Administration of mircoRNA‐135b‐reinforced exosomes derived from MSCs ameliorates glucocorticoid‐induced osteonecrosis of femoral head (ONFH) in rats

Abstract Exosomes were found to exert a therapeutic effect in the treatment of osteonecrosis of the femoral head (ONFH), while miR‐135b was shown to play an important role in the development of ONFH. In this study, we investigated the effects of concomitant administration of exosomes and miR‐135b on the treatment of ONFH. A rat mode of ONFH was established. TEM, Western blotting and nanoparticle analysis were used to characterize the exosomes collected from human‐induced pluripotent stem cell–derived mesenchymal stem cells (hiPS‐MSC‐Exos). Micro‐CT was used to observe the trabecular bone structure of the femoral head. Real‐time PCR, Western blot analysis, IHC assay, TUNEL assay, MTT assay and flow cytometry were performed to detect the effect of hiPS‐MSC‐Exos and miR‐135b on cell apoptosis and the expression of PDCD4/caspase‐3/OCN. Moreover, computational analysis and luciferase assay were conducted to identify the regulatory relationship between PDCD4 mRNA and miR‐135b. The hiPS‐MSC‐Exos collected in this study displayed a spheroidal morphology with sizes ranging from 20 to 100 nm and a mean concentration of 1 × 1012 particles/mL. During the treatment of ONFH, the administration of hiPS‐MSC‐Exos and miR‐135b alleviated the magnitude of bone loss. Furthermore, the treatment of MG‐63 and U‐2 cells with hiPS‐MSC‐Exos and miR‐135b could promote cell proliferation and inhibit cell apoptosis. Moreover, PDCD4 mRNA was identified as a virtual target gene of miR‐135b. HiPS‐MSC‐Exos exerted positive effects during the treatment of ONFH, and the administration of miR‐135b could reinforce the effect of hiPS‐MSC‐Exos by inhibiting the expression of PDCD4.

the loss of structural integrity in the femoral head. 2 It has been reported that the apoptosis of osteocytes becomes more obvious in glucocorticoid-induced ON than that in alcohol-induced ON. In addition, the apoptosis of osteocytes is absent in ON triggered by sickle cell disease or traumatic events. 3 It has also been found that the apoptosis of osteocytes and osteoblasts is enhanced in ONFH compared with that in subcapital fracture or regular OA. 4 As small liposomes with a diameter of <100 nm, exosomes are generated by the invagination of endosomal membrane in multi-vesicular bodies (MVBs). Subsequently, exosomes are released into the extracellular space via fusion to the plasma membrane. 5 These extracellular vehicles (EVs) are implicated in the transportation of many essential materials, including proteins, RNAs and cytokines. 1 In addition, the content in exosomes is protected against degradation by the exosomal membrane. Furthermore, specific ligands located on the surface of exosomes can deliver biochemical materials into target cells by binding to these cells, thus stimulating corresponding biological functions. 6 In addition, it has been shown that the serum level of exosome is significantly reduced in patients suffering from steroid-induced ONFH. The ROC analysis has also shown that the level of serum exosome may be used as a diagnostic marker in steroid-induced ONFH. 7 As a novel tumour inhibitor, programmed cell death protein 4 (PDCD4) is involved in programmed cell death. Located on chromosome 10q24, the PDCD4 gene plays important roles and its allelic mutations are frequently observed in human cancers. The up-regulation in PDCD4 expression is seen during apoptosis, indicating that the reduction in PDCD4 expression may render cancer cells anti-apoptotic. 8 Furthermore, the expression of miR-206 is up-regulated in SANFH specimens along with down-regulated PDCD4 expression. 9 Therefore, miR-206 may promote the development of SANFH by inhibiting the proliferation of osteoblasts and by inducing their apoptosis, which in turn is dependent on the function of PDCD4. 9 As small and non-coding RNAs, microRNAs (miRNAs) play essential roles in the post-transcriptional regulation of gene expression. 10 By binding to the complementary sequences in the 3′-untranslated regions (3′-UTRs) of their target mRNAs, miRNAs induce the translational repression or degradation of their target genes. 11 It has been shown that miRNAs are present in exosomes during the osteogenic differentiation of bone marrow derived stromal cells (BMSCs). In addition, the expression of miR-302b, miR-299-5p, miR-219, miR-203, miR-135b, miR-148a, miR-218, miR-199b and let-7a is significantly elevated in exosome samples collected from human BMSCs. 12 The up-regulation in miR-135b was also shown to impair the osteogenic differentiation of BMSCs in patients with multiple myeloma, and it also exhibited protective effect in the development of ONFH via activating the AMPK. 13,14 It has also been found that the administration of exosomes collected from human-induced pluripotent stem cell-derived mesenchymal stem cells (hiPS-MSC-Exos) can promote the angiogenesis in ischaemic tissues. 15 These results suggested that hiPS-MSC-Exos may be used as a powerful tool in the treatment of ischaemic diseases, such as ONFH. 16,17 It was also shown that hiPS-MSCs-Exos can prevent bone loss by promoting the angiogenesis in ONFH. 18 It has been reported that the deregulation of miRNAs is involved in the pathogenesis of ONFH. 19 In particular, miR-135b has been shown to play an important role in the development of ONFH. 14 Furthermore, exosomes were found to exert a therapeutic effect in the treatment of ONFH. 16,17 In this study, we isolated hiPS-MSC-Exos and administered them with miR-135b in a rat animal model of ONFH, so as to investigate their therapeutic effect in the treatment of ONFH as well as relevant underlying molecular mechanisms.

| Isolation and characterization of exosomes
Exosomes were isolated from human MSCs using a miRCURY Exosome Isolation Kit (Exiqon) following the kit instructions. In brief, human MSCs were incubated in high-glucose DMEM containing 10% of foetal bovine serum and 1% penicillin/streptomycin (100 unit/mL streptomycin and 100 μg/mL penicillin). The culture conditions were 37°C, 5% CO 2 and saturated humidity. After reaching 80% confluency, the SMSCs were rinsed with PBS before being cultured in a FBS-free MesenGro hMSC medium for 48 hours. Subsequently, the cells were centrifugated respectively at 300 g and 2000 g (10 minutes each) and the supernatant was pre-filtered through a 0.22 µm membrane filter (Millipore), ultra-centrifuged using a 15 mL Amicon Ultra-15 Centrifugal Filter Unit (Millipore), and finally ultracentrifuged for 1 hour at 100 000 g in a sterile Ultra-Clear™ tube (BD). Subsequently, the morphology and distribution of extracted exosomes were measured using Western blotting, transmission electron microscopy (TEM), and dynamic light scattering (DLS), respectively. The expression of markers including CD9, CD63 and CD81 on the surface of the exosomes was measured using Western blotting to confirm the correct identity of these isolated exosomes.

| Animal model and grouping
A total of 60 healthy adult female Sprague-Dawley (SD) rats weighing between 200 and 300 g were purchased from the experimental animal centre of our institute. In order to investigate the therapeutic effect of exosomes and miR-135b on the treatment of ONFH, a rat model of ONFH was established and the rats were randomly divided into a SHAM group (sham-operated rats, n = 15), an ONFH group (rats induced of ONFH by GC, n = 15), an ONFH + EXO group (ONFH rats treated with exogenous MSC-Exos, n = 15), and an ONFH + EXO+miR-135b group (ONFH rats treated with both exogenous MSC-Exos and miR-135b, n = 15). In the ONFH group, methylprednisolone (MPS, Sigma-Aldrich) at a dose of 20 mg/kg/d was given once daily via intramuscular injection for three weeks to induce ONFH. After the course of injections was finished, the rats were fed with a standard diet and were allowed of free activity for another 3 weeks. In the ONFH + EXO group, ONFH rats were given an administration of 1 × 10 11 exosomes (100 μL of hiPS-

MSC-Exos) resuspended in 200 μL of PBS via tail vein injection on
a daily basis. In the ONFH + EXO + miR-135b group, ONFH rats were given daily administration of 1 × 10 11 exosomes and miR-135b simultaneously via tail vein injection. All animal experiments were approved by the Institutional Ethics Committee and were conducted in accordance with NIH's Laboratory Animal Care and Use Guide.

| Analysis and characterization of bone structure using micro-computed tomography (micro-CT)
A high-resolution micro-CT system (SkyScan 1272, Bruker micro-CT) was used to detect the structural characteristics of femoral heads collected from ONFH rats. The collected bone tissues were cut into thin slices and wrapped in a wax film coated with dental wax to prevent drying or bone movement during the scanning process. The voltage of X-ray was set to 50 kV and the light beam was

| RNA isolation and real-time PCR
Total RNA content from collected tissue and cell samples was extracted using a TRIzol reagent kit (Invitrogen), and the concentration and purity of RNA were determined via the detection of its optical density value. According to the instructions of a PrimeScript RT reagent Kit (Takara), RNA was reversely transcribed into cDNA. The total volume of RT reaction system is 25 μL and the reaction conditions were set as the following: reverse transcription at 37°C for 3 cycles (15 min each cycle), and inactivation of reverse transcriptase for 5 seconds at 85°C. Real-time PCR was carried out using an ABI 7900 real-time PCR system (ABI). The reaction conditions were as follows: pre-denaturation at 95°C for 4 minutes, and 40 cycles of denaturation at 94°C for 30 seconds, annealing at 58°C for 30 seconds, and extension at 72°C for 1 minute, followed by a final extension cycle at 72°C for 7 minutes. The expression of miR-135b and PDCD4 mRNA was calculated using the 2 −ΔΔCt method 40 and β-actin/U6 was used as the internal reference. In this study, the primer sequences used were: miR-135b-forward: 5′-GGCTTTTCATTCCTATGTG-3′; miR-135b-reverse: 5′-GAACATGTCTGCGTATCTC-3′; PDCD4-forward: 5′-ACTG TGCCAACCAGTCCAAAGG-3′; PDCD4-reverse: 5′-CCTCCACATC ATACACCTGTCC-3′; beta-actin-forward: 5′-CACCATTGGCAATGA GCGGTTC-3′; beta-actin-reverse: 5′-AGGTCTTTGCGGATGTCCA CGT-3′.

| Vector construction, mutagenesis and luciferase assay
The 3′ UTR of PDCD4 mRNA containing the binding site of miR-135b was amplified by PCR and cloned into a pcDNA dualluciferase vector (Promega). At the same time, site-directed mutagenesis was carried out in the miR-135b binding site to obtain mutant PDCD4, which was also cloned into the pcDNA vector.
In the next step, MG-63 and U-2 cells were co-transfected with miR-135b mimics + wild type/mutant PDCD4 using Lipofectamine 2000. At 48 hours post transfection, the luciferase activity of transfected cells was detected using a dual-luciferase detection kit (Promega) on a luminometer.

| Western blot analysis
Cell and tissue samples were lysed in a RIPA buffer and then centrifuged at 4℃ and 12 000 g for 15 minutes to harvest total protein, which was separated by 10% SDS-PAGE (Bio-Rad) and transferred onto a polyvinylidene difluoride (PVDF) membrane (Millipore). The PVDF membrane was then blocked in 5% skimmed milk and then incubated at 4°C overnight with primary anti-PDCD4 and anti-caspase antibodies (Abcam), followed by incubation with HRP-labelled secondary antibodies for 60 minutes at room temperature. An enhanced chemiluminescence instrument was utilized to visualize the expression of target proteins.

| Apoptosis analysis
The apoptotic profiles of cell and tissue samples were measured using TUNEL and flow cytometry assays, respectively. For flow cytometry assay, an Annexin V-FITC/PI staining kit (Invitrogen) was used according the kit instructions. For TUNEL assay, a TUNEL kit (Thermo Fisher) was used.

| MTT assay
The proliferation of cells was measured using an MTT assay according the kit instructions.

| Immunohistochemistry
An immunohistochemistry assay was carried out to observe the ex-

| Statistical analysis
SPSS (IBM) was used to statistically analyse the data. All results were shown as mean ± standard deviation (SD). Student's t test was used to compare two groups, while one-way ANOVA followed by Tukey's test as the post hoc test was used to compare multiple groups. A P value of <.05 was considered statistically significant.

| Establishment of a rat model of ONFH and the characterization of hiPS-MSC-Exos
During the characterization of hiPS-MSC-Exos, they displayed a spheroidal morphology with sizes ranging from 20 to 100 nm ( Figure 1A). In a subsequent Western blot analysis to evaluate the expression of surface markers, including CD9, CD63 and CD81, on these exosomes, the abnormal enrichment of these proteins was observed ( Figure 1B). Meanwhile, the results of nanoparticle analysis ( Figure 1C) also showed the exosomes had a mean concentration of 1 × 10 12 particles/mL and sizes ranging from 20 to 100 nm.

| Differentiated expressions of miR-135b, PDCD4 mRNA/protein and caspase-3 in different animal groups
The relative expression of miR-135b and PDCD4 mRNA was determined via real-time PCR. As shown in Figure 3A and compared with that in the SHAM group, the relative expression of miR-135b was significantly up-regulated in the ONFH + EXO group and evidently suppressed in the ONFH group, while the expression of miR-135b was further increased in the ONFH + EXO + miR-135b group. Interestingly, the expression of PDCD4 mRNA showed an opposite trend among these groups ( Figure 3B). In addition, the administration of exosomes alone failed to completely alleviate the dysregulated expression of PDCD4 mRNA, while the concomitant administration of exosomes and miR-135b fully inhibited the excessive expression of PDCD4 mRNA. Furthermore, in the Western blot analysis ( Figure 3C) and IHC assay ( Figure 4A), the ONFH group also presented a higher intensity of PDCD4 protein, while the concomitant treatment of exosomes and miR-135b greatly reduced the intensity of PDCD4 protein. Additionally, the protein intensity of caspase-3 ( Figure 3D) showed the same trend as that of PDCD4 protein. Therefore, it can be concluded that the expression of miR-135b, F I G U R E 2 Effect of hiPS-MSC-Exos and miR-135b on bone loss (*P value < .05 compared with the SHAM group; **P value < .05 compared with the ONFH group; # P value < .05 compared with the ONFH + EXO group). A, Quantification of BV/TV in the SHAM group, ONFH group, ONFH + EXO group and ONFH + EXO + miR-135b group; B, Quantification of trabecular thickness in the SHAM group, ONFH group, ONFH + EXO group and ONFH + EXO + miR-135b group; C, Quantification of bone surface area/bone volume in the SHAM group, ONFH group, ONFH + EXO group and ONFH + EXO + miR-135b group; D, Number of trabecular in the SHAM group, ONFH group, ONFH + EXO group and ONFH + EXO + miR-135b group PDCD4 and caspase-3 is affected during the treatment of ONFH by exosomes and miR-135b.

| In vivo effects of hiPS-MSC-Exos and miR-135b on osteocyte apoptosis and OCN level
In the TUNEL assay ( Figure 4B), the apoptosis index of osteocytes was the highest in the ONFH group, while the administration of exosomes reduced the value of apoptosis index in ONFH rats to a certain degree. In addition, the co-administration of exosomes and miR-135b further reduced the value of apoptosis index in ONFH rats. Also, as shown in Figure 5, the IHC results of the expression of OCN in the rat groups also indicated that the concomitant treatment of exosomes and miR-135b evidently promoted the suppressed level of OCN protein in ONFH rats. Therefore, the concomitant administration of exosomes and miR-135b could most significantly alleviate the apoptosis of osteocytes while obstructing the down-regulation of OCN in the treatment of ONFH.

| In vitro effects of hiPS-MSC-Exos and miR-135b on cell apoptosis and the expression of PDCD4/ caspase-3
In the MTT assay, the proliferation of MG-63 cells treated with hiPS-MSC-Exos or hiPS-MSC-Exos + miR-135b ( Figure 6A)  of Western blot analysis also showed that the protein expression of F I G U R E 3 Expression of miR-135b, PDCD4 mRNA/protein and caspase-3 in the SHAM group, ONFH group, ONFH + EXO group and ONFH + EXO+miR-135b group (*P value < .05 compared with the SHAM group; **P value < .05 compared with the ONFH group; # P value < .05 compared with the ONFH + EXO group). A, Real-time PCR results regarding the relative expression of miR-135b in the SHAM group, ONFH group, ONFH + EXO group and ONFH + EXO + miR-135b group; B, Real-time PCR results regarding the relative expression of PDCD4 mRNA in the SHAM group, ONFH group, ONFH + EXO group and ONFH + EXO + miR-135b group; C, Western blot results regarding the expression of PDCD4 protein in the SHAM group, ONFH group, ONFH + EXO group and ONFH + EXO + miR-135b group; D, Western blot results regarding the expression of caspase-3 in the SHAM group, ONFH group, ONFH + EXO group and ONFH + EXO + miR-135b group caspase-3 and PDCD4 ( Figure 6D) was evidently reduced in the presence of hiPS-MSC-Exos and miR-135b. Similar results were also observed in U-2 cells (Figure 7), indicating that the therapeutic effect of exosomes and miR-135b in the treatment of ONFH may be associated with their ability to regulate the expression of PDCD4 and caspase-3.

| The establishment of a miR-135b/PDCD4 molecular pathway
Computational analysis was carried out to identified a 'seed sequence' of miR-135b on the 3′UTR of PDCD4 mRNA ( Figure 8A).
In addition, the results of luciferase assay only showed reduced luciferase activity in MG-63 ( Figure 8B) and U-2 ( Figure 8C) cells cotransfected with wild-type PDCD4 mRNA and miR-135b. Therefore, it can be concluded that PDCD4 mRNA is targeted by miR-135b and the reinforcing effect of miR-135b in the administration of hiPS-MSC-Exos is mediated by its inhibition on PDCD4expression.

| D ISCUSS I ON
In this study, a rat model of ONFH was established and treated with MSC-Exos or EXO-miR-135b. It was found that bone indicators, in- MVBs. 22 At first, exosomes were thought as containers used to ship out unwanted membranous proteins or intracellular components from reticulocytes. 23 Recently, exosomes have been shown F I G U R E 4 PDCD4 protein expression and osteocyte apoptosis in the animal groups. A, IHC results regarding the expression of PDCD4 protein in the SHAM group, ONFH group, ONFH + EXO group and ONFH + EXO + miR-135b group. B, TUNEL assay results regarding the apoptosis index of osteocytes in the SHAM group, ONFH group, ONFH + EXO group and ONFH + EXO + miR-135b group F I G U R E 5 IHC results regarding the expression of OCN in the SHAM group, ONFH group, ONFH + EXO group and ONFH + EXO + miR-135b group to play important roles in the development of autoimmune diseases and cancer. Therefore, exosomes are now considered as a novel therapeutic tool that can be used in the treatment of neurological disorders including Parkinson's disease. 24 In particular, exosomes originated from several types of cells are shown to act as carriers of signalling transduction by transporting cell-specific miRNAs, DNAs, RNAs, lipids and proteins. 25 In this study, we showed that Furthermore, exosomes released from transfected BMSCs can enhance bone repair in steroid-induced ONFH. 26 Increased expression of miR-135b is observed in many cancers. [27][28][29] Real-time PCR and microarray analysis both showed that the expression of miR-135b is significantly up-regulated in non-small cell lung cancer and head and neck squamous cell carcinoma. 27,28 Studies also demonstrated that the expression of miR-135b is significantly up-regulated in both carcinomas and adenomas. 29 Moreover, the expression of miR-135b is up-regulated during the malignant transformation of normal tissues, indicating that the deregulation of miR-135b is involved in the onset of CRC. Located on 1q32.1, miR-135b is encoded by the LEMD1 gene. Interestingly, the DNA copy number of 1q32.1 usually increases during the progression of CRC, while the expression of LEMD1 is highly increased in CRC tissues. 30,31 Furthermore, past studies indicated that miR-135b can promote the malignant transformation of colorectal tissues and is required for the proliferation of placental cells. 32 It was also reported that miR-135b F I G U R E 6 Effects of hiPS-MSC-Exos and miR-135b on cell apoptosis and the expression of PDCD4/caspase-3 in MG-63 cells (*P value < .05 compared with the control group; **P value < .05 compared with the EXO group). A, MTT assay results regarding the proliferative capacity of MG-63 cells treated with hiPS-MSC-Exos, hiPS-MSC-Exos + miR-135b and the control; B, Flow cytometry results regarding the percentage of apoptotic cells in MG-63 cells treated with hiPS-MSC-Exos, hiPS-MSC-Exos + miR-135b and the control; C, Real-time PCR results regarding the relative expression of PDCD4 mRNA in MG-63 cells treated with hiPS-MSC-Exos, hiPS-MSC-Exos + miR-135b and the control; D, Western blot results regarding the expression of PDCD4 protein and caspase-3 in MG-63 cells treated with hiPS-MSC-Exos, hiPS-MSC-Exos + miR-135b and the control is implicated in the progression of many cancers. 33 In particular, overexpression of miR-135b is observed in lung, breast, and colon cancers. 33 It was also shown that, as a tumour inhibitor involved in the regulation of cell invasion, cell migration, and c-Myc expression, miR-135b is apparently down-regulated in osteosarcoma. 34 It has also been shown that the silence of miR-135b expression induces the apoptosis of CRC cells, although the relationship between miR-135b expression and the development of chemo-resistance in CRC cells remains unknown. 35 It is suspected that miR-135b can reduce the chemosensitivity and inhibit the apoptosis of colorectal cancer cells in vivo. 33  The up-regulation of PDCD4 was discovered in cells undergoing apoptosis, and PDCD4 has been shown to act as a novel tumour inhibitor and its down-regulation is observed in many cancers. 36,37 Furthermore, as a gene ubiquitously expressed in normal tissues, PDCD4 is up-regulated in both healthy and apoptotic cells. 37 In addition, the overexpression or silencing of PDCD4 can greatly impact cell survival and growth. 38 Although the mechanisms underlying the effect of PDCD4 remain unknown, it is believed that the PDCD4 protein in the cytosol can interact with eukaryotic translation initiation factor 4A (eIF4A) through its MA-3 domain and suppress the activity of its helicase, thus suppressing cap-dependent translation. 39 In addition, PDCD4 controls gene transcription by suppressing the activity of activator protein-1 (AP-1) in the nucleus.

| CON CLUS ION
In summary, we demonstrated here for the first time that miR-135b derived from exosomes secreted from MSCs alleviated the severity of ONFH in rats by reducing the level of PDCD4-induced apoptosis of osteoblasts. We also showed that, by directly targeting the 3'UTR of PDCD4, miR-135b derived from exosomes secreted from MSCs suppressed PDCD4 expression and hence played an important role in the treatment of ONFH. F I G U R E 7 Effects of hiPS-MSC-Exos and miR-135b on cell apoptosis and the expression of PDCD4/caspase-3 in U-2 cells (*P value < .05 compared with the control group; **P value < .05 compared with the EXO group). A, MTT assay results regarding the proliferative capacity of U-2 cells treated with hiPS-MSC-Exos, hiPS-MSC-Exos + miR-135b and the control; B, Flow cytometry results regarding the percentage of apoptotic cells in U-2 cells treated with hiPS-MSC-Exos, hiPS-MSC-Exos + miR-135b and the control; C, Real-time PCR results regarding the relative expression of PDCD4 mRNA in U-2 cells treated with hiPS-MSC-Exos, hiPS-MSC-Exos + miR-135b and the control; D, Western blot results regarding the expression of PDCD4 protein and caspase-3 in U-2 cells treated with hiPS-MSC-Exos, hiPS-MSC-Exos + miR-135b and the control

DATA AVA I L A B I L I T Y S TAT E M E N T
The data sets generated during and/or analysed during the current study are available from the corresponding author on reasonable request.