A circulating exosomal microRNA panel as a novel biomarker for monitoring post‐transplant renal graft function

Abstract Accurate and effective biomarkers for continuous monitoring of graft function are needed after kidney transplantation. The aim of this study was to establish a circulating exosomal miRNA panel as non‐invasive biomarker for kidney transplant recipients. Plasma exosomes of 58 kidney transplant recipients and 27 healthy controls were extracted by gel exclusion chromatography and characterized by transmission electron microscopy, nanoparticle tracking analysis and Western blotting. Post‐transplant renal graft function was evaluated by estimated glomerular filtration rate (eGFR). Quantitative real‐time polymerase chain reaction was used to determine the expression of exosomal microRNAs (miRNAs). Exosomal miR‐21, miR‐210 and miR‐4639 showed negative correlations with eGFR in the training set and were selected for further analysis. In the validation set, miR‐21, miR‐210 and miR‐4639 showed the capability to discriminate between subjects with chronic allograft dysfunction (eGFR < 60 mL/min/1.73 m2) and those with normal graft function (eGFR > 90 mL/min/1.73 m2). Three‐miRNA panel exhibited higher accuracy compared with individual miRNAs or double indicators. One‐year follow‐up revealed a stable recovery of allograft function in subjects with low calculated score from three‐miRNA panel (below the optimal cut‐off value). In conclusion, a unique circulating exosomal miRNA panel was identified as an effective biomarker for monitoring post‐transplant renal graft function in this study.

standard test is histological diagnosis with a renal transplant biopsy. 6 However, there are some disadvantages of low specificity and sensitivity or invasiveness during the evaluation. Cr is derived from the non-enzymatic dehydration of skeletal muscle creatine, which itself is generated from amino acids in the liver. 7 Thus, numerous factors such as muscle mass and turnover, sex, diet, race, liver function and medication use can influence serum Cr concentration. 8 In kidney transplant recipients, serum Cr concentration can be affected due to the long-term use of corticosteroids, infection, acute rejection and previous prolonged haemodialysis therapy. 7 As a result, the Cr-based eGFR estimation equation is also flawed, and the evaluation equation itself is not perfect. 9 Moreover, proteinuria can be affected by exercise and diet. 10 Kidney biopsies are considered to be the gold standard for evaluating allograft dysfunction. However, renal biopsy cannot be used to monitor the progression of injury because it is an invasive procedure and cannot be performed serially. 11 Furthermore, the histological evaluation of biopsies is subjective and samples removed from one segment of the transplanted kidney may not represent the whole graft. 12 Therefore, it is necessary to find a sensitive and non-invasive biomarker for the continuous monitoring of graft function after kidney transplantation.

MicroRNAs (miRNAs) are a group of small non-coding RNAs that
can regulate up to 60% of gene expression in mammals by binding to the 3′ untranslated region (3′-UTR) of the target messenger RNA (mRNA) involved in many diseases. 13 Exosomes are small (40-160 nm) membrane vesicles of endocytic origin that are released into the extracellular environment on fusion of multivesicular bodies with the plasma membrane. 14,15 Exosomal miRNAs are proved to be stably expressed in serum, plasma, urine, saliva and other body fluids. 16 Many studies have indicated that levels of exosomal miR-NAs are associated with renal function. For instance, the differential expression of five miRNAs (miR-32, miR-107, miR-142-3p, miR-204 and miR-211) in patients with chronic allograft dysfunction was confirmed by using an independent set of kidney tissue samples and paired urine samples. 17 Moreover, five miRNAs (miR-200b, miR-375, miR-423-5p, miR-193b and miR-345) were identified as potential biomarkers for monitoring allograft function in the urine samples of renal transplant recipients. 18 However, the role of circulating exosomal miRNAs in the monitoring of post-transplant renal graft function has not been fully figured out.
In this study, we examined correlations between exosomal miRNA levels and eGFR in cohorts of kidney transplant recipients and healthy controls. A circulating exosomal miRNA panel was established as the non-invasive biomarker for monitoring of post-transplant renal graft function in the 1-year follow-up. The flow chart for the study design is illustrated in Figure 1.

| Patients and samples
A total of 58 kidney transplant recipients and 27 healthy controls were enrolled in the study (Figure 1). Participants were enrolled between January 2017 and October 2018 in Third Hospital of Soochow University (Changzhou, China). No living donors, HIV-positive patients and/or re-transplant patients were included. The allograft function was evaluated by eGFR, which is calculated as the "CKD-EPI equation". 19 For the screening of 12 exosomal miRNAs and the validation of three exosomal miRNAs, patients' plasma samples were collected 3 months after renal transplantation (at study entry). In the follow-up study, plasma samples were collected at months 3, 6 and F I G U R E 1 An overview of the experimental design 12 after study entry. Two millilitre blood samples were collected from all patients and healthy controls without breakfast in the early morning. Within 2 hours, plasma separation was accomplished by centrifugation at 3200 g for 5 minutes to completely remove cell debris. The supernatant plasma was collected and stored at −80°C until analysis. The study was approved by the ethics committee of Soochow University. Written informed consent was obtained from all patients. Basic characteristics of participants are shown in Table 1.

| Exosome extraction
Plasma exosomes were extracted by gel exclusion chromatography (Exo-spin™; Cell Guidance Systems), strictly in accordance with kit instructions. Briefly, 200 μL plasma was centrifuged at 20 000 g for 30 minutes to remove cell debris. Supernatant was transferred to a new centrifuge tube and ½ volume of Exo-spin™ buffer was added.
After incubating at 4°C for at least 1 hour, the mixture was centrifuged at 20 000 g for 1 hour. Plasma exosomes were re-suspended in 100 μL phosphate-buffered saline (PBS), transferred to the top of the Exo-spin column and centrifuged at 50 g for 60 seconds. Eluate was discarded, and additional 200 µL PBS was added to the top of the column. The purified plasma exosomes were harvested in the eluate by centrifuging at 50 g for 60 seconds.

| Characterization of plasma exosomes
Plasma exosomes were applied to 200-mesh nickel grids and precipitated for several minutes. Samples were stained with 2% phosphotungstic acid for 1 minute. After drying at room temperature for several minutes, exosomes were imaged by a transmission electron microscope (H-7650; Hitachi High-Tech) at 80 kV. The particle size analysis of exosomes was detected by Nanoparticle Tracking System

| RNA isolation and quantification
Exosomal miRNAs were extracted using TRIzol LS reagent The qRT-PCR assay was conducted by TB Green™ Premix Ex Taq™ (Takara Biomedical Technology). qRT-PCR was conducted on ABI 7500 system (Applied Biosystems). The miR-16 expression was used as endogenous control because it is consistently expressed in exosomes from plasma samples. Relative miRNA expression was calculated by the 2 −ΔC t method in which ΔC t was calculated as C t (miRNA of interest) − C t (reference gene). Data are expressed as the mean ± SD, number (percentage) or median (10%-90% percentiles) when appropriate. Correlations between variables were calculated using Spearman's rank-order correlations, and the diagnostic performance of biomarkers was evaluated by ROC curves. All P-values were two-tailed and P < .05 was considered to indicate a statistically significant difference.  Note: Data were presented as the mean (SD).

| Correlations between exosomal miRNA levels and eGFR
As illustrated in Figure 1  Note: P-value <.05 was defined as statistically significant and showed in bold values.
Based on the analyses of the training set, three different exosomal miRNAs (miR-21-5p, miR-210-3p and miR-4639-5p) correlated significantly with eGFR (Table 2). Thus, these three miRNAs were further examined by qRT-PCR in a larger cohort of validation set including 36 kidney transplant recipients and 17 matched healthy controls.
Consistent with the results from the training set, miR-21-5p, miR-210-3p and miR-4639-5p were found to be correlated with eGFR.
To evaluate whether these three selected exosomal miR-

| Establishment of a predictive diagnostic miRNA panel and longitudinal eGFR analysis
MiR-21-5p, miR-210-3p and miR-4639-5p were combined into panels to further evaluate their diagnostic potential for renal function.
Logistic regression model was applied to combine exosomal miRNAs into two-miRNA panels or three-miRNA panel with the samples from eGFR < 60 and eGFR ≥ 90 (mL/min/1.73 m 2 ) groups. The optimal cut-off values, AUC, 95% confidence intervals (CI), sensitivities and specificities for each analysis were summarized in Table 3. Pearson's correlation coefficient (rho) is shown. eGFR, estimated glomerular filtration rate. ****P < .0001 The group 1 was defined as those with a score above the optimal cut-off value (>0.43, n = 18), and the group 2 was defined as those with a score below the optimal cut-off value (<0.43, n = 11) ( Figure 5B). Then, the changing rates of eGFR levels were compared between the two groups in the following 12 months. The eGFR level of each individual was collected at the time-point of 3, 6 and 12 months during follow-up. We found that individuals in group 2 (calculated score <cut-off) had significantly elevated eGFR levels compared with those in group 1 (calculated score >cut-off) ( Figure 5C). Our longitudinal analysis implied that the score of three-miRNA panel may predict future eGFR recovery and the improvement of post-transplant renal graft function.

| D ISCUSS I ON
In the past 20 years, with the application of highly effective immunosuppressive drugs, major progress has been made in extending graft and patients' survival after kidney transplantation. Nevertheless, long-term graft survival is still suboptimal due to both immunologic and non-immunologic factors, including ischaemia/reperfusion injury, untreated or ineffective clinical and subclinical rejection, nephrotoxicity of calcineurin inhibitors and existed donor diseases. 3,4 Therefore, it is imperative to investigate specific and non-invasive biomarkers for continuous monitoring post-transplant renal graft function, which may help to predict disease progression and determine therapeutic strategies.
Exosomes are tiny vesicles released from cells and widely found in body fluids such as blood, urine and saliva. 16 Accumulating evidence has demonstrated that exosomes contain a large number of molecules including protein, lipids, mRNAs and miRNAs. 15,21 These molecules carry a large amount of intracellular biological information that is closely related to disease status. 14,22 Due to complete membrane structures, exosomes are less disturbed by the external environment and carry small molecules with good stability. 23 Thus, molecules in exosomes such as miRNAs can be referred as non-invasive from those with eGFR ≥ 90 (mL/min/1.73 m 2 ). eGFR, estimated glomerular filtration rate; ROC curve, receiver operating characteristic curve; AUC, area under the curve. *P < .05, **P < .01, ***P < .001 biomarkers for the detection of renal diseases. 16   (mL/min/1.73 m 2 ) were divided into two groups according to the optimal cut-off value of 3-miRNA panel. The change of eGFR levels in the following 12 mo was presented (mean ± SEM). eGFR, estimated glomerular filtration rate; ROC curve, receiver operating characteristic curve; AUC, area under the curve; *P < .05 injury. 37 These results were consistent with our observations that patients with impaired renal function tended to have elevated miR-21 expression in plasma exosomes.
MiR-210 is another hypoxia/ischaemia-associated miRNA that regulates cellular events in the kidney by targeting multiple genes. Lorenzen et al showed that miR-210 level was strongly altered in urine of the patients with acute renal allograft rejection. 38 Deregulated miR-210 level was associated with higher decline in GFR after 1-year transplantation. 38 It is also reported that circulating miR-210 could predict survival in critically ill patients with acute kidney injury, 39 indicating a clinical application of miR-210 in disease monitoring. MiR-4639 is a newly discovered miRNA that participates in cellular oxidative stress responses. 40 MiR-4639 is enriched in exosomes of human plasma that may facilitate biomarker discovery.
In the present study, the combination of three miRNAs as a panel exhibited a better diagnostic potential compared with individual miRNA or two-miRNA panels ( Table 3 Similarly, urinary exosomes are mostly derived from cells in contact with the renal tubule lumen, such as renal tubular epithelial cells. 43 Thus, urine is also an appropriate non-invasive biofluid for exosomal studies. However, by deep sequencing analysis, Lesley Cheng et al found that the number and abundance of miRNAs in cell-free urine exosomes were significantly lower than those in plasma exosomes. 44 Only 12 miRNAs were abundantly expressed from 2.5 mL of cell-free urine, while 1 mL of plasma can contain more than 500 high-abundance miRNAs. 45,46 This phenomenon may be related to high RNase activity in the bladder. 47 Moreover, miRNAs in plasma exosomes can be accurate quantified by controlling sample volume and detecting miRNA internal control gene. However, the amount of exosomes in urine can be easily affected by the water intake of the body. Due to the difference in the composition of morning urine samples and urine samples at other time-points, morning urine samples should be obtained for analysis, which may bring inconvenience.
Therefore, in this study, plasma was used as the source for isolating exosomal miRNAs for further analysis.
The exosomal miRNA panel proposed in this study has some advantages comparing to the existing methods. First, examine exosomes isolated from plasma is a non-invasive procedure. Second,

| CON CLUS IONS
In summary, this work revealed that miR-21, miR-210 and miR-

ACK N OWLED G EM ENTS
The authors thank the patients who contributed to this study.

CO N FLI C T O F I NTE R E S T
The authors confirm that there are no conflicts of interest. The authors had full access to all of the data in this study and take complete responsibility for the integrity of the data and the accuracy of the data analysis.

E TH I C A L A PPROVA L
The study was approved by the ethics committee of Soochow University. Informed consents were obtained from all participants.

DATA AVA I L A B I L I T Y S TAT E M E N T
The data that support the findings of this study are available from the corresponding author upon reasonable request.