MicroRNAs in Injury and Repair

Abstract

Organ damage and resulting pathologies often involve multiple deregulated pathways. MicroRNAs (miRNAs) are short, non-coding RNAs that regulate a multitude of genes at the post-transcriptional level. Since their discovery over two decades ago, miRNAs have been established as key players in the molecular mechanisms of mammalian biology including the maintenance of normal homeostasis and the regulation of disease pathogenesis. In recent years, there has been substantial progress in innovative techniques to measure miRNAs along with advances in targeted delivery of agents modulating their expression. This has expanded the scope of miRNAs from being important mediators of cell signaling to becoming viable quantitative biomarkers and therapeutic targets. Currently, miRNA therapeutics are in clinical trials for multiple disease areas and vast numbers of patents have been filed for miRNAs involved in various pathological states. In this review, we summarize miRNAs involved in organ injury and repair, specifically with regards to organs that are the most susceptible to injury: the liver, heart and kidney. In addition, we review the current state of knowledge on miRNA biology, miRNA biomarkers and nucleotide-based therapeutics designed to target miRNAs to prevent organ injury and promote repair.

INTRODUCTION

MicroRNAs (miRNAs) are short, non-coding RNAs that regulate gene expression at the post-transcriptional level. In recent years, miRNAs have been shown to be integral in regulating not just normal homeostasis, development and physiology but also disease pathogenesis in terms of initiation, progression or resolution of the disease[1, 2]. According to miRBase version 21, there are more than 2500 annotated human miRNAs that are each predicted to regulate many mRNAs, and each mRNA can be targeted by a large number of different miRNAs[1, 3, 4]. Importantly, most miRNAs are evolutionarily conserved, making the translation of findings from preclinical models more likely than some other molecular targets. These fundamental characteristics allow miRNAs to serve as potential biomarkers of disease progression and targets for therapeutic intervention.

MiRNAs are expressed from genomic DNA and processed in multiple steps before becoming ~22 nucleotide mature miRNAs (Figure 1). First, miRNAs are transcribed by RNA polymerase II and can originate from discrete miRNA-coding genes or from within introns of protein-coding genes (deemed “miRtrons”)[5]. Many diverse transcriptional regulators have been shown to contribute to miRNA expression including HIF1α, SMADs, p53, MYC, NF-κB and CREB[6]. After transcription, long stem-loop structures called primary miRNAs (pri-miRNAs) undergo cleavage in the nucleus by an RNase-III-type enzyme, Drosha, to form 60-to-70-nucleotide precursor miRNAs (pre-miRNAs). Alternatively, miRtrons can bypass Drosha and form pre-miRNAs directly through spliceosome-mediated processing[7]. The Ran GTPase Exportin-5 complex then shuttles all pre-miRNAs through nuclear pore complexes into the cytosol. Pre-miRNAs are further processed by Dicer to generate miRNA duplexes. Then, miRNA duplexes will dissociate with the help of Argonaute 2 and passenger strands may degrade while the guide strands can remain active. Mature miRNAs then form the RNA-induced silencing complex (RISC), which include additional proteins necessary for mRNA silencing activity[8]. Mature miRNAs will canonically match their “seed regions” (nucleotides 2 through 8 on the 5′ end of the miRNA) to a complementary region on the 3′ untranslated region (UTR) of a target mRNA[2]. However, evidence has emerged that miRNAs can bind to 5′ UTRs and within the mRNA coding sequence[9, 10]. These interactions promote mRNA degradation or inhibit protein synthesis and are responsible for regulating gene expression[11]. In addition, though most miRNAs act in the cytosol to regulate translation some mature miRNAs can reenter the nucleus to regulate transcription, miRNA expression and RNA processing[12, 13].

Figure 1. MiRNA expression, processing and function are highly organized.

MicroRNA biogenesis begins with transcription of genomic DNA by RNA polymerase II and can be mediated by many common transcription factors. Often, large RNA stem-loop structures called pri-miRNAs are formed directly. Drosha then cleaves pri-miRNAs to form intermediate stem-loop structures called pre-miRNAs. Alternatively, pre-miRNAs can form through an intermediate called a “miRtron” from splicing of mRNA. Exportin 5 (XPO5) then transports pre-miRNAs through nuclear pore complexes (NPC) into the cytosol. In the cytosol, Dicer cleaves pre-miRNAs to form microRNA duplexes. Argonaute 2 (AGO2) then mediates the dissociation of the miRNA duplex to form two mature miRNAs. These two miRNAs are deemed “5p” or “3p” depending on whether they originate from the 5′ or 3′ end of the pre-miRNA, respectively. Depending on the context or specific miRNA, mature miRNAs will either assemble within RNA-induced silencing complexes (RISC), degrade or may reenter the nucleus via Importin 8 (IPO8) to regulate transcription. After RISC assembly, miRNAs can bind to mRNAs with their complementary seed region to the 3′ untranslated region (UTR), 5′ UTR or the open reading frame (ORF). MiRNAs then regulate protein production either by promoting mRNA degradation or by inhibiting translation.

When organs experience injury or disease, affected cells may release specific miRNAs into circulation and miRNA biomarkers can be detected in the blood, urine and various other body fluids[14]. MiRNA biomarkers are remarkably stable due to packaging within microvesicles and/or binding to RISC protein components or high-density lipoproteins[15]. In addition, nucleotide-based therapeutics have been developed that can either mimic or inhibit miRNA expression in vivo. MiRNA therapeutics can resist degradation in systemic circulation due to modifications to the nucleotide backbone or by being packaged within microvesicles before dosing[16–18]. Clinical trials are currently underway for a miR-122 inhibitor to treat hepatitis C virus (HCV)[19], a miR-21 inhibitor to treat Alport nephropathy[20] and a miR-34a mimic to treat liver cancer[21]. These aspects highlight the transformative potential of miRNAs as mechanistic biomarkers and therapeutic targets in translational medicine. In this review, we discuss the role of miRNAs in organ damage and repair with a particular focus on the liver, heart and kidney, which are organs that are most susceptible to injury.

MiRNAs IN LIVER INJURY AND REPAIR

MiRNAs in liver injury

Drug induced liver injury (DILI) accounts for more than half of all cases of acute liver failure in many Western countries and acetaminophen (APAP) overdose is a major cause in the United States[22]. DILI is also a major reason for attrition during drug development and is one of the main causes of withdrawing drugs from the market[23]. Other causes of liver injury include chronic viral hepatitis C (HCV) and non-alcoholic fatty liver disease (NAFLD), which are two of the leading causes of chronic liver disease worldwide[24, 25]. The liver encounters toxic stress due to its anatomical location and first-pass metabolism and biotransformation that results in generation of toxic reactive metabolites. Current methods of detecting liver damage include measuring aminotransferases, such as alanine aminotransferase (ALT), and total bilirubin in the blood. However, these methods lack sensitivity and specificity, especially with regards to early detection and prognoses of liver disease, and other factors can affect the quantities of these markers in the blood[26]. Therefore, miRNAs have been explored as biomarkers and as mediators in the mechanisms of pathogenesis for a number of these liver ailments.

MiR-122

MiR-122 is the most abundant miRNA in the liver and has been implicated in basic liver physiology, disease and regeneration[27]. MiR-122 has been shown to be expressed up to 100,000-times higher in the normal liver compared to other tissues such as brain, heart, kidney and lung[28]. MiR-122 knockout mice were shown to develop steatohepatitis, fibrosis and hepatocellular carcinoma, which indicates the protective role of miR-122 in pathogenesis of liver disease [29, 30]. Furthermore, miR-122 has emerged as a promising biomarker of liver damage and can be measured in blood (Figure 2). MiR-122 has been demonstrated to be more sensitive and with a greater dynamic range than ALT in both animals and humans. In mice subjected to APAP-induced liver injury, miR-122 was decreased in liver tissue but highly increased in the plasma in a dose-dependent manner[31]. In humans with APAP-induced hepatotoxicity, miR-122 was significantly increased in the sera compared to healthy controls and has shown its utility to diagnose APAP toxicity in at least one clinical case[32, 33].

Figure 2. MiR-122 is a biomarker and therapeutic target in liver disease.

MiR-122 is the most abundant miRNA in the liver and can increase or decrease in the liver tissue depending on the injury or disease. In diverse forms of liver injury and disease, miR-122 is secreted into the blood within small vesicles, dead cells or bound to RISC proteins or high-density lipoproteins. MiR-122 has been successfully utilized as a biomarker of various forms of liver injury and liver diseases such as HCV. Because of its involvement in HCV replication, miR-122 has also been successfully targeted pharmacologically with an inhibitor, Mirvirsen, which is currently in clinical trials.

Importantly, miR-122 upregulation in the blood is responsive to diverse etiologies of DILI. Specifically, miR-122 allowed for early detection of liver injury following exposure alpha-naphthylisothiocyanate[34, 35], methapyriline[35], tri-chlorobromomethane[28], carbon tetrachloride (CCl4)[28] and dioxin[36]. In addition, miR-122 has been used as a diagnostic biomarker in other disease settings including NAFLD[37], alcoholic liver disease[38] and chronic HCV infection[39, 40]. In HCV, miR-122 was found to play a critical role in promoting the viral life cycle and when inhibited interferes with viral replication[19]. Thus, a miR-122 inhibitor, Miravirsen, is currently in clinical trials to treat HCV in human patients[41]. Though many of the targets of miR-122 are currently unknown, miR-122 is clearly a critical miRNA in the liver and is promising to diagnose and treat liver diseases.

MiR-192

MiR-192 is highly enriched in the liver and elevated after injury, but it is also expressed highly in other tissues including small intestine, colon and kidney, which is different from the liver-specific miR-122[42]. Following diverse etiologies of DILI in both animal and in human sera, miR-192 has been shown to track with miR-122 expression; however, miR-192 is generally not as highly expressed as miR-122[31, 32, 43, 44]. In addition, miR-192 and miR-122 were effective serum biomarkers of liver injury in rats that experienced acute liver transplantation rejection[45]. In a recent study, miR-192 was increased in the sera of humans with acute liver failure and in mice subjected to hepatic ischemia-reperfusion injury or liver injury induced by CCl4[46]. In addition, miR-192 in sera was correlated with increasing liver damage. However, miR-192 was downregulated in liver tissues of mice and humans with liver injury. In a mouse hepatoma cell line, miR-192 was shown to target Zeb2, an anti-apoptotic gene, and increased susceptibility to oxidative stress[46].

MiR-125b

MiR-125b-5p was increased in the blood of patients with APAP overdose and HCV[47, 48]. Recently, miR-125b overexpression was shown to prevent liver injury and improve the survival of mice following APAP-induced liver failure[49]. To uncover the role of miR-125b in hepatocyte survival an unbiased screen of 302 miRNA mimics was performed using primary mouse hepatocytes. The cells were transfected with miRNA mimic library, subjected to APAP toxicity and analyzed for improvements in cell viability, where miR-125b was found as potential hit. Remarkably, administration of a hepatocyte-specific pri-miR-125b vector into mice before APAP overdose protected them from liver failure and death by targeting Kelch-like ECH-associated protein 1 (Keap1). Lastly, miR-125b was shown to be cytoprotective in human hepatocytes and was downregulated in biopsies from patients with acute liver failure[49]. This suggests that a miR-125b mimic may be therapeutically beneficial in patients with increased risk of liver failure.

MiR-34a

MiR-34a [50] is an established tumor suppressor implicated in promoting apoptosis and p21 expression in cancer cells[51]. Though miR-34a is not liver-specific, it has been shown to be upregulated at the termination phase of liver regeneration in rats and inhibit proliferation by targeting inhibin αB and Met when overexpressed in rat hepatocytes[52]. In addition, miR-34a has been shown to be significantly upregulated in animals and humans with Non-Alcoholic Fatty Liver Disease (NAFLD)[53, 54]. In rat hepatocytes, miR-34a overexpression increased apoptosis when challenged with ursodeoxycholic acid via directly targeting Sirtuin 1 (SIRT1), an NAD-dependent deacetylase for p53[54]. In an in vivo study, miR-34a was shown to target SIRT1 and peroxisome proliferator-activated receptor-α (PPARα), an essential modulator of lipid transport and metabolism[55]. Importantly, when injected with a miR-34a inhibitor, mice with NAFLD had less lipid accumulation and improvements in levels of steatosis[55]. Taken together, miR-34a is dysregulated in liver disease and may be targeted therapeutically to promote hepatocyte survival and limit steatosis.

MiRNAs in liver development and regeneration

Despite the liver’s vulnerability to toxic insult, it is also the one mammalian organ that is capable of fully regenerating to achieve baseline size and function after acute or chronic injury[56]. Normally, hepatocytes are highly differentiated and quiescent and will only enter the cell cycle once initiated by ischemia, toxic exposure or infection. Interestingly, mice with hepatocyte-specific Dicer knockout exhibit normal development and hepatic function at birth, suggesting the absence of mature miRNAs does not greatly affect liver development[57]. However, within four months after birth these mice experience progressive liver damage, decreased function and aberrant hepatocyte proliferation and apoptosis[57]. In addition, these mice were shown to have deficiencies in correct liver zonation, metabolic pathways and tumor suppression[58, 59]. This suggests miRNAs are critical in basic liver function. Furthermore, specific miRNAs have been found that regulate processes of liver regeneration as well as the response toxic stress as will be discussed below (Table 1).

Table 1.

Additional miRNAs involved in in vivo liver regeneration

miRNA	Expression during liver regeneration	Target(s)	Model	Effect of miRNA	Citation
miR-23b	downregulated at termination phase	SMAD3	Rats	Inhibition of miRNA inhibited liver cell proliferation	[135]
miR-26a	downregulated	Cyclin D2*, Cyclin E2*	Mice	Inhibition of miRNA promotes hepatocyte proliferation	[136]
miR-127	downregulated	BCL2*, SETD8*	Rats	Inhibition of miRNA promoted liver cell proliferation	[137]
miR-221	upregulated	p27, p57, Arnt, Puma	Mice	Overexpression of miRNA promoted hepatocyte proliferation	[138–140]
miR-382	upregulated	PTEN	Mice, humans	Overexpression of miRNA promoted liver cell proliferation	[141, 142]

Asterisk (*) represents that only correlation between miRNA and target was shown, not direct interaction.

MiR-21

MiR-21 is among the most upregulated miRNAs during liver regeneration. Previously, miR-21 has been identified as an onco-miRNA that has multiple targets that normally inhibit proliferation. These targets include phosphatase and tensin homolog (PTEN), B-cell translocation gene 2 (BTG2), and Ras homolog gene family member B (RhoB) [60–63]. Therefore, when miR-21-specific antisense oligonucleotides were injected into mice following two-thirds hepatectomy, early liver regeneration was impaired because of a decrease in hepatocyte proliferation[63]. In this study, miR-21 was found to target RhoB and increase Akt and mamalian target of rapamycin (mTOR) signaling during early regeneration, which increased global translation in the cell via increased eukaryotic initiation factor (eIF)-4F activation. Also, upregulation of miR-21 increased cell cycle progression through G1 to S phase via accelerated translation of cyclin D1 mRNA[63]. A later study showed miR-21 was upregulated during in vitro hepatocyte proliferation and also targeted PTEN to enhance PI3K/Akt activity[60]. Interestingly, another study showed that deoxycholic acid (DCA) – a toxic bile acid that may lead to NAFLD – inhibited miR-21 expression in rats. When miR-21 was overexpressed in hepatocytes in vitro DCA-induced cell death was reduced[64]. Overall, evidence suggests that miR-21 promotes liver regeneration and may protect from pathological liver diseases such as NAFLD.

MiRNAs IN HEART INJURY AND REPAIR

MiRNAs in heart injury

Many drugs and environmental chemicals are known to induce cardiotoxicity specifically by targeting cardiomyocytes and can contribute to the development of coronary heart disease[65]. Cardiotoxicity is a major obstacle to drug development largely because of the failure to predict it in preclinical studies[66]. Currently, pre-clinical and clinical cardiotoxicity biomarkers include measuring the electrical efficiency of the heart using electrocardiography, visualizing the heart with echocardiograms and hERG assays to assess QT prolongation. In addition, methods for assessing cardiac dysfunction include measuring cardiac troponins or performing endomyocardial biopsies, which are invasive[67]. Therefore, miRNAs have been studied as being more sensitive, specific and stable biomarkers of heart disease (Table 2). In addition, many groups have looked at miRNAs that respond to direct cardiac injury and are involved in mechanisms of cardiomyocyte cell death.

Table 2.

MiRNA biomarkers that respond to cardiac injury.

miRNA	Expression during heart injury	Biofluid	Model	Citation
miR-486-3p	upregulated in MI	sera	human	[143]
miR-150-3p	upregulated in MI	sera	human	[143]
miR-126-3p	downregulated in MI	sera	human	[143]
miR-26a-5p	downregulated in MI	sera	human	[143]
miR-191-5p	downregulated in MI	sera	human	[143]
miR-1	upregulated in MI	plasma	human, mice	[144, 145]
miR-499	upregulated in MI	plasma	human, mice	[145–147]
miR-133a	upregulated in MI	plasma	human, mice	[145]
miR-133b	upregulated in MI	plasma	human, mice	[145]
miR-122	downregulated in MI	plasma	human	[145]
miR-375	downregulated in MI	plasma	human	[145]
miR-208b	upregulated in MI	plasma	human	[147]

MiR-34a

In a mouse model of chronic cardiotoxicity using doxorubicin, miRNA sequencing identified that miR-34a was the only miRNA upregulated across five time points and showed higher expression with increasing doses of doxorubicin[68]. In a different study, miR-34a was increased in the serum exosomes of patients who suffered myocardial infarction (MI) and developed heart failure within one year, showing its potential as a prognostic biomarker[69]. In addition, miR-34a has been implicated in aging-associated cardiomyocyte cell death and when inhibited in mice after MI reduced cell death and improved recovery of cardiac function[70]. Taken together, miR-34a acts similarly in the heart as was previously discussed in the liver by promoting cell death and tissue dysfunction.

MiR-208a

MiR-208a is a cardiac-specific miRNA that is generated from the intronic region of the gene that codes for a major protein that regulates myocardial contractility in the adult heart, α-myosin heavy chain (α-MHC)[71]. MiR-208a has been shown to regulate critical cardiac transcription factors and is required for proper cardiac function[72]. Specifically, overexpression of miR-208a resulted in increased α-MHC expression and was associated with arrhythmias, fibrosis and hypertrophy in mice[72]. Following isoproteronol-induced cardiac injury in rats, miR-208a was shown to have stable upregulation in plasma and was superior to cardiac troponins in indicating injury[73, 74]. MiR-208a was also increased in the human heart tissue of patients who died from MI[75]. Importantly, miR-208a was upregulated in plasma and exhibited a greater sensitivity than cardiac troponins in patients following MI[76]. As such, miR-208a has also been evaluated as a possible therapeutic target for heart injury. In one study, inhibiting miR-208a before doxorubicin treatment prevented cardiomyocyte apoptosis and improved cardiac function as determined by echocardiography[77]. In addition, in a rat model of congestive heart failure, miR-208a inhibition prevented pathological cardiac remodeling and improved cardiac function and survival[78].

MiRNAs in heart development and regeneration

During development, the heart increases in size due to proliferation of cardiomyocytes. Despite some limited renewal throughout life, mammalian cardiomyocytes generally exit the cell cycle after birth and any further increase in heart size is due to enlargement of existing cells[79]. The failure of cardiomyocytes to proliferate after injury can lead to persistent thinning of the myocardial wall, an expansion of cardiac fibroblasts that can lead to fibrosis and an overall decrease in cardiac function. MiRNAs have been shown to be necessary for proper heart development and cardiac function. Previously, mice with cardiac progenitor cell-specific Dicer deletion were shown to be embryonically lethal due to malfunctions in ventricular myocardium development, outflow tract morphogenesis and chamber septation[80, 81]. Likewise, three week-old juveniles with myocardium-specific, tamoxifen-inducible Dicer knockout resulted in ventricular remodeling and premature death within one week. Also using these mice, it was shown that inducing Dicer knockout in adults led to severe hypertrophy, myofiber disarray and ventricular fibrosis[82]. Recently, many miRNAs have been identified as being critical for the normal development of the heart and in enhancing processes of repair after injury.

MiR-15 family

In mice, there is short period (<7 days) after birth when the heart can sustain ischemic injury and cardiomyocytes will proliferate to facilitate full regeneration; however, these regeneration mechanisms are silenced after this period[83]. Previously, the miR-15 family was shown to help regulate this process (Figure 3). The miR-15 family is composed of six members (including miR-15a, miR-15b, miR-16-1, miR-16-2, miR-195 and miR-497) that all have the same seed sequence. Interestingly, the miR-15 family generally, and miR-195 specifically, are among the most highly expressed miRNAs immediately after the postnatal period when mice lose the ability to regenerate the heart. Overexpression of miR-195 has been shown to promote even earlier premature cardiomyocyte cell cycle withdrawal by targeting checkpoint kinase 1 (Chek1), which regulates the progression through G2/M and spindle assembly checkpoints in the cell cycle[84]. Conversely, inhibition of miR-15b and miR-16 prolonged the postnatal regeneration period by increasing the expression of Chek1[85]. In addition, the miR-15 family is highly upregulated after MI and was shown to promote apoptosis by targeting B-cell lymphoma 2 (Bcl-2) and led to mitochondrial dysfunction by targeting ADP-ribosylation factor-like protein 2 (Arl2). Impressively, using anti-miRs complementary to the miR-15 family was able to protect adult mice and pigs from MI[86].

Figure 3. Diverse approaches have been developed using miRNAs that can promote heart regeneration.

Cardiomyocytes (CMs) may be the target of injury from drugs, environmental chemicals or from myocardial infarction (MI). Following injury, cardiomyocytes die and do not normally recover. Many different approaches have been developed with specific miRNAs or combinations of miRNAs that promote cardiac regeneration and recovery.

miR-17~92 cluster

Identified as the first onco-miRNA (“oncomiR-1”), the miR-17~92 cluster has been implicated in promoting cardiac differentiation and proliferation (Figure 3)[87]. This polycistronic cluster encodes six mature miRNAs (including miR-17, miR-18a, miR-19a, miR-19b-1, miR-20a and miR-92-1), all of which are expressed differentially due to selective post-transcriptional processing mechanisms[88]. Using miR-17~92 knockout mice and in vitro knockdown strategies, the cluster was shown to target cardiac progenitor genes and promoted proper myocardial differentiation[89]. Interestingly, cardiac-specific miR-17~92 knockout mice were shown to have smaller hearts due to a decrease in total cardiomyocytes, but most mice survived until adulthood despite decreased cardiac function[90]. Conversely, cardiac-specific miR-17~92 transgenic mice had increased postnatal cardiomyocyte proliferation through direct PTEN suppression that resulted in larger hearts as well as cardiomyocyte proliferation into adulthood. Importantly, transgenic mice were protected from MI as shown by echocardiogram and increased cardiomyocyte proliferation[90].

miR-99/100 & Let-7a/c

Unlike mammals, many lower vertebrates including amphibians and fish retain the ability of cardiac regeneration throughout their entire lives[91, 92]. Specifically, adult zebrafish can sustain an amputation of up to 20% of the ventricle and the remaining cardiomyocytes will still dedifferentiate, proliferate and redifferentiate to completely regenerate the heart[92, 93]. Some groups have postulated that this ability may still be present in mammals and active during development but is somehow silenced in adulthood. A recent study explored this concept further by analyzing miRNAs in regenerating zebrafish hearts post-amputation[94]. They found two specific miRNA families, miR-99/100 and Let-7a/c, that were conserved in higher vertebrates and downregulated during zebrafish heart regeneration (Figure 3). These miRNAs were shown to target SMARCA5 and FNTβ which are key regulators of differentiation. In mice, however, miR-99/100 and Let-7a/c levels increase and SMARCA5 and FNTβ decrease during development, but do not change following induction of MI in adults. Incredibly, inhibiting miR-99/100 after MI was sufficient to induce dedifferentiation and proliferation of cardiomyocytes, reduce infarct size and improve cardiac function[94].

Combination of miRNAs

Instead of increasing proliferation of surviving cardiomyocytes after injury, some groups have focused on inducing cardiac fibroblasts or embryonic stem cells (ESCs) to become cardiomyocytes and injecting them into the injured areas of the heart. In this case, however, combinations of miRNAs have been examined that regulate cardiomyocyte differentiation. Specifically, a set of miRNAs deemed “miR combo” (including miR-1, miR-133, miR-208a and miR-299) have been utilized due to their established roles in cardiac-specific development and function (Figure 3)[95]. Previously, both MiR-1 and miR-133 were shown to be co-transcribed, specific to cardiac muscle and skeletal muscle and integral for promoting proper heart development[96–98]. Also, ESCs have been transfected with miR-1 and improved cardiac function when transplanted following MI[99]. However, there is evidence of miR-1 and miR-133 having opposing functions in cardiomyocyte differentiation, suggesting they likely have different but perhaps complementary roles in cardiac functions[100, 101]. The other members of miR combo, MiR-208 and miR-499, are intronic miRNAs both within the α- and β-MHC primary transcript and have been shown to control MHC expression[72, 102]. Through empirical experimentation, miR combo was discovered to induce significant fibroblast reprogramming to cardiomyocyte-like cells in vitro[95]. Importantly, following MI and the subsequent injection of lentivirus-encoding miR combo in mice, cardiac fibroblasts became positive for cardiomyocyte markers. Injection of miR-1 alone was sufficient to promote this cardiac fibroblasts reprogramming in vivo, but the addition of miR-133, 208 and 499 – especially when JAK inhibitor was included – further improved these effects and resulted in improved cardiac function[103].

MiRNAS IN KIDNEY INJURY AND REPAIR

MiRNAs in kidney injury

The kidney is a critical organ for the excretion of waste products from the body and therefore, like the liver, is a main target of toxicity or ischemic insult. As a result, acute kidney injury (AKI) is growing as a major problem for clinicians around the world[104]. In fact, a recent meta-analysis found that AKI occurred in one-in-five adults and one-in-three children worldwide who were hospitalized with acute illnesses[105]. AKI is believed to be increasing in part due to growth in the aging population and health events such as diabetes mellitus, cardiovascular disease, invasive surgeries, sepsis, chronic kidney disease (which predisposes people to AKI) and exposure to environmental toxicants and nephrotoxic drugs[106]. Therefore, because no specific treatments have yet been developed to ameliorate AKI, miRNAs have been investigated that may have a role. Proximal tubule-specific Dicer knockout mice were shown to be protected from renal ischemia reperfusion injury (IRI), suggesting that mature miRNAs are critical in regulating pathways that mediate the response to kidney injury[107]. Many specific miRNAs have also been identified that regulate kidney injury and offer therapeutic benefits when modulated (Table 3).

Table 3.

Additional miRNAs described in kidney injury.

miRNA	Expression during kidney injury	Target(s)	Model	Effect of miRNA	Citation
miR-24	upregulated by IRI	H2AX, HO-1	Mice	Inhibition of miRNA reduced tubular injury, decreased fibrosis & increased survival	[148]
miR-34a	upregulated by cisplatin & IRI	ATG4B	Mice, mouse PTECs	Inhibition of miRNA increased apoptosis of cisplatin-treated mouse PTECs & decreased autophagy	[149, 150]
miR-155	upregulated following IRI, UUO & gentamycin-induced injury	c-Fos*	Rats, mice	Knockout mice had increased kidney injury & decreased function	[113, 151]
miR-489	upregulated following IRI	PARP1	Mice	Inhibition of miRNA increased apoptosis in cells, exacerbated kidney injury	[152]
miR-494	upregulated following IRI	ATF3	Mice	Overexpression of miRNA increased apoptosis and inflammation, decreasing renal function	[153]

Asterisk (*) represents that only correlation between miRNA and target was shown, not direct interaction. AA: aristolochic acid; IRI: ischemia-reperfusion injury; PTECs: proximal tubular epithelial cells; UUO: unilateral ureteral obstruction

Current methods of detecting AKI rely on non-specific and insensitive biomarkers such as serum creatinine (SCr) and blood urea nitrogen (BUN)[108]. In 2008, FDA and European Medicines Agency qualified seven biomarkers for kidney toxicity in preclinical studies and the evaluation of these in humans is currently ongoing[109]. These studies provided compelling evidence that showed proteins excreted in the urine, such as Kidney Injury Molecule-1 (KIM-1), outperform BUN and SCr and correlate with histopathology in the rats. This has prompted investigators to conduct additional discovery studies to identify miRNAs as sensitive biomarkers or potentially as companion biomarkers that can also provide mechanistic information regarding pathway dysregulation[110].

MiR-21

MiR-21 is the most studied miRNA in kidney disease and is promising as both a biomarker and therapeutic target (Figure 4). Mice with miR-21 deletion or treated with miR-21 inhibitors after the induction of unilateral ureteral obstruction (UUO)-induced kidney injury were protected from fibrosis development[111]. In this study, miR-21 was shown to directly target PPARα and Mpv17I, an inhibitor or reactive oxygen species (ROS) generation in the mitochondria, and therefore miR-21 knockout mice had increased lipid oxidation, higher activation of metabolic pathways and decreased ROS following injury. In a mouse model of Alport syndrome, a genetic disorder involving progressive fibrosis that results in end-stage renal failure in young adults, miR-21 inhibition significantly reduced the development of fibrosis[112]. Because of these promising preclinical results, a miR-21 inhibitor, RG-012, is currently in Phase 2 clinical trials for this indication[20]. In addition, our lab was among the first to measure miR-21 in the urine supernatant in mice and patients following AKI[113]. Later, our lab was able to effectively differentiate AKI patients from healthy patients by measuring miR-21 combined with miR-200c, miR-423 and miR-4640 in urine[114]. Since then, miR-21 has been found to be a reliable biomarker of drug-induced kidney injury[110] and fibrosis[115], and was associated with kidney injury from multiple chronic diseases including type 2 diabetes and IgA nephropathy[116, 117].

Figure 4. MiR-21 mediates kidney injury and causes the progression of fibrosis.

Following diverse forms of kidney injury, miR-21 is upregulated in kidney cells including the tubular epithelium, glomerular cells, myofibroblasts and infiltrating macrophages. MiR-21 mediates kidney injury by decreasing lipid oxidation, deregulating metabolic processes and increasing reactive oxygen species (ROS). In addition, miR-21 is secreted in the urine within small vesicles, dead cells or bound to RISC proteins or high-density lipoproteins. MiR-21 can be measured in the sediment or supernatant of urine samples as a biomarker of kidney injury. Because of the role of miR-21 in promoting fibrosis, a miR-21 inhibitor, RG-012, is in clinical trials to treat Alport nephropathy.

Additional miRNA biomarkers for AKI

Investigators looking at miRNA biomarkers of AKI have focused largely on measuring changes in the urine, since the kidney’s role is to produce urine and its collection is noninvasive. In addition, miRNAs are remarkably stable in this medium[118]. Multiple groups have described urinary miRNA expression changes in preclinical rodent models and patients with drug-induced kidney injury[110]. In one study, male Wistar rats treated with a single dose of the nephrotoxicant cisplatin were shown to have a greater than 20-fold increase in 11 miRNAs in the urine – including miR-15, miR-16, miR-20a, miR-192, miR-193 and miR-210 – over multiple time points[119]. Another study using cisplatin in Sprague-Dawley rats found let-7g-5p, miR-93, miR-191a and miR-192 significantly increased in the urine after injury[120]. Though kidney biopsies are rarely acquired in the clinic, sequencing studies conducted on biopsies of human kidney allografts have suggested specific miRNAs that may be used as biomarkers of acute rejection. In a study comparing 12 acute rejection biopsies to 21 normal, six miRNAs were differentially regulated including miR-142-5p, miR-155 and miR-223 (all upregulated) and miR-10b, miR-30a-3p and let-7c (all downregulated)[121]. In a later study, several miRNAs including miR-21, miR-142-3p, miR-142-5p and the miR-506 cluster were significantly increased in allograft biopsies with tubulointerstitial fibrosis compared to normal while miR-30a, miR-30d and miR-30e were all significantly downregulated[122]. Though more cross-validation and translational studies are needed, these studies show the potential of miRNA biomarkers for AKI.

MiRNAs in kidney development and regeneration

Following an episode of AKI the kidney has a remarkable ability to regenerate, though repeated injury or incomplete repair could lead to chronic kidney disease[123]. During mammalian development, Six2-positive progenitor cells give rise to all of the diverse cell types that constitute the nephron[124]. Previous studies in mice using Six2-specific Dicer ablation have shown that miRNAs are critical in regulating proper nephrogenesis[125]. In addition, Six2-specific miR-17~92 knockout mice had decreased nephrogenesis and developed albuminuria and glomerulosclerosis by three months[126]. Likewise, Dicer knockouts in other kidney cell subtypes have had differential effects on kidney function. For example, proximal tubule-specific Dicer knockout mice had no apparent effects on kidney function under normal conditions[107]. Conversely, when Dicer was knocked out in podocytes (cells that are essential in forming the glomerular filtration barrier), mice appeared to develop normally but rapidly acquired significant glomerular and tubular injury, proteinuria and died within four weeks after birth[127]. Ablating Dicer in juxtaglomerular cells (cells in the glomerulus that are responsible for renin synthesis and release) resulted in decreased cell number, renin signaling and lowered blood pressure[128]. Taken together, miRNAs are clearly critical in development but knowledge about specific miRNAs that regulate processes of regeneration is limited.

Compared to other organs discussed in this review, more is known about miRNAs that mediate kidney injury than about specific miRNAs involved in kidney repair. Nonetheless, some efforts have been initiated through temporal miRNA sequencing experiments following AKI to identify miRNAs that are elevated during periods of repair. Our lab recently conducted small RNA sequencing in mice subjected to folic acid-induced nephropathy and identified 108 miRNAs that were differentially regulated across seven time points[129]. Using principal component analysis, we found that there were unique expression patterns of miRNAs in the injury phase versus the repair and fibrotic phases. In addition, at early time points following injury when repair processes are initiated (days 1 through 3), miR-18a and miR-132 were highly expressed throughout the renal cortex and in the injured tubules. Similar temporal expression patterns were seen in renal IRI and UUO models. Independent of our lab, a different miRNA sequencing experiment conducted on multiple early time points following induction of renal IRI identified miR-18a as possibly being involved in repair[130]. Interestingly, in a different study miR-132 inhibition following UUO was shown to increase the number of proliferating proximal tubular cells and reduce the development of fibrosis, but inhibition of miR-132 in vitro resulted in less proliferation[131]. In addition to these miRNAs, a recent study showed that miR-687 induced cell cycle progression by targeting PTEN following IRI in mice[132]. Interestingly, however, inhibiting miR-687 resulted in less tubular apoptosis and improved kidney function. Much more work must be conducted on kidney regeneration in general and how miRNAs specifically mediate these processes.

OTHER miRNAs INVOLVED ORGAN INJURY AND REPAIR

In addition to the liver, heart and kidney, there is emerging evidence of miRNAs offering therapeutic potential in the injury and repair of other organs. For example, miRNAs have been identified that are associated with the injury and repair of lungs, skin and the central nervous system (Table 4).

Table 4.

MiRNAs that regulate injury and repair processes in other organs.

miRNA LUNG	Expression	Target(s)	Model	Effect of miRNA	Citation
miR-146a	Upregulated by LPS	IRAK-1, TRAF-6	Rats	Overexpression of miRNA suppressed pro-inflammatory signals in alveolar macrophages	[154]
miR-155	Upregulated in ventilator- induced injury	C/EBPα	Mice	Inhibition of miRNA decreased BAL, cells in BAL and pro-inflammatory cytokines	[155, 156]
miR-21	Upregulated by ventilator- induced injury	BMPR2*, PTEN*	Mice	Inhibition of miRNA decreased BAL and improved oxygen exchange	[155]
miR-127	Downregulated by bleomycin	CD64	Mice	Overexpression reduced pro-inflammatory signals in vivo	[157]
SKIN
miR-31	Upregulated by wounding & TGF-α	EMP-1	Human	Overexpression of miRNA induced keratinocyte proliferation & migration	[158]
miR-132	Upregulated by wounding & TGF-α	HB-EGF	Human, mouse	Inhibition of miRNA prevented epithelialization in human and mouse wound model	[159]
miR-200b	Downregulated by wounding	GATA2, VEFGR2	Mice	Inhibition of miRNA mediated angiogenesis in vivo	[160]
miR-21	Upregulated in chronic wounds	LepR, EGR3	Rats, human	Overexpression of miRNA inhibited epithelialization in human & rat wound model	[161]
CENTRAL NERVOUS SYSTEM
miR-21	Upregulated by TBI	PTEN*	Rat	Inhibition of miRNA increased apoptosis and decreased angiogenesis after injury	[162]
miR-23b	Downregulated in plasma of TBI patients & brain of rats	ATG12	Rats, human	Overexpression of miRNA reduced apoptosis, lesion volume & edema in brain	[163]
miR-223	Upregulated in injured spinal cord	GluR2*	Rats	Inhibition of miRNA reduced apoptosis and improved motor function	[164]
miR-497	Upregulated by cerebral ischemia	Bcl-2, Bcl-w	Mice	Inhibition of miRNA reduced infarct volume, improve neurological score	[165]

Asterisk (*) represents that only correlation between miRNA and target was shown, not direct interaction. BAL: bronchoalveolar lavage; LPS: lipopolysaccharide; TBL: traumatic brain injury

FUTURE PERSPECTIVES

The rapid growth in miRNA research has made clear that miRNAs are critical players in the molecular mechanisms of mammalian biology. There is ample evidence that miRNAs can serve as stable and specific biomarkers of various disease states and efficacious targets to manipulate therapeutically. As a result, there has been a surge of patent applications that is approaching 500 annually and there are many miRNA therapeutics in various stages of clinical development[133]. Still, challenges exist in oligonucleotide therapeutic development and delivery. Because of the ability of miRNAs to regulate multiple pathways and their different functions depending on the tissue and context, identifying miRNAs that can be modulated efficaciously and safely in patients may be difficult. In addition, although different miRNA targeting strategies are being developed there are currently no examples of miRNA-specific oral therapeutics, and half-life and specificity remains an important issue. Nonetheless, chemically-modified oligonucleotides and liposomal delivery methods have made enormous progress in recent years. Diverse strategies of therapeutic delivery include restoring miRNA function with mimics or overexpression vectors or inhibiting miRNAs with antisense oligonucleotides, vectors that express miRNA “sponges” or small molecule inhibitors that inhibit miRNA pathways[134]. Moreover, proprietary chemical modifications to nucleotide backbones have been developed that confer resistance to serum nucleases and have improved target-binding affinities. Taken together, manipulating miRNAs in human disease settings offers a promising path forward, particularly with regards to organ injury and repair.