HK-2 cell lines (American Type Culture Collection) and MSCs (American Type Culture Collection) were maintained in RPMI-1640 medium (Thermo Fisher Scientific, Inc.), supplemented with 10% FBS (Gibco; Thermo Fisher Scientific, Inc.), 1% penicillin (Thermo Fisher Scientific, Inc.) and 1% streptomycin (Thermo Fisher Scientific, Inc.) in a condition with 5% CO₂ and 37°C. To mimic AKI in vitro, HK-2 cells were treated with 20 µM cisplatin (MedChemExpress) for 48 h according to previous refs. (8,27).

HK-2 cells or MSCs (5×10³ cells/well) were transfected with miR-1184 agomir or agomir-negative control (NC) using Lipofectamine^® 2000 (Thermo Fisher Scientific, Inc.) at 37°C for 48 h. miR-1184 agomir (50 nM) and agomir-NC (50 nM) were obtained from Shanghai GenePharma Co., Ltd. The sequences were as follows: miR-1184 agomir, 5′-CCUUCGGUAGUUCAGCGACGUCC-3′ and agomir-NC, 5′-UUCUCCGAACGUGUCACGUUU-3′. After 48 h of transfection, cells were used in subsequent experiments.

For FOXO4 overexpression, MSCs (5×10³ cells/well) were transfected with pcDNA3.1 (1 µg/µl; Shanghai GenePharma Co., Ltd.) or pcDNA3.1-FOXO4 (1 µg/µl; Shanghai GenePharma Co., Ltd.) using Lipofectamine 2000 for 48 h at 37°C, according to the manufacturer's instructions. After 48 h of transfection, cells were used in subsequent experiments.

The differentially expressed miRNAs were presented in a volcano plot and a heatmap using the GSE53771 dataset (28) from the Gene Expression Omnibus (GEO) database (https://www.ncbi.nlm.nih.gov/geo). In addition, the differentially expressed miRNAs were screened using R analysis (29). Using a t-test, P<0.05 and fold-change >1.5 or <0.667 were considered as differentially expressed miRNAs. On the other hand, the functions of miRNA-targeted mRNAs in terms of ‘Cellular Components’ and ‘Biological Processes’ were investigated by Gene Ontology (GO) analysis (http://www.geneontology.org). Pathway analysis was performed to define biological pathways by using the KEGG orthology-based annotation system (KOBAS; version 3.0; http://kobas.cbi.pku.edu.cn/index.php).

The downstream target of miR-1184 was predicted using TargetScan (version 7.2; http://www.targetscan.org/vert_72/) and miRDB (version 2.0; http://www.mirdb.org/).

Briefly, MSCs were maintained in RPMI-1640 medium until they reached 80% confluence. The medium was then replaced with the serum-free medium. The supernatants were centrifuged for 1 h (300 × g for 15 min at 4°C, 2,000 × g for 15 min at 4°C and 10,000 × g for 30 min at °C) following 48 h of culture. Subsequently, the supernatants were filtrated and collected to extract exosomes using ultracentrifugation (2,000 × g for 15 min at 4°C). Western blot analysis was used to detect the exosome isolation, and the detailed protocol was in accordance with a previous study (25). The particle sizes were detected by Nanoparticle tracking analysis (NTA).

A total of ~0.3 ml supernatant was loaded into the sample chamber of an LM10 NanoSight unit (NanoSight, Ltd.) and three videos of either 30 or 60 sec were recorded of each sample. Data analysis was performed using NTA 2.1 software (NanoSight, Ltd.). In NTA, the paths of unlabeled particles acting as point scatterers, undergoing Brownian motion in a 0.25 ml chamber through which a 635-nm laser beam was passed, was determined from a video recording, with the mean squared displacement determined for each possible particle. The diffusion coefficient and sphere-equivalent hydrodynamic radius were subsequently determined using the Stokes-Einstein equation (25).

MSCs (5×10⁴ per well) were seeded overnight. Subsequently, cells were labeled for 24 h at 4°C with phalloidin (1:1,000; Abcam; cat. no. ab176753) or PKH26 red membrane dye (1:1,000; Biolab Co., Ltd.; cat. no. HR9070). The nuclei were stained with 5 µl/ml DAPI (Beyotime Institute of Biotechnology). The results were observed under a fluorescence microscope (magnification, ×200; Olympus Corporation).

RIPA lysis buffer (Beyotime Institute of Biotechnology) was used to extract total protein from the HK-2 cells. A BCA protein kit (Thermo Fisher Scientific, Inc.) was used to quantify the total protein. SDS-PAGE (10%) was used to separate the protein (40 µg per lane), and the protein was then transferred to PVDF membranes (Thermo Fisher Scientific, Inc.). After blocking with 5% skimmed milk at room temperature for 1 h, the membranes were incubated overnight at 4°C with anti-CD63 (cat. no. ab134045; 1:1,000), anti-CD81 (cat. no. ab109201; 1:1,000), anti-Bax (cat. no. ab32503; 1:1,000), anti-Bcl-2 (cat. no. ab32124; 1:1,000), anti-cleaved caspase-3 (cat. no. ab32042; 1:1,000), anti-TSG101 (cat. no. ab125011; 1:1,000), anti-FOXO4 (cat. no. ab128908; 1:1,000), anti-p27 Kip1 (cat. no. ab32034; 1:1,000), anti-CDK2 (cat. no. ab32147; 1:1,000) and anti-β-actin (cat. no. ab8227; 1:1,000) primary antibodies (all Abcam). Subsequently, the membranes were incubated with secondary anti-rabbit antibodies (HRP-conjugated; Abcam; cat. no. ab7090; 1:5,000) for 1 h at room temperature. Protein bands were visualized using an ECL kit (Thermo Fisher Scientific, Inc.). β-actin was used as the loading control. All antibodies were purchased from Abcam. ImageJ software (version 6.0; National Institutes of Health) was used for densitometry.

TRIzol^® reagent (Invitrogen; Thermo Fisher Scientific, Inc.) was used to isolate total RNA from HK-2 cells or MSCs. The PrimeScript RT Reagent Kit (Takara Bio, Inc.) was used to reverse transcribe total RNA into cDNA, according to the manufacturer's protocol. Subsequently, qPCR was performed using the SYBR Premix Ex Taq II kit (ELK Biotechnology). RT-qPCR reactions were performed under the following protocol: Initial denaturation for 2 min at 94°C, followed by 35 cycles (30 sec at 94°C and 45 sec at 55°C). The following primer pairs were used for RT-qPCR: miR-1184 forward, 5′-CTGGACTGAGCCGTGCTAC-3′ and reverse, 5′-CTCAACTGGTGTCGTGGAGTC-3′; and U6 forward, 5′-CTCGCTTCGGCAGCACAT-3′ and reverse, 5′-AACGCTTCACGAATTTGCGT-3′. The 2^−ΔΔCq method (30) was used to quantify the data. U6 was used as an internal control.

HK-2 cells (5×10³ cells/well) were seeded and treated with cisplatin (20 µM), MSC-Exo^{miR-1184 agomir} or cisplatin + MSC-Exo^{miR-1184 agomir} for 72 h at 37°C. Subsequently, CCK-8 reagent (10 µl; Beyotime Institute of Biotechnology) was added to the cells for 2 h at 37°C. The absorbance (450 nm) was measured using a microplate reader (Thermo Fisher Scientific, Inc.).

HK-2 cells were trypsinized, washed with PBS, resuspended in Annexin V Binding Buffer, and stained with 5 µl FITC and 5 µl PI for 15 min in the dark. The cells were the analyzed using a flow cytometer (FACSLyric™; BD Biosciences) to assess the incidence of cell apoptosis (early + late apoptosis). FlowJo (version 10.6.2; FlowJo LLC) was used to analyze the data.

The FOXO4 3′-untranslated region (UTR) containing putative miR-1184 binding sites were obtained from Sangon Biotech Co., Ltd., and cloned into pmirGLO Dual-Luciferase miRNA Target Expression Vectors (Promega Corporation) to construct FOXO4 wild-type (WT) or mutant (MUT) reporter vectors. The mutated 3′-UTR was generated using a site directed mutagenesis kit (Sangon Biotech Co., Ltd.). FOXO4 (WT) or FOXO4 (MUT) were transfected into HK-2 cells (5×10³) together with NC or miR-1184 agomir using Lipofectamine 2000. After 48 h of transfection, relative luciferase activities were then analyzed using a Dual-Glo Luciferase Assay System (Promega Corporation). Renilla luciferase activity was used for normalization.

HK-2 cell supernatants were collected by centrifugation (500 × g, 10 min, 4°C). Subsequently, the levels of TNF-α (cat. no. H052-1) and IL-1β (cat. no. H002) were investigated using ELISA kits (Nanjing Jiancheng Bioengineering Institute), according to the manufacturers' protocols.

In brief, HK-2 cells were harvested, fixed with 75% ethanol on ice for 20 min, permeabilized with 0.25% Triton X-100 and stained with PI/RNase (BD Pharmingen; BD Biosciences). Following incubation at 4°C for 15 min, cells were analyzed using a flow cytometer (BD FACSAria III; BD Biosciences). The data were quantified using FlowJo software (version 3.0; FlowJo LLC).

Each group was examined in three independent experiments and all data are expressed as the mean ± SD. Western blot analysis, RT-qPCR, flow cytometry, CCK-8 assay and immunofluorescence staining were repeated three times. An unpaired Student's t-test was used to analyze the differences between two groups, and one-way ANOVA followed by Tukey's test were used to analyze differences among multiple groups (>2 groups, using GraphPad Prism 7; GraphPad Software, Inc.). P<0.05 was considered to indicate a statistically significant difference.

To detect differentially expressed miRNAs in AKI, bioinformatics analysis was performed. As indicated in Fig. 1A and B, differentially expressed miRNAs (miR-1269a, miR-1184, miR-299-5p, miR-411, miR-451 and miR-499) are presented using a volcano plot and heatmap. In addition, the six differentially expressed miRNAs (miR-1269a, miR-1184, miR-299-5p, miR-411, miR-451 and miR-499; miR-1269a and miR-1184 were downregulated, while the other four miRNAs were upregulated) in AKI are presented (Fig. 1C). Moreover, GO and pathway analyses were performed to determine the most common ‘Biological Process’ of miR-1184. The results indicated that miR-1184 was enriched in the following ‘Cellular Component’ terms: ‘Brush border membrane’, ‘main axon’, ‘platelet dense granule’, ‘membrane region’, ‘pigment granule’, ‘melanosome’, ‘myelin sheath’, ‘membrane raft’, ‘membrane microdomain’ and ‘microfibril’ (Fig. 2A). miR-1184 was enriched in the following ‘Molecular Function’ terms: ‘Protein domain specific binding’, ‘nucleocytoplasmic carrier activity’, ‘disordered domain specific binding’, ‘β-catenin binding’, ‘protein C-terminus binding’, ‘nuclear import signal receptor activity’, ‘RNA polymerase II general transcription initiation factor binding’, ‘DNA-binding transcription activator activity’ and ‘RNA polymerase II-specific’ (Fig. 2A). miR-1184 was enriched in the following ‘Biological Process’ terms: ‘Modulation of process of other organism involved in symbiotic interaction’, ‘modulation by symbiont of host process’, ‘melanocyte differentiation’, ‘modulation of process of other organism’, ‘gland morphogenesis’, ‘modulation by virus of host process’, ‘pigmentation’, ‘pigment cell differentiation’, ‘regulation of DNA-binding transcription factor activity’ and ‘regulation of viral process’ (Fig. 2A). Notably, the most enriched ‘Cellular Component’ was ‘brush border membrane’, the most enriched ‘Molecular Function’ was ‘protein domain specific binding’ and the most enriched ‘Biological Process’ was ‘modulation of process of other organism involved in symbiotic interaction’ (Fig. 2A). Pathway analysis revealed that the ‘Hedgehog signaling pathway’, ‘pathways in cancer’, ‘sphingolipid signaling pathway’, ‘antifolate resistance’, ‘oocyte meiosis’, ‘TGF-beta signaling pathway’, ‘FOXO signaling pathway’, ‘prostate cancer’, ‘gastric cancer’, ‘pantothenate and CoA biosynthesis’, ‘cholinergic synapse’ and ‘cysteine and methionine metabolism’ were associated with the development of AKI (Fig. 2B). Based on the aforementioned data, miR-1184 was selected for further analysis.

In order to detect the efficiency of cell transfection and exosome isolation, RT-qPCR and NTA was performed. As shown in Fig. 3A, the expression of miR-1184 in MSCs was significantly upregulated by miR-1184 agomir, and NTA revealed a similar size distribution of exosomes (Fig. 3B). Moreover, exosomal proteins (TSG101, CD63 and CD81) were highly expressed in exosomes derived from MSCs, while they were expressed at low levels in MSCs (Fig. 3C). Moreover, the expression of miR-1184 in exosomes derived from MSCs was significantly upregulated by miR-1184 agomir (Fig. 3D), and MSC-derived exosomes labeled with fluorescent PKH26 were internalized by unstained MSCs (Fig. 3E). Furthermore, the expression level of miR-1184 in MSCs was significantly increased by exosomes derived from miR-1184 agomir-treated MSCs (Fig. 3F). Taken together, the data demonstrated that exosomes were successfully isolated from MSCs.

In order to detect the effects of exosomes on AKI in vitro, a CCK-8 assay was used. As illustrated in Fig. 4A, the viability of the HK-2 cells was significantly increased when the cells were co-cultured with exosomes derived from miR-1184 agomir-treated MSCs, and exosomes expressing miR-1184 significantly reversed the cisplatin-induced inhibition of cell viability. In addition, cisplatin significantly induced HK-2 cell apoptosis, while this phenomenon was reversed in the presence of exosomes derived from miR-1184 agomir-treated MSCs (Fig. 4B). Moreover, cisplatin significantly upregulated the expression levels of Bax and cleaved caspase-3 and downregulated the protein expression level of Bcl-2, while this phenomenon was reversed by exosomes derived from miR-1184 agomir-treated MSCs (Fig. 4C-F). Taken together, the results demonstrated that exosomes derived from miR-1184 agomir-treated MSCs significantly reversed cisplatin-induced HK-2 cell growth inhibition.

To explore the potential target of miR-1184, TargetScan and miRDB databases were used. As shown in Fig. 5A, FOXO4 may be the target of miR-1184, and the relative luciferase activity in the WT-FOXO4 group was significantly decreased by miR-1184 agomir (Fig. 5B). Moreover, the expression level of FOXO4 in HK-2 cells was significantly inhibited by the overexpression of miR-1184 (Fig. 5C). Furthermore, the levels of IL-1β and TNF-α in the supernatants of HK-2 cells were significantly upregulated by cisplatin, while this phenomenon was partially reversed by MSC-Exo^{miR-1184 agomir} (Fig. 5D and E). Therefore, miR-1184 directly targeted FOXO4 in HK-2 cells.

In order to further explore the underlying mechanisms through which exosomes or miR-1184 agomir regulate cisplatin-induced HK-2 cell injury, western blot analysis was performed. As indicated in Figs. 6A-C and S1A-C the protein expression levels of FOXO1 and p27 Kip1 in HK-2 cells were significantly upregulated by cisplatin. By contrast, cisplatin significantly inhibited the expression level of CDK2 (Figs. 6D and S1D). Furthermore, the effect of cisplatin on these three proteins was reversed by MSC-Exo^{miR-1184 agomir} (Figs. 6A-D and S1A-D). In addition, cisplatin-induced G₁ arrest was reversed by MSC-Exo^{miR-1184 agomir} (Fig. 6E). The expression of FOXO4 in MSCs was significantly upregulated by transfection with the FOXO4-overexpression vector (Fig. 6F). Meanwhile, the anti-inflammatory effect of exosomes derived from miR-1184 agomir-treated MSCs on cisplatin-treated HK-2 cells was rescued in the presence of exosomes derived from MSCs co-treated with miR-1184 agomir and FOXO4 overexpression (Fig. 6G). Collectively, the results demonstrated that exosomes derived from miR-1184 agomir-treated MSCs significantly induced G₁ arrest in HK-2 cells via the mediation of FOXO1, p27 Kip1 and CDK2.