Human bone marrow derived stromal cells (HMSCs) were obtained from Lonza and cultured in αMEM basal media (Gibco) supplemented with 20% fetal bovine serum (FBS, Gibco), 1% l-glutamate (Gibco) and 1% antibiotic, antimycotic solution (Gibco) (Huang et al., 2020b). Primary bone marrow derived monocytes were isolated from mouse marrow as per our previously published protocols (Kang et al., 2020a; Kang et al., 2020b; Kang et al., 2022). The monocytes were differentiated into macrophages using macrophage colony stimulating factor (M-CSF) and subsequently polarized into pro-inflammatory (M1) or pro-reparative (M2) macrophages using LPS/IFNγ and IL-4 respectively as per previously published methodology (Kang et al., 2020a; Kang et al., 2020b; Kang et al., 2022).

EV-specific overexpression of miRNA-424 was carried out as per our published protocol (Mathew et al., 2023). Lentiviral particles containing a mammalian dual promoter vector that encodes for miRNA-424 followed by the EV targeting sequence under the control of EF1α promoter and a GFP marker under the control of SV40 promoter was custom generated by System Biosciences LLC and used for HMSC transduction. The vector was designed to target the 5p strand of the miRNA to the EVs. Positively transfected cells were selected for stable expression using puromycin. In parallel, a control vector without the miRNA was also transduced in HMSCs and similarly selected (referred to as GFP EV). The increase in expression of miR-424 3p and 5p fragments was verified by qRT-PCR from both cell and EV isolates as per our published protocols for cellular and EV miRNA quantitation (Huang et al., 2020b).

Functionally engineered EVs (FEEs) were isolated from the miR-424 expressing HMSCs as per standardized protocols (Narayanan et al., 2016; Huang et al., 2019; Huang et al., 2020b). Before isolation, the HMSCs were washed in serum free basal medium and cultured for 24 h in serum free basal medium. The conditioned medium was harvested and centrifuged at 3000xg to remove cellular debris. The medium was then concentrated five-fold (20% of original volume) using a 100 kDa cutoff spin filter (Millipore) and EVs were isolated from this concentrated medium using the ExoQuick TC isolation reagent (System Biosciences LLC). EVs isolated from the cells were resuspended in PBS at a concentration of 1.8 × 10¹⁰ particles/ml. The isolated EVs were characterized for the presence of markers CD63 (1/250, Abcam), CD81 (1/250, Abcam) and TSG101 (1/250, Abcam) by immunoblotting. Nano particle tracking analysis (NTA) was performed to obtain the size distribution of the isolated EVs.

Primary bone marrow macrophages and HMSCs were cultured as described above in 96 well tissue culture plates (10,000 cells/well). GFP EVs and miR-424 EV were fluorescently labeled as per published protocols (Huang et al., 2020a; Huang et al., 2021). Dose dependent endocytosis was performed using GFP EVs and miR-424 EV using similar amounts of EVs. For qualitative experiments, the cells were cultured in 12 well dishes containing cover glass (50,000 cells/well). The cells were treated with 1.8 × 10⁹ fluorescently labeled GFP or miR-424 EV for 2 h at 37°C. The cells were fixed in formalin, permeabilized counterstained for actin using phalloidin TRITC (Sigma) and imaged using a Zeiss LSM 710 Meta confocal microscope.

EV functionality was assessed at the level of MSC differentiation and BMP2 pathway activation by the miR-424 EV in comparison with vector control GFP EVs. MSC differentiation was assessed first by alizarin red staining. Briefly 100,000 HMSCs were seeded in 6 well culture dishes and subjected to osteogenic differentiation for 2 weeks using a combination of ascorbic acid, β-glycerophosphate and dexamethasone as per previously published protocols (Kang et al., 2022) in the presence/absence of control (GFP EVs) or miR-424 EV (3.6 × 10⁹ EVs/well). The culture medium was changed once in 2 days and the EVs were added to culture each time the medium was changed. Post 2-week culture, the cells were fixed and subjected to alizarin red staining to analyze the presence of calcium deposits qualitatively.

A BMP2 response element luciferase assay was performed to analyze the triggering of BMP2 pathway. Briefly, HMSCs (50,000/well) seeded in 12 well plates were transfected with an SBE12 luciferase reporter plasmid that contains a BMP2 response element as per previously published methodology (Huang et al., 2020b). The transfected cells (48 h post transfection) were then treated with control or miR-424 EV (1.8 × 10⁹ EVs/well) as well as 250 ng/ml rh BMP2 (positive control). 48 h post transfection, the total protein from the cells were harvested and luciferase activity was measured using a BioTek microtiter plate reader as per published methodology (Huang et al., 2020b). For these experiments n = 4 replicates.

The ability of the miR-424 EV to trigger MSC differentiation was evaluated by qRT-PCR for osteoinductive gene expression. For this experiment, HMSCs were seeded in 12 well culture dishes and treated with miR-424 EV (24 h post seeding) as described above (n = 4). 72 h post treatment, RNA was isolated from untreated and treated wells and subjected to qRT-PCR for osteoinductive genes (VEGFa, TGFβ1, osterix, Runx2, BMP2, BMP6 and BMP9). Data were represented as mean fold change in gene expression over control.

A similar experiment to the one described above was performed to evaluate SMAD 1/5/8 phosphorylation by miR-424 EV. Here, total protein was isolated from the cells and immunoblotting was performed for phosphorylated SMAD 1/5/8 24 and 48 h post treatment with the EVs or rhBMP2. Tubulin served as loading control. The blots were imaged using a Licor Fluorescence imager. For quantitative measurements, an in-cell western was performed in 96 well plates (10,000 cell/well and 3.6 × 10⁸ EVs/well, n = 6).

The effect of control and miR-424 EV on macrophages was evaluated in vitro. Primary mouse bone marrow derived macrophages were polarized towards inflammatory M1-like phenotype or reparative M2-like phenotype as described earlier in the presence/absence of control (GFP-EVs) or miR-424 EV. The expression levels of pro-inflammatory markers (IL-1β, TNFα and MCP-1) and reparative phenotype markers (Ym-1 and Arg-1) in M1 and M2 polarized macrophages respectively were analyzed by qRT-PCR as per established and published methodology (Kang et al., 2020a; Kang et al., 2020b; Kang et al., 2022).

All animal experiments were performed in accordance with protocols approved by the UIC animal care committee (ACC, Assurance No: A3460.01). All groups and time points contained 6 defects per group. Briefly, the rats were anesthetized intraperitoneally using Ketamine (80–100 mg/kg)/Xylazine (10 mg/kg). Using aseptic technique, a vertical incision was made in the head at the midline to expose the calvarial bone. The connective tissue was removed and two 5 mm calvarial defects were created bilaterally in the calvarium without dura perforation using a trephine burr. Each defect was covered by a collagen plug (5 mm diameter, 1 mm thickness) containing 9 × 10⁹ FEEs/control EVs/no EVs (PBS). Days 1, 3 and 7 post-surgeries, the tissue around the defects were extracted for qualitative and quantitative IHC. The tissues were fixed in neutral buffered formalin, embedded in paraffin and 5 μm sections were stained for Arg1 and iNOS as markers for reparative and inflammatory cells. The number of positive cells per field of view was counted using ImageJ software and used to obtain quantitative data (n = 6). Four- and 8-week post-surgery, the rats were euthanized by carbon dioxide asphyxiation and cervical dislocation. The calvaria were dissected, fixed in neutral buffered 10% formalin and scanned using a μCT scanner. The data from the scan was quantitatively analyzed using a custom MatLab software to obtain BV/TV data for all the groups and time points. The samples were then decalcified in 10% EDTA solution, embedded in paraffin, and 10 μm sections were subjected to H&E staining and immunohistochemistry (IHC) for bone morphogenetic protein 2 (BMP2, 1/250, Abcam), bone sialoprotein (BSP, 1/250, Abcam), dentin matrix protein 1 (DMP1, 1/250, Santacruz Biotechnology) and osteocalcin (OCN, 1/250, Abcam).

The normal distribution of the data obtained from the experiments was evaluated using the Shapiro-Wilk test. For experiments involving two groups, student’s t-test with a confidence interval of 95% was utilized. For the experiments involving comparison of more than two groups, one-way ANOVA was performed with a confidence interval of 95%. Pairwise comparisons were performed using Tukey’s ad hoc test with a confidence interval of 95%.

The expression levels of miR-424 3p and 5p strands were verified by qRT-PCR with respect to EVs from vector control (GFP EV). Figure 1A shows these results. A differential expression pattern of the strands between cellular and EV expression was noted indicating the EV-specific targeting of the miRNA 5p strand. The expression of both membranous (CD63, CD81) and intra EV (TSG101) marker proteins was verified by immunoblotting. The absence of intracellular proteins was verified by probing for tubulin using cell lysate as a control. Results presented in Figure 1B indicate that both vector control and miR-424 EV express EV marker proteins. The size distribution of the isolated EVs was evaluated by nanoparticle tracking analysis (NTA). Results indicated a similar size distribution between naïve MSC EVs, vector control GFP EVs and miR-424 EV indicating that EV-specific overexpression of miR-424 does not alter EV physical properties, marker expression and size distribution.

We have published previously that MSC EVs are endocytosed by multiple cell types in a dose dependent and saturable manner (Huang et al., 2020a). Therefore, we verified if altering the EVs by either transfecting the vector or the miRNA construct alters its endocytic ability. The ability of the engineered EVs to be endocytosed by MSCs was verified qualitatively and quantitatively by immunocytochemistry and quantitative fluorescence measurements. Figure 2 shows the results of these experiments. Figure 2A shows the endocytosis of fluorescently labeled (green) control (GFP-EVs, left panel) and miR-424 EV (right panel) by HMSCs. The cells were counterstained using phalloidin-tritc for actin labeling and the nuclei were stained using DAPI. Figure 2B shows a dose dependent and saturable endocytosis of fluorescently labeled control and miR-424 EV by HMSCs.

miR-424 EV were generated to trigger osteogenic differentiation of naïve MSCs by enhancing the activity of the SMAD 1/5/8 signaling cascade. This property was evaluated in vitro qualitatively and quantitatively. First, the ability of the EVs to enhance the differentiation of HMSCs was verified by evaluating calcium deposition under the influence of osteogenic differentiation medium. Figure 3A shows the results of alizarin red staining of control and miR-424 EV treated HMSCs under differentiation conditions after 2 weeks in culture. Macroscopic (3A) and microscopic (3B) evaluation indicates that miR-424 EV promote enhanced calcium deposition under differentiation conditions compared to controls (differentiation conditions and differentiation conditions with control GFP EVs). Undifferentiated cells did not show the presence of calcium deposits. Following this, the ability of the EVs to induce the expression of osteoinductive genes was evaluated by qRT-PCR following 72 h of treatment of HMSCs with miR-424 EV compared to control. Results presented in Figure 3C shows that miR-424 EV triggered a significant increase (denoted by *) in the expression levels of several osteoinductive genes that include growth and transcription factors.

To verify the induction of the SMAD 1/5/8 signaling cascade, we first evaluated the activity of the pathway using a luciferase assay that incorporates the SBE12 BMP2 response element. Our laboratory has utilized in earlier studies to evaluate the triggering of this pathway (Huang et al., 2020b). Figure 4A shows the results of this experiment. As is evident from the figure, positive control rhBMP2 triggered increased luciferase activity 48 h post treatment compared to untreated controls. The only other group to trigger significant increase in luciferase activity was the miR-424 EV group. This increase was statistically significant (n = 4, p < 0.05, Tukey’s test post ANOVA) when compared to untreated control group (*) as well as the control EV group (#). miR-424 belongs to a cluster of miRNAs that enhance endogenous SMAD 1/5/8 phosphorylation (Vimalraj et al., 2014; Xiao et al., 2015; Baptista et al., 2018; Pande et al., 2018; Ma et al., 2020). Therefore, we performed immunoblotting for detecting phosphorylated SMAD 1/5/8 in HMSCs treated with miR-424 EVs. rhBMP2 was used as a positive control for these experiments. Results presented in Figure 4B show that 24 and 48 h post treatment with either rhBMP2 or miR-424 EV, SMAD phosphorylation was increased. Quantitative measurement of this phenomenon was performed using an in-cell assay with tubulin as an internal control. Images and the corresponding normalized quantitation (n = 6) presented in Figure 4C indicate that miR-424 EV treatment significantly enhanced the presence of phosphorylated SMAD 1/5/8 in HMSCs. Taken together, the results presented in Figures 3, 4 collectively indicate that miR-424 EV can trigger osteogenic differentiation of HMSCs in vitro by potentiating the BMP2 signaling cascade.

MSC derived EVs possess inherent immunomodulatory properties that are designed to reduce inflammation by negatively regulating pro-inflammatory cytokines. Therefore, it is important for engineered EVs to maintain this property whilst also possessing enhanced pathway-specific activity they are engineered to target. We compared the activity of the miR-424 EV along with control EVs on primary mouse bone marrow derived macrophages. Both the control and miR-424 EV were dose dependently and saturably endocytosed by macrophages (Figure 5A). When macrophages were polarized to M1 phenotype in the presence/absence of EVs, both control and miR-424 EV triggered similar responses by significantly reducing the expression levels of IL-1β, TNFα and MCP-1 (Figure 5B). When similar treatment was evaluated on M2 polarized macrophages, both control and miR-424 EV significantly enhanced Ym-1 expression indicating similar in vitro activity on macrophages.

We evaluated two different properties of miR-424 EV. First, we evaluated the immunomodulatory properties of the engineered EVs in vivo by performing immunohistochemical evaluation of pro-inflammatory marker iNOS and pro-reparative macrophage marker Arg1. The representative images for post-operative days 1, 3 and 7 for iNOS and Arg-1 and their respective quantitation are provided in Figures 6A,B respectively. Results presented in Figure 6A show that miR-424 EV significantly reduced the number of iNOS positive cells at day 1 compared to untreated control group as well as the GFP-EV group. However, while both EV groups reduced the number of iNOS positive cells at days 3 and 7 compared to controls, no significant differences were observed between them at these time points. On the other hand, miR-424 EV significantly enhanced the number of Arg1 positive cells at compared to both control groups at all time points.

Quantitative μCT was performed to evaluate bone regeneration 4- and 8-week post-surgery. Representative images presented in Figure 7A show increased bone formation in the miR-424 groups at 4 weeks compared to other control groups. Volumetric quantitation shown in Figure 7A shows that miR-424 EV triggered a significant increase in bone volume in the defects at both 4 and 8 weeks indicating their ability to enhance bone repair. Histological evaluation by means of H&E staining (Figure 7B) corroborated the μCT results with the miR-424 EV treated group showing increased presence of bone (arrows) compared to the control groups at 4 and 8 weeks. Finally, IHC was performed to evaluate the expression levels of osteogenic proteins in the defects. Results presented in Figure 8 indicate the expression of BMP2, BSP, DMP1 and OCN in representative sections from all groups at 4 and 8 weeks. While the increased presence of bone tissue in the miR-424 EV group is evident from these images, due to the varying degrees of the presence of connective tissue, the expression levels could not be quantitatively discerned. However, it is evident from the images that both EV treated groups show enhanced expression of the markers compared to the control group.

Collectively, the results presented in Figures 6–8 indicate that the engineered miR-424 EV possess immunomodulatory properties on par with control EVs, show enhanced propensity to trigger/recruit reparative macrophages (Arg1 positive cells) and promote enhanced bone formation.