Background
Chronic kidney disease (CKD) represents a major public health problem and is an important cause of global mortality [
1]. Studies have documented the high prevalence of CKD worldwide [
2‐
4]. Patients with CKD are at increased risk of end-stage renal disease (ESRD), cardiovascular disease and death [
5]. Tubulointerstitial injury, such as interstitial fibrosis (IF) and tubular atrophy (TA), an inevitable consequence of various CKD, tightly correlates with CKD progression [
6‐
10]. However, assessment of IF/TA is only achievable with renal biopsies, which are taken from patients by an invasive procedure. Developing a noninvasive surrogate biomarker that can monitor the development of IF/TA and predict the progression of CKD is urgently needed.
MicroRNAs (miRNAs), a class of ~ 22-nt non-coding RNAs, play important roles in various physiologic and pathologic processes through post-transcriptional gene silencing [
11]. Recent studies have demonstrated that miRNAs in urine, serum, plasma and other body fluids can serve as useful biomarkers for various diseases and tissue injuries [
12‐
16]. Studies by our group and others have demonstrated the potential of miRNAs in urine or plasma as indicators to evaluate renal ischemia–reperfusion injury [
17], glomerular disease activity [
18], post-transplant renal graft function [
19,
20], glomerular injury [
21] and renal fibrosis [
22]. Our previous study also showed a positive correlation between urinary miR-196a level and disease activity of focal segmental glomerulosclerosis (FSGS) [
23], a common cause of CKD and ESRD. In addition, our recent study showed that miR-196a is predominantly expressed in kidney, with 74.3% of total mouse miR-196a being distributed in kidneys and plays an essential role in renal fibrosis. Kidney miR-196a was significantly decreased in unilateral ureteral obstruction mouse model and in patients with renal fibrosis, while elevating renal miR-196a level strongly attenuated renal fibrosis. Further study showed that miR-196a prevented renal fibrosis via suppressing the activity of transforming growth factor-beta/Smad signaling pathway [
24]. Studies by other groups showed that miR-196a is involved in tissue fibrosis by regulating the expression of type I collagen and NFκB inflammatory signaling pathway [
25‐
27]. We postulated that miR-196a level in urine may serve as a noninvasive biomarker for predicting disease progression in CKD patients.
In the present study, FSGS cohorts were used as the representative example of CKD. To illustrate the origin of urinary miR-196a, we measured the paired urinary and plasma miR-196a levels simultaneously from 100 each of FSGS patients with nephrotic-range proteinuria (FSGS-A), complete remission (FSGS-CR) and normal controls (NCs). To prove that urinary miR-196a is a biomarker associated with kidney injury and can be used for risk stratification of ESRD and predicting disease progression, we assessed the relationship of urinary miR-196a level with IF/TA and the risk of progression to ESRD using another separate cohort consisting of 231 patients with FSGS. Our study has identified that urinary miR-196a is a kidney-derived, and injury related biomarker for predicting the progression of CKD.
Discussion
In the present study, we have shown that urinary miR-196a is a kidney-derived biomarker for renal injury. The negative correlation between intrarenal miR-196a and IF/TA observed in the biopsy specimens from FSGS patients also supports that urinary miR-196a is a renal injury marker. Our results further demonstrate that urinary miR-196a may serve as a biomarker for predicting the disease progression in patients with FSGS.
We previously found that FSGS-A patients had an elevation of urinary miR-196a level compared to FSGS-CR patients or healthy donors [
23]. In addition, in the steroid-responsive patients, urinary miR-196a decreased to normal levels after steroid treatment. However, the origin of urinary miR-196a was unknown. In order to identify the source of urinary miR-196a, we measured both urinary and plasma levels of miR-196a using paired urine and plasma samples from the same FSGS-A patients, FSGS-CR patients and NCs. The correlations of urinary and plasma miR-196a levels with the disease activity of FSGS were respectively analyzed. Our results showed that urinary miR-196a can serve as a biomarker for the discrimination of FSGS-A patients from FSGS-CR patients and NCs, which is in agreement with our previous study. In contrast, plasma miR-196a levels showed no difference among FSGS-A patients, FSGS-CR patients and NCs, demonstrating that plasma miR-196a level is not associated with the disease activity and that elevation of urinary miR-196a level in FSGS-A patients is not derived from circulation but mainly from kidney tissue because we have previously shown that miR-196a is predominantly expressed in kidney [
24].
Renal fibrosis is a hallmark of CKD progression. Previous studies by others and us have demonstrated that miR-196a is a kidney-enriched miRNA and plays an important role in tissue fibrosis [
24‐
26,
38]. Downregulation of miR-196a lead to tissue fibrosis through directly upregulating the expression of type I collagen; conversely, overexpression of miR-196a resulted in the downregulation of type I collagen in fibroblasts [
25,
26]. Moreover, expression levels of miR-196a was shown to affect the activity of NFκB inflammatory signaling pathway via direct interaction with its target genes IKKα and IKKβ in fibroblasts [
27]. Our previous study has also shown that miR-196a is expressed in glomerular, tubular, as well as infiltrated inflammatory cells and that renal tubular miR-196a can mitigate renal fibrosis via targeting transforming growth factor-β receptor 2 [
24]. In the present study, we examined the expression of intrarenal miR-196a and explored the correlation of intrarenal miR-196a and renal fibrosis in FSGS patients. The results confirmed the downregulation of miR-196a in kidney tissue of FSGS patients, which was consistent with the results from a mouse model [
24]. In addition, our results showed that intrarenal miR-196a was negatively correlated with the severity of renal fibrosis.
Previous studies have showed tubular cell loss, inflammatory cells infiltration, fibroblast proliferation, inflammatory pathways activation in the process of renal fibrosis [
39‐
41]. It is also known that the expression of miR-196a in inflammatory cells is much lower than that in renal tubular cells [
24] and that miR-196a is downregulated in fibroblast cells when inflammatory pathways are activated [
27]. These may contribute to the decrease of intrarenal miR-196a during renal fibrosis. Considering that urinary miR-196a was mainly derived from kidney, we presumed that increase of urinary miR-196a was due to passive leakage of renal miR-196a into urine from injured tubular cells or active secretion by renal cells via microvesicles [
42,
43] or a microvesicle-free RNA-binding protein-dependent pathway [
44]. Supportively, several studies have demonstrated that, under certain stimulation, miRNAs are present at higher concentrations in extracellular fluids although their expression is decreased in cells or tissues. For example, studies have reported that expression of miR-10b, miR-125b and miR-145 are decreased in breast cancer but elevated in serum of the patients compared to healthy controls [
45‐
47]. It was also shown that miR-24, miR-29a, miR-150 and miR-222 were downregulated in THP-1 cells in response to the stimulus of oleic acid/palmitic acid; however, the levels of these miRNAs in the supernatant of THP-1 cells were increased [
48]. Although the exact molecular basis of the elevation of urine miR-196a during renal injury remains unclear, detection of urinary miR-196a provides us a non-invasive method to monitor renal fibrosis and predict disease progression in patients with CKD.
To assess the predictive value of urinary miR-196a on risk stratification of ESRD and disease progression of FSGS patients, we set the progression to ESRD or 40% eGFR reduction as composite end points, which reflect the current concept of clinical trial end points in CKD [
29]. The cohort of patients in this part had relatively long prospective follow-up time periods (at least 36 months for those patients who did not progress to ESRD). We found that urinary miR-196a levels at baseline can distinguish patients who progressed to ESRD from those who maintained a stable renal function. We also analyzed the correlations of urinary miR-196a level with various clinical parameters and observed significant correlations of urinary miR-196a level with proteinuria and eGFR. More interestingly, we found that urinary miR-196a level significantly correlated with renal fibrosis. Patients with severe renal fibrosis had higher urinary miR-196a levels compared to those with less severe renal fibrosis. We next performed Kaplan–Meier curve and determined that those in a higher urinary miR-196a tertile were at higher risk of progression to ESRD. Furthermore, by multivariate Cox regression analysis, we found a significant correlation between the tertiles of urinary miR-196a and renal outcome even after adjustment for traditional risk factors age, sex, proteinuria and eGFR. Urinary miR-196a as a continuous variable also remained significantly associated with progression to ESRD after adjusted for traditional risk factors. The strategy had been used to analyze other biomarkers such as soluble urokinase-type plasminogen activator receptor, growth differentiation factor-15 and endotrophin, which were used to predict CKD progression [
35‐
37].
More and more studies have shown that biomarker panels can significantly improved prediction accuracy compared with a single biomarker [
49‐
51]. We found that combining urinary miR-196a level with the above-mentioned standard risk factors (proteinuria and eGFR) resulted in substantially improved predictive power for renal outcome with significantly higher c-statistic and lower AIC. Time-dependent ROC analysis also showed that addition of urinary miR-196a was superior to proteinuria and eGFR in combination in terms of predictive power for renal outcome. Moreover, miR-196a is kidney-enriched and participates in the process of renal fibrosis, urinary miR-196a can be therefore considered to be a mechanism-based biomarker for risk stratification and predicting disease progression in patients with CKD.
miR-196a has also been shown to play a role in other kidney diseases. Wang et al. [
52] showed that miR-196a is downregulated markedly in mice with diabetic nephropathy (DN) thereby regulating high glucose-induced mesangial cell hypertrophy by targeting p27
kipI. Our previous study showed that urinary miR-196a levels are associated with DN and membranous nephropathy (MN) [
23]. These results suggest that urinary miR-196a might also have a prognostic value in patients with DN and MN. Of course, such a postulation needs to be further validated.
Several previous studies have explored the potentials of miRNA levels as biomarkers for CKD progression. Wang et al. showed that the levels of miR-192 in kidney and miR-200b in urine, two microRNAs that are involved in CKD [
53,
54], are significantly correlated with the rate of eGFR decline in patients with IgA nephropathy [
55,
56]. miR-21 and miR-216a are also known to play a role in CKD development [
57,
58]. Szeto et al. found that urinary miR-21 and miR-216a expression correlated with the rate of eGFR decline and the risk of progression to ESRD in CKD patients [
59]. These results indicate that miRNAs involved in CKD development also have the potential to be a prognostic biomarkers of CKD. However, there were several limitations in these studies, including (1) multivariate analysis was not conducted to determine the prognostic value of these miRNAs given the small sample size in these studies; (2) the authors did not test whether the miRNAs quantification could provide additional prognostic accuracy after adjusting for the degree of proteinuria; and (3) FSGS patients were not included in these studies. Therefore, the prognostic potential for CKD progression of these miRNAs warrants further studies. Urinary miR-196a dysregulation is common in several types of CKD (FSGS, DN and MN), thus providing a new promising prognostic option for CKD.
On the other hand, there are a few limitations in this study. First, our findings were based on Chinese patients with FSGS and it may be difficult to generalize our results to other races. Second, the present study focused on a single miRNA level in urine and the predictive value of urinary miR-196a level in other kidney diseases remains unknown.
Authors’ contributions
Research conception and study design: ZL and CmZ; Research and data analysis: CmZ, WL, and LX; Manuscript preparation: CmZ, SS, KZ, and ZL; Clinical data collection: CmZ, SC, and XW; Pathologic score: SL and CZ; Research sample preparation: CmZ, and CxZ. All authors have read the journal’s authorship agreement and agree with the statements. Each author contributed important intellectual content during manuscript drafting or revision and accepts accountability for the overall work by ensuring that questions pertaining to the accuracy or integrity of any portion of the work are appropriately investigated and resolved. All authors read and approved the final manuscript.