive, especially now that the first miRNA-targeted drug (miravirsen for the treatment of hepatitis C) has entered a phase II clinical trial.¹³⁵ So far, the potential of miRNAs as an effective antifibrotic therapy has been demonstrated only in experimental models of CKD because many safety concerns must be resolved, from the delivery method to the adverse effects on alternative templates. Instead, a more impending application is the detection and quantification of circulating miRNAs as a novel non-invasive, repeatable method to identify and monitor the degree of disease. However, because most miRNAs are highly pleiotropic and act differently depending on the cell type, a single miRNA is unlikely to be able to diagnose and predict a form of CKD. Instead, a network of correlated miRNAs must be considered.

 

Nature Reviews Nephrology | Review

Piera Trionfini^1, Ariela Benigni^1, Giuseppe Remuzzi^1,

Published online 11 November 2014