Identifying Key MicroRNAs Targeted by Narenmandula in a Rodent Nephropathy Model

Background Untreated nephropathy can progress to renal failure. The traditional Mongolian remedy Narenmandula regulates the kidney “yang.” This study aimed to identify key microRNAs (miRNAs) targeted by Narenmandula in a rat model of nephropathy. Methods Fifteen rats exhibiting normal renal function were randomized to three study arms. Nephropathy was induced in n = 10 rats using doxorubicin hydrochloride, followed by either Narenmandula treatment (treatment group) or no treatment (control group). In n = 5 rats, no doxorubicin was given and renal function remained unchanged (healthy group). Microarray analysis identified miRNAs which were differentially expressed (DE-miRNAs) between groups. Target genes of DE-miRNAs were predicted using miRWalk version 2.0, followed by enrichment analysis using DAVID, and construction of the miRNA coregulatory network using Cytoscape. Results Nephropathy was successfully induced, with doxorubicin resulting in differential expression of 3645 miRNAs (1324 upregulated and 2321 downregulated). Narenmandula treatment induced differential expression of a total of 159 miRNAs (102 upregulated and 57 downregulated). Upregulated DE-miRNAs (e.g., miR-497-5p, miR-195-5p, miR-181a-5p, miR-181c-5p, and miR-30e-5p) and downregulated DE-miRNAs (e.g., miR-330-3p and miR-214-3p) regulated a high number of target genes. Moreover, the miRNA pairs (e.g., miR-195-5p—miR-497-5p, miR-181a-5p—miR-181c-5p, and miR-30e-5p—miR-30a-5p) coregulated a high number of genes. Enrichment analysis indicated functional synergy between miR-30e-5p—miR-30a-3p, miR-34a-5p—miR-30e-5p, miR-30e-5p—miR-195-3p, and miR-30a-3p—miR-195-3p pairs. Conclusion Narenmandula may modulate doxorubicin-induced nephropathy via targeting miR-497-5p, miR-195-5p, miR-181a-5p, miR-181c-5p, miR-30e-5p, miR-330-3p, miR-214-3p, miR-34a-5p, miR-30a-3p, and miR-30a-5p.


Introduction
Chronic nephropathy is a progressive disorder, which when untreated can result in renal failure, including consequent uremia and death [1,2]. e main risk factors for nephropathy are glomerular immunoglobulin A (IgA) deposition, analgesic consumption, xanthine oxidase deficiency, and cytotoxic chemotherapeutic drugs [3]. Globally, nephropathy impacts quality of life in millions of patients, and thousands of these require life-saving renal transplantation [4]. A more comprehensive understanding of the pathogenesis of nephropathy will contribute to development of more effective therapies and improved outcomes for nephropathy patients.
In recent years, nephropathy mechanisms have been partially elucidated by several studies. For example, in reflux nephropathy (RN), angiotensin-converting enzyme (ACE) overexpression induces tubulointerstitial damage via increased extracellular matrix (ECM) component production; ACE inhibition may thus suppress renal fibrosis in this context [5]. By suppressing activation of the v-akt murine thymoma viral oncogene homolog (AKT)/phosphatase and tensin homolog (PTEN) pathway, miR-21 inhibition restrains tubular cell and podocyte fibrogenesis in IgA nephropathy (IgAN) [6,7]. Furthermore, miR-146a plays an anti-inflammatory role in the context of diabetic nephropathy (DN) [8], and miR-18a-5p modulates autophagy by regulating the ataxia telangiectasia mutated (ATM) gene, which may be prophylactic or therapeutic in the context of DN [9]. Moreover, miRNAs have exhibited utility as biomarkers and/ or therapeutic targets in the context of renal disorders such as IgA nephropathy and DN [10][11][12]. Nevertheless, pathogenesis of nephropathies remains incompletely understood. e traditional Mongolian remedy Narenmandula (also known as Sheng Yang Shi Yi Wei Wan) is a combination of pomegranate, cinnamon, cardamom, humble pie, yellow essence, safflower, wintergreen fruit, bamboo, asparagus, Bletilla, and Tribulus terrestris [13]. e major traditional uses of Narenmandula include regulating the kidney "yang", promoting capillary refill and digestion, and relieving diarrhea and edema [14]. Although polysaccharides present in Narenmandula exhibit antioxidant activity [15], potential mechanisms of Narenmandula in opposing nephropathy have not been investigated.
In the present study, doxorubicin hydrochloride was used to induce nephropathy in rats, followed by treatment with Narenmandula. Microarray-based miRNA expression profiling of healthy, treatment, and control groups was followed by differential expression analysis, prediction of genes targeted by differentially expressed miRNAs (DE-miRNAs), enrichment analysis, and miRNA coregulatory network construction. Results regarding key miRNAs modulated by Narenmandula may contribute to identifying novel therapeutic targets in the context of nephropathy.

Model Establishment and Sample
Acquisition. Study protocols were approved by the Ethical Committee of Affiliated Hospital of Inner Mongolia University for the Nationalities, Tongliao, Inner Mongolia Autonomous of China. Rats were obtained from Yisi Laboratory Animal Technology Co., Ltd. (Jilin, Changchun, China). After a 5-day acclimation period, 15 rats exhibiting normal total 24 h urine protein content were selected for randomization to a healthy group (receiving 6.5 mL/kg normal saline), treatment group (receiving 6.5 mg/kg doxorubicin hydrochloride, Shenzhen Arcandor's Pharmaceutical Co., Ltd., Guangdong, Shenzhen, China) followed after a 4-day period by 3.0 g Narenmandula (Tong Kang Pharmaceutical Co., Ltd., Hebei, Anguo, China) once per day for 21 consecutive days), and a control group (receiving doxorubicin followed by distilled water rather than Narenmandula). Each group contained n � 5 rats. Saline and doxorubicin were administered intravenously by injection into the tail caudal vein, while Narenmandula was administered intragastrically. After the final dose of Narenmandula or water, all rats were fasted for 12 h prior to anesthetization with pentobarbital. Animals were then euthanized via abdominal aortic phlebotomy.

Microarray Analysis and Data
Preprocessing. An miRNA 4.0 Array (Affymetrix, Santa Clara, CA) was used to determine whole blood miRNA expression profiles of rats in each group. Expression data have been deposited in the Gene Expression Omnibus (GEO) database (accession number GSE123776). e Robust Multichip Average (RMA) algorithm in R package "affy" [16] was used in conjunction with CEL result files to conduct background correction (integrating probe and probeset signals) and normalization (minimizing the effect of between-sample biological variability).

Model Validation: Comparing Healthy and Treatment
Groups. R package "limma" [17] was used to determine whether any miRNAs were differentially expressed between the healthy and treatment groups, using as screening criteria |log 2 fold-change (FC)| > 0.58 and p value ≤0.05. Rodent nephropathy model success was defined as the presence of significant DE-miRNAs.

Target Gene Enrichment
Analysis. Using DAVID version 6.8 (with classification stringency set to medium) [31], Gene Ontology (GO) [32] and Kyoto Encyclopedia of Genes and Genomes (KEGG) [33] terms were predicted for target genes of miRNAs differentially expressed between treatment and control groups. e threshold for significance was set at p value <0.05.

Construction of the miRNA Coregulatory Network.
Based on shared targets of miRNA pairs, an miRNA coregulatory network was constructed using Cytoscape version 3.4 [34]. For miRNA pairs coregulating the highest number of target genes, GO enrichment analysis was conducted. e threshold for significance was set at p value <0.05. Functional synergy was defined as significant enrichment of biological process (BP) terms between common targets of an miRNA pair [35].

Model Validation: Comparing Healthy and Treatment
Groups. When comparing healthy and treatment groups, 3645 miRNAs were significantly differentially expressed (1324 upregulated and 2321 downregulated in the treatment group; Figure 1), suggesting successful induction of nephropathy.

Comparing Treatment and Control Groups and Prediction of Target Genes of Identified DE-miRNAs.
When comparing treatment and control groups, 159 miRNAs were significantly differentially expressed (102 upregulated and 57 downregulated in the treatment group; Figure 2). Based on FC, the top ten most upregulated and downregulated DE-miRNAs are shown in Table 1.

Target Gene Enrichment Analysis.
Biological process and KEGG biological pathway enrichment analysis of genes targeted by the top ten most significantly up-and downregulated DE-miRNAs ( Figure 3) revealed that genes targeted by upregulated miRNAs are mainly involved in response to hormone stimulus (GO BP) and ubiquitinmediated proteolysis (KEGG pathway) (Figure 3(a)) and that genes targeted by downregulated miRNAs are mainly involved in the phosphorus metabolic process (GO BP) and Wnt signaling pathway (KEGG pathway) (Figure 3(b)).
Melatonin is known to alleviate endothelial-to-mesenchymal transition (EMT) of glomerular endothelial cells during DN by altering expression of miR-497 [36] (upregulated by Narenmandula in the present study). Overexpression of miR-497 can significantly repress proliferation, migration, and invasiveness of renal cancer cells and may serve as a prognostic factor and therapeutic target for this tumor type [37]. e present study demonstrates upregulation of miR-195 by Narenmandula. Decreased miR-195 expression prevents mesangial cell apoptosis, and miR-195 may also exert antiapoptotic effects during early DN [38]. e immunosuppressive drug cyclosporine A (CsA) is nephrotoxic, producing renal injury and fibrosis, but miR-181c (upregulated by Narenmandula) may help protect renal tissues against this [39]. However, miR-181a inhibition alleviates 5-fluorouracil-(5-FU-) associated nephrotoxicity and may represent a promising target for treatment of chemotherapy-induced nephrotoxicity [40]. Furthermore, miR-181a overexpression contributes to clear cell renal cell carcinoma (ccRCC) progression via modulating Krüppel-like factor 6 (KLF6) expression; targeting miR-181a may have therapeutic potential in this context [41,42]. Also upregulated by Narenmandula in the present study, miR-30e exhibits low-level expression in DN, and overexpression can prevent progression of DN to renal fibrosis [43]. We speculate that miR-497-5p, miR-195-5p, miR-181a-5p, miR-181c-5p, and miR-30e-5p may be implicated in protective effects of Narenmandula in the context of nephropathy.

Data Availability
Datasets generated by this study are available from the Gene Expression Omnibus (GEO) database (accession number GSE123776).

Conflicts of Interest
e authors declare that there are no conflicts of interest.