Cellular and Molecular Pathways Leading to External Root Resorption

Abstract

External apical root resorption during orthodontic treatment implicates specific molecular pathways that orchestrate nonphysiologic cellular activation. To date, a substantial number of in vitro and in vivo molecular, genomic, and proteomic studies have supplied data that provide new insights into root resorption. Recent mechanisms and developments reviewed here include the role of the cellular component—specifically, the balance of CD68⁺, iNOS⁺ M1- and CD68⁺, CD163⁺ M2-like macrophages associated with root resorption and root surface repair processes linked to the expression of the M1-associated proinflammatory cytokine tumor necrosis factor, inducible nitric oxide synthase, the M1 activator interferon γ, the M2 activator interleukin 4, and M2-associated anti-inflammatory interleukin 10 and arginase I. Insights into the role of mesenchymal dental pulp cells in attenuating dentin resorption in homeostasis are also reviewed. Data on recently deciphered molecular pathways are reviewed at the level of (1) clastic cell adhesion in the external apical root resorption process and the specific role of α/β integrins, osteopontin, and related extracellular matrix proteins; (2) clastic cell fusion and activation by the RANKL/RANK/OPG and ATP-P2RX7-IL1 pathways; and (3) regulatory mechanisms of root resorption repair by cementum at the proteomic and transcriptomic levels.

Introduction

Cellular Components and Adaptations during Root Resorption

External apical root resorption (EARR) following orthodontic treatment is a pathologic side effect of the undesirable activity of specific clastic cells on the root surface that can usually be discerned with routine radiography (Brezniak et al. 2004; Aminoshariae et al. 2016).

Several effector cells are implicated in the root resorption process—specifically, the odontoclasts, the morphologic and functional characteristics of which are extremely similar to those of the osteoclast (Wang and McCauley 2011). Resident periodontal ligament (PDL) and bone marrow–derived circulating mononuclear hematopoietic precursor cells may be directed toward odontoclast differentiation from a semiquiescent state to one of superspecialization, necessary to promote alveolar bone as well as root resorption (Hienz et al. 2015). Cell commitment and mononuclear cell fusion are regulated by E-cadherin, which is essential for the cell-to-cell adhesion that mediates fusion of immature osteoclasts precursors (Segeletz and Hoflack 2016). Activated odontoclasts/osteoclasts attach to the mineral matrix, forming a sealing zone and adopting a polarized morphology with ruffled border secretors of proteases that initiate mineral resorption (Georgess et al. 2014). A V-ATPase pump carries the protons produced by carbonic anhydrase (CA) II to the ruffled border membrane, releasing them into the resorption pit and generating an acidic microenvironment that is completed by chloride transport (Matsumoto et al. 2014). The tartrate-resistant acid phosphatase (TRAP) enzyme is responsible for the elimination of endocytosed material. Finally, the resorption process ends by degrading the organic component. The activity of cathepsins B, E, K, L, S and matrix metalloproteinases (MMPs) 1, 2, 3, 13 hydrolyze the collagen-rich organic bone matrix (Teitelbaum 2007).

As a self-regulatory mechanism, active osteoclasts/odontoclasts progressively increase cytoplasmic calcium, mediated by 2 intracellular calcium-binding proteins: calmodulin and calcitonin. Calcitonin modulates the phosphorylation of protein tyrosine kinase 2 (Pyk2) and proto-oncogene tyrosine-protein kinase Src (Src) and their intracellular organization in the clastic cell (Shyu et al. 2007), which in turn ultimately deactivates the cell’s resorptive capabilities by blocking proton extrusion, decreasing osteopontin expression, and altering the ability of the clastic cell to bind to the podosome. This results in its detachment from the mineral surface (Shyu et al. 2007).

Resorption follows a similar process in bone and dentin (Rumpler et al. 2013). To this respect, microarray analysis of sequential expression of genes in mature active osteoclasts during bone or dentin resorption found greater overexpression of genes associated with cell fusion (CD9, P2X purinergic receptor ligand-gated ion channel 7 [P2RX7], ADAM8, cytoskeleton-related genes such as β-actin, actinin, filamin A, and tubulins [TUB-beta, TUB-alpha, MTMR2]) and genes associated with activity (receptor activator of nuclear factor Kβ [RANK], TRAP, CA II, and ATPase II) in clastic cells cultured on dentin substrate as compared with bone (Rumpler et al. 2013).

In conjunction with this, a recent study reported the critical influence of macrophages on the root resorption process (He, Kou, Yang, et al. 2015). Two distinct in vitro phenotypes are described: classically activated macrophages (M1), or “killer” macrophages, and alternatively activated macrophages (M2), or “healer” macrophages (Novak and Koh 2013). A remarkable plasticity can be observed in the switch from M1 to M2 polarization states depending on the different conditions in the cellular microenvironment (He, Kou, Luo, et al. 2015), which allows them to mediate inflammation and tissue homeostasis. In this respect, increased numbers of the M1 versus M2 cell type is likely to be associated with root resorption (increased M1:M2 ratio). The mechanism explaining this is that M1 macrophages promote inflammation by secreting proinflammatory cytokines (tumor necrosis factor alpha [TNF-α]) with upregulation of nitric oxide production, while M2 macrophages have an inhibitory effect on inflammation mediated by interleukin 10 (IL-10) and arginase I (Hunter et al. 2010; He, Kou, Luo, et al. 2015). Likewise, M1 macrophages are activated by Th1 cytokines, such as interferon γ, while M2 macrophages are regulated by Th2 cytokines (IL-4 or IL-13) (Davies et al. 2013; He, Kou, Luo, et al. 2015). Under prolonged orthodontic force, root resorption lesions were found concomitant with an increase in CD68+ and iNOS-positive M1-like macrophages with upregulation of interferon γ and TNF-α. When the orthodontic force was removed, there was visibly decreased root resorption, with an increased number of CD68+ CD163+ M2-like macrophages, upregulation of IL-4 and anti-inflammatory IL-10 cytokines, and a decreased number of M1-type cells. The severity of root resorption was partly attenuated when a decrease in the ratio of M1:M2 macrophages was detected (He, Kou, Luo, et al. 2015; Fig. 1).

Figure 1.

Schematic illustration of the critical influence of macrophages on the root resorption process. Remarkable plasticity can be observed in the switch from M1 to M2 polarization states, depending on the conditions in the cellular microenvironment, by secreting several proinflammatory cytokines. M1: CD68+, inducible nitric oxide synthase–positive (iNOS+) macrophages. M2: CD68+, CD163+ type 2 macrophages. PDLCs, periodontal ligament cells; IFN, interferon; TNF, tumor necrosis factor; NO, nitric oxid; IL, interleukin; BSP, bone sialoprotein; SOD, superoxide dismutase; OPG, osteoprotegerin; AMB, ameloblastin; DSP, dentin sialoprotein; AMG, amelogenin; OPN, osteopontin; TRAP, tartrate-resistant acid phosphatase; RANKL, receptor activator for nuclear factor κ B ligand.

The biological reaction induced by orthodontic force is not limited to regional cellular recruitment. Instead, the mononuclear phagocyte system in peripheral blood and spleen reservoir monocytes noticeably decreases from days 1 to 3 and then recovers on day 7 after strain application, as shown by flow cytometry (Zeng et al. 2015). Moreover, this decrease in the systemic inflammatory monocytes correlates with an increase in regional monocytes colocalizing with the TRAP+ osteoclasts adjacent to the root surface in the “compressed” PDL.

Molecular Pathways Leading to Root Resorption Process in Orthodontics

Clastic Cell Fusion and Activation: RANKL/RANK/OPG and IL1-Related Pathways

Different cell types and a wide range of cytokines and molecular factors intervene in the maturation stages of odontoclast/osteoclast cell differentiation, inducing greater or less specialization, rate of proliferation, life span, and range of activity (Hayashi et al. 2012; Lee et al. 2015), thereby determining root resorption activity during orthodontics (Fig. 2). Cytokine regulation is ultimately more decisive in modeling and remodeling related to the trabecular bone, which is closer to the cytokine-rich bone marrow than to the cortical bone through which the root is usually displaced (Teitelbaum 2007). TNF-α cannot stimulate osteoclastogenesis on bone slices independently; rather, it requires the presence of permissive levels of RANKL or RANKL pretreated with bone marrow macrophages (BMMs) for macrophages to become functional resorbing osteoclasts (Jules et al. 2010; Yamashita et al. 2015). The RANK receptor contains an IVVY (535-538) motif that controls the lineage commitment and differentiation by modulating osteoclast-associated genes into a more receptive state where they can be stimulated by TNF. In this respect, osteoclastogenesis induced by permissive levels of TNF is conditioned by RANKL to commit the BMMs to the osteoclast lineage, and RANKL regulates lineage specialization via the IVVY motif. The IVVY motif also plays a critical role in promoting root resorption mediated by the influence of the proinflammatory cytokine IL-1. IL-1 cannot directly induce the osteoclast/odontoclast formation required for resorptive activity. RANKL makes the osteoclast/odontoclast genes encoding MMP-9, cathepsin K, TRAP, and CA II in the BMMs responsive to IL-1 (Jules et al. 2012). Research has provided further insights into the molecular mechanisms involved in these results. The IVVY motif in the RANK receptor has been shown to play a vital role in IL-1-mediated osteoclast commitment by reprogramming the 4 key osteoclast marker and NFATc1 genes to an IL-1-inducible state. IL-1 alone is therefore unable to trigger expression of NFATc1, the key transcriptional regulator of osteoclastogenesis. However, under permissive levels of RANKL, IL-1 upregulates expression mediated by MyD88, an essential factor of the IL-1 receptor I signaling pathway. Therefore, activation of IL-1-mediated osteoclastogenesis/odontoclastogenesis requires the RANKL and has also been suggested as reducing osteoprotegerin (OPG) secretion (Fujisaki et al. 2007).

Figure 2.

Development schema of hematopoietic precursor cell differentiation into differentiated osteoclasts and odontoclasts, which are fused polykaryons arising from numerous individual cells. Specialization and maturation occur on alveolar bone from peripheral mononuclear cells with traits of the macrophage lineage. M-CSF (CSF-1) and RANKL are essential during lineage allocation and maturation. Some differences might be observed between clastic activity in bone or dentin substrate. OPG, osteoprotegerin; TGF, tumor growth factor; TNF, tumor necrosis factor; NO, nitric oxide; CSF, macrophage colony-stimulating factor; IL1ra, interleukin 1 receptor antagonist; TRAP, tartrate-resistant acid phosphatase; RANKL, receptor activator for nuclear factor κ B ligand; MMP, matrix metalloproteinase.

Also supporting the key role of the IL-1 pathway in root resorption, P2X purinergic receptor, ligand-gated ion channel, 7 (P2x7^-/-) knockout mice were found to have more root resorption than wild-type mice under orthodontic loading (Viecilli et al. 2009). The receptor is found mostly on the cell surface of macrophages involved in reception of the ATP signal. When orthodontic loading is applied, necrotic tissue deriving from cellular apoptosis leads to ATP release from damaged cells and its binding to the P2x7 receptor (Barberà-Cremades et al. 2016). Activation of this receptor leads in turn to autocrine and paracrine cell stimulation with overexpression of IL-1 cytokines and other inflammatory-related molecules. Consistent with this, P2x7^-/- mice—even when stimulated with lipopolysaccharides and subjected to adenosine triphosphate (ATP) in vivo—were unable to produce significant levels of IL-1β. The molecules released all function as chemoattractants for neutrophils and lymphocytes responsible for eliminating apoptotic cells and necrotic tissue, so allowing the root to move through the bone and orthodontic tooth movement to take place (Barberà-Cremades et al. 2016).

Results from recent studies (Kikuta et al. 2015) have stated that root resorption is mediated by Notch signaling in response to orthodontic forces of high magnitude, stimulating the process of root surface resorption via IL-6 as well as RANKL production by PDL cells. A number of studies have proposed that the Notch signaling pathway is involved in osteoblast and osteoclast commitment (Yamada et al. 2003); it has also been reported that suppression of Notch signaling by a selective supressor inhibits osteoclastogenesis facilitated by RANKL (Fukushima et al. 2008), whereas the presence of immobilized Jagged1 and the intracellular ectopic expression of Notch2 promotes osteoclastogenesis mediated by RANKL. In the orthodontic context, results suggest that Notch signaling stimulates pathologic root surface resorption via RANKL and IL-6 upregulation. It has been demonstrated furthermore that EARR induced by orthodontic treatment concurrent with certain biological scenarios, like some allergic diseases, is mediated by expression of RANKL with leukotriene B4, a potent lipid mediator of allergic inflammation (Murata et al. 2013).

In line with findings about RANKL-mediated root resorption in orthodontics that followed from the conclusions of other theories (Tyrovola 2015), compressive forces ranging from 0.5 to 3.0 g/cm² applied sequentially to cells of the PDL compartment for 24 h induced overexpression of RANKL and a parallel decrease in OPG levels at the 2.0-g/cm² threshold. Specifically, between the 2.0- and 3.0-g/cm² range of force, the increase in RANKL and decrease in OPG levels occurred in a force-dependent manner. With reference to root resorption, just as a nonlinear association between the root surface resorption and RANKL/OPG ratios was observed in the crevicular fluid of rats undergoing orthodontic tooth movement (Tyrovola et al. 2010), in terms of orthodontic tooth movement in young and adult humans, the OPG:RANKL ratio in the crevicular fluid varied according to the magnitude of the compressive force and in relation to time in a nonlinear way (Nishijima et al. 2006; Yamaguchi et al. 2006). Specifically, research results observed after applying compressive forces (up to 2.0 g/cm²) to cells of the PDL compartment, as retrieved from patients with aggressive EARR and healthy matched controls, indicated that the ratios of soluble RANKL/OPG were more accentuated in patients affected by EARR (Yamaguchi et al. 2006). So, root resorption can be defined qualitatively by the equation f(x) = αχ2 + βχ+γ, where root resorption correlates with the ratio between OPG and RANKL (parabolic correlation; Tyrovola 2015). The importance of these experimental results (correlation curves) is that they help provide a qualitative explanation, whereby we can conclude that OPG/RANKL fluctuates in a discontinuous way during tooth movement mediated by orthodontic force, depending on the force magnitude and in relation to time. These results partially explain the root resorption process and tooth movement following the mechanostat theory adapted from Frost’s (1987) orthopedic theory. Nonetheless, these mathematical models cannot precisely define the fact that root surface resorption represents the cellular action of osteoclasts/odontoclasts nor that the variation seen with different genetic backgrounds challenge a qualitative model that concentrates on the RANKL/OPG ratio (Iglesias-Linares et al. 2014; Sharab et al. 2015).

Clastic Cell Adhesion: Integrins and Extracellular Matrix Proteins

Proper adhesion of the clastic cell to the mineral substrate enables activation of all the intracellular machinery necessary to degrade the mineral component (Warren et al. 2015). Odontoclast attachment to the mineral surface is mediated through podosomes, which contain bands of actin filaments and F-actin and actin monomers and which allow further polarization of the clastic cells and integrin-mediated physical interaction with the extracellular matrix (Georgess et al. 2014). The integrins are αβ heterodimers, many of whose extracellular domains bind to proteins in the mineral component matrix and their intracellular components to cytoskeleton-organizing and signaling molecules. Mineral-resorbing cells express a main binding α_vβ₃ heterodimer (McHugh et al. 2000) with an appetite for the amino acid motif arginine-glycine–aspartic acid (RGD), whose sequence is found in different proteins in bone and dentin, such as osteopontin and dentin sialoprotein (Fisher et al. 2001). Osteopontin is a major noncollagenous acidic phosphorylated glycoprotein containing the RGD motif that is able to interact with α_vβ₃ and other integrins and cause the odontoclast to adhere to the root surface, promoting cell attachment through integrins and CD44 at the onset of physiologic or pathologic resorption (Denhardt et al. 2001). Other dentin extracellular matrix proteins, such as bone sialoprotein (BSP), have been found to affect RANKL-mediated bone resorption, by promoting osteoclast proliferation and survival and minimizing osteoclast programmed death in clastic precursors (Valverde et al. 2005).

Interestingly, clastic cells have been found to adhere some 45% better to root dentin than to bone (Rumpler et al. 2013). A longer half-life of the actin rings of clastic cells has also been reported on dentin slices compared with bone. Similarly, root dentin induces a sharper increase in the number of resorption pits, with 3,134 pits/cm² in dentin, compared with 449 pits/cm² in bone (Geblinger et al. 2010). When dentin slices were used to test resorption of deciduous or permanent incisors, clastic cells exhibited a marked upregulation of cathepsin K and MMP9 genes, and there was greater increase in the resorbed areas of the deciduous dentine as compared with cells cultured on permanent dentin slices (Varghese et al. 2006). These data from in vitro experiments evidence that the dentin substrate shows much greater potential than bone for inducing the genesis and maturation of new clastic cells. A possible explanation for this different appetite and resorption behavior is that dentin contains more matrix proteins of noncollagenous origin, such as osteopontin, when compared with bone (Azari et al. 2011). The absence of regulatory osteocytes and osteocyte proteins in dentin may also account for the quantitative differences (Cabahug-Zuckerman et al. 2016). Pretreatment with osteopontin or dentin sialoprotein has been reported to induce an increase in IL-1 (Yao et al. 2008). Recent reports highlighted the fact that osteopontin is involved in osteoclast development and osteolysis, orchestrating the secretion of pro- or anti-inflammatory cytokines from macrophages, as occurred with IL-1β in vivo and in vitro (Shimizu et al. 2010).

In humans, mechanical stress, such as orthodontic force, was shown to induce osteopontin expression by the PDL cells mediated by the Rho kinase pathway (Wongkhantee et al. 2008). The authors suggested that mechanical strain induces the release of ATP, which in turn modulates Rho kinase upregulation facilitated by the P2Y1 purinoceptor, which provokes overexpression of osteopontin. Therefore, it was suggested that, when induced by mechanical stress, ATP might be an important factor mediating root surface resorption, as through the P2X7 receptor already mentioned.

The osteopontin^-/- knockout mouse in vivo model was clearly in agreement with this (Chung et al. 2008). After application of orthodontic forces, transgenic mice were not affected by root resorption, nor did they show an increase in the number of odontoclasts, unlike wild-type mice, which developed EARR secondary to an increase in the number of TRAP+ odontoclasts. Interestingly, the same study found that a deficiency of osteopontin did not affect the number of TRAP+ osteoclasts located in the alveolar bone on the side of the direction of orthodontic tooth movement, which were comparable to those observed in the wild-type control animals. Other authors, however, found some decrease in the size and number of osteoclasts, which would suggest impaired cell fusion during bone modeling. All the results taken together suggest that the absence of osteopontin secretion suppresses or specifically mediates root resorption secondary to orthodontic strain.

Another matrix adhesion protein expressed in teeth, ameloblastin, plays an essential role in mediating root resorption (Lu et al. 2013). In vivo upregulation of this protein induced severe root resorption in transgenic mice, while the in vitro expansion of clastic cells pretreated with ameloblastin showed a 2-fold increase in cell adhesion, increased spreading, and podosome and actin ring formation in clastic cells. In this respect, ameloblastin seems to initiate an integrin-dependent cascade of extracellular matrix signals that lead to increased ERK1/2 and AKT phosphorylation and the upregulation of genes regulating proliferation and differentiation, which affect the number and maturation of clastic cells and affect their resorptive activity by regulating cell adhesion and actin polymerization (Sriarj et al. 2009).

These effects, mediated by specific noncollagenous proteins in dentin, led to the hypothesis that it was extracts of organic matrix components in dentin that modulated clastic activity (Sriarj et al. 2009). Findings showed that this may have been true of the increase in clastic cell adhesion; nevertheless, the organic matrix of dentin did not itself seem to cause the proresorptive effect on clastic cell activity, even when organic matrix from different substrates was used (Duplat et al. 2007; Sriarj et al. 2009). Bone marrow osteoclasts pretreated with extract of dentin organic matrix and cultured on ivory slices exhibited decreased resorptive activity (Sriarj et al. 2009). Downregulation of the mRNA levels of TRAP, v-ATPase, CTR, cathepsin K, and MMP-9 is observed on the pretreated ivory slices, with a reduced resorption surface area, indicating that noncollagenous proteins embedded in the dentin may play a leading role mediating odontoclast/osteoclast adhesion to the surface of the dentin. These data further suggest that other compounds in the dentin may be interfering with odontoclast/osteoclast activity, although not with the fusion and maturation processes, since odontoclast/osteoclast numbers did not seem to be affected. In association with this, it was reported that a c-Src kinase (involved in integrin formation)/Pyk2/Cbl complex that affects osteoclast migration also showed reduced activity in clastic cells after exposure to dentin extracts. The results suggested that certain compounds in the dentin could also intervene in odontoclast/osteoclast migration (Destaing et al. 2008).

In this respect, it should be remembered that various growth factors secreted during predentin formation—TGF-β, IGF-I, and BMP-2—remain in the dentin after mineralization. Consistent with this, TGF-β1 and TGF-β2 have been shown to potentially reduce the resorptive activity of clastic cells by enhancing OPG levels and downregulating RANKL mRNA (Schwartz et al. 2000) so that any of these or other dentin matrix proteins could partly affect odontoclast/osteoclast activity during mineral resorption (Varghese et al. 2006). Other research showed that dentin extracts promoted a potent stimulus to migration that was time and dose dependent and induced progressive cell maturation in peritoneal macrophages harvested from naïve and thioglycollate-injected mice. In addition, the authors showed that dentin extracts induced upregulation of IL-1β, TNF-α, nitric oxide, and hydrogen peroxide. The data suggested the possible involvement of organic components of dentin in inflammatory events/disorders, with their release at root resorption sites (Crane et al. 2016).

Reparative Capabilities of Mineralized Tissue: Role of Cementum

Cementum is commonly regarded as an antiresorptive barrier because it lacks a mineral-remodeling process. It remains to be elucidated whether the antiresorptive properties derive from a tissue component or cellular antiresorptive signaling associated with resident cells or even whether they are the result of the anatomic distance from the vascular and clastic precursor supply.

In vivo regulation in the cementocyte matrix has shown that the extracellular component exhibits fewer canalicular connections, irregular spacing, and lacunar shape as compared with osteocytes in alveolar bone (Zhao, Foster, and Bonewald 2016). At the transcriptomic level, cementocytes have shown an in vivo expression profile similar to that of the osteocytes, being able to express dentin matrix protein 1, Sost/sclerostin, E11/gp38/podoplanin, Tnfrsf11b (OPG), and Tnfsf11 (RANKL; Zhao, Nociti, et al. 2016). Despite similarities, in vitro experiments have shown that cementocytes express significantly higher levels of Tnfrsf11b and lower levels of Tnfsf11a mRNA than osteocytes from either the alveolar or long bones, resulting in a higher OPG:RANKL ratio, which constitutes a direct osteoclast inhibitory stimulus (Xiong et al. 2015). More important, when subjected to mechanical stimuli, osteocytes and cementocytes reveal different critical responses: while the former induce and increase Tnfsf11 expression, the latter significantly increase Tnfrsf11b and significantly reduce Tnfsf11. The higher OPG:RANKL ratio in cementocytes suggests a protective effect that inhibits cementum resorption or remodeling by Tnfsf11 (Walker et al. 2008; Xiong et al. 2015). Furthermore, in Sost^-/- mice, lack of sclerostin expression led to increased cellular cementum formation. Of substantial interest is the fact that cementocytes mimic osteocyte behavior in bone and repress Sost mRNA in cementum when subjected to fluid flow shear stress, so leading to new cellular cementum formation by modulating Wnt activity through sclerostin inhibition. This was also reported in a study of genetic strain of mice with elevated Wnt signaling (OCN-Cre;Wls(fl/fl)) that exhibited thinner cementum, demonstrating that the Wnt signaling pathway regulated cementum homeostasis. In these animals, RANKL expression was upregulated and OPG expression partially inhibited, modulating an increase in the resorptive activity of clastic cells on the dental root (Lim et al. 2014).

At the proteomic level (Salmon et al. 2013), studies of human dental cementum compared with alveolar bone revealed that up to 318 proteins in common can be found in both tissues. Nevertheless, 83 proteins were found exclusively in the cementum, implicated in different cellular pathways. Of these proteins, superoxide dismutase 3 (SOD3) was found strongly expressed in cementoblast and cementocytes. This SOD is a member of the SOD family that acts as a critical cellular protection against O2− and peroxynitrite. Interestingly, SOD3 was immunolocalized in cervical root cementoblasts and around apical third cementocytes (Salmon et al. 2013). These regions are well known as areas sensitive to root resorption; hence, regarding SOD3 localization and its function, this protein could play a key role as a “defense” from oxidative stress during cementum maintenance.

Other matrix proteins are also involved, as seen in Bsp^–/– mice, where BSP played a nonredundant role in acellular cementum formation in the root. So, loss of BSP caused extensive root resorption mediated by increased RANKL due to defective mineralization of the acellular cementum layer (Foster et al. 2013), which is in turn critical for homeostasis and the integrity of the root. Other matrix proteins have been reported as playing a key role in protecting the root against resorption (Bosshardt 2005). Some authors showed that restricted expression of amelogenins by the epithelial rests of Malassez in the periodontal region between the alveolar bone and the cementum layer indicates that these proteins may protect against the pathologic destruction of the cementum (Hatakeyama et al. 2003). Lack of expression of the gene coding for this protein induced a RANKL-mediated increase in the resorption process at the root cementum; the observed effects of the absence of amelogenins in the PDL region suggested that this particular type of protein plays a critical role in protecting against root resorption so that the RANKL/RANK/OPG axis seems to be involved in the regulation of resorption of cementum, too (Hatakeyama et al 2003; Le et al. 2016). Other authors (Zheng et al. 2015) have hypothesized that attenuation of dentin resorption and its reparative effect is controlled by some cellular component of the mesenchymal dental pulp stem cells, which are in turn also found in the PDL stem cells. Stem cells from dental pulp yielded ~20-fold-lower RANKL expression in clastic cells and a simultaneous 2-fold increase in osteoprotegerin expression compared with controls, resulting in a RANKL/OPG ratio of 41:1. The study therefore suggested that some degree of root resorption may be controlled by dental stem cells of mesenchymal origin, which may be studied further in the context of orthodontic root resorption.

Conclusion

Many highly complex autocrine and paracrine chemical signaling pathways mediate the 3 key cellular aspects of cell fusion/activation, clastic cell adhesion, and mineralized tissue reparative capabilities in a series of self-regulated cellular events of activation. Based on the recent literature, the molecular pathway leading to root resorption is not likely to be one and dependent on a single regulating factor. In contrast to this, it is highly likely that several of the molecular pathways and factors mentioned interact in crosstalk and influence effector cells for resorption at the level of fusion, activation, and cell adhesion. Similarly, differences in remineralization and even root tissue formation may account for differences in repair capabilities or the susceptibility of the dental root to pathologic resorption by clastic cells.

Author Contributions

A. Iglesias-Linares and J.K. Hartsfield Jr, contributed to conception, design, data acquisition, analysis, and interpretation, drafted and critically revised the manuscript. All authors gave final approval and agree to be accountable for all aspects of the work.