Articles
D.P. SARMENT * R. GAPSKI ** O. ANUSAKSATHIEN *** Q.M. JIN **** W.V. GIANNOBILE *****
*Department of Periodontics, Prevention, Geriatrics and Center for Biorestoration of Oral Health,
University of Michigan, Ann Arbor, Michigan, Etats-Unis
**Department of Periodontics, Prevention, Geriatrics,
University of Michigan, Ann Arbor, Michigan, Etats-Unis
***Department of Periodontics, Prevention, Geriatrics and Center for Biorestoration of Oral Health,
University of Michigan, Ann Arbor, Michigan, Etats-Unis
****Department of Periodontics, Prevention, Geriatrics,
University of Michigan, Ann Arbor, Michigan, Etats-Unis
*****Department of Periodontics, Prevention, Geriatrics,
University of Michigan, Ann Arbor, Michigan, Etats-Unis
Wound healing is a multi-step complex process, which aims at repairing the damaged tissue by a scar or a functional organ (fig. 1). All wounds follow a three-step process : inflammation, granulation, and matrix remodeling (Clark, 1988). Periodontal wound healing is unique in the body in that...
Periodontal diseases are characterized by the loss of alveolar bone, periodontal ligament and cementum integrity. While resective procedures were the only treatment of choice until the early 70s, regeneration of supporting structures has been a major aim of periodontal therapy since then. However, predictability of results remains questionable since multiple cell types and a complex microflora affect regenerative outcomes. Although principles of guided tissue regeneration remain valid, better understanding of wound repair has led to the development of novel bioengineering paradigms for potential periodontal applications. The emerging field of tissue engineering combines the life sciences with engineering for the reconstruction of tissues and organs. This approach is being utilized in many areas of medicine and more recently in dentistry.
In this chapter, we layout the goals of bioengineering for periodontal regeneration highlighting recent advances using signaling molecules such as growth factors. This review will conclude with an introduction of emerging applications for periodontal engineering such as cell and gene therapy.
Wound healing is a multi-step complex process, which aims at repairing the damaged tissue by a scar or a functional organ (fig. 1). All wounds follow a three-step process : inflammation, granulation, and matrix remodeling (Clark, 1988). Periodontal wound healing is unique in the body in that nonmineralized and mineralized tissues are in close proximity and compete to reoccupy the tissue defect. In addition, wound margins face a non-vascularized dental root which does not contribute a blood supply. Furthermore, the regeneration of tissues occurs while a diverse periodontal microflora affects the perimucosal interface. Finally, many of the cells, matrix components, and signaling molecules involved in periodontal tissue repair also contribute to the development of periodontal tissues. Thus regeneration is considered a « recapitulation of development ».
For the tissue engineering of periodontal structures, complex processes occur in concert relevant to wound repair and are discussed in this issue : periodontal stem cells (Amar), inflammation (Graves), macromolecules/components of the extracellular matrix (Bartold), root biomodification (Caton), and molecular factors (Aukhil). Polypeptide growth factors play crucial roles in all of these areas and will be discussed in the context of the reengineering of lost periodontal support.
Guided tissue regeneration (GTR) is based on the concept that multiple cell lineages are competing to invade the periodontal wound. Epithelial and connective tissue cells do not permit periodontal regeneration and must be excluded from the compartment in order to selectively allow osteoblasts, PDL cells and cementoblasts to proliferate and reoccupy the tissue defect (Melcher, 1976). A cell occlusive barrier is utilized to achieve such a goal, and must remain long enough for the periodontal attachment to reform. The field of tissue engineering has utilized aspects of GTR for the development of novel biomaterials for cell, protein and gene delivery (Vacanti and Lange, 1999). Polymer scaffolds composed of poly (lactic acid), poly (glycolic acid) and their copolymers have been used extensively in GTR for tissue engineering (Alsberg et al., 2001 ; Kellomaki et al., 2000). These polymers undergo controllable hydrolytic degradation into natural metabolites allowing a variety of porosities for cell and tissue ingrowth as well as serving as a reservoir for growth factors. By designing highly porous scaffolds with a high surface area/volume ratio, mass transport among cells with the scaffold and surrounding host tissue provides an environment conducive to tissue regeneration (Murphy and Mooney, 1999). The use of barriers are important for the attachment and migration of host or transplanted cells with enough mechanical stability to resist cellular contractile forces during tissue neogenesis (Kim et al., 1999) (fig. 2).
Many bioengineered materials under investigation are taking advantage of polypeptide growth factors (GFs). GFs are natural mediators that regulate the proliferation, migration, and synthesis of most all cell types (reviewed in Wikesjö and Selvig, 2000 ; McCauley and Somerman, 1998 ; Anusaksathien et al., 2002). GFs are synthesized in a proform and stored in extracellular granules. After activation, they are released to stimulate the producing cells (autocrine), neighboring cells (paracrine), or distant cells via the blood circulation (endocrine). When binding to a receptor, a complex cascade of signals is transduced targeting the nucleus. These signal transduction mechanisms are reviewed in detail (Anusaksathien et al., 2002). The role of various GFs in periodontal tissue regeneration remains an active area of research. A brief overview of the most well-studied growth factors for periodontal regeneration is described below (fig. 3 and table 1).
Granules of platelets are a source of PDGF but may be produced by many cell types. There are 4 isoforms of PDGF (-A, -B, -C, and -D) although most periodontal studies have investigated -A and -B chains (Boyan et al., 1994). PDGFs exert multiple biological responses, including mitogenesis and chemotaxis of PDL fibroblasts (Matsuda et al., 1992 ; Oates et al., 1993), cementoblasts (Saygin et al., 2000a ; Strayhorn et al., 1999) and osteoblasts (Canalis, 1981).
There is evidence that PDGF has potential for enhancing periodontal wound healing. PDGF receptors are induced in wounded tissues whereas they are not expressed in unwounded epithelia (Antoniades et al., 1991 ; Green et al., 1997). In addition, PDGF has little effect on the differentiation of mesenchymal cells in the periodontal wound while it primarily stimulates periodontal ligament fibroblasts to proliferate on root surfaces (Gamal et Mailhot, 2000 ; Wang et al., 1994). Interestingly, PDGF stimulatory effect is selective for PDL cells, while possessing minimal effects on gingival fibroblasts (Mumford et al., 2001), although these cells do express PDGF receptors (Parkar et al., 2001). A single bolus delivery of PDGF alone or combined with insulinlike growth factor-1 (IGF-1) for a transient period appears to be sufficient to enhance the regenerative process, potentially due the fact that many critical events involved in wound repair occur within the first few days (Lynch et al., 1991). PDGF has also shown positive stimulatory effects on periodontal regeneration in preclinical nonhuman primate models (Giannobile et al., 1994 ; Giannobile et al., 1996 ; Rutherford et al., 1992) and in a multicenter human trial (Howell et al., 1997).
Bone morphogenetic proteins (BMPs) belong to the large superfamily of transforming growth factor ß (TGF ß) proteins (Reddi, 2001). BMPs are powerful regulators of cartilage and bone formation during embryonic development and regeneration in post-natal life. Some BMPs also participate in the development and repair of extraskeletal tissues (Nifuji and Noda, 1999) and organs such as the brain, kidney, and nerves (Thomadakis et al., 1999). A striking and discriminatory feature of some of these proteins is their ability to induce de novo endochondral osteogenesis in ectopic sites (e.g., skin or muscle) (Sakou, 1998). BMP activities are not only modulated through gene expression and protein processing, but by the interaction with antagonists such as noggin (Aspenberg et al., 2001).
Preclinical animal models have shown a potent effect of recombinant human (rh) BMP-2 on bone apposition to implant surfaces, as compared to carriers alone (Cochran et al., 1999 ; Sykaras et al., 2001). Similarly, Sigurdsson et al. (2001) demonstrated extensive alveolar bone augmentation around titanium implants when rhBMP-2 was applied in a collagen sponge. The clinical use of BMP-2 in humans has been recently reviewed (Valentin-Opran et al., 2002). Recombinant human BMP-2 has been safely applied for implant site development (Cochran et al., 2000) and for sinus floor elevation in human trials (Boyne et al., 1997), but quantitative analysis was not reported in these preliminary studies. Using animal models for sinus floor augmentation and simultaneous implant placement, Terheyden et al. (1999) found that bone to implant contact increased by two fold over control and Margolin et al. (1998) found similar increases in bone mineral density, using BMP-7 - or osteogenic protein-1 (OP-1) - compared to the carrier alone.
BMPs have also shown potent effects in stimulating periodontal tissue repair in several experimental animal model systems (Giannobile et al., 1998b ; King et al., 1997 ; Kinoshita et al., 1997 ; Sigurdsson et al., 1995 ; Sigurdsson et al., 1996). In most of these studies the regeneration of bone and cementum were predictably regenerated in large critical size alveolar bone defects. Human trials are being completed to examine the efficacy of these approaches for regeneration of chronic periodontal lesions.
Basic fibroblast growth factor (bFGF or FGF-2) is a member of a heparinbinding family that possesses potent angiogenic properties. FGF-2 is mitogenic and chemotactic for endothelial cells, fibroblasts (Folkman and Klagsbrun, 1987) as well as periodontally-derived cells (Terranova et al., 1989). Amongst other origins, they are synthesized by inflammatory cells and are stored in the extra-cellular matrix by binding to heparan sulfate proteoglycans. FGF-2 has been extensively studied for its role in dermal wound healing both in animals and in human clinical trials (Abraham and Klagsbrun, 1996). More recently, periodontal models reveal a potential benefit of FGF-2 for closure of class III furcations (Rossa et al., 2000 ; Takayama et al., 2001) or for regeneration of intrabony defects (Murakami et al., 1999). To date, no human trials are on-going using FGF-2 for periodontal repair.
A new strategy in promoting periodontal regeneration is to mimic the specific events that occur on the development of supporting tissues during tooth organogenesis. It has been shown that inner cells from the Hertwig's epithelial root sheath (apical extension of the dental organ) have a secretory stage prior to cementum formation, suggesting that epithelial-mesenchymal interactions are essential for formation of the periodontium (Fincham et al., 1995 ; Hakki et al., 2001). Slavkin and Boyde (1975) hypothesized that these cells release enamel-related proteins and that they are involved in the formation of acellular cementum. This concept has been confirmed by others (Araujo and Lindhe, 1998 ; Hammarström, 1997 ; Lindskog, 1982a ; Lindskog, 1982b ; Owens, 1978 ; Schonfeld and Slavkin, 1977). Studies in non-human primates have also demonstrated the formation of cellular cementum in infrabony defects when treated with enamel matrix proteins (EMD) (Sculean et al., 2000). EMD is mainly composed of amelogenins, constituting about 90 % of the matrix (Brookes et al., 1995) while the other 10 % of the « non-amelogenin » group include proline-rich enamel proteins (Fukae and Tanabe, 1987), tuft proteins (Robinson et al., 1975) and tuftelin (Deutsch et al., 1991). It has been demonstrated that a preparation of crude porcine enamel matrix could initiate the formation of a tissue in monkeys that was histologically identical to acellular, extrinsic fibrillar cementum. In the control defects, a poorly attached, hard cellular tissue similar to immature bone was formed (Gestrelius et al., 1997). Therefore, although it is still necessary to further clarify the role of EMD in epithelial-mesenchymal interactions, these proteins have proved to be a potentially useful tool for periodontal regeneration. Initial animal experiments have demonstrated near complete regeneration of acellular cementum and new bone after application of low molecular weight amelogenins. High molecular weight proteins of the enamel matrix produced very little new cementum. Propylene glycol alginate (PGA) also demonstrated to be a suitable vehicle in conjunction with the amelogenin fraction (Gestrelius et al., 1997). The pH-and temperature-dependent viscosity of PGA allows EMD to adsorb to denuded root surfaces and to form insoluble aggregates in vivo (Gestrelius et al., 1997). Regeneration of tissues takes place without down-growth of epithelial cells, suggesting that the biochemical environment may prevent epithelial cells to migrate along the treated root surfaces (Gestrelius et al., 1997). In recent years, several clinical studies have been conducted using EMD for multiple periodontal indications such as treatment of infrabony defects (Bratthall et al., 2001 ; Froum et al., 2001 ; Heijl et al., 1997 ; Okuda et al., 2000 ; Parodi et al., 2000 ; Pietruska et al., 2001 ; Sculean et al., 1999b ; Tonetti et al., 2002 ; Zetterström et al., 1997), in conjunction with GTR (Pontoriero et al., 1999 ; Sculean et al., 1999a ; Sculean et al., 2000 ; Sculean et al., 2001a, Sculean et al., 2001b), in conjunction with bone grafts (Pietruska, 2001), together with gingival curettage (Wennström and Lindhe, 2002) and for root coverage surgeries (Hagewald et al., 2002 ; Modica et al., 2000). Clinical trials comparing guided tissue regeneration (GTR) with EMD have generally found no differences in clinical parameters in the treatment of infrabony defects (Pietruska et al., 2001 ; Pontoriero et al., 1999 ; Sculean et al., 1999c ; Sculean et al., 2001b ; Silvestri et al., 2000). In addition, the combination of GTR with EMD has shown no additional effect in clinical parameters when compared to each component alone (Sculean et al., 2001a). Interestingly, human histology reports have failed to show evidence of a « new attachment » when EMD was used, contradicting early animal studies (Parodi et al., 2000). Sculean et al. (1999a) demonstrated in humans that the formation of new attachment was not always followed by bone regeneration in EMD-treated defects while in GTR cases, the formation of a varying amount of new bone was noted. These data are in agreement with some in vitro studies, where EMD has blocked the mineralization process in murine follicle cells (Hakki et al., 2001). Therefore, although the concept of mimicking the stages of tooth formation for periodontal regeneration has a promising future, the true mechanism of EMD must be clarified, especially when this material is utilized in vivo.
Enhancement of cell attachment through molecular mechanisms is another approach to regenerate periodontal wounds. In tissues, collagen is an abundant extracellular scaffold for cell attachment and migration, and it also modulates cell differentiation and morphogenesis by chemical stimuli and transmission of mechanical forces (Pavlin and Gluhak-Heinrich, 2001). A specific cellbinding domain of type I collagen, composed by a 15 amino acid sequence, has been synthesized (P-15) and utilized in conjunction with bone grafting. P-15 is a very small synthetic fragment of the α1 chain of type I collagen and has been proposed to be uniquely involved in the binding of cells such as fibroblasts and osteoblasts. One gram of anorganic bovine derived HA bone matrix (ABM) in conjunction with 200 ng of P-15 has been shown in vitro to enhance attachment of cells (Seyedin, 1989). Another in vitro study has demonstrated greater numbers periodontal ligament fibroblasts (PDLF) in the presence of ABM/P-15 compared to ABM alone (Bhatnagar et al., 1999). In another study, only the initial attachment of cultured human PDLF to ABM was enhanced by the P-15 peptide (Lallier et al., 2001). However, such combination demonstrated inferior capability of PDLF attachment when compared to bone allografts (Lallier et al., 2001). Therefore, the combination ABM/P-15 seems to enhance cell attachment when compared with ABM alone, but lesser results compared to DFDBA/FDBA has raised questions about its clinical relevance. Interestingly, multi-center trials have evaluated the efficacy of ABM/P-15 in treatment of human osseous defects and the short-term results were significantly better than DFDBA and open debridement (Yukna et al., 1998) or ABM alone (Yukna et al., 2000). It is difficult to conclude if the significantly greater defect fill of this material was due to its « cell attachment » properties or mainly based on the scaffold properties of the ABM. Long-term studies have also been criticized by lacking controls or comparison groups (Yukna et al., 2002). Results to date suggest that ABM/P-15 has potential to augment periodontal regeneration/repair (table 2).
To date, cell therapy has not been employed to any great extent in periodontology. However, when one considers autologous bone grafting for example, it is a form of cell therapy. True, ex vivo approaches have not been used extensively. Recent applications for the use of development and characterization of tissue-engineered human oral mucosa equivalent have been described by Izumi et al. (2000). The approach of ex vivo transfer of cells to scaffolds for propagation and subsequent transplantation into periodontal defects (fig. 4) offers potential for tissue engineering. Others have worked on the use cryopreserved dermal implants for tissue engineering of skin and periodontal soft tissues (reviewed in Naughton and Applegate, 2002 ; McGuire MK, unpublished).
Somerman et al. have also demonstrated the potential usage of cell therapy using cloned cementoblasts (D'Errico et al., 1999 ; Saygin et al., 2000b). In a series of reports, the group has cloned and characterized a cementoblast cell line, which possesses many of the phenotypic characteristics of tooth lining cells in vivo. Early evidence suggests that cementum engineering can be influenced by the proper delivery of cementoblasts via polymer sponges and delivered ectopically or to periodontal defects (Jin et al., 2002a ; Zhao et al., 2001). By utilizing strategies to better understand the basic biologic mechanisms involved in cementogenesis, methods to improve periodontal wound healing may be developed.
A major challenge with current growth factor delivery systems to periodontal tissues is the extremely short duration of growth factor bioactivity in the wounds. The factors remain in the periodontal defect for a short duration, presumably due to proteolytic breakdown, receptormediated endocytosis, and the solubility of the delivery vehicle. Therefore, the use of DNA delivery systems may serve as another method of targeting proteins to a wound site and improving bioavailability of GFs in the defect (Giannobile, 1999).
The transduction of appropriate targandcells represents the critical first step in gene therapy ; consequently, the development of methods of gene transfer suitable for different forms of therapy has been a major focus of research. The single common feature of the methods is the efficient delivery of genes into cells (Mulligan, 1993). In the case of retroviral vectors and adeno-associated viral vectors, the transferred DNA sequences are stably integrated into the chromosomal DNA of the targandcell. Other methods of gene transfer result primarily in the introduction of DNA sequences into the nucleus in an unintegrated form. These methods, which result in high, but transient gene expression, have predominantly been considered for use in vivo gene therapies in which genetic material is directly transferred into cells and tissues of the patient (as opposed to ex vivo therapies requiring transgene expansion from a tissue specimen). Examples of methods of gene delivery for transient expression of genes include adenovirus and DNA-lipid complexes (Kozarsky and Wilson, 1993). Since the regulation of wound repair occurs in a controlled fashion over a short period of time, a goal of gene therapy in a compromised wound such as chronic periodontitis may be accelerated by an elevated and sustained production of GFs. Several groups have delivered GF genes to healing skin (Eming et al., 1999), bone (Fang et al., 1996) and periodontal (Giannobile et al., 1998a) wounds using plasmid DNA. Eriksson et al. (1998) have developed unique methods of transducing wounds by an in vivo microseeding technique which we have adapted to periodontal wounds. Our group has achieved the targeted, long term delivery of PDGF genes to cells derived form the periodontium such as gingival fibroblasts, PDL cells, osteoblasts, and cementoblasts (Giannobile et al., 2001 ; Zhu et al., 2001). In addition, we have shown that PDGF gene transfer sustains PDGF signaling (Chen and Giannobile, 2002) and promotes in vitro gingival wound repair (Anusaksathien et al., 2002) (fig. 5). We have also recently demonstrated that ex vivo gene transfer of skin fibroblasts transduced by adenovirus encoding BMP-7 stimulates chrondrogenesis and osteogenesis in large periodontal bone defects (Jin et al., 2002b).
Future studies using cell and gene therapy will need to closely compare the expression pattern of transduced genes in the promotion of periodontal repair to assess the potential advantages of this approach above that of conventional protein delivery. The use of cell and gene therapy may have possible benefits such as in control of the host response, anti-infective therapy or delivery of angiogenic factors.
While maintaining basic periodontal treatment principles, better understanding of tooth development and wound healing has lead to improvement in regenerative protocols and development of novel biomaterials. This nascent research area of bioengineering tissues and organs by using cells, scaffolds and signaling molecules is greatly transforming regenerative biology. Development of tissue engineering offers an exciting future for periodontology.
The authors appreciate the assistance of Ms. Beverly Sutton for the preparation of the manuscript. These studies were supported by NIH/NIDCR grants DE 11960 and DE 13397 to WVG.
Send reprints requests to
William V. GIANNOBILE : Department of Periodontics, Prevention, Geriatrics - University of Michigan - 1011 N. University Ave - Ann Arbor, MI 48109-1078 - ETATS-UNIS.