Eighty-four male C57BL/6 mice (6–8 weeks old, body weight 20–22 g; Chengdu Dossy Experimental Animals, Co. Ltd., Chengdu, Sichuan Province, China) were raised following standard guidelines. The Animal Ethics Committee of West China Hospital, Sichuan University, Chengdu, China (no. 2019141A) approved this study. All experimental procedures followed the National Institutes of Health Guide for the Care and Use of Laboratory Animals.

For the first experiment, mice were randomly divided into sham, ischemia (I), ischemia and cold preservation (IC), and IRI groups. For the I group (n = 6), the bilateral kidney pedicles were clamped simultaneously with a noninvasive artery clamp for 30 min, and then the kidneys were removed for further experiments. For the IC group (n = 9), the clamps were removed after 30 min of ischemia, and then kidneys were removed and cold preserved in 4 °C UW preservation solution (6, 12, and 24 h, respectively) for the follow-up experiments. For the IRI model (n = 9), the clamps were removed after ischemia and reperfusion were performed for 24, 48, or 72 h, and then mice were euthanized to obtain the kidneys. The kidney pedicles were not clamped for the sham group (n = 6), and the other operations were the same as in the I group.

For the second experiment, Cy3-labeled miR-92a agomir or agomir NC (10 μL/g body weight, GenePharma) were injected through the tail vein of mice 24 h before modeling (n = 6 for each group), and all mice were sacrificed 24 h after modeling. The miR-92a agomir and agomir NC were synthesized by GenePharma Co. Ltd. (Shanghai, China). For the in vivo test, a lipofectamine–agomir complex was prepared by mixing miRNA agomir (20 nmol) with lipofectamine 2000 (200 μL) and injected into mice via the tail vein (10 μL/g body weight).

For the third experiment, the mice were subdivided into the following seven groups (n = 6 for each group): sham group; IC + agomir NC (IC–NC); IRI + NC (IR–NC); use agomir before ischemia in IC group (ag before IC); use agomir before ischemia in IRI group (ag before IR); use agomir before cold preservation in IC group (ag during IC); and use agomir after IRI in IRI group (ag after IR). Mice in the first five groups were given agomir or agomir NC 24 h before and sacrificed after modeling. In the ag during IC group, kidneys were removed after 30 min ischemia and preserved in 4 ℃ UW preservation solution containing agomir for 6 h. In the ag after IR group, agomir was given after ischemia. To dynamically observe the effect of agomir on renal function, serum creatinine, and urea nitrogen of mice in the sham group, IR-NC group, ag before IR, and ag after IR were measured at 48 h, 3 d, and 7 d after modeling. Blood and kidney tissues of mice were collected for the following molecular and pathological examination. The level of serum creatinine (Scr) was detected by the creatine oxidase method (Creatinine Assay Kit, Nanjing Jiancheng Bioengineering Institute, Nanjing, Jiangsu, China), and the urea nitrogen was measured by the urease method (C013-2, Nanjing Jiancheng Bioengineering Institute).

HK-2 cells were purchased from the American Type Culture Collection (ATCC, CRL-2190). HK-2 cells were cultured in serum-free DMEM/F12 medium (1552510, Gibco, Thermo Fisher Scientific Inc. Waltham, MA, USA). After synchronization, the cells were divided into three groups: control group (Con), hypoxia group (Hypo), hypoxia and reoxygenation group (Reo 6, 12, and 24 h). First, HK-2 cells were placed in an oxygen deprivation chamber (Billups-Rothenberg), and 95% N2 and 5% CO2 were mixed to induce hypoxia. The HK-2 cells were then incubated in a cell incubator at 37 ℃ for 24 h. HK-2 cells in the Reo groups were restored to normal oxygen conditions (complete culture base plus aerobic 6, 12, and 24 h) after hypoxia. To explore the effects of miR92-a on HK-2 apoptosis and autophagy under hypoxia conditions, agomir NC, miR-92a agomir (25 nM), and miR-92a antagomir (50 nM) synthesized by GenePharma Co. Ltd. (Shanghai, China), were used to intervene in the Con, Hypo, and Reo group cells, respectively.

The histopathological changes in the kidneys were detected by hematoxylin–eosin (H&E) staining. The Pathology Laboratory of West China Hospital (Sichuan University, Chengdu, Sichuan Province, China) assisted us in processing pathological sections. Three slices of each mouse were statistically analyzed at ×400 magnification.

The microstructure of kidneys was observed by transmission electron microscope (TEM). The kidney slice (approximately 0.5 × 0.5 cm) was fixed in 3.0% glutaraldehyde and sent to Chengdu Li Lai Biotechnology Co. Ltd (Chengdu, Sichuan Province, China) for analysis. A JEM-1400PLUS transmission electron microscope (JEOL Ltd., Tokyo, Japan) was used for observation.

Terminal deoxynucleotidyl transferase-mediated dUTP nick end labeling (TUNEL, Roche, 11684795910) was used to detect the apoptosis of kidneys. Briefly, paraffin-embedded sections were routinely dewaxed and hydrated. Next, these slides were permeabilized by proteinase K solution. After washing with phosphate buffer solution, all slides were refixed and equilibrated with equilibration buffer for 5–10 min. Then, these slices were labeled with TdT reaction mix in a dark and moist box for 60 min. The slice was placed in 2× Saline Sodium Citrate buffer (SSC) to stop the reaction, and 4′,6-diamidino-2-phenylindole (DAPI) nuclear stain was added in the mounting medium, and analyzed. Localized green fluorescence of apoptotic cells was detected in the blue background by fluorescence microscopy (ZEN 2012 Lite, ZEISS).

The apoptosis of HK-2 cells was detected by flow cytometry. Cells in each group were digested and collected with trypsin (15050065, Gibco, Thermo Fisher Scientific Inc.) without EDTA. Then, the cells were resuspended with 100 μL binding buffer after washing with phosphate buffer solution (PBS), and 3 μL Annexin V and 3 μL PI (Annexin V-FITC staining kit, Beyotime Biotechnology Co. Ltd., Haimen, Jiangsu, China) were added then incubated for 20 min at room temperature. The cells were collected by centrifugation, resuscitated by PBS, and then detected by flow cytometry.

Total ribonucleic acid (RNA) of the kidneys was isolated using Trizol Reagent (G3013, Wuhan Servicebio Technology Co. Ltd., Wuhan, Hubei, China) and reverse transcribed to cDNA with the Revert Aid First Strand cDNA Synthesis Kit (no. K1622, Thermo Fisher Scientific Inc.). Real-time-qPCR was performed using FastStart Universal SYBR Green Master (ROX) (04913914001, Roche). All of the procedures were performed according to the kit instructions. U6 messenger RNA (mRNA) was used to standardize the target genes, and the relative quantification of PCR products was calculated using the 2−ΔΔCT method. The associated gene primer sequences are shown in Additional file 1: Table S1.

The kidneys were cut and homogenized. After the concentration of the protein isolated by centrifugation was determined (BCA, Beyotime Biotechnology Co. Ltd.), equal amounts of protein were loaded on 10.0% Tris–glycine gels (120915132, Nanjing BERKE Biology Technology Co. Ltd., Nanjing, Jiangsu, China) for electrophoresis. Proteins were wet-transferred to polyvinylidene fluoride membranes (Merck Millipore, Billerica, MA, USA) with standard procedures. The primary antibodies used were as follows: Beclin 1 (1:1000, PAJ557Hu01, Wuhan Servicebio Technology Co. Ltd.), JNK (1:1000, PAB156Hu01, Cloud-clone Co. Ltd., Wuhan, Hubei, China), LC3II/I (1:1000, 3868, Cell Signaling Technology, Danvers, MA, USA), MEK4 (1:1000, PAD564Hu01, Cloud-clone Co. Ltd.), active caspase 3 (1:2000, 9664, Cell Signaling Technology), GAPDH (1:3000, T0004, Affinity Biosciences, Cincinnati, OH, USA), p-JNK (1:1000, 4668, Cell Signaling Technology). GAPDH was used to standardize the protein expression. The images were collected by a Bio-Rad ChemiDoc MP (Bio-Rad, Berkeley, CA, USA) and then quantified by Image J software (National Institute of Health, Bethesda, MD, USA).

To detect the luciferase activity of cells in each group,the Firefly and Renilla luciferase activities of each well were detected on a full-function microplate detector. The cells were divided into five groups: (1) miR-92a-3p mimic + pGL3-control + PLR-TK; (2) miR-92a-3p mimic + pGL3-MAPK8-1798–1805 + PLR-TK; (3) miR-92a-3p mimic + pGL3-MAPK8-1798–1805-mut + PLR-TK; (4) miR-92a-3p mimic + pGL3- MAP2K4-105–112 + PLR-TK; and (5) miR-92a-3p mimic + pGL3- MAP2K4-105–112-mut + PLR-TK. According to the above experimental groups, different plasmids were transfected for 48 h. All operations were then carried out following Promega’s double luciferase report kit (E1960, Promega Corporation, Madison, WI, USA).

The autophagic flux was detected by confocal microscopy. The cells were infected with mRFP-GFP-LC3 adenovirus (HB-AP210 0001, Hanbio Biotechnology Co. Ltd., Shanghai, China) 24 h before establishing the hypoxia–reoxygenation model. After modeling, the cells were collected, fixed, sealed, and then analyzed by confocal photography. MRFP-GFP-LC3, GFP, and mRFP expressed in adenovirus containing fluorescent protein were used to label and track LC3. The decrease of GFP can indicate the fusion of lysosome and autophagosome to form autophagy–lysosome. Yellow spots represent autophagosomes and red spots represent autophagy lysosomes. The number of spots can reflect the intensity of autophagy flow.

Continuity variables were presented as mean ± standard deviation (SD) and analyzed by GraphPad Prism 5 software (GraphPad Software, San Diego, CA, USA). A one-way ANOVA followed by a Tukey’s test, was used to compare the differences of all column pairs among multiple groups. For the difference between the two groups, a Student’s t-test was applied. P < 0.05 was considered to be significant.