- Email: [email protected]
Intravital microscopic studies of angiogenesis during bone defect healing in mice calvaria
Injury, Int. J. Care Injured 42 (2011) 765–771
Contents lists available at ScienceDirect
Injury journal homepage: www.elsevier.com/locate/injury
Intravital microscopic studies of angiogenesis during bone defect healing in mice calvaria J.H. Holstein a,b,c,*, S.C. Becker a,b, M. Fiedler a,b, P. Garcia a,b,c, T. Histing a,b,c, M. Klein a,b, M.W. Laschke b,c, M. Corsten b, T. Pohlemann a,c, M.D. Menger b,c a b c
Department of Trauma, Hand & Reconstructive Surgery, University of Saarland, Homburg/Saar, Germany Institute for Clinical & Experimental Surgery, University of Saarland, Homburg/Saar, Germany Collaborative Research Center AO Foundation, University of Saarland, Homburg/Saar, Germany
A R T I C L E I N F O
A B S T R A C T
Article history: Accepted 10 November 2010
Purpose: Due to the great availability of specific antibodies, gene-targeted animals and knockout strains, mouse models came into the focus of musculoskeletal research. Herein, we introduce a calvarian defect model in mice that allows the repetitive analysis of blood vessel formation during bone repair by intravital microscopy. Methods: The right parietal calvaria of 20 adult CD-1 mice were exposed by skin excision. Under continuous irrigation, a circular defect (Ø0.75 mm) was drilled into the calvarium without penetrating the inner cortical shell. A circular glass (Ø12 mm; thickness 0.15 mm) was fixed by two microscrews (M1; length 2 mm) to cover the bone defect. Angiogenesis was analysed by intravital microscopy at days 0, 3, 6, 9, 12, 15, 18 and 21. In addition, bone repair was evaluated by histomorphometry at days 3, 6, 9 and 15. Immunohistochemical stainings for the angiogenic growth factor vascular endothelial growth factor (VEGF) and the cell proliferation marker proliferating cell nuclear antigen (PCNA) were performed to assess angiogenic and proliferative activity during healing of the calvarian defect. Results: Histomorphometry showed a typical pattern of intramembranous bone repair. Osseous bridging of the defect was observed at day 9. This was associated with the formation of a neo-periosteum, which covered the new woven bone and contained a dense network of newly formed blood vessels. At day 9, particularly cells of the neo-periosteum showed intense staining for VEGF, whilst PCNA-positive staining was found mainly in osteoblasts. At day 15, the major fraction of fibrous tissue was replaced by bone undergoing extensive remodelling. Intravital microscopy revealed an increase of vascular density between days 3 and 15. Blood vessel diameters showed an increase between days 3 and 9 and a subsequent decrease between days 9 and 21. Conclusions: The present calvarian defect model provides a powerful tool to evaluate the process of angiogenesis during intramembranous bone repair in mice. ß 2010 Elsevier Ltd. All rights reserved.
Keywords: Bone repair Angiogenesis Calvarium Intravital microscopy Mice
Angiogenesis has been identified as one key factor determining the outcome of bone repair. Accordingly, inhibition of angiogenic growth factors leads to severe alteration of bone repair, whilst their administration is capable of stimulating bone formation.5,8,9 The investigation of molecular and cellular mechanisms promoting the interaction between blood vessel formation and bone regeneration represents an important topic of current musculoskeletal research. Different imaging techniques, including laser Doppler flowmetry, magnetic resonance imaging (MRI), and orthogonal polarisation imaging, have been used to analyse the role of
* Corresponding author at: Department of Trauma, Hand and Reconstructive Surgery, University of Saarland, D-66421 Homburg/Saar, Germany. Tel.: +49 6841 16 31501; fax: +49 6841 16 31503. E-mail address: [email protected] (J.H. Holstein). 0020–1383/$ – see front matter ß 2010 Elsevier Ltd. All rights reserved. doi:10.1016/j.injury.2010.11.020
angiogenesis during bone repair.6,12,20 Besides, immunohistochemistry provides an additional tool to quantify blood vessel density within regenerating bone tissue. In contrast to these imaging and histological techniques, intravital microscopy bears the great advantage to allow repeated real-time analyses of the microcirculation and new blood vessel formation over a prolonged period of time.16 For this reason, several bone-chamber models have been introduced to assess angiogenesis during bone healing. As early as 1959, Branemark developed a hollow-screw model for intravital microscopy in bone.3 This original bone-chamber model, which later was modified by McCuskey et al., Albrektsson and Albrektsson, Winet and others, allows a long-term observation of the microcirculation in newly formed bone up to 1 year.15,1,19 So far, the hollow-screw model has not been applied in animals smaller than rabbits. In contrast to rabbits, however, small rodents and, especially, mice bear the advantage of a broad spectrum of
766
J.H. Holstein et al. / Injury, Int. J. Care Injured 42 (2011) 765–771
specific antibodies, gene-targeted animals and knockout strains available to study molecular and cellular aspects of angiogenesis during bone regeneration. Accordingly, there is an essential need to apply the powerful tool of intravital microscopy to the analysis of angiogenesis and bone repair in mice. Herein, we introduce a calvarian defect model, which allows intravital microscopic studies of the microcirculation and new blood vessel formation during bone defect healing in mice. The study was approved by the local governmental animal care committee and was conducted in accordance with the German legislation on protection of animals and the National Institutes of Health (NIH) Guidelines for the Care and Use of Laboratory Animals. Methods Animals For the present study, we used adult female CD-1 mice with a mean age of 20 weeks and a body weight of 30–40 g. During the experiments, the mice had free access to tap water and standard pellet food (Altromin, Lage, Germany). Preparation of the calvarian defect model For the preparation of the calvarian defect model, the mice were anaesthetised by intraperitoneal injection of ketamine (75 mg kg body weight 1; Pharmacia GmbH, Erlangen, Germany) and xylazine 2% (25 mg kg body weight 1; Bayer, Leverkusen, Germany). After shaving of the scalp, the mice were fixed on a custom-made support. A circular excision (Ø10 mm) of the scalp and the underlying periosteum was performed to expose the frontal and both parietal aspects of the calvarium. Using a microtrephine, a circular defect (Ø0.75 mm) was drilled under continuous irrigation into the right parietal calvarium, proximately to the sagittal and coronal sutures. Penetration of the inner cortical shell was avoided to prevent an injury of the meninges adhering underneath the calvarium. To cover the defect, a circular glass (12 mm; thickness 0.15 mm) was fixed to the bone by two microscrews (M1.0; length 2 mm). These screws were inserted into the right and left parietal bone at a distance of 3 mm to the sagittal suture. After preparation of the bone chamber, the circular skin defect was tightened by a purse-string suture (Fig. 1). Intravital fluorescence microscopy For intravital microscopy, the mice were anaesthetised and fixed on a custom-made support as reported above. Intravital fluorescence microscopy was performed after retrobulbar IV injection of the plasma tracer fluorescein isothiocyanate (FITC)-labelled dextran 150 000 (5%, 0.1 ml). Epi-illumination was achieved with a 100-W mercury lamp using a fluorescence filter for FITC (Zeiss Axiotech Microscope; Zeiss, Oberkochen, Germany). The microscopic images were recorded using a charge-coupled device (CCD) camera (FK 6990; Pieper, Schwerte, Germany), displayed on a 14-in. video screen (KV-14C T1E; Sony, Tokyo, Japan) and stored on video tapes for off-line analysis. The images were assessed quantitatively using the computer-assisted off-line analysis system CapImage (Zeintl, Heidelberg, Germany).2,13 Defect size and angiogenesis were evaluated using a 10 long-distance objective. Microvessel density and blood vessel diameter were used as parameters of angiogenesis. The defect area was marked manually and calculated by the image analysis system in millimetres.2 Blood vessel density was calculated as the length of red blood cell-perfused blood vessels per observation area, and is given as per millimetre. The observation area was defined as the bone defect area at day 0. Blood vessel diameter was
Fig. 1. Preparation of the calvarian defect model (A and B). A circular defect (1) is drilled into the right parietal calvarium (2) proximately to the sagittal and coronal sutures (3 and 4). Penetration of the inner cortical shell is avoided to prevent an injury of the meninges adhering underneath the calvarium (5). To cover the defect a
J.H. Holstein et al. / Injury, Int. J. Care Injured 42 (2011) 765–771
767
Fig. 2. Intravital fluorescence microscopy of the bone defect (dotted circle) after injection of the plasma tracer fluorescein isothiocyanate (FITC)-labelled dextran 150 000 at days 3, 6, 9, and 15 (A, C, E, and G). At days 3 and 6 most of the defects show complete lack of blood vessel perfusion (A and C). In contrast, at days 9 and 15 a dense network of newly formed blood vessels including numerous blood vessel sprouts is observed (E and G). This network of blood vessels covers not only the defect but also the entire calvarium within the bone chamber including the surrounding bone surface. The defect area does not change significantly until day 6 (A and C), but is found decreased at day 9 (E) and no longer detectably by intravital microscopy at day 15 (G). Histology shows a typical pattern of intramembranous bone repair (B, D, F, and H). The defect (bd) of the calvarian bone (cb) is filled with fibrous tissue (ft) at day 6 (C). At day 9, we found extensive periosteal bone formation bridging the defect (F). The new woven bone (wb) is covered by a fibrous tissue membrane demonstrating the characteristics of periosteum (p). Of interest, numerous blood vessels are visible within this neo-periosteum. At day 9, some remaining fibrous tissue is detectable in the centre of the defect (F), whilst at day 15 all fibrous tissue within the defect is replaced by bone undergoing remodelling (H). Scale bars represent 0.5 mm (A, C, E, and G) and 0.25 mm (B, D, F, and H).
measured perpendicularly to the vessel path. For this purpose, the mean blood vessel diameter of 15 characteristic blood vessels, which
were selected randomly using an observation grid, was calculated. Intravital fluorescence microscopy was performed at days 0 (day of chamber preparation), 3, 6, 9, 12, 15, 18 and 21.
circular glass (6) is fixed to the bone by two micro screws (7). After preparation of the bone chamber (dotted circle), the skin defect is tightened by a purse-string suture (C). The first intravital fluorescence microscopy at day 0 (D) shows some preexisting blood vessels (arrows) of the inner cortical shell at the bottom of the defect (dotted circle). Scale bar represents 0.5 mm.
Histology Histological sections of the calvaria were used to analyse bone healing of the defect. After sacrifice of the animals, the calvaria
768
J.H. Holstein et al. / Injury, Int. J. Care Injured 42 (2011) 765–771
were fixed in 4% phosphate-buffered formalin for 24 h, decalcified in 10% ethylene diamine tetraacetic acid (EDTA) solution for 5 weeks and embedded in paraffin. Sagittal sections of 5-mm thickness were stained according to the trichrome method. Those sections of each specimen which showed the largest diameter of the bone defect, were analysed using a 10 long-distance objective (Olympus BX60 Microscope; Olympus, Tokyo, Japan; Zeiss Axio Cam and Axio Vision 3.1; Carl Zeiss, Oberkochen, Germany). The images were digitised and assessed quantitatively using the computer-assisted off-line analysis system ImageJ (NIH, Bethesda, MD, USA). The total defect area and the bone and fibrous tissue areas were marked manually and calculated by the image-analysis system. The tissue fractions were given in percent of the total defect area. Bone and fibrous tissue fractions of the defects were quantified at days 3, 6, 9 and 15.
blood vessel formation (Fig. 2(C)). In some animals, first sprouts appeared at day 6, whilst all animals showed a dense blood vessel network including numerous sprouts at day 9. Blood vessel density increased until day 15 and slightly decreased during the following observation period until day 21. Blood vessel diameters increased until day 9 and subsequently decreased thereafter (Figs. 2 and 3). In addition to the microscopy of the microvasculature, we analysed the size of the bone defects. The defect area did not change significantly between days 0 and 6, but decreased between days 6 and 18. The defect was no longer detectable by intravital fluorescence microscopy at days 18 and 21. At these late time points, a dense blood vessel network covered the calvarium within the bone chamber, including the former defect area as well as the surrounding bone surface (Figs. 2 and 3). Histology
Immunohistochemistry
All data are given as mean standard error of the mean (SEM). After proving normal distribution and equal variance, comparison between repeated measures of the intravital microscopic analyses was performed by one-way analysis of variance (ANOVA) followed by Tukey test for those outcome variables, which were normally distributed (blood vessel density) or by ANOVA on ranks followed by Tukey test for those outcome variables, which were not normally distributed (defect area and blood vessel diameter). Statistics were performed using the SigmaStat software package (Jandel, San Rafael, CA, USA). A p value <0.05 was considered to indicate significant differences. Results Intravital fluorescence microscopy Using the calvarian defect model, defect size and new blood vessel formation were analysed at days 0, 3, 6, 9, 12, 15, 18 and 21. The first intravital fluorescence microscopy at day 0 showed some pre-existing blood vessels at the surface of the inner cortical shell (Fig. 1(D)). At day 3, no blood-perfused microvessels could be detected (Fig. 2(A)). The pre-existing blood vessels at the surface of the inner cortical shell, observed at day 0, may not have been visible at this time point due to extensive oedema. More probably, however, regression of these pre-existing blood vessels has occurred. This latter view is supported by histology (Fig. 2(B)), demonstrating lack of tissue and blood vessels at the surface of the defect. At day 6, many of the defects still showed a complete lack of
Defect area [mm2]
0.6
0.4
a
0.2
c c
c
c
18
21
0.0 0
3
6
9
12
15
d
Blood vessel density [mm-1]
Statistics
Histology showed a typical pattern of intramembranous bone repair. Accordingly, the calvarian bone defect was filled predominately with fibrous tissue at day 6. Three days later, we observed extensive periosteal bone formation bridging the defect. The new
15
c
d
c
c 10
5
0 0
Blood vessel diameter [µm]
To evaluate cell proliferation and angiogenic activity within the healing calvarian defect, immunohistochemical staining for proliferating cell nuclear antigen (PCNA) and vascular endothelial growth factor (VEGF) was performed. For this purpose, tissue sections were deparaffinised in xylene and rehydrated in a descending, graded series of alcohol. Endogenous peroxidase was blocked by 3% H2O2. Antigen retrieval was achieved by treating specimens with boric acid (17 h; 60 8C). After blocking unspecific binding sites with phosphate-buffered saline (PBS) and goat normal serum, sections were incubated overnight with mouse monoclonal anti-PCNA (1:50 PBS; Dako Cytomation, Glostrup, Denmark) or with mouse monoclonal anti-VEGF (1:100 PBS; Abcam, Cambridge, UK) antibodies. Peroxidase-conjugated goat anti-mouse antibodies (1:200; Amersham Biosciences, Buckinghamshire, UK) were used as secondary antibodies. Diaminobenzidine (DAB) (Dako Cytomation) served as the chromogen and Mayer’s hemalum as the counterstain.
3
6
9
12
c
b
15
18
21
18
21
30
b
20
10
0 0
3
6
9
12
15
Day Fig. 3. Analyses of defect area, blood vessel density, and blood vessel diameter by intravital fluorescence microscopy at days 0, 3, 6, 9, 12, 15, 18, and 21. All data are given as means standard error of the mean (SEM). ap < 0.05 versus day 0, bp < 0.05 versus days 0 and 3, cp < 0.05 versus days 0, 3, and 6, and dp < 0.05 versus days 0, 3, 6, and 9.
J.H. Holstein et al. / Injury, Int. J. Care Injured 42 (2011) 765–771
Tissue fraction
100%
769
Immunohistochemistry Immunohistochemistry revealed positive staining for VEGF in the neo-periosteum, the new woven bone and the fibrous tissue fraction within the healing defect at days 9 and 15. The most intensive staining for VEGF was found in cells of the neoperiosteum at day 9 (Fig. 5). Positive staining for the cell proliferation marker PCNA was found predominately in osteoblasts at day 9 (Fig. 6).
75%
50%
25%
0% 3
6
9
15
Day Fig. 4. Histomorphometric analysis of the different tissue fractions within the bone defect including fibrous tissue (grey fraction of the columns) and woven bone (black fractions of the columns). The remaining defect area without new tissue formation is represented by the white fraction of the columns.
woven bone was covered by a fibrous tissue membrane demonstrating characteristics of a periosteum. Of interest, numerous blood vessels were visible within this neo-periosteum. At day 9, some remaining fibrous tissue was detectable in the centre of the defect, whilst at day 15 all fibrous tissue within the defect was replaced by bone undergoing remodelling. Thus, the histomorphometric analysis of the healing defect revealed a peek fibrous tissue formation at day 6, whereas the bone fraction increased continuously during the whole observation period (Figs. 2 and 4).
Discussion Herein, we studied the process of angiogenesis during bone healing of a calvarian defect over an observation period of 3 weeks. A duration of 3 weeks was chosen as the study period because pilot experiments have shown a steady state of blood vessel density and diameter as well as a closure of the defect after day 18. So far, calvarian bone chambers in mice have been used for the implantation of cytokine-loaded gels or scaffolds.4,11 In the present study, we analysed the physiological process of angiogenesis in a non-grafted bone defect. In addition, we assessed healing of the bone defect by histomorphometry and immunohistochemistry at several time points to achieve further information on the type of bone healing, the composition of the different tissue fractions within the healing defect and the angiogenic and proliferative activity during bone repair. This information is not provided by previous studies using intravital microscopy in calvarian bone chambers.4,11
Fig. 5. Immunohistochemical stainings of the angiogenic growth factor VEGF. Figures A, C, and E provide an overview of the total defect, whilst figures B, D, and F show the surface area of the healing defect with higher magnification. At days 9 and 15 numerous VEGF-positive cells are found in the neo-periosteum, the new woven bone, and the fibrous tissue fraction within the healing defect. The most intense staining for VEGF is found in cells of the neo-periosteum at day 9 (arrows). Scale bars represent 250 mm (A, C, and E) and 75 mm (B, D, and F).
770
J.H. Holstein et al. / Injury, Int. J. Care Injured 42 (2011) 765–771
Fig. 6. Immunohistochemical stainings of the cell proliferation marker PCNA. Figures A, C, and E provide an overview of the total defect, whilst figures B, D, and F show the surface area of the healing defect with higher magnification. Positive staining for PCNA was found predominately in osteoblasts at day 9 (arrows). Scale bars represent 250 mm (A, C, and E) and 75 mm (B, D, and F).
In previous studies, Dellian et al. and Klenke et al. created a fullthickness defect of the calvaria exposing the meninges.4,11 Accordingly, the new blood vessels, which were growing into the bone defect, derived from the pia mater. By contrast, in the present model, angiogenesis is not affected by the meningeal vascularisation because we avoided the penetration of the inner cortical shell. Thus, the present model may be more representative for the study of defect healing in flat bones apart from the calvarium. In the previous studies, cyanoacrylate glue and a methylmethacrylate polymer were used to fix the cover glass of the chamber.4,11 Because we feel that these fixation materials might interfere with bone repair, we modified the technique and used stainless steel screws. In 2005, Hansen-Algenstaedt et al. developed a highly sophisticated bone-chamber model that allows intravital microscopy of the microcirculation during bone repair in the mouse femur.7 In addition, the authors evaluated the formation of calcified tissue by using the fluorescence marker, oxytetracycline. One major limitation of this study, however, is the lack of histological analyses. In accordance, no detailed information is provided on the type of bone repair (intramembranous versus endochondral ossification) and on the type of tissue, hosting the newly formed blood vessels assessed by intravital microscopy. The histological analyses of the present study showed a typical pattern of intramembranous bone healing. PCNA staining showed high cell proliferative action, particularly of osteoblasts. In addition, we found a neo-periosteum, which contained a high number of blood vessels. Of interest, numerous cells of the neoperiosteum showed positive staining for VEGF, indicating a high angiogenic activity and representing most probably the cause for the high number of blood vessels. As the neo-periosteum covers
the surface of the defect, we conclude that these are the blood vessels visible by intravital microscopy. This view is supported by the fact that first blood vessels are detectable by intravital microscopy at the same time the formation of the neo-periosteum is visible by histology. This poses the question of whether the microscopy of periosteal but not intraosseous blood vessels represents a drawback of the herein-introduced model. As early as 1974, Rhinelander showed that blood vessels of the external callus originate from the periosteal circulation.17 Today, it is well accepted that periosteal blood supply is an important prerequisite determining the outcome of bone repair.21 Therefore, we feel that the evaluation of the microcirculation in the periosteum is rather an interesting aspect than a limitation of the present calvarian window model. Quantitative analysis of blood vessel parameters by intravital fluorescence microscopy revealed a continuous increase of blood vessel density until day 15, indicating the formation of a new vascular network covering the bone defect. Of interest, new blood vessels were also detectable at the bone surface surrounding the defect, whilst histological analyses showed periosteal bone formation at the rim of the defect. These results demonstrate that the creation of a calvarian defect leads to an angiogenic and osteogenic response not only within the defect, but also in the neighbourhood of the injury. Blood vessel density and diameter decreased after having reached a peak at days 15 and 9, respectively. These results correspond with data on angiogenesis during wound healing and indicate the maturation of the newly formed blood vessels during the later healing phase.14,18 The herein-introduced model bears some limitations. It has to be considered that the clinician encounters, in the majority of cases, fractures and bone defects of long bones rather than those of flat bones such as the calvarium. A general limitation that has to be
J.H. Holstein et al. / Injury, Int. J. Care Injured 42 (2011) 765–771
addressed in all chamber models is the influence of the cover glass on the healing process of the bone. In accordance, it cannot be excluded that mechanical, osteoconductive or pro-inflammatory properties of the cover glass might indirectly affect also angiogenesis during bone repair.10 A third aspect that has to be discussed is the character of the bone defect created in the present model. We did not penetrate the inner tabula of the calvarium. Therefore, it might be argued that a monocortical injury does not represent a typical situation seen in patients suffering from bone defects. Although we are aware of this potential limitation, our position of not to perforate the inner tabula of the calvaria was to avoid an injury of the meninges and to exclude an affection of bone repair and angiogenesis by the meningeal vasculature. Conclusions In conclusion, we could establish a calvarian defect model that allows the repetitive analysis of periosteal angiogenesis during intramembranous bone repair in mice. In the light of the great availability of specific antibodies, gene-targeted animals and knockout strains in mice, the present model represents a powerful tool to study cellular and molecular aspects of angiogenesis during bone defect healing. Conflict of interest All authors have no conflicts of interest. Acknowledgements We greatly appreciate Janine Becker for excellent technical assistance. The study was financed by the Institute for Clinical and Experimental Surgery, University of Saarland. References 1. Albrektsson T, Albrektsson B. Microcirculation in grafted bone. A chamber technique for vital microscopy of rabbit bone transplants. Acta Orthop Scand 1978;49:1–7.
771
2. Amon M, Menger MD, Vollmar B. Heme oxygenase and nitric oxide synthase mediate cooling-associated protection against tnf-alpha-induced microcirculatory dysfunction and apoptotic cell death. FASEB J 2003;17:175–85. 3. Branemark PI. Vital microscopy of bone marrow in rabbit. Scand J Clin Lab Invest 1959;11:1–82. 4. Dellian M, Witwer BP, Salehi HA, et al. Quantitation and physiological characterization of angiogenic vessels in mice: effect of basic fibroblast growth factor, vascular endothelial growth factor/vascular permeability factor, and host microenvironment. Am J Pathol 1996;149:59–71. 5. Geiger F, Bertram H, Berger I, et al. Vascular endothelial growth factor geneactivated matrix (vegf165-gam) enhances osteogenesis and angiogenesis in large segmental bone defects. J Bone Miner Res 2005;20:2028–35. 6. Groner W, Winkelman JW, Harris AG, et al. Orthogonal polarization spectral imaging: a new method for study of the microcirculation. Nat Med 1999; 5:1209–12. 7. Hansen-Algenstaedt N, Schaefer C, Wolfram L, et al. Femur window—a new approach to microcirculation of living bone in situ. J Orthop Res 2005;23: 1073–82. 8. Hausman MR, Schaffler MB, Majeska RJ. Prevention of fracture healing in rats by an inhibitor of angiogenesis. Bone 2001;29:560–4. 9. Holstein JH, Klein M, Garcia P, et al. Rapamycin affects early fracture healing in mice. Br J Pharmacol 2008;154:1055–62. 10. Karlsson KH. Bone implants—a challenge to materials science. Ann Chir Gynaecol 1999;88:226–35. 11. Klenke FM, Liu Y, Yuan H, et al. Impact of pore size on the vascularization and osseointegration of ceramic bone substitutes in vivo. J Biomed Mater Res A 2008;85:777–86. 12. Kowalski MJ, Schemitsch EH, Kregor PJ, et al. Effect of periosteal stripping on cortical bone perfusion: a laser Doppler study in sheep. Calcif Tissue Int 1996;59:24–6. 13. Laschke MW, Haufel JM, Scheuer C, Menger MD. Angiogenic and inflammatory host response to surgical meshes of different mesh architecture and polymer composition. J Biomed Mater Res B Appl Biomater 2009;91:497–507. 14. Lindenblatt N, Calcagni M, Contaldo C, et al. A new model for studying the revascularization of skin grafts in vivo: the role of angiogenesis. Plast Reconstr Surg 2008;122:1669–80. 15. McCuskey RS, McClugage SG, Younker WJ. Microscopy of living bone marrow in situ. Blood 1971;38:87–95. 16. Menger MD, Lehr HA. Scope and perspectives of intravital microscopy—bridge over from in vitro to in vivo. Immunol Today 1993;14:519–22. 17. Rhinelander FW. Tibial blood supply in relation to fracture healing. Clin Orthop Relat Res 1974;34–81. 18. Sorg H, Krueger C, Vollmar B. Intravital insights in skin wound healing using the mouse dorsal skin fold chamber. J Anat 2007;211:810–8. 19. Winet H. A horizontal intravital microscope-plus-bone chamber system for observing bone microcirculation. Microvasc Res 1989;37:105–14. 20. Yoshiura T, Wu O, Sorensen AG. Advanced MR techniques: diffusion MR imaging, perfusion MR imaging, and spectroscopy. Neuroimaging Clin N Am 1999;9:439–53. 21. Zhang X, Awad HA, O’Keefe RJ, et al. A perspective: engineering periosteum for structural bone graft healing. Clin Orthop Relat Res 2008;466:1777–87.
Contents lists available at ScienceDirect
Injury journal homepage: www.elsevier.com/locate/injury
Intravital microscopic studies of angiogenesis during bone defect healing in mice calvaria J.H. Holstein a,b,c,*, S.C. Becker a,b, M. Fiedler a,b, P. Garcia a,b,c, T. Histing a,b,c, M. Klein a,b, M.W. Laschke b,c, M. Corsten b, T. Pohlemann a,c, M.D. Menger b,c a b c
Department of Trauma, Hand & Reconstructive Surgery, University of Saarland, Homburg/Saar, Germany Institute for Clinical & Experimental Surgery, University of Saarland, Homburg/Saar, Germany Collaborative Research Center AO Foundation, University of Saarland, Homburg/Saar, Germany
A R T I C L E I N F O
A B S T R A C T
Article history: Accepted 10 November 2010
Purpose: Due to the great availability of specific antibodies, gene-targeted animals and knockout strains, mouse models came into the focus of musculoskeletal research. Herein, we introduce a calvarian defect model in mice that allows the repetitive analysis of blood vessel formation during bone repair by intravital microscopy. Methods: The right parietal calvaria of 20 adult CD-1 mice were exposed by skin excision. Under continuous irrigation, a circular defect (Ø0.75 mm) was drilled into the calvarium without penetrating the inner cortical shell. A circular glass (Ø12 mm; thickness 0.15 mm) was fixed by two microscrews (M1; length 2 mm) to cover the bone defect. Angiogenesis was analysed by intravital microscopy at days 0, 3, 6, 9, 12, 15, 18 and 21. In addition, bone repair was evaluated by histomorphometry at days 3, 6, 9 and 15. Immunohistochemical stainings for the angiogenic growth factor vascular endothelial growth factor (VEGF) and the cell proliferation marker proliferating cell nuclear antigen (PCNA) were performed to assess angiogenic and proliferative activity during healing of the calvarian defect. Results: Histomorphometry showed a typical pattern of intramembranous bone repair. Osseous bridging of the defect was observed at day 9. This was associated with the formation of a neo-periosteum, which covered the new woven bone and contained a dense network of newly formed blood vessels. At day 9, particularly cells of the neo-periosteum showed intense staining for VEGF, whilst PCNA-positive staining was found mainly in osteoblasts. At day 15, the major fraction of fibrous tissue was replaced by bone undergoing extensive remodelling. Intravital microscopy revealed an increase of vascular density between days 3 and 15. Blood vessel diameters showed an increase between days 3 and 9 and a subsequent decrease between days 9 and 21. Conclusions: The present calvarian defect model provides a powerful tool to evaluate the process of angiogenesis during intramembranous bone repair in mice. ß 2010 Elsevier Ltd. All rights reserved.
Keywords: Bone repair Angiogenesis Calvarium Intravital microscopy Mice
Angiogenesis has been identified as one key factor determining the outcome of bone repair. Accordingly, inhibition of angiogenic growth factors leads to severe alteration of bone repair, whilst their administration is capable of stimulating bone formation.5,8,9 The investigation of molecular and cellular mechanisms promoting the interaction between blood vessel formation and bone regeneration represents an important topic of current musculoskeletal research. Different imaging techniques, including laser Doppler flowmetry, magnetic resonance imaging (MRI), and orthogonal polarisation imaging, have been used to analyse the role of
* Corresponding author at: Department of Trauma, Hand and Reconstructive Surgery, University of Saarland, D-66421 Homburg/Saar, Germany. Tel.: +49 6841 16 31501; fax: +49 6841 16 31503. E-mail address: [email protected] (J.H. Holstein). 0020–1383/$ – see front matter ß 2010 Elsevier Ltd. All rights reserved. doi:10.1016/j.injury.2010.11.020
angiogenesis during bone repair.6,12,20 Besides, immunohistochemistry provides an additional tool to quantify blood vessel density within regenerating bone tissue. In contrast to these imaging and histological techniques, intravital microscopy bears the great advantage to allow repeated real-time analyses of the microcirculation and new blood vessel formation over a prolonged period of time.16 For this reason, several bone-chamber models have been introduced to assess angiogenesis during bone healing. As early as 1959, Branemark developed a hollow-screw model for intravital microscopy in bone.3 This original bone-chamber model, which later was modified by McCuskey et al., Albrektsson and Albrektsson, Winet and others, allows a long-term observation of the microcirculation in newly formed bone up to 1 year.15,1,19 So far, the hollow-screw model has not been applied in animals smaller than rabbits. In contrast to rabbits, however, small rodents and, especially, mice bear the advantage of a broad spectrum of
766
J.H. Holstein et al. / Injury, Int. J. Care Injured 42 (2011) 765–771
specific antibodies, gene-targeted animals and knockout strains available to study molecular and cellular aspects of angiogenesis during bone regeneration. Accordingly, there is an essential need to apply the powerful tool of intravital microscopy to the analysis of angiogenesis and bone repair in mice. Herein, we introduce a calvarian defect model, which allows intravital microscopic studies of the microcirculation and new blood vessel formation during bone defect healing in mice. The study was approved by the local governmental animal care committee and was conducted in accordance with the German legislation on protection of animals and the National Institutes of Health (NIH) Guidelines for the Care and Use of Laboratory Animals. Methods Animals For the present study, we used adult female CD-1 mice with a mean age of 20 weeks and a body weight of 30–40 g. During the experiments, the mice had free access to tap water and standard pellet food (Altromin, Lage, Germany). Preparation of the calvarian defect model For the preparation of the calvarian defect model, the mice were anaesthetised by intraperitoneal injection of ketamine (75 mg kg body weight 1; Pharmacia GmbH, Erlangen, Germany) and xylazine 2% (25 mg kg body weight 1; Bayer, Leverkusen, Germany). After shaving of the scalp, the mice were fixed on a custom-made support. A circular excision (Ø10 mm) of the scalp and the underlying periosteum was performed to expose the frontal and both parietal aspects of the calvarium. Using a microtrephine, a circular defect (Ø0.75 mm) was drilled under continuous irrigation into the right parietal calvarium, proximately to the sagittal and coronal sutures. Penetration of the inner cortical shell was avoided to prevent an injury of the meninges adhering underneath the calvarium. To cover the defect, a circular glass (12 mm; thickness 0.15 mm) was fixed to the bone by two microscrews (M1.0; length 2 mm). These screws were inserted into the right and left parietal bone at a distance of 3 mm to the sagittal suture. After preparation of the bone chamber, the circular skin defect was tightened by a purse-string suture (Fig. 1). Intravital fluorescence microscopy For intravital microscopy, the mice were anaesthetised and fixed on a custom-made support as reported above. Intravital fluorescence microscopy was performed after retrobulbar IV injection of the plasma tracer fluorescein isothiocyanate (FITC)-labelled dextran 150 000 (5%, 0.1 ml). Epi-illumination was achieved with a 100-W mercury lamp using a fluorescence filter for FITC (Zeiss Axiotech Microscope; Zeiss, Oberkochen, Germany). The microscopic images were recorded using a charge-coupled device (CCD) camera (FK 6990; Pieper, Schwerte, Germany), displayed on a 14-in. video screen (KV-14C T1E; Sony, Tokyo, Japan) and stored on video tapes for off-line analysis. The images were assessed quantitatively using the computer-assisted off-line analysis system CapImage (Zeintl, Heidelberg, Germany).2,13 Defect size and angiogenesis were evaluated using a 10 long-distance objective. Microvessel density and blood vessel diameter were used as parameters of angiogenesis. The defect area was marked manually and calculated by the image analysis system in millimetres.2 Blood vessel density was calculated as the length of red blood cell-perfused blood vessels per observation area, and is given as per millimetre. The observation area was defined as the bone defect area at day 0. Blood vessel diameter was
Fig. 1. Preparation of the calvarian defect model (A and B). A circular defect (1) is drilled into the right parietal calvarium (2) proximately to the sagittal and coronal sutures (3 and 4). Penetration of the inner cortical shell is avoided to prevent an injury of the meninges adhering underneath the calvarium (5). To cover the defect a
J.H. Holstein et al. / Injury, Int. J. Care Injured 42 (2011) 765–771
767
Fig. 2. Intravital fluorescence microscopy of the bone defect (dotted circle) after injection of the plasma tracer fluorescein isothiocyanate (FITC)-labelled dextran 150 000 at days 3, 6, 9, and 15 (A, C, E, and G). At days 3 and 6 most of the defects show complete lack of blood vessel perfusion (A and C). In contrast, at days 9 and 15 a dense network of newly formed blood vessels including numerous blood vessel sprouts is observed (E and G). This network of blood vessels covers not only the defect but also the entire calvarium within the bone chamber including the surrounding bone surface. The defect area does not change significantly until day 6 (A and C), but is found decreased at day 9 (E) and no longer detectably by intravital microscopy at day 15 (G). Histology shows a typical pattern of intramembranous bone repair (B, D, F, and H). The defect (bd) of the calvarian bone (cb) is filled with fibrous tissue (ft) at day 6 (C). At day 9, we found extensive periosteal bone formation bridging the defect (F). The new woven bone (wb) is covered by a fibrous tissue membrane demonstrating the characteristics of periosteum (p). Of interest, numerous blood vessels are visible within this neo-periosteum. At day 9, some remaining fibrous tissue is detectable in the centre of the defect (F), whilst at day 15 all fibrous tissue within the defect is replaced by bone undergoing remodelling (H). Scale bars represent 0.5 mm (A, C, E, and G) and 0.25 mm (B, D, F, and H).
measured perpendicularly to the vessel path. For this purpose, the mean blood vessel diameter of 15 characteristic blood vessels, which
were selected randomly using an observation grid, was calculated. Intravital fluorescence microscopy was performed at days 0 (day of chamber preparation), 3, 6, 9, 12, 15, 18 and 21.
circular glass (6) is fixed to the bone by two micro screws (7). After preparation of the bone chamber (dotted circle), the skin defect is tightened by a purse-string suture (C). The first intravital fluorescence microscopy at day 0 (D) shows some preexisting blood vessels (arrows) of the inner cortical shell at the bottom of the defect (dotted circle). Scale bar represents 0.5 mm.
Histology Histological sections of the calvaria were used to analyse bone healing of the defect. After sacrifice of the animals, the calvaria
768
J.H. Holstein et al. / Injury, Int. J. Care Injured 42 (2011) 765–771
were fixed in 4% phosphate-buffered formalin for 24 h, decalcified in 10% ethylene diamine tetraacetic acid (EDTA) solution for 5 weeks and embedded in paraffin. Sagittal sections of 5-mm thickness were stained according to the trichrome method. Those sections of each specimen which showed the largest diameter of the bone defect, were analysed using a 10 long-distance objective (Olympus BX60 Microscope; Olympus, Tokyo, Japan; Zeiss Axio Cam and Axio Vision 3.1; Carl Zeiss, Oberkochen, Germany). The images were digitised and assessed quantitatively using the computer-assisted off-line analysis system ImageJ (NIH, Bethesda, MD, USA). The total defect area and the bone and fibrous tissue areas were marked manually and calculated by the image-analysis system. The tissue fractions were given in percent of the total defect area. Bone and fibrous tissue fractions of the defects were quantified at days 3, 6, 9 and 15.
blood vessel formation (Fig. 2(C)). In some animals, first sprouts appeared at day 6, whilst all animals showed a dense blood vessel network including numerous sprouts at day 9. Blood vessel density increased until day 15 and slightly decreased during the following observation period until day 21. Blood vessel diameters increased until day 9 and subsequently decreased thereafter (Figs. 2 and 3). In addition to the microscopy of the microvasculature, we analysed the size of the bone defects. The defect area did not change significantly between days 0 and 6, but decreased between days 6 and 18. The defect was no longer detectable by intravital fluorescence microscopy at days 18 and 21. At these late time points, a dense blood vessel network covered the calvarium within the bone chamber, including the former defect area as well as the surrounding bone surface (Figs. 2 and 3). Histology
Immunohistochemistry
All data are given as mean standard error of the mean (SEM). After proving normal distribution and equal variance, comparison between repeated measures of the intravital microscopic analyses was performed by one-way analysis of variance (ANOVA) followed by Tukey test for those outcome variables, which were normally distributed (blood vessel density) or by ANOVA on ranks followed by Tukey test for those outcome variables, which were not normally distributed (defect area and blood vessel diameter). Statistics were performed using the SigmaStat software package (Jandel, San Rafael, CA, USA). A p value <0.05 was considered to indicate significant differences. Results Intravital fluorescence microscopy Using the calvarian defect model, defect size and new blood vessel formation were analysed at days 0, 3, 6, 9, 12, 15, 18 and 21. The first intravital fluorescence microscopy at day 0 showed some pre-existing blood vessels at the surface of the inner cortical shell (Fig. 1(D)). At day 3, no blood-perfused microvessels could be detected (Fig. 2(A)). The pre-existing blood vessels at the surface of the inner cortical shell, observed at day 0, may not have been visible at this time point due to extensive oedema. More probably, however, regression of these pre-existing blood vessels has occurred. This latter view is supported by histology (Fig. 2(B)), demonstrating lack of tissue and blood vessels at the surface of the defect. At day 6, many of the defects still showed a complete lack of
Defect area [mm2]
0.6
0.4
a
0.2
c c
c
c
18
21
0.0 0
3
6
9
12
15
d
Blood vessel density [mm-1]
Statistics
Histology showed a typical pattern of intramembranous bone repair. Accordingly, the calvarian bone defect was filled predominately with fibrous tissue at day 6. Three days later, we observed extensive periosteal bone formation bridging the defect. The new
15
c
d
c
c 10
5
0 0
Blood vessel diameter [µm]
To evaluate cell proliferation and angiogenic activity within the healing calvarian defect, immunohistochemical staining for proliferating cell nuclear antigen (PCNA) and vascular endothelial growth factor (VEGF) was performed. For this purpose, tissue sections were deparaffinised in xylene and rehydrated in a descending, graded series of alcohol. Endogenous peroxidase was blocked by 3% H2O2. Antigen retrieval was achieved by treating specimens with boric acid (17 h; 60 8C). After blocking unspecific binding sites with phosphate-buffered saline (PBS) and goat normal serum, sections were incubated overnight with mouse monoclonal anti-PCNA (1:50 PBS; Dako Cytomation, Glostrup, Denmark) or with mouse monoclonal anti-VEGF (1:100 PBS; Abcam, Cambridge, UK) antibodies. Peroxidase-conjugated goat anti-mouse antibodies (1:200; Amersham Biosciences, Buckinghamshire, UK) were used as secondary antibodies. Diaminobenzidine (DAB) (Dako Cytomation) served as the chromogen and Mayer’s hemalum as the counterstain.
3
6
9
12
c
b
15
18
21
18
21
30
b
20
10
0 0
3
6
9
12
15
Day Fig. 3. Analyses of defect area, blood vessel density, and blood vessel diameter by intravital fluorescence microscopy at days 0, 3, 6, 9, 12, 15, 18, and 21. All data are given as means standard error of the mean (SEM). ap < 0.05 versus day 0, bp < 0.05 versus days 0 and 3, cp < 0.05 versus days 0, 3, and 6, and dp < 0.05 versus days 0, 3, 6, and 9.
J.H. Holstein et al. / Injury, Int. J. Care Injured 42 (2011) 765–771
Tissue fraction
100%
769
Immunohistochemistry Immunohistochemistry revealed positive staining for VEGF in the neo-periosteum, the new woven bone and the fibrous tissue fraction within the healing defect at days 9 and 15. The most intensive staining for VEGF was found in cells of the neoperiosteum at day 9 (Fig. 5). Positive staining for the cell proliferation marker PCNA was found predominately in osteoblasts at day 9 (Fig. 6).
75%
50%
25%
0% 3
6
9
15
Day Fig. 4. Histomorphometric analysis of the different tissue fractions within the bone defect including fibrous tissue (grey fraction of the columns) and woven bone (black fractions of the columns). The remaining defect area without new tissue formation is represented by the white fraction of the columns.
woven bone was covered by a fibrous tissue membrane demonstrating characteristics of a periosteum. Of interest, numerous blood vessels were visible within this neo-periosteum. At day 9, some remaining fibrous tissue was detectable in the centre of the defect, whilst at day 15 all fibrous tissue within the defect was replaced by bone undergoing remodelling. Thus, the histomorphometric analysis of the healing defect revealed a peek fibrous tissue formation at day 6, whereas the bone fraction increased continuously during the whole observation period (Figs. 2 and 4).
Discussion Herein, we studied the process of angiogenesis during bone healing of a calvarian defect over an observation period of 3 weeks. A duration of 3 weeks was chosen as the study period because pilot experiments have shown a steady state of blood vessel density and diameter as well as a closure of the defect after day 18. So far, calvarian bone chambers in mice have been used for the implantation of cytokine-loaded gels or scaffolds.4,11 In the present study, we analysed the physiological process of angiogenesis in a non-grafted bone defect. In addition, we assessed healing of the bone defect by histomorphometry and immunohistochemistry at several time points to achieve further information on the type of bone healing, the composition of the different tissue fractions within the healing defect and the angiogenic and proliferative activity during bone repair. This information is not provided by previous studies using intravital microscopy in calvarian bone chambers.4,11
Fig. 5. Immunohistochemical stainings of the angiogenic growth factor VEGF. Figures A, C, and E provide an overview of the total defect, whilst figures B, D, and F show the surface area of the healing defect with higher magnification. At days 9 and 15 numerous VEGF-positive cells are found in the neo-periosteum, the new woven bone, and the fibrous tissue fraction within the healing defect. The most intense staining for VEGF is found in cells of the neo-periosteum at day 9 (arrows). Scale bars represent 250 mm (A, C, and E) and 75 mm (B, D, and F).
770
J.H. Holstein et al. / Injury, Int. J. Care Injured 42 (2011) 765–771
Fig. 6. Immunohistochemical stainings of the cell proliferation marker PCNA. Figures A, C, and E provide an overview of the total defect, whilst figures B, D, and F show the surface area of the healing defect with higher magnification. Positive staining for PCNA was found predominately in osteoblasts at day 9 (arrows). Scale bars represent 250 mm (A, C, and E) and 75 mm (B, D, and F).
In previous studies, Dellian et al. and Klenke et al. created a fullthickness defect of the calvaria exposing the meninges.4,11 Accordingly, the new blood vessels, which were growing into the bone defect, derived from the pia mater. By contrast, in the present model, angiogenesis is not affected by the meningeal vascularisation because we avoided the penetration of the inner cortical shell. Thus, the present model may be more representative for the study of defect healing in flat bones apart from the calvarium. In the previous studies, cyanoacrylate glue and a methylmethacrylate polymer were used to fix the cover glass of the chamber.4,11 Because we feel that these fixation materials might interfere with bone repair, we modified the technique and used stainless steel screws. In 2005, Hansen-Algenstaedt et al. developed a highly sophisticated bone-chamber model that allows intravital microscopy of the microcirculation during bone repair in the mouse femur.7 In addition, the authors evaluated the formation of calcified tissue by using the fluorescence marker, oxytetracycline. One major limitation of this study, however, is the lack of histological analyses. In accordance, no detailed information is provided on the type of bone repair (intramembranous versus endochondral ossification) and on the type of tissue, hosting the newly formed blood vessels assessed by intravital microscopy. The histological analyses of the present study showed a typical pattern of intramembranous bone healing. PCNA staining showed high cell proliferative action, particularly of osteoblasts. In addition, we found a neo-periosteum, which contained a high number of blood vessels. Of interest, numerous cells of the neoperiosteum showed positive staining for VEGF, indicating a high angiogenic activity and representing most probably the cause for the high number of blood vessels. As the neo-periosteum covers
the surface of the defect, we conclude that these are the blood vessels visible by intravital microscopy. This view is supported by the fact that first blood vessels are detectable by intravital microscopy at the same time the formation of the neo-periosteum is visible by histology. This poses the question of whether the microscopy of periosteal but not intraosseous blood vessels represents a drawback of the herein-introduced model. As early as 1974, Rhinelander showed that blood vessels of the external callus originate from the periosteal circulation.17 Today, it is well accepted that periosteal blood supply is an important prerequisite determining the outcome of bone repair.21 Therefore, we feel that the evaluation of the microcirculation in the periosteum is rather an interesting aspect than a limitation of the present calvarian window model. Quantitative analysis of blood vessel parameters by intravital fluorescence microscopy revealed a continuous increase of blood vessel density until day 15, indicating the formation of a new vascular network covering the bone defect. Of interest, new blood vessels were also detectable at the bone surface surrounding the defect, whilst histological analyses showed periosteal bone formation at the rim of the defect. These results demonstrate that the creation of a calvarian defect leads to an angiogenic and osteogenic response not only within the defect, but also in the neighbourhood of the injury. Blood vessel density and diameter decreased after having reached a peak at days 15 and 9, respectively. These results correspond with data on angiogenesis during wound healing and indicate the maturation of the newly formed blood vessels during the later healing phase.14,18 The herein-introduced model bears some limitations. It has to be considered that the clinician encounters, in the majority of cases, fractures and bone defects of long bones rather than those of flat bones such as the calvarium. A general limitation that has to be
J.H. Holstein et al. / Injury, Int. J. Care Injured 42 (2011) 765–771
addressed in all chamber models is the influence of the cover glass on the healing process of the bone. In accordance, it cannot be excluded that mechanical, osteoconductive or pro-inflammatory properties of the cover glass might indirectly affect also angiogenesis during bone repair.10 A third aspect that has to be discussed is the character of the bone defect created in the present model. We did not penetrate the inner tabula of the calvarium. Therefore, it might be argued that a monocortical injury does not represent a typical situation seen in patients suffering from bone defects. Although we are aware of this potential limitation, our position of not to perforate the inner tabula of the calvaria was to avoid an injury of the meninges and to exclude an affection of bone repair and angiogenesis by the meningeal vasculature. Conclusions In conclusion, we could establish a calvarian defect model that allows the repetitive analysis of periosteal angiogenesis during intramembranous bone repair in mice. In the light of the great availability of specific antibodies, gene-targeted animals and knockout strains in mice, the present model represents a powerful tool to study cellular and molecular aspects of angiogenesis during bone defect healing. Conflict of interest All authors have no conflicts of interest. Acknowledgements We greatly appreciate Janine Becker for excellent technical assistance. The study was financed by the Institute for Clinical and Experimental Surgery, University of Saarland. References 1. Albrektsson T, Albrektsson B. Microcirculation in grafted bone. A chamber technique for vital microscopy of rabbit bone transplants. Acta Orthop Scand 1978;49:1–7.
771
2. Amon M, Menger MD, Vollmar B. Heme oxygenase and nitric oxide synthase mediate cooling-associated protection against tnf-alpha-induced microcirculatory dysfunction and apoptotic cell death. FASEB J 2003;17:175–85. 3. Branemark PI. Vital microscopy of bone marrow in rabbit. Scand J Clin Lab Invest 1959;11:1–82. 4. Dellian M, Witwer BP, Salehi HA, et al. Quantitation and physiological characterization of angiogenic vessels in mice: effect of basic fibroblast growth factor, vascular endothelial growth factor/vascular permeability factor, and host microenvironment. Am J Pathol 1996;149:59–71. 5. Geiger F, Bertram H, Berger I, et al. Vascular endothelial growth factor geneactivated matrix (vegf165-gam) enhances osteogenesis and angiogenesis in large segmental bone defects. J Bone Miner Res 2005;20:2028–35. 6. Groner W, Winkelman JW, Harris AG, et al. Orthogonal polarization spectral imaging: a new method for study of the microcirculation. Nat Med 1999; 5:1209–12. 7. Hansen-Algenstaedt N, Schaefer C, Wolfram L, et al. Femur window—a new approach to microcirculation of living bone in situ. J Orthop Res 2005;23: 1073–82. 8. Hausman MR, Schaffler MB, Majeska RJ. Prevention of fracture healing in rats by an inhibitor of angiogenesis. Bone 2001;29:560–4. 9. Holstein JH, Klein M, Garcia P, et al. Rapamycin affects early fracture healing in mice. Br J Pharmacol 2008;154:1055–62. 10. Karlsson KH. Bone implants—a challenge to materials science. Ann Chir Gynaecol 1999;88:226–35. 11. Klenke FM, Liu Y, Yuan H, et al. Impact of pore size on the vascularization and osseointegration of ceramic bone substitutes in vivo. J Biomed Mater Res A 2008;85:777–86. 12. Kowalski MJ, Schemitsch EH, Kregor PJ, et al. Effect of periosteal stripping on cortical bone perfusion: a laser Doppler study in sheep. Calcif Tissue Int 1996;59:24–6. 13. Laschke MW, Haufel JM, Scheuer C, Menger MD. Angiogenic and inflammatory host response to surgical meshes of different mesh architecture and polymer composition. J Biomed Mater Res B Appl Biomater 2009;91:497–507. 14. Lindenblatt N, Calcagni M, Contaldo C, et al. A new model for studying the revascularization of skin grafts in vivo: the role of angiogenesis. Plast Reconstr Surg 2008;122:1669–80. 15. McCuskey RS, McClugage SG, Younker WJ. Microscopy of living bone marrow in situ. Blood 1971;38:87–95. 16. Menger MD, Lehr HA. Scope and perspectives of intravital microscopy—bridge over from in vitro to in vivo. Immunol Today 1993;14:519–22. 17. Rhinelander FW. Tibial blood supply in relation to fracture healing. Clin Orthop Relat Res 1974;34–81. 18. Sorg H, Krueger C, Vollmar B. Intravital insights in skin wound healing using the mouse dorsal skin fold chamber. J Anat 2007;211:810–8. 19. Winet H. A horizontal intravital microscope-plus-bone chamber system for observing bone microcirculation. Microvasc Res 1989;37:105–14. 20. Yoshiura T, Wu O, Sorensen AG. Advanced MR techniques: diffusion MR imaging, perfusion MR imaging, and spectroscopy. Neuroimaging Clin N Am 1999;9:439–53. 21. Zhang X, Awad HA, O’Keefe RJ, et al. A perspective: engineering periosteum for structural bone graft healing. Clin Orthop Relat Res 2008;466:1777–87.