Systemic SHED transplantation rescues impaired BMMSC function through telomerase activity in recipient OVX mice
In this study, female C57BL/6 mice (10 weeks old) were ovariectomized and used to understand whether estrogen deficiency affects BMMSCs. The CFU-F and BrdU assays showed enhanced colony formation and decreased proliferation, respectively, in BMMSCs derived from OVX mice (OVX-BMMSCs) in comparison to that of BMMSCs derived from Sham-mice (Figures 1a, 1b). FCM results revealed that only CD146 and CD73 levels were decreased in OVX-BMMSCs compared to Sham-BMMSCs (Figure 1c). Furthermore, Alizarin Red staining and RT-qPCR revealed that OVX-BMMSCs exhibited impaired osteogenesis in osteogenic inductive conditions as observed by a reduction in formation of mineralized nodules and mRNA levels for osteogenic markers, including Runt-related transcription factor 2 (Runx2) and bone gamma carboxyglutamic acid protein (Bglap) (Figures 1d–1f). OVX-BMMSCs also showed decreased expression of the osteoprotective factor Semaphorin-3a (Sema3a) under these conditions (Figure 1g). Meanwhile, results from CFU-F, BrdU, and FCM assays showed that SHED transplantation rescued the abnormally functioning recipient BMMSCs (SHED transplanted mice derived BMMSCs: SHED-BMMSCs) as seen by decreased colony formation as well as increased cell proliferation, and CD146 and CD73 levels. We also detected increased mineralized nodule formation and mRNA levels of Runx2, Bglap, and Sema3a under osteogenic conditions, indicating recovery of the impaired osteogenic function of recipient SHED-BMMSCs (Figures 1a–1g). Thus, SHED transplantation appears to rescue the impaired function of recipient BMMSCs and bone reduction in OVX mice.
Since low telomerase levels have been shown to regulate osteogenesis by BMMSCs [19,22], we next examined the difference in telomerase activity between OVX-BMMSCs and Sham-BMMSCs. OVX-BMMSCs showed a significant decrease in telomerase activity in comparison with that of Sham-BMMSCs as determined via the real-time RQ-TRAP (Figure 1h). Interestingly, SHED transplantation recovered the telomerase activity of recipient BMMSCs in OVX mice (Figure 1h).
Additionally, TERT has been shown to play an essential role in regulating telomerase activity [16]. We, therefore examined whether TERT contributes to the improved telomerase activity in recipient OVX-BMMSCs, and confirmed, via RT-qPCR, rescue of the Tert mRNA level in OVX-BMMSCs upon SHED transplantation (Figure 1i). Further, although SHED were detected in the bone marrow seven days after the infusion; CFSE labeling showed no significant difference in the number of SHED in the recipient bone and bone marrow (Figure 1j). Hence, these results indicate that the status of TERT-associated telomerase activity is crucial for the SHED transplantation-mediated rescue of impaired recipient BMMSC functioning and bone reduction in OVX mice.
Systemic SHED transplantation recovers bone loss in postmenopausal OVX mice
Postmenopausal osteoporosis is a common systemic skeletal disease in elderly women that results from an imbalance between bone resorption by osteoclasts and bone formation by osteoblasts, thereby leading to reduced BMD and deteriorated bone microarchitecture that is a high risk factor for fragility fractures [31]. Systemic SHED transplantation has been found to ameliorate bone reduction in an ovariectomized mouse model for postmenopausal osteoporosis [12,15]. Hence, in the current study, OVX mice received SHED (0.1×106/10 g body weight) two days post-surgery and were used to determine the osteoporotic phenotype of the L3 four weeks post-transplantation. Increased BMD, BV/TV, Tb.N, and Tb.Th were observed by micro-CT, indicating rescue of the osteoporotic phenotype in the treated OVX mice as compared to the control mice (Additional file 1: Supplementary Figure 1).
Enhanced in vivo osteoclast activity in OVX mice was rescued four weeks after SHED transplantation was performed, as evidenced by the observed decrease osteoclast number as well as serum levels of the sRANKL and CTX following TRAP staining and ELISA (Additional file 1: Supplementary Figure 2). Further, the in vitro osteoclast activity of OVX mice was rescued four weeks after SHED transplantation as indicated by the decrease in TRAP-positive multinuclear cells (MNCs) and mRNA levels of the osteoclast markers, including receptor activator for nuclear factor κB (Rank), nuclear factor of activated T-cell (Nfatc1), and cathepsin K (Ctsk) by TRAP staining and RT-qPCR following co-culturing of calvarial osteoblasts from wild-type newborn C57BL/6 mice stimulated with prostaglandin E2 (Additional file 1: Supplementary Figure 3). These results suggest that systemic SHED infusion rescues osteoporotic phenotypes and improves the differentiation and function of osteoclast-lineage cells.
Infusion of SHED-secreted EVs ameliorate bone reduction in OVX mice
Since SHED transplantation increased Tert mRNA levels to rescue telomerase activity in recipient BMMSCs, we hypothesized that trophic factors released from SHED contribute to the SHED transplantation-mediated rescue of impaired Tert mRNA expression, telomerase activity, and bone loss in OVX mice. Particularly, trophic factors within MSC-secreting EVs mediate cell-cell communication to deliver extracellular signals, thereby leading to therapeutic effects [32,33]. Thus, we isolated EVs from the SHED-conditioned medium harvested from a 3-day culture and examined the characteristics of the SHED-EVs. Results show that SHED-EVs were 57–272 nm in diameter with a mean of 86±2.5 nm, as determined using particle tracking analysis (Figure 2a). The concentration of SHED-EVs in the CM was 1.6×109 ± 4.8×108 particles/mL. Moreover, SHED-EVs significantly expressed exosome markers, such as CD9 (83.3 ± 2.5%), CD63 (42.8 ± 3.3%), and CD81 (85.9 ± 6.2%), and reduced expression of the MSC surface marker CD90 (4.1 ± 1.8%) as detected by FCM (Figure 2b). Western blotting showed an enrichment of CD63 and CD81 in the SHED-EVs, however, the same was not observed for calnexin (an endoplasmic reticulum marker), as compared to that SHED (Figure 2c). We also detected miRNAs within SHED-EVs (the concentration of miRNAs and small RNAs were 2.8 ± 0.46 ng/mL and 4.6 ± 0.83 ng/mL, respectively; the proportion of miRNA content in the small RNA was 61.1 ± 2.4%; Figure 2d). The total protein content of SHED-EVs was 828.3 ± 65.8 mg/mL. Next, we depleted the small RNA content, especially miRNAs, of SHED-EVs by treating with RNase (5 U/mL) for 30 min at 37 °C (resulting concentrations of miRNAs and small RNAs were 1.5 ± 0.43 ng/mL and 2.3 ± 0.28 ng/mL, respectively; the proportion of miRNA content in the small RNA was 62.8 ± 3.8%; Figure 2e); RNase treatment did not affect the particle size of the SHED-EVs (Figure 2f) nor their membrane surface phenotype and total protein content (data not shown). These findings indicate that SHED-EVs released into the CM are miRNA-enriched exosomes.
Next, to determine whether EVs released from SHED are an alternative to SHED themselves in postmenopausal osteoporosis therapy, we intravenously infused SHED-EVs (100 μg/mouse) into OVX mice two days post-surgery. After four weeks of intravenously administering SHED-EVs, the osteoporotic phenotype of OVX mice was observed to be rescued as indicated by the marked increase in BMD, BV/TV, Tb.N, and Tb.Th by micro-CT analysis (Figures 3a–3c). The enhanced in vivo and in vitro osteoclast activity in OVX mice was also rescued four weeks after administering SHED-EVs, as indicated by the decrease in the TRAP-positive cells and serum levels of sRANKL and CTX (Additional file 1: Supplementary Figures 4a–c) as well as reduced TRAP-positive MNCs and mRNA levels of Rank, Nfatc1 and Ctsk (Additional file 1: Supplementary Figures 5a–c). Moreover, to investigate whether RNAs within SHED-EVs play a role in their therapeutic effect, we systemically infused OVX mice with SHED-EVs pretreated with or without RNase for 30 min at 37 °C and examined the osteoporotic phenotype after four weeks. RNase-pretreated SHED-EVs showed no significant recovery of the osteoporotic phenotype, differentiation, or function of osteoclast-lineage cells (Figure 3, Additional file 1: Supplementary Figures 4, 5). These findings indicate that RNAs, specifically miRNAs, in SHED-EVs contribute to bone loss rescue in OVX mice.
Systemic administration of SHED-EVs rescue impaired BMMSC function via telomerase activity in recipient OVX mice
We next examined the effect of SHED-EV infusion on impaired BMMSCs in recipient OVX mice and found that recipient BMMSCs recovered Tert mRNA levels and telomerase activity, as determined via RT-qPCR and RQ-TRAP, respectively (Figures 4a, 4b). Furthermore, infusion of SHED-EVs improved CFU-F formation, cell proliferation, as well as CD146 and CD73 expression in recipient BMMSCs, as detected by the CFU-F, BrdU-labeling, and FCM assays, respectively (Figures 4c–4e). Infusion of SHED-EVs also markedly rescued impaired osteogenesis as indicated by the increase in mineralized nodule formation and mRNA levels of Runx2, Bglap, and Sema3a by Alizarin Red staining and RT-qPCR under osteogenic conditions (Figures 4f–4i). However, RNase-treated SHED-EVs did not rescue colony formation, cell proliferation, telomerase activity, osteogenesis, or Sema3a expression in recipient BMMSCs (Figure 4).
SHED-EVs ameliorate the properties of hBMMSCs
To determine the direct effects of SHED-EVs on BMMSCs in mediating cell-cell communication, we incubated hBMMSCs with or without SHED-EVs. Fluorescence microscopy and FCM revealed uptake of SHED-EVs into hBMMSCs (Figures 5a, 5b). SHED-EVs significantly increased the mRNA levels of TERT and telomerase activity when compared to that in MOCK-hBMMSCs, by RT-qPCR and RQ-TRAP, respectively (Figures 5c, 5d). FCM and BrdU results showed that SHED-EVs upregulated the expression of CD146 and cell proliferation, respectively (Figures 5e, 5f). SHED-EVs also enhanced the in vitro osteogenic functions of hBMMSCs, as indicated by increased mineralized nodule formation and mRNA levels of RUNX2, BGLAP, and SEMA3A as determined via Alizarin Red staining and RT-qPCR under osteogenic conditions (Figures 5g–5j).
Lastly, to examine whether RNAs from SHED-EVs regulate SHED-EV-mediated enhancement of hBMMSC properties, we assayed hBMMSCs incubated with RNase-pretreated SHED-EVs and found that RNase-pretreated SHED-EVs attenuated the SHED-EV-mediated rescue of TERT mRNA levels, telomerase activity, CD146 expression, and cell proliferation in hBMMSCs (Figures 6a–6d). RNase-pretreated SHED-EVs also attenuated SHED-EV-mediated rescue of mineralized nodule formation, upregulation of RUNX2, BGLAP, and SEMA3A mRNA levels, and enhanced de novo bone formation in hBMMSCs (Figures 6e–6h). Finally, SHED-EV-pretreated hBMMSCs exhibited a significant increase in bone formation when implanted into immunocompromised mice subcutaneously using HA/TCP as a carrier (Figures 6i, 6j). RNase-pretreated SHED-EVs also attenuated enhanced de novo bone formation in hBMMSCs (Figure 6i, 6j).