Cisplatin-induced acute kidney injury is alleviated by BMSCs-derived exosome via mmu-miR-874-3p-mediated activation of the wnt/β-catenin signaling pathway

Abstract

Background

Acute kidney injury (AKI) results from cisplatin chemotherapeutic agents in 30 %–46 % of patients, but clinically effective preventive and therapeutic approaches are lacking. Bone marrow mesenchymal stem cells-derived exosomes (BMSCs-exo) have potential in tissue repair, but the mechanism by which they attenuate cisplatin-induced kidney injury is unknown.

Objective

To explore the therapeutic effect of BMSCs-exo on cisplatin-induced AKI and to analyze the key molecular mechanism involved.

Methods and materials

BMSCs-exo were extracted via ultracentrifugation and identified via transmission electron microscopy, nanoparticle analysis and Western blot. C57BL/6 mice were divided into a control group (Con), a cisplatin model group (Cis), and a BMSCs-exo treatment group (BMSCs-exo), and renal function was dynamically tested. PAS staining was used to observe histopathological changes in mouse kidney tissues, while immunohistochemistry was employed to assess the expression levels of Wnt4, β-catenin, FZD5, CD31, and the tubular injury markers NGAL and KIM1. Western blot was used to detect the expression of Wnt4, β-catenin, FZD5 and CD31. High-throughput sequencing was used to screen for differential miRNAs, and GO/KEGG enrichment analysis of target genes was performed.

Results

Blood creatinine and urea nitrogen levels were significantly higher in the Cis group than in the Con group, and renal tubular epithelial cells exhibited necrosis, confirming successful AKI model establishment. BMSCs-exo alleviated renal dysfunction, histopathological alterations, and tubular injury in vivo, as evidenced by NGAL and KIM1 expression. We further demonstrated that BMSCs-exo specifically localized to the injured kidney. MiRNA sequencing of renal tissues from the Con, Cis and BMSCs-exo groups identified mmu-miR-874-3p—enriched in Wnt signaling and angiogenesis pathways—as a key mediator of the renoprotective effects of BMSCs-exo, with FZD5 as its downstream target. Moreover, treatment with BMSCs-exo markedly prevented microvascular loss. In the BMSCs-exo group, Wnt4, β-catenin and CD31 expression were upregulated, whereas FZD5 expression was downregulated, consistent with the immunohistochemistry results.

Conclusions

BMSCs-exo protect kidneys against cisplatin-induced AKI(Cis-AKI) by attenuating injury to the renal microvasculature and tubule epithelial cells, primarily through mmu-miR-874-3p-mediated inhibition of FZD5 activation and promotion of Wnt/β-catenin pathway activation.

1. Introduction

Acute kidney injury (AKI) is a clinical syndrome characterized by a dramatic, short-term decrease in renal function. In intensive care units (ICUs), AKI occurs in approximately 33 %–66 % of patients, 10 %–15 % of whom require renal replacement therapy (KRT), but the mortality rate of such patients is still as high as 50 % [1]. Cisplatin, a highly effective chemotherapeutic agent for solid tumors, can cause 30 %–46 % of cancer patients to develop pharmacological AKI [2,3], but the mechanism of its nephrotoxicity has not been fully elucidated, and there is still a lack of effective prevention and treatment, which makes exploring novel therapeutic strategies highly important. During glomerular filtration and tubular secretion, cisplatin accumulates in the kidneys, especially in renal proximal tubular epithelial cells, where its concentration can be up to five times greater than its blood concentration, further leading to proximal tubule injury, specifically in the S3 segment [4]. Although cisplatin is associated with a high incidence of nephrotoxicity, the mechanism of its nephrotoxicity has not been fully elucidated. Cell-based therapy is deemed a promising approach for AKI treatment, and multiple studies have verified the therapeutic effects of stem cells. Nevertheless, cell-based therapy is also recognized as having side effects that require further evaluation. Previous studies have shown that BMSC-derived exosomes exert therapeutic effects through antiapoptotic pathways in disease models such as spinal cord injury, cerebral ischemia‒reperfusion injury and AKI [[5], [6], [7]]. Exosomes (exos), small vesicles with a diameter of 30–150 nm, are among the key paracrine effectors of mesenchymal stem cells (MSCs). They play a regulatory role through their contents, such as proteins, microRNAs (miRNAs), and long noncoding RNAs [8,9]. Among these, miRNAs are considered the most diverse. Owing to their properties, MSCs-derived exosomes can be used as carriers to transport miRNAs to recipient cells [10], showing significant therapeutic potential. Studies have shown that miRNAs are involved in the differentiation, proliferation, apoptosis, and decomposition of vascular endothelial cells and renal tubular epithelial cells and play important roles in different AKI models [6,11]. However, their role and mechanisms in cell-free therapy still need further study. With ongoing research, MSCs-derived exosomes have exhibited marked therapeutic potential for treating various diseases. Specifically, the role of exosomes from bone marrow stem cells (BMSCs) in renal-related disorders has attracted increasing interest. For example, BMSC-derived exosomal miRNA (miR)-34c-5p has been shown to inhibit core fucosylation in the unilateral ureteral obstruction (UUO) kidney during renal fibrosis [12]. Similarly, curcumin-loaded BMSC-derived exosomes mediate sepsis-associated AKI (SA-AKI) through the inhibition of RTEC injury through the secretion of FTO to reduce OXSR1 expression [13]. In addition, BMSCs protect against renal ischemia‒reperfusion injury by secreting exosomes loaded with miR-199a-5p that target BIP to inhibit endoplasmic reticulum stress during renal ischemia‒reperfusion injury [14]. Despite these promising findings, the specific mechanisms by which BMSC-derived exosomes influence cisplatin-induced AKI and related conditions warrant further exploration. The Wnt signaling pathway plays an important role in the repair of tissue injury [15,16], but controversy still exists as to whether the Wnt/β-catenin pathway is involved in the protection against renal injury by cisplatin. There are both studies supporting its protective role [[15], [16], [17], [18]] and evidence that it may exacerbate injury. To complicate matters, this pathway also mediates cisplatin resistance in tumor cells [[19], [20], [21]], which further highlights the need for mechanistic studies. On the basis of these findings, the present study was conducted to systematically analyze the differential expression of miRNAs in renal tubular cells and the changes in the Wnt/β-catenin signaling pathway after the exosomal intervention of BMSCs by establishing a mouse model of cisplatin-induced AKI [3,16]. Therefore, in this study, by establishing a mouse model of cisplatin-induced AKI, we systematically analyzed the differential expression of miRNAs in renal tissues and changes in the Wnt/β-catenin signaling pathway after the intervention of exosomes from BMSCs and explored whether the exosomes affect the Wnt signaling pathway through paracrine miRNAs to construct a molecular network in which the exosomes attenuate the renal injury caused by cisplatin and to provide a new theory for providing a new theoretical basis for clinical prevention and control strategies.

2. Methods

2.1. BMSCs culture

C57BL/6 mouse BMSCs were purchased from Seastar Biologicals (Guangzhou, China). The BMSCs were placed in 25 cm² culture flasks and cultured at 37 °C, 5 % CO₂, and 90 % humidity using Seastar specialized medium (Suzhou, China) according to the supplier's instructions. The medium was changed every 3 days. The experiments were performed using 6–8 generations of passaged BMSCs. The company's criteria for the identification of BMSCs were positive for CD29, CD44, and Ly-6AE and negative for CD31 and CD117.

2.2. Isolation and identification of BMSCs-exo

When the confluence of the BMSCs reached 80 %, the serum-free medium was changed, and the culture was continued for 48 h. The culture supernatant was collected and centrifuged at 500×g for 20 min to remove cellular debris. Then, the supernatant was concentrated by centrifugation at 500×g for 100 min in an ultrafiltration tube, followed by centrifugation at 10,000×g for 20 min to remove large vesicles and centrifugation at 100,000×g for 180 min in an ultracentrifuge to obtain the exosome precipitate. After the precipitate was resuspended in sterile PBS, centrifugation at 100,000×g for 180 min was repeated for washing, and the final precipitate was resuspended in an appropriate amount of PBS and stored at −80 °C. The morphology of the exosomes was observed via transmission electron microscopy (TEM; JEOL, Tokyo, Japan). The diameter and distribution of the exosomes were measured via a nanoparticle tracking analyzer (NTA, Zetaview PMX-120, Germany). The exosome surface marker proteins CD63, CD9, CD81 and calnexin were detected via Western blot. The protein content of the exosomes was determined via a BCA protein quantification kit (Thermo, USA).

2.3. Animal experimentation

Adult male C57BL/6 mice (8–10 weeks old, body weight 20–25 g) were purchased from Guangdong Provincial Medical Laboratory Animal Center (License No. sCXK (GD) 2022-0002). The mice were housed at the Experimental Animal Center of Shenzhen University (License No. SYXK(GD)2022-0302) under controlled conditions of constant temperature (24 °C) and relative humidity (40 %–80 %) and were maintained under a 12-h light/dark cycle. The experimental mice (n = 10/group) were randomly divided into the following experimental groups: control group (Con): 4 ml/kg 0.9 % saline was injected intraperitoneally; cisplatin group (Cis): 20 mg/kg cisplatin was injected intraperitoneally (Jiangsu Haosun Pharmaceutical Group Co., Ltd.), which was used to construct the model of AKI; cisplatin + BMSCs-exo group (BMSCs-exo): 20 mg/kg cisplatin was injected intraperitoneally for the construction of an AKI model; after cisplatin injection, 100 μg of exosomes/mouse was injected via the tail vein. After cisplatin injection, blood was collected from the mice every 24 h through the medial canthus vein. At 72 h after cisplatin injection, all the mice were sacrificed, and blood and tissue samples were collected to assess renal function and tissue damage. The blood was centrifuged at 3000×g for 10 min, and the serum was separated and stored at −80 °C for testing. Both kidneys were immediately removed and cut into four parts along the coronal plane: two of the kidney tissues were rapidly frozen in liquid nitrogen and stored at −80 °C for subsequent protein analysis; the other part of the kidney tissues were fixed in 4 % paraformaldehyde solution and embedded in paraffin wax for light microscopic observation and immunohistochemical study. All animal experiments were conducted in accordance with the Basel Declaration and approved by the Ethics Committee for Animal Experiments of the Medical College of Shenzhen University (IACUC-2023–00181).

2.4. In vivo animal imaging

To trace BMSCs-exo in vivo, exosomes were labeled with PKH26 fluorescent dye. PKH26-labeled exosomes were obtained by adding 2.5 μl of PKH26 to 200 μl of exosome suspension, incubating at 37 °C for 5 min, adding culture medium and centrifuging at 100,000×g for 70 min. In the Con group (n = 5/group) and the Cis group (n = 5/group), each mouse received a tail vein injection of 100 μg of PKH-26-labeled exosomes after 0.9 % saline or cisplatin injection, and was imaged for in vivo biodistribution (λex/λem = 551/567 nm) via an ABL X6pro In Vivo Imaging System (Tanon, Shanghai, China) at 24 h, 48 h, and 72 h.

2.5. Renal function tests and histopathological observations

Renal function was assessed by measuring serum creatinine (sCr) and blood urea nitrogen (BUN) using a fully automated biochemical analyzer (Hitachi QA36, Tokyo, Japan) at 24 h, 48 h and 72 h after cisplatin injection. Kidney tissues were paraffin-embedded and sectioned at 4 μm thickness. Renal pathological damage was evaluated by periodic acid–Schiff (PAS) staining. Ten nonoverlapping fields (20 × magnification) were randomly selected under a light microscope, and semiquantitative scoring was performed based on the degree of renal tubular injury. Observed pathological features—tubular dilatation, tubular necrosis, protein cast formation, cell debris in the lumen and intraepithelial vacuolar degeneration of renal tubular epithelial cells—were scored according to the proportion of injured area within each field: 0 (no injury), 1 (<10 %), 2 (10–25 %), 3 (26–50 %), 4 (51–75 %), and 5 (>75 %). The score for each field was recorded as the quantitative measure of pathological damage.

2.6. RNA extraction and cDNA library construction

Two micrograms of total RNA per sample was taken for small RNA library preparation. The sequencing libraries were constructed via the NEBNext® Multiplex Small RNA Library Prep Set for Illumina® (NEB, USA) according to the instructions: total RNA was used as the starting sample, and cDNA was synthesized via reverse transcription by directly ligating the junctions to the ends of the small RNAs; the cDNA libraries were separated into 140–150 bp DNA fragments and recovered via PAGE, and the cDNA libraries were finally obtained. The cDNA library was obtained via PAGE to separate 140–150 bp DNA fragments and recover them.

2.7. miRNA sequencing and analysis

Raw RNA reads were obtained via high-throughput sequencing via the Illumina NextSeq 550 platform. The raw RNA read sequences were subjected to splice removal and quality control via Cutadapt and Trimmomatic. The resulting clean reads were aligned to the mouse reference genome (GRCm39) via bowtie to filter nonmouse reads as well as reads aligned to exons. All reads were aligned to the Rfam database via BLASTN (2.6.0) and bowtie to filter nonmiRNA sequences (rRNA, sRNA, snRNA, snoRNA, and snoRNA). snRNA, snoRNA, and mRNA). All reads were compared to the Repbase database via RepeatMasker to remove duplicate sequences. Known miRNA sequences were obtained from the miRBase database, and known miRNAs were identified via mirDeep2 (v2.0.0.8) software. Mature miRNAs were quantified to obtain counts, which were normalized to CPM (read count per million) values. Differential analysis was performed via edgeR. The miRNA target genes were predicted via the miRanda algorithm. For GO function analysis and KEGG pathway analysis, we used clusterProfiler.

2.8. Western blot

Total protein was extracted from mouse kidney samples via RIPA lysis buffer (Biyuntian, Shanghai, China) containing a protease inhibitor mixture (MCE, New Jersey, USA). The protein concentration was determined via a Pierce™ BCA protein quantification kit (Thermo Fisher Scientific, Massachusetts, USA). Thirty micrograms of protein samples were separated via sodium dodecyl sulfate‒polyacrylamide gel electrophoresis (SDS‒PAGE). The initial electrophoresis voltage was set at 50 V, and the voltage was increased to 80 V after the samples had completely entered the separation gel. Electrophoresis was continued for approximately 80 min until the protein bands migrated to approximately 0.5 cm from the bottom of the gel. The proteins were then transferred to a polyvinylidene difluoride (PVDF) membrane at a constant current of 300 mA in an ice bath for 1 h. After membrane transfer, the PVDF membrane was incubated with FZD5 (1:1000), Wnt4 (1:1000), β-catenin (1:1000), CD31(1:1000) and GAPDH (1:20,000) (Cell Signaling Technology, CST, USA) in 1 × TBST buffer for 5 min each. Next, the membrane was incubated with a horseradish peroxidase (HRP)-labeled rabbit/mouse-derived secondary antibody (1:5000, CST, USA) for 1 h at room temperature and washed again three times (5 min each time) with 1 × TBST buffer after completion of incubation. GAPDH was used as an internal reference protein, and the membranes were finally imaged and processed via a chemiluminescent immunoblotting analysis system.

2.9. Immunohistochemistry

Kidney tissues were fixed in 4 % paraformaldehyde solution at room temperature overnight, dehydrated, made transparent and then processed for paraffin embedding following standard procedures. The embedded tissues were cut into 4 μm thick sections, which were wrapped in a polymer of lysine. Immunohistochemical analysis was performed as previously reported. For immunohistochemical analysis, tissue slices were subjected to antigen retrieval by microwaving for 10 or 15 min in 10 mM sodium citrate buffer (pH 6.0). The tissue slices were immersed in 3 % hydrogen peroxide for 10 min to block endogenous peroxidase activity. The sections were incubated with a 1:50 dilution of primary antibodies. The primary antibodies used were as follows: a 1:200 dilution of anti-neutrophil gelatinase-associated lipocalin (NGAL), anti-kidney injury molecule 1 (KIM1), anti-CD31, anti-FZD5, anti-β-catenin antibody and a 1:50 dilution of Wnt4 antibody(CST,USA). The sections were treated simultaneously without the primary antibodies as a negative control overnight at 4 °C. The sections were washed three times with PBS and incubated with biotin-conjugated goat anti-rabbit IgG (Invitrogen) for 30 min at room temperature. The sections were incubated with secondary antibodies at 37 °C for 30 min. The slices were incubated with DAB (Invitrogen) followed by examination for 3 or 10 min under a microscope. The sections were counterstained with hematoxylin.

2.10. qPCR

Total RNA was extracted via TRIzol reagent (Invitrogen) and reverse transcribed into cDNA via HiScript® IIQ Select RT SuperMix for qPCR (Vazyme, Inc.). Real-time PCR was performed using 2∗Q3 SYBR qPCR master mix (TOLOBIO, Inc.) for real-time PCR. The primers used are shown in Table 1. GAPDH was used as an internal control. All PCR products were determined via the 2-Δ Δ CT method.

Table 1.

Primers for real-time PCR.

Gene	Forward5’ -3′	Reverse5’ -3′
GAPDH mmu-miR-874-3p	CCTCGTCCCGTAGACAAAATG ACACTCCAGCTGGG CTGCCCTGGCCCGAGG	TGAGGTCAATGAAGGGGTCGT TGGTGTCGTGGAGTCG

2.11. Statistical analyses

Statistical analysis of the data was performed via SPSS software (version 21.0, IBM Corp., Armonk, NY) and GraphPad Prism 8.0.1 (GraphPad, San Diego, CA, USA). The data are presented as the means ± SDs, and a p value of less than 0.05 was considered to indicate statistical significance. Differences among groups were analyzed using one-way analysis of variance (ANOVA) followed by Tukey's post hoc test.

All groups consisted of n = 10 mice per group (Con, Cis, and BMSCs-exo) and animals were randomly assigned at the start of the experiment. For each figure/panel, the actual sample size (n) is indicated in the manuscript: Fig. 2A–D (sCr, BUN,PAS and histopathological scores) used n = 10/group; Fig. 2E and F and Fig. 6 (IHC for NGAL, KIM-1, Wnt4, β-catenin, FZD5 and CD31) used n = 3/group; Fig. 2G (in vivo and exosome imaging) used n = 5/group; Fig. 4 (qPCR for miR-874-3p) used n = 3/group; and Fig. 5 (Western blots for Wnt4, β-catenin, FZD5 and CD31) used n = 3/group. With the exception of the in vivo imaging (n = 5/group), these smaller sample sizes represent randomly selected subsets drawn from the full cohort of n = 10 animals per group. All animals were included in the study and no animals or data were excluded (there were no missing data). Importantly, all biological replicates are independent individuals — each data point originates from a different mouse, and repeated measurements from the same animal were not treated as independent replicates.

Fig. 2.

Repair effects of BMSCs-exo on a cisplatin-induced acute kidney injury model. The renal function of the mice was assessed by measuring the serum creatinine (sCr) and blood urea nitrogen (BUN) concentrations separately. The sCr and BUN concentrations gradually increased in the Cis group, and the increase was significantly inhibited in the BMSCs-exo group. The most significant decrease was observed at 48 h and 72 h. The data are expressed as the means ± standard deviations (n = 10/group). A) Serum creatinine (sCr); B) blood urea nitrogen (BUN); C) representative PAS staining of renal sections from the Con, Cis, and BMSCs-exo groups, original magnification of the images in the lower panel, × 20, scale bars, 100 μm and original magnification of the images in the upper panel, × 40, scale bars, 50 μm; D) The histopathological scoring of renal tubular injury was defined as tubular dilatation, tubular necrosis, protein cast formation, cell debris in the lumen and intraepithelial vacuolar degeneration of renal tubular epithelial cells. Histopathologic scores were based on the percentage of affected tubules in renal sections, as described in the Materials and Methods. The data are expressed as the means±standard deviations (n = 10/group). ∗∗∗∗P < 0.0001; ∗∗P < 0.001; ∗P < 0.05; ns, no significant difference. E) Representative images of immunohistochemistry staining of NGAL in kidneys from different groups, × 40; scale bar, 50 μm) after 72 h of cisplatin administration (n = 3/group). F) Representative images of immunohistochemistry staining of KIM1 in kidneys from different groups, × 40; scale bar, 50 μm) after 72 h of cisplatin administration (n = 3/group). G) Representative in vivo fluorescence imaging of control and treated mice following tail-vein injection of PKH26-labeled BMSCs-exo (n = 5/group).

Fig. 6.

Kidney tissues were collected 72 h after cisplatin injection for immunohistochemical staining to analyze the expression and localization of proteins related to the Wnt/β-catenin pathway. A) Expression of Wnt4: positive signals (brownish color) were localized in the cytoplasm and mesenchyme of renal tubular epithelial cells, with stronger expression in the Con group and significantly weakened expression in the Cis, and the intensity of the expression was restored and enhanced after the intervention with BMSCs-exo, which was consistent with the trend of the upregulation of Wnt4 protein expression observed via Western blotting (n = 3/group). B) Expression of β-catenin: The intranuclear positive signal (indicated by arrows) was weak in the control group, further reduced in the Cis, and significantly enhanced after BMSCs-exo intervention, which was consistent with the results of intranuclear β-catenin accumulation detected by Western blotting (n = 3/group). C) FZD5 expression: the positive signal was distributed in the vascular endothelium and renal tubular basement membrane, with stronger expression in the Cis, and the signal intensity was significantly reduced after BMSC-exo intervention, which was in line with the trend of downregulation of the FZD5 protein shown by Western blotting (n = 3/group). D) CD31 expression: Positive signals (brown) predominantly localized to vascular endothelial cells. Controls displayed robust CD31 expression, indicative of intact vascular networks. Cisplatin treatment markedly attenuated CD31 signals, suggesting that vascular integrity was impaired. BMSCs-exo restored and enhanced CD31 expression (n = 3/group), mirroring the recovery trends of Wnt4 and β-catenin.

Fig. 4.

To assess the regulatory effect of BMSCs-exo on cisplatin-induced kidney injury, the expression level of miR-874-3p in each group was detected via qPCR. The results revealed that miR-874-3p expression was significantly lower in the Cis group than in the Con group, whereas its expression level was significantly restored after BMSCs-exo intervention, suggesting that the exosomes effectively reversed cisplatin-induced miRNA inhibition. Con vs Cis, ∗∗P < 0.01 between the indicated groups; Cis vs BMSC-exos, ∗P < 0.05 between the indicated groups; Con vs BMSCs-exo, ns between the indicated groups(n = 3/group).

Fig. 5.

Tissue proteins were extracted 72 h after cisplatin injection, and the expression levels of Wnt/β-catenin pathway-related proteins were detected via Western blot analysis. A) Expression of Wnt4, with GADPH serving as an upsampling control. The Wnt4/GADPH relative densities were expressed as the means ± standard deviations; for Wnt4, Con vs Cis, ∗P < 0.05; Con vs BMSCs-exo, intergroup ns shown; Cis vs BMSCs-exo, ∗P < 0.05; B) Expression of β-catenin, with GADPH serving as an upsampling control. The β-catenin/GADPH relative densities are expressed as the means ± standard deviations; β-catenin in Con vs Cis, ns between groups shown; Con vs BMSCs-exo, ∗∗∗P < 0.001; Cis vs BMSCs-exo, ∗∗P < 0.01. C) Expression of FZD5, with GADPH serving as an upsampling control. The FZD5/GADPH relative densities are expressed as the means ± standard deviations; Con vs. Cis, ns between groups; Con vs. BMSCs-exo, ∗∗P < 0.01; Cis vs. BMSCs-exo, ∗∗P < 0.01. D) Expression of CD31, with GADPH serving as an upsampling control. The CD31/GADPH relative densities are expressed as the means ± standard deviations; Con vs Cis, ∗∗P < 0.01; Con vs. BMSCs-exo, ∗∗P < 0.01; Cis vs. BMSCs-exo, ∗∗∗∗P < 0.0001,(n = 3 biological replicates per group,one mouse per lane).

3. Results

3.1. Identification of BMSCs-exo

To characterize the exosomes, electron microscopy was used to study their morphological characteristics, which were characterized by their shape and size (Fig. 1A), with particle diameters ranging from 30 nm to 150 nm and a double-membrane chaotic structure, and the results of NTA tracing revealed that 87 % of the vesicle diameters were concentrated in the range from 30 nm to 150 nm (Fig. 1B), which was consistent with the range of the particle size distribution of BMSCs-exo. exos, which was consistent with the particle size distribution range of the BMSCs-exo. Differences in exosome-associated marker expression between BMSCs group and BMSCs-exo group were examined via Western blot. CD63, CD9, and CD81 were expressed at higher levels in the BMSCs-exo group than in the BMSCs group. Conversely, Calnexin expression was lower in BMSCs-exo group compared to BMSCs group (Fig. 1C), which indicated that the particles in the supernatant of the BMSCs were exosomes.

Fig. 1.

Identification of BMSCs-exo. A) The shape and size of the BMSCs-exo were detected via electron microscopy (scale bars, 200 nm). Particles ranging from 30 to 150 nm in diameter were observed in a double-layered membranous teato-like structure. B) Measurement of the size of BMSCs-exo via NTA revealed that 87 % of the vesicles were 30–150 nm in diameter. C) The expression levels of exosome markers were detected via Western blotting. The expression levels of CD63, CD9 and CD81 were significantly greater in the BMSCs-exo group than in the BMSCs group. The expression levels of Calnexin were lower in BMSCs-exo group compared to BMSCs group.

3.2. Protective effects of BMSCs-exo against cisplatin-induced acute kidney injury at the pathological and biochemical levels

To evaluate the therapeutic effect of BMSCs-exo on cisplatin-induced AKI, the present study was conducted to construct a mouse model of AKI by intraperitoneal injection of cisplatin, and the results were analyzed on the basis of biochemical indices and pathological characterization. All the serum samples were used to analyze the levels of BUN and sCr. As shown in Fig. 2, BMSCs-exo exerted nephroprotective effects. Seventy-two hours after injection, the sCr level in the BMSCs-exo group was significantly lower than that in the Cis group (P < 0.0001) (Fig. 2A); 48 h after injection, the BUN level in the BMSCs-exo group was already significantly lower than that in the Cis group (P < 0.0001) (Fig. 2B), and the above results indicated that the addition of BMSCs-exo had a significant therapeutic effect.

Pathological examination revealed that kidneys in the Cis group exhibited hallmark features of drug-induced injury, including extensive tubular dilation, tubular necrosis, tubular cast formation, intratubular cellular debris, and intraepithelial vacuolar degeneration of tubular epithelial cells. In contrast, BMSCs-exo treatment markedly ameliorated these histopathological changes, as evidenced by preserved tubular architecture, reduced cast formation, and attenuated epithelial vacuolar degeneration (Fig. 2C). Semiquantitative renal injury scoring confirmed a significant reduction in pathology scores in the BMSCs-exo group versus the Cis group (Fig. 2D), consistent with NGAL and KIM1 immunohistochemical staining (Fig. 2E and F). These results indicate that BMSCs-exo infusion provides robust protection against cisplatin-induced renal histopathological damage.

3.3. In vivo tracing of PKH26-labeled BMSCs-exo reveals cisplatin-induced kidney-targeted enrichment in mice with acute kidney injury

Dynamic monitoring via an in vivo fluorescence imaging system revealed that PKH26-labeled BMSCs-exo injected via the tail vein into cisplatin-induced AKI model mice exhibited widespread systemic distribution. In the model group, fluorescent signals were detected in liver, lung, kidney, spleen and brain tissues at 24 h after administration; but the injured kidney showed marked signal enhancement at 48 h and reached peak intensity at 72 h, whereas no significant renal fluorescence was observed in the control group after injection (Fig. 2G). These results indicate that BMSCs-exo have kidney-specific targeting ability in the pathological microenvironment, and that their enrichment kinetics at the site of injury suggest they may participate in tissue repair via a paracrine mechanism.

3.4. Bone marrow mesenchymal stem cells-derived exosomes alleviate cisplatin-induced acute kidney injury through key miRNA regulation

By miRNA sequencing analysis of the Con, Cis and BMSCs-exo groups, we identified 42 significantly differentially expressed miRNAs (Fig. 3A), of which 41 miRNAs were differentially expressed between the Cis and Con groups, and 26 upregulated and 15 downregulated miRNAs were enriched in the Cis group (Fig. 3B). Meanwhile, the BMSCs-exo intervention reversed the cisplatin-induced miRNA dysregulation, and the expression of 6 miRNAs was significantly altered in the BMSCs-exo group compared with the Cis group, of which 2 were upregulated and 4 were downregulated (Fig. 3C). Further analysis revealed that mmu-miR-344-3p, mmu-miR-124-3p, mmu-miR-874-3p, mmu-novel-32, and mmu-miR-132-3p were significantly differentially expressed between the Cis and Con groups, suggesting that these miRNAs may be core mediators of exosomes for repairing renal injury.

Fig. 3.

Differential miRNA expression profiles, target gene interaction network and functional enrichment analysis A) Heatmap of significant differential miRNA expression (key miRNAs common to both the Cis vs Con and BMSCs-exo vs Cis groups are marked in red); B) Differential miRNAs between the Cis vs Con groups; C) Target gene interaction network of the BMSCs-exo vs Cis intergroup differential miRNAs; D) Target gene interaction network of the Cis vs Con differential miRNAs; E) GO functional enrichment of the Cis vs Con differential miRNA target genes; F) KEGG pathway enrichment of the Cis vs Con differential miRNA target genes; G) Target gene interaction network of the BMSCs-exo vs Cis differential miRNAs; H) GO function enrichment of the BMSCs-exo vs Cis differential miRNA target genes; I) KEGG pathway enrichment of the BMSCs-exo vs Cis differential miRNA target genes.

3.5. BMSCs-exo ameliorate cisplatin-induced kidney injury through miR-874-3p-mediated modulation of angiogenesis

To elucidate the molecular mechanism of cisplatin-induced acute kidney injury by BMSCs-exo treatment, in this study, we systematically predicted the target genes in the Cis group versus the Con group (41 differential miRNAs) as well as in the BMSCs-exo group versus the Cis group (6 differential miRNAs) and constructed a functional interaction network via the miRanda algorithm. The results revealed that a total of 3285 target genes were predicted by 29 miRNAs in the Cis group versus the Con group, whereas 4 miRNAs in the BMSCs-exo group versus the Cis group corresponded to 1435 target genes. GO and KEGG enrichment analyses revealed that the differential miRNA target genes compared between the two groups had significant functional overlap, in which the Wnt/β-catenin signaling pathway and the angiogenesis-related pathway were significantly enriched in both groups (Fig. 3 E, F, H, I). Notably, Table 2 was constructed on the basis of the following principle: from all the differential miRNA-predicted target genes, key genes involved in both the Wnt/β-catenin signaling pathway and angiogenesis pathway were screened out, and their targeting relationships with miRNAs were verified via an algorithm. The results revealed that mmu-miR-874-3p had the largest number of target genes and covered the core nodes of the above two pathways, including Nkd1, a negative regulator of the Wnt pathway; Tcf7, a member of the transcription complex; and Ephb2 and Tie1, angiogenesis-related receptors (Table 2). Further analysis revealed that among the 41 miRNAs differentially expressed between the Cis group and the Con group, most of the target genes related to the Wnt/β-catenin and angiogenesis pathways were directly regulated by mmu-miR-874-3p, whereas the proportion of the 6 miRNAs whose expression was reversed after BMSCs-exo intervention increased. These findings suggest that mmu-miR-874-3p constitutes a dual molecular mechanism by which BMSCs-exo repair kidney injury by simultaneously targeting the Wnt/β-catenin signaling pathway. The above results systematically revealed the molecular mechanism by which BMSCs-exo synergistically regulate the multipathway network by delivering mmu-miR-874-3p.

Table 2.

MiRNA target genes involved in angiogenesis and the Wnt signaling pathway.

miRNA	Gene	Associated pathways	Other evidence
mmu-miR-874-3p	Vangl2	Wnt Signaling Pathway	miRWalk
mmu-miR-874-3p	Tcf7	Wnt Signaling Pathway
mmu-miR-874-3p	Rac3	Wnt Signaling Pathway
mmu-miR-874-3p	Plcb3	Wnt Signaling Pathway
mmu-miR-874-3p	Fzd8	Wnt Signaling Pathway
mmu-miR-874-3p	Wnt16	Wnt Signaling Pathway
mmu-miR-874-3p	Nkd1	Wnt Signaling Pathway	TargetScan8.0, miRWalk
mmu-miR-874-3p	Fzd5	Angiogenesis
mmu-miR-874-3p	Scg2	Angiogenesis
mmu-miR-874-3p	Col18a1	Angiogenesis
mmu-miR-874-3p	Meis1	Angiogenesis
mmu-miR-874-3p	Gdf2	Angiogenesis
mmu-miR-874-3p	Aggf1	Angiogenesis
mmu-miR-874-3p	Tal1	Angiogenesis	miRWalk
mmu-miR-874-3p	Tie1	Angiogenesis
mmu-miR-874-3p	Ephb2	Angiogenesis	TargetScan8.0, miRWalk
mmu-miR-874-3p	Egf	Angiogenesis
mmu-miR-874-3p	Col4a2	Angiogenesis
mmu-miR-874-3p	C1galt1	Angiogenesis
mmu-miR-874-3p	Thy1	Angiogenesis	miRWalk
mmu-miR-874-3p	Cspg4	Angiogenesis
mmu-miR-874-3p	Parva	Angiogenesis

3.6. Mmu-miR-874-3p mRNA expression in renal tissues from each group

To clarify the therapeutic effect of BMSCs-exo on cisplatin-induced acute kidney injury, this study detected the expression level of miR-874-3p in each group via qPCR. The results revealed that the expression level of miR-874-3p was significantly greater in the renal tissues of the BMSCs-exo intervention group than in those of the Cis group(Fig. 4), which was consistent with the pre-miRNA sequencing data.

3.7. BMSCs-exo attenuate cisplatin-induced renal injury by activating the Wnt4/β-catenin axis, suppressing FZD5, and enhancing angiogenesis

BMSCs-exo significantly ameliorated cisplatin-induced kidney injury by modulating key proteins of the Wnt pathway. Through target prediction analysis, FZD5 was identified as the direct target of miR-874-3p, and it is involved in the regulation of both the Wnt signaling pathway (by inhibiting β-catenin nuclear translocation) and angiogenesis; thus, FZD5 was selected as the core research subject. Compared with that in the Cis group, FZD5 protein expression was significantly downregulated in the BMSCs-exo intervention group, whereas β-catenin intranuclear accumulation was markedly increased, and Wnt4 expression was synchronously increased. The above results indicated that BMSCs-exo targeted and inhibited FZD5 by delivering miR-874-3p, relieved its inhibitory effect on the classical Wnt pathway, and thus activated the Wnt4/β-catenin repair signaling axis, which provided direct molecular evidence for the alleviation of renal injury by exosomes. Furthermore, we analyzed CD31, an endothelial marker closely associated with angiogenesis, by Western blotting and found that its expression was lowest in the cisplatin model group, intermediate in the untreated control group, and highest in the BMSCs-exo treatment group, further supporting a role for BMSCs-exo in promoting renal microvascular repair(Fig. 5).

3.8. Immunohistochemical results revealed that BMSCs-exo alleviate cisplatin-induced kidney injury by modulating the wnt/β-catenin pathway

Immunohistochemical results revealed that Wnt4 protein expression was significantly decreased in the renal tissues of the Cis group, and the positive signal was confined mainly to the cytoplasm of renal tubular epithelial cells, whereas the intensity of Wnt4 expression was significantly restored and widely distributed in the renal cortex and medullary mesenchymal stroma after intervention with BMSCs-exo, which was in line with the trend of the upregulation of proteins detected by Western blotting. β-catenin showed a weak positive signal in the nucleus of renal tubular epithelial cells in the Cis group. The positive signal in the nucleus of epithelial cells was weak, and after treatment with BMSCs-exo, the deposition of brown granules in the nucleus increased significantly, suggesting that β-catenin nuclear translocation was activated, further supporting the observation of β-catenin accumulation in the nucleus via Western blotting. In addition, FZD5 protein was highly expressed in the vascular endothelium and renal tubular basement membrane region in the cisplatin group, whereas the intensity of its positive signal was significantly reduced after BMSCs-exo treatment, which was consistent with the downregulation of FZD5 protein expression shown by Western blotting (Fig. 6). These findings suggest that BMSCs-exo can remodel the repair microenvironment of renal tissues by synergistically upregulating the classical Wnt pathway (Wnt4/β-catenin) and potentially inhibiting the expression of the nonclassical pathway receptor FZD5 to attenuate Cis-AKI. CD31 expression was robust in the control group, indicating an intact vascular network. Cisplatin treatment markedly attenuated CD31 signaling, suggesting impaired vascular integrity. BMSCs-exo restored and enhanced CD31 expression, reflecting a trend toward restoration of Wnt4 and β-catenin. These findings are consistent with bioinformatics data emphasizing the regulation of angiogenesis-related genes and further support that BMSC-exo-derived miR-874-3p promotes vascular repair while activating Wnt/β-catenin, which together improve renal microcirculation and functional recovery.

4. Discussion

Cis-AKI is an urgent clinical challenge in tumor chemotherapy, and its core mechanisms are closely related to renal tubular epithelial cell apoptosis, oxidative stress, and microvascular injury [1,[22], [23], [24]]. In the present study, we demonstrated that BMSCs-exo significantly ameliorated renal function and histopathological injury in Cis-AKI mice by modulating the miR-874-3p-Wnt/β-catenin signaling axis.

Exosomes, as the core mediators of the paracrine effects of stem cells, have become the focus of research in the treatment of AKI because of their ability to circumvent the risk of cell transplantation and efficiently deliver bioactive molecules [6,[25], [26], [27], [28]]. We observed that sCr and BUN levels in Cis-AKI mice were significantly reduced, renal tubular epithelial cell necrosis was markedly reduced and attenuated the cisplatin-induced increases in the renal tubular injury markers NGAL and KIM1 after 72 h of BMSCs-exo intervention, a result that is consistent with previous studies on the mechanism by which exosomes modulate endothelial function, promote neovascularization and maintain microcirculatory integrity through miRNAs [29]. Meanwhile, exosomal function depends on internal miRNAs [22,30], playing an important role by binding to target cell genes [31]. Thus, we performed RNA sequencing to compare miRNAs expression differences among groups, preliminarily identified candidate miRNAs, and conducted GO functional analysis and KEGG pathway analysis to explore downstream signaling mechanisms. We screened 47 differentially expressed miRNAs between the BMSCs-exo group and the Cis group, among which mmu-miR-874-3p was significantly down-regulated after cisplatin injury, and its expression was restored to a near-normal level by exosome intervention, suggesting that this miRNA is a key molecule mediating the nephroprotective effect. Further analysis showed that mmu-miR-874-3p may exert a dual regulatory function by targeting the Wnt pathway and angiogenesis-related genes. Research studies have shown that miRNAs dysregulation and the aberrant activation/inhibition of the Wnt pathway highly associated with in various different biological processes such as hypoxia, cell proliferation, and cell death in cisplatin induced AKI. It was commonly known that Wnt signaling cascades were usually triggered through the secreted Wnt ligands binding to Frizzled (FZD) receptor proteins. It was also speculated that mmu-miR-874-3p regulated the angiogenesis by targeting FZD5(Table 2).

miR-874-3p is the mature product of miR-874. A growing number of studies have shown that miR-874 has important parts in diseases via targeting different genes. MSC-exos-derived miR-874 can protect renal ischemia-reperfusion injury via targeting RIPK [6]. Jun Ge et al., reported that miR-874-3p was downregulated in AKI patients and overexpressing miR-874-3p attenuated oxidative stress and inflammation in both I/R- and LPS induced AKI mice to recover the function of the kidney [32]. Kangling Xie et al. Showed overexpression of miR-874-3p can inhibit CXCL12 expression to promote angiogenesis and inhibit inflammatory factor release in ischemic stroke mice by activating the Wnt/β-catenin pathway [33]. Moreover, the potential mechanism under this process might be related to FZD5/Wnt signaling pathway. Western Blot and immunohistochemistry results showed the supplementation of exogenous mmu-miR-874-3p inhibited FZD5 expression, upregulated CD31, thereby promoting angiogenesis. Cisplatin, in addition to causing tubular cell injury by reducing medullary blood flow in the kidneys, can also damage the glomeruli. And the results of Western blot analysis for CD31 showed that its expression was significantly increased in the BMSCs-exo group, indicating a marked ability to repair vascular injury. The endothelial dysfunction it induces leads to vasoconstriction, which further exacerbates hypoxic injury to the capillaries. This ultimately induces AKI through ischemic damage [3]. Indeed, we also observed a reduced density of glomerular endothelial cells and peritubular capillaries following cisplatin treatment. This reduction was restored after exosome transplantation. The Wnt/β-catenin signaling pathway is reactivated post-injury to trigger the repair process [34,35]. As a downstream effector of the canonical Wnt pathway, β-catenin can be activated by different Wnt isoforms. Reports indicate that in cisplatin exposure, Wnt4 levels can be restored via BMSCs-CM treatment and play a key role in kidney tubulogenesis [17]. In the early phase of renal IRI, Wnt4 is near tubule regeneration sites and regulates β - catenin/Cyclin D1 [36]. The expression of β-catenin, factors involved in the Wnt/β-catenin signaling pathway, and Wnt4 were checked using Western blot and immunohistochemistry. Results showed that wnt4 and β-catenin expression were reduced significantly in model mouse. However, we observed that the activation effects of mmu-miR-874-3p on wnt4 and β-catenin expression were followed after reduced FZD5 expression by BMSCs-exo. The dual upregulation of β-catenin and Wnt4 in this study suggests that exosomes may enhance classical pathway activity through the following mechanisms: miR-874-3p targets and inhibits Nkd1 (a transcriptional repressor of β-catenin), deregulating its negative regulation of β-catenin and facilitating its nuclear translocation; moreover, the intranuclear accumulation of β-catenin may activate the expression of classical Wnt ligands (e.g., Wnt4) through transcription, forming a positive feedback loop that further reinforces repair signaling [15]. This self-reinforcing signaling network can continuously drive the expression of antiapoptotic genes and promote the proliferation and regeneration of renal tubular epithelial cells, thereby accelerating injury repair. FZD5, a coreceptor for nonclassical Wnt pathways (e.g., Wnt5a signaling), was significantly downregulated in the BMSCs-exo group, but its main function was to inhibit classical β-catenin signaling through activation of the RhoA/ROCK or Wnt/Ca²⁺ pathways [[37], [38], [39], [40]]. Thus, downregulation of FZD5 may indirectly promote the nuclear translocation of β-catenin by reducing the inhibitory effect of nonclassical pathways [38,39,41]. This phenomenon does not contradict the direct activation of the classical pathway but reflects the precise regulatory ability of exosomes to coordinate the balance of the pathway through multiple targets.

Although the present study revealed the molecular mechanism by which BMSCs-exo ameliorate renal injury via the miR-874-3p–β-catenin axis, the following limitations remain: First, this study did not include a healthy + exo (exosome-only) control group, and thus we cannot fully exclude baseline effects of exosomes on normal renal structure or function; Second, exosome-only toxicity and safety in the absence of injury were not systematically assessed in our cisplatin-AKI model. Third, the targeted delivery efficiency of exosomal miRNAs and the mechanism of renal-specific enrichment have not yet been clarified, and integrin-mediated organ homing effects may be involved. Fourth, β-catenin may enhance cisplatin resistance in the tumor microenvironment [42], and the potential risk of BMSCs-exo needs to be evaluated in tumor transplantation models. Notably, Yu et al. [6]reported that human umbilical-cord MSC-derived exosomes carrying miR-874-3p mitigated tubular epithelial cell injury by targeting RIPK1/PGAM5; their study included an “Exo” (sham + exosome) cohort in an in-vivo UUO model and complementary in-vitro experiments in cisplatin-injured HK-2 cells, observing protective effects in injured kidneys without obvious structural or functional alterations in sham kidneys. However, it should be noted that Yu et al.'s in vivo findings were obtained in a UUO model, not a cisplatin-induced AKI model; to the best of our knowledge, studies specifically reporting safety assessments with healthy + exo controls in cisplatin-AKI animal models remain relatively limited. The inclusion of such an exosome-only sham group in animal experiments therefore provides a practical example of how baseline exosome effects and safety can be evaluated; the absence of a healthy + exo cohort in our cisplatin-AKI animal experiments remains a key limitation, because without it we cannot exclude the possibility of baseline biological effects or subtle toxicity of BMSCs-exo in uninjured kidneys nor definitively ascribe observed benefits solely to therapeutic effects in injured animals. Therefore, future in-vivo studies should incorporate healthy + exo groups and perform comprehensive safety and specificity assessments (renal histology, serum creatinine/BUN, urinary biomarkers, and longer-term follow-up), and could further resolve the dual role of miR-874-3p in repair versus tumor biology by using miR-874-3p knockout/overexpression exosomes combined with renal conditional β-catenin knockout mouse models.

5. Conclusions

BMSCs-exo attenuated Cis-AKI by delivering mmu-miR-874-3p, which activated the Wnt/β-catenin pathway and inhibited FZD5-mediated nonclassical signaling via mmu-miR-874-3p. This dual regulation enhances β-catenin nuclear translocation and Wnt4 expression while regulating angiogenesis-related targets, which together restore renal function and attenuate tubular injury. These findings suggest that BMSCs-exo constitute a promising miRNA-based therapeutic strategy for AKI.

Consent to participate

Not applicable.

Author contributions

Deyang Kong and Zhuohang Yang were involved in the conception and design; Xiaoting Zhang and Yifan Song participated in the animal experiments; Hao Liu，Zhanci Ou ，Yanping Li and Umer Anayyat participated in the analysis and interpretation of the data; Zhuohang Yang participated in the drafting of the paper; Shuo Pang and Xiaomei Wang participated in the critical revision of its intellectual content and ultimately approved the version for publication; and all the authors read and approved the final manuscript.

Declarations

The ethical approval and consent to participate in this study protocol were reviewed and approved by the Animal Experimentation Ethics Committee of the Medical Department at Shenzhen University, with approval number: IACUC-202300181.

Consent to publish declarations

Not applicable.

Data availability

Data available on request from the authors.

Data supporting the results of this study are available from the corresponding author Wang, upon reasonable request.

Funding

The present work was supported by Shenzhen Science and Technology R&D Funding for Basic Research (General Project) (JCYJ20220530164802006), Research Project on High-Quality Development of Public Hospitals Affiliated to Bao'an District (BAGZL2024189) and Research Project on Medical and Health Sciences of Bao'an District Medical Association in Shenzhen (BAYXH2024059).

Declaration of competing interest

The authors declare that there are no known financial or personal relationships that could have appeared to influence the work reported in this manuscript.