The authors have declared that no competing interests exist.
Conceived and designed the experiments: KZ CYZ JWY. Performed the experiments: NW YZ LJ. Analyzed the data: DHL. Wrote the paper: KZ.
The steadily increasing incidence of kidney injury is a significant threat to human health. The current tools available for the early detection of kidney injury, however, have limited sensitivity or specificity. Thus, the development of novel biomarkers to detect early kidney injury is of high importance. Employing mouse renal ischemia-reperfusion and streptozotocin (STZ)-induced renal injury as acute and chronic kidney injury model, respectively, we assessed the alteration of microRNA (miRNA) in mouse urine, serum and kidney tissue by TaqMan probe-based qRT-PCR assay. Our results demonstrated that kidney-enriched microRNA-10a (miR-10a) and microRNA-30d (miR-30d) were readily detected in mouse urine and the levels of urinary miR-10a and miR-30d were positively correlated with the degree of kidney injury induced by renal ischemia-reperfusion or STZ diabetes. In contrast, no such alteration of miR-10a and miR-30d levels was observed in mouse serum after kidney injury. Compared with the blood urea nitrogen (BUN) assay, the test for urinary miR-10a and miR-30d levels was more sensitive for the detection of acute kidney injury. Furthermore, the substantial elevation of the urinary miR-10a and miR-30d levels was also observed in focal segmental glomerulosclerosis (FSGS) patients compared to healthy donors. In conclusion, the present study collectively demonstrates that urinary miR-10a and miR-30d represent a novel noninvasive, sensitive, specific and potentially high-throughput method for detecting renal injury.
The incidence of kidney injuries, including diabetic nephropathy and drug-induced side effects, is steadily increasing worldwide
MicroRNAs (miRNAs) are ∼22-nt long noncoding RNA molecules that have emerged as a new class of gene regulators at the posttranscriptional level. It is widely believed that miRNAs regulate nearly 30% of protein-coding genes and are involved in almost every aspect of developmental, pathogenic and tumorigenesis processes. Accumulating evidence demonstrates that miRNAs serve as mediators in chronic kidney disease
In the present study, we hypothesized that the levels of specific circulating miRNA species could be used to detect and monitor the pathological development associated with kidney injuries. Using different mouse renal injury models, we reported that miR-10a and miR-30d were readily detected in urine and that their levels specifically correlated with mouse kidney injury induced by renal ischemia-reperfusion or STZ treatment. The elevation of the urinary levels of miR-10a and miR-30d was also confirmed in urine samples from patients with focal segmental glomerulosclerosis (FSGS).
All animal experimental procedures were carried out in accordance with the National Institutes of Health Guide for the Care and Use of Laboratory Animals and were approved by the Animal Care Committee of Nanjing University (Nanjing, China).
Male C57BL/6J mice (6–8 weeks old, 22–25 g) were assigned into three groups with ten mice per group: the sham, unilateral and bilateral ischemia-reperfusion (I/R) groups. Renal I/R was performed according to an established procedure
Renal function was monitored by measuring the BUN concentration using analytical kits according to the protocols specified by the manufacturer (Rongsheng Biotech, Shanghai, China). Mouse blood samples were collected when mice were sacrificed. Samples were centrifuged to separate plasma.
Renal histology was examined by hematoxylin/eosin staining. Briefly, kidneys were collected and fixed in 10% phosphate-buffered formalin at 4°C overnight. The kidneys were subsequently paraffin embedded and sectioned (4 µm). Sections were stained with hematoxylin/eosin for general histology. Images of representative fields were recorded.
Total RNA was extracted from 100 µL samples (serum or urine). In brief, 100 µL sample was mixed with 300 µL diethylpyrocarbonate-treated water, 200 µL acid phenol, 200 µL chloroform. The mixture was vortex-mixed vigorously and centrifuged at 12000 g for 15 min at room temperature. After phase separation, the aqueous layer (∼400 µL) was mixed with 2 volumes of isopropyl alcohol and 0.1 volumes of 3 mol/L sodium acetate (pH 5.3). This solution was stored at −20°C for 1 h. The RNA pellet was collected by centrifugation at 16000 g for 20 min at 4°C. The resulting RNA pellet was washed once with 750 mL/L ethanol and air-dried for 10 min at room temperature. Finally, the pellet was dissolved in 20 µL of ribonuclease-free water.
Kidney injury was established in STZ-induced diabetic mice as previously described
The present study was approved by the Institutional Review Board of Nanjing University (Nanjing, China) and written informed consent was obtained from each participant. Urine and blood samples (10–15 mL) were collected from 16 healthy donors and 16 FSGS patients (
Clinical features | Normal (Mean ± SEM) | FSGS (Mean ± SEM) |
Age (year) | 29.1±3.7 | 26.6±2.9 |
DD (month) | - | 4.78±0.81 |
BP (mmHg) | 128.5±3.1/79.6±3.3 | 135.1±1.5/84.5±2.6 |
Upro (g/24 h) | 0.27±0.12 | 6.86±1.03 |
URBC (×104) | 2.78±1.06 | 6.81±2.32 |
Scr (mg/dl) | 0.86±0.13 | 0.80±0.04 |
BUN (mg/dl) | 14.6±1.61 | 15.3±1.3 |
Ccr (ml/min) | 108.7±9.79 | 90.3±15.83 |
TG (mmol/l) | 1.47±0.63 | 3.39±0.76 |
TC (mmol/l) | 4.38±1.64 | 9.83±1.10 |
DD: Duration of the disease; BP: Blood pressure; Upro: Urinary protein; URBC: Urinary red blood cell; Scr: Serum creatinine; BUN: Blood urea nitrogen; Ccr: Creatinine clearance. TG: Triglyceride; TC: Total cholesterol.
, p<0.05;
, p<0.01.
Total RNA including miRNA was isolated from various tissues of C57BL/6J mice and mouse or human urine and sera using TRIzol reagent (Invitrogen, CA) as previously described
For each experiment, qRT-PCR assays were performed in triplicate. The data shown are represented as means ± SEM for at least three independent experiments. Differences were considered statistically significant at p<0.05, as assessed using Student's
The role of miRNA in kidney function has been illustrated previously by investigators through the specific knockdown of Dicer activity in podocyte cells
RNA from various organs of normal C57BL6 mice was extracted, and the miRNA levels were determined by qRT-PCR assay using U6 as an internal control.
Based on the tissue distribution of miRNAs determined by qRT-PCR assays, miR-10a and miR-30d were selected as kidney-specific miRNAs for further analysis. Next, we tested whether miR-10a and miR-30d are released into animal urine under normal and injury conditions. In this experiment, we employed unilateral and bilateral renal ischemia/reperfusion (I/R) in mice as a kidney injury model. Urine samples from normal male C57BL/6J mice (6–8 weeks old, 22–25 g) and male C57BL/6J mice with kidney injuries were collected, and absolute levels of miR-10a and miR-30d were assessed. As shown in
Conversely, we found that miRNAs were normally expressed in mouse urine at 0.1–15.0 fmol/L range and were readily detected by the qRT-PCR assay. More importantly, the levels of urinary miR-10a and miR-30d were significantly increased in mice with either unilateral ischemia/reperfusion or bilateral ischemia/reperfusion. Compared with the sham control group, unilateral and bilateral ischemia/reperfusion kidney resulted in 5.5- and 19.5-fold elevations of urinary miR-10a, respectively (
Previous studies by our group and others have demonstrated that the alteration of the serum miRNA expression profile can be used as a molecular fingerprint for various diseases
Next we determined the levels of miR-10a and miR-30d in mouse kidney tissue with or without renal I/R. As shown in
Chronic diabetic conditions generally lead to kidney injury
Note that the levels of urinary miR-10a (
Kidney/Body weight(mg·g−1) | Blood glucose(mmol·L−1) | 24 h proteinuria(mg) | |
Control | 9.862±1.034 | 6.488±0.568 | 0.290±0.141 |
STZ (1 month) | 11.463±1.165 |
26.321±5.304 |
0.541±0.276 |
STZ (2 months) | 13.025±1.094 |
27.274±4.841 |
3.274±0.604 |
Compared to the control group,
To find out whether the elevation of urinary miR-10a and miR-30d also occurs in patient with kidney injuries, we assessed the levels of urinary miR-10a and miR-30d in FSGS patients. Renal dysfunction of FSGS patients was also monitored by measuring blood pressure, BUN, serum creatinine, uric acid, urinary protein, urinary red blood cell, serum creatinine, creatinine clearance, triglyceride and total cholesterol levels. As shown in
Urine samples were collected from 16 healthy donors and 16 FSGS patients. Note that, compared with healthy donors, FSGS patients had strikingly elevated levels of urinary miR-10a and miR-30d. **, p<0.01.
Urine is regarded as a source for biomarkers, given its easy availability and reduced complexity when compared with serum. Many urinary proteins, such as neutrophil gelatinase-associated lipocalin (NGAL)
To serve as a biomarker, urinary miRNAs must be present at a considerable concentration to be readily detected. Our study clearly demonstrated that urinary miRNAs are stable and can be reliably extracted and assayed by qRT-PCR. By comparing the levels of miRNA in sera and urine, we found that kidney-enriched miRNAs, such as miR-10a and miR-30d, were present in urine, and their concentrations were approximately 1/10 of those in sera. Importantly, when kidney injury occurred, the levels of miR-10a and miR-30d in urine were strikingly elevated, while their levels in the serum were not increased. These results strongly suggest that urinary miR-10a and miR-30d can serve as ideal biomarkers for kidney injury. We used both I/R-induced acute kidney injury and STZ diabetes-induced chronic kidney injury animal models and showed that changes in the levels of urinary miR-10a and miR-30d occurred as a result of renal damage. Therefore, the relative concentration of kidney-enriched miRNA species has a potential to become a sensitive indicator for detecting and monitoring kidney injury (
Because miR-10a and miR-30d are enriched in kidney tissue (
We demonstrated using well-established models that kidney injuries can be precisely detected and monitored using a small number of circulating miRNA species. In our study, this miRNA-based method was more sensitive and probably more reliable than the current BUN method for detecting kidney injury. Elevation of the urinary miR-10a and miR-30d levels can be detected in mice with unilateral I/R in which the protein levels were not changed, suggesting that the urinary miR-10a and miR-30d levels can reflect mild or early kidney injury. In addition, urinary miR-10a and miR-30d are highly enriched to the kidney; therefore, the elevation of these miRNAs may be directly linked to the injuries of kidney. This hypothesis is supported by our observation that the elevation of miR-10a and miR-30d concentrations occurred only in urine and not in serum when mice were treated with renal I/R. Interestingly, recent study by Lorenzen et al.
The role of tissue miR-10a and miR-30d in kidney function also strengthens our conclusion that urinary miR-10a and miR-30d can serve as indicators for kidney injury. Previous studies have shown that miR-10a can target IL-12/IL-23p40 expression
Interestingly, although the chronic hyperglycemia caused an elevation of urinary miR-10a and miR-30d likely due to the kidney damage, a short period of high blood glucose exposure did not increase the level of these kidney-specific miRNAs in urine. By challenging 12 h–fasting mice with an intraperitoneal injection of glucose (2 g/kg of body weight), we found no elevation of urinary miR-10a and miR-30d within 1–3 h (data not shown). These results may suggest that the increase in urine miRNA levels in the diabetes mouse model is not due to a glycosuria-induced osmotic diuresis.
The results for the human urine samples further confirmed the feasibility of using the urinary miR-10a and miR-30d levels to detect kidney injury in humans. As shown in
(TIF)
(TIF)
(TIF)
Relative enrichment of miRNAs in various mouse tissues detected by Solexa sequencing. Nine mouse organs including kidney, liver, spleen, brain, intestine, lung, etc. were assayed. After normalization, the miRNAs with total counts from in nine mouse organs >2000 were selected. The ratio of miRNAs in other tissues such as lung, brain, etc. was not shown due to the limited space.
(DOC)