Ophiopogonin D promotes bone regeneration by stimulating CD31hiEMCNhi vessel formation

Abstract Objectives CD31hiEMCNhi vessels (CD31, also known as PECAM1 [platelet and endothelial cell adhesion molecule 1]; EMCN, endomucin), which are strongly positive for CD31 and endomucin, couple angiogenesis and osteogenesis. However, the role of CD31hiEMCNhi vessels in bone regeneration remains unknown. In the present study, we investigated the role of CD31hiEMCNhi vessels in the process of bone regeneration. Materials and Methods We used endothelial‐specific Krüppel like factor 3 (Klf3) knockout mice and ophiopogonin D treatment to interfere with CD31hiEMCNhi vessel formation. We constructed a bone regeneration model by surgical ablation of the trabecular bone. Immunofluorescence and micro‐computed tomography (CT) were used to detect CD31hiEMCNhi vessels and bone formation. Results CD31hiEMCNhi vessels participate in the process of bone regeneration, such that endothelial‐specific Klf3 knockout mice showed increased CD31hiEMCNhi vessels and osteoprogenitors in the bone regeneration area, and further accelerated bone formation. We also demonstrated that the natural compound, ophiopogonin D, acts as a KLF3 inhibitor to promote vessels formation both in vitro and in vivo. Administration of ophiopogonin D increased the abundance of CD31hiEmcnhi vessels and accelerated bone healing. Conclusions Our findings confirmed the important role of CD31hiEmcnhi vessels in bone regeneration and provided a new target to treat bone fracture or promote bone regeneration.

a specific microenvironment that decides the fate of progenitor cells. [15][16][17][18][19][20] Therefore, it is important to investigate the effects of vessels on tissue regeneration. CD31 hi EMCN hi vessels (CD31, also known as PECAM1 [platelet and endothelial cell adhesion molecule 1]; EMCN, endomucin), which are strongly positive for CD31 and endomucin, are specific vessels in the skeletal system that couple angiogenesis and osteogenesis. 18 Our previous studies demonstrated that inducing CD31 hi EMCN hi vessels could stimulate bone formation. [21][22][23] Angiogenesis coupled with osteogenesis plays an important role in bone metabolism; however, the role of CD31 hi EMCN hi vessels in bone regeneration remains unknown.
Our previous study demonstrated that mutant REG1CP (regenerating family member 1 gamma) increased the formation of the CD31 hi Emcn hi endothelium in bone marrow by binding to Krüppellike factor 3 (KLF3) to inhibit its activity. We also identified a natural compound, ophiopogonin D, which functions as a KLF3 inhibitor. 23 Administration of ophiopogonin D increased the abundance of CD31 hi Emcn hi vessels and bone formation. 23 In the present study, we expanded our research and demonstrated that ophiopogonin D acts as a KLF3 inhibitor to promote vessel formation and further stimulate bone regeneration in young (3 months old) and middle-aged (12 months old) mice.

| Mice
To specific knockout Klf3 in endothelium, we crossed mice carrying loxP-flanked Klf3 alleles (Klf3 flox/flox ) with Cdh5-Cre transgenics to get were purchased from Jackson Laboratory, and loxP-flanked Klf3 mice were purchased from Cyagen Biosciences Inc (China). The bone regeneration model was established as described before. 24,25 A longitudinal incision was made on each knee to expose the femoral condyle by patella dislocation. Then, we used a dental drill to make a hole was at the intercondylar notch of the femur. A 0.6-mm-diameter Kirschner wire was placed from the proximal end of the femur. We confirmed the bone ablation of trabecular bone by radiography. Bone samples were collected 1 week after the surgery. For ophiopogonin D treatment experiment, mice under surgical ablation of trabecular bone were intraperitoneally treated with ophiopogonin D at dosage of 20 mg/kg every day for 7 days. All mice we used were C57BL/6J background.
Male mice at indicated age were used in our experiments. All mice were maintained in standard, specific pathogen-free facility of the Laboratory Animal Research Center of Central South University.

| Isolate primary BMSCs
We isolated primary BMSCs as reported previously. 26,27 We collected all bone marrow cells and incubated them with FITC-, APC-and PE-conjugated antibodies which recognized mouse

| Osteoclasts differentiation assay
Osteoclasts differentiation assay was performed as reported previously. 28 Monocytes and macrophages were collected from bone marrow of mice by flushing the marrow space of femora and tibiae.

| Tube formation assay
Tube formation assay was performed as reported previously. 29 Endothelial cell tube formation assay was conducted in 48-well plates precoated with Matrigel (BD). 1 × 10 5 cells were seeded per well after 1-hour serum starvation. After 5-, 7-, 9-and 12-hour incubation at 37°C, the tube formation of HMECs was observed and the number of tube branches was quantified by counting 4 random microscope visual fields in each well.

| Wound healing assay
Human microvascular endothelial cells or BMSCs were grown to confluency. A linear wound was made by scraping a non-opening Pasteur pipette across the confluent cell layer. 5, 7, 9 and 12 hours after wound, the migrate cells were observed and measured by counting 4 random microscope visual fields in each well.

| qRT-PCR analysis
Total RNA from cells was extracted using TRIzol reagent (Invitrogen).

| Flow cytometry
We isolated femora and tibia from mice, then crushed the metaphysis region in ice-cold PBS. Bone pieces were digested using 1 mg/mL type I A collagenase at 37°C for 20 minutes to obtain single-cell suspensions. Then, the cells were counted and incubated for 45 minutes at 4°C with endomucin antibody (Santa Cruz, SC-65495, 1:100) and APC-conjugated CD31 antibody (R&D Systems, FAB3628A, 1:100). We performed acquisition on a fluorescence-activated cell sorting (FACS) FACScan cytometer (BD Immunocytometry Systems).

| μCT analysis
We used the high-resolution μCT (Skyscan 1172, Bruker microCT, Kontich, Belgium) to perform the µCT analysis. The scanner was set

| Histochemistry
Histochemistry staining was performed as reported previously. 30,31 Femora were dissected from mice. After fixing overnight with 10% formalin at 4°C, the samples were decalcified at 4°C using 10% EDTA (pH 7.4) for 21 days and then embedded in paraffin. Fourmicrometre-thick femora were used for staining. The slides were processed for TRAP, and HE staining was performed using a standard protocol (Sigma-Aldrich).

| Immunocytochemistry
Immunocytochemistry staining was performed as reported previously. 32 Femora were dissected from mice. After fixing overnight with 10% formalin at 4°C, the samples were decalcified at 4°C using 10% EDTA (pH 7.4) for 21 days and then embedded in paraffin. Four-micrometre-thick femora were used for staining. The slides were stained with individual primary antibodies to OCN (Takara Bio Inc M137) at 4°C overnight. Horseradish peroxidasestreptavidin detection system (Dako) was used to detect immunoactivity. Then, we counterstained the sections with haematoxylin (Sigma-Aldrich).

| Immunofluorescence
Immunofluorescence staining was performed as reported previously. 33,34 Femora were dissected from mice. After fixing with ice-cold 4% paraformaldehyde solution for 4 hours, the samples were decalcified in 0.5 mol/L EDTA (pH 7.4) at 4℃ for 24 hours (1and 3-month-old mice) or for 48 hours (12-month-old mice). The bone samples were then incubated in 20% sucrose and 2% polyvinylpyrrolidone (PVP) solution overnight, as described previously. 35 For CD31 hi EMCN hi vessels staining, the tissues were embedded

| Statistics
The data are presented as the mean ± SD; For comparisons of two groups, two-tailed Student's t test was used. For comparisons of multiple groups, one-way ANOVA was used. Differences were considered significant at P < .05. No randomization or blinding was used, and no animals were excluded from analysis. Sample sizes were selected on the basis of previous experiments.

| The amount of CD31 hi EMCN hi vessels decreased during ageing but increased in bone regeneration
As animals age, the decrease in the number of osterix-positive (OSX + ) osteoprogenitors and bone mass correlates with the pronounced reduction in CD31 hi Emcn hi vessels (Figure 1A-C and G-H). The amount of CD31 hi Emcn hi vessels, which mainly appear in the metaphysis just below the growth plate in 1 month old, was significantly decreased in adults (3 months old) and was nearly absent in aged (12 months old) mice ( Figure 1A,B). Flow cytometry analysis confirmed the age-dependent reduction of CD31 hi Emcn hi endothelial cells ( Figure 1D,E). To investigate the role of CD31 hi EMCN hi vessels in bone regeneration, we constructed a bone regeneration model after surgical ablation of the trabecular bone. 24,25 We found that abundant CD31 hi EMCN hi vessels, as well as alkaline phosphatasepositive (ALP + ) osteoprogenitors and OSX + osteoprogenitors, emerged in the bone regeneration area in the 3-month-old mice at 1 week after surgery ( Figure 1F and Figure S1). The amount of CD31 hi EMCN hi vessels in the femur of the 12-month-old mice was also increased at 1 week after trabecular bone ablation ( Figure 1F).
The bone volume in the regeneration area was increased together with the number of CD31 hi EMCN hi vessels and osteoprogenitors cells ( Figure 1G-H and Figure S1). These results indicated that the CD31 hi EMCN hi vessels might play an important role in bone regeneration.

| Ophiopogonin D acts as a KLF3 inhibitor and promotes vessels formation in vitro
Our previous research demonstrated that KLF3 represses the expression of JUNB (JunB proto-oncogene, AP-1 transcription factor subunit) as well as its downstream factor VEGFA (vascular endothelial growth factor A) in endothelial cells, and further inhibited the process of angiogenesis. 23 We also identified a natural compound from Radix Ophiopogon japonicus, ophiopogonin D, which acts as a KLF3 inhibitor to abolish the transcriptional repression function of KLF3 and further increase the expression of JUNB and VEGFA in endothelial cells. 23 As expected, we observed that CD31, EMCN,

| Ophiopogonin D does not affect BMSC migration and osteoblastic differentiation in vitro
To investigate the role of ophiopogonin D in bone formation in vitro, we isolated primary bone marrow-derived stem cells

| Ophiopogonin D treatment promotes CD31 hi EMCN hi vessel and bone formation during bone regeneration
To investigate whether treatment with ophiopogonin D could promote CD31 hi EMCN hi vessel formation and further stimulate bone regeneration in vivo, 3-month-old and 12-month-old C57BL/6J mice, after surgical ablation of the trabecular bone, were treated intraperitoneally with ophiopogonin D at 20 mg/kg every day for 1 week.
This treatment did not affect the body weight of either group of mice ( Figure 5A and Figure

| D ISCUSS I ON
Blood vessels play an important role in tissue regeneration by recruiting progenitor cells. 36 and VEGFA, and further repressed angiogenesis. 23 We also demonstrated that specific knockout of Klf3 in endothelial cells increased the formation of CD31 hi EMCN hi vessels and increased bone formation. 23 Here, we extended our study and confirmed that specific knockout of Klf3 in endothelial cells also could stimulate CD31 hi EMCN hi vessel formation after bone injury and could accelerate bone regeneration.
Ophiopogonin D is a natural compound isolated from the traditional Chinese herbal agent Radix Ophiopogon japonicus, 49 which showed anti-osteoporotic effects in a murine ovariectomized (OVX) model. 50 Previously, we identified that ophiopogonin D could bind to KLF3 and suppress its function, thus further promoting angiogenesis. 23 In the present study, we further demonstrated that ophiopogonin D also could increase CD31 hi EMCN hi vessel formation and bone regeneration after bone injury.
In summary, we revealed the important role of CD31 hi EMCN hi vessels in bone regeneration and demonstrated the natural compound ophiopogonin D could increase the number of CD31 hi EMCN hi vessels and promoted bone regeneration after bone injury. Our findings provide a potentially novel strategy to accelerate bone healing.

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

CO N FLI C T O F I NTE R E S T
The authors have declared that no conflict of interest exists. lysed results, co-wrote the manuscript, and is the guarantor of this work and has full access to all the data in this study and takes the responsibility for data accuracy.

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 within the article and Supplementary Files or available from the authors upon request.