Preparation of Deproteinized Human Bone and Its Mixtures with Bio-Glass and Tricalcium Phosphate – Innovative Bioactive Materials for Skeletal Tissue Regeneration

Magdalena Cieslik1, Jacek Nocon2, Jan Rauch3, Tadeusz Cieslik4, Anna ¥losarczyk5, Maria Borczuch-Łączka6 and Aleksander Owczarek7 1Faculty and Institute of Stomatological Materials Science, Medical University of Silesia, Katowice, Bytom, 2Private Dentistry Practice, Oberhausen, 3NZOZ – Specialist Dentistry Clinic, Wadowice, 4Faculty and Clinic of Oral and Maxillofacial Surgery, Medical University of Silesia, Katowice, 5Faculty of Glass Technology and Amorphous Coatings, AGH Krakow University of Science and Technology, Krakow, 6Faculty of Ceramic Technology, AGH Krakow University of Science and Technology, Krakow, 7Division of Statistics, Medical University of Silesia, Katowice, Sosnowiec, 1,3,4,5,6,7Poland 2Germany


Introduction
Repair of the skeletal system is one of the principal research problems in medical science is closely associated with the field of material engineering. The reasons for using bone implants and grafts include injuries, infections, neoplasms and other hard tissue lesions. Bone replacement materials are predominantly used in medical disciplines such as dentistry, dental surgery, maxillofacial surgery and plastic surgery, as well as in orthopedics and traumatology (Barradas et al., 2011;Kao & Scott, 2007;Precheur, 2007).
From a biological, immunological, and legal point of view, autogenous bone grafting still remains a very popular method in reconstruction following skeletal loss (Block, 2002;Giannoudis et al., 2005). Factors considered in the selection of the source of the bone graft include, among others, the ease of surgical access and the volume of bone mass required (Precheur, 2007). The type of autogenous bone used as a graft (cortical bone vs. cancellous bone) should also be considered. For instance, the higher content of morphogenetic proteins (BMPs) in cortical bone means that grafts of this type induce the process of bone growth more effectively than cancellous bone grafts. Nonetheless, skeletal reconstruction with autogenous bone grafts always requires additional surgical manipulations that constitute an 371 tricalcium phosphate (betaTCP, betaCa 3 (PO 4 ) 2 ) is another calcium phosphate that has been used successfully in bone substitution. Its mineralogical analogue is whitlockite (TCP) (Barradas et al., 2011). Similar to HAp, TCP is characterized by high biocompatibility. In comparison to hydroxyapatite materials, it has a higher solubility in vitro and a resultant higher susceptibility to resorption and biodegradation in the environment of the living organism (Bohner, 2010;Henkel et al., 2006). TCP is considered an osteoinductive material, which stimulates the processes of bone reconstruction (Wang et al., 2009). Many studies, mostly dealing with the gradual, controlled resorption rate of calcium phosphate ceramic, resulted in the design of the second polymorphic variant of TCP, namely TCP (Zima et al., 2010). Similar to beta-TCP, it is formed as a result of the non-stoichiometric heating of HAp with a well-defined temperature and defined kinetics of thermal processing. Compared to beta-TCP, TCP has an approximately five-fold higher susceptibility to resorption in living tissues, along with higher biocompatibility (Oonishi et al., 1999). The high degree of osteointegration of TCP-based bioceramic is plausibly the result of the higher solubility of Ca 3 (PO 4 ) 2 than of HAp or TCP. Delivered within the ceramic, calcium and phosphate ions constitute the material for synthesizing the layer composed of non-stoichiometric hydroxyapatite that binds the implant to bone (Kon et al., 1995). Some authors suggest that TCP can be cytotoxic, probably due to pH changes that are induced in vitro (Santos et al., 2002). These suggestions, however, were not confirmed during in vivo studies, and TCP is included in many commercially available bone cements. Since the 1990s, biphasic HAp-TCP ceramic (BCP biphasic calcium phosphate) has also been used in the reconstruction of skeletal defects (Deculsi, 1998, Deculsi et al., 2003Fellah et al., 2008;Schwarz et al., 2007).
Bio-glass and apatite-wollastonite glass-ceramic also play an important role amongst the bioactive materials used in the processes of bone synthesis. The biological activity of glass and glass-derived crystalline materials (glass-ceramic materials) is mostly exploited for the manufacture of surface-active implants or implants that can be resorbed in the human body (Giannoudis et al., 2005). The composition of surface-active materials is selected in order to enable interactions between the physiological environment and certain components of the implant. As a result, living tissue is bound to the implant surface. Resorbable glass is characterized by its high content of chemical elements involved in metabolic processes of the human body. In both cases, living tissues can infiltrate the implant and bind it directly to the bone. These processes result not only from the chemical composition of the implants but also from the specific nature of the contained glass substance. The binding abilities of bioactive glass and glass-ceramic implants to skeletal tissue have been the subject of many studies. These studies confirmed the usefulness of such materials in tissue engineering, where they can be used in cell culture (Ferreira, 2007;Xynos et al., 2000). Their application in reconstructive surgery is reflected by their ability to stimulate and supporting bone reconstruction. In view of their confirmed ability to directly bind to skeletal tissue, they have been used successfully in various stomatological disciplines, including implantology, dental surgery and periodontology. Moreover, modern surface engineering has allowed the coating of metal implants with bioactive glass. Such coatings either protect surfaces of metal alloys against corrosion and wear, or stimulate the processes of bone formation in their surroundings. Additionally, resorbable glass can be used as a drug carrier, providing prolonged release of an active substance. Furthermore, some attempts have been made to combine various materials with each other to manufacture glass-containing bioactive composites, used as the components of binders, among others (Bellucci et al., 2011). These 372 experiments were aimed at obtaining biomaterials with better durability parameters and optimal biological characteristics. Various types of bio-glass and glass-ceramics may differ in terms of their biological properties, depending on the technique for their synthesis (Bellucci et al., 2011;Yuan et al. 2001). Amongst various methods of synthesis, the advantages of the chemical sol-gel method are worth noting. This technique enables material whose biological activity is greater than products manufactured using other processing methods to be obtained (Ravarian et al., 2010). Moreover, the sol-gel method does not require high temperature processing. Additionally, this technique allows the production of material with a strictly defined texture and parameters. Due to the possibility of manufacturing porous forms, bio-glass and glass-derived crystalline materials are morphologically similar to bone; this facilitates their infiltration by bone cells at the implantation site. Additionally, the porous structure of the implant enables the supply of fluids and nutrients required for growth to the newly formed bone, as well as the elimination of metabolites (Yuan et al., 2001). Finally, the resorption rate of the implant material is also determined by the degree of its porosity.
In view of the complexity of the biological environment at the implantation site, one principle problem of biomaterial engineering pertains to issues associated with the biodegradation and bioresorption of implanted biomaterials. Determination of degradation time and rate, and the kinetics of this process in the human body, constitute a significant challenge in the design of new implant materials (including bioceramic) used for the purposes of skeletal tissue substitution (Daculsi et al., 2003;Giannoudis et al., 2005). The controlled degradation rate of the implant along with the associated reconstruction of skeletal tissue should result in the formation of bone tissue resembling its natural structure as closely as possible, both at the biological and physico-mechanical levels. One of many methods allowing for the controlled biodegradation of bone replacement materials is the design of implants made of various composites or mixtures. Combining two or more different materials results in the manufacture of an absolutely new composite biomaterial, which is frequently superior in terms of biological and mechanical properties. Due to its specific properties, bioceramic is very frequently used as a basic component of ceramicceramic, metal-ceramic and polymer-ceramic systems (Thomas et al., 2005). Hydroxyapatite is very frequently included in bone composites, mostly due to its chemical similarity to the natural components of bone. Some studies have confirmed the positive effects of hydroxyapatite used in combination with metals, ceramic or polymers (Abu Bakar et al., 2003;Bellucci et al., 2011;Choi et al., 2004;Daculsi et al., 2003). The results of research on HAp-containing composites based on resorbable polymers, mostly polylactide (PLA), polyglycolide (PGA), co-polymer of glycolide and lactide (PLGA) or collagen [Cie lik et al., Nagata et al., 2005], seem particularly interesting. These studies confirmed the possible application of such composites as binding and reconstructive elements in reconstructive surgery, as culture media in tissue and genetic engineering, and as drug carriers (Wei & Ma, 2004;Nagata et al., 2003). Although hydroxyapatite is characterized by high biocompatibility, its reactivity with living skeletal tissue is relatively low. In view of these findings, attempts to design composites that combine this biomaterial with markedly more biologically active bio-glass propose an interesting solution. Such a composite can stimulate osteogenesis, leading to the rapid formation of new skeletal tissue around the implant (Yuan et al., 2001;Ravarian et al., 2010). Biological tests of SiO 2 -CaO-Mg/natural HAp (bovine bone) composite obtained by means of thermal plasma processing have confirmed its lack of toxicity. Moreover, the glass-ceramic contained in this composite was confirmed to stimulate the growth and proliferation of human fibroblasts (Yoganand et al., 2010). Attempts at bone replacement with materials obtained by combining autogenous bone and deproteinized bovine bone constitute another example of the potential optimization of biological conditions for new skeletal tissue growth within a skeletal defect. The inclusion of autogenous material with osteoinductive properties in such composites results in the enhanced formation of better quality bone Pripatnanont et al., 2009;Thorwarth et al., 2006;Thuaksuban et al., 2010). In vitro cellular studies have confirmed that allogenic materials of human origin (demineralized bone matrix, deproteinized bone) may also have some osteoinductive activity. This activity was confirmed by an increase in alkaline phosphatase (ALP), osteocalcin (OC) and Ca 2+ concentrations observed in human bone marrow stromal osteoprogenitor cells (hBMSCs) cultured for three weeks in medium based on human bone components ).
The increasing demand for bone replacement materials stimulated us to design original composite materials for use in the regeneration of skeletal defects, characterized by both osteoinductive and osteoconductive properties. In designing such composites, we have used well-described biomaterials that have been applied successfully to skeletal defect regeneration, namely bio-glass (BG) and tricalcium phosphate (TCP). Lyophilized human bone obtained from a tissue bank served as a base for designing three types of mixtures: 1) human bone/bio-glass; 2) human bone/TCP; and 3) human bone/bio-glass/TCP. Such variants of material combinations enabled us to analyze the effects of particular components on the process of bone formation, and specifically on its dynamics and on the quality of newly formed bone. Also, the selection of the components included among the analyzed materials (mixtures) was not accidental. We have assumed that both bio-glass and tricalcium phosphate will activate and stimulate osteoblasts to dynamically grow on a biological scaffold of lyophilized human bone, thereby providing optimal biological conditions for the formation of full value bone.

Material
The materials used in the study included lyophilized, deproteinized human bone (B)group B, and its mixtures with: 1) bio-glass (BG) in a proportion of 80:20% weight ratio (B:BG) -group B+BG; 2) tricalcium phosphate (TCP) in a proportion of 80:20% weight ratio (B:TCP) -group B+TCP; 3) bio-glass and tricalcium phosphate in a proportion of 70:15:15% weight ratio (B:BG:TCP) -group B+BG+TCP. The deproteinized human bone and all other components used in the mixtures were in a granulated form with diameters ranging from 0.3 to 0.5 mm.
Human bone used in this study was obtained from the Tissue Bank of the Regional Center of Blood Donation and Treatment in Katowice (Poland). It was cancellous bone subjected to lyophilization, deep freezing and irradiation sterilization with a dose of 35 kGy.
The bio-glass was made from the CaO-SiO 2 -P 2 O 5 system with the use of the sol-gel technology, in the laboratory of the Department of Glass and Amorphous Coatings at the AGH Science and Technology University in Cracow, Poland. Its high-calcium A2 variety was used (54% mol. CaO) with a density of 2.9082 g/cm³, a dominating glassy phase and the beginnings of apatite crystallization. The thermal treatment of the bio-glass was performed at a temperature of 800˚C, and its specific surface area (calculated by BET method) amounted to 57.8166 m 2 /g.
Resorbable, monophasic TCP ceramic -a salt of trialkaline orthophosphoric acid Ca 3 (PO 4 ) 2 -was synthesized from powdered components obtained by means of wet synthesis in the Bioceramic Laboratory of the Department of Ceramic and Fire Resistant Material Technology at the AGH Science and Technology University in Cracow, Poland ( lósarczyk & Paszkiewicz, 2005;Zima at al., 2010). The reagents included CaO (obtained by means of calcination of Ca(OH) 2 -pure for analysis; MERCK, Poland), and H 3 PO 4 (pure for analysis; POCH, Poland).

In vivo animal experiments
The study was carried out on a group of 48 guinea pigs, with an equal number of both sexes, and weights ranging from 500 to 600 grams. The animals were divided randomly into four groups, which corresponded to 12 animals for each studied composite material (6 males and 6 females). The animal experiments were performed on the 7th, 14th, and 21st days of the study, as well as after the 4th, 8th and 12th experimental weeks. Two guinea pigs (one male and one female) were examined in each experiment. All animal surgical procedures were performed at the Central Experimental Animal Farm at the Medical University of Silesia and were granted permission by the university's Bioethical Board for Experimental Animals.
Before starting any surgical procedures, the animals received general anesthesia with thiopental (0.4 g/kg b.w.). Bone defects (6 mm in diameter and 3 mm in depth) were formed bilaterally on the external surface of the mandibular trunk, 2 mm below its lower edge (between the radices of the incisive and molar teeth) with the aid of a rosette burr placed in the straight hand-piece of a dental machine (Fig.1a). Depending on the experimental group, bone defects on the right side of the mandible were filled with: 1) a preparation of deproteinized human bone (group B), 2) a mixture of deproteinized human bone and bioglass (group B+BG), 3) a mixture of deproteinized human bone and tricalcium phosphate (group B+TCP), or 4) a mixture of deproteinized human bone with bio-glass and tricalcium phosphate (group B+BG+TCP). Before implantation into the mandibular bone defect, each material was mixed with the animal's blood obtained from the surgical wound (Fig. 1b)  proportion of 25:75% weight ratio (examined material to blood). Blood served as a binder, supplying the implanted material with organic components. Additionally, it prevented the displacement of the implanted granules from within the bone defect, and facilitated their insertion. The defect on the left side of the mandible was left to fill postoperatively with clotted blood and undergo spontaneous healing (Fig.1c), and was considered a control to the right mandibular defect in each animal (control group). The bilateral wounds were closed in multiple layers with Dexon 4.0 sutures.

Methods of examination
On each examination day, all guinea pigs were clinically evaluated for surgical wound healing and general condition. Additionally, radiographs were taken in order to assess the regeneration of mandibular bone defects. Newly formed skeletal tissue in the defects was also analyzed quantitatively and qualitatively in terms of bone mineral density at the implantation site (computed tomography and radiographic densitometry). Following euthanasia with Morbital (sodium pentobarbital 133.3 mg/mL and pentobarbital 26.7 mg/mL, 1-2 mL/kg b.w.), the operated area and surrounding tissues were examined macroscopically. Moreover, tissue specimens from the euthanized animals were subjected to histopathologic analysis of the skeletal tissue and bone marrow at the implantation site and at its periphery, including the healing rate of the skeletal tissue and the soft and hard tissue reactions to the implanted material. Histopathologic analysis also included organs involved in detoxification of the body (i.e. liver and kidneys).
Radiographs of the mandibular trunk with the sites of the bone defects were taken with a Heliodent type MD/D -3195 nr 051692 apparatus (SIEMENS), and AGFA DENTUS M2 CONFORT films for axial pictures (75 mm x 65mm), using the following exposure conditions: 0.16 s, 7 mA and 60 kV.
The tissue specimens taken for the histopathologic examinations were preserved in a 10% solution of buffered formalin. The osseous tissue was decalcified either in a 10% solution of disodium versenate, or electrolytically in Romeis liquid (80 mL of hydrochloric acid + 100 mL of formic acid, diluted to make 1000 mL with water) using PW23 bone decalcifier and an electric current of 0.5A. Next, all the tissues were routinely processed in Technicon's Duo autotechnicon using the sequence of 96% alcohol, acetone and xylene. They were then embedded in paraplast. The obtained cubes were then shaved on a Microm HM335E rotating microtome. The shavings (4-6 microns in thickness) were put onto basic slides, deparaffinized, and stained with hematoxylin and eosin (H&E). They were subsequently mounted with Canada balsam. Histologic slides were analyzed under an Olympus BX50 light microscope equipped with a set for optical (Olympus SC35) and digital microphotography (Olympus Canmedia C-5050 Zoom), at 40 to 400 x magnification. Additionally, the slides were consulted by other histologists using a dual-head Nikon Labophot-2 microscope.
Bone mineral density (BMD) was determined by means of dual-energy X-ray absorptiometry (DXA) with a DPX-L densitometer (LUNAR Radiation Corporation, Madison, USA) using Small Animal Appendicular scanning. Total bone mineral density (BMD; g/cm 2 ) was determined on the basis of examination of the entire area of the transverse cross-section of the bone defect (28.26 mm 2 ), analyzed as 0.5 mm 2 to 3 mm 2 sections. One hundred and twenty

www.intechopen.com
Tissue Regeneration -From Basic Biology to Clinical Application 376 BMD measurements were taken in each group (corresponding to 20 measurements per experimental day). In most cases, reproducible results were not included in further analysis; o n l y u n i q u e v a l u e s a n d t h e m o s t f r e q u e n t reproducible values were analyzed, which corresponded to 10 BMD measurements for each day of examination. Additionally, the BMD of the normal mandibular trunk of the guinea pig was determined for the purpose of comparative analysis. Based on 10 consecutive measurements, this value was estimated at 0.51 ± 0.001 g/cm 2 . During the measurements, the densitometer was regularly calibrated and controlled according to the manufacturer's recommendations. Both DXA measurements and BMD result analysis were performed by the same investigator.
Additionally, computed tomography (CT) was used to determine the bone mineral density, expressed in Hounsfield units (HU). These measurements were taken using a Somatom Emotion 6 scanner (Siemens; exposure parameters: 13.4 s, 14 mA, 130 kV). Transverse crosssections of the mandible were visualized in 2 mm slices. Then, cross-sections including the area of the bone defect were selected and analyzed using Volume Viewer software. In each experimental group, six measurements were taken for each examination day. Additionally, six measurements of the normal mandible were taken for the purpose of comparative analysis. Based on these measurements, the normal bone mineral density was determined to be 1218 ± 15.2 HU.
The results of bone density are presented as mean values ± standard deviation. Variables distribution was evaluated by the Shapiro-Wilk test. Homogeneity of variance was assessed by the Levene test. ANOVA for repeated measurements with contrasts analysis were done to assess time and preparation of deproteinized human bone type interaction. The Mauchley test was done to check sphericity. Differences were considered to be statistically significant at p<0.05. All calculations were performed using the commercially available statistical package Statistica 9.0.

Clinical observations
Throughout the entire study period, no complications in surgical wound healing were observed in animals of any experimental group. The guinea pigs were calm, which suggested a lack of pain. The animals ate and drank water normally, and neither scraped against the cage nor scratched their wounds during the entire postoperative period. Also, wound dehiscence was not observed in any of the groups.
Tissue edema over the surgical skin wounds resolved 3 to 5 days following surgery and was replaced by protrusions of the tissue. There were no signs of excessive fluid accumulation around the wound or of hematoma formation, but in some animals skin redness was observed around the stitches. In most animals, tissue protrusions persisted until the 21st day after surgery. The stitches were removed 10 to 14 days after surgery. Over the entire study period, the animals gained weight gradually (in a statistically insignificant manner).

Macroscopic examination
Up to 14th experimental day, the sites of the implanted bone defects were clearly distinguishable from the surrounding tissues as clear, oval protrusions covered with delicate tissue, which could be compressed elastically. Additionally, clearly distinguishable white granulation was visible throughout the superficial tissue in animals of the B+BG and B+BG+TCP groups. Probably, these granules corresponded to the bio-glass particles included in the material implanted in these groups. In the control group, the site of the bone defect was visible as a clear protrusion up to the 7th experimental day. On the next day of examination (the 14th day), however, the site was covered with a tissue whose coloration was darker in comparison to the surrounding tissues.
After the 3rd experimental week, the protrusion visible over the implantation site in groups B+BG, B+TCP and B+BG+TCP was markedly smaller in size, and covered with a hard tissue of more compact texture. Coloration of the tissues covering the implantation site still differed from the color of normal bone. In group BG, white granulation was still visible throughout the superficial tissue. On the 21st day after surgery, the coloration of the bone defect implanted with deproteinized human bone (group B) resembled the color of the surrounding tissues more closely than in previous periods. In both group B and in the control group, clearly distinguishable small areas in the form of a dark-colored spot were visible within the defects. Additionally, tissue protrusion observed over the defects in the control group increased in size when compared to previous periods.
Four weeks following implantation, the area of the bone defect in group B+BG was slightly smaller in size but still clearly distinguishable and differed in color from the surrounding tissues. Similar changes were observed in the B+TCP group; the implantation site in this group was concave, but still appeared hard when compressed. The tissue visible over the implantation site in group B+BG+TCP resembled the surrounding normal tissues the most closely when compared to the other groups. In group BG, white granulation was still visible throughout the superficial tissue covering the bone defect. In the control group, on the other hand, a large protrusion was still visible over the implantation site, with a small darkcolored spot at its center.
After the 8th and 12th weeks of experiment, the implantation sites of the group B animals were still clearly distinguishable on macroscopic examination; they were protruding and differed in color from the surrounding tissues. Protrusions over the implantation sites were also visible in group B+BG, but only up to the 8th week of the study. After this time, the implantation site was hardly distinguishable from the normal tissues, and only the presence of granulation, which was hardly visible throughout the superficial tissue, enabled its visual identification. In group B+BG+TCP, the implantation site could also only be localized due to the subtle appearance of white granulation as early as after 8 weeks of the experiment. In group B+TCP, no concavity was observed over the implantation site beginning from the 8th week of the study, and the tissue covering the defect only slightly differed in color from the surrounding normal bone. After 12 weeks of observation, the tissue over the bone defects in this group had an identical appearance to the surrounding tissues. On the last examination day, the site of the bone defect was distinguishable only in the control group. Although the implantation site was covered with hard tissue, minute concavities and dark spots were still visible on its surface.

Radiological examinations
Material-related differences in the rates of new skeletal tissue formation were revealed as early as after the 7th day of the study. In group B, spherical translucencies with regular edges were observed on radiographic images, with a size corresponding to the size of the bone defect. The initial process of bone reconstruction, manifested by a foggy appearance of the implantation site, was observed no earlier than after the second experimental week (Fig. 2a). At that same time point, initial signs of bone formation were also observed on radiographic images taken in group B+BG+TCP (Fig. 5a), while in group B+BG this phenomenon was already visible after 7 days (Fig. 3a). Irregular translucencies at the implantation site were seen in this group, but a small shadow was observed in the central zone, whose area and intensity increased with time. Radiologic findings suggested that the beginning of new tissue formation was most delayed in group B+TCP (Fig. 4a) and in the controls (Fig. 6a). In these groups, a distinct shadow was observed no earlier than after the 3rd week of the experiment; this shadow was larger and more intense in B+TCP group.
After the 4th week of the study, radiographic images taken in group B revealed the nearly complete formation of new skeletal tissue. At this point in time, indistinct translucencies were seen only at the periphery of the skeletal defect, suggesting osteogenesis was ongoing in this area. After 8 weeks of the experiment, bone reconstruction was complete in this group. The whole defect was excessively shadowed, suggesting that tissue with greater mineralization was present in this area when compared to normal bone (Fig. 2b).
As with group B, in group B+TCP the mineralization of the implantation site was complete after the 8th week of the study, as suggested by a fully shadowed area of the bone defect visible on radiographic images (Fig. 4b). At the same time, a small translucency was visible in the superior-medial aspect of the bone defect in group B+BG. In this group, the process of bone formation was completed no earlier than after the 12th week of the experiment. This was confirmed by the excessive mineralization of skeletal tissue, manifested radiographically as a distinct shadow at the implantation site (Fig. 3b).
In the B+BG+TCP group, a homogenous shadow was visible at the implantation site as early as after the 8th week of the experiment, and an incomplete process of skeletal tissue regeneration was suggested only by the presence of spotted translucencies. However, the site of the bone defect could not be distinguished from the surrounding tissues until at least the 12th week of the study (Fig. 5b). In the control group, the process of bone formation was markedly delayed, and after the 8th week of the study a distinct, longitudinal translucency could still be seen at the site of the bone defect. Moreover, noncalcified areas were still visible on the last examination day despite newly formed skeletal tissue present within the entire area of the bone defect (Fig. 6b).

Radiographic densitometry
A gradual increase in BMD was observed in all experimental groups and in the controls. The highest BMD value was observed after the 8th week of the study in bone defects implanted with B+TCP (0.400.05 g/cm 2 ) and B+BG+TCP (0.390.05 g/cm 2 ) (Table 1). However, these values were still lower than the normal BMD of the mandibular trunk determined in guinea pigs (0.51± 0.001 g/cm 2 ). The most regular increase in BMD was observed in defects implanted with B+BG+TCP, followed by B and B+BG. Some irregularities in the time profiles of BMD changes were noted, however, in animals of group B+TCP and in the controls, suggesting inhomogeneous formation of new skeletal tissue (Fig. 7).

BMD [g/cm 2 ]
Additionally, statistical analysis revealed intergroup differences in the time profiles of BMD changes. The most pronounced differences were observed between groups B and B+BG -0.98, followed by B+BG vs. B+TCP -0.79, and B vs. B+TCP -0.78. Slight differences in the time profiles were noted between the B+BG+TCP group and other material groups, and no significant differences were observed when experimental groups were compared to the control group (Table 2). Moreover, a relative increase in BMD between the 1st and 12th experimental weeks was calculated for each group (Table 3). The highest relative increase in BMD was observed in bone defects in group B+BG+TCP (25.49) but the increase in the B+TCP group was only slightly lower (24.14). The lowest relative increase in BMD between the 1st and 12th experimental weeks was observed in animals of group B+BG (9.98); the time plot of BMD changes in this group was closest to a vertical line.

Fig. 7. Time profile of changes in bone mineral density (BMD) determined radiographically in bone defects implanted with various materials and in control defects
Due to non-sphericity (Mauchley test: 0.42; p<0.001) only multivariable tests were used. We confirmed that there were statistically significant changes in BMD with time (F = 894.17; p<0.001). Moreover, the interaction between time and the type of material used (group) was also statistically significant (F = 35.59; p<0.001). pronounced difference was observed between groups B+BG and B+TCP (4.33), while the lowest difference pertained to groups B+TCP and B+BG+TCP. After 12 weeks of the experiment, the most pronounced difference in BMD of the bone defect was noted between the control group and the B+BG+TCP group, while the difference between the B and B+TCP groups was the lowest. We have also observed statistically significant differences in the growth of BMD with time between the analyzed groups (F=8.15; p<0.001). The smallest changes in BMD were yielded by B+BG (0.060.01), then the control group (0.080.01) and B (0.110.02). The largest changes were observed for B+TCP (0.140.01) and B+BG+TCP (0.160.01). For all paired comparisons, statistically significant differences were noted (p<0.01). Table 5 summarizes the values of bone density in the experimental groups and in the controls as determined by CT. A gradual increase in the density of healing bone defects was observed in all studied groups. After 12 weeks of the experiment, the highest values of CT bone density were observed in the B+BG (1014.8±53.9 HU) and B+BG+TCP (941.2±28.9 HU) groups, while the lowest values were noted in the controls (812.3±21.8 HU). However, the highest determined values of CT bone density were still lower compared to the bone density of the normal mandible (1218 ± 15.2 HU).

CT Bone Density [HU]
Group The most homogeneous increase in CT bone density values was observed in groups B+BG and B+TCP, and in the controls. Some irregularities in the time profile of bone density were, by contrast, noted in animals of the B and B+BG+TCP groups (Fig. 8).

Fig. 8. Time profile of changes in CT bone density (expressed in Hounsfield units, HU) in bone defects implanted with various materials and in control defects
Due to non-sphericity (Mauchley test: 0.0030; p<0.001) only multivariable tests were used. We confirmed that there were statistically significant changes in Hounsfield Units with time (F = 1244.88; p<0.001). Moreover, the interaction between time and the type of material used (group) was also statistically significant (F = 65.35; p<0.001).
When the time profiles of CT bone density changes were compared between the studied groups, the only significant differences observed were between groups B and B+BG -0.86, and B+TCP and B+BG+TCP -0.30 (Table 6). The most evident intragroup differences in CT bone density determined in the 1st and 12th experimental weeks were observed in the control group (29.77), but the differences were only slightly less pronounced in groups B+TCP (24.10) and B+BG+TCP (22.94). The least evident differences between the two analyzed time points were noted in the B (10.67) and B+BG (16.77) groups (Table 7). After seven days of the experiment, the most pronounced intergroup differences in CT bone density were observed between the control group and groups B and BG. The least pronounced intergroup differences noted in this period pertained to groups B and B+BG.

Profile comparison (p-values)
After 12 weeks of the study, the most evident intergroup differences in CT bone density were again observed between the controls and group B+BG, while the lowest differences were noted between groups B and B+TCP.

Histopathologic analysis
In all analyzed groups, after 7 days of the study, mandibular bone defects had filled with immature fibrous tissue. Numerous, minute granules of deproteinized human bone (with no signs of activity) were seen in this tissue (group B). Additionally, depending on the implant composition, foggy fractions of bio-glass or linearly cracked deposits of tricalcium phosphate could be seen (in groups B+BG, B+TCP and B+BG+TCP). At the same time, active reconstruction of skeletal tissue was observed at the entire periphery of the bone defects implanted with B, B+TCP and B+BG+TCP, as suggested by the presence of ground substance (osteoid) containing numerous immature bone trabeculae covered with osteoblasts. In controls, as well as in groups B and B+BG, blood clots along with the remnants of necrotic bone trabeculae could be seen in the bone marrow at the base of the bone defect. In groups where bio-glass was implanted into the bone defect (B+BG and B+BG+TCP), fragments of this material formed pseudocystic structures that were covered with a thin connective tissue capsule comprised of fibroblasts, fibrocytes and single giant polynuclear cells.
After two weeks of the experiment, the implanted bone defects were filled with mature fibrous connective tissue that contained collagen fibers. Only in group B could immature connective tissue with numerous blood vessels be seen at the periphery of the defect. Immature bone trabeculae were visible in this tissue, surrounded with osteoblasts. Depending on the implant composition, fragments of deproteinized bone granules, bio-glass and/or tricalcium phosphate were seen in the fibrous connective tissue. Around BG and TCP particles, distinct, thick-walled pseudocystic structures could be observed, containing numerous giant polynuclear cells. Additionally, giant polynuclear cells were frequently visible on the surface of deproteinized bone but they did not form distinct capsule-like linear structures in this location. Fragmentation of some TCP particles could be observed due to infiltration by cells composing the previously mentioned cystic structures. Active osteogenesis was evident in all analyzed groups, as manifested by the pronounced growth of numerous immature bone trabeculae covered with osteoblasts. This process was particularly intensive around the particles of implanted material.
In group B, advanced reconstruction of bone defects was observed after three weeks of the study. Most bone trabeculae filling the defect were mature with either no or very little osteoblastic activity. Linearly placed osteoblasts, or even osteoclasts, could be seen on the surface of remaining trabeculae (Fig. 9). A similar advancement in the bone formation process was observed in B+TCP specimens: osteoid was formed on the base of connective fibrous tissue along with the intensive growth of numerous bone trabeculae. Some of these trabeculae were mature already and showed no signs of cellular activity. Skeletal tissue regeneration in this group was particularly enhanced around the fragments of deproteinized bone and tricalcium phosphate (Fig. 10). In contrast to B particles, TCP particles showed signs of dilution and structural fragmentation. Only a few giant polynuclear cells could be seen around the implants, and the previously observed cystic structures of TCP had only residual character. After three weeks of the study, slightly less advanced processes of skeletal defect healing were observed in histologic specimens from groups B+BG and B+BG+TCP. Growth of immature skeletal tissue was observed in these groups on the "scaffold" of deproteinized bone, giving the impression of the implant being "incorporated" into the growing bone trabeculae. Only single giant polynuclear cells could be seen on the surface of specimens from group B+BG. Additionally, thin, linearly placed cells could be observed around some bio-glass particles, forming pseudocystic structures. These structures were visible mostly in areas directly adjacent to f i b r o u s c o n n e c t i v e t i s s u e . N u m e r o u s b o n e t r a b e c u l a e w e r e o b s e r v e d a r o u n d t h e B+BG+TCP mixture particles, as well as in the fibrous connective tissue between the particles (Fig. 11). Some trabeculae were mature and showed no signs of osteoblastic activity on their surfaces. Some implant particles in this group were covered with giant polynuclear cells forming structures resembling foreign body granulomas. In the control group, all osteoid trabeculae were surrounded by osteoblasts still showing signs of osteoblastic activity.
After four weeks of the study, only the bone defects in group B were nearly completely filled with mature, compact skeletal tissue. This tissue contained numerous "incorporated" granules of deproteinized human bone. Growing bone trabeculae with surface signs of osteoblastic activity could only be seen in a narrow layer of fibrous connective tissue located between the newly formed bone and the bottom of the defect. At the same time, mature cancellous skeletal tissue could only be observed at the periphery of bone defects in groups B+TCP and B+BG+TCP. "Incorporated" fragments of the implanted material, mostly human bone, were visible in this tissue. The central part of the defects was still filled with fibrous connective tissue, showing signs of the ongoing process of bone formation. This area also contained B particles and a few particles of TCP ceramic. These fragments of tricalcium phosphate were covered with a pseudo-capsule comprised of giant polynuclear cells, and were gradually fragmented and resorbed. After four weeks of the experiment, numerous mature bone trabeculae with no signs of cellular activity were observed in specimens from group B+BG (Fig. 12). Some of these trabeculae developed around B and BG particles, giving the impression of "incorporating" implanted material into the structure of reconstructed bone. Some bone defects were still partially filled with mature fibrous connective tissue containing numerous collagen fibers along with particles of implanted material. Additionally, pseudocystic structures could be observed around the BG particles, comprised of linearly placed giant polynuclear cells.
Histopathologic examination performed after eight weeks of the study revealed the completed process of bone healing in group B. Bone structure was fully regenerated, and mature compact bone and cancellous bone could be observed at the defect site, along with normal bone marrow (Fig. 13). Observations made in the B+BG+TCP group after the 8th and 12th weeks of the experiment gave similar findings. In this group bone defects were also filled with mature compact and cancellous skeletal tissue, with "incorporated" particles of bio-glass and deproteinized bone still visible (Fig. 14). These particles showed no signs of activity, and were covered with thin fibrous capsules. In the 8th experimental week, these particles could also be observed in fibrous connective tissue. After the 8 th week of the study, the process of osteogenesis was still incomplete in B+BG and B+TCP specimens. In defects implanted with bio-glass, some areas, usually peripheral ones, were still filled with fibrous connective tissue containing numerous collagen fibers. This connective tissue showed signs of ongoing bone formation: maturating or mature bone trabeculae, along with bio-glass particles forming pseudocystic structures (Fig. 15). Particles of deproteinized bone were more rarely evidenced. At the same time, continued bone formation was observed in the central part of bone defects in group B+TCP (Fig. 16). This process was particularly intensive around the B particles, giving the characteristic impression of "incorporating" this material into newly formed bone. After eight weeks of the study, only some bone trabeculae in the control group showed signs of osteoblastic activity, while other areas of the defect contained mature trabeculae. After 12 weeks of the study, the process of skeletal tissue regeneration in specimens from groups B+BG and B+TCP was still not complete. Although the defects were nearly filled in their entirety with mature compact and cancellous bone, mature fibrous connective tissue containing numerous collagen fibers and showing signs of ongoing osteogenesis could still be seen in the superficial zone (Fig. 17) were revealed in fibrous connective tissue (Fig. 18). In the control group, specimens obtained in the 12th week of the study contained completely matured and fully mineralized bone.

Discussion
Due to the limited regenerative ability of skeletal tissue, bone grafting or the implantation of bone derivatives or bone replacement materials is required for the complete healing of large bone defects, whether the result of surgical removal of skeletal cysts or tumors, or caused by other skeletal disorders (Precheur, 2007). The most satisfactory results in stimulating skeletal tissue regeneration have been reported after using autogenous grafts. After being implanted into the bone defect, autogenous grafts can induce all the basic mechanisms responsible for bone reconstruction, i.e. osteogenesis, osteoinduction and osteoconduction (Giannoudis et al., 2005;Merkx et al., 2003). Osteogenic activity is associated with the presence of osteoprogenitor cells in the periosteum, endosteum and bone marrow. In the case of free bone grafts, some of the cells located most superficially may survive and are involved in regeneration processes. The results of bone healing stimulation are definitely most satisfactory when autogenous cancellous bone chips are implanted, since this type of bone contains a high number of osteoprogenitor cells. Osteogenic activity can only be observed in fresh bone grafts. Osteoinduction is associated with the presence of so-called bone morphogenetic proteins (BMP) in the bone matrix. These proteins are released during bone remodeling, and can stimulate minimally differentiated connective tissue cells surrounding the graft to transform into osteoblasts (Barradas et al., 2011). Both fresh autogenic bone grafts and the allogenic grafts obtained from tissue bank (especially when frozen and partly decalcified) have osteoinductive properties. Additionally, BMP preparations can be obtained by extraction from bones, and as of recent, also by biotechnological synthesis in recombinant form. Allogenic bones are very frequently lyophilized, which depletes them of BMP. Additionally, such grafts lose their immunogenic properties due to irradiation sterilization and deep freezing (Bohner, 2010;Liu et al., 2008). This process of sterilization results in decreased durability of the material, and the preserved bone matrix has only osteoconductive properties. It is degraded by osteoclasts with the simultaneous formation of woven bone, which is further transformed into lamellar bone through the process of osteoclasia. Such grafts have been shown to undergo revascularization and remodelingsimilar to autogenous grafts, but at a slower rate. Apart from bone implants, organic and inorganic alloplastic bone replacement materials also have osteoconductive properties. Combined with growth factors and autogenous barrier membranes, they are frequently used as basic elements in the process of guided bone regeneration (GBR) (Kao & Scott, 2007;Schwarz et al., 2007).
For various reasons, different bone materials are frequently combined with each other or with alloplastic biomaterials. As a result, biologically improved material compositions are obtained, some of which positively influence the bone formation processes. Combination of natural hydroxyapatite with chitosane resulted in a composite with osteoconductive properties. In the presence of this composite, tibial consolidation in rabbits was observed as early as 12 weeks after implantation, and complete healing was observed after 16 weeks of the study (Yuan et al., 2008). In another study, a composite based on bovine bone with the addition of bio-glass showed no cytotoxicity to human fibroblasts. Moreover, a crystalline carbonated apatite phase was developed on the sample surface as early as 12 days after immersion in simulated body fluid (Yoganad e al., 2010). Another example of the positive effects of combining deproteinized bovine bone with autogenous bone comes from a study in which such a material was used for the regeneration of bone defects in the frontal part of the porcine skull. The presence of autogenous bone in the mixture was the basis for the osteoinductive properties of the material and the more favorable biological conditions for bone growth when compared to deproteinized bovine bone alone (Thorwarth et al., 2006). Similarly, more satisfactory clinical results were reported when deproteinized bovine bone was used in combination with autogenous bone in the management of alveoschisis in humans, instead of bone autograft alone (Thuaksuban et al., 2010). Experiments on rabbits have also given interesting results. It was revealed that the addition of deproteinized bovine bone to autogenous grafts increased the mean optical density of newly formed skeletal tissue, with a simultaneous decrease in its content in bone defects (skullcap). The opposite effects were observed when autogenous bone graft was used alone (Pripatnanont et al., 2009). Using a mixture of allogenic bone and deproteinized bovine bone (BioOss ® /Orthoblast II ® ) for the purposes of maxillary sinus lift did not have results as satisfactory as with the application of deproteinized animal bone or synthetic bone (Osteon ® ) alone. The individual use of one of these two implant materials was associated with a higher percentage of newly formed osseous fraction collected from the lateral sinus at 4 and 6 months post-operatively .
In this study we have combined allogenic bone with artificially obtained biomaterials (bioglass -BG, and/or beta-tricalcium phosphate -TCP) in order to form bone replacement material with improved biological characteristics. Reference materials for comparative analysis of the studied mixtures (B+BG, B+TCP, B+BG+TCP) included lyophilized human bone (B) and the clotted blood of experimental animals. Both clinical observations and further macroscopic, radiographic and histopathologic examinations confirmed that bone defects healed normally in the presence of all studied biomaterials. However, the type of implanted mixture modulated the kinetics of bone formation and the quality of newly formed bone. Bone regeneration was induced markedly earlier whenever biologically active bio-glass was included in the implanted mixture (B+BG, B+BG+TCP). In groups where bioglass was implanted, irregular shadows were observed on radiographic images of the bone defect sites as early as after two weeks of the study. Probably, these radiographic changes resulted from ongoing reparative processes within the bone. This was additionally confirmed on histopathologic analysis, which revealed intense bone formation processes as early as three weeks after the implantation of BG-containing material. However, bone density measurements (BMD and CT bone density) taken in the early period of this study confirmed the superior quality of newly formed bone only in case of the B+BG mixture. It is plausible that the lower bone densities determined for B+BG+TCP implants resulted from the low content of bio-glass in this mixture. Moreover, as confirmed by histopathologic analysis, resorption of beta-tricalcium phosphate contained in B+BG+TCP already began in the early period of this study. Nonetheless, in the later period of this study, increases in BMD and CT bone density of B+BG implanted bone were markedly lower. As a result, after 12 weeks of the experiment, the defects filled with this mixture were characterized by the lowest BMD values, and histopathologic examination confirmed ongoing bone formation. The final result of bone regeneration was markedly better in the case of defects implanted with B+BG+TCP. In the 12 th week, histopathologic analysis revealed mature skeletal tissue (both compact and cancellous bone) at the implantation sites, and this finding was confirmed on radiographic examination. Additionally, new bone formed using B+BG+TCP implantation was characterized by the highest BMD and relatively high CT bone density. Therefore, this regenerated bone most closely resembled the normal skeletal tissue of experimental animals of all mixtures examined. In the 12 th w e e k o f t h i s s t u d y , b o n e formation processes were still observed in B+TCP implanted defects. Although the BMD of tissue formed on the basis of this implant was higher than in the B+BG implanted bone, it was still lower than in the B+BG+TCP group. Notably, in both the 1 st and 12 th experimental weeks, only slight differences in BMD and CT bone density were observed between the B+TCP and B+BG+TCP mixtures. Undoubtedly, the process of bone defect regeneration was completed the earliest in group B. Histopathologic studies confirmed that bone defects in this group were filled with mature skeletal tissue with no signs of osteoblastic activity as early as after eight weeks of the study. Early completion of skeletal healing was also confirmed by the radiographic images taken in this group. In the 12 th experimental week, BMD of bone defects implanted with human bone alone was higher in comparison to defects filled with B+TCP and B+BG+TCP, in contrast to the early period of this study when BMD and CT bone density in group B were among the highest.
Results of this study suggest that the quality of bone formed in the late period of the experiment was poorest in the control group, as suggested by the low bone density of defects healing on the basis of clotted blood. Since histopathologic studies confirmed that bone formation in controls was complete in week 12, one should not expect further increases in bone density in this group. It is likely that bone density would increase further, however, with prolonged observation of the B+BG and B+TCP groups, since the last densitometric measurements in these groups were taken on incompletely matured bone.

Conclusion
This in vivo animal study revealed that both lyophilized human bone (B) and mixtures formed on its basis (B+BG, B+TCP, B+BG+TCP) are fully biocompatible materials. We have confirmed that, in the presence of these materials, there is a possibility of forming normal, mature skeletal tissue in mandibular bone defects of guinea pigs. This was possible mostly thanks to the inclusion of deproteinized human bone in analyzed mixtures. Due to its osteoconductive properties, the presence of human bone resulted in favorable biological conditions that promoted skeletal regeneration. Therefore, deproteinized bone formed a scaffold to support the growth of osteogenic cells within the defects. Furthermore, the addition of alloplastic materials, bio-glass and beta-tricalcium phosphate, markedly influenced the rate of bone formation and the quality of newly formed bone. The most satisfactory results were observed in the case of lyophilized human bone mixed with bioglass and beta-tricalcium phosphate (B+BG+TCP). The group implanted with this material was the only one in which fully matured compact and cancellous bone was observed on histopathologic examination performed after 12 weeks of the study. The particles of bioglass and beta-tricalcium phosphate included in the mixture induced the processes of bone formation and stimulated the growth of osteogenic cells. As a result of these initiated biological processes, defragmented and resorbed -TCP was gradually replaced with newly formed skeletal structures. Such a course of bone formation process modulated the quality of newly formed bone, as confirmed by high BMD and CT bone density values determined after 12 weeks in B+BG+TCP implanted defects. As confirmed on histopathologic examination, throughout the three-month period of this study, skeletal regeneration was not completed in defects implanted with B+BG and B+TCP mixtures. This incomplete regeneration was reflected by the lower bone density values observed in these groups when compared to the B+BG+TCP group. In conclusion, the results of this study confirmed our initial assumptions. We have revealed that, in addition to a proper course of bone formation processes, another important outcome in the presence of various bone replacement materials is the high quality of newly formed bone. In our opinion, these two aforementioned results were positively achieved by the mixture of lyophilized human bone with bio-glass and tricalcium phosphate.