Tonsil-derived mesenchymal stem cell-embedded in situ crosslinkable gelatin hydrogel therapy recovers postmenopausal osteoporosis through bone regeneration

Abstract

We investigated therapeutic potential of human tonsil-derived mesenchymal stem cells (TMSC) subcutaneously delivered to ovariectomized (OVX) mice for developing more safe and effective therapy for osteoporosis. TMSC were isolated from tonsil tissues of children undergoing tonsillectomy, and TMSC-embedded in situ crosslinkable gelatin-hydroxyphenyl propionic acid hydrogel (TMSC-GHH) or TMSC alone were delivered subcutaneously to the dorsa of OVX mice. After 3 months, three-dimensionally reconstructed micro-computed tomographic images revealed better recovery of the femoral heads in OVX mice treated with TMSC-GHH. Serum osteocalcin and alkaline phosphatase were also recovered, indicating bone formation only in TMSC-GHH-treated mice, and absence in hypercalcemia or other severe macroscopic deformities showed biocompatibility of TMSC-GHH. Additionally, visceral fat reduction effects by TMSC-GHH further supported their therapeutic potential. TMSC provided therapeutic benefits toward osteoporosis only when embedded in GHH, and showed potential as a supplement or alternative to current therapies.

Fig 1. General characteristics of the OVX mouse model.

(a) Body weight changes of OVX and non-OVX mice were traced from 8 weeks of age, when OVX was performed, to the end of the experimental period. (b) Serum OCN concentrations were measured in sera collected from non-OVX and OVX mice at 0 and 3 months after OVX. (c) 3D-constructed microCT images of the femoral head trabeculae of non-OVX and OVX mice at necropsy. (d) BMD and BV/TV values of the femoral head trabecular regions at necropsy from non-OVX and OVX mice, as analyzed from microCT images. Statistical significance is marked as ***P < 0.001.

Fig 2. Schematic process used to prepare TMSC-GHH for injection, and overall experimental timeline.

(a) GHPA polymers were enzymatically in situ crosslinked to form hydrogel, and TMSC were embedded in the hydrogel meshwork. (b) For TMSC-GHH injections, one solution containing GHPA polymers, HRP, and TMSC was mixed with another solution containing GHPA polymers and H₂O₂ at the time of injections. (c) Experimental procedures including the injection schedule and allotted durations.

Fig 3. In vitro viability assay of TMSC-GHH over a 3-week period, and macroscopic morphology of TMSC-GHH implants.

(a) TMSC-embedded GHH were plated on a 24-well plate, and LIVE/DEAD^® assays were performed at designated times. Horizontal sections of the entire depth of the TMSC-GHH were viewed under a microscope (×10 magnification). Live cells appear green (top row) and dead cells appear red (bottom row). (b) Fluorescence signals from LIVE/DEAD^® assays were quantified and expressed in arbitrary units (a.u.). (c) TMSC-GHH was injected subcutaneously in the dorsum. (d) Three months after treatment, TMSC-GHH-treated mice were anesthetized, and the subcutaneous layer of the dorsum was exposed to photograph the remaining hydrogel and nearby blood vessels.

Fig 4. MicroCT images and BMD of femoral head trabeculae at the experimental endpoint.

(a) Representative horizontal and coronal cross-sectional images capturing the proximal end of a femur as well as a 3D-constructed image of the femoral head trabecular bone for each group. (b) BMD values calculated from the microCT images of femoral head trabeculae. Statistical significance is marked as *P < 0.05 against Untreated group, and §P < 0.05 against Estrogen group.

Fig 5. Serum OCN, ALP, and calcium levels of OVX mice.

(a,b) Measurements of OCN and ALP levels from sera sampled at 3 months after first treatment. (c) The serum total calcium levels without adverse effects. Statistical significance is marked as *P<0.05 and **P<0.01.

Fig 6. Postmortem kidney and liver samples from OVX mice.

(a) Macroscopic morphology of representative postmortem kidneys and liver obtained from each group at necropsy. (b) Mean body weight of each group 3 months after first treatment. (c) Mean body weight-adjusted kidney mass values at 3 months after treatment. (d) Mean body weight-adjusted liver mass values at 3 months after first treatment. Statistical significance is marked as *P < 0.05.

Fig 7. Body weights and visceral fat masses of OVX mice at the experimental endpoint.

Visceral fat pads from periovarian and parametrial regions were collected and weighed at necropsy, which took place 3 months after treatment. (a) Mean visceral fat mass of each group. (b) Mean body weight-adjusted visceral fat mass values. (c) Representative macroscopic observations from each group of internal organs and fat pads after the peritoneum was cut open at necropsy. Statistical significance is marked as *P < 0.05 and **P < 0.01.

Fig 8. Schematic representation of bone regeneration by TMSC-embedded GHH.

(a) Bilateral OVX osteoporosis mouse model was used for investigating bone formation effect of TMSC-embedded GHH. (b) GHPA polymers were enzymatically in situ crosslinked and formed GHH, and TMSC-embedded GHH was subcutaneously injected into OVX mice. (c) Improved microCT results were observed in TMSC-GHH-treated mice 3 months after the initial treatment.