Urinary miR-16 transactivated by C/EBPβ reduces kidney function after ischemia/reperfusion–induced injury

Ischemia-reperfusion (I/R) induced acute kidney injury (AKI) is regulated by transcriptional factors and microRNAs (miRs). However, modulation of miRs by transcriptional factors has not been characterized in AKI. Here, we found that urinary miR-16 was 100-fold higher in AKI patients. MiR-16 was detected earlier than creatinine in mouse after I/R. Using TargetScan, the 3′UTR of B-cell lymphoma 2 (BCL-2) was found complementary to miR-16 to decrease the fluorescent reporter activity. Overexpression of miR-16 in mice significantly attenuated renal function and increased TUNEL activity in epithelium tubule cells. The CCAAT enhancer binding protein beta (C/EBP-β) increased the expression of miR-16 after I/R injury. The ChIP and luciferase promoter assay indicated that about −1.0 kb to −0.5 kb upstream of miR-16 genome promoter region containing C/EBP-β binding motif transcriptionally regulated miR-16 expression. Meanwhile, the level of pri-miR-16 was higher in mice infected with lentivirus containing C/EBP-β compared with wild-type (WT) mice and overexpression of C/EBP-β in the kidney of WT mice reduced kidney function, increased kidney apoptosis, and elevated urinary miR-16 level. Our results indicated that miR-16 was transactivated by C/EBP-β resulting in aggravated I/R induced AKI and that urinary miR-16 may serve as a potential biomarker for AKI.

Patients and urine collection. A total of 18 human serum and urine samples were obtained immediately when patients were admitted to the Tzu Chi General Hospital (Hualien, Taiwan). These samples were collected from October 2013 to August 2014. This research project was approved by the Institutional Review Board of the Tzu Chi General Hospital and informed consents were obtained from all patients then samples were collected in accordance with the approved protocols and guidelines. The samples were collected from eleven critical patients who developed AKI (acute kidney injury), defined as ≧1.5 fold increase in serum creatinine in compliance with RIFLE-AKIN criteria. ICU control samples were collected from seven critical patients who did not develop AKI. The relevant information of these patients was listed in Table 1. In addition, serum and urine samples were obtained from six healthy volunteers. Urine samples of 25-35 ml were collected from the participating subjects. All samples were frozen at − 80 °C until use. Quantitative qPCR and reverse transcriptase-PCR. Quantitation of the relative mRNA and miRNA abundance was performed using Applied Biosystems Step One Plus (Applied Biosystems, Foster City, USA). For the detection of miR-16 expression, we used TaqMan MicroRNA assays (Life Technologies) using small nuclear U6B (RNU6B) RNA as an internal standard. To quantify the amount of mRNA (BCL-2 and CEBP-β ) and pri-miR-16, we used TaqMan Gene Expression Assay using GAPDH as an internal control. Samples were tested in triplicate, and differences of threshold cycles between target genes and house-keeping genes (GAPDH in mRNA and U6 in miRNA) were calculated by the 2 −ΔΔCT method using a control group as the calibrator according to the manufacturer's user manual. . 293T Cells (4 × 106) were seeded into 10-cm 2 culture dishes in 5.5 ml of the medium and transfected the following day with 2 μ g of pMD-G plasmid, 8 μ g of pCMV8.9 plasmid, and 12 μ g of Lenti-miR16 vector plasmid (GeneCopoeia, MD, USA), antisense-miR-16 (SBI CA, USA) or Lenti-CEBβ . The medium was collected on day 2 and 3 post transfection and concentrated using the Vivapure LentiSELECT40 Kit (Sartorius Stedim Biotech, Aubagne Cedex, France).

Whole-mount
Intrarenal pelvic injection of lentivirus. The procedure was as described previously 14 . Mice were anesthetized with intraperitoneal pentobarbital (50 mg/kg). The renal artery, renal vein, and ureter were clamped at the same time just below the renal pelvis before transfection. Recombinant lentivirus, or PBS was slowly injected into the left renal artery with the use of a 30-G needle, subsequently the needle was removed and the ureter was declamped. After 2 weeks, mice were euthanized, and the kidneys were removed and homogenized for designated experiments.
Histopathology. Mouse kidneys were fixed in 10% buffered formalin overnight at 4 °C and processed with paraffin fixation. Sections were stained with hematoxylin and eosin. Apoptosis in renal tissues was identified using the TUNEL (terminal deoxynucleotidyl transferase-mediated dUTP-biotin nick end labeling) assay with an ApopTag In Situ Apoptosis Detection kit (S7160, Chemicon, Darmstadt, Germany) and counterstained with DAPI (SouthernBiotech, AL, USA) following the manufacturer's instructions.

Measurement of biochemical parameters.
At the end of reperfusion, 500-μ l blood samples were collected via the tail vein. Samples were centrifuged at 6000 × g for 3 min to separate the serum from the cells. Biochemical parameters were measured in serum within 24 hrs.
Chromatin immunoprecipitation assay. One day after lenti-pSin or lenti-hCEBPβ transfection, 293T cells were incubated under conditions of normoxia or hypoxia (1% O 2 ) for 4 hrs followed by 4 hrs of reoxygenation. The cells were fixed in 1% formaldehyde, and the ChIP assay was conducted using the Upstate protocol (Millipore). Chromatin was immunoprecipitated with anti-CEBPβ antibody (Santa Cruz Biotechnology). The purified DNA was detected using standard PCR with the following primer pairs for the miR16 promoter region, 5′-GGTCCAACAGATAATTTACCCAACAAGGC-3′ (forward) and 5′-CCACCGCGTGGAGCCCTATAAAG-3′ (reverse).

Statistical analysis.
Values are expressed as means ± SEM from at least three experiments. The statistical significance was analyzed using ANOVA followed by the Tukey test for the in vivo experiments. Samples from the patients were analyzed using the Rank Sum test. A value of P < 0.05 was considered statistically significant.

Results
Increased urinary miR-16 concentration reflected AKI condition. Previous study has shown that urinary miR-494 can be used as an indicator for AKI 14 . Therefore, we explored the possibility if new miRs could become alternative indicators for AKI. MiRs array profiling of urine samples from normal and AKI patients were compared. Figure 1A shows that 30 miRNAs were significantly modulated by more than two-fold, within these seven miRNAs were upregulated (let-7d, life-26-3p, miR-16, miR-451, miR-486-5p, miR-518e* , miR-720) and twenty one were downregulated. We were particularly interested in miR-16, a previously reported tumor suppressor in leukemia 19 , which showed the largest increase in AKI patients compared to normal subjects and also revealed consistent result with real time polymerase chain reaction (qPCR) (Fig. 1B). Furthermore, urine and serum samples were collected from the other seven patients in intensive care unit (ICU) with AKI, seven ICU patients without AKI, and four healthy volunteers. The serum miR-16 levels among the healthy volunteers, ICU patients with and without AKI did not differ significantly (Fig. 1C). However, the urinary level of miR-16 of ICU patients with AKI was significantly higher than those without AKI and normal volunteers (Fig. 1D, left). These results were consistent with the known urinary AKI marker NGAL (Fig. 1D, right). Thus, increased urinary miR-16 level reflects the condition of AKI.
Urinary miR-16 was induced earlier than traditional kidney injury marker, urea or creatinine, in a mouse I/R renal failure model. To delineate whether urinary miR-16 is related to the kidney, the content of miR-16 in various mouse organs was measured. Using semi-quantitative qPCR, we found that miR-16 was expressed highly in the heart, moderately in the kidneys, testes and lung, and lowest in the liver, brain and spleen ( Fig. 2A). Our previous study indicated that the appearance of urinary miRs may result from kidney injury like I/R 14 , therefore, I/R mouse model was used. Figure 2B shows that serum urea and creatinine level were increased after ischemia followed by reperfusion for 3 hrs, however, urinary miR-16 was significantly increased after reperfusion for 1 hr. The elevated urinary miR-16 appeared earlier than the elevated serum urea and creatinine (Fig. 2C). These results were consistent with the known urinary AKI marker NGAL (Fig. 2D). Dicer and associated miRs have been reported to involve in I/R injury in the kidney 20 , and I/R injury is a major cause of AKI meaning miR-16 may come from immature miR-16, pri-miR-16, after I/R injury. Figure 2E (left) indicates that pri-miR-16 was rapidly declined from 1 hr lasting to 6 hrs after reperfusion, in contrast, the level of miR-16 was increased from 1 hr to 6 hrs after reperfusion in mice (Fig. 2E right). In situ hybridization also revealed that the expression of miR-16 was predominantly within the tubular epithelial cytosol after reperfusion for 3 hrs (Fig. 2F).

Role of miR-16 in I/R-induced renal function and apoptotic response.
To further address the role of miR-16 in I/R injury, we compared the renal function (as assessed by blood urea nitrogen and creatinine) of mice with or without overexpression of either miR-16 or antisense-miR-16 in the presence or absence of renal I/R. As shown in Fig. 4A, overexpression of miR-16 significantly increased the mRNA level of miR-16 in the kidney without or with I/R compared to lentivirus GFP injection alone and overexpression of miR-16 also induced renal dysfunction (Fig. 4B), increased numbers of apoptotic TUNEL-positive tubular epithelial cells (Fig. 4C), elevated activity of the cleaved, active form of caspase-3 ( Fig. 4D). High expression level of miR-16 was observed only in acute kidney injury mice (Fig. 2E,F). Thus, overexpression of miR-16 can induce kidney dysfunction which is accompanied by elevated miR-16 level in the urine. Figure 4E shows that significantly higher level of urinary miR-16 was detected in mice infused with lentivirus containing miR-16 compared to lenti-pSin controls after I/R. In contrast, overexpression of antisense-miR-16 in mice attenuated I/R-induced renal dysfunction compared to lenti-pSin controls (Fig. 4F). Likewise, urinary miR-16 level was reduced by in vivo lentivirus-mediated antisense-miR-16 gene transfer into the kidneys (Fig. 4G). These results suggest that miR-16 plays important pathophysiological roles in the kidneys after I/R and urinary miR-16 level may be a biomarker for kidney injury. gene (Fig. 5A). To evaluate pri-miR16 level, 293T cells were transfected with ATF3, PPARα -RXRα , NF-κ B, C/EBP-β plasmid separately, and only overexpression of NF-κ B and C/EBP-β permitted evaluation of pri-miR-16 transcription level without H/R (Supplemental Fig. S1 and Fig. 5C I). To further confirm the relationship between C/EBP-β and miR-16 after H/R, both C/EBP-β and miR-16 were all increased in 293T cells after 24 hrs of reoxygention (Fig. 5B). After H/R, the pri-miR-16 level was dramatically reduced accompanied by an increase of miR-16 in 293T cells transfected with C/EBP-β compared to pcDNA3 control (Fig. 5C I,II). In addition, the target gene's expression level of miR-16, BCL-2, was inhibited after H/R under C/EBP-β overexpression (Fig. 5C III). In addition, to examine whether the C/EBP-β protein was located in the nucleus and associated with the C/EBP-β binding region, a ChIP assay was performed. Figure 5D indicates that overexpression of C/EBP-β increased the amount of C/EBP-β protein recruited to the miR-16 promoter region compared to the scrambled control. To further determine the activity of C/EBP-β -induced pri-miR-16 transcription on the upstream of miR-16 genome promoter, serious deletion of miR-16 genome promoters were constructed (− 2.0, − 1.0 and − 0.5 kb) and assayed for luciferase activity in 293T cells with or without C/EBP-β treatment. Figure 5E shows that significantly reduced luciferase activity was observed in the miR-16 promoter − 1.0 to − 0.5 kb region, but not in the − 0.5 to − 0.0 kb region (Fig. 5E). These results suggest that C/EBP-β interacts with the miR-16 regulatory region.

MiR-16 was necessary for C/EBP-β to block BCL-2 and to induce kidney epithelium cell apoptosis resulting in kidney dysfunction.
Pri-miR-16 is transcriptionally regulated directly by C/EBP-β and targets BCL-2 by the in vitro assay. Therefore, whether C/EBP-β is expressed in the kidney during I/R, meaning that C/EBP-β blocks BCL-2 protein expression through upregulation of pri-miR-16 and that this is necessary for kidney dysfunction homeostasis, was investigated. qPCR indicated that C/EBP-β mRNA levels were significantly increased 1 hr after reperfusion and lasted for 48 hrs (Fig. 6A). Consistent with qPCR, immunohistochemistry revealed that increased translocation of C/EBP-β into nucleus was found 1 hr after reperfusion (showing purple staining) on both tubular epithelial and glomerular cells (Fig. 6B). Whether the pathophysiological role of C/EBP-β induced miR-16 in I/R injury is via inhibition of the BCL-2 pathway, the gain-of-function of  lentivirus-mediated gene transfer was used to overexpress C/EBP-β in the kidneys of sham or I/R mice. As shown in Fig. 7A,B, both C/EBP-β and pri-miR-16 expression were increased in the kidney tissue infused with lentivirus containing C/EBP-β compared to the lenti-pSin (self-inactivating vector) indicating that miR-16 was indeed transcriptionally regulated by C/EBP-β . Furthermore, overexpression of C/EBP-β significantly increased I/R-induced renal dysfunction compared to without overexpression and the sham group (Fig. 7C). Enhanced renal dysfunction with overexpression of C/EBP-β was accompanied by increased numbers of apoptotic TUNEL-positive tubular epithelial cells, elevated activity of the cleaved, active form of caspase-3 and urinary miR-16 level (Fig. 7D,E). These results suggest that inhibition of anti-apoptotic gene BCL-2 by C/EBP-β induced miR-16 pathway together with increasing urinary miR-16 level may play important pathophysiological roles in the kidneys after I/R.

Discussion
MiRs exert a prominent homeostatic plasticity in the kidney, such as tubular epithelium cells apoptosis, calcium handling 22 , podocyte function 23 and renin-producing juxtaglomerular cells 24 , that are regulated by transcriptional  factors 25 but not including C/EBP family. Previous studies have shown that C/EBP-β is an important regulator in adipocyte differentiation, metabolism syndrome, ER stress and inflammatory cascade 26 , but studies of miRs regulation in the kidney have not been reported. Our results demonstrated a novel mechanism for C/EBP-β to regulate miR-16 in the kidney pathophysiology. Normally, C/EBP-β presents in the cytosol, and after I/R, it translocates into the nucleus and binds to the promoter region of pri-miR-16 genome resulting in the elevated level of pri-miR-16 in the nucleus and its subsequent translocation into the cytosol. After I/R, increasing amount of mature miR-16 are cleaved from pri-miR-16 by Dicer, which then inhibit one of the anti-apoptotic protein, BCL-2, leading to severe kidney injury (Fig. 8). We also showed that miR-16 was expressed at higher levels in the urines of AKI patients in the ICU compared to patients without AKI or normal individuals. In addition, urinary miR-16 is an earlier and non-invasive indicator compared to creatinine. Most miR-16 studies have been focused on cancer and conflicting results have been reported. In breast cancer and esophageal squamous cell carcinoma, miR-16 suppressed cell apoptosis while promoted growth by regulating RECK, SOX6 or RPS6KB1 27,28 . But in gastric adenocarcinoma 29 and human nasopharyngeal carcinoma 30 , miR-16 played an inhibitory role in cancer activation, suggesting that miR-16 targets different genes on various organs resulting in differential clinical outcomes. In our study on miR-16 targeted genes, we mainly focused on apoptosis signaling pathway gene, BCL-2. However, using integrative analysis of targetome (1,195 targets), TGF-β , PI3K-Akt, p53, GnRH, MAPK and ubiquitin mediated proteolysis signaling pathway were also found in the miR-16 regulatory network 31 . In addition, renal cell carcinoma related signaling pathways molecules, like PI3kinase, VEGFR2, VEGF, EIF4E and RAS, all are miR-16 target genes 32 . Because anti-apoptotic effects of PI3kinase, VEGFR2, VEGF, EIF4E, RAS may result from its anti-apoptosis or anti-oxidative effect after I/R, inhibiting these genes by miR-16 would result in greater apoptosis and hence larger damage to the kidneys. Therefore, these miR-16 related targeted genes may also involve in I/R induced nephropathy. Our in vitro H/R experiments revealed that elevated miR-16 level was accompanied by a decrease in BCL-2 protein expression (Fig. 3D right panel). However, BCL-2 mRNA expression rebounded at 24 hrs after reoxygenation (Fig. 3C,D left panel). We suggested that except for miR-16 there are many miRs like miR-15b, 302b, 497 may also will influence BCL-2 level in vitro 33,34 ,whether BCL-2 be influenced by these miRs after H/R in vitro will be evaluated later. In addition, in Fig. 3E transfected with shRNA of miR-16 only in vitro did not effect on BCL-2 expression may also be effected by theses miRs except for miR-16. The other possibility suggested that up-regulation of Bcl-2 mRNA transcription but not link protein level after reoxygenation 24 hrs which may be regulated by some factors such as Cyclic AMP (cAMP) Response Element Binding Protein (CREB) and cAMP-Responsive CREB Coactivator-2 35 . But the results both from in vitro overexpression with lenti-miR-16 (Fig. 3E) and mice infused with shRNA-miR-16 (Fig. 4) definitely demonstrated the inhibition effect of miR-16 to BCL-2 and protected kidney function effect from I/R suggestion miR-16 actually plays an important pathophysiology role in kidney.
Many circulating miRs have been screened out for diagnosis and prognosis of various kinds of cancers including lung, colorectal, ovarian, pancreatic and other cancers 36 . It has been shown that miR-16-5p may be a prospective biomarker for gastric cancer and its progression by the variation of its plasma level 37 . However, most of these miR-16 studies failed to determine whether it is related to kidney dysfunction. In our study, results from clinical patients definitely indicated that up-expression of urinary miR-16 may be served as a biomarker for AKI patients. MiRs in the urine have been described as possible biomarkers for hepatotoxicity 38 , acute myocardial infarction 39 or AKI 14 . In addition, Mall et al. showed that miR-16 and miR-21 are relatively stable in the urine under a variety of storage conditions, which supports their utility as urinary biomarkers 40 . We also demonstrated that miR-16 can be detected by using capped gold nanoslit SPR in a microfluidic chip after freeze-throw four times (Supplemental Fig. S2).
Lorenzenz et al. also demonstrated that the levels of circulating miRs including miR-16, 24, 1244, 620, 320, 30d, let 7f and let 7b all were upregulated in patients with AKI compared to healthy controls or patients with acute myocardial infarction (non-AKI) 41 . High level of urinary miR-16 in AKI patients may come from the breakdown of tubular cells mediated by decreased BCL-2, resulting in apoptosis or necrosis 42 , which may explain why high level of urinary miR-16 were found in our clinical AKI patients (Fig. 1A,B). Whether other miRs, including 24, 1244, 620, 320, 30d, let 7f and let 7b, are all elevated in the urine of AKI patients require further study. The other possibility of high urinary miR-16 may result from increased miR-16 expression in the podocytes, which perturbs the actin cytoskeleton, and increases the release of exosomes containing miR-16 as previously reported 43 (Fig. 8). Finally, Collino et al. showed that mesenchymal stromal cell-derived extracellular vesicles, also called exosome, carrying miRs or mRNA can repair AKI induced kidney injury 44 . Whether urinary miR-16 may exist as extracellular vesicle form called microparticle miR-16 (Fig. 8) and inhibit BCL-2 activity by endocytosis will be evaluated later.
Promoter analysis indicated that upstream genome region of miR-16 contains NFκ -β , PPARα -RXRα , ATF3/CRE and C/REBP-β binding sites, but except for NFκ -β and C/REBP-β , the other transcription factors failed to induce miR-16 (Supplemental Fig. S1). In the present study, we identified miR-16 as a novel direct target of C/EBP-β in tubular epithelium cells (Fig. 5). High expression of miR-16 regulated by NFκ -β in gastric cancer has been observed 45 , therefore, we cannot rule out the possibility that miR-16 regulated by NFκ -β may target BCL-2 during kidney I/R. Whether epigenetic modification of C/REBP-β is involved in miR-16 transcription, previous emerging evidence indicates that C/REBP-β can activate EF2-regulated gene by recruiting coactivator p300 46 and acetylation by p300 resulting in C/EBP-β -mediated IL-6 and TGF-1 expression 47 . In this study, we demonstrated that miR-16 was regulated transcriptionally by C/EBP-β and whether other miRs are regulated by C/EBP-β resulting in kidney dysfunction will be evaluated by comparing the microRNA array analysis from the knock down of C/EBP-β in 293T cells after H/R 24 hours.
Our findings provide new insights for the C/EBPβ -mediated microRNA induced kidney dysfunction progression, which is the key step to disrupt renal epithelium cells resulting in elevated urinary miR-16 level. In this report, we show that C/EBP-β upregulates miR-16, and miR-16 blocks one of the anti-apoptotic genes, BCL-2, after I/R. We cannot rule out the possibility of miR-16 targeting other proapoptotic and antiapoptotic genes like programmed cell death 11 (PDCD11) and SOCS6. To our knowledge, this is the first report that shows the mechanism for urinary miR-16 levels enhancement by C/EBP-β after I/R in the kidney. Overexpression of epigenetic C/EBP-β by lentivirus can increase urinary miR-16 after I/R in the kidney. Finally, urinary miR-16 is stable and expressed earlier than creatinine, which makes it a useful indicator for AKI patients.