• rats;
  • bone and vascular histomorphometry;
  • neutralizing vascular endothelial growth factor antibody;
  • Flt1/R1;
  • KDR/R2


  1. Top of page
  2. Abstract
  7. Acknowledgements

Physiological angiogenesis during bone remodeling is undefined. Treadmill-running rats displayed bone marrow angiogenesis concomitant with bone formation increase and resorption decrease and upregulation of VEGF and its R1 receptor mRNA in proximal tibia. VEGF blockade over 5 weeks of training fully prevented the exercise-induced bone mass gain.

Introduction: We investigated the role of vascular endothelial growth factor (VEGF) and angiogenesis in the osteogenic response to exercise.

Materials and Methods: Nine-week-old male Wistar rats were treadmill-trained at 60% VO2max for various periods. Bone and vascular histomorphometry was performed after 2- and 5-week experiments. On-line RT PCR for VEGF and its receptors R1 and R2 was done after a 10-day experiment. In the 5-week experiment, running rats received either a VEGF inhibitory antibody or a placebo.

Results: After 2 weeks, tibial BMD did not change; however, vessel number in the proximal metaphysis increased by 20% in running versus sedentary rats. In running rats, vessel number correlated positively (r = 0.88) with bone formation rate and negatively (r = −0.85) with active resorption surfaces. After 10 days of training, upregulation of VEGF and VEGF receptor R1 mRNA was detected in periosteum and metaphyseal bone. VEGF blockade in 5-week trained rats fully prevented the exercise-induced increase in metaphyseal BMD (9%) and cancellous bone volume (BV/TV; 25%), as well as the increased vessel number (25%). In 5-week placebo-treated running rats, bone formation rate returned to initial values, whereas osteoclastic surfaces continued to decline compared with both sedentary and anti-VEGF-treated running rats.

Conclusion: VEGF signaling-mediated bone angiogenesis is tightly related to exercise-induced bone cellular uncoupling and is indispensable for bone gain induced by exercise.


  1. Top of page
  2. Abstract
  7. Acknowledgements

PHYSICAL TRAINING INCREASES bone mass and strength.(1–4) Although it is thought that exercise produces forces that provide the stimulus for adaptation of bone, the exact mechanisms that lead to tissue adaptation are far from being fully understood.

Contrary to bone, angiogenesis in exercised skeletal muscle leading to a functional increase in capillarity has been more thoroughly studied.(5) This response in skeletal muscle is mediated by a number of angiogenic factors including, most importantly, vascular endothelial growth factor (VEGF). For example, an acute bout of endurance training causes an increase in VEGF expression.(6,7) However, mechanisms for the role of VEGF in modulation of physiological angiogenesis in skeletal muscle by mechanical forces are less well understood.(8) Conversely, inactivity and detraining result in loss of muscle capillaries.(9)

Angiogenesis is an essential component of skeletal development and repair. During endochondral ossification in late embryonic development and in rapid postnatal growth, a major event is the invasion of the cartilage of the growth plate regions of long bones with new capillaries from existing blood vessels.(10) After growth plate closure in adults, the angiogenic switch can transiently be turned on during bone modeling in response to fracture or other pathological conditions such as osteoarthritis.(11,12) Administration of angiogenesis inhibitor completely prevents fracture healing by suppressing formation of both callus and periosteal woven bone.(13) Blood supply and formation of new vasculature might also be important for skeletal integrity (i.e., during bone remodeling). Indeed, in relatively healthy older women, an increased rate of bone loss at the hip and calcaneus is associated with decreased blood flow in the lower extremities(14) and bone mineral decreases in the leg with unilateral occlusive arterial disease.(15)

A variety of growth factors and cytokines including VEGF, the fibroblast growth factor (FGF) family, insulin growth factor-1 (IGF-1), epidermal growth factor (EGF), platelet-derived growth factor-A (PDGF-A), and the transforming growth factor-β (TGF-β) family are involved in bone angiogenesis.(16–20) Among these factors, the endothelial cell-specific mitogen VEGF plays a key role for normal and abnormal angiogenesis.(21,22) Alternative splicing of primary transcript from a single VEGF pre-mRNA produces at least five different isoforms of VEGF, three of them being more prevalent (VEGF 120, 164, and 188) in mice.(23,24) All members of the VEGF family have the common ability to stimulate endothelial cell migration and proliferation, proteolytic activity, and capillary morphogenesis.(21) The action of VEGF is primarily mediated by two major receptor tyrosine kinases: VEGF receptor (VEGFR)1 (Flt-1) and VEGFR2 (KDR/Flk-1). Mice lacking VEGF isoforms 164 and 188 showed delayed invasion of new vessels into the perichondrium and the primary ossification center, which shows the significant role of VEGF at both early and late stages of cartilage vascularization.(25,26) Gerber et al.(27) showed that soluble VEGF receptor chimeric protein almost completely suppressed blood vessel invasion associated with impaired cancellous bone formation and with decreased recruitment and/or differentiation of chondroclasts. Cessation of the anti-VEGF treatment restored the invasion of capillary, bone growth, and resorption of the hypertrophic cartilage; it also normalized the growth plate architecture. It has been shown recently that adenovirus-mediated VEGF gene transfer induces bone formation by increasing osteoblast activity in the intact rabbit femur.(28)

We hypothesized that VEGF-mediated vascular adaptation is a required event for bone adaptation during physical exercise. Previous studies in our laboratory have shown that vessel number was decreased during bone loss induced either by hindlimb unloading or by ovariectomy.(29,30) In this study, we aimed to explore morphological bone vascular adaptation during exercise-induced bone gain and to evaluate whether VEGF signaling plays a key role in this bone angiogenesis process.


  1. Top of page
  2. Abstract
  7. Acknowledgements

Animals and training program

Nine-week-old male Wistar rats, weighing 315 ± 5 g, were acclimatized for 1 week and then began the exercise running program. The light/dark cycle was 12/12 h, with lights on from 7:00 a.m. to 7:00 p.m. The rats were group housed (four per cage) and were allowed free access to water and chow. Body weights were recorded twice a week throughout the experiment.

On the first day, rats from the RUN group ran 15 minutes at a speed of 20 m/minute on a motor-driven treadmill (Quinton Instrument, Seattle, WA, USA). During the first week of training, the duration of each running session was progressively increased up to 90 minutes, 5 days/week. By the fifth week of training, rats ran 90 minutes at a speed of 30 m/minute on the level. This regimen has been previously shown to require 60% of maximal O2 consumption(31) and produce an osteogenic response after 5 weeks.

The procedure for the care and killing of the animals was in accordance with the European Community standards on the care and use of laboratory animals (Ministère de l'Agriculture, France, Authorization 04827).

Experimental design

First experiment:

We first designed a 14-day training experiment to observe bone and vascular alterations occurring before the detectable exercise-induced bone gain. Twenty-four rats were randomly divided into three groups of eight rats each: baseline (Base), running exercise (RUN), and sedentary control (SED) groups.

Second experiment:

Our aim was to study mRNA expression of VEGF and two members of the VEGF receptor family, VEGFR2 and VEGFR1, in periosteum and cancellous bone after a 10-day training program. We wanted to analyze the possible changes in the VEGF pathway occurring before bone angiogenesis adaptation. Six rats (RUN) ran for 10 days according to the protocol previously described, while another group of six weight-matched rats were used as sedentary controls (SED).

Third experiment:

We performed a 5-week VEGF-blockade experiment to study whether VEGF-mediated angiogenesis was required for exercise-induced bone gain. At baseline, BMD and BMC were evaluated on 30 9-week-old rats at the proximal metaphysis of the left tibia using DXA (Lunar PIXImus densitometer, Lunar, Paris, France) under anesthesia (5 mg/kg ketamine/xylazine solution IP). Rats were randomly distributed into three groups of 10 animals: VEGF blockade exercise (RUN anti-VEGF), exercise control (RUN), and sedentary control (SED) groups. The rats in both exercise groups were submitted to a 5-week treadmill-running program as describe above.

The rats in the anti-VEGF (RUN anti-VEGF) group were injected intraperitoneally twice a week with 760 μg anti-VEGF antibody at a dose of 2.5 mg/kg. This dose can inhibit the mitogenic activity of 1 ng/ml of VEGF for human, bovine, or rat endothelial cells with an IC50 of 20 μg/ml. This is a rabbit polyclonal antibody raised against human VEGF (isoform 165).(32,33) The control rats in both the RUN and SED groups were injected with the same volume of saline. Repeated DXA measurements were performed once a week, starting in the second week of training.

For the first and third experiments, the animals were double labeled with tetracycline (20 mg/kg body weight [BW]) and calcein (10 mg/kg BW) by intraperitoneal injection at 5 and 1 day before killing, respectively.

Vascular network quantification

At time of killing by intraperitoneal injection of pentobarbital (60 mg/kg BW), rats were injected subcutaneously with heparin (200 UI/kg) to prevent blood coagulation. Then, a mixture of barium sulfate and India ink was injected into the inferior vena cava to opacify the bone vascular network according to Barou et al.(34) Briefly, the heart was exposed; the left ventricle and the inferior vena cava were both catheterized with a polyethylene catheter. After right auricle section, 90 ml of heparinized saline solution was infused into the left ventricle for 10 minutes to wash out blood from vessels. The animals were infused with 4% paraformaldehyde solution (30 ml) through the left ventricle. The mixture of barium sulfate and India ink (25 g/100 ml) was injected into the vena cava. The infusion was considered complete when the animals were fully colored in black by the injected product from head to tail.

Vascular and bone histomorphometry

The left proximal tibia was excised and fixed in 10% formaldehyde for 24 h at 4°C. Specimens were dehydrated in absolute acetone and embedded in methylmethacrylate at low temperature according to the method developed in our laboratory. The central plane of the proximal part of the tibia was sliced frontally with a microtome (Reichert-Jung Polycut, Heidelberg, Germany). Two 8-μm-thick sections per sample were stained with toluidine blue and used for the evaluation of vascular parameters. The vessel number and area were quantified in the secondary spongiosa (IISP) of proximal metaphyses using an ocular integrator with 100-point grid at 250×. CVs for vascular number and area were 6.68% and 7.37%, respectively.

On the same left proximal tibias used for vascular evaluation, five 8-μm-thick sections were stained with Goldner's trichrome. They were used for measurement in IISP of several parameters according to the ASBMR histomorphometry nomenclature(35) using an automatic image analyzer (BIOCOM, Lyon, France): bone volume (BV/TV, %), mean trabecular thickness (Tb.Th, μm), mean trabecular number (Tb.N, /mm2), and mean trabecular separation (Tb.Sp, μm). Five 8-μm-thick sections were stained with TRACP to measure active osteoclastic surfaces (Oc.S/BS, %). Histodynamic parameters were determined on five unstained, 12-μm-thick sections under UV light: mineral apposition rate (MAR, μm/day), single-labeled surface (sLS/BS, %), and double-labeled surface (dLS/BS, %). Mineralizing surface per bone surface (MS/BS, %) was calculated by adding dLS/BS and one-half sLS/BS. Bone formation rate (BFR/BS, μm3/μm2/day) was calculated as the product of MS/BS and MAR. These parameters of bone resorption and formation were measured with a semiautomatic system made of a digitizing table (Summasketch-Summagraphics, Paris, France) connected to a personal computer and to a Reichert Polyvar microscope equipped with a drawing system (Camera Lucida; Reichert-Jung Polyvar).

Measurement of the mRNA of VEGF and its receptors, VEGFR2 and VEGFR1, by RT-PCR

RNA extraction and purification:

Total RNA was extracted according to the method of Chomczynski and Sacchi.(36) Immediately after the end of the last training session, the periosteum of the two tibias (second experiment) was harvested after complete removal of muscles and ligaments. The bone marrow was washed out from the left tibia with solution D, and proximal metaphysis of left tibia was collected. The sample was homogenized in a sterile tube containing 5 ml of solution D. The samples were rapidly frozen in liquid nitrogen and stored at −80°C until RNA extraction. The content of total RNA was quantified by spectrophotometry. Ten micrograms of RNA was treated with 5 U RQ1 RNase-free DNase (Promega, Charbonnières, France) for 30 minutes at 37°C, followed by two extractions (one in phenol/chloroform and one in chloroform). RNA pellets were dissolved in 0.3 M sodium acetate, precipitated with ethanol, and redissolved with 40 μl distilled water. The content of total RNA was quantified again.

Table Table 1.. Primer Sequences
Thumbnail image of

Quantitative RT-PCR with LightCycler system

Two micrograms of total RNA was reverse transcribed in a 20-μl reaction system using a first-strand cDNA synthesis kit for RT-PCR (Roche Molecular Biochemicals, Meylan, France) according to the supplier's recommendations. Quantification of mRNA of rat VEGF and its receptors was performed by RT-PCR on the LightCycler system (Roche Molecular Biochemicals). Primers were designed as listed in Table 1. A pair of VEGF primers was designed in the common region of all VEGF isoforms to observe a possible change in VEGF expression related to exercise, regardless of the isoform types (Fig. 1A). This ensured that there is only one PCR product when the specificity of the amplified PCR product was appraised by the melting curve program analysis. PCR was performed using LightCycler-FastStart DNA Master SYBR Green 1 reaction mixture. PCR amplification of the housekeeping gene, rat glyceraldehyde-3-phosphate dehydrogenase (GAPDH), was done for each sample as a control of sample loading and to allow normalization between samples. The experimental protocol consists of denaturation at 95°C for 8 minutes, denaturation of 40 cycles at 95°C for 15 s, annealing for 10 s at the tested optimum temperature for each targeted gene, and elongation at 72°C from 6 to 13 s according to the length of the targeted gene. A standard curve was constructed with serial dilutions of one cDNA sample (1/10, 1/20, 1/40, 1/80, and 1/160). Each sample was duplicated, and each PCR amplification included a nontemplate as well as a negative control. The relative expression in each sample was calculated with respect to the standard calibration curve obtained with the quantification program. The melting curve was analyzed to assess the specificity of the amplified PCR product. PCR products were confirmed as single bands using gel electrophoresis, the sizes of the amplified fragment being consistent with the sizes of the target genes (Fig. 1B).

thumbnail image

Figure FIG. 1.. (A) Sequence model of rat VEGF mRNA isoforms. The VEGF gene consists of eight exons. There are five different isoforms of VEGF in mice produced by alternative splicing of primary transcript from a single VEGF gene. The published mRNA sequences of all rat VEGF isoforms have a common region of 420 bps, including exons 1-5 (GenBank number: AF 215725, AF 215726, and AF 222779), and a common sequence, exon 8, consisting of 21 bps. The designed forward primer and reverse primer are in the region of exons 1 and 4, respectively. The amplified products are the total of all VEGF isoforms; using these products, we can observe whether there is a total VEGF expression change related to exercise, regardless of the type of VEGF isoforms. This design will also ensure there is only one PCR product so as not to confuse the possible genomic DNA contamination when the specificity of the amplified PCR product was appraised by the melting curve program analysis. (B) Check for the specificity of the amplified PCR products by electrophoresis after real time PCR. There is only one band in each sample, which is consistent with the size of the target gene. Line 1, DNA marker 1000; line 2, GAPDH; line 3, VEGF; line 4, VEGFR1; line 5, VEGFR2; line 6, negative sample.

Download figure to PowerPoint

Statistical analysis

Comparisons among all the groups within each experiment were made using ANOVA. Comparisons between two specific treatments were made using a two-tailed Student t-test. Statistical significance was p < 0.05. Results are expressed as mean ± SD. Correlations between vascular and bone cellular activity parameters were evaluated with a two-tailed Spearman correlation coefficient.

Table Table 2.. Bone Histomorphometric Parameters in the Secondary Spongiosa of the Left Proximal Tibia
Thumbnail image of


  1. Top of page
  2. Abstract
  7. Acknowledgements

Vascular adaptation occurred before bone gain in the IISP of proximal tibia

The bone mass parameters in IISP of the proximal tibia were not yet modified after the 2-week running program; there were no differences in BV/TV, Tb.Th, Tb.N, or Tb.Sp between the RUN group and the SED control groups (Table 2). BV/TV was higher and Tb.Sp was lower in the RUN group and the SED group than in Base. Two weeks of treadmill training increased the vascular number in the tibial IISP by 19% compared with the SED group, whereas the total vascular area showed no difference (Fig. 2A). The ratio of vessel area to vessel number, representing the mean vessel area, decreased by 18% in RUN compared with SED group (Fig. 2B).

thumbnail image

Figure FIG. 2.. Two weeks of training induced a remarkable vascular adaptation in secondary spongiosa of tibia. (A) Vascular number (p = 0.023, one-way ANOVA) and vascular area. Exercise increased vessel number in secondary spongiosa by 20% compared with the SED control (p = 0.041, two-tailed t-test.). RUN vs. Base (p = 0.02). However, total vascular area showed no significant change (p = 0.308). (B) The ratio of vascular area to vascular number decreased in RUN rat (p = 0.05), indicating that the area per vessel of the cancellous bone in the RUN group is smaller than the SED group. *p < 0.05 compared with the SED control group;#p < 0.05 compared with the Base group, two-tailed t-test; RUN vs. SED, p = 0.035; RUN vs. Base, p = 0.14.

Download figure to PowerPoint

Ten-day treadmill training upregulated mRNA expression of VEGF and its receptors in cancellous bone and periosteum

Figure 3 shows the relative expression of VEGF and its receptors to GAPDH. After 10 days of exercise, expression of VEGF and VEGFR1 mRNAs increased in cancellous bone by 150% and 80%, respectively. Although the expression of VEGFR2 mRNA seemed to increase, the difference between the RUN and SED control groups did not reach statistical significance. Similarly, in the periosteum expression of VEGF and VEGFR1, mRNAs was increased by 86% and 92%, respectively, whereas VEGFR2 mRNA expression did not change.

thumbnail image

Figure FIG. 3.. Exercise upregulates expression of VEGF and its receptors in tibial proximal metaphysis at (A) the cancellous level and (B) tibial periosteum, with data expressed as relative mRNA expression units to GAPDH (mean ± SD). Ten-day training significantly upregulated expression of total VEGF and VEGFR1 by 153% and 83%, respectively (p < 0.05), in the cancellous bone of the tibia. The training increased VEGFR2 mRNA by 26% in the cancellous bone but did not show statistical significance with SED control. Similar upregulation of VEGF (86%) and VEGFR1 (92%) in periosteum was observed in the trained rats (p < 0.05), whereas VEGFR2 mRNA in periosteum showed no change. *p < 0.05 compared with the SED control group, two-tailed t-test.

Download figure to PowerPoint

VEGF blockade with VEGF antibody completely prevented bone gain and bone vascular adaptation induced by 5-week exercise

Sequential analysis of proximal tibia BMD showed that treadmill training induced an increase in bone mass in a time-dependent manner (Fig. 4A) observed as early as the third week in the RUN control group. After 5 weeks, BMD increased by 9% in RUN compared with SED (p = 0.011). VEGF antagonist administration completely prevented the exercise-induced bone gain. Proximal tibia BMD in the RUN anti-VEGF group was similar to that of the SED group and significantly lower than that of the RUN group at 5 weeks.

thumbnail image

Figure FIG. 4.. (A) VEGF blockade with VEGF antibody completely prevented the exercise-induced bone gain shown by dynamic analysis of BMD in metaphysis of left proximal tibia. At 5 weeks, the trained rats achieved a 9% increase in BMD compared with the SED control group (p = 0.011). The change in BMD of VEGF-blocked rats was similar to that observed in SED controls and significantly lower than that observed in the untreated trained rats at 5 weeks (p = 0.023). (B) Similar preventive effect of VEGF blocking was observed on BMC, but group means were not significantly different. *p < 0.05 compared with the SED control group;#p < 0.05 compared with the VEGF blocking group, two-tailed t-test.

Download figure to PowerPoint

Histomorphometric analysis showed that VEGF blockade exerted similar preventive effects on cancellous bone gain induced by exercise (Table 2). In left tibia IISP, BV/TV and Tb.N increased by 25% and 21%, respectively, compared with the SED group after 5 weeks of training. This cancellous bone gain was associated with thicker trabeculae and reduced trabecular separation. BV/TV in the RUN anti-VEGF group was markedly lower than that of the RUN group.

Alterations in intramedullary vessel number and average area per vessel induced by 5-week exercise were fully prevented by VEGF blockade (Figs. 5A and 5B).

thumbnail image

Figure FIG. 5.. VEGF blockade completely prevented exercise-induced bone vascular adaptation. (A) Five-week training increased vascular number in secondary spongiosae of the tibia by 26% compared with the SED control (p = 0.017), although it did not affect the total vascular area. VEGF blockade completely prevented an increase in the exercise-induced bone vascular number. (B) The training significantly decreased the average area per vessel in the cancellous bone of tibia by 17.7% compared with SED control group (p = 0.002). VEGF blockade also decreased the average area per vessel in the exercise-induced bone vascular number. *p < 0.05 compared with the SED control group;#p < 0.05 compared with the VEGF blockade group, two-tailed t-test.

Download figure to PowerPoint

Relationship between bone angiogenesis and bone cellular activities in the IISP of the tibial metaphysis

As expected, histomorphometry revealed that 2 weeks of exercise resulted in a significant increase in bone formation (MAR, dLS/BS, and BFR/BS) and a decrease in bone resorption (active osteoclastic surfaces; Table 2). After 5 weeks, however, differences in these bone formation parameters were no longer observed between RUN and SED groups. Osteoblastic activities in the RUN anti-VEGF group did not differ from the RUN and SED groups. In contrast, Oc.S./BS were still significantly lower after 5 weeks in the RUN than in the SED groups. Administration of the anti-VEGF antibody completely prevented this exercise-induced decrease in osteoclastic resorption. The longitudinal growth rates evaluated in 5-week SED, RUN, and RUN anti-VEGF groups did not vary significantly (77.7 ± 6.4, 83.8 ± 7.3, and 85.7 ± 8.1 μm/day, respectively).

At 2 weeks, MAR, dLS/BS, and BFR/BS were positively correlated with vessel number in IISP of tibial metaphysis in the RUN group, whereas only MAR showed significant positive correlation with vessel number in the SED group (Table 2; Fig. 6). In addition, vessel number was negatively correlated with Oc.S/BS in the RUN group only (Fig. 6D). In contrast, after 5 weeks of exercise, vessel number in the tibial IISP exhibited no relationship with any parameter of bone formation and resorption (Table 2).

thumbnail image

Figure FIG. 6.. Relationship between bone angiogenesis and bone cellular activities in the secondary spongiosa of the tibial metaphysis. At 2 weeks, the MAR, the dLS/BS, and BFR showed significantly positive correlation with vessel number in the secondary spongiosa of the tibial metaphysis in RUN group, whereas only MAR showed significantly positive correlation with vessel number in the SED control group. In addition, there was a negative correlation between vessel number and active eroded osteoclast surfaces in RUN group at 14 days, but we did not find this negative correlation in the SED control group. At 5 weeks, the vessel number in the tibial IISP did not show significant correlation with any parameter of bone formation or bone resorption in any group of SED, RUN, and VEGF (data not shown). Analysis was by two-tailed Spearman correlation coefficient.

Download figure to PowerPoint


  1. Top of page
  2. Abstract
  7. Acknowledgements

We confirmed the kinetics of exercise-induced bone alterations as previously shown in our laboratory.(31) After 2 weeks of treadmill training, cancellous bone formation at the tibial metaphysis was increased, and resorption activity was decreased; however, BMD and cancellous BV/TV were not yet altered. During this adaptation phase, such bone cellular uncoupling leads to positive bone balance and greater bone mass, as seen after the third week of running training. After 5 weeks of training, bone formation activity was no longer elevated, but bone resorption remained depressed compared with SED controls. These cellular events might contribute to maintaining a positive balance responsible for the thicker trabeculae.

To our knowledge, this is the first study that has directly correlated physiologic loading with markers of angiogenesis and vascular morphometry in bone during physical exercise. Indeed, we found in proximal tibia that 10 days of running at 60% VO2max induced significant increases in VEGF and VEGFR1 mRNA expression in cancellous bone and in the number of intramedullary vessels, which increased 20% after 2 weeks and 26% after 5 weeks. We did not assess vessel morphology in the periosteum but we did verify an increased VEGF and VEGFR1 mRNAs expression after 10 days of running, suggesting that running training induces angiogenic activity through VEGF upregulation at the periosteal surface of midshaft bone as well. Profound microcirculatory disturbances have been reported after closed tibial fracture in both muscle and periosteum.(37) Predominance of periosteal blood circulation to the bone cortex compared with centromedullary vascularization has been emphasized,(38) suggesting that cortical perfusion is also altered during physical exercise.

Electrical stimulation of ischemic muscles can promote angiogenesis using VEGFR2 upregulation, although VEGFR1 expression is elevated in ischemic nonstimulated muscles.(39) It is known that VEGFR1 is critical to vessel survival and enhances endothelial migration,(40,41) whereas VEGFR2 promotes vasoproliferation.(42) In our study, there was a significant increase in VEGFR1 mRNA expression and a trend to increased VEGFR2 mRNA expression after a 10-day running period. Interestingly, it has been shown by in situ hybridization that osteoblasts strongly express VEGFR1, whereas VEGFR2 is minimal in C57BL6 mice.(43) Exercise also significantly upregulates VEGFR1 mRNA in the rat gastrocnemius muscle.(44)

To better assess the VEGF importance in exercise-induced bone gain, we treated RUN animals with blocking VEGF antibody over a 5-week period. We first found that VEGF blockade fully prevented the exercise-induced bone vascular adaptation, because vascular morphological parameters were similar in trained rats with VEGF blockade and in SED control animals. We also showed that VEGF blockade fully prevented the exercise-induced cortical and cancellous bone gains, as seen by longitudinal BMD assessment from 2 to 5 weeks, as well as by histomorphometric measures at the end of the experiment. These findings clearly suggest that the gains in bone are coupled to VEGF-mediated events. However, an indirect effect mediated by vascular-induced events on the adjacent muscle tissue cannot be fully ruled out.

Interestingly, during the bone adaptation phase to treadmill training, that is, during the first 2 weeks, the intramedullary vessel number correlated positively with bone formation parameters and negatively with active resorption surfaces. This probably constitutes a coordinated adjustment devoted to meet the bone metabolic requirements that are presumably upregulated by physical activity. Several lines of evidence in the literature support this proposed causal link between angiogenesis and bone formation activity. VEGF is expressed in fracture callus in much the same temporal and spatial pattern as during long bone development.(45,46) In an endochondral ossification model (induced by long bone fracture), treatment with a neutralizing VEGF receptor decreased angiogenesis, bone formation, and callus mineralization; the same treatment also inhibited healing of a cortical bone defect (intramembranous ossification model).(47) This is consistent with reports of expression of VEGF and its receptors, VEGFR1 and VEGFR2, in chondro-osteogenic lineage cells(25,48–50) and of a direct autocrine role for VEGF in osteoblast differentiation and chemotactic migration.(50–52) Sprouting microvessel endothelial cells are accompanied by vascular pericytes that undergo osteogenic differentiation in vivo and in vitro. In addition, VEGF may act as a central mediator for other factors. For example, inhibition of VEGF blocks the angiogenic activity of basic fibroblast growth factor (bFGF) and bone morphogenetic protein (BMP)-2(53,54) and induction of osteoblast differentiation by matrix metalloproteinase 7 (MMP7/osteogenic protein-1 [OP-1]).(55) Furthermore, most osteoinductive factors,(54,56,57) including mechanical strain,(58,59) stimulate osteoblastic VEGF production.

It is noteworthy that treatment with the VEGF antibody had no apparent effect on bone formation activity in proximal tibia cancellous bone or on longitudinal growth rate when assessed after 5 weeks. Thus, in these conditions, VEGF blockade seems to affect only those parameters undergoing active adaptation to the exercise stimulus and not parameters that have stabilized in normal range.

The negative correlation between vessel number and bone resorption parameters seen after 2 weeks of training and the decreased active resorption surfaces associated with increased VEGF expression might seem more puzzling. Decreased osteoclastic resorption is a well-known adaptation to physical exercise in growing rats.(2–4,31) Osteoclasts are also target cells for VEGF; they express both distinct VEGF receptors, VEGFR1 and VEGFR2.(60) During skeletal development, presence of chondroclastic/osteoclastic cells in hypertrophic cartilage is always accompanied by the invasion of blood vessels(25,26); the recruitment of osteoclasts into hypertrophic cartilage requires VEGF, which acts as a chemoattractant for osteoclasts.(61) Mice lacking VEGF isoforms 165 and 188 exhibit significantly reduced osteoclastic resorption.(25,26) In vitro and in vivo studies have verified that VEGF enhances osteoclastic bone resorption and survival of mature osteoclasts.(60,62) However, other cytokines produced by endothelial cells such as endothelin and NO inhibit osteoclastic resorption(63–65); expression of these factors is not accounted for in this study. Moreover, resorption surfaces have been found to decrease in the rabbit femur after VEGF gene transfer.(28) Consistent with these findings, we showed that administration of a VEGF antagonist completely inhibited the exercise-induced decrease of osteoclastic resorption.

In conclusion, our findings indicate that upregulation of VEGF and its receptor, VEGF-R1, at the cancellous and periosteal envelopes mediate bone angiogenesis that is tightly coupled to increased osteoblastic activity and decreased osteoclastic resorption during the early adaptation to exercise. These vascular events are indispensable for exercise-induced bone gain, because a neutralizing VEGF antibody fully prevented bone vascular events and bone adaptation to treadmill training.


  1. Top of page
  2. Abstract
  7. Acknowledgements

The authors thank Veronique Coxam for the treadmill. This study was supported by INSERM. INSERM also provided a post doc salary to ZY. This study was funded by ERISTO Contract 14232/00/NL/SH and St Etienne University.


  1. Top of page
  2. Abstract
  7. Acknowledgements
  • 1
    Rubin C, Turner AS, Bain S, Mallinckrodt C, McLeod K 2001 Anabolism. Low mechanical signals strengthen long bones. Nature 412: 603604.
  • 2
    Notomi T, Okimoto N, Okazaki Y, Tanaka Y, Nakamura T, Suzuki M 2001 Effects of tower climbing exercise on bone mass, strength, and turnover in growing rats. J Bone Miner Res 16: 166174.
  • 3
    Notomi T, Lee SJ, Okimoto N, Okazaki Y, Takamoto T, Nakamura T, Suzuki M 2000 Effects of resistance exercise training on mass, strength, and turnover of bone in growing rats. Eur J Appl Physiol 82: 268274.
  • 4
    Iwamoto J, Yeh JK, Aloia JF 1999 Differential effect of treadmill exercise on three cancellous bone sites in the young growing rat. Bone 24: 163169.
  • 5
    Prior BM, Lloyd PG, Yang HT, Terjung RL 2003 Exercise-induced vascular remodelling. Exerc Sport Sci Rev 31: 2633.
  • 6
    Richardson RS, Wagner H, Mudaliar SR, Henry R, Noyszewski EA, Wagner PD 1999 Human VEGF gene expression in skeletal muscle: Effect of acute normoxic and hypoxic exercise. Am J Physiol 277: H2247H2252.
  • 7
    Hiscock N, Fischer CP, Pilegaard H, Pedersen BK 2003 Vascular endothelial growth factor mRNA expression and arteriovenous balance in response to prolonged, submaximal exercise in humans. Am J Physiol Heart Circ Physiol 285: H1759H1763.
  • 8
    Brown MD, Hudlicka O 2003 Modulation of physiological angiogenesis in skeletal muscle by mechanical forces: Involvement of VEGF and metalloproteinases. Angiogenesis 6: 114.
  • 9
    Mujika I, Padilal S 2001 Muscular characteristics of detraining in humans. Med Sci Sports Exerc 33: 12971303.
  • 10
    Gerber HP, Ferrara N 2000 Angiogenesis and bone growth. Trends Cardiovasc Med 10: 223228.
  • 11
    Glowacki J 1998 Angiogenesis in fracture repair. Clin Orthop 355 (Suppl): S82S89.
  • 12
    Fenwick SA, Gregg PJ, Rooney P 1999 Osteoarthritic cartilage loses its ability to remain avascular. Osteoarthritis Cartilage 7: 441452.
  • 13
    Hausman MR, Schaffler MB, Majeska RJ 2001 Prevention of fracture healing in rats by an inhibitor of angiogenesis. Bone 29: 560564.
  • 14
    Vogt MT, Cauley JA, Kuller LH, Nevitt MC 1997 Bone mineral density and blood flow to the lower extremities: The study of osteoporotic fractures. J Bone Miner Res 12: 283289.
  • 15
    Laroche M, Moulinier L, Leger P, Lefebvre D, Mazières B, Boccalon H 2003 Bone mineral decrease in the leg with unilateral occlusive arterial disease. Clin Exp Rheumatol 21: 103106.
  • 16
    Garcia-Ramirez M, Toran N, Andaluz P, Carrascosa A, Audi L 2000 Vascular endothelial growth factor is expressed in human fetal growth cartilage. J Bone Miner Res 15: 534540.
  • 17
    Ornit DM, Marie PJ 2000 FGF signaling pathways in endochondral and intramembranous bone development and human genetic disease. Genes Dev 16: 14461465.
  • 18
    Fiorelli G, Orlando C, Benvenuti S, Franceschelli F, Bianchi S, Pioli P, Tanini A, Serio M, Bartucci F, Brandi ML 1994 Characterization, regulation, and function of specific cell membrane receptors for insulin-like growth factor I on bone endothelial cells. J Bone Miner Res 9: 329337.
  • 19
    Horner A, Bord S, Kemp P, Grainger D, Compston JE 1996 Distribution of platelet-derived growth factor (PDGF) A chain mRNA, protein, and PDGF-alpha receptor in rapidly forming human bone. Bone 19: 353362.
  • 20
    Dickson KM, Bergeron JJ, Philip A, O'Connor-McCourt M, Warshawsky H 2001 Localization of specific binding sites for 125I-TGF-beta1 to fenestrated endothelium in bone and anastomosing capillary networks in enamel organ suggests a role for TGF-beta1 in angiogenesis. Calcif Tissue Int 68: 304315.
  • 21
    Ferrara N, Gerber HP, LeCouter J 2003 The biology of VEGF and its receptors. Nat Med 9: 669676.
  • 22
    Zachary I 2001 Signalling mechanisms mediating vascular protective actions of vascular endothelial growth factor. Am J Physiol Cell Physiol 280: C1375C1386.
  • 23
    Tober KL, Cannon RE, Spalding JW, Oberyszyn TM, Parrett ML, Rackoff AI, Oberyszyn AS, Tennant RW, Robertson FM 1998 Comparative expression of novel vascular endothelial growth factor/vascular permeability factor transcripts in skin, papillomas, and carcinomas of v-Ha-ras Tg.AC transgenic mice and FVB/N mice. Biochem Biophys Res Commun 247: 644653.
  • 24
    Zhang L, Conejo-Garcia JR, Yang N, Huang W, Mohamed-Hadley A, Yao W, Benencia F, Coukos G 2002 Different effects of glucose starvation on expression and stability of VEGF mRNA isoforms in murine ovarian cancer cells. Biochem Biophys Res Commun 292: 860868.
  • 25
    Zelzer E, McLean W, Ng YS, Fukai N, Reginato AM, Lovejoy S, D'Amore PA, Olsen BR 2002 Skeletal defects in VEGF (120/120) mice reveal multiple roles for VEGF in skeletogenesis. Development 129: 18931904.
  • 26
    Maes C, Carmeliet P, Moermans K, Stockmans I, Smets N, Collen D, Bouillon R, Carmeliet G 2002 Impaired angiogenesis and endochondral bone formation in mice lacking the vascular endothelial growth factor isoforms VEGF (164) and VEGF (188). Mech Dev 111: 6173.
  • 27
    Gerber HP, Vu TH, Ryan AM, Kowalski J, Werb Z, Ferrara N 1999 VEGF couples hypertrophic cartilage remodeling, ossification and angiogenesis during endochondral bone formation. Nat Med 5: 623628.
  • 28
    Hiltunen MO, Ruuskanen M, Huuskonen J, Mahonen AJ, Ahonen M, Rutanen J, Kosma VM, Mahonen A, Kroger H, Yla-Herttuala S 2003 Adenovirus-mediated VEGF-A gene transfer induces bone formation in vivo. FASEB J 17: 11471149.
  • 29
    Barou O, Vico L, Alexandre C, Lafage-Proust MH 2000 Early bone vascular alterations in the rat hindlimb unloading model. J Bone Miner Res 15: S301. (abstract)
  • 30
    Mekraldi S, Lafage-Proust MH, Bloomfield S, Alexandre C, Vico L 2003 Changes in vasoactive factors associated with altered vessel morphology in the tibial metaphysis during ovariectomy-induced bone loss in rats. Bone 32: 630641.
  • 31
    Bourrin S, Palle S, Pupier R, Vico L, Alexandre C 1995 Effect of physical training on bone adaptation in three zones of the rat tibia. J Bone Miner Res 10: 17451752.
  • 32
    Guerrin M, Moukadiri H, Chollet P, Moro F, Dutt K, Malecaze F, Plouët J 1995 Vasculotropin/vascular endothelial growth factor is an autocrine growth factor for human retinal pigment epithelial cells cultured in vitro. J Cell Physiol 164: 385394.
  • 33
    Berard M, Sordello S, Ortega N, Carrier JL, Peyri N, Wassef M, Bertrand N, Enjolras O, Drouet L, Plouet J 1997 Vascular endothelial growth factor confers a growth advantage in vitro and in vivo to stromal cells cultured from neonatal hemangiomas. Am J Pathol 150: 13151326.
  • 34
    Barou O, Mekraldi S, Vico L, Boivin G, Alexandre C, Lafage-Proust MH 2002 Relationships between trabecular bone remodeling and bone vascularization: A quantitative study. Bone 30: 604612.
  • 35
    Parfitt AM, Drezner MK, Glorieux FH, Kanis JA, Malluche H, Meunier PJ, Ott SM, Recker RR 1987 Bone histomorphometry: Standarization of nomenclature, symbols and units. J Bone Miner Res 2: 595610.
  • 36
    Chomczynski P, Sacchi N 1987 Single-step method of RNA isolation by acid guanidinium thiocyanate-phenol-chloroform extraction. Anal Biochem 162: 156159.
  • 37
    Zhang L, Bail H, Mittlmeier T, Haas NP, Schaser KD 2003 Immediate microcirculatory derangements in skeletal muscle and periosteum after closed tibial fracture. J Trauma 54: 979985.
  • 38
    Chanavaz M 1995 Anatomy and histophysiology of the periosteum: Quantification of the periosteal blood supply to the adjacent bone with 85Sr and gamma spectrometry. J Oral Implantol 21: 214219.
  • 39
    Milkiewicz M, Hudlicka O, Verhaeg J, Egginton S, Brown MD 2003 Differential expression of Flk-1 and Flt-1 in rat skeletal muscle in response to chronic ischaemia: Favorable effect of muscle activity. Clin Sci (Lond) 105: 473482.
  • 40
    Waltenberger J, Claesson-Welsh L, Siegbahn A, Shibuya M, Heldin CH 1994 Different signal transduction properties of KDR and Flt-1, two receptors for vascular endothelial growth factor. J Biol Chem 269: 2698826995.
  • 41
    Shih SC, Ju M, Liu N, Smith LE 2003 Selective stimulation of VEGFR-1 prevents oxygen-induced retinal vascular degeneration in retinopathy of prematurity. J Clin Invest 112: 5057.
  • 42
    Millauer B, Shawver LK, Plate KH, Risau W, Ullrich A 1994 Glioblastoma growth inhibited in vivo by a dominant-negative Flk-1 mutant. Nature 367: 576579.
  • 43
    Gerber HP, Vu TH, Ryan AM, Kowalski J, Werb Z, Ferrara N 1999 VEGF couples hypertrophic cartilage remodeling, ossification and angiogenesis during endochondral bone formation. Nat Med 5: 623628.
  • 44
    Gavin TP, Wagner PD 2002 Attenuation of the exercise-induced increase in skeletal muscle Flt-1 mRNA by nitric oxide synthase inhibition. Acta Physiol Scand 175: 201209.
  • 45
    Ferguson C, Alpern E, Miclau T, Helms JA 1999 Does adult fracture repair recapitulate embryonic skeletal formation? Mech Dev 87: 5766.
  • 46
    Tatsuyama K, Maezawa Y, Bab H, Imamura Y, Fukuda M 2000 Expression of various growth factors for cell proliferation and cytodifferentiation during fracture repair of bone. Eur J Histochem 44: 269278.
  • 47
    Street J, Bao M, deGuzman L, Bunting S, Peale FV Jr, Ferrara N, Steinmetz H, Hoeffel J, Cleland JL, Daugherty A, van Bruggen N, Redmond HP, Carano RA, Filvaroff EH 2002 Vascular endothelial growth factor stimulates bone repair by promoting angiogenesis and bone turnover. Proc Natl Acad Sci USA 99: 96569661.
  • 48
    Deckers MM, Karperien M, van der Bent C, Yamashita T, Papapoulos SE, Lowik CW 2000 Expression of vascular endothelial growth factors and their receptors during osteoblast differentiation. Endocrinology 141: 16671674.
  • 49
    Saadeh PB, Mehrara BJ, Steinbrech DS, Spector JA, Greenwald JA, Chin GS, Ueno H, Gittes GK, Longaker MT 2000 Mechanisms of fibroblast growth factor-2 modulation of vascular endothelial growth factor expression by osteoblastic cells. Endocrinology 141: 20752083.
  • 50
    Wang DS, Miura M, Demura H, Sato K 1997 Anabolic effects of 1,25-dihydroxyvitamin D3 on osteoblasts are enhanced by vascular endothelial growth factor produced by osteoblasts and by growth factors produced by endothelial cells. Endocrinology 138: 29532962.
  • 51
    Mayr-Wohlfart U, Waltenberger J, Hausser H, Kessler S, Gunther KP, Dehio C, Puhl W, Brenner RE 2002 Vascular endothelial growth factor stimulates chemotactic migration of primary human osteoblasts. Bone 30: 472477.
  • 52
    Midy V, Plouet J 1994 Vasculortropin/vascular endothelial growth factor induces differention in cultured osteoblasts. Biochem Biophys Res Commun 199: 380386.
  • 53
    Claffey KP, Abrams K, Shih SC, Brown LF, Mullen A, Keough M 2001 Fibroblast growth factor 2 activation of stromal cell vascular endothelial growth factor expression and angiogenesis. Lab Invest 81: 6175.
  • 54
    Deckers MM, van Bezooijen RL, van der Horst G, Hoogendam J, van Der Bent C, Papapoulos SE, Lowik CW 2002 Bone morphogenetic proteins stimulate angiogenesis through osteoblast-derived vascular endothelial growth factor A. Endocrinology 143: 15451553.
  • 55
    Yeh LC, Lee JC 1999 Osteogenic protein-1 increases gene expression of vascular endothelial growth factor in primary cultures of fetal rat calvaria cells. Mol Cell Endocrinol 153: 113124.
  • 56
    Chua CC, Hamdy RC, Chua BH 2000 Mechanism of transforming growth factor-beta1-induced expression of vascular endothelial growth factor in murine osteoblastic MC3T3-E1 cells. Biochem Biophys Acta 1497: 6976.
  • 57
    Kozawa O, Matsuno H, Uematsu TJ 2001 Involvement of p70 S6 kinase in bone morphogenetic protein signaling: Vascular endothelial growth factor synthesis by bone morphogenetic protein-4 in osteoblasts. Cell Biochem 81: 430436.
  • 58
    Kohno S, Kaku M, Tsutsui K, Motokawa M, Ohtani J, Tenjo K, Tohma Y, Tokimasa C, Fujita T, Kawata T, Tanne K 2003 Expression of vascular endothelial growth factor and the effects on bone remodeling during experimental tooth movement. J Dent Res 82: 177182.
  • 59
    Fong KD, Nacamuli RP, Loboa EG, Henderson JH, Fang TD, Song HBM, Cowan CM, Warren SM, Carter DR, Longaker MT 2003 Equibiaxial tensile strain affects calvarial osteoblast biology. J Craniomaxillofac Surg 14: 348355.
  • 60
    Nakagawa M, Kaneda T, Arakawa T, Morita S, Sato T, Yomada T, Hanada K, Kumegawa M, Hakeda Y 2000 Vascular endothelial growth factor (VEGF) directly enhances osteoclastic bone resorption and survival of mature osteoclasts. FEBS Lett 473: 161164.
  • 61
    Engsig MT, Chen QJ, Vu TH, Pedersen AC, Therkidsen B, Lund LR, Henriksen K, Lenhard T, Foged NT, Werb Z, Delaisse JM 2000 Matrix metalloproteinase 9 and vascular endothelial growth factor are essential for osteoclast recruitment into developing long bones. J Cell Biol 151: 879889.
  • 62
    Matsumoto Y, Tanaka K, Hirata G, Hanada M, Matsuda S, Shuto T, Iwamoto Y 2002 Possible involvement of the vascular endothelial growth factor-Flt-1-focal adhesion kinase pathway in chemotaxis and the cell proliferation of osteoclast precursor cells in arthritic joints. J Immunol 168: 58245831.
  • 63
    Chiao JW, Moonga BS, Yang YM, Kancherla R, Mittelman A, Wu-Wong JR, Ahmed T 2000 Endothelin-1 from prostate cancer cells is enhanced by bone contact which blocks osteoclastic bone resorption. Br J Cancer 83: 360365.
  • 64
    Alam AS, Gallagher A, Shankar V, Ghatei MA, Datta HK, Huang CL, Moonga BS, Chambers TJ, Bloom SR, Zaidi M 1992 Endothelin inhibits osteoclastic bone resorption by a direct effect on cell motility: Implications for the vascular control of bone resorption. Endocrinology 130: 36173624.
  • 65
    MacIntyre I, Zaidi M, Alam AS, Datta HK, Moonga BS, Lidbury PS, Hecker M, Vane JR 1991 Osteoclastic inhibition: An action of nitric oxide not mediated by cyclic GMP. Proc Natl Acad Sci USA 88: 29362940.