Delivery of miR-15b-5p via magnetic nanoparticle-enhanced bone marrow mesenchymal stem cell-derived extracellular vesicles mitigates diabetic osteoporosis by targeting GFAP

Abstract

Diabetic osteoporosis (DO) presents significant clinical challenges. This study aimed to investigate the potential of magnetic nanoparticle-enhanced extracellular vesicles (GMNP_E-EVs) derived from bone marrow mesenchymal stem cells (BMSCs) to deliver miR-15b-5p, thereby targeting and downregulating glial fibrillary acidic protein (GFAP) expression in rat DO models. Data was sourced from DO-related RNA-seq datasets combined with GEO and GeneCards databases. Rat primary BMSCs, bone marrow-derived macrophages (BMMs), and osteoclasts were isolated and cultured. EVs were separated, and GMNP_E targeting EVs were synthesized. Bioinformatic analysis revealed a high GFAP expression in DO-related RNA-seq and GSE26168 datasets for disease models. Experimental results confirmed elevated GFAP in rat DO bone tissues, promoting osteoclast differentiation. miR-15b-5p was identified as a GFAP inhibitor, but was significantly downregulated in DO and enriched in BMSC-derived EVs. In vitro experiments showed that GMNP_E-EVs could transfer miR-15b-5p to osteoclasts, downregulating GFAP and inhibiting osteoclast differentiation. In vivo tests confirmed the therapeutic potential of this approach in alleviating rat DO. Collectively, GMNP_E-EVs can effectively deliver miR-15b-5p to osteoclasts, downregulating GFAP expression, and hence, offering a therapeutic strategy for rat DO.

Keywords: Bone marrow mesenchymal stem cell; Diabetic osteoporosis; Extracellular vesicle; GFAP; Magnetic nanoparticle; Osteoclast differentiation; miR-15b-5p.

Fig. 1

Selection of essential genes related to DO in the database. Note: (A) Volcano plot of DEGs analysis based on RNA-seq data set (Control group: n = 3; Case group: n = 3). Red dots represent up-regulated genes, blue dots represent downregulated genes, and gray dots represent genes with no difference; (B) Volcano plot of DEGs analysis based on GSE26168 data set (Control group: n = 4; Case group: n = 4); (C) Venn diagram showing the intersection of DEGs in the GSE56116 and GSE35958 microarray data sets with diabetes-related genes; (D) Heatmap of the 38 intersection genes in the RNA-seq data set (Control group: n = 3; Case group: n = 3); (E) Heatmap of the 38 intersection genes in the GSE26168 data set (Control group: n = 4; Case group: n = 4); (F) Differential expression of 5 intersection genes in RNA-seq data (Control group: n = 3; Case group: n = 3); (G) Bar chart of differential expression of GFAP in the RNA-seq data set; (H) Protein–protein interaction (PPI) network diagram of GFAP interactions; (I) KEGG functional enrichment analysis, including molecular function (MF), biological process (BP), and cell component (CC)

Fig. 2

Influence of GFAP on osteoclast differentiation. Note: (A-C) RT-qPCR and Western blot analysis of GFAP expression levels at different time points in BMMs stimulated with RANKL; (D-E) TRAP staining to detect osteoclast differentiation in different groups of BMMs (MNC ≥ 3 nuclei were considered positive), scale bar = 100 μm; (F-G) RT-qPCR and Western blot analysis of TRAP, NFATC1, MMP9, and CTSK expression levels in different groups of BMMs. * indicates P < 0.05 compared to the control group, ** indicates P < 0.01, *** indicates P < 0.001. The cell experiments were repeated three times

Fig. 3

Influence of GFAP on osteoporosis in DO rats. Note: (A-B) RT-qPCR (A) and Western blot (B-C) analysis of GFAP expression levels in bone tissues of different groups of rats; (D) Micro-CT analysis of femurs in different groups of rats; (E–H) Statistical analysis of BV/TV, Tb. N, Tb.Th, and BMD in femurs of different groups as shown in panel D; (I) TRAP staining to detect the number of osteoclasts in femoral tissues of different groups, with arrows pointing to positive cells, scale bar = 200 μm; (J) Statistical analysis of the number of positive osteoclast cells shown in panel I; (K-L) ELISA to measure the levels of bone resorption markers TRAP5b (K) and CTX-I (L) in serum of different groups of rats. Each group consisted of 6 rats. * indicates P < 0.05 compared to the control group, ** indicates P < 0.01, *** indicates P < 0.001

Fig. 4

EVs-mediated regulation of GFAP expression by miR-15b-5p. Note: (A) Prediction of miRNA associated with GFAP using the Norwalk, minimap, and miRDB databases, with the middle section representing the intersection of the three data sets; (B) RT-qPCR analysis of miR-15b-5p expression levels in bone tissues of different groups of rats; (C-D) Dual-luciferase reporter gene experiment to verify the targeted binding relationship between miR-15b-5p and GFAP; (E) RT-qPCR analysis of miR-15b-5p expression in BMSCs and EVs; (F) Fluorescence microscopy to detect Cy3-labeled fluorescence signals (Scale bar = 20 μm); (G) RT-qPCR analysis of miR-15b-5p and GFAP expression levels in osteoclasts; (H-I) Western blot analysis of GFAP expression levels in different groups of osteoclasts. Each group consisted of 6 rats. The cell experiments were repeated three times. * indicates P < 0.05 compared to the control group, ** indicates P < 0.01, *** indicates P < 0.001

Fig. 5

Characterization of GMNP_E-EVs. Note: (A) Schematic diagram of GMNP_E-EVs formation; (B) Infrared spectroscopy analysis of the chemical composition at each step of GMNP_E formation; (C) Elemental analysis by energy-dispersive X-ray spectroscopy (EDS); (D) SEM observation of MNP, GMNP, and GMNP_E (scale bars = 1 μm); (E) Particle size distribution of GMNP_E at key steps of formation; (F) Changes in Zeta potential of GMNP_E before and after stepwise modification; (G) Magnetic hysteresis loop of GMNP_E; (H) Confocal microscopy imaging of anti-CD63 on GMNP_E surface (bar = 25 μm); (I) Measurement of human serum albumin (HSA) adsorption on GMNP_E and GMNP_N nanoparticles after incubation at 37 °C for 3 days, FITC-labeled HSA, fluorescence spectroscopy used to measure the adsorption amount of HSA; (J) TGA analysis of antibody content in GMNP_E. All experiments were repeated three times. * indicates difference between two groups with p < 0.05, ** indicates difference with p < 0.01, *** indicates difference with p < 0.001

Fig. 6

Enrichment of GMNP_E-EVs in DO rat bone tissue. Note: (A) TEM imaging showing the clear core–shell corona structure of GMNP_E nanoparticles and the double-layer membrane structure of EVs, top left bar = 100 nm, top right bar = 20 nm, bottom right bar = 10 nm; (B) Representative confocal microscopy imaging of GMNP_E-EVs, co-localization of MNPs (red) and EVs (green), bar = 25 μm; (C) Binding capacity of GMNP_E and GMNP_N with EVs; (D) Western blot analysis of CD63 expression in GMNP_N-EVs, GMNP_E-EVs, and EVs lysates; (E) CCK-8 assay to evaluate the effect of nanoparticles on osteoclast proliferation; (F) Confocal microscopy images showing the localization of GMNP_E-EVs in cells, bar = 25 μm; (G) Confocal microscopy images showing the localization of GMNP_N-EVs and GMNP_E-EVs in rat body, bar = 25 μm; (H) Relative fluorescence intensity of EVs in image G; (I) Immunofluorescence detection of EVs (green) and GMNP_E (red)-EVs (green) signals in brain, heart, lung, liver, spleen, and kidney tissue cells, bar = 25 μm. Each group consisted of 3 rats, and cell experiments were repeated 3 times. * indicates difference between two groups, ** indicates difference with p < 0.01, *** indicates difference with p < 0.001

Fig. 7

Effect of miR-15b-5p-GMNP_E-EVs targeted inhibition of GFAP on osteoclastogenesis. Note: (A) Gene expression levels of GFAP and miR-15b-5p in transfected cells detected by RT-qPCR; (B) CCK-8 assay to assess the proliferation of osteoclasts co-cultured with GMNP_E-EVs; (C-D) TRAP staining to assess osteoclast differentiation of BMMs after transfection (MNC ≥ 3 nuclei were considered positive); (E–G) Gene expression levels of TRAP, NFATC1, MMP9, and CTSK in transfected cells measured by RT-qPCR and Western blot. All cell experiments were repeated three times. * indicates difference between two groups with p < 0.05, ** indicates difference with p < 0.01, *** indicates difference with p < 0.001

Fig. 8

Influence of miR-15b-5p-GMNP_E-EVs targeting GFAP in DO rats. Note: (A-B) RT-qPCR analysis of miR-15b-5p (A) and GFAP (B) expression in DO rat bone tissue; (C) Observation of trabecular bone area in femoral sections of DO rats using H&E staining; (D) Quantification of absorption area in femoral sections of DO rats from graph B; (E–F) Quantitative analysis of TRAP-positive cells in bone sections; (G) Micro-CT evaluation of bone formation in femoral tissue of DO rats; (H) Statistical analysis of BV/TV, Tb. N, Tb.Th, and BMD; (I-J) ELISA detection of serum TRAP 5b (I) and CTX-1 (J) levels in rats. Each group consisted of 6 rats. * indicates difference between two groups with p < 0.05, ** indicates difference with p < 0.01, *** indicates difference with p < 0.001

Fig. 9

Molecular mechanism diagram of MNPs loaded with BMSCs-derived EVs delivering miR-15b-5p to regulate GFAP expression and promote osteoclastogenesis in DO progression