The authors have no conflict of interest.
Reduced Expression of Platelet Endothelial Cell Adhesion Molecule-1 in Bone Marrow Cells in Mice After Skeletal Unloading†
Article first published online: 4 JAN 2005
Copyright © 2005 ASBMR
Journal of Bone and Mineral Research
Volume 20, Issue 6, pages 1002–1010, June 2005
How to Cite
Sakuma-Zenke, M., Sakai, A., Nakayamada, S., Kunugita, N., Tabata, T., Uchida, S., Tanaka, S., Mori, T., Nakai, K., Tanaka, Y. and Nakamura, T. (2005), Reduced Expression of Platelet Endothelial Cell Adhesion Molecule-1 in Bone Marrow Cells in Mice After Skeletal Unloading. J Bone Miner Res, 20: 1002–1010. doi: 10.1359/JBMR.050102
- Issue published online: 4 DEC 2009
- Article first published online: 4 JAN 2005
- Manuscript Accepted: 29 DEC 2004
- Manuscript Revised: 26 AUG 2004
- Manuscript Received: 6 JUN 2004
- endothelial cell;
- alkaline phosphatase activity;
- colony-forming unit-fibroblastic
One week of tail suspension significantly decreased the expression of PECAM-1 in mouse tibial bone marrow cells but not those of a number of other vascular factors. Anti-PECAM-1 antibody suppressed both ALP+ CFU-f formation and ALP production under co-culture of the osteoblastic cell line and the PECAM-1+ endothelial cell line. This study suggests that the reduced ALP activity after skeletal unloading is related to downregulation of PECAM-1 expression in bone marrow cells in mice.
Introduction: Vascular factors play a role in bone development and regeneration. We tested the hypothesis that skeletal unloading reduces osteogenic potential by inhibiting the molecules related to angiogenesis and/or vasculogenesis in bone marrow cells.
Materials and Methods: Eight-week-old male mice were assigned to three groups after acclimatization for 1 week: ground control (GC), tail suspension (TS), and reloading after 7-day TS (RL). Bilateral tibial and humeral samples were used for analyses. MC3T3-E1, a mouse osteoblastic cell line, and EOMA and ISOS-1, mouse endothelial cell lines, were also used.
Results: Flow cytometric analysis revealed that 7-day TS significantly decreased the expression of platelet endothelial cell adhesion molecule-1 (PECAM-1, CD31) in tibial bone marrow cells, but not those of angiopoietin-1, angiopoietin-2, Flk-1 (vascular endothelial growth factor receptor-2), and vascular endothelial cadherin. The expression of PECAM-1 in tibial marrow cells was reduced at day 3 of TS to 80% and still showed significantly low levels at day 7 of TS to 72% of that at the respective days of GC. This decreased expression of PECAM-1 after 7-day TS showed the GC level at 5-day reloading after 7-day TS. However, the expression of PECAM-1 in humeral marrow cells (internal bone marrow control) after TS and RL remained unchanged and equivalent to that of GC. The expression level of PECAM-1 mRNA was significantly lower at day 7 of TS to 62% of that in GC. Double labeling analyses revealed that PECAM-1+ cells mostly consisted of endothelial cells and partially of granulocytes. In bone marrow cell cultures, the formation of alkaline phosphatase (ALP)+ colony forming units-fibroblastic was significantly reduced in the presence of anti-PECAM-1 antibody in the medium compared with the presence of immunoglobulin G (0.025 times as much as ALP production with immunoglobulin G). ALP production by cultured MC3T3-E1 was enhanced in combination with PECAM-1+ EOMA (1.8 times as much as ALP production by MC3T3-E1 alone), but not in combination with PECAM-1− ISOS-1. Anti-PECAM-1 antibody inhibited the increase in ALP production under co-culture with EOMA.
Conclusions: Our data show that the reduced ALP activity after skeletal unloading is closely correlated with reduced expression of PECAM-1 in bone marrow cells. We speculate that the loss of osteogenic potential after skeletal unloading is caused by the suppression of PECAM-1 signaling on endothelial cellular surface.
Skeletal unloading induces bone loss in loaded bones in humans(1) and animals.(2–7) Tail-suspended mice have been used as a model to simulate conditions of no mechanical stimulation on loaded bones, such as space flight and long-term bed rest.(5–7) Histomorphometric analyses have shown that unloading for 1 week reduced the bone formation rate, but this change was recovered to the normal loading level after reloading.(3,7) These data were closely associated with changes in alkaline phosphatase (ALP)+ colony forming units-fibroblastic (CFU-f) formation in bone marrow cell culture.(7) Unloading by tail-suspension inhibited proliferation(8–10) and differentiation(10,11) of osteoprogenitor cells in vitro. The basic mechanism underlying inhibition of proliferation and differentiation of osteoprogenitor cells after skeletal unloading has not been elucidated.
We have shown that vascular endothelial growth factor (VEGF) is expressed along with its receptors, Flt-1 (VEGF receptor-1) and Flk-1 (VEGF receptor-2), during the healing process of bone and bone marrow after drill-hole injury.(12) A recent gene-targeting study showed that not only angiogenesis but also bone formation and skeletal development are impaired in mice lacking VEGF164 and VEGF188.(13) Vascular factors play a role in bone development and regeneration, and there seems to be developmental reciprocity between vascular factors and osteogenic potential. We considered therefore that the molecules related to angiogenesis and/or vasculogenesis may in part mediate the decrease in trabecular bone formation after skeletal unloading. However, the vascular signals for bone response to mechanical stimuli are not defined.
Bone adaptation to lack of mechanical loading requires bone cells to integrate mechanical signals into appropriate changes in bone architecture. Fujiwara and colleagues(14–16) reported that the platelet endothelial cell adhesion molecule-1 (PECAM-1, CD31), a cell adhesion molecule localized in the interendothelial cell adhesion site, is tyrosine-phosphorylated when endothelial cells are exposed to physiological levels of fluid shear stress. These findings suggest the possible role of PECAM-1 in mechanosensing by endothelial cells. We postulated that bone adaptation to lack of mechanical loading through PECAM-1 signaling has important implications for local regulation of trabecular bone turnover.
Thus, we hypothesized that skeletal unloading inhibits local factors related to angiogenesis and/or vasculogenesis in bone marrow cells, thereby reducing osteogenic potential in hind limbs of tail-suspended mice. In this study, we showed that reduced ALP activity after skeletal unloading is closely related to reduced expression of PECAM-1 in bone marrow cells in mice.
MATERIALS AND METHODS
C57BL/6NCrj male 8-week-old mice were purchased from Charles River Japan (Tokyo, Japan). The mice were acclimatized for 1 week during which they were fed standard rodent chow (CE-2; Clea Japan, Tokyo, Japan) containing 1.25% calcium, 1.06% phosphorus, and 2.0 IU/g vitamin D3. Mice were allowed free access to food and water and were housed individually in cages in a temperature-regulated environment (temperature, 24 ± 1°C; humidity, 55 ± 5%) that was illuminated from 7:00 a.m. to 7:00 p.m. The body weights of the mice at 8 weeks of age ranged from 18 to 22 g. The mice were divided randomly into body weight-matched groups. Tail suspension (TS) of mice was performed as described previously by our laboratory.(7) TS mice were allowed free access to food (CE-2) and water. Reloading (RL) of mice began by stopping TS after a 7-day unloading. Ground control (GC) mice with normal loading were fed the average amounts of food taken by TS mice and RL mice. All experimental protocols were approved in advance by the Ethics Review Committee for Animal Experimentation of the University of Occupational and Environmental Health.
The mice were killed by cervical dislocation. To determine the effect of unloading on changes in molecules related to angiogenesis and/or vasculogenesis in unloaded limbs, we performed flow cytometric analyses on tibial bone marrow cells 1 week after TS and GC. The numbers of mice were 15 for PECAM-1 and 5 each for angiopoietin-1, angiopoietin-2, Flk-1, and vascular endothelial cadherin (VE-cadherin, CD144) in the respective groups of TS and GC. The PECAM-1 expression was also monitored in humeral bone marrow cells (n = 15 for TS and GC each) obtained from loaded forelimbs as each internal bone marrow control. After confirming that PECAM-1 expression significantly changed at day 7, we analyzed its expression in tibial and humeral bone marrow cells at day 3 of TS (n = 4), day 3 of RL (n = 9), day 5 of RL (n = 4), day 3 of GC (n = 3), day 10 of GC (n = 5), and day 12 of GC (n = 4). We studied mRNA expression of PECAM-1 in bone marrow cells at day 7 of TS and GC (n = 4 for TS and GC each). To clarify the origin of the cells expressing PECAM-1, we performed flow cytometric analyses on tibial bone marrow cells 1 week after TS and GC (n = 5 for TS and GC each) using double stain of PECAM-1 and other cell surface proteins, which were VE-cadherin (n = 5), c-fms (CD115, macrophage colony-stimulating factor receptor; n = 5), Flk-1 (n = 5), CD33 (n = 5), and CD11b (n = 5). We performed primary bone marrow cell cultures using anti-PECAM-1 antibody to evaluate the shutdown effect on tibial bone marrow capacity for bone cell development (n = 5). We also measured the number of ALP+ CFU-f. To determine the effect of PECAM-1 on osteoblasts, we co-cultured osteoblastic cell line with or without PECAM-1-expressing or -nonexpressing endothelial cell lines (n = 3). We also reconfirmed PECAM-1 action on ALP production in an osteoblastic cell line, using anti-PECAM-1 antibody (n = 4).
Antibodies and other reagents
The following antibodies were used as purified immunoglobulin (Ig) in preparation of staining and analysis of cell surface or cytoplasmic molecules: anti-mouse PECAM-1 monoclonal antibody (mAb), anti-mouse CD11b mAb (Beckman Coulter, Tokyo, Japan), angiopoietin-1 polyclonal antibody (pAb), angiopoietin-2 pAb, CD33 pAb, normal mouse IgG, fluorescein-isothiocyanate (FITC)- or phycoerythrin (PE)-labeled donkey anti-goat IgG pAb (all were from Santa Cruz Biotechnology, Santa Cruz, CA, USA), anti-mouse Flk-1 mAb, anti-mouse VE-cadherin mAb, FITC-labeled polyclonal goat anti-rat Ig pAb, anti-mouse CD32/16 (Fc III/II receptor) mAb (PharMingen, San Diego, CA, USA), anti-mouse c-fms mAb, PE-labeled F(ab)2 goat anti-rat IgG mAb (Cosmo Bio, Tokyo, Japan), anti-mouse ALP mAb (R&D Systems, Minneapolis, MN, USA), and control mAb thy1.2 (Becton Dickinson, San Jose, CA, USA).
Evaluation of molecules related to angiogenesis and/or vasculogenesis in bone marrow cells by flow cytometry
Bone marrow cells were obtained from tibias and humeri in each animal and used for staining and flow cytometric analyses using the methods described previously.(17) Briefly, the bone marrow was flushed out from the proximal cut end with 2 ml of α-MEM (Nakarai, Kyoto, Japan). The bone marrow cells at a density of 5 × 105 cells/ml were incubated with anti-mouse CD32/16 mAb in PBS for 30 minutes at 4°C to block the Fc receptor. The cells were incubated with PE-labeled anti-PECAM-1 mAb, PE-labeled anti-Flk-1 mAb, FITC-labeled anti-VE-cadherin mAb, anti-angiopoietin-1 pAb, or anti-angiopoietin-2 pAb in PBS for 30 minutes at 4°C and then washed with PBS. For angiopoietin-1 and -2, the cells were further incubated with FITC-labeled donkey anti-goat IgG antibody for 30 minutes at 4°C and washed with PBS. Dead cells were excluded by staining with propidium iodide (Sigma Chemical, St Louis, MO, USA). Stained cells were analyzed on a flow cytometer (EPICS XL; Coulter Co., Tokyo, Japan). Quantification of cell surface antigen on a single cell was calculated using standard beads (QIFKIT; DAKO Japan, Kyoto, Japan). PE- or FITC-labeled donkey anti-goat IgG was used as a negative control.
Characterization of PECAM-1+ cells
Flashed bone marrow cells obtained from tibias were double stained with PE-labeled and FITC-labeled antibodies of anti-PECAM-1 mAb in combination with either of anti-mouse VE-cadherin mAb, anti-mouse c-fms mAb, anti-mouse Flk-1 mAb, CD33 pAb, or anti-mouse CD11b mAb (the former three were used as endothelial markers and the latter two were used as granulocyte markers). The cells were prepared by the same methods as above. The cells were incubated with the combination of FITC-labeled anti-PECAM-1 mAb and either PE-labeled CD11b mAb, PE-labeled Flk-1 mAb, no-labeled c-fms mAb, or no-labeled CD33 pAb or with the combination of PE-labeled anti-PECAM-1 mAb and no-labeled VE-cadherin mAb in PBS for 30 minutes at 4°C and then washed with PBS. For the cases of the no-labeled first antibodies, cells were further incubated with FITC-labeled polyclonal goat anti-rat Ig pAb for VE-cadherin or PE-labeled secondary antibody (F[ab]2 goat anti-rat IgG mAb for c-fms, donkey anti-goat IgG pAb for CD33) for 30 minutes at 4°C and then washed with PBS. Stained cells were analyzed on a flow cytometer (EPICS XL; Coulter Co.). Quantification of cell surface antigen on a single cell was calculated using standard beads (QIFKIT; DAKO Japan). PE- or FITC-labeled donkey anti-goat IgG was used as a negative control.
Quantitative real-time RT-PCRs
Bone marrow cells were obtained from tibias of 7-day TS and GC mice and used for isolating total RNA and quantitative RT-PCR analyses using the methods described previously.(18) Briefly, the bone marrow was flushed out from the proximal cut end with 2 ml of PBS. The bone marrow cells were frozen in liquid nitrogen. Total RNA was extracted using an acid guanidinium thiocyanate-phenol-chloroform method after homogenizing.(19) The isolated RNA was cleaned up using the RNeasy kit (Qiagen, Hilden, Germany). First-strand cDNA was reverse-transcribed from total RNA (1 μg) using Moloney murine leukemia reverse transcriptase (SuperScript; Life Technologies, Rockville, MD, USA) and oligo(dT) 12-18 primer (Life Technologies). Quantitative RT-PCR analysis was performed using an Optical System Interface software (version 3.0; Bio-Rad). The quantitative RT-PCR for PECAM-1 was performed in 20 μl with ∼7.5 ng cDNA, 0.5 pM primers, and 10 μl iQ SYBR Green Supermix (Bio-Rad) containing 0.4 mM of each dNTP (dATP, dCTP, dGTP, and dTTP), iTaq DNA polymerase, and SYBR Green 1 dye. The sequences of primers were PECAM-1 (245 bp) sense, 5′-TGCAGGAGTCCTTCTCCACT-3′; antisense, 5′-ACGGTTTGATTCCACTTTGC-3′; β-actin (639 bp) sense, 5′-TTGAGACCTTCAACACCCCAG-3′; antisense, 5′-ACTTGCGCTCAGGAGGAGCAA-3′. These primers were designed by Primers 3 software (Whitehead Institute/MIT Center of Genome Research, Cambridge, MA, USA). β-actin was used as an internal control. The amplification conditions were an initial 3 minutes at 95°C, and then 40 cycles of denaturation at 95°C for 30 s, annealing at 60°C for 30 s, and extension at 72°C for 30 s. β-actin was annealed at 65°C for 30 s. PCR reactions were performed in four independent experiments. The expression levels of mRNA were normalized using β-actin as a housekeeping gene and expressed as a relative value to the baseline control over the time-course.
CFU-f assay from bone marrow cells
Marrow cultures were initiated using the method of Maniatopoulos et al.(20) Tibial and femoral bone marrow was flushed out from the proximal cut end, a single-cell suspension was prepared by repeated aspiration, and all bone marrow cells were counted in a cell counter (F-820; Sysmex Co., Kobe, Japan). The obtained cells were disseminated into a 6-well plate (Corning, New York, NY, USA) at a concentration of 5 × 105 cells/ml in 2 ml of α-MEM supplemented with 15% FCS (Gibco, New York, NY, USA), 2.0 g/liter NaHCO3, 100 μg/ml streptomycin and 100 U/ml penicillin, 1.25 U/ml nystatin (Sigma), and 10 mM of sodium β-glycerophosphate (Sigma). Cells were cultured at 37°C in a humidified atmosphere of 5% CO2 in air for 10 days. The medium was changed twice a week. Anti-mouse PECAM-1 mAb (2 μg/ml) and normal mouse IgG (2 μg/ml) was added at the beginning of the cultures and at each time after medium change. After culturing for 10 days, CFU-f colonies were fixed with 10% neutral buffered formalin and stained for ALP by an enzyme histochemical procedure using BCIP-NBT kit (Nakarai). The dishes were scanned with a Hewlett Packard ScanJet, the image was stored on computer, and stained areas were manually circumscribed and quantitated using NIH Image 1.6. Colonies comprising >50 cells at 5-fold magnification were defined as CFU-f.
Co-culture study of osteoblastic and endothelial cell lines
MC3T3E-1, an osteoblastic cell line, was provided by the Riken Cell Bank (Tsukuba, Japan). EOMA, an endothelial cell line from murine hemangioendothelioma, was purchased from American Type Culture Collection (Manassas, VA, USA). ISOS-1, an endothelial cell line from murine angiosarcoma, was provided by Cell Resource Center for Biomedical Research, Tohoku University (Sendai, Japan). MC3T3-E1, EOMA, and ISOS-1 cells were cultured in DMEM (Gibco) containing 10% FCS (Gibco) and streptomycin/penicillin (10 units/ml; Sigma Aldrich) in 25-cm2 culture flasks (Falcon, Lincoln Park, NJ, USA) in a humidified 5% CO2 atmosphere. Cells were passed by trypsinization using 0.25% Trypsin/1 mM EDTA-4Na; (Invitrogen Life Technologies, Carlsbad, CA, USA), followed by dilution (5- to 10-fold) in DMEM containing 10% FCS.
Co-culture of osteoblastic and endothelial cells
After cultured cells reached subconfluence, the medium was changed to DMEM containing 1% FCS at the day before assay. MC3T3-E1 cells (1 × 105 cells/ml) were co-cultured with EOMA cells (1 × 105 cells/ml) or ISOS-1 cells (1 × 105 cells/ml) in DMEM without FCS at 37°C for 24 h in 6-well plates. Anti-PECAM-1 mAb was added at concentrations of 0, 0.1, 1.0, or 10 μg/ml into the medium of MC3T3-E1 and EOMA co-cultures. After incubation, the cultured cells were washed twice with PBS and treated with 0.25% trypsin for 1 minute at 37°C. DMEM containing 10% FCS was added to stop trypsinization. The cells were harvested from the wells, washed with PBS, and settled in media suitable for the following experiments.
ALP production of cultured cells
Flow cytometric analysis of MC3T3-E1 cells was performed using a FACScan (Becton Dickinson, Mountain View, CA, USA), following the procedure described previously.(21) Briefly, 2 × 105 cells were incubated with negative control mAb thy-1.2 or anti-ALP mAb in FACS media consisting of Hank's balanced salt solution (Nissui), 0.5% human serum albumin (HSA; Mitsubishi Pharma, Osaka, Japan) and 0.2% NaN3 (Sigma Aldrich) for 30 minutes at 4°C. Cytoplasmic antigens of MC3T3-E1 cells pretreated with cell permeabilization kit (Caltag, Burlingame, CA, USA) were stained by anti-ALP mAb in FACS media for 30 minutes at 4°C. After washing the cells three times with FACS media, they were further incubated with FITC-labeled goat anti-mouse IgG Ab for 30 minutes at 4°C. The mAbs-stained cells were detected using FACScan. ALP production on a single cell was quantified using standard beads (QIFKIT; DAKO Japan) as described previously.(22,23) The data were used for the construction of the calibration curve (mean fluorescence intensity [MFI]) against antibody-binding capacity (ABC). The cell specimen was analyzed on the FACScan, and ABC was calculated by interpolation on the calibration curve. When the green fluorescence laser detector was set at the 450 level in the used FACScan, ABC = 414.45 × exp(0.0092 × MFI) (R2 = 0.9999). Subsequently, specific ABC (SABC) was obtained after corrections for background, apparent ABC of the negative control mAb thy − 1.2. SABC corresponds to the mean number of accessible antigenic sites per cell, referred to as antigen density and expressed as molecules per cell.
Results are expressed as the mean ± SE. Differences between groups were examined for statistical significance using the Mann-Whitney U-test. A p value <0.05 denoted the presence of a statistically significant difference.
Body weight gain did not differ between GC mice and TS mice (data not shown). All mice remained healthy throughout the experimental period.
PECAM-1 expression after unloading
The cell surface antigens on the flushed tibial bone marrow cells after TS were identified in mice. The 7-day unloading significantly reduced the percentage of PECAM-1 expression cells in bone marrow cells compared with that of GC (Fig. 1). There were no significant differences in angiopoietin-1, angiopoietin-2, Flk-1, and VE-cadherin between TS and GC. PECAM-1 expression decreased in tibial bone marrow cells at days 3 and 7 of unloading compared with that at the respective days of GC (Fig. 2A). The expression level of PECAM-1 was lower after 3-day RL and similar after 5-day RL compared with the respective values at days 10 and 12 of GC. On the other hand, PECAM-1 expression was maintained in humeral bone marrow cells obtained from loaded forelimbs as internal bone marrow control after TS and RL (Fig. 2B). The expression level of PECAM-1 mRNA in TS was significantly lower at 7 days compared with that in GC (Fig. 3).
Characterization of PECAM-1+ cells
All the PECAM-1+ cells expressed VE-cadherin and c-fms (Fig. 4A). Almost 100% of Flk-1+ cells expressed PECAM-1 (Fig. 4B). Unloading significantly reduced PECAM-1+ cells in VE-cadherin+ cells. The cells stained by CD33 and CD11b partially expressed PECAM-1.
The formation of ALP+ CFU-f was significantly reduced under the medium condition of supplementation of anti-PECAM-1 antibody compared with that of IgG (Fig. 5).
Co-culture study of PECAM-1+ or PECAM-1− endothelial cell lines and osteoblastic cell line
ALP production by the cultured mouse osteoblastic cell line MC3T3-E1 was enhanced in the presence of the mouse endothelial cell line EOMA, positive for PECAM-1 expression, but not in the presence of the mouse endothelial cell line ISOS-1, negative for PECAM-1 expression (Figs. 6A and 6B). Anti-PECAM-1 antibody dose-dependently inhibited the increase in ALP production of MC3T3-E1 under co-culture with EOMA (Fig. 6C).
Our flow cytometric analyses clearly showed that the unloading state suppressed PECAM-1 expression in the tibial bone marrow cells, whereas unloading did not alter the expression levels of angiopoietin-1, angiopoietin-2, Flk-1, and VE-cadherin. The decreased expression of PECAM-1 after 7-day unloading showed the normal loaded control level after a 5-day reloading. ALP activity was reduced in close association with decreased signaling of PECAM-1, judging from the data that anti-mouse PECAM-1 antibody suppressed both ALP+ CFU-f formation and ALP production under the co-culture of the osteoblastic cell line and the PECAM-1+ endothelial cell line. The loss of ALP activity after skeletal unloading is closely associated with the suppression of PECAM-1 signaling on the bone marrow cellular surface. Our double-labeling analyses showed that PECAM-1+ cells mostly consisted of endothelial cells and partially of granulocytes. All of the PECAM-1+ cells expressed VE-cadherin and c-fms, markers of endothelial cells.(24,25) We consider that PECAM-1 is located on endothelial cell surfaces in bone where PECAM-1 can be affected by unloading and mediate the mechanical sensing signals in bone. A nearly total disappearance of ALP+ CFU-f was presented in Fig. 5. This was a remarkable effect compared with all other experiments. Under cell culture conditions, bone marrow cells obtained from the tibias had a lower ability of proliferation compared with the cells of the established cell lines. We considered that this is the reason for the markedly suppressive effect in the ALP+ CFU-f study compared with the effect in the co-culture study of osteoblastic and endothelial cell lines, although we used similar concentrations of anti-PECAM-1 antibody (2 μg/ml for the ALP+ CFU-f study and 0.1-10 μg/ml for the cell line experiment).
We previously reported that trabecular bone loss caused by skeletal unloading was related to facilitation of intracellular p53-p21 signaling in bone marrow cells.(26) Unloading significantly increased the percentage of hypoploid bone marrow cells, which are considered to reflect predominantly apoptotic cells, relative to that in loaded groups in p53+/+ mice. However, there was no significant difference in ploidy between the unloaded and loaded groups in p53−/− mice. In this regard, Sho et al.(27) reported that apoptosis of endothelial cells was induced by a reduction in blood flow. Osawa et al.(14) reported rapid tyrosine phosphorylation of PECAM-1 in endothelial cells exposed to fluid shear stress. Osmotic changes also induced similar PECAM-1 and extracellular signal-regulated kinase phosphorylation with nearly identical kinetics.(15,16) Therefore, they stated that both fluid shear stress and osmotic changes might activate the same mechanosignaling cascade by perturbing the cell membrane mechanically. However, the molecular mechanism for the biological response to unloading remains largely unknown. We speculate that unloading causes lack of mechanical stimulation in bone tissue, and PECAM-1 serves as a mechanosensor on responded endothelial cells.
PECAM-1, as well as angiopoietin-1, angiopoietin-2, Flk-1, and VE-cadherin, is expressed on the endothelial cell surface. Our results showed that unloading reduced only the expression of PECAM-1, but not the others, in bone marrow cells. Furthermore, PECAM-1 expression was decreased in unloaded tibial marrow cells but not loaded humeral marrow cells. Thus, we consider that unloading does not decrease the number of endothelial cells in bone marrow cells, but results in downregulation of PECAM-1 in the endothelial cell. Therefore, PECAM-1 signaling in TS mice reflects the change of skeletal loading. Duncan et al.(28) reported that PECAM-1-disrupted mice are viable and undergo normal vascular development, suggesting that PECAM-1 is not critical for vasculogenesis. In this experiment, PECAM-1, one of cell adhesion molecules, could play a role in mediators of mechanical sensing signals in bone but not in vascularization. We consider that an increase in the vascularization of the bone during unloading could not directly lead to prevention of bone loss after skeletal unloading.
Our previous studies showed that trabecular bone formation rate in the proximal tibia was significantly decreased to ∼43% of the normal loaded control level by 7-day skeletal unloading, showed a steady-state level from 7 to 14 days after unloading, and recovered to the normal loaded control level after a 14-day reloading.(7,26) The same study used bone marrow cell culture and revealed that 7-day unloading decreased ALP+ CFU-f formation to 58% and mineralized nodule formation to 21% of the normal loaded control levels, which recovered to the normal loaded control level after reloading.(7,26) The reduction of osteogenic potential in bone marrow cells suggests a reduction of bone formation rate and the loss of trabecular bone volume under the unloaded condition. We consider that the shutdown of PECAM-1 signal in bone marrow cells leads to the decrease in bone formation rate and loss of trabecular bone volume, because the medium supplemented with anti-PECAM-1 antibody markedly suppressed ALP+ CFU-f formation in this study.
ALP is produced under the isolated culture of MC3T3-E1. The ALP production by MC3T3-E1 was enhanced by the PECAM-1+ cell line EOMA, but not by the PECAM-1+ cell line ISOS-1. Anti-PECAM-1 antibody suppressed the increase of ALP production under co-culture of MC3T3-E1 and EOMA. These results are compatible with those of the sequential changes of PECAM-1 expression in this study and ALP+ CFU-f formation in our previous study(7) after unloading and reloading. In fact, osteoblastic cells usually coexist with endothelial cells in the environment of bone marrow. It is reported that endothelial mesenchymal cells interact through gap junction communication for the development of bone-forming cells,(29) whereas the expression level of PECAM-1 on endothelial cells may adjust osteogenic potential of bone marrow cells to mechanical stimulation. Arihiro and Inai(30) reported that PECAM-1 expression in osteosarcoma with no bone metastasis and osteosarcoma with bone metastasis was noted in 10% and 75% of cases, respectively. Formation of metastatic foci of osteosarcoma cells in other bones was significantly correlated with expression of PECAM-1. Thus, we consider that the expression of PECAM-1 on endothelial cells directly or indirectly activates osteogenic potential through endothelial cell migration,(31) adhesion,(32) and differentiation.(30)
PECAM-1 acts as an adhesion molecule itself as well as a regulatory molecule of integrin.(21,33) We consider that PECAM-1 on endothelial cells signals osteoblastic cells through heterophilic binding with some ligands, such as αvβ3 integrin(31,34,35) and CD38.(36,37) PECAM-1 is persistently expressed at high levels on endothelial cells, independent of the organ or tissue analyzed or of the type of vessel. Human CD38 is a cell surface molecule involved in the regulation of lymphocyte adhesion to endothelial cells.(36) CD38 expression is found in bone tissue as well.(37) PECAM-1 and CD38 cognate interactions modulate heterotypic adhesion and implement cytoplasmic calcium fluxes. Other effects include the synthesis of messages for cytokines, markedly increased on receptor-ligand interactions.(21,33)
In conclusion, we have shown in this study that reduced ALP activity after skeletal unloading closely relates to downregulation of PECAM-1 in bone marrow cells in mice. This is the first description that PECAM-1 plays a crucial role in bone adjustment for skeletal unloading. We need further experiments of in vivo administration of the PECAM-1 monoclonal antibody and/or the TS model with PECAM-1 transgenic mice to implicate PECAM-1 expression as a causal link between skeletal unloading and bone formation.
The authors thank Dr Tadashi Murakami, Dr Makoto Watanuki, Erika Kobayashi, Keiko Shigemoto-Kouno, and Mayumi Shirahashi from the University of Occupational and Environmental Health for constructive advice and technical assistance. This work was supported in part by Grants-in-Aid from the Japan Ministry of Education, Science, Sports and Culture (Scientific Research on Priority Areas B 12137210 to TN, Scientific Research B 16390446 to TN, and Scientific Research B 11470315 to AS).
- 11978 Inhibition of bone formation during space flight. Science 201: 1138–1141.,
- 21984 Effects of simulated weightlessness on bone mineral metabolism. Endocrinology 114: 2264–2270., ,
- 31991 Bone histomorphometric comparison of rat tibial metaphysis after 7-day tail suspension vs. 7-day spaceflight. Aviat Space Environ Med 62: 26–31., , ,
- 41982 Skeletal abnormalities in rats induced by simulated weightlessness. Metab Bone Dis Relat Res 4: 69–75.,
- 51992 The physical and mechanical effects of suspension-induced osteopenia on mouse long bones. J Biomech 25: 489–499., , ,
- 61994 Contribution of dietary and loading changes to the effects of suspension on mouse femora. J Exp Zool 269: 277–285., , , , ,
- 71999 Trabecular bone turnover and bone marrow cell development in tail-suspended mice. J Bone Miner Res 14: 1596–1604., , , , , , ,
- 81993 Skeletal unloading in rat decreases proliferation of rat bone and marrow-derived osteoblastic cells. Am J Physiol 264: E790–E799., , , , , ,
- 91994 Insulin-like growth factor-I increases trabecular bone formation and osteoblastic cell proliferation in unloaded rats. Endocrinology 134: 1031–1038., , , ,
- 101997 Skeletal unloading inhibits the in vitro proliferation and differentiation of rat osteoprogenitor cells. Am J Physiol 273: E1133–E1139., , ,
- 111995 Rat tail suspension reduces messenger RNA level for growth factors and osteopontin and decreases the osteoblastic differentiation of bone marrow stromal cells. J Bone Miner Res 10: 415–423., , , , , ,
- 122003 Vascular endothelial growth factor is expressed along with its receptors during the healing process of bone and bone marrow after drill-hole injury in rats. Bone 32: 491–501., , , , , , ,
- 132002 Impaired angiogenesis and endochondral bone formation in mice lacking the vascular endothelial growth factor isoforms VEGF164 and VEGF188. Mech Dev 111: 61–73., , , , , , ,
- 141997 Tyrosine phosphorylation of platelet endothelial cell adhesion molecule-1 (PECAM-1, CD31) in mechanically stimulated vascular endothelial cells. Eur J Cell Biol 72: 229–237., , , ,
- 152001 Is PECAM-1 a mechanoresponsive molecule. Cell Struct Funct 26: 11–17., , , ,
- 162002 Evidence for a role of platelet endothelial cell adhesion molecule-1 in endothelial cell mechanosignal transduction: Is it a mechanoresponsive molecule? J Cell Biol 158: 773–785., , ,
- 171998 Bone marrow cell development and trabecular bone dynamics after ovariectomy in ddy mice. Bone 23: 443–451., , , , , , ,
- 182004 Prostaglandin E2 receptor (EP4) selective agonist (ONO-4819.CD) accelerates bone repair of femoral cortex after drill-hole injury associated with local upregulation of bone turnover in mature rats. Bone 34: 940–948., , , , , , ,
- 191987 Single-step method of RNA isolation by acid guanidinium thiocyanate-phenol-chloroform extraction. Anal Biochem 162: 156–159.,
- 201988 Bone formation in vitro by stromal cells obtained from bone marrow of young rats. Cell Tissue Res 254: 317–330., ,
- 211992 CD31 expressed on distinctive T cell subsets is a preferential amplifier of β1 integrin-mediated adhesion. J Exp Med 176: 245–253., , , , , , , , ,
- 221999 H-Ras signals to cytoskeletal machinery in induction of integrin-mediated adhesion of T cells. J Immunol 163: 6209–6216., , , , , , , , , , , ,
- 231998 Constitutive chemokine production results in activation of leukocyte function-associated antigen-1 on adult T-cell leukemia cells. Blood 91: 3909–3919., , , , , , , , , ,
- 242003 MCSF expression is induced in healing myocardial infarcts and may regulate monocyte and endothelial cell phenotype. Am J Physiol 285: H483–H492., , , , , , ,
- 252002 Macrophage colony stimulating factor modulates the development of hematopoiesis by stimulating the differentiation of endothelial cells in the AGM region. Blood 99: 2360–2368., , , ,
- 262002 Disruption of the p53 gene results in preserved trabecular bone mass and bone formation after mechanical unloading. J Bone Miner Res 17: 119–127., , , , , ,
- 272001 Blood flow decrease induces apoptosis of endothelial cells in previously dilated arteries resulting from chronic high blood flow. Arterioscler Thromb Vasc Biol 21: 1139–1145., , , , ,
- 281999 Genetic evidence for functional redundancy of platelet/endothelial cell adhesion molecule-1 (PECAM-1): CD31-deficient mice reveal PECAM-1-dependent and PECAM-1-independent functions. J Immunol 162: 3022–3030., , , , , , , , , , , , ,
- 292002 Effect of HUVEC on human osteoprogenitor cell differentiation needs heterotypic gap junction communication. Am J Physiol Cell Physiol 282: C775–C785., , , , , ,
- 302001 Expression of CD31, Met/hepatocyte growth factor receptor and bone morphogenetic protein in bone metastasis of osteosarcoma. Pathol Int 51: 100–106.,
- 311999 Neutrophils sense flow-generated stress and direct their migration through αvβ3-integrin. Am J Physiol 276: H858–H864., , ,
- 321997 Role of cadherins and plakoglobin in interendothelial adhesion under resting conditions and shear stress. Am J Physiol 273: H2396–H2405., ,
- 332002 PECAM-1 (CD31) homophilic interaction up-regulates α6β1 on transmigrated neutrophils in vivo and plays a functional role in the ability of α6 integrins to mediate leukocyte migration through the perivascular basement membrane. J Exp Med 196: 1201–1211., , , ,
- 341995 CD31/PECAM-1 is a ligand for αvβ3 integrin involved in adhesion of leukocytes to endothelium. J Cell Biol 130: 451–460., , , , , ,
- 351996 Identification of αvβ3 as a heterotypic ligand for CD31/PECAM-1. J Cell Sci 109: 437–445., , , , , ,
- 361998 Human CD38 (ADP-ribosyl cyclase) is a counter-receptor of CD31, an Ig superfamily member. J Immunol 160: 395–402., , , , , , , ,
- 371998 Analysis of the distribution of human CD38 and of its ligand CD31 in normal tissues. J Biol Regul Homeost Agents 12: 81–91., , , , , , , ,