Roles of microRNA-192 in diabetic nephropathy: the clinical applications and mechanisms of action

Diabetic nephropathy (DN) is one of the most common and intractable microvascular complications of diabetes worldwide, serving as the main cause of terminal renal disease. Due to the lack of early specific symptoms and diagnostic markers, DN severely threatens the sufferer’s life. MicroRNA-192 (miR-192) was early identified in human renal cortical tissue and stored and excreted in urine as microvesicles. MiR-192 was found to be involved in the development of DN. For the first time, the present review summarized all the current evidence on the topic of the roles of miR-192 in DN. Finally, 28 studies (ten clinical trials and eighteen experimental studies) were eligible for thorough reviewing. Most of the clinical trials (7/10, 70%) indicated miR-192 might be a protective factor for DN development and progression, while the majority of experimental studies (14/18, 78%) suggested miR-192 might be a pathogenic factor for DN. Mechanistically, miR-192 interacts with various direct targeted proteins (i.e., ZEB1, ZEB2, SIP1, GLP1R, and Egr1) and signaling cascades (i.e., SMAD/TGF-β and PTEN/PI3K/AKT), together contribute to the pathogenesis of DN through epithelial-to-mesenchymal transition (EMT), extracellular matrix deposition, and fibrosis formation. The current review highlights the dual role of miR-192 in the development of DN. Low serum miR-192 expression could be applied for the early prediction of DN (the early stage of DN), while the high miR-192 level in renal tissues and urine may imply the progression of DN (the late stage of DN). Further investigations are still warranted to illustrate this inconsistent phenomenon, which may facilitate promoting the therapeutic applications of miR-192 in predicting and treating DN.


Introduction
Diabetic nephropathy (DN), a serious renal disease, is found to be correlated with the development of proteinuria, glomerular enlargement, reduction in glomerular filtration, and renal fibrosis (1). It affects roughly 9% of the global adult population in those suffering from either type 1 (T1DM) or type 2 diabetes mellitus (T2DM). In China, the prevalence of diabetes in adults was recorded at about 11% (2), while the prevalence of DN was up to 21.8% (3). There is a positive association between DN and the risk of the development of terminal renal disease (4). In diabetic patients with terminal kidney disease, the five-year survival rate is only 20%. There is a significant increase in deaths and morbidity associated with DN-associated heart failure with unchanged ejection fraction (5,6). Renal replacement therapy is necessary in a substantial proportion of cases due to there are few effective treatments. At present, the therapeutic strategies for DN aim to reduce its progression but most of them turned out to be ineffective (7). As a result, an effective, rapid, and non-invasive method for detecting DN early and predicting its prognosis is critical. In the past two decades, multiple microRNAs (miRNAs) have been identified to involve in the pathophysiological action in DN.
MiRNAs are short, endogenous, and noncoding RNA molecules consisting of 19-24 nucleotides. A miRNA achieves its biological function by binding to the 3' untranslated region of a target gene mRNA. By promoting the degradation of mRNA or causing translational repression, miRNAs effectively regulate gene expression post-transcriptionally (8, 9). MiRNAs are involved in the regulation of multiple cellular biological processes, either physiologic or pathologic conditions (e.g., proliferation, differentiation, programmed death, apoptosis, or passive cell death) (8). The dysregulation and aberrant expression of several miRNAs are also found to be associated with the development and progression of DN, such as miR -21, miR-126, miR-29, miR-192, miR-214, miR-342, and miR-192 (10, 11). According to the current literature, miR-192 is one of the most investigated miRNAs with aberrant expression in DN. Mounting clinical studies demonstrated that abnormal miR-192 expression level was detected in the majority of patients with DN. In addition, the molecular mechanisms of miR-192 action are found to be complex and multidirectional in the development of DN.
Since the pivotal role of miR-192 in DN has received increasing attention from the investigators, it is, therefore, necessary to summarize all the current evidence on this topic through a comprehensive review, which may help to develop a better understanding of the prognostic and predictive role of miR-192 in DN.
Currently, no relevant review article has been published for summarizing the clinical implications and the molecular mechanisms of miR-192 in DN. As a result, we conducted this comprehensive review based on the current evidence.
Finally, twenty-eight studies (31-57) (58) published in 2007-2022 were included in this review for further analysis. A routine data collection form was used to collect relevant information from the included studies, e.g., the first author name, article publication year, research subject (patients, cells, or animals), expression of miR-192 (up-regulation or down-regulation), involved molecular mechanisms, target genes, and the main findings within each eligible study. The study design for the 28 included studies was either clinical trials (10 studies) or experimental studies (18 studies). The sample size of the clinical trials ranged from 6 to 602 participants. Diverse clinical specimens were applied in the clinical studies, including serum, urine, and kidney tissues. According to the experimental studies, miR-192 was found to interact with various direct targeted proteins (i.e., ZEB1, ZEB2, SIP1, GLP1R, and Egr1) and signaling cascades (i.e., SMAD/TGF-b and PTEN/PI3K/AKT), together contributed to the pathogenesis of DN through epithelial-to-mesenchymal transition (EMT), extracellular matrix deposition, and fibrosis formation. The characteristics of eligible studies reporting miR-192 in DN were listed in Tables 1, 2. Figure 1 showed the main molecular mechanisms of miR-192 expression in the development of DN.  The protective effect of AC on DN might be associated with the reduction of the expression level of miR-192 and upregulation of ZEB2. (Continued)

Clinical significances of miR-192 in DN
Ten included clinical studies provided clinical information on miR-192 expression levels in DN. Most of these clinical trials (eight studies, 7/10, 70%) demonstrated that miR-192 was down-regulated in adult DN patients. Inconsistent with this finding, one included study (39) reported an opposite result that a high level of miR-192 in serum was found in pediatric DN patients as compared to the healthy controls. One study (31) showed that no significant difference was observed in the miR-192 expression between T2DM subjects with and without DN. Two studies reported the expression of miR-192 in urine. One study showed the results in both serum and urine.

miR-192 expression in the serum of patients with DN
Eight included studies investigated the expression level of miR-192 in venous blood collection (serum) of DN patients. Six out of eight studies (75%) reported the miR-192 expression was downregulated in patients with DN.
The pathogenesis of DN causes by various factors, such as high glucose and lipid metabolism disorder, which may induce extracellular matrix (ECM) accumulation, filtration membrane damage, and renal interstitial fibrosis (59). Transforming growth factor-b1 (TGF-b1) promotes the synthesis of ECM, prevents its degradation, and accumulates ECM by promoting adhesion between cells and matrix (60). Fibronectin (FN) is one of the key components of ECM, which can be applied to assess the extent of ECM accumulation. Ma et al. (33) recruited 464 patients and found that the miR-192 level in the DN group (especially in the albuminuria group) was significantly lower than the healthy controls (P< 0.05). The expression of miR-192 was negatively correlated with TGF-b1, FN, and urine albumin creatinine ratio (UACR). The authors concluded that there was a positive association between the low level of miR-192 and nephritic fibrosis in DN.
Histopathological examination is the gold standard for identifying multiple diseases. Krupa et al. reported that a low level of miR-192 was associated with the promotion of kidney fibrosis in DN (42). However, this finding was derived from kidney tissue samples, which was hard to implement for clinical practice. Mounting evidence shows that non-invasive sampling methods (i.e., serum and urine examination) are promising for the early detection of various diseases, including DN. Yang et al. (34) investigated 283 DN patients and found that the miR-192 level in serum significantly decreased in patients with DN, whereas in urine it increased with the progression of DN. Both of the differences were statistically significant. Therefore, the authors suggested that both serum and urinary miR-192 could be a potential biomarker of DN.
Though microalbuminuria was found to be an early sign of DN, some researchers wonder about its ability to precisely detect the disease progression of DN. So, it is necessary to find a diagnostic and predictive marker for DN. Given that circulating miR-192 might represent a potential novel source of non-invasive biomarkers for DN, more and more were designed for validating this finding. KAFAJI et al. (35) assessed 85 DN patients in Manama and found that lower miR-192 expression was observed in DN patients compared with healthy controls (P< 0.05). As a result, blood-based miR-192 was considered to serve as an effective biomarker for the early detection of DN. In line with this finding,  (34) reported that the miR-192 level in urine was increased, whereas in serum it decreased with the progression of DN (n=283). They further indicated that a combination with high levels of urine miR-192 and low levels of serum miR-192 had a higher specificity and lower misdiagnosis rate. However, Smith et al. (38) observed an opposite result that low miR-192 expression might be associated with the progression of DN. The authors detected that miR-192 expression fell from a 1.54-fold change in the control cohort in DN patients. MiR-192 was found to be specifically expressed in renal cortical tissue, and stored and excreted in urine as microvesicles (64). Therefore, kidney injury caused by DN might induce more microvesicles containing miR-192 through the urine. Nevertheless, since it is still being debated on the expression level of urinary miR-192 in DN patients, its diagnostic and predictive effects on DN still require further exploration.

Molecular mechanisms of miR-192 IN DN
The aforementioned clinical studies suggested a causal association between miR-192 expression and DN, exploring the biological function of miR-192 and its potential mechanisms in DN might be profound for the researchers. MiR-192 was downregulated in the serum of DN patients, which was reported in the majority of the clinical studies (6/8, 75%). However, the vast majority of experimental studies (14/18, 78%) demonstrated that miR-192 was up-regulated in a cell or animal model of DN (high glucose). Therefore, whether miR-192 played a protective or pathogenic role in the development of DN was still controversial among different studies. Since miRNAs function by interacting with their target genes, we summarized the 18 included in vitro and in vivo studies employing the different miR-192-targeted genes.

Roles of ZEB1 and ZEB2 underlie the effect of miR-192 in DN genesis
An essential role for EMT exists in renal interstitial fibrosis in DN (65). The formation and deposition of EMT and ECM in renal tubular epithelial cells attribute to the pathogenesis of DN. The physiological function of miR-192 was found to be closely correlated to EMT. It is known that Zinc finger E-box-binding homeobox (ZEB) plays a key role in EMT, which is closely related to many aspects of life (66). ZEB family mainly comprises two proteins ZEB1 and ZEB2. Two zinc finger clusters in ZEB1 bind to DNA sequences specific to its function. There are mainly binding sites in the zinc finger region of the N-terminal zinc finger or the zinc finger region of the C-terminal zinc finger. In the downstream region of the zinc finger of ZEB1, phosphorylated receptor-activated SMAD is bound, so ZEB1 could regulate the signaling pathway of TGFb (67). ZEB1 interacts with various miRNAs (i.e., miR-200c and miR-205), and together mediate the corresponding signaling pathways, such as TGFb, hippo pathway, and wingless/integrated (Wnt). It is known that EMT plays a key role in the development of fibrosis, while ZEB1 is an important transcription factor of EMT. During the process of EMT, ZEB1 effectively promotes cell proliferation, migration, collagen synthesis, and fibrosis formation (68).
Five included studies demonstrated that ZEB1/ZEB2 might be a direct target of miR-192. Deshpande et al. (45) conducted an experimental study in cells and animals and subsequently validated it in human kidney tissues. They found an increased expression level of miR-192 and p53 and a decreased level of ZEB2 (the direct target for miR-192). TGF-b could promote the transcriptional activation of p53 through miR-192. This study showed that the TGF-b1-induced feedback amplification circuit between p53 and miR-192/ZEB2 contributed greatly to the pathogenesis of DN. Similar to this finding, Chen et al. (53) observed that miR-192 was increased in glomerular mesangial cells cultured with high glucose. High glucose levels could regulate both ZEB1 and monocyte chemotactic protein-1 (MCP-1) expression by upregulating the level of miR-192. Though lacking a luciferase reporter gene examination, the expression of miR-192 significantly increased after the targeting silencing of ZEB1, indicating ZEB1 might be the target of miR-192. miR-192/ZEB1 was considered to involve in the occurrence of the inflammatory reaction in DN development. Consistently, Putta et al. (44) also detected a high level of miR-192 in the mice and cell models of DN. Zeb1/2 (miR-192 targeted genes) were decreased by means of the TGF-b signaling. They further found that locked nucleic acid (LNA)-modified inhibitor of miR-192 dramatically increased Zeb1/2 expression and reduced the level of collagen, TGF-b, and fibronectin, therefore alleviating DN. Consistently, Yu et al. (54) reported that miR-192 elevated in a mouse model of DN. They next found that metformin or alcohol extract of Coreopsis tinctoria Nutt (AC) significantly decreased the expression of miR-192, which could alleviate the degree of renal fibrosis. The protective effect of AC on DN might be associated with the reduction of the expression level of miR-192 and upregulation of miR-192-targeted gene ZEB2. This protective effect may be attributed to the indirect modulation of the activity of the PTEN/PI3K/AKT pathway. The above four studies confirmed that miR-192 was up-regulated under high glucose conditions. Contrary to the above three included studies, Krupa et al. (42) reported that miR-192 was low in both patient tissue and HK-2 cells of the DN model. The author observed that upregulated ZEB1/ZEB2 (potential targeted genes for miR-192) and TGF-b and decreased levels of E-cadherin in the pathogenesis of DN. The downregulated miR-192 level could increase fibrosis and decline GFR in DN, which might induce by enhancing TGF-bmediated downregulation of E-cadherin in proximal tubular cells. The above evidence indicated that ZEB1/ZEB2-associated EMT might contribute to the development of miR-21-mediated DN and the progression of miR-21-mediated DN.

Roles of SIP1 underlie the effect of miR-192 in DN genesis
Smad-interacting protein 1 (SIP1), one of the two-handed zinc-finger proteins and the transcriptional regulators for E-cadherin expression, can interact with activated SMAD transcriptional cofactors (69,70). Smads are the primary mediators of TGFb signaling, which modulates the activity of SIP1 as a transcriptional repressor. In the present review, three experimental studies identified SIP1 might be the direct target for miR-192. All three studies demonstrated that miR-192 was upregulated in numerous in-vivo and in vitro DN models. A previous study developed by Kato et al. (41) showed that SIP1 and dEF1 were down-regulated and Col1a2 was increased in DN mouse and mesangial cells with high glucose. In this study, the authors found that TGF-b led to the down-regulation of SIP1 (via miR-192) and dEF1 could cooperate to reinforce Col1a2 expression via derepression at E-box elements. dEF1 is an important inhibitor of E-cadherin. The cross-talk between E-box repressors (dEF1 and SIP1) contributes to the TGF-b-mediated collagen regulation in the pathogenesis of DN. Mao et al. (52) reported that astragaloside IV decreased miR−192, TGF-b1, Smad3, a-SMA, and collagen type 1, and elevated Smad7 in rats and RMCs cells model of DN. The therapeutic effect of Astragaloside IV on DN might be associated with the repression of excessive mesangial proliferation and renal fibrosis via the TGF-b1/Smad/miR-192 pathway. A more recent study (57) showed that TGF-b and miR-192 expression significantly elevated and SIP1 decreased in the DN group. Interestingly, the administration of troxerutin and insulin significantly reversed this tendency, indicating the renal-protective effects derived from these agents might be attributed to the inhibition of miR-192 expression and the elevation of SIP1 expression.

Other targeted genes of miR-192 in DN genesis
Glucagon-like peptide-1 receptor (GLP1R) exists in various organs, including the liver, brain, pancreas, gut, and hypothalamus (71). In addition, GLP1R was also found to be highly expressed in the kidney, especially in renal tubular epithelial cells (72). Mounting evidence demonstrates that GLP1R is a pivotal pharmacological target for T2DM (73). Thus, GLP1R was believed to regulate both non-diabetic and diabetic renal fibrosis (74). Jia et al. (49) conducted an in-vitro study on HK-2 cells in the high glucose condition. The authors indicated the level of miR-192 was down-regulated and p53 expression was upregulated. They found that exendin-4 downregulated cellular and secreted miR-192, therefore increasing the expression of GLP1R in a p53-dependent manner, which might ameliorate renal fibrosis. A subsequent in-vitro (HK-2) and in-vivo (rats) study (55) also suggested that miR-192-5p was elevated and its targeted gene GLP-1R was decreased in high-glucose-incubated human renal tubular epithelial cells and rat renal fibroblasts. The researchers next found that Icariin and dihydrotestosterone improved diabetic renal tubulointerstitial fibrosis by restoring autophagy via the miR-192-5p/GLP-1R pathway.
Early growth response factor 1 (Egr1), a transcription factor binding to DNA, has been found to abnormally express in diabetic mice (75). Besides, Egr1 is associated with the development of renal fibrosis (76). Multiple mechanisms are involved in the actions of Egr1 in diabetes mellitus-associated renal fibrosis, including elevating the expression of TGF-b, promoting the proliferation of mesangial cells, and accelerating the EMT process of renal tubular epithelial cells (77, 78). Thus, targeting with Egr1 might have promising effects for treating DN. Liu et al. (50) reported that miR-192 was down-regulated in both HK-2 cell and rat models of DN. Subsequently, the authors observed a tendency of high levels of Egr1 and low levels of TGF-b1 and FN in both cell and animal models. The biological effects exerted by miR-192 that could inhibit the progression of DN and protect DN rats from renal interstitial fibrosis might be mediated by decreasing TGF-b1 and FN. This study indicated that miR-192 could serve as an innovative and prospective therapeutic target for DN.

Perspectives
In this review, conflicting results were identified among the included studies about the expression of miR-192 among serum, urine, renal tissues samples. Based on the 28 eligible studies, the level of miR-192 tend towards decrease in serum but increase in urine and renal tissues under the disease condition of diabetic nephropathy. One of the probable explanations was that miR-192 might release from other tissue types thus changing the actual level. Numerous studies (79,80) indicated that serum miR-192 was downregulated in patients or animal models with diabetes mellitus. Consistently, we also found that serum miR-192 was decreased in diabetic nephropathy. It is known that the blood travel almost solely through the renal system. Therefore, the small molecules contained in the blood, like miR-192, may accumulate in the kidneys. Therefore, there is constant kidney accumulation of miR-192, leading to the high level of miR-192 in renal tissues and subsequently excrete with the urine. As a result, the expression of miR-192 may elevate in both kidney and the urine. On the other hand, miR-192 was found to be abundant in the renal glomerular and tubulointerstitial fibrosis. We speculated that renal fibrosis might encapsulate the small molecule "miR-192", leading to an accumulation of miR-192 in the renal tissues. As aforementioned, serum miR-192 was downregulated in patients or animal models with diabetes mellitus, while diabetic nephropathy was considered to the vascular complication of diabetes mellitus. Thus, low serum miR-192 may be the early manifestation of diabetic nephropathy. Gradually, however, an increase of miR-192 level in the renal tissues and urine was detected with increased loading of circulating miR-192. According to this assumption, higher expression of miR-192 in both renal tissues and urine may indicate the progression of diabetic nephropathy. According to the above hypotheses, low serum miR-192 might be helpful for early diagnosis of DN, while higher miR-192 level in renal tissues and urine may suggest the progression of DN.
The challenges with the current knowledge about miR-192 in DN may be the inconsistent results of the expression of miR-192 in serum, urine, and renal tissues samples among the different included studies. However, these conflicting results might cause by the different stages of DN. Low serum miR-192 could be applied for the early prediction of DN (the early stage of DN), while the high miR-192 level in renal tissues and urine may imply the progression of DN (the late stage of DN). This hypothesis should be investigated in near future, especially for the well-designed clinical trials.

Conclusion
To the best of our knowledge, this is the first review to summarize all the current evidence on the topic of the roles of miR-192 in DN. According to the 28 included studies, most of the clinical trials indicated miR-192 might be a protective factor for DN development and progression, while the majority of experimental studies suggested miR-192 might be a pathogenic factor for DN. Mechanistically, miR-192 interacts with various direct targeted proteins (i.e., ZEB1, ZEB2, SIP1, GLP1R, and Egr1) and signaling cascades, together contributing to the pathogenesis of DN. Since the results are conflicting and controversial between clinical trials and experimental studies, further investigations are still warranted to illustrate this inconsistent phenomenon. A clear understanding of the clinicopathological features and the molecular mechanisms may provide new insights into the therapeutic applications of miR-192 in predicting and treating DN. This review highlights the dual role of miR-192 in the pathogenesis of DN, which still needs more studies to verify its role in serving as a biomarker.
Author contributions XW, JL, and JC contributed to conceive and design the study. SZ performed the systematic searching. JL and HL extracted the data. XW and JC wrote the manuscript. XW and CC supervised the manuscript. All authors contributed to the article and approved the submitted version.