Master of Science in Oral Biology, University of California, Los Angeles, 1995
Bone regeneration is a major limiting factor in periodontal regenerative therapy. The concept of guided tissue regeneration has provided the strongest evidence that tissue healing can be influenced exogenously by wound management. At the cellular level, the existence of osteoprogenitors with the capacity to produce bone in the wound site and the potential to exogenously influence the behavior of these cells offers the opportunity to further enhance regenerative wound healing. Chitosan, with a chemical structure similar to hyaluronate, has been implicated as a wound healing agent. The purpose of this research was to evaluate the effect of chitosan on osteoblast differentiation and bone formation in vitro. Mesenchymal stem cells were harvested from fetal mouse calvaria (Swiss Webster mice) prior to osteoblast differentiation and calcification (12-13 days in utero). Stem cells were seeded into six-wee culture plates at a density of 350,000 cells per well. Using this model, it is possible to quantify the influence of chemical mediators on osteoprogenitor differentiation and osteogenesis. Experimental wells were pretreated with 200m1 chitosan (2mg/ml in 0.2% acetic acid vehicle). Control wells were pretreated with 200m1 vehicle (0.2% acetic acid only) or remained untreated. Cells were allowed to grow under optimal conditions for 14 days. Cell cultures were fixed with glutaraldehyde and stained with VonKossa stain to identify bone forming colonies. Positive staining colonies were identified and counted under light microscopy. Histologic cross-sections identified osteoblasts and confirmed bone formation. Examination of control wells revealed 3.6 ± 0.6 colonies per well while experimental wells revealed 6.2 ± 1.2 colonies per well (p< 0.01). Computer assisted image analysis of the area of bone formed by control colonies was 0.34 ± 0.09 (relative units) while that of experimental colonies was 0.39 ± 0.06 (relative units) per bone forming colony (p=0.4691). The results of this in vitro experiment suggest that chitosan potentiates the differentiation of osteoprogenitors and facilitates the formation of bone.
The long term goal of this research is to investigate the optimal conditions for enhancing complete regeneration of periodontal, pert-implant and alveolar ridge supporting tissues. The osseous tissue component is the main structural supporting tissue of the periodontium and seems to determine the extent to which tissue regeneration will occur. For this reason, studies related to the activation and growth of bone forming cells are fundamental to the objectives of this research goal.
Cells with the potential for osteogenesis (osteoprogenitors) are stimulated to form bone by local factors or signals in the microenvironment. Chemical mediators or substances that enhance bone formation are thought to be conducive to perioodontal, pert-implant and alveolar ridge regeneration. The influence of chitosan (poly-N-acetyl glucosaminoglycan), an interesting biopolymer, on periodontal bone regeneration is of particular interest. This study investigates the effect of chitosan on osteoprogenitor differentiation and osteogenesis in vitro.
Pertinent concepts to be outlined as an introductory basis to this research include alveolar bone loss resulting from periodontal disease, periodontal therapy, periodontal wound healing, osseous wound healing and the regenerative potential of periodontal tissues. The biochemistry of chitosan and evidence of its influence on the wound healing process is offered to support the premise of this research.
A. Periodontal Disease, Bone Loss, Therapy and Regeneration
The periodontium is a unique structure in the human body. It is the only place where epithelium must create a barrier against a protruding, hard, mineralized structure in a harsh, wet environment that is teeming with microorganisms. The barrier created at the gingival-sulcus by gingival epithelium and connective tissue attachment is the weakest link in the integument preventing the invasion of microorganisms into the periodontal ligament. Periodontal disease is a bacterial induced inflammatory reaction that challenges that tissue barrier. Plaque (periodontal bacteria) accumulation on the tooth surface elicits an inflammatory response in the adjacent soft and hard tissues of the periodontium. An acute inflammatory reaction, often short-lived or nonexistent, is succeeded by a chronic inflammation. If persistent, the result is an inflammatory reaction that causes breakdown of the tissues within the gingival sulcus and leads to destruction of the periodontium.
Gingivitis, a chronic inflammatory disease that effects nearly all adults and most adolescents at some time in their lives, is defined as inflammation localized to the supracrestal gingival tissue (1, 2). Tissue damage is confined to the gingiva coronal to the transeptal and crestal fibers. An intact connective tissue apparatus between the root surface cementum and supporting bone, prevents down growth of gingival epithelium. By definition, gingivitis does not involve connective tissue attachment breakdown nor bone loss. Periodontal pocket formation, if any, is manifest by gingival swelling and not the result of tissue destruction. Treatment is simple and predictable. Professional cleaning and improvement in daily oral hygiene results in decreased inflammation and the associated gingival swelling. Complete restoration of gingival tissues and preservation of supporting periodontal structures can be expected. For this reason, gingivitis is considered a reversible periodontal disease process.
Periodontitis is a chronic inflammatory disease that effects a large percentage of the adult population in the United States (2). Human periodontitis comprises a heterogeneous group of infectious diseases of the periodontium that lead to destruction of the supporting tissues and is the leading cause of tooth loss. A variety of clinical periodontal disease entities have been described and are believed to differ with respect to bacterial etiology, host response and clinical disease progression (2). Most commonly, periodontal diseases are chronic in nature, slowly progressive diseases with mild to moderate bone loss occurring over a period of years. Some, less prevalent, periodontal diseases are more acute in nature and cause rapid destruction of periodontal supporting bone over a shorter period. Regardless of differences, the various forms of periodontal disease share the common characteristic of complex host-parasite interaction with an inflammatory component that results in destruction of periodontal supporting tissues (2). When inflammation progresses epically to the level of the supporting alveolus, bone is lost which ultimately reduces support for the tooth. Treatment of periodontitis is not as simple nor as predictable as that for gingivitis. The outcome is greatly dependent on the amount of remaining alveolar bone support and the morphology of the resultant periodontal defect.
Historically, treatment of periodontitis has been geared towards reducing inflammation and maintaining the health of the existing, albeit compromised structures. The goal of eliminating periodontal disease and maintaining teeth had been routinely attempted via methods of scaling, root planing, good oral hygiene and flap surgery. Osseous resective surgery techniques were developed to reduce osseous defects and pocket depth at a time when regeneration was not feasible (3). These therapeutic techniques were used to reduce periodontal pocket depth around teeth at the expense of osseous tissue. It was felt that teeth could be maintained better if pocket depth was minimized. This approach to treatment often meant accepting compromised esthetics, root sensitivity and weakened dental support. In other words, periodontally diseased teeth were treated and "maintained" for as long as possible. Untreated and severely periodontally compromised teeth were inevitably lost. Periodontal prosthetics (dental restorations designed to sustain periodontally weakened teeth) were used to replace missing teeth and help support those compromised teeth that remained.
When they were able to maintain their teeth, periodontal patients suffered unesthetic appearance of long teeth with dark spaces, food traps, root exposure with sensitivity and increased susceptibility to dental caries and loose teeth that were often less functional than before. In spite of these problems, keeping natural teeth was equated to treatment success.
Without treatment, the adverse effects (pocket formation) of periodontitis serve to perpetuate the disease process by creating areas that harbor pathogenic plaque and become increasingly more difficult to debride. The inflammatory destruction proceeds and bone loss continues. Teeth that suffer significant alveolar bone loss cannot support the forces of mastication and are ultimately lost. Although plaque-induced periodontal inflammation and tissue destruction completely resolves following the extraction of diseased teeth, alveolar bone loss may continue. Immediately following removal of teeth, the dimensions of the alveolar process collapse and will continue to resorb over time. The degree of alveolar ridge resorption becomes increasingly more problematic for patients wearing prosthetic tooth replacements, especially lower complete dentures. Dentures become less retentive and more unstable as the severity of alveolar bone resorption continues.
With the advent and recent successes enjoyed by endosseous dental implants, the quantity and quality of bone remaining in edentulous ridges has become increasingly more important. Controlled studies, case reports and practical experience show that endosseous dental implants are most successful when placed in abundant, dense bone. The edentulous anterior mandible is a prime example of an area of dense bone for the placement of dental implants. The anterior mandible has enjoyed the greatest success with regard to osseointegrated dental implants. Bone volume is equally important to the placement and success of dental implants. Areas of alveolar ridge that have lost bone height and width are less desirable for placement of endosseous dental implants because the surface area supported by bone is diminished. Unfortunately, patients with very porous bone and inadequate bone volume are not good candidates for endosseous dental implants. For this reason, alveolar bone regeneration via re-growth and grafting has become a highly desirable treatment goal.
Although endosseous dental implants have gained tremendous acceptability and achieved great success in restoring teeth for endentulous and partially edentulous patients, they also suffer from bone loss. Under some circumstances, dental implants seem to suffer from a chronic inflammatory disease similar to periodontitis (i). Although the nature and etiology has been debated, the so-called "peri-implantitis" can result in a progressive circumferential bone loss around the affected implant. Continued peri-irnplant bone loss leads to mobility and failure of the implant(s). Although significant research and treatment has been directed at the regeneration of bone around the "failing" dental implant, many implants suffering from bone loss are ultimately removed. Subsequently, the patient and surgeon are faced with the dilemma of deciding whether to replace the dental implant in an area that has suffered still more bone loss.
For these reasons, bone maintenance and bone regeneration have become essential concepts in the treatment of periodontal disease, in the healing of tooth extraction sockets and in the utilization of dental implants. Today, great efforts are being made not only to maintain and prevent bone loss but also to augment and regenerate bone around teeth and implants and to rebuild edentulous ridges. Ideally, therapy for periodontally diseased teeth (or implants) would result not only in arrest of disease, but also in regeneration of supporting periodontal tissues. Our understanding of the wound healing process and factors that facilitate the bone regenerative outcome are important to the future advances of periodontal regenerative therapy.
B. Principles of Wound Healing
1. Periodontal Wound Healing
Periodontal wound healing outcomes, as defined by the American Academy of Periodontology's Glossary of Periodontic Terms (1986), are as follows:
Repair: Healing of a wound by tissue that does not fully restore the architecture or the function of the part.
Regeneration: Reproduction or reconstitution of a lost or injured part.
Healing of the periodontal supporting tissues lost by periodontitis is complex due to the different tissue types involved. Melcher (1976) described the four tissue cell types contributing to healing and proposed a theory on the repair potential of the periodontium (5). He proposed the following possible scenarios based on the predominate cell type populating the root surface.
Periodontal supporting structures are composed of five distinct tissues, namely gingival epithelium, gingival connective tissue, periodontal ligament, alveolar bone and cementum. Melcher's (1976) model described above does not discuss the contribution of the cementum. The gingival epithelium and connective tissue fibroblasts generally have high mitotic activity and continue to proliferate even in the absence of tissue damage (6). The cells of the periodontal ligament, alveolar bone and cementum, on the other hand, have decreased mitotic activity in the absence of tissue injury but retain their ability to proliferate as the situation demands (6).
The most common periodontal wound healing events do not fully restore architecture nor function. Thus, by definition, periodontal wounds repair rather than regenerate periodontal structures. Examples of periodontal repair include, long junctional epithelial attachments, gingival connective tissue repair without bone fill and osseous healing to the root surface without intervening periodontal ligament fibers (ankylosis). These common periodontal wound healing results support Melcher's wound healing theories based on the type of tissue that predominately occupies the wound space. These results are inconsistent with the goals of regeneration. From an ideal perspective, wound healing events that result in repair might be considered failed attempts at regeneration. The dominance of certain tissues types in the wound healing process precludes periodontal wounds from regenerating normal tissue architecture and function. In other words, the sequence of events, magnitude of response and cells involved in those healing processes were not optimal for regeneration of periodontal structures.
Wound healing that leads to periodontal regeneration is a complex process of events involving biomechanisms that result in the formation of cementum on the root, a functional periodontal ligament with inserting fibers and alveolar bone. If the complex sequence of events are disordered, excessive, lacking or nonexistent, repair rather than regeneration is the result. In fact, repair is overwhelmingly the most common end result of periodontal wound healing. It has only been within the last few decades that periodontal regeneration has been considered a possible outcome. As mentioned previously, many periodontal therapies were based on arresting disease and creating the opportunity for tissues to repair in a manner consistent with health and maintenance. In other words, regeneration was not a goal because it was not considered, by most, to be predictable or even feasible.
In order to understand why wound healing results in repair, or more importantly, how wound healing events can result in regeneration, one must consider the complexity of the cells and biomechanisms involved. Periodontal wound healing is a battle between cells with the potential to regenerate normal structures and forces that interfere with that process. The forces that interfere with regeneration include those that are exogenously introduced as well as those that are native to the wound healing environment.
Just as laws of nature tend to seek randomness, the forces that disrupt periodontal regeneration tend to prevail in periodontal wound healing. External influences such as plaque and the harsh oral environment are often responsible for wound healing that results in repair. Bacterial plaque is not only responsible for the original breakdown but is also a major factor that inhibits regeneration. Micro-movement and agitation of periodontal wounds that result in repeated tissue disruption are also factors that favor repair over regeneration. The oral cavity is thus an environment with many forces that obstruct periodontal regeneration.
Elimination of these negative forces together with excellent oral hygiene and wound stabilization could improve the likelihood that regeneration would occur. However, even if they could be eliminated, there are wound healing events that prevail over the sequence of events that could lead to regeneration. In the absence of plaque and wound instability, periodontal regeneration requires an orchestrated series of events such as cell differentiation, migration, proliferation and production of structura1 tissue components. Cells with the potential to become cementoblasts and osteoblasts are essential if true regeneration is going to occur. Once differentiated, these cells must be capable of movement, proliferation, production and intercellular communication. Cementoblasts, fibroblasts and osteablasts communicate with each other and are influenced by the extracellular matrix and proteins which make up their environment. Any miscommunication between cells and influences from the extracellular matrix may alter the wound healing pathway and lead towards repair rather that regeneration.
The literature is replete with evidence that these regenerative events can and do occur (7, 8, 9,10,11). Each example tends to show that under selected conditions (e.g. when using a barrier membrane to exclude epithelium), regeneration of periodontal structures is possible. Unfortunately, we have not been able to determine the therapeutic modalities that consistently lead to complete regeneration in all circumstances.
Complete periodontal regeneration is hindered by the differing rates of proliferation, migration, and synthesis of the contributing tissue components. The gingival epithelium and gingival connective tissues may proliferate more rapidly and occupy the wound spaces before the alveolar bone forming cells have a chance to form new bone. If osteoprogenitor cells were given space and the opportunity to divide, migrate and produce osteoid, then bone could be formed and the periodontal framework might have a better opportunity to rebuild. This theory is supported by studies employing the concepts of guided tissue regeneration. Karring, et. al. (12) and Nyman, et. al. (13, 14) have demonstrated the repair potential of periodontal tissues in a series of well designed experiments. They concluded that the only cells with the capacity to form a new connective tissue attachment are cells originating from the periodontal ligament. These histologic and clinical studies gave credence to the possibility of achieving new connective tissue attachment to a previously diseased root surface.
Periodontal ligament cells have been shown to have some osteoblast-like characteristics (15, 16) and contain phenotypes that can differentiate into osteoblasts or cementoblasts (17). It has also been demonstrated that periodontal ligament cells can form mineralized nodules in vitro (18, 19). Proliferation, migration and differentiation of periodontal ligament cells are influenced by growth factors and proteoglycans (7).
Based on the theory that the periodontal ligament cells retain the greatest capacity to regenerate a connective tissue attachment, many studies and clinical treatments have utilized therapies for guided or protected proliferation of these cells (7). The most widely studied and utilized therapy generated from this theory is the use of barrier membranes to prevent epithelium and gingival connective tissues from occupying the wound space (7, 8), thus, selectively permitting periodontal ligament cells to proliferate and regenerate a connective tissue attachment. These materials have enjoyed t remendous popularity and widespread use because of their perceived beneficial effects. However, there are limits as to when this procedure is indicated based on the amount of potentially new connective tissue attachment and bone that can be formed. Regeneration procedures are least predictable in cases that have lost bone in the horizontal dimension or when the osseous defect is large. The reason for this limited result probably lies in the different potential of each tissue type and limited ability of osteoprogenitor cells to proliferate and build a matrix for hard tissue reconstruction.
Although regeneration occurs in the gingival epithelium and localized areas of the wound margins, complete restoration of the damaged periodontal tissues is not easy to accomplish. This is primarily due to the lack of bone or the "framework" of the periodontal supporting tissues. Periodontal wound healing events consist of dot formation which establishes an initial matrix for cellular migration, followed by clot degradation and matrix reorganization which leads to the final tissue composition whether it be repair or regeneration. In order to achieve periodontal wound healing that results in regeneration, the respective cells of each tissue type must proliferate, migrate and produce matrix in an organized manner and in an appropriate position relative to one another to restore the lost tissue structures. The wound matrix (clot) must be sustained throughout the healing process.
Healing of gingival tissue, following a gingivectomy, involves the formation of granulation tissue by gingival fibroblasts followed by the migration of epithelial cells (6). Epithelial cells continue to migrate until stopped by cell-to-cell or cell-to-barrier inhibition. Since epithelial cells do not encounter other epithelial cells on the surface of the tooth, the intact fiber apparatus suspends cell migration (6). The regenerated epithelial cells differentiate into junctional, sulcular and gingival epithelium depending on their location. As long as alveolar bone and connective tissue are mostly intact, restoration of periodontal tissues at an apical level can be expected. However, areas of the osseous defect that are further away from the bone (e.g. coronal aspect) are less likely to regenerate bone. Epithelium and gingival connective tissues heal the wound and occupy those areas before osteoblasts have an opportunity to build and calcify a matrix.
Bone regeneration is a major limiting factor in periodontal regenerative therapy. Even under optimal conditions, bone fails to regenerate completely. Without tissue separation or similar intervention therapy, bone always forms slower and to a lesser extent than gingival connective tissues and epithelium. Soft tissues quickly migrate and proliferate to occupy wound spaces before osseous tissues. Various substances have been used in periodontics in an attempt to stimulate, induce and/or conduct the regeneration of bone. These substances range from chemical mediators (growth factors) to bone grafts (autologous, heterologous, alloplastic). Many of these substances have enjoyed only limited success. The most likely reason (for limited success) is that we have not yet determined the optimal therapeutic conditions nor the proper sequence of stimulatory effectors. In fact, because of the complexity of cellular activity, it may be some time before we can appreciate all of the essential elements of osteogenesis and its potential for regeneration.
Historically, the ability of bone to repair and the use of bone transplants (grafts) to stimulate or augment that repair was long a subject of debate, a debate that centered on whether only osteoblasts or osteogenic type cells were capable of producing new bone (20, 21, 22, 23) or whether less differentiated stem cells could be induced to differentiate into osteoblasts which would cause osteogenesis when transplanted. The pre-osteoblast contenders included chondroblasts, white blood cells, connective tissue cells and cells lining vascular canals. It was Muller, in 1836, who first demonstrated that undifferentiated cells from embryonic connective tissue cells could deposit bone (24). The existence of progenitor stem cells with the capacity to become osteoblasts and form bone provides a substantially greater potential for regeneration than the limited number of terminally differentiated osteoblasts existing at the wound site could possibly offer. The question that remains is: What are the optimal-conditions for progenitor cell stimulation at the wound site?
Many strides have been made toward achieving periodontal tissue regeneration. Recent advances have led to a greater understanding of the requirements for and the possibility of regenerative periodontal therapy. The concept of guided tissue regeneration (GTR) has been employed to successfully prevent epithelium and gingival connective tissues from migrating into and occupying the space where we would like to have bone growth (25, 26, 27, 28). Since significant advances have been made with soft tissue regeneration, bone regeneration appears to be a critically limiting factor in our ability to regenerate periodontal (peri-implant) osseous tissues. Bone regeneration remains the most desirable yet least predictable (controllable) aspect of regenerative wound healing.
Guided tissue regeneration of periodontal tissues has received great interest and support for the keatment of periodontal pockets and bony defects. Current theory holds that space maintenance and clot stabilization are important to the regeneration of periodontal tissues (especially bone). GTR has been shown to enhance new connective tissue attachment to root surfaces and to increase the amount of bone growth in periodontal bony defects by preventing down-growth of gingival epithelium and allowing the proliferation of the more desirable progenitor cells (namely bone and periodontal ligament). This method of treatment is supported by the following:
2. Soft Tissue Wound Healing
Soft tissue wounds resulting from kaumatic, pathologic or surgical injury to living tissues represent an anatomic or functional disruption in the continuity of the tissue. The healing process is a complex and integrated sequence of events initiated by the stimulus of injury. Healing of the wound requires a well orchestrated coordination between many specialized cell types to restore structural and functional integrity.
The disrupted epithelium responds, within hours, to injury by dedifferentiation, proliferation and migration across the wound surface. Chemical mediators and a lack of contact inhibition stimulate epithelial cells to lose their firm attachment to the underlying connective tissue and neighboring cells (30). The process of epithelial migration continues until complete coverage over the exposed connective tissue is achieved. Once a single layer of epithelium covers the entire wound, the epithelial cells begin to form multiple layers by mitotic division.
At the time of injury, blood vessels are disrupted causing local hemorrhage. Hemostatic mechanisms are initiated to establish a dot with the aggregation of platelets and the deposition of fibrin. Thus, the initial wound matrix is composed of crosslinked fibrin strands which provide a scaffold for the influx of cells. An inflammatory response follows the release of polypeptide growth factors from platelets and injured cells into the wound site. The inflammatory response may also be initiated and perpetuated by a bacterial contamination or infection of the wound.
The acute phase of inflammation is characterized by an influx of polymorphonuclear neutrophils (PMNs) into the wound site within 12-24 hours. These cells provide antibacterial activity and initial wound debridement. They do not appear to be necessary for wound healing because neutropenic patients heal normally in the absence of infection (31). The acute inflammatory phase is followed by chronic inflammation predominated by lymphocytes and macrophages. These cells, especially the macrophage, direct the subsequent migration of fibroblasts into the wound by releasing chemotactic polypeptide growth factors. The fibrin extracellular matrix is degraded and replaced by proteoglycans forming the next provisional wound matrix.
Granulation tissue formation begins in the wound site within 3-4 days and consists of a dense population of macrophages and lymphocytes. The tissue is highly vascular and loose in its connective tissue organization which facilitates the migration of fibroblasts into the wound. Fibroblasts are critical for the production of collagen, elastin, fibronectin, glycosaminoglycans (GAGs) and enzymes that makeup granulation tissue (32). Endothelial cells simultaneously migrate into the granulation tissue establishing neovascularization for the delivery of oxygen and nutrients and the removal of debris and metabolic waste.
During the early phases of granulation tissue formation, hyaluronic acid (one of the GAGs characterized by repeating disaccharide units of glucouronic acid and hexosamine) is a predominant component of the provisional wound extracellular matrix. Recent evidence suggests that hyaluronic acid plays a significant role in the differentiation, migration and proliferation of cells during the wound healing process (33).
The next phase of wound healing is marked by an increase in the number of cells in the wound site as a result of proliferation of those cells that occupy the wound. Hyaluronidase activity increases to remove the hyaluronate wound matrix. Collagen production increases and replaces the proteoglycan extracellular matrix. The final phase of wound healing occurs over several months with the synthesis, crosslinking and remodeling of collagen. A mature collagen scar results with tensile strength approaching that of normal tissue.
3. Hard Tissue Wound Healing
Like soft tissue wound healing, the healing process of bone fractures or defects is a complex and integrated sequence of events initiated by the stimulus of injury. Healing of a bone injury, like soft tissue, requires an integrated coordination between specialized cell types that work to restore structural and functional integrity. Specifically, three principle cells, osteoblasts, osteocytes and osteoclasts are involved in bone formation and remodeling.
Simply stated, the sequence of events in the repair of bone are similar to that of soft tissue repair except calcification of the bone wound matrix occurs. Injury to the bone causes a break in blood vessels of the periosteum, endosteurn and haversian canals. Hemorrhage activates platelets and a clot is formed with fibrin deposition. The resulting hematoma fills the osseous wound space and serves as the initial matrix or framework for the ingrowth of cells and new blood vessels. The clot organizes into a soft callus that extends beyond the injury for anchorage but does not provide structural rigidity. An inflammatory response ensues and stimulates the proliferation and migration of cells into the wound matrix producing granulation tissue. As the wound matures, a provisional hard tissue callus gradually replaces the granulation tissue. The wound matrix is irregular and quickly formed. The callus undergoes mineralization to form woven bone.
The immature bone is subsequently replaced during the remodeling phase by lamellar bone consisting of regularly arranged collagen bundles. Remodeling continues via osteoclast resorption of immature woven bone and osteoblast synthesis of osteoid which mineralizes to form dense lamellar bone. The remodeling process culminates in an consistent lamellar pattern due to the organized manner of resorption and replacement of immature bone.
Bone healing, unlike that of other tissues, does not result in scarring unless signiffcant displacement of structure occurred. Even in the event of significant displacement, bone continues to remodel until demarcations and excessive protuberances are removed. It appears that the extensive remodeling process and perhaps the calciffcation process decreases the likelihood of bone scarring. In the absence of complications, such as infection, inadequate blood supply, interposition of soft tissue or excessive movement, regeneration of bone within the confines of viable bone structure is possible and predictable.
Bone grafts have been used successfully to improve healing of bone fractures and defects. These have been particularly useful in the craniofacial region to assist bone healing, augment bone defects and alter facial contours. The healing principles of bone grafts are similar to the wound healing described with some differences.
Osteogenesis and osteoconduction are the primary mechanisms by which bone grafts heal. Osteogenesis is the formation of new bone by surviving cells within the graft. Grafts must be well vascularized in order for bone forming cells within the graft to survive. Osteoconduction is the process by which blood vessels and cells grow into the graft and use it as a scaffold or matrix to form new bone. The process whereby dead bone is resorbed and replaced by new bone is referred to as "creeping substitution" by some investigators.
Osteoinduction is the process of transformation of local undifferentiated cells into bone forming cells. The most notable example of an osteoinductive substance is the family of bone rnorphogenic protein or BMPs. These proteins or growth factors are released from the resorbing bone matrix and stimulate mesenchymal stem cells to differentiate and form bone (34).
C. Wound Healing: Repair versus Regeneration
Normal adult wound healing, as described under soft tissue wound healing above, is considered the norm and most often results in repair or scar formation as opposed to regeneration of the structure and function of damaged tissues. Adult mammals have a very limited capacity for tissue regeneration. True regeneration of wounded tissues in the adult appears to be limited by fibroplasia or scar formation as a result of collagen deposition and remodeling. Minor success in regenerating the distal tips of severed fingers in young individuals by repeated removal of the wound surface epithelium has been reported (35). Similar findings have been reported in studies with mice (36). These reports suggest that some potential for regeneration exists and encourages further research in this area. In the absence of wound manipulation, the normal adult wound healing process impairs the potential for tissue regeneration through its deposition of collagen.
Some lower vertebrates are quite adept at regenerating injured tissue as exemplified by the newt which can completely regenerate an amputated limb. Following amputation of a limb, an epithelial cap forms under which the cells dedifferentiate and form a "blastema." This is followed by cellular proliferation, migration and differentiation. The formation of the blastema is correlated with a hyaluronate rich, collagen poor extracellular matrix (37, 38, 39). It appears from these studies that increased hyaluronate content in the extracellular matrix is associated with cellular dedifferentiation, proliferation migration and ultimately regeneration.
Fetal wound healing differs markedly from adult wound healing in several ways. Fetal wounds typically heal faster and regenerate the structure and function of injured tissues rather than repairing them with scar tissue. The most significant findings in studies of fetal wound healing are the absence of acute inflammation, minimal fibroblast proliferation, a matrix rich in hyaluronic acid and very little deposition of collagen (40). As compared to adult wounds, the composition of fetal wounds dramatically increase in hyaluronate concentration while decreasing in collagen content. These findings suggest that a hyaluronic acid wound matrix facilitates normal tissue regeneration whereas collagen deposition hinders normal tissue regeneration and leads to scar formation.
Certain invertebrates retain the capacity to regenerate limbs that have been amputated. Autotomy and regeneration crustacean limbs are two processes that are closely linked, in that autotomy is the process by which a limb can be lost with minimal damage to the crab and regeneration is the process by which the appendage can be re-grown (41). Chitin plays a pivotal role in these processes. In the course of regeneration, a flexible limb bud is developed from the breaking plane. It consists of a flexible chitinous sac, which contains a small regenerated limb, folded twice upon itself. At ecdysis the chitinous sac is shed with the cast integument, and the regenerated limb straightens and becomes functional though still smaller than before its loss. The full size is regained at succeeding molts. This process of regeneration following autotomy is described in detail by Skinner (1985). The presence of chitin surrounding the limb bud as it develops suggests that the chitin content of the wound matrix plays a pivotal role in the process of regenerating the lost appendage.
The regenerative potential of invertebrates, lower vertebrates and fetal wounds suggests a mechanism or chemical mediating substance present in those wounds yet absent from adult wounds which have little capacity for regeneration. One possible candidate is the hyaluronic acid (or chitin) that increases dramatically in the extracellular matrix of lower vertebrates and fetal wounds (or invertebrates). In contrast to the increased hyaluronic acid (or chitin) content observed in those wounds, adults wounds reveal only a slight increase in hyaluronic acid content for a few days followed by a dramatic decrease in hyaluronic acid. It is felt that the increased content of hyaluronic acid in the wound matrix is conducive to cellular movement because its large mass blocks cell to cell contacts and so prevents contact inhibition.
D. Biochemistry of Chitin and Chitosan
Chitin is second, only to cellulose, as the most abundant natural biopolymer. It is an important structural component of the exoskeleton of invertebrates (e.g. shrimp, crab, lobster), cell wall of fungi and the cuticle of insects. The very stable polysaccharide is a linear polymer of N-acetyl-D-glucosamine units joined in 1,4 b glucosidic linkages, the minimum descriptive unit being the disaccharide chitobiose. Chitin bears a close resemblance to cellulose, the major structural polysaccharide of plants which consists of D-glucose chains in 1,4 b linkages. Chitin, like cellulose, adopts a highly ordered chain conformation which gives rise to characteristic x-ray diffraction patterns (42). It was first prepared, at least in a concentrated form, by Braconnot in 1811 by the removal of the alkali - soluble material derived from some higher fungi (42).
Chitosan is a derivative of chitin made by treating it with hot strong alkali. Analysis of this process has shown abundant deacetylation of side chains, which is manifest by a decrease in CO-OH bonds and an increase in OH-HO bonds. The resultant chitosan (1-4, 2-amino-2-deoxy-b-D-glucan) is a polycationic complex carbohydrate. Biodegradable and non-toxic, chitosan has a molecular weight of 800-1,500 Kd. Chitosan's availability in a variety of useful forms and its unique chemical and biological properties make it a very attractive biomaterial.
Chitin, which is processed to form chitosan, is most abundantly found in arthropod exoskeletons, fungi and plant cell walls. It is degraded via chitinase or lysozyme digestion. The degradation product is glucosamine which is a natural monosaccharide that can be used by mammalian cells as an energy source.
Chitosan's ability to be made into gels, films, membranes, fibers and beads as well as powders, flakes or solutions has led to many commercial and biomedical applications. This novel biomolecule may have many applications in dentistry and medicine due to its potential to effect the blood- tissue interface.
E. Biomedical Applications of Chitosan
Chitosan, with a chemical structure similar to hyaluronate, has been implicated as a wound healing agent in mammals. (Figure 1) The beneficial role of chitin and its derivative, chitosan, in potentiating wound healing in mammals have been surmised since antiquity and studied extensively in the last 25 years. It has also been reported to function as a unique hemostatic agent. Although its mechanism of action is not known, chitosan is thought to play a role in cellular migration and tissue organization during the wound healing process. Research experiments using chitosan in various animal models have demonstrated its ability to improve hemostasis, decrease fibroplasia, facilitate osteogenesis and enhance tissue regeneration.
Studies have also reported the use of chitosan as a suture material, (43, 44) a wound dressing and an adjunct to drug delivery. Allen, et. al., reported effective burn treatments in rats using chitosan acetate (45). Tough protective chitosan acetate films had the healing advantages of good oxygen permeability, high water absorptivity and slow enzymatic (lysozyme) degradation, thus avoiding the need for repeated removal. A partially deacetylated chitin (chitosan) was found to be a good substrate for lysozyme and to lend itself to applications as a biodegradable material. Specifically, chitosan has been used as a biodegradable carrier for delayed release of pharmaceuticals (46).
Chitosan is a hemostatic substance that can be applied to open wounds in surgery. Chitosan forms a coagulum in contact with defibrinated blood, heparinized blood and washed red blood cells (47). Several studies have shown chitosan's safety and efficacy as a topical hemostatic and wound healing agent in various animal models (47, 48, 49, 50, 51, 52). It achieves hemostasis independent of normal dotting mechanisms or platelets via cellular aggregation or clogging (47, 49, 50, 51, 52). This high molecular weight, polycationic biomolecule appears to adhere red blood cell membranes together (50, 51, 52). When knitted DeBakey vascular grafts were treated with Chitosan, they were impermeable to blood. Examination of these grafts at 24 hours revealed no re-bleeding. Animals with control vascular grafts exsanguinated (53).
The scientific basis for the utility of the monomer sugar N--acetylglucosamine in the promotion of wound healing was documented by Reynolds, et. al. in 1960 (54). By the 1970s, N-acetylglucosamine enhanced rates of wound healing were reported by Balassa, Prudden and coworkers and became the subject of several US Patents (55, 56, 57). Chitosan which is easier to use was also studied and found to be effective in wound healing stimulation by Balassa and Prudden (58).
The effect of chitosan on wound healing has been evaluated by several investigators using various animal models. The wound healing acceleration properties of chitin and chitosan are well assessed in the literature (59). Minami, et. al. noted accelerated regeneration of tissue with no visible scarring in treating various types of infected livestock wounds with chitin/polyester non-woven dressings, chitin/cotton and chitosan-cotton wound filling materials (60). The degree of acceleration of the wound healing process was determined in animal tests by measuring the tensile strength of the newly formed tissue of the wound. Similar results were observed with domestic pets utilizing the same materials. Brzeski, et. al. reported 99% healing effectiveness in livestock postoperative and traumatic wounds, new and old (non-healing) and >90% healing of infectious hoof inflammation using a topical chitosan spray. Clinical results using the same product on human chronic leg ulcers have shown inaeased granulation and epidermis formation and accelerated healing (61).
Generally, chitosan has been proposed to enhance wound healing by inhibiting fibroplasia and promoting organized tissue reconstruction. Chitosan applied to abdominal skin and subcutaneous incisions resulted in healing without fibroplasia or scarring (Malette W.G., et. al., 1986). Wound healing with chitosan proceeds in a manner that reconstructs tissues in normal architecture without fibroplasia or scarring (47, 62). Muzzarelli, et. al. used chitosan and chitosan ascorbate to replace dura mater in cats and reported complete polymer disappearance and normal tissue regeneration in 60 days. Malette, et. al. showed that treating various dog tissues with chitosan solution resulted in the inhibition of fibroplasia (scarring) and enhanced tissue regeneration (53).
From a tissue regeneration perspective, it is particularly interesting to note that Van der Lei, et al. found that chitosan treated Goretex³ vascular grafts formed a unique blood clot layer that enhanced the proliferation of endothelial cells (progenitor cells from the lumen of an arterial graft) as compared to the untreated Goretex³ graft (63). Unlike the typical clot with platelets and fibrin, the unique blood clot formed by chitosan consisted of a homogeneous layer of red blood cells that promoted the propagation of endothelium and smooth muscle over the clot layer. The chitosan treated Goretex³ formed a clot layer that completely covered the surface of the graft while the untreated Goretex³ graft material was not covered by any form of a clot. Examination at one, two, three and four months showed the chitosan treated grafts to be encased in smooth muscle with a living endothelial cell lining (47). As is well known and expected, the untreated grafts were well tolerated but surrounded by fibrous connective tissue (i.e. scar). It is undetermined whether the regeneration of tissues lining the graft were induced by the presence of chitosan or the layer of red blood cells aggregated by chitosan. In either case, chitosan appears to have prevented fibroplasia while enhancing the regeneration of normal tissues by progenitor cells present at the wound margin.
The effect of chitosan on bone wound healing has been examined. Wounds were made through the cortex and into the marrow of dog radii. Saline treated wounds healed by typical osteoblastic-osteoclastic sequence with callus formation while chitosan treated wounds healed by direct cortical formation without a callus (64). Malette, et. al. presented early evidence of enhanced leg bone regeneration in dogs using chitosan (65). Borah, et. al. demonstrated actual bone growth induction by N-acetyl chitosan in large metacarpal - fibular defect areas in a rabbit model (66). This healing was sufficiently vigorous to eliminate the need for surgical bone grafting (67).
From their work with regeneration of various types of animal tissue, Muzzarelli, et. al. indicated that chitosan derivatives stimulate the regrowth of bone tissue and could be a useful reconstructive aIternative (68). Recent research has shown that methylpyrrolidinone chitosan with the bioactivity of chitosan and the lysozyme hydrolytic susceptibility of pyrrolidinone, is a potent osteoconductive agent in dental surgical and bone defect repair applications (68). It is postulated that this highly hydrophilically modified chitosan, with its N-acetylglucosamine units so similar to the biologic glycosaminoglycans, binds fibroblast growth factors stimulating angiogenesis and osteoblast-like cell proliferation (68).
Some oral wound healing applications of chitosan have been reported in humans. Sapelli, et. al. reported in 1986 on the dental applications of chitosan powders in promoting healing of periodontal pockets and large intrabony defects resulting from tooth extraction (69). Muzzarelli, et. al. (1989) applied chitosan to oral wound healing in humans and found beneficial results. They applied chitosan to periodontal wounds and found that it decreased fibroplasia and enhanced cell proliferation and tissue organization of periodontal soft tissues. Muzzarelli, et. al. reported on the use of chitosan ascorbate gel to enhance reconstruction of periodontal tissue in humans, showing that it enhanced the normal cell proliferation and organization of reconstructed tissue (70). Muzzarelli, et. al. (1993) also applied chitosan to jaw bone defects (extraction sites and apicoectomy sites) and concluded, by radiography and biopsy, that it enhanced normal bone formation. Unfortunately, there were no controls in their study.
Chitosan has been used in combination with other materials to enhance bone growth. Ito (71) developed a self hardening mixture of chitosan and hydroxyapatite and proposed its use as a bone filling paste for the treatment of periodontal defects or augmentation of edentulous ridges. Kawakami, et. al. (72) found that the chitosan-bonded hydroxyapatite paste enhanced osteoconduction when applied to the tibia of adult Japanese white rabbits as compared to controls.
Chitosan also accelerates the growth of tissue in culture and promotes tissue growth in a multilevel configuration rather than in a monolayer. Malette, et. al. (1986) have had interesting results from their tissue culture studies. Fibroblasts grown without chitosan are spindle shaped and have long, extending processes. But, fibroblasts grown in the presence of chitosan did not adhere well and displayed many short or exploratory processes (65). Ultimately, the cells rounded up and died. Unlike fibroblasts, endothelial cells and myocardial cells flourished in chitosan. Endothelial cells grown with chitosan grew faster and assumed configuration faster than cultures without chitosan (65). Chitosan treated cardiac myocytes formed three dimensional, synchronously beating masses in 48 hours (65). Myocytes grown without chitosan proliferated slower and only formed a monolayer. All of these cell cultures with chitosan resisted contamination with mycoplasma.
F. Specific Aims:
The objective of this project was to evaluate the effect of chitosan (poly-N-acetyl glucosaminoglycan) on the differentiation, proliferation and growth of osteoprogenitors in cell culture. The specific aims of this study were to:
The hypothesis tested was directly related to the specific aims of this research project. Specifically, that chitosan enhances the differentiation of mesenchymal stem cells into bone forming colonies and that chitosan increases the amount of bone formed by bone colonies.
No previous studies have been done which specifically relates to an investigation of the effect of chitosan (poly-N-acetyl glucosaminoglycan) on bone colony formation by mesenchymal stem cells. . Having a structure similar to glycosaminoglycans, chitosan may serve as a substrate for osteoprogenitor cells to form bone colonies or a scaffolding for bone nodule construction. This project is based on that premise and focuses on the effect of poly-N-acetyl glucosaminoglycan (chitosan) on bone colony formation by cells obtained from fetal mouse calvaria. The results will be interpreted and used to assess chitosan's potential as a therapeutic moiety for periodontal, peri-implant or alveolar bone regeneration.
A. Chitosan
Chitosan (Hoechst lloussel Phamaceuticals, Inc., Somerville, N.J.) was obtained in a clear, slightly viscous solution. Each milliliter contained 2 mg chitosan in 0.2% acetic acid solution. Thus, solutions consisted of a 0.2% chitosan in 0.2% acetic acid. Vehicle (control) solutions were obtained which consisted of the same concentration of acetic acid (0.2% acetic acid) but did not contain chitosan. Prior to use solutions were stored at room temperature.
Chitosan is a partially deacetylated, predominately unbranched, high molecular weight (range 800-1,500 Kd) polyglucosamine biopolymer. The chitosan obtained in solution was processed from the exoskeleton of crabs. It is slightly hydrophobic (except in pH ranges below 5.0) which is the reason for using an acetic acid vehicle solution. When in solution, chitosan has a- slight positive charge.
B. Preliminary Studies (preparation of Culture Wells)
Prior to this experiment, mesenchymal stem cells had not been cultured in the presence of chitosan. The relative insolubility of chitosan in a non-acidic pH solution posed a potential problem for its use in tissue culture. Thus, some preliminary testing was necessary to establish the feasibility of successfully running this experiment. The first question that needed to be answered prior to initiating the proposed experiment was to determine the method of introducing chitosan into the cell cultures. The second question that needed to be answered was whether mesenchymal stem cells could survive in the presence of chitosan. Previous testing with gingival fibroblasts yielded very poor growth (73). If the same were true for mesenchymal stem cells, no comparisons would be possible. Both of these important preliminary concerns were tested and the respective questions were answered in the following manner:
Since chitosan would not remain in solution under normal pH levels and since chitosan was thought to provide a substrate for tissue organization and growth, the chitosan solution was applied to the bottom of the culture wells and the acetic acid vehicle was evaporated prior to seeding the mesenchymal stem cells. As a result, chitosan was introduced to the system as a thin, dry coating on the bottom of the well without the acetic acid. This method of substance application to the bottom of the culture plate is commonly used by research technicians as well as commercial companies to provide a chemical substrate for cell culture.
Serial dilutions of poly-N-acetyl glucosaminoglycan (chitosan) were used to determine the ideal (and critical) concentration of chitosan for growth and proliferation of mesenchymal stem cells (osteoprogenitor cells). Chitosan (2 mg/ml in 0.2% acetic acid; MW 800-1500 kd) was plated on 35 mm Petri dishes in a laminar flow hood using sterile techniques. The amount of chitosan solution applied to the test culture plates ranged from one milliliter (2mg chitosan) down to 0.005 milliliter (1mg chitosan). (Table 1) The coated Petri dishes were allowed to sit in the laminar flow hood without lids under UV light until the acetic acid was completely evaporated. As a control, the vehicle solution (0.2% acetic acid without chitosan) was tested in the same fashion. Untreated plates were also used as controls. Thus, preliminary test plates consisted of those prepared with various amounts of chitosan and vehicle coatings and those without any pre-treatment.
Mesenchymal stem cells were obtained in the manner as described by Marvaso and Bernard (74) and seeded on preliminary test plates. The growth media was monitored for color changes to guard against acidic pH during all phases of the experiment. Cell growth was evaluated daily to observe differences in chitosan versus control plates as well as to detect problems with cell growth in the chitosan treated plates. Following a 14 day period of cell growth under optimal conditions, plates were fixed, stained and embedded as described below. Cell growth and bone colony formation was evaluated at each concentration. An optimal concentration of chitosan that permitted growth and promoted osteogenesis was selected for an experimental range to be tested.
Having established the optimal concentration that the stem cells can grow and proliferate in the presence of poly-N-acetyl glucosaminoglycan (chitosan), the following experimental tests were carried out.
C. Animals
Bernard and Pease (1969) reported that calcification begins in mouse calvaria during the fourteenth day in utero. In order to be sure that calcification had not begun, Swiss Webster mice which were pregnant 12-13 days (Figure 2) were used to obtain mesenchymal tissue (75). A total of six runs, consisting of 8 to 10 six-well culture plates (each run), were used to determine the effect of chitosan on osteogenesis. Every six-well culture plate consisted of alternating chitosan and control treated wells to insure an even distribution. Each culture well was seeded with 350,000 cells. Six time-pregnant (12-13 day) mice were used for each run. Each time-pregnant mouse produced 10 to 12 fetuses and yielded 3 to 5 million mesenchymal cells from the collected calvaria. This experiment utilized 36 time-pregnant mice (6 time-pregnant mice for each of 6 runs) to determine the effect of chitosan on osteogenesis.
Pregnant (12-13 days) Swiss Webster mice weighing approximately 25 grams were used as the source of fetal mouse calvaria. The mice were anesthetized and kept alive until all fetuses were harvested.
D. Surgical Procedures
Each time-pregnant (Simonsen) Swiss Webster mouse was anesthetized with 0.4 cc (50mg/ml) sodium Nembutal via intra-peritoneal injection. The abdomen was shaved and prepped with bedadine. A single vertical incision was carefully made through the abdominal wall to expose the placenta without damaging abdominal contents. Fetuses were aseptically extracted. Calvaria were dissected from epithelium and brain tissues and placed into medium (Figure 3a)(Figure 3b)(Figure 4a). The tissues were kept cool on ice throughout the collection procedures.
E. Mesenchymal Stem Cells (Osteoprogenitors)
Mesenchymal stem cells with osteogenic potential were used in this study to evaluate the influence of chitosan on osteoprogenitors. The method utilized to obtain osteoprogenitor cells was developed by Bernard and Marvaso (74) and has been well used by Bellows and others (76, 77, 78). Bernard and Pease (1969) and Marvaso and Bernard (1977) demonstrated that mesenchymal cells obtained prior to 14 days (in utero development) contained stem cells with the potential to become osteoblasts and form bone colonies (74, 75). This method established that cells isolated by enzymatic digestion from fetal mouse calvaria can be used as a unit colony assay for bone nodule-forming cells and thus can be directly applied to studies on osteoprogenitor cell differentiation. These previous studies have shown that osteoprogenitor cells, while low in number, can be consistently cultured from calvarial populations. Furthermore, it has been shown that some osteoprogenitor cell differentiation can be influenced by exogenous agents (77, 78) (e.g. glucocorticoid hormones). This effect can be measured by in vitro assay.
F. Preparation of Cells for Tissue Culture
The protocol designed by Marvaso and Bernard (74) uses enzyme digestion to release undifferentiated mesenchymal cells with the potential to become osteoprogenitor cells from the calvaria of fetal mice (Figure 4b). Mesenchymal stem cells derived from fetal mouse calvaria were utilized because they have the potential to become osteoblasts that form bone in culture. The influence of exogenously introduced substances (e.g. chitosan) can then be tested in cell culture for their effect on these stem cells.
The tissue obtained from several fetuses was pooled and transferred to Hank's balanced salt solution (HBSS) without calcium or magnesium. The pooled mesenchymal tissue was mechanically cut into small pieces (lxlmm2) and digested with collagenase in dextrose for 90 minutes at 37C. Fifty milligrams of crude collagenase and 25 milligrams of dextrose were added to 10 milliliters of Hank's buffered salt solution. The digestion solution was filter sterilized through a 0.2 mfilter.
G. Tissue Culture
Digested tissue was centrifuged, washed and re-suspended in BGJb medium (Gibco Laboratories, Life Technologies, Inc., Grand Island, New York 14073) supplemented with 0.1% L-glutamine, 0.1% penicillin-streptomycin, 0.1% fungizone and 10% heat inactivated fetal calf serum. The cell suspension was counted with a haemocytometer. Aliquots containing 350,000 cells were seeded in each well of the six-well culture plates (with or without chitosan) and allowed to grow at 37C in 100% humidity with 95% air: 5% C02 for 14 days.
Mesenchymal stem cells with osteogenic potential were harvested from fetal mouse calvaria (12-13 days in utero) via sterile dissection followed by enzyme digestion and seeded onto six-well culture plates with or without the presence of chitosan. The mesenchymal stem cells were cultured under optimal conditions for fourteen days before fixation, staining and embedment (74).
H. Fixation, Staining and Embedding
Tissue growth and colony formation was monitored daily under phase contrast microscopy. After fourteen days, growth of bone colonies was stopped by fixation. As described by Marvaso and Bernard (1977), bone cultures were fixed in 2.5% glutaraldehyde in 0.1M sodium cacodylate buffer at pH 7.4 for two hours. They were washed twice with 0.1M sodium cacodylate buffer and stained with Von Kossa reagents. Then, dehydrated with alcohol and embedded in Epon.
Plastic plates were removed from the Epon embedded tissues and thin sections were cut through representative bone cell colonies. Cross-sections were stained with 1% aqueous toluidine blue in 1% borax. Photomicrographs were taken for analysis.
I. Light Microscopy
After fixation and staining, the bone colonies formed were evaluated, by light microscopy, for quantity, size and morphology. Quantitative and qualitative differences of bone colonies formed in the presence of chitosan were compared to those formed without chitosan (controls). Bone forming colonies (identified by Von Kossa staining) were counted under light microscopy. Representative colonies were sectioned, stained and further examined under light microscopy to confirm the presence of osteoblast type cells and bone formation.
Analysis of test cultures as compared to control cultures were descriptive and quantitative. The descriptive analysis included bone colony size, morphology and density. The quantitative assessment (number and size of colonies formed) of chitosan induced bone colony formation was compared to control colony formation. The statistical analysis was done using a paired student t-test.
J. Image Analysis
Bone forming colonies were analyzed with the aid of computer measurement with an image analysis program (Image 1 Analysis). Images were fed into the computer via a converter from the microscope to the computer (IBM 486). Following microscope calibration, the relative surface area of bone colonies were measured. The statistical analysis was done using a paired student t-test.
A. Preliminary Studies (Effect of Culture Well Preparation)
The method of applying chitosan to the bottom of the plate appeared to be successful with regard to having an effect on the mesenchymal stem cells. In higher concentrations of chitosan, there was little to no growth observed while in the very low concentrations of chitosan, cell growth was moderate (Table 1). Total cell growth (proliferation) was always less in the chitosan treated plates than in the control plates. Cell growth in the culture plates that were coated with the vehicle solution (0.2% acetic acid without chitosan) was similar in all respects to the cell growth observed in the untreated culture plates. Thus, the changes in cell growth observed in the chitosan treated plates was due to the chitosan and not a result of the acetic acid vehicle solution. Furthermore, since cell growth in the vehicle coated plates closely resembled the cell growth in the untreated plates, either vehicle coated or untreated plates could serve as adequate controls for experimental purposes.
Based on cell growth rates relative to the concentration of chitosan in preliminary studies, a concentration of 200 mg chitosan per well was selected and used for all experimental wells in this study. The highest dose of chitosan with the greatest amount of cell growth was selected for this study in order to achieve maximum effect without severely hindering the growth of cells. Lower dosages may not have provided a significant effect. While, higher dosages of chitosan prevented cell attachment and growth which made the evaluation of its effects impossible.
B. Light Microscopy
Examination by light microscopy of control wells revealed a heterogeneous population of cells that achieved confluency within five to seven days. The morphology of these cells was consistent with fibroblasts, endothelial cells and osteoblasts. Although, colonies that formed bone were found randomly distributed around the well, (Figure 5a)(Figure 5b) most of the osseous colonies were concentrated near the center of the well. These control colonies seemed to be fairly consistent in size and shape. Histologic evaluation of a representative (control) bone forming colony in cross section revealed a fibroblast cell layer supporting osteoblasts, osteoid and a calcified matrix (Figure 6). Quantitative analysis revealed 3.6 ± 0.6 bone forming colonies per control well (Table 2).
Examination of chitosan treated wells revealed a heterogeneous population of cells that did not achieve confluency. The morphology of these cells was consistent with fibroblasts, endothelial cells and osteoblasts. However, the number and distribution of cells attached to the well bottom appeared to be limited by the amount and location of chitosan in the well. There were distinct islands of fibroblast attachment and growth with surrounding areas that inhibited fibroblast attachment and growth.
After fourteen days in culture, bone forming colonies were found only in areas where fibroblast type cells were attached. Most of the experimental osseous colonies were concentrated at the periphery of the cellular islands near the chitosan (Figure 7a)(Figure 7b). Chitosan treated colonies appeared to vary in size and shape more than the control colonies (Figure 8a)(Figure 8b). Histologic evaluation of a representative (chitosan) bone forming colony in cross section revealed a fibroblast cell layer supporting osteoblasts, osteoid and a calcified matrix (Figure 9). Quantitative analysis revealed 6.2 ± 1.2 bone forming colonies per chitosan well (Table 2). Although fibroblasts were confined to distinct areas of growth by chitosan, they appeared to begin crossing the barrier and spreading on adjacent areas near the end of 14 days in culture.
The number of cells attached to the bottom of the well appeared to be limited by the amount of chitosan in the well. Fibroblast type cells (required for bone colonies to attach in vitro) formed isolated masses (presumably in areas where chitosan was not present). Initially after seeding (1-5 days), clear round cell blastomer type colonies (osteoblasts are probably included here) were seen attached to all areas of the well. In areas where fibroblasts were not attached, these round blastomer type cells (possibly osteogenic colonies) proliferated three dimensionally on the plate until their size could no longer tolerate the weak attachment to the well bottom and they were washed away (Figure 10a)(Figure 10b)(Figure 11a). Some colonies adhered and proliferated to become very large (Figure 11b). Very few of these three dimensional colonies were retained after multiple washing and fixation (Figure 12a). One very small albeit similar colony that did remain attached stained positive for calcification (Figure 12b). In vitro, bone forming colonies are not able to maintain attachment to the plastic without the assistance of other cells (e.g. fibroblasts). Bone forming cells (osteoblasts) are anchorage dependent and rely on other cells with the ability to attach themselves to the well bottom in culture.
C. Image Analysis
Image 1 Analysis software was used to analyze microscopic images of bone forming colonies captured and converted onto a computer monitor via an image converter. The surface area of the bone colony matrix that was calcified was measured and calibrated on the computer. Following microscope calibration, the relative surface area of calcified matrix within each bone colony were measured.
The average size of the calcified matrix in control colonies was 0.34 ± 0.09 (relative units). The average size of the calcified matrix in chitosan treated colonies was 0.39 ± 0.06 (relative units). (Table 3) Statistical analysis using a paired student t-test did not reveal a statistically significant difference in size between bone forming colonies grown in the presence of chitosan and those cultured under control conditions. However, several chitosan treated bone forming colonies were larger than the largest control treated colonies. Examples of Image 1 Analysis measurements are shown in comparison to the respective light microscopy photomicrographs (Figure 13a)(Figure 13b)(Figure 14a)(Figure 14b)(Figure 15a)(Figure 15b)(Figure 16a)(Figure 16b).
Chitosan, with a chemical structure similar to hyaluronate, has been implicated as a wound healing agent in mammals. Several investigators have reported enhanced wound healing by via inhibition of fibroplasia and promotion of organized tissue reconstruction (47, 53, 62, 63). This research evaluated the effect of chitosan on osteoprogenitor stem cells and osteogenesis. The results of this in vitro study have demonstrated that chitosan nearly doubles the differentiation of mesenchymal stem cells into osteoblasts as compared to controls and the quantity of bone formed by those colonies is congruently doubled. The amount of bone formed per individual chitosan treated bone forming colony was not found to be significantly different than control colonies.
Clearly chitosan had an inhibitory effect on fibroblast cell attachment and proliferation in culture. None of the chitosan treated wells reached confluency whereas all control wells were confluent within seven days. Many clear, round cell blastomer type colonies with morphologic characteristics that are consistent with osteoblasts were observed growing on chitosan treated wells in areas void of fibroblasts. No such cells or colonies were observed in the control wells. Unfortunately, these cells were not retained throughout the culture and fixation period. If these colonies represent osteogenic bone forming colonies, the effect of chitosan on osteogenesis may be substantially greater.
It is possible to speculate from this study that chitosan enhances the formation of bone by bone forming colonies. Although the difference between the average amount of calcified matrix produced per bone colony was not found to be statistically significant, several bone colonies (calcified matrix) in chitosan treated wells were found to produce nearly twice as much bone as bone colonies in control wells. None of the bone forming colonies found in control wells produced as much bone as the larger chitosan treated colonies. An abundance of colonies were similar in size. Thus, the average surface area of calcified matrix was statistically similar. This may be explained by the dominance of early colonies as compared to "late" colonies. The few larger chitosan colonies may have grown faster or started earlier. This difference in size may have been significant if the growth period were longer than two weeks and a greater difference could be appreciated.
The effect of chitosan on bone wound healing has been examined in various models with notable results (64, 65, 68, 70). Malette, et. al. showed that chitosan enhanced leg bone regeneration in dogs (65). Borah, et. al. demonstrated actual bone growth induction by chitosan in large metacarpal/fibular defect areas in a rabbit model (66). Some oral wound healing applications of chitosan have also been reported from human studies. Sapelli, et. al. reported that chitosan powders promoted healing of large intrabony defects resulting from tooth extraction (69).
Chitosan has been used in combination with other materials to enhance bone growth. Ito developed a self hardening mixture of chitosan and hydroxyapatite and proposed its use as a bone filling paste for the treatment of periodontal defects or augmentation of edentulous ridges (71). Kawakami, et. al. (72) found that the chitosan-bonded hydroxyapatite paste enhanced osteoconduction when applied to the tibia of adult Japanese white rabbits as compared to controls. Muzzarelli, et. al. reported on a novel modified chitosan covalently linked to imidazole groups that is more effective than other derivatives in activating bone tissue formation and mineralization (68).
The results of the present study are consistent with reports from other investigators regarding the regenerative potential of chitosan on bone wound healing. The design of this study did not permit an evaluation of chitosan's mechanism of action. Having structural characteristics similar to the glycosaminoglycans found in extracellular matrices of many tissues, it seems that chitosan may mimic their functional behavior. Chitosan could be a primer on which normal tissue architecture is organized. Having a structure similar to hyaluronic acid, a ubiquitous glycosaminoglycan, chitosan may serve as a substrate for osteoprogenitor cells to form bone colonies or a scaffolding for bone nodule construction. Different cell types (fibroblasts versus osteoblasts) may have different responses to various concentrations of hyaluronic acid or analogs such as chitosan. Chitosan may bind with cell surface receptors to stimulate the differentiation of progenitor stem cells into osteoblasts which form bone. On the other hand, chitosan may interfere with the function and interaction of other cell types (e.g. fibroblasts) that inhibit bone colony formation and osteogenesis. Chitosan may bind with fibroblast cell surface receptors and cause an inhibitory effect thus, preventing fibroblast proliferation and fibrous collagen deposition or fibroplasia.
Regardless of the mechanism, chitosan enhances the number of bone forming colonies and the amount of bone formed as compared to controls. The meaning of these results will need to be further investigated with experiments designed to evaluate whether chitosan's effect on osteoprogenitors and osteogenesis is direct (i.e. osteoblast stimulus), indirect (i.e. fibroblast inhibitor), a combination of both or some other mechanism. In any case, it appears from this study that chitosan has a positive effect on bone colony formation and osteogenesis. The results of this in vitro experiment suggests that chitosan (poly-N-acetyl glucosamine) facilitates "selective tissue regeneration" when applied to mesenchymal stem cells. Chitosan selectively inhibits fibroblasts while enhancing the potential for osteoblast differentiation and bone formation.
Chitosan's ability to be made into gels, films, membranes, fibers and beads as well as powders, flakes or solutions has already led to many commercial and biomedical applications. This novel biomolecule may have potential use as an adjunct to periodontal, peri-implant and alveolar bone regeneration. As an example, the biodegradability of chitosan and its ability to be produced in layers offer the potential to be used as a resorbable tissue separating barrier that enhances the differentiation and proliferation of osteoprogenitors while inhibiting fibroplasia. The results of this research support the need for further investigation of this interesting biopolymer in the area of periodontal, peri-implant and alveolar bone regeneration.