Periosteum progenitors could stimulate bone regeneration in aged murine bone defect model

Abstract Periosteal stem cells are critical for bone regeneration, while the numbers will decrease with age. This study focused on whether Prx1+ cell, a kind of periosteal stem cell, could stimulate bone regeneration in aged mice. Four weeks and 12 months old Prx1CreER‐GFP; Rosa26tdTomato mice were used to reveal the degree of Prx1+ cells participating in the femoral fracture healing procedure. One week, 8 weeks, 12 and 24 months old Prx1CreER‐GFP mice were used to analyse the real‐time distribution of Prx1+ cells. Twelve months old C57BL/6 male mice (n = 96) were used to create the bone defect model and, respectively, received hydrogel, hydrogel with Prx1− mesenchymal stem cells and hydrogel with Prx1+ cells. H&E staining, Synchrotron radiation‐microcomputed tomography and mechanical test were used to analyse the healing results. The results showed that tdTomato+ cells were involved in bone regeneration, especially in young mice. At the same time, GFP+ cells decreased significantly with age. The Prx1+ cells group could significantly improve bone regeneration in the murine bone defect model via directly differentiating into osteoblasts and had better osteogenic differentiation ability than Prx1− mesenchymal stem cells. Our finding revealed that the quantity of Prx1+ cells might account for decreased bone regeneration ability in aged mice, and transplantation of Prx1+ cells could improve bone regeneration at the bone defect site.

chymal stem cells and hydrogel with Prx1 + cells. H&E staining, Synchrotron radiationmicrocomputed tomography and mechanical test were used to analyse the healing results. The results showed that tdTomato + cells were involved in bone regeneration, especially in young mice. At the same time, GFP + cells decreased significantly with age. The Prx1 + cells group could significantly improve bone regeneration in the murine bone defect model via directly differentiating into osteoblasts and had better osteogenic differentiation ability than Prx1 − mesenchymal stem cells. Our finding revealed that the quantity of Prx1 + cells might account for decreased bone regeneration ability in aged mice, and transplantation of Prx1 + cells could improve bone regeneration at the bone defect site.

K E Y W O R D S
aged mice, bone defect, bone regeneration, periosteum derived cells, Prx1 − MSC, Prx1 + MSC Bone regeneration occurs by two major ossification processes, endochondral ossification in which the skeletal element first develops as a cartilaginous template that is subsequently replaced by bone, and intramembranous ossification in which mesenchymal cells directly differentiate into bone-forming osteoblasts. The ossification process does not require pre-existing cartilage. 7,8 Both procedures require the replenishment of the osteogenic or chondrogenic progenitor cells that participate in bone or cartilage formation during normal development and under pathologic conditions, such as fracture healing. 9,10 In general, osteogenic progenitors distribute in various bone compartments along the bone's outer surface within the periosteum and the inner surface of bone within the endosteum. 11,12 Histological, periosteum is composed of at least two layers, outer fibrous layer and inner cambium layer. 13 The outer fibrous layer mainly contains fibroblastic cells, while the inner cambium layer contains several types of cells, such as fibroblasts, mesenchymal stem cells, osteogenic or chondrogenic progenitors. 14 Many works have attempted to investigate the periosteal stem cells and determine the cell types. [15][16][17] However, the identity of periosteal stem cells remains unclear.
A previous study found that Prx1 was a paired-related homeobox gene expressed in a subset of periosteal cells in the cambium layer surrounding the long bones and cartilage. 18 These Prx1 positive cells (Prx1 + MSCs) have the potential of osteogenic and chondrogenic differentiation and are essential for limb development and bone regeneration. [18][19][20] Duchamp de Lageneste et al found that Prx1 + MSCs can efficiently contribute to cartilage and bone regeneration, while periostin was essential for maintaining the Prx1 + MSCs pool. 21 Therefore, we wondered that if the quantity of Prx1 + MSCs would decrease with age and affect bone regeneration ability in aged mice.
In this study, we verified the role of Prx1 + MSCs in the young murine bone fracture model and its distribution characterization with age. We isolated Prx1 + MSCs and Prx1 negative mesenchymal stem cells (Prx1 − MSCs) from murine periosteum and used them to enhance bone defect regeneration in aged murine bone defect model. We also compared the osteogenic ability and proliferation ability between Prx1 + MSCs and Prx1 − MSCs. We found that Prx1 + MSCs had better osteogenic ability than Prx1 − MSCs and could significantly improve bone regeneration in aged murine bone defect model.

| Animal
Prx1CreER-GFP (Stock No. 029211) and Rosa26 tdTomato (Stock No. 007909) mice were purchased from Jackson Laboratory. All the mice were housed in our animal facility in our university with controlled temperature and light cycles (24°C and 12/12 light cycle). This animal study was reviewed and approved by our Institutional Animal Care and Use Committees (No. 201703222).

| Lineage tracing analysis
To verify the specific distribution of Prx1 + MSC on the periosteum of the femoral shaft with ageing, 1 week, 8 weeks, 12 and 24 months old mice (n = 3 per group) were killed, and the femurs were harvested for immunofluorescence analysis. To investigate the degree of Prx1 + MSC participating in bone regeneration in young and aged mice, 4 weeks and 12 months old Prx1CreER-GFP;Rosa26 tdTomato mice (n = 3 per group) were received intraperitoneal injections of 75 mg/kg bodyweight tamoxifen (Sigma-Aldrich) for 5 days before closed left femoral bone fracture was made. 20 At 2 weeks after surgery, the mice were killed, and the femurs were harvested for immunofluorescence.

| Cells isolation
Periosteal mesenchymal stem cells were isolated from C57BL/6 background transgenic mice using a modified method based on previously literature. 21,22 Briefly, hindlimbs were disconnected from the trunk of 4 weeks old Prx1CreER-GFP mice, and the entire attached soft tissues were removed from the bone. After the epiphysis of tibias and femurs on both sides was cut off, the bone marrow was

| Osteogenic differentiation
Cells (1 × 10 5 per well) are seeded into 0.1% gelatin-coated 12-well culture plate. After cells reached 60%-70% confluence, the medium was replaced to OriCell™ C57BL/6 Mouse Mesenchymal Stem Cell Osteogenic Differentiation Medium (Cyagen Biosciences). The medium was refreshed twice a week. After 3 weeks of osteogenic induction, the cells were fixed in 4% paraformaldehyde for 30 minutes and stained with Alizarin Red for 5 minutes. To quantify the osteogenic differentiation, we used the 10% cetyl pyridinium chloride (Sigma-Aldrich) to solubilize the stain for 20 minutes. Then, the OD values of solutions were measured at 560 nm. Besides, cell samples were collected after 7 days of osteogenic induction for ALP staining, ALP activity assay and qRT-PCR.

| ALP staining and activity
After 7 days of osteogenic induction, the cell culture supernatants were collected and centrifuged to remove cell debris, in which ALP activity was then determined using an ALP activity detection kit

| Real-time qPCR
Total RNA was extracted using the TRIzol reagent (Invitrogen), and cDNA was synthesized with a GoScript™ Reverse Transcription System (Promega) according to the manufacturer's instructions. PCR reactions were performed on ABI PRISM ® 7900HT System (Applied Biosystems) with GoTaq ® qPCR Master Mix (Promega). Expression data were uniformly normalized to β-actin, and the relative expression was calculated using the 2 −ΔΔC t method. The primer sequences employed in the current study were listed in Table 1.

| Femoral defect model and cell transplantation
The surgical procedure was performed as previous report. 23 Briefly, 12 months old male C57BL/6 mice were anaesthetized with pentobarbital, and the skin incision was made over the thigh. Then, the left femur surface was exposed, and a hole (0.8 mm in diameter) was drilled into one cortex without drilling into the opposite cortex in the middle shaft. For cell transplantation, every 1 × 10 6 cells were embedded in 20 μL hydrogel using a Flexcell ® Thermacol ® Kit and 20 μL cell hydrogel mixed component was transplanted into the defect site. After surgery, mice were allowed to move freely. At 2 and 4 weeks after surgery, radiographic analysis, histological analysis and biomechanical tests were used to assess the bone regeneration.

| Synchrotron radiation-microcomputed tomography (SR-μCT) analysis
At 2 and 4 weeks after surgery, the left femurs were harvested and fixed in 4% paraformaldehyde. The microstructure of newly formed bone in the bone defect area was evaluated by SR-μCT based on previous report. 24 Each sample was fixed in a table that allowed 180° rotation at the centre of the rotary stage during scanning. Then, we set beam energy, exposure time and sample-to-detector distance at 18.0 keV, 0.5 seconds and 5.0 cm. All the 720 radiographic projections were imaged by the CCD detector with a pixel size of 3.25 μm.
At the same time, dark-field and flat-field images were also captured to reduce the ring artefact during reconstruction. After the projections were transformed into 8-bit slices, the phase retrieval of projected images was performed by PITRE software written by BL13W1. According to the previous report, the bone was extracted from soft tissue using a fixed threshold segmentation after a median filter reduced noise. Morphological parameters of the newly formed bone at the defect site, such as bone volume to total volume ratio (BV/TV) and trabecular thickness (Tb·Th), were calculated.

| Histology
After radiographic assay, fixed samples were decalcified in EDTA, dehydrated in gradient ethanol, embedded in paraffin and then cut into 5 μm slices. The sections were stained with haematoxylin and eosin for general histology analyses.

| Biomechanical testing
The femurs' mechanical properties with drill-hole defects were examined at 2 and 4 weeks after surgery using a three-point bending test.
The intact contralateral femur was also tested as an internal control.

| Immunofluorescence
To trace the cell fate of Prx1 + MSCs and Prx1 − MSCs, they were labelled

| Statistical analysis
All quantitative data were presented as mean ± standard deviation. ANOVA was conducted, followed by Bonferroni multiple comparison post hoc test for comparing variables among groups using GraphPad Prism 7 software. The differences were considered to be statistically significant when P < .05.   Figure 2B). They showed a spindle-like morphology ( Figure 2C) and had the same proliferation ability ( Figure 2D). higher than that in Prx1 − MSC (P < .05; Figure 3D). The osteogenic related gene showed higher expression in the Prx1 + MSC group than that in Prx1 − MSC group, and there was a significant difference between these two groups in ALP (P < .05), Runx2 (P < .05), SP7 (P < .05; Figure 3E).

| Histology analyses
At 2 weeks after surgery, H&E staining showed that the bone de-

| Mechanical test
During the mechanical testing, all samples were cracked at the defect part, and no one was excluded. At 2 weeks after surgery, the Prx1 + MSC group and Prx1 − MSC group exhibited a higher value of failure load when compared with the control group (P < .01 for all), but no significant difference was found between the hydrogel and control group (P > .05). The Prx1 + MSC group also showed a higher value of failure load than the Prx1 − MSC group. At the same time, the Prx1 + MSC group showed a higher value of stiffness than the control group (P < .01), but no significant difference was found among the other groups (P > .05 for all). At 4 weeks after surgery, failure load and stiffness in the four groups increased significantly, and no significant difference was found among them (P > .05 for all; Figure 6).

| Prx1 + MSC could involve bone regeneration via intramembranous ossification
To investigate the fate of the transplanted cells, 12 months old mice were implanted with Prx1 + MSCs and Prx1 + MSCs, which were permanently labelled with GFP using lentivirus. Immunofluorescence showed that the transplanted periosteal stem cells could survive in the healing site and improve bone regeneration via directly differentiating into osteoblasts ( Figure 7A). More dentin matrix protein 1 (DMP1) was found in the Prx1 + MSC group than in Prx1 − MSC group, indicating that Prx1 + MSC was better than Prx1 − MSC on enhancing bone regeneration in aged mice ( Figure 7B). This effect was consistent with the result in vitro.

| D ISCUSS I ON
In this study, we found that Prx1 + MSC were mainly localized within the periosteum and highly participated in bone defect regeneration in young mice, while its number would decrease with age. Prx1 + MSC had better osteogenic differentiation potency than Prx1 − MSC. We According to a previous study, most of the newly formed bone in the marrow cavity will be absorbed, and newly formed bone in the cortical gap will be remodelled comparable into compact bone at post-operative 2 weeks in young mice. 23 Compared with this, the newly formed woven bone remodelling was incomplete in the control group with aged mice, as demonstrated by histology and SR-μCT. The results indicated that bone regeneration ability was indeed impaired in aged mice. In general, senescence is defined by a minimum age of at least 18 months in mice. 34

| CON CLUS ION
In conclusion, transplanting PDCs into the bone defect site in 12 months old mice could stimulate bone regeneration. The decreased bone regeneration ability in aged mice might be related to the dropping of Prx1 + MSC number within the periosteum.

ACK N OWLED G EM ENTS
This work was supported by the National Natural Science

CO N FLI C T O F I NTE R E S T
The authors declare no competing financial interests.

DATA AVA I L A B I L I T Y S TAT E M E N T
The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions.