Abstract
Tooth vitality and health are related to the presence of a living connective tissue, the dental pulp (DP), in the center of the dental organ. The DP contains the tooth immune defence system that is activated against invading oral cariogenic bacteria during the caries process and the tissue repair/regeneration machinery involved following microorganisms’ eradication. However, penetration of oral bacteria into the DP often leads to complete tissue destruction and colonization of the endodontic space by microorganisms. Classical endodontic therapies consist of disinfecting then sealing the endodontic space with a gutta percha-based material. However, re-infections of the endodontic space by oral bacteria can occur, owing to the lack of tightness of the material. Recent findings suggest that regenerating a fully functional pulp tissue may be an ideal therapeutic solution to maintain a tooth defence system that will detect and help manage future injuries. The objective of this paper was to explain the different revascularization and regeneration strategies that have been proposed to reconstitute a living DP tissue and to discuss the main challenges that have to be resolved to improve these therapeutic strategies.
Keywords
Introduction

Schematic representation of a human tooth.

Strategies for the endodontic treatment of an infected/necrosed tooth. Today, after the removal of the affected dental pulp, the endodontic space is filled with a gutta percha-based inert material, leaving the tooth devitalized and unable to defend itself. Strategies to revitalize the tooth include the revascularization strategy, which consists of evoking bleeding into the endodontic space through the root apical foramen, formation of a blood clot, and replacement of the blood clot with a new tissue built by recruited periodontal stem/progenitor cells. The “cell-free regeneration” strategy consists of injecting, into the endodontic space, a scaffold functionalized with growth factors (GF), in order to recruit periodontal stem/progenitor cells that will populate the scaffold and are expected to differentiate into the various cell types of the DP. The “cell-based regeneration” strategy consists of injecting, into the endodontic space, a combination of scaffold and stem/precursor cells able to differentiate into dentin-forming odontoblasts. In this case, factors responsible for endodontic space angiogenesis will be brought by implanted cells.
The dental pulp (DP) is the specialized loose connective tissue situated in the center of the tooth (Fig. 1). It is responsible for tooth vitality and health, but also for pain sensation, immune defence and tissue repair/regeneration after tooth injury. It is completely surrounded by mineralized dentin except at the root tip where a small apical foramen allows for the entry of blood vessels and nerve fibers into the DP. The external part of dentin is covered at the crown level by enamel, a highly mineralized and tight tissue that protects dentin and pulp from oral injuries, including invasion by oral microorganisms. At the root level, dentin is covered by cementum, a mineralized tissue that is part of the tooth attachment system to the jaw bone. DP may be injured through accidental trauma, excessive wear, or invasion by acidogenic oral bacterial thus leading to a caries lesion [1]. In the latter case, penetration of bacteria through enamel (or cementum) then dentin triggers defence mechanisms within the DP tissue which rapidly develop and ultimately leads, in the absence of bacteria removal by the dental practitioner, to acute irreversible immune/inflammatory events that totally destroy the DP. The tooth can no more defend itself and, to prevent complete contamination of the pulp (or endodontic) space and microorganism dissemination to the periodontal area, the infected/necrosed DP has to be removed by the dental practitioner. The endodontic space is then disinfected and sealed with a gutta percha-based filling material to prevent subsequent recolonization by oral microorganisms (Fig. 2). However, failure of the disinfection process and/or lack of filling tightness often occur owing to the complexity of the endodontic space anatomy and the presence of thousands of intradentinal tubules connected with the DP. Both can lead to the re-infection of the endodontic space and to related immune/inflammatory events in the periodontal area in contact with the root apical foramen [2,3]. Thus, a risk of bacterial dissemination throughout the body always remains after classical endodontic treatments and justify the contraindication of such treatments in immunocompromised patients or those with a risk of infective endocarditis. Another drawback of current endodontic therapies is that the gutta percha-based filling material and the sealing cement are more or less tolerated and can trigger periapical chronic inflammations and/or local allergies responsible for chronic pain. But above all the absence of a living tissue in the endodontic space will prevent the sensing, by the pulp immune system, of a future invasion of enamel, cementum and dentin tissues by oral cariogenic bacteria. Therefore, authors have suggested that regenerating a living pulp tissue structurally and functionally identical to the original one to revitalize the tooth, instead of merely filling the endodontic space with an inert material, would be an ideal therapeutic solution. Indeed, it could provide immunosurveillance to the endodontic space and trigger an immune anti-infectious response to microorganisms when needed. Three innovative therapeutic strategies have been proposed to reconstitute a pulp tissue into the endodontic space (Fig. 2) [4]. The first one is a strategy of “revascularization” of the endodontic space. It consists of inducing a bleeding into the emptied endodontic space from the periodontal area situated next the root foramen. This bleeding is expected to fill the endodontic space with blood, with the hope that the formed blood clot will then be rapidly vascularized and replaced with host periodontal stem/precursor cells that will regenerate a fully structured and functional DP tissue. Results have shown that, even if a connective tissue can be formed in the endodontic space, this tissue is not identical to a normal DP and mostly resembles periodontal tissue [5,6]. Especially, the absence of a fully differentiated odontoblast layer at the pulp-dentin interface is very harmful, since dentin can no more be deposited to increase the robustness of the tooth and protect the new DP from external irritants. In addition, the absence of odontoblasts, that constitute the DP early alarm system against external injuries, will prevent the early sensing of these injuries. The second strategy, that we call the “cell-free regeneration” approach, relies on the injection, into the endodontic space, of a scaffold functionalized with active molecules such as growth factors (GF). The latter will be progressively released through the root foramen to attract host periodontal stem/precursor cells that hopefully have the potential to differentiate into DP cells for regenerating a whole DP tissue appropriately vascularized, innervated and able to defend itself [7–9]. The cell-free approach is considered a relatively easy way to regenerate the DP since there is no need to isolate stem/precursor cells from the donor and to expand them in vitro [10–13]. The third strategy is a “cell-based regeneration” approach based on the assumption that the host does not contain, in the periodontal area, cells capable of giving rise to all types of differentiated pulp cells, in particular to odontoblasts, and that stem/precursor cells have to be added to the biomaterial before injection to subsequently obtain complete DP regeneration. These cells, that could be autologous or allogeneic, will be isolated and expanded according to good manufacturing practices in order to create a cell-based medicinal product [10,14]. The cell-based strategy has already been used in human craniofacial therapy for regenerating bone tissue with stem cells derived from bone marrow, adipose tissue and DP [15,16]. The cell-based approach is nevertheless challenging, owing to the necessity to include, into the biomaterial that plays the role of a scaffold, cells that have to be isolated and manipulated in good manufacturing practice conditions, to avoid the risks of immune response and tumorigenicity brought by the implanted cells [6,17]. The aim of this paper is to review the main challenges of human DP revitalization strategies, focusing primarily on tooth preparation, scaffold properties, growth factors and stem/precursor cells.
One major challenge for DP revascularization and regeneration strategies is the correct vascularization of the endodontic space that can only be achieved by the penetration of a sufficient number of periodontal blood vessels into the endodontic space through the root foramen. Too small a size of the foramen will thus limit or prevent the recruitment of angiogenic periodontal cells, and recent studies have reported that the root foramen must possess a minimal diameter of 0.7 to 1.1 mm to obtain a good vascularization of the endodontic space [14,18–20]. The large size of their root apical foramen has first made developing, immature teeth of young patients a good indication for the use of the revascularization strategy [21]. Appropriate technical procedures to enlarge the much narrower root apical foramen of mature teeth without damaging the apical periodontal tissue are required for applying this strategy to adult patients.
Once the irreversibly damaged DP is removed during the endodontic treatment, the tooth is no more vascularized and protected by the immune system. Microorganisms can thus easily colonize and proliferate into the endodontic space and adjacent dentin tubules [1]. However, DP formation can only occur if the endodontic space is totally disinfected to prevent abscess formation into the induced blood clot (revascularization strategy) or in the scaffold (cell-free or cell-based regeneration strategies) [22]. Disinfection procedures using sodium hypochlorite (NaOCl) are considered the most suitable ones to remove microorganisms from the endodontic space both in classical gutta percha-based endodontic treatments and during revascularization and regeneration approaches. However, denaturation of dentin matrix proteins by NaOCl was shown to prevent the differentiation of odontoblasts along the dentin wall (see below). Therapeutic strategies of human teeth revascularization have also proposed the use of a triple antibiotic paste associating metronidazole, ciprofloxacin and minocycline for two to three weeks before endodontic space filling with gutta-percha [14]. However, several authors have reported tooth discoloration related to the presence of minocycline [23], and they have suggested either to remove minocycline or to substitute it with clindamycin, cephalosporin or amoxicillin [14,24].
Dentin contains many bioactive molecules that are entrapped into the mineralizing matrix during the dentinogenic process and stored in the mature tissue. These molecules can be exposed and released from the tissue upon demineralization, which reactivates them. Among them, GF of the transforming growth factor- beta superfamily have been shown to induce the differentiation of stem/precursor mesenchymal pulp cells into odontoblasts and it is now admitted that chemical exposure of these factors to the dentin surface promotes the differentiation of odontoblasts and the deposition of new dentin layer at the dentin surface [25]. However, NaOCl, which is the gold standard for endodontic space disinfection, denaturates exposed dentin proteins, which prevents their differentiation action on the cells. Several studies have shown that ethylenediaminetetraacetic acid (EDTA) softly demineralizes dentin and exposes dentin-stored bioactive molecules without affecting them, which allows for odontoblast differentiation of stem/precursor cells that contact them [25]. For this reason, recent clinical revascularization protocols propose the alternate use of NaOCl and EDTA, with a final step of dentin treatment with EDTA [5,14,25,26].
Scaffold properties
In tissue regeneration strategies, a biomaterial in the form of a scaffold is specifically designed to promote adhesion, survival, proliferation, migration and/or differentiation of the incorporated cells. Its properties will be adapted in function of the nature of the tissue to regenerate [27]. In a general way, an ideal scaffold should be biocompatible, mimic the extracellular matrix to recreate the natural environment of the cells and sustain the physiological needs of the regenerating tissue. It should also be degraded in a controlled and reproducible manner without the release of cytotoxic by-products, and with a kinetics allowing for its coordinated replacement by the newly formed tissue [24,28]. Scaffold properties such as pore distribution and size are key points for cell attachment, proliferation and differentiation [9]. Unlike revascularization, DP regeneration strategies require the injection of a scaffold containing or not stem/precursor cells into the endodontic space with a fine needle. In the case of the cell-free approach, the scaffold has to be functionalized by incorporating one or several bioactive molecules that will diffuse through the root foramen up to the host periodontal area to attract cells able to give rise to all DP cell populations. In the cell-based approach, such molecules are not indispensable since they can be directly produced by cells incorporated into the scaffold. Within the field of DP regeneration, numerous scaffolds have been proposed without a clear consensus [6,9], and designing new, more suitable ones appear clearly warranted, together with comparative studies with existing scaffolds.
Growth factors
Recruitment of periodontal stem/precursor cells into the endodontic area, as well as their commitment towards the various DP cell lineages, may require the action of inductive GF released by the scaffold or the dentin matrix [29]. Among them, fibroblast growth factor (FGF)-2, vascular endothelial growth factor (VEGF) and platelet-derived growth factor (PDGF) were reported to promote angiogenesis and DP regeneration [11]. Treatment of mesenchymal stem cells (MSCs) with FGF-2 before their implantation enhanced angiogenesis in tissue constructs via VEGF and hepatocyte growth factor release [30]. Similarly, hypoxic treatment of DP-MSCs greatly improved angiogenesis in DP tissue through the increased secretion of angiogenic factors [30–32]. Regarding cell differentiation, bone morphogenetic proteins (BMP)-2 and BMP-4 synergistically promoted odontoblast differentiation and dentin formation [33], whereas the granulocyte-colony stimulating factor induced the mobilization of DP stem cells having high proliferation and differentiation capacities [34].
Cells
Numerous studies have demonstrated that the DP contains several populations of MSCs that can be collected with minimal invasiveness [35–37]. Dental MSCs have also been isolated from the apical papilla (the most apical, embryonic-like pulp tissue in developing teeth), human exfoliated deciduous teeth and periodontal tissues [38]. MSCs from extra-oral sources such as adipose tissue or bone marrow have been also proposed for DP regeneration but, even if they are able to generate a connective pulp-like tissue, it is still not clear if these cells are able to differentiate into odontoblasts (Fig. 3) [39].

In vivo analysis of DP regeneration. (A) Schematic representation of an animal model allowing for endodontic space revitalization. In this example, the endodontic space of a human root dentin fragment was sealed at its cervical tip by Biodentine® cement, and then a combination of adipose tissue MSCs/collagen-based scaffold was injected into the endodontic space through the enlarged root apical foramen (≈1.2 mm diameter). (B) Two weeks after the subcutaneous engraftment of the cellularized dentin fragment into the back of an immunocompromised mouse, a connective tissue was observed in the endodontic space, but no odontoblast layer.
Multiple passages are often necessary to obtain a clinical-scale amount of MSCs, but they are often accompanied by a slow-down of proliferation rate, progressive cell senescence, genomic instability, and loss of multipotentiality that prevents future cell differentiation [40]. Culture medium composition and culture conditions also clearly influence cell phenotype and fate [9,40–42]. For example, FGF-2 and hypoxia priming before grafting have been associated with increased stemness markers’ expression and differentiation potentiality [29,30].
A major concern during tissue regeneration is the possible migration of implanted cells out of the endodontic space and their dissemination throughout the body. It is thus crucial to perform a follow-up to determine the risk of spreading, ectopic localization and possible tumorigenicity of these cells. In a recent study, Souron and collaborators examined the fate of rat DP cells implanted in molars and observed that these cells remained localized inside the endodontic space even after 3 weeks of implantation [43].
Immunomodulation properties
One crucial challenge in DP regeneration is the host immune response against the scaffold and/or implanted cells, which can lead to the destruction and/or the rejection of the grafted tissue by the host immune system [27]. Interestingly, DP-MSCs, like all MSC populations, were reported to be immunosuppressive and anti-inflammatory [44–47]. Even if the engraftment procedure may induce inflammation, moderate inflammation was reported to promote odontoblastic differentiation [1,29]. Understanding of the interplay between DP inflammation and reparation/regeneration will be fundamental to the clinical translation to humans of DP regeneration protocols designed in animal models [48].
Follow-up of the regenerated tissue
Clinical outcomes of new DP tissue formation, like for other endodontic therapies, are traditionally focused on the lack of periodontal inflammation, pain or swelling, in association with radiographic signs of periodontal tissue health [49,50]. However, regenerative strategies should additionally be focused on the functional and histological mimesis of the reconstructed tissue [49,50]. In this context, a positive response to vitality testing could indicate a more organized vital pulp tissue [21].
Conclusions
Over the last decade, substantial progress was achieved in the fields of DP revascularization and regeneration [51]. DP revascularization has been translated into the clinics and is now recommended by the American Association of Endodontists for the treatment of immature teeth [21,48]. However, there are still a lot of challenges to resolve before obtaining a true regeneration of the DP tissue into the human endodontic space. In particular, the impossibility to obtain a well-organized layer of dentin-producing mature odontoblasts lining the dentin wall without the contribution of exogenous competent cells is probably one of the most crucial one [50]. A more complete odontoblast differentiation could be obtained by applying the knowledge of the mechanisms activated during reparative dentinogenesis [9,10]. If cell-free and cell-based regeneration strategies have shown encouraging results in animal models, recruitment of cells from the host is a relevant concept that may offer easier clinical translation, in relation to the many difficulties related to the preparation of human stem/precursor cells in the cell-based approach [14]. One of these difficulties will undoubtedly be the creation of financially affordable GMP facilities to produce clinical-grade cells suitable for clinical application in human [10,24,29,52]. Results from the phase I clinical trial currently under progress in Japan will be very helpful to evaluate the true potentiality of human DP-MSCs in DP-regeneration [34].
Footnotes
Acknowledgements
This work was supported by CNRS, the French Ministry of Higher Education and Research, the French Institute for Odontological Research (IFRO), the “Société Française de Rhumatologie”, and the “Gueules Cassées” Foundation.
Conflict of interest
The authors have no conflict of interest to report.
