The use of autologous mesenchymal bone marrow stem cells absorbed in fibrin clot for the regeneration of injured lower jawbones in rats

Introduction 

Despite advances in oral and maxillofacial surgery, traumatology and orthopedics, full repair of cartilaginous and bone tissues in mammals is still a challenging problem, because large bone defects cannot regenerate in a natural way. The usage of stem cells for such purposes is therefore referred to as “regenerative biology” and “regenerative medicine”. This is a rapidly developing area of research, with high hopes for more effective treatment of bone injuries. New therapeutic approaches with stem cells may be helpful or even replace standard surgical methods when the latter are ineffective [3, 4].

The application of mesenchymal stem cells (MSC) leads to faster regeneration of injured bones and an increase in bone density, when compared to natural restorative processes [6, 7, 20]. 

The products of fibrin degradation lead to more rapid restoration of surgical bone defects in experiments and in clinical practice.  It is noteworthy that fibrin not only stimulates the migration of fibroblasts but also accelerates the synthesis of connective tissue [8-11].

Hence, for acceleration of repair we performed an experimental study of jawbone regeneration in rats treated by introduction of the following substances into the injured areas: (1) autologous platelet-enriched fibrin clot (PEFC); (2) a suspension of autologous mesenchymal stem cells from a bone marrow origin (AMSC) in a culture medium; and (3) PEFC with absorbed AMSC.

The effects of these different methods upon bone and marrow repair were revealed with light microscopy and X-ray densitometry. 

Materials and methods

Male six-month old Wag rats (180–200 g) were used in this study. All manipulations with the animals were carried out under general ether inhalation anesthesia, in a clean operating room, in compliance with international “Regulations for Working with Experimental Animals”. At least 8 rats were studied at each time-point of the observation period.

Damaged bone tissue model in the experiment 

We used an original experimental model of bone damage and repair, as described elsewhere [11]. In brief, the surgery was performed under general ether inhalation anesthesia, under aseptic conditions, after skin preparation with ethyl alcohol. A 1.5–2 cm longitudinal incision was made along the inferior edge of the lower jaw. By means of a blunt technique and a raspatory, the mastication muscle was exfoliated and the lower jawbone surface was exposed at the jaw angle. A round hole (2mm in diameter) was made at the outside jaw angle bone with a dental drill. The defect passed through all layers of the bone but was not connected with the oral cavity.

Animals were divided into the following four groups according to the method used to regenerate the damaged lower jawbones:

1.    In the first group (58 rats), the natural events of bone repair were followed. The hole in the bone was covered with the mastication muscle, and the soft tissues were sutured in layers with vicryl.

2.    The second group (56 rats) represented the regeneration after inserting PEFC into the hole. After densely filling the hole with PEFC, it was covered with the mastication muscle and the skin was sutured with vicryl.

3.    In the third group (58 rats) the course of repair was observed after the introduction of an AMSC suspension in an α-МЕМ medium (100 mcl, 10⁶ cells per 1 ml) into the hole in the lower jaw.

4.    In the fourth group (60 rats) we used PEFC with AMSC. Prepared PEFC was immediately dipped into the AMSC suspension in culture medium (10⁶ cells per 1 ml) for 2 hours, as living cells are attached to any solid substrate.

PEFC preparation 

Several rats of the same breed were decapitated, and 2–7 ml of blood was collected in sterile glass tubes. The blood was centrifuged at 2800 rpm for 12 minutes [8-11]. Then the upper layer (platelet-enriched fibrin clot or fibrin clot with platelets) was placed in sterile Petri dishes and was stored for a few hours at 37°C until use. Immediately before use, PEFC fragments were cut with sterile scissors to the necessary size. 

AMSC preparation

AMSC were separated by flushing out bone marrow from the epiphysis of Wag-male rat femoral bones. Individual cell suspensions were placed into plastic vials (“Nunc”, Denmark). After 48 hours the unattached cells were poured off. Adherent cells were cultivated in an α-МЕМ medium with 10% embryonic calf serum (“Biolot”, Russia) at 37°C in a СО₂ incubator with 5% СО₂ and 100% humidity. The medium was changed every three days. During sub-cultures, the monolayer cultures were dispersed at a density of 1000–5000 cells/cm² (depending on the growth effects of the embryonic serum). Standard solutions of EDTA and trypsin were used for MSC yielding. 

Using the common approaches of light and fluorescent microscopy and standard methods of cytological staining and immunochemistry, we revealed the general characteristics of the cultured marrow cells to be as follows:

• CD90+, CD105+, CD34-, CD45-,
• Plastic-adherent cell populations in vitro,
• Fibroblast-like morphology of cells through the entire period of culture,
• The cells were capable of several passages in culture,
• Form colonies of fibroblast-like cells after sub-culture of low cell density, 
• Ability to differentiate into bone cells in the presence of lineage-specific factors.

However, these physical, morphological, and phenotypic signs are not specific criteria for precise identification of AMSC. The ability of AMSC to differentiate in vitro into bone, fat, and cartilage is the only criteria to determine a prospective population of stem cells. 

Induction of osteogenic differentiation of mesenchymal stem cells in vitro: AMSC possess the ability to differentiate into the cells of bone tissue under reproducible conditions [3-6, 13, 14, 20]. Therefore, the osteogenic differentiation of mesenchymal stems cells is commonly tested in vitro. Osteogenic differentiation is a typical method of developing most AMSC in a culture. To induce osteogenic differentiation, 0.1 µM deoxymetazone, 50 µM ascorbic acid and 10 mM β-glycerophosphate (“Sigma”, US) were applied.

The osteogenic differentiation was defined by two histochemical markers: active alkaline phosphatase and mineralization of the intercellular matrix by calcium ions. Cytochemical detection of the alkaline phosphatase was made with Nitrotetrazolium Blue in the presence of 5-bromo-4–chloro-3-indolyl, a substrate for alkaline phosphatase. Calcium deposits in the intercellular matrix were registered by Alizarin Red staining.

The animals were removed from the experiment at the first, second, third, fourth, and fifth week after the operation. Fragments of the lower jaws with the experimental injuries were fixed in a 4% solution of paraformaldehyde in phosphate buffer (рН 7.4) for at least 24 hours.

The bone fragments of the rats’ lower jaws devoid of skin, muscles, and connective tissue were examined using X-ray densitometry. The software “Tomodent” (Anvisystem, Russia) installed on a radiovisiographic computer device “Heliodont+” (Herona, Germany, 2010) allowed evaluation of bone tissue density in conventional units, as a ratio of retrieved data of the bone density in the damaged area to the measurements for contralateral undamaged bone fragments.

The big lower jaw fragments were then decalcified in “Biodek R” solution (Bio Optica, Milano, Italy) for 24 hours, dehydrated in a gradient of increasing concentrations of ethanol, clarified in xylene and embedded into histoplast perpendicularly aperture in a bone so that the sections passed in parallel to the outer bone surface. Sections 5–7 microns thick were stained with hematoxylin and eosin and studied with an Axioimager M1 light microscope (Carl Zeiss, Germany), at an optical magnification of up to 1200x.

To research the area of structures of red bone marrow on a section of the bottom jawbone, we applied the square test system combined on the computer screen with the image received from a digital camera on a microscope. Using a lens with an increase of 5 times, the final test square area was equal to 16,900 microns (square party was equal to 130 microns).

Statistical analysis of results was performed with MS Excel’s applied statistical program (Microsoft, USA). Both arithmetic means and standard deviations were determined. The differences between the mean values were considered to be significant at p≤0.05, using Student’s coefficient.

Results

Natural course of bone regeneration:

One week after the injury, the hole was partially filled with blood, some fragments of friable connective tissue, and structures of granulation tissue. These events represented the initial stages of tissue repair in the bone defect area, as evidenced by de novo formation of separate bone and cartilage islands among the granulation tissues (Figure 1a, b).

Two weeks after injury the bone defect was completely closed with conjugated islands of new bone tissue. Some cartilage tissue was also present among the newly formed bone structures, especially in the middle of the former hole.

Three weeks after injury the hole was entirely filled by the newly formed bone tissue. Multiple, randomly arranged trabeculae (callus structures) were the only evidence of the surgical intervention. In some cases (3 out of 12), bone marrow cavities were already formed by that time. In most cases, at 4 to 5 weeks, a post-injury callus remained as the only sign that the operation had been performed.

Statistical analysis of densitometric data of bone regeneration in lower jawbones of rats by natural healing has revealed that bone density in the damaged fragment was 12.1%, 11%, 10.5% and 9.3% lower than in the corresponding healthy fragment on the contralateral side, when measured respectively, at week 1, 2, 3, 4, and 5 after surgery (Table 1) (Figure 2a).

Defects Filled with PEFC: 

One week after the injury, the hole was completely filled with conjugated islands of the newly formed bone (Figure 1c, d).

In most cases, in two weeks after the bone injury, the hole was absolutely closed with newly formed bone tissue. 

In three weeks after the bone injury, the hole was absolutely closed with newly formed bone tissue with chaotically arranged trabeculae of the formed callus and cavities with bone marrow. 

In 4 and 5 weeks after the injury, the hole was absolutely closed with newly formed bone tissue, as it was with the natural healing.

After PEFC use the bone density in the defect was 9.5% lower than the undamaged contralateral side in the first week, and 4.9% lower in the second week (Figure 2b) (Table 1).

Table 1. Bone density in defect of the rat's lower jaw in comparison with intact tissues of contralateral undamaged bone fragments (S±σ)

Postoperative periods	Regeneration Process
	Natural healing course	After PRFC application	After AMSC application	After PRFC+AMSC application
1	2	3	4	5
1 Week	0.892±0.053*	0.913±0.017*	0.928±0.044	0.917±0.037*
2 Weeks	0.901±0.035*	0.953±0.021*	0.903±0.046*	0.905±0.057
3 Weeks	0.905±0.02*	0.949±0.036	0.96±0.086	0.927±0.04
4 Weeks	0.912±0.059	0.942±0.048	0.932±0.052	0.922±0.032*
5 Weeks	0.915±0.016*5	0.924±0.063	0.856±0.028*	0.978±0.022
Notes: * –  data, significantly different from the intact bone (р ≤ 0.05), 2, 3, 4, 5 – data, significantly different from each other in this columns (р < 0.05)
Cell Ther Transplant. 2012;3:e.000101.01. doi:10.3205/ctt-2012-en-000101.01-table1

Healing of defects filled with AMSC suspension:

One week after the injury the bone defects were completely filled with blood; typical granulations were found between the blood clot and margins of the defect. Notably, some signs of bone formation were also detectable, presenting as separate islands of new bone and cartilage in the defect area. The tissues within the damaged area were similar to those in controls (normal regeneration) at the same time point. However, it is noteworthy that there were many more blood vessels penetrating the bone hole, as well as in the granulation structures.

After two weeks, the defects in lower jawbone were completely replaced by new bone and cartilage tissue, with a great number of conjugated bone cell islands and a plethora of thin-walled blood vessels. It's notable that formation of red bone marrow (with hematopoietic cells) structures took place by this period (Figure 1e, f). By this time a significant acceleration of repair processes had been observed, thus resulting in rapid development and regeneration of hematopoietic tissue such as bone marrow in the bone callus. 

Within the subsequent period, the bone tissue islands fused together, forming osseous callus structures, along with progressive restoration of red bone marrow.

Statistical differences between bone densities after AMSC treatment and those at the intact contralateral side were found only during the second and fifth weeks: 10.7% and 16.8% less respectively. However, five weeks after surgical injury, X-ray density of the bone defect after AMSC treatment alone was even less than in the natural regenerative process (Figure 2c) (Table 1). 

The terms of occurrence and development of cavities containing red bone marrow in the compared groups are presented in Table 2. The area of such structures after application of AMSC at week 3 did not differ to a statistically significant amount from the intact control. During the healing course with other influences the size of cavities with marrow began to correspond to control only at the 4-week point.

Table 2. The occurrence and development of red bone marrow in the rat's lower jaw

Post- operative periods	Regeneration process
Natural healing course
	Number of animals in group  (n)	Number of animals with restoration of bone marrow  (n(%))	The size of cavities with marrow in a casual section of the bone  (mm2, S±σ)	Number of animals in group  (n)	Number of animals with restoration of bone marrow  (n(%))	The size of cavities with marrow in a casual section of the bone  (mm2, S±σ)
1	2	3	4	5	6	7
1 Week	12	-	-*	12	-	-*
2 Weeks	12	-	-*9	12	-	-*9
3Weeks	12	3(25)	0.225±0.035*	12	4(33.3)	0.232±0.021*
4 Weeks	12	11(91.7)	0.269±0.024	12	8(66.7)	0.268±0.029
5 Weeks	10	10(100)	0.326±0.041	8	8(100)	0.336±0.023
After AMSC application
	Number of animals in group  (n)	Number of animals with restoration of bone marrow  (n(%))	The size of cavities with marrow in a casual section of bone  (mm2,S±σ)	Number of animals in group  (n)	Number of animals with restoration of bone marrow  (n(%))	The size of cavities with marrow in a casual section of the bone  (mm2, S±σ)
1	8	9	10	11	12	13
1 Week	12	-	-*	12	-	-*
2 Weeks	12	8(66.7)	0.203±0.013*4, 7, 13	12	-	-*9
3Weeks	12	12(100)	0.287±0.032	12	5(41.7)	0.226±0.035*
4 Weeks	12	11(91.7)	0.34±0.032	12	10(83.3)	0.316±0.025
5 Weeks	10	10(100)	0.355±0.041	12	12(100)	0.344±0.03
Notes: * –  data, significantly different from the intact bone (0.339±0.04 mm2; р≤0.05), 4, 7, 10, 13 – data, significantly different from each other in this columns (р < 0.05)
Cell Ther Transplant. 2012;3:e.000101.01. doi:10.3205/ctt-2012-en-000101.01-table2

Defects Filled with PEFC and AMSC:

In a week after the surgery and PEFC and AMSC taken together, in most cases, more than 2/3 of the defect of the lower jawbone was filled with newly formed bone tissue. However, the tissue was separated from the “old bone” (the defect edge) by connective tissue with granulations. It is likely that in these cases the bone formation starts and progresses rapidly in the defect center and gradually comes up to the defect edges where granulations still exist. In the areas of the defect that had still not been filled with bone tissue, there were granulations with a large number of blood vessels (Figure 1g, h).

Two weeks after the surgery, in most cases, the defect was completely closed with new bone and cartilage tissues. 

In 3–5 weeks after PEFC with AMSC use, the defect of the lower jawbone was fully closed with the bone tissue. The presence of callus structures was the only difference from the undamaged contralateral jawbone.

After PEFC with AMSC use, statistically significant differences of the bone tissue density from the contralateral fragment of the lower jawbone were found only during the first and fourth weeks – 9.1% and 8.5% lower respectively. Moreover, in this group of animals five weeks after the operation, there was a significant difference in the density of bone tissue – 6.9% higher compared to the natural regenerative process at the same time (Figure 2d) (Table 1).

Discussion

According to published scientific literature, fibrin in tissues reduces the intensity of the inflammatory process and limits the spread of infection [8-11]. That is, the introduction of PEFC in the wound cavity can protect the surrounding tissues from the dissemination of microorganisms, and from excessive exposure to the lysosomal enzymes of phagocytes. By limiting this destruction, the regenerative processes in tissues begin earlier, and there are lower amounts of antigens and detritus, so the wound is cleansed rapidly. 

Plasma and fibrin clot contain many cytokines and high concentrations of growth factors: platelet-derived growth factor (PDGF), transforming growth factor beta (TGF-b), platelet-derived epidermal growth factor, platelet-derived angiogenesis factor, insulin-like growth factor 1 (IGF-1), and platelet-derived factor 4. These releases cause migration and division of all mesenchymal cells (including chondrocytes and mesenchymal stem cells) and epithelial cells, stimulate collagen synthesis and the formation of connective tissues matrix [1, 2, 16].

In addition, the fibrin clot acts as a matrix on which the migration of leukocytes (neutrophils), endotheliocytes, and fibroblasts occurs. Migrating on fibrin, neutrophils more rapidly reach all parts of the wound, even if the wound is covered with layers of pus and detritus. Thus, antigenic substances (microorganisms and detritus) are cleansed from tissues more rapidly. Moreover, when neutrophils migrate on fibrin clot they partially dissolve the fibrin clot with their own enzymes, so as a result even a dense fibrin clot starts to resemble a net. Fibroblasts, located on the fibrin net [1, 2, 16], begin to synthesize collagen, not only from the bottom of the wound, but also from its cavity. Thus the scar tissue forms more rapidly.

Therefore, using PEFC contributes to more rapid regeneration of the damaged fragment of the lower jawbone.

In the course of the natural regenerative process of the damaged lower jawbone, the tissue regeneration starts from the edges of the holes. Pre-existing osteoblasts and stem cells migrate from residual bone tissue [12, 17] and periosteum [5] in the blood clot. As a result, separate isolated foci of osteogenesis may be detected in the blood clot filling the bone defect as early as 1 week after the surgical injury.

After the introduction of the AMSC suspension, in our opinion, a large number of stem cells immediately appear in all parts of the artificial bone hole. In this case, there is no lag period for stem cells to migrate to the damaged fragment and to penetrate directly into the damaged tissues. 

Since the mesenchymal stem cells in our experiments were of bone marrow origin, it is likely that some hematopoietic cells were present among the stem cells. Apparently, the activity of these cell precursors enables rapid and early regeneration of hematopoietic structures, i.e., red bone marrow. It's necessary to note an opportunity of AMSC differentiation into hematopoietic stem cells [15].

According to the literature, we expected an acceleration of the damaged bone repair and, respectively, its higher density in comparison with results of jaw densitometry obtained in the cases of non-modified repair [6, 7, 20]. 

Most likely, the lower tissue density registered at the 2nd and 5th week after surgery may be ascribed to the development of red bone marrow, in accordance with densitometry data. During the first week after injury, active bone regeneration (mineralization) took place, and there was a noticeable increase in the bone tissue density. Thereafter, red bone marrow was formed in the cavities, and, therefore, bone density could be decreased. After AMSC application, the formation of bone marrow seemed then to be more rapid and pronounced. Therefore, bone density could decrease in the damaged fragment and became even lower than the healing without using stem cells. It is necessary to take into consideration the possible lowering of the strength properties of the regenerated bone due to the presence of large cavities filled with bone marrow.  

One week after using PEFC with AMSC, the lower jawbone defect was mostly filled with newly formed bone tissue. It's likely that in these cases the bone formation began in the center of the defect, and not along the edges. After two weeks, the lower jawbone defect was completely filled with new bone tissue. In the following period the artificial hole had disappeared – the only trace of it was the callus structures. 

According to scientific literature data, good results were obtained when the combination of MSC and platelet-enriched plasma were used for acceleration of bone tissue regeneration and the implant osteointegration. Such plasma can maintain the bone tissue growth and act as a matrix for bone growth from MSC [13, 14, 19]. It is based on the fact that cytokines of megakaryocytes influence MSC differentiation. Moreover, the interaction of thrombopoietic structures and MSC stimulates endochondral ossification [18]. MSC with plasma and platelets can be delivered to the areas of regenerating bone and cartilaginous tissues by means of injection [19]. The modification of platelet-enriched plasma is a fibrin gel, glue, and sponge, which can be used very effectively together with AMSC. 

Most likely when PEFC is used together with AMSC, the synthesis of the two optimizes bone defect regeneration. The stem cells in fibrin clot fill the entire defect more or less evenly. These cells do not migrate from the place of introduction (actively or passively, as happens when AMSC is used alone). Cytokines, platelet factors, and especially megakaryocytes in PEFC stimulate the proliferation of AMSC as well as its differentiation in the direction of osteogenesis. As a result, the most rapid and successful regeneration of damaged bone tissue is achieved.

Conclusion

According to the experimental data, the best results in bone regeneration were achieved by applying PEFC with AMSC to the bone defect. After one week, the hole in the lower jawbone was mostly filled with formed bone tissue. It is more likely, in this case, that the combined characteristics of fibrin and stem cells optimize bone defect regeneration. The bones regenerated significantly faster than when PEFC and AMSC were used separately. Evidently, the bone formation starts in the center of the defect rather than from the edges. Stem cells in fibrin clot spread throughout the defect and fill it more or less evenly and completely. As a result, the most rapid and successful regeneration of the bone tissue defect is achieved. The development of red bone marrow in the bone callus occurs much earlier after the use of AMSC during surgery than in the other courses of influence on tissue repair. Formation of cavities with functional bone marrow may cause a decrease in tissue density within the 4th and 5th weeks after injury with AMSC application at the site of damage. 

Acknowledgements

This work was financially supported by the Fundamental Research Program of the Presidium of RAS “Fundamental Science - Medicine” (project № 21.31 “Development of technologies for process control of bone-tissue regeneration with biodegradable polymers application”).

References

1. Anitua E, Sanchez M, Nurden A, et al. New insights into and novel applications for platelet-rich fibrin therapies. Trends in Biotechnology. 2006:24(5):227-234. doi:10.1016/j.tibtech.2006.02.010.

2. Anitua E, Sanchez M, Nurden AT, et al. Autologous fibrin matrices: a potential source of biological mediators that modulate tendon cell activities. Journal of Biomedical Materials Research. Part A. 2006:77(2):285-293. doi:10.1002/jbm.a.30585.

3. Chanda D, Kumar S and Ponnazhagan S. Therapeutic potential of adult bone marrow-derived mesenchymal stem cells in diseases of the skeleton. Journal of Cellular Biochemistry. 2010:111(2):249-257. doi:10.1002/jcb.22701.

4. Clines GA. Prospects for osteoprogenitor stem cells in fracture repair and osteoporosis. Current Opinion in Organ Transplantation. 2010:15(1):73-78. pmid:19935065.

5. Hayashi O, Katsube Y, Hirose M, et al. Comparison of osteogenic ability of rat mesenchymal stem cells from bone marrow, periosteum, and adipose tissue. Calcified Tissue International. 2008:82(3):238-247. doi:10.1007/s00223-008-9112-y.

6. Kallai I, Lenthe van GH, Ruffoni D, et al. Quantitative, structural, and image-based mechanical analysis of nonunion fracture repaired by genetically engineered mesenchymal stem cells. Journal of Biomechanics. 2010:43(12):2315-2320.

7. Kumar S, Wan C, Ramaswamy G, et al. Mesenchymal stem cells expressing osteogenic and angiogenic factors synergistically enhance bone formation in a mouse model of segmental bone defect. Molecular Therapy: the Journal of the American Society of Gene Therapy. 2010:18(5):1026-1034. doi:10.1038/mt.2009.315.

8. Maiborodin IV, Kolesnikov IS, Sheplev BV and Ragimova TM. Application of fibrin and its preparations in stomatology. Stomatologiia (Mosk). 2008:87(6):75-77.

9. Maiborodin IV, Kolesnikov IS, Sheplev BV, et al. Granulomatous inflammation after fibrin preparation use. Morphological letters. 2007:(3-4):116-118.

10. Maiborodin IV, Kolesnikov IS, Sheplev BV, et al. Adjusting gingival tissues morphology after dental implantation with fibrin use. Stomatologiia (Mosk). 2009:88(1):9-13.

11. Maiborodin I, Shevela A, Perrin T, et al. Experimental results of the fibrin clot use to accelerate the regeneration of damaged bone in the rat lower jaw. Surgical Science. 2010:1(1):1-6. doi:10.4236/ss.2010.11001.

12. Neumann K, Dehne T, Endres M, et al. Chondrogenic differentiation capacity of human mesenchymal progenitor cells derived from subchondral cortico-spongious bone. Journal of Orthopaedic Research: Official Publication of the Orthopaedic Research Society. 2008:26(11):1449-1456.

13. Niemeyer P, Fechner K, Milz S, et al. Comparison of mesenchymal stem cells from bone marrow and adipose tissue for bone regeneration in a critical size defect of the sheep tibia and the influence of platelet-rich plasma. Biomaterials. 2010:31(13):3572-3579.

14. Pieri F, Lucarelli E, Corinaldesi G, et al. Effect of mesenchymal stem cells and platelet-rich plasma on the healing of standardized bone defects in the alveolar ridge: a comparative histomorphometric study in minipigs. Journal of Oral and Maxillofacial Surgery: Official Journal of the American Association of Oral and Maxillofacial Surgeons. 2009:67(2):265-272.

15. Ratajczak MZ, Kucia M, Reca R, et al. Stem cell plasticity revisited: CXCR4-positive cells expressing mRNA for early muscle, liver and neural cells 'hide out' in the bone marrow. Leukemia: Official Journal of the Leukemia Society of America, Leukemia Research Fund, U.K. 2004:18(1):29-40. doi:10.1038/sj.leu.2403184.

16. Schwartz-Arad D, Levin L and Aba M. The use of platelet rich plasma (PRP) and platelet rich fibrin (PRP) extracts in dental implantology and oral surgery. Refu’at Ha-peh Veha-shinayim. 2007:24(1):51-55,84.

17. Steenhuis P, Carr KM, Pettway GJ and Ignelzi MA Jr. Osteogenic and adipogenic cell fractions isolated from postnatal mouse calvaria. Cells, Tissues, Organs. 2009:190(3):150-157. doi:10.1159/000187633.

18. Sumiyoshi K, Kubota S, Furuta RA, et al. Thrombopoietic-mesenchymal interaction that may facilitate both endochondral ossification and platelet maturation via CCN2. Journal of Cell Communication and Signaling. 2010:4(1):5-14. doi:10.1007/s12079-009-0067-1.

19. Yoshimi R, Yamada Y, Ito K, et al. Self-assembling peptide nanofiber scaffolds, platelet-rich plasma, and mesenchymal stem cells for injectable bone regeneration with tissue engineering. The Journal of Craniofacial Surgery. 2009:20(5):1523-1530. doi:10.1097/SCS.0b013e3181b09b7e.

20. Zhang ZY, Teoh SH, Chong MS, et al. Neo-vascularization and bone formation mediated by fetal mesenchymal stem cell tissue-engineered bone grafts in critical-size femoral defects. Biomaterials. 2010:31(4):608-620.