The human miR-21 gene has 3,433 nucleotides, located in chromosome 17q23.2, an intergenic region with the 3' UTR end of the Transmembrane Protein 49 (TMEM 49) (also known as Human Vacuole Membrane Protein 1, VMP-1) (15). miRNAs share a common biosynthetic pathway. The biogenic processing steps of miR-21 were first characterized in 2004 (16). miR-21 gene has two transcription sites, T1 (the minor transcription site) and T2 (the primary transcription site), and both are utilized for the initiation of transcription. RNA Pol II enzyme initiates miR-21 transcription and synthesizes full-length (~3.5-kb) primary miR-21 (pri-miR-21) in the nucleus (17). Following transcription, primary miR-21 requires two cleavage steps to become a mature and functional miRNA. First, pri-miR-21 is cleaved into ≈70-nt-long intermediate precursor miR-21(pre-miR-21) by the microprocessor complex that contains RNase III Drosha and a dsRNA binding protein DGCR8 (18, 19). Exportin 5 transports pre-miR-21 in a Ran-GTPase-dependent manner from nuclear to cytoplasm, where maturation can be completed (20). Upon export to the cytoplasm, a second RNase III enzyme, Dicer, removes the terminal loop from pre-miRNA and releases a small 22-nucleotide miRNA duplex. Following miRNA duplex loading, the precursor RNA-induced silencing complex (pre-RISC) quickly degrades the “passenger” strand to form a mature miRNA- RISC complex (21).

It is shown that most human protein-coding genes contain one or more miRNA-binding sites, so that miRNAs strictly regulate protein expression (22). In miRNA- RISC complex, miRNAs serve as target recognition purpose. “miRNA seed” is a domain at 2–7 nucleotides position of the 5' end of miRNAs. It recognizes the 3' untranslated region (UTR) of mRNAs, and miRNAs bind to UTR with complementary base pairing fashion. Once locked on an mRNA, the RISC silencing complex promotes downregulation of the protein by repressing or degrading the mRNA (23).

miR-21 is proven to be involved in the pathogenesis of cardiovascular diseases through regulating downstream target genes. To date, several genes are identified as miR-21 targets in cardiomyopathy. The most recognized two target genes that involve in anti-apoptosis are programmed cell death 4 (PDCD4) (24) and phosphatase and tensin homolog deleted from chromosome 10 (PTEN) (25). In cardiac fibrosis, the primary three target genes are SMAD family member 7(Smad7) (26) and sprouty1/2 (SPRY1/2) (27, 28). Besides, some other targets are also suggested that contain miR-21 specific binding sites, including peroxisome proliferator-activated receptor alpha(PPAR-α) (29), a-kinase anchoring protein 8 (Akap8), and BRCA1 associated RING domain 1 (Bard1) (30), Jagged 1 (31), sorbin and SH3 domain-containing protein 2 (SORBS2) and PDZ and LIM domain 5 (PDLIM5) (32).

In vitro studies, two main cell types, cardiomyocytes and cardiac fibroblasts are used to illustrate the miR-21 expression pattern and function.

In most cases, miR-21 expression levels were elevated in stress conditions such as oxidative stress, high glucose, and hypoxia (59–62). However, oxygen-glucose deprivation (OGD) and palmitate exposure can considerably decrease miR-21 expression levels in cardiomyocytes (63, 64).

Furthermore, murine models were applied to study miR-21 expression profiles in cardiomyopathy showing different miR-21 responses to cardiac damage. miR-21 was down-regulated in infarcted areas but was upregulated in border areas in the acute myocardial infarction (AMI) rat model (33). In the myocardial infarction (MI) mouse model miR-21 expression changed in a time-dependent manner, increased in the first and second week but unchanged in the fourth week (34). Besides, two independent studies in mice models showed opposite miR-21 expression trends. One was observed an increase of miR-21 in the infarcted zone after MI, and the other showed a decrease of miR-21 after AMI (35, 36). Similarly, the reverse expression trends were also observed in ischemia/reperfusion (I/R) murine injury models. In the I/R injury mouse model, miR-21 was upregulated in the heart (55, 56), whilst heart miR-21 expression level decreased in the myocardial I/R injury rat model (57).

In the clinical studies, the expression level of miR-21 was measured in the blood and heart tissue of patients suffering from different heart diseases. The results showed divergent expression trends. In the case of atrial fibrillation (AF), it is suggested that miR-21 might be involved in the pathogenesis of atrial fibrosis due to its increased expression in left atrial tissue and positive correlation with atrial collagen content (44). Another study also showed that miR-21 was highly expressed in human atrial myocytes from patients with chronic atrial fibrillation compared to the sinus rhythm group (45). By contrast, different independent studies observed significantly decreased miR-21 levels in plasma or serum in AF patients (46, 47). Likewise, the reverse trends of miR-21 expression were also found in patients with coronary artery diseases. miR-21 expression levels were upregulated in plasma (6, 39–41), but dramatically decreased in peripheral blood mononuclear cells (PBMCs) (42).

In heart failure, for example, the upward trend expression of miR-21 was observed in circulation (48), but there was a decrease of miR-21 expression in serum in symptomatic heart failure patients (49). More surprisingly, patients with acute heart failure showed increased plasma miR-21 levels between admission and discharge of hospital, then decreased following clinical compensation period (50). Further, this time-dependent expression of circulating miR-21 occurred in patients with post-myocardial infarction as well. miR-21 decreased in the first 2 days and increased on the 5th day, then returned to the control level in the latter point of post-myocardial infarction (37). In another study, upregulated serum miR-21 expression was observed in elderly patients with acute myocardial infarction (AMI) (38).

In other cases, various cardiomyopathies, including dilated cardiomyopathy, left ventricular ejection fraction, and ischemic cardiomyopathy, upregulated miR-21 expression levels were shown in heart tissue and circulation (53, 54, 58, 65). In addition, increased miR-21 level was suggested as a new biomarker of cardiovascular disease-related premature death to predict patients at great risk of adverse outcomes (66, 67). However, unchanged expression miR-21 level was also reported in patients with hypertrophic cardiomyopathy (53).

In short, it has been shown that the distinct miR-21 expression pattern in one cardiac disease depending on different sample sources. Moreover, inconsistent trends were also observed from the same source of samples (such as plasma) in the same cardiac diseases.

Positive functions of miR-21 were studied in a range of heart diseases by both gains-of-function and loss-of-function. In myocardial infarction (MI), infarct heart size was decreased, and heart function improved after transfection of miR-21 extracellular vesicles (EVs) or lentivirus overexpressing miR-21. The further mechanism study revealed that miR-21 downregulated programmed cell death 4 (PDCD4) and apoptosis markers in cardiomyocytes, including bcl-2-like protein 4 (bax) and caspase-3 (34, 36). The similar protective effects of miR-21 also showed in myocardial ischemia-reperfusion (IR), diabetic cardiomyopathy, and hypertrophic cardiomyopathy. miR-21 attenuates cardiac injury and promotes cardiac myocytes survival through targeting PDCD4, gelsolin, and well-known tumor suppressor gene PTEN (52, 56, 70). Apart from anti-apoptotic function, miR-21 plays an anti-inflammatory role in cardiomyopathy as well. miR-21 mimics notably decreased the rate of tumor necrosis factor (TNF)-α-induced apoptosis in human cardiomyocytes by activating the JNK/p38/caspase-3 signaling pathway (38). In the I/R injury study, the overexpression of miR-21 effectively inhibited the TLR4/NF-κB pathway that reduced the level of myocardial apoptosis and inhibited the release of pro-inflammatory factors (57). Furthermore, a study has presented a new targeted therapy that miR-21 nanoparticle delivery into cardiac macrophages could promote angiogenesis, reduce hypertrophy, fibrosis, and cell apoptosis in the remote myocardium (68). miR-21 also can promote the proliferation of cardiac stem cells (CSCs). miR-21 up-regulation with miR-21 mimics efficiently accelerated cell viability and proliferation in CSCs and protected them from oxidative injury by activating PTEN/PI3K/Akt pathway (72, 76). It is also proven that miR-21 can decrease cell apoptosis, inhibit viral progeny release in the heart, and protect it from coxsackievirus B3 infection through MAP2K3/P38 MAPK signaling pathway (77). Overall, miR-21 has beneficial effects on diseased cardiac tissues mainly by inhibiting PTEN and PDCD4 and promoting cell survival and proliferation.

miR-21 has been proposed as a biomarker in a variety of heart diseases. A clinical biomarker is defined as “an indicator of normal biological processes, pathogenesis, or responses to an exposure or therapeutic intervention” (89). For instance, troponin I (cTnI) and T (cTnT) are unique heart proteins and are specific and sensitive biomarkers of myocardial injury in various cardiovascular disorders (90). Ideal biomarkers have the potential to provide valuable information that aids the diagnosis of a disease and predicts disease progression and outcomes. Bradford Hill's Guidelines can be applied to establish and evaluate pre-biomarkers (91). Here, we list the first three criteria to show whether or not miR-21 can be a biomarker of cardiomyopathies.

The first essential requirement is that there should be strong associations between a biomarker and the outcome of a clinical disorder. Undoubtedly, miR-21 regulates key pathways and tightly relates to cardiac pathogenesis, such as cardiac fibrosis. Expression changes of miR-21 were observed in many studies, and the pathological roles were demonstrated by both gain- and loss- of-function study.

Additionally, consistent findings should be observed by different individuals in different places with different samples. miR-21 is not fulfilled with this point. The inconsistent expression levels and contradictory functions have been reported throughout the last two decades. miR-21 expression levels in circulation and heart tissue were various in independent studies. A plausible reason is that there are no standard reference values for miR-21 expression in physiological and pathological conditions. The expression levels in control groups might vary from subject to subject.

Most importantly, an ideal marker is supposed to express at a specific site and associate with a particular disease. miR-21 is not a specific miRNA in the heart. miR-21 is expressed abundantly and ubiquitously in some other multiple types of tissues, such as the liver, kidney, nervous system, and immune cells (92). Not only does diseased heart tissue contribute to the circulation level of miR-21, but other tissues, especially those globally involved in various diseases, namely immune cells, also secret miR-21 to respond to pathological stimuli via inflammation pathways (93). In addition, miR-21 is not a disease-specific marker either. miR-21 is suggested as a marker in various cardiovascular diseases, including heart failure, atrial fibrillation, and dilated cardiomyopathy. Apart from cardiovascular diseases, numerous studies implicated miR-21 as a biomarker in many other diseases ranging from cancer, renal fibrosis, and diabetes (94). So far, miR-21 cannot be considered an ideal biomarker for a specific cardiomyopathy.

With increasing interest in miRNAs, miRNA-based drug development is already in progress. Over the last decade, several biopharmaceutical industries have launched miRNA projects in their development pipeline. For example, the first miRNA-targeted drug for treating hepatitis C virus (HCV) infection sponsored by Santaris Pharma in 2012 is miravirsen. It is an LNA-modified antimiR oligonucleotide, targets miR122 (95). miRNA-based drugs have two main types: AntagomiRs, which can specifically bind to target miRNAs in a complementary fashion and inhibit their functions; another is synthetic miRNA mimics, which can restore the reduced expression of miRNAs in disease conditions (96). Currently, there are going on many miR-21-based diagnostics and therapeutic clinical trials in a range of diseases, such as Predictive and Prognostic Value of Inflammatory Markers in Stage IV Colorectal Cancer (NCT04149613), miR-200b & miR-21 in Diabetic Wounds recruiting (NCT02581098), and Study of Lademirsen (SAR339375) in Patients with Alport Syndrome (HERA) (NCT02855268). Of note, miRNA-based drugs are currently not available for diagnostics and therapeutic purposes.

Similarly, miR-21 has emerged as a potential therapeutic target in cardiomyopathy due to its participation in the pathogenesis of many cardiac diseases via regulating cardiomyocytes apoptosis and cardiac fibrosis. Numerous preclinical studies have been carried out to illustrate its therapeutic outcomes in cardiomyopathies. For instance, a study demonstrated that miR-21 overexpression by nucleofection in human cardiac fibroblasts with the expression vector leads to upregulation of collagen expression. In contrast, miR-21 Inhibition by LNA anti-miR-21 reduced collagen expression and alleviated cardiac fibrosis (82).

However, developing miR-21 targeted treatment for cardiomyopathies has several challenges. Firstly, It is challenging to apply repression or restoration, because the confusion of miR-21 expression pattern in cardiomyopathies. Besides, it is problematic for targeted therapy that miR-21 is not an organ or tissue-specific miRNA, not even cell type-specific miRNA. Hence, issues with off-target effects and undesired on-target effects have to be taken into account (97).

Despite the considerable understanding of miR-21 regulation, further studies require to obtain sufficient knowledge of miR-21 in its spatial and temporal regulation. miR-21 expression is tightly regulated by multiple factors. Several promotors can effectively induce miR-21 expression at the transcriptional level, including AP-1, STAT3, Ras, ERK1/2, and EGFR (98–102). Oppositely, some transcriptional suppressors have also been reported. For instance, miR-21 transcription was repressed by NFI, C/EBP, Gfi1, and estrogen receptor (98, 103, 104). Moreover, miR-21 expression is regulated at the post-translational level. TGFβ and BMP4 (a member of the TGFβ superfamily) upregulate pre-mir-21 expression and SMADs (SMAD1/5 and SMAD2/3) mediate pri-mir-21 expression (105). On the contrary, BMP6 (a member of the TGFβ superfamily) has been shown to inhibit miR-21 expression (106). In addition, miR-21 expression can be modulated by other non-coding RNA associated with various molecular pathways. Tumor-Associated Long ncRNA Expressed on Chromosome 2(TALNEC2) was highly expressed in myocardial ischemic patients. Overexpression of TALNEC2 downregulated miR-21 and aggravated hypoxic injury by miR-21/PDCD4-mediated activation of the Wnt/β-catenin signaling pathway in myocardial ischemic injury (107). In cardiovascular diseases, LncRNA MEG3 is downregulated and negatively correlates with miR-21 levels. MEG3 overexpression suppresses endothelial cell proliferation and migration and reduces proteoglycan, type I, and V collagen expression through the decrease of miR-21 (108).