Abstract
Odontoblasts are dentin-secreting cells that survive for the whole life of a healthy tooth. Once teeth are completely erupted, odontoblasts transform into a mature stage that allows for their functional conservation for decades, while maintaining the capacity for secondary and reactionary dentin secretion. Odontoblasts are also critically involved in the transmission of sensory stimuli from the dentin-pulp complex and in the cellular defense against pathogens. Their longevity is sustained by an elaborate autophagic-lysosomal system that ensures organelle and protein renewal. However, progressive dysfunction of this system, in part caused by lipofuscin accumulation, reduces the fitness of odontoblasts and eventually impairs their dentin maintenance capacity. Here we review the functional activities assumed by mature odontoblasts throughout life. Understanding the biological basis of age-related changes in human odontoblasts is crucial to improving tooth preservation in the elderly.
Introduction
Odontoblasts derive from cranial neural crest cells that emerge early during vertebrate evolution (Chai et al., 2000; Hall and Gillis, 2013). Their secretory product, dentin, is one of the first manifestations of extracellular matrix based on biomineralized collagen. This hard compound appeared in the Ordovician period in jawless fish with a dermal skeleton, whose tooth-like denticles contained dentinous tissues similar to the orthodentin that forms the bulk of contemporary vertebrate teeth (Smith and Sansom, 2000; Kawasaki and Weiss, 2008). In mammals, odontoblasts are long-lived post-mitotic cells organized in a continuous cellular palisade at the dentin-pulp interface, where they maintain pre-dentin and dentin apposition for the whole life of a tooth. Together with an impressive network of trigeminal nerve fibers, they form a complex sensory organ, detecting and transmitting changes in temperature, mechanical stimuli, and pain (Byers et al., 2003; Magloire et al., 2010; Farahani et al., 2011).
Increasing life expectancy in humans, exceeding 80 years in several developed countries, means that ever more individuals are reaching old age. From an oral health perspective, the increase in human longevity implies the necessity for prolonged tooth preservation (Tsakos, 2011). Understanding the mechanisms and biological foundation of age-related changes affecting odontoblasts and dentin apposition along life may create novel foundations for dental therapy and the application of adequate protective measures, depending on the age of the tooth.
This article reviews the different activities of odontoblasts during their life cycle, with a focus on their secretory, sensory, and defensive functions. In addition, it aims to define the homeostatic mechanisms that maintain mature odontoblasts for several decades, describing their autophagic-lysosomal system as one of the main pathways for organelle turnover in this cell (Couve, 1986; Couve and Schmachtenberg, 2011; Couve et al., 2012). Finally, the consequences of odontoblast aging for tooth preservation are discussed.
Dentin-secretory Activity of Odontoblasts
Odontoblasts differentiate from embryonic ectomesenchymal cells derived from the cranial neural crest during the initial events of tooth development (Koussoulakou et al., 2009). The development of pre-odontoblasts into terminally differentiated odontoblasts results in highly polarized cells with a columnar shape, about 50 µm in height (Couve, 1986). They are connected by junctional complexes forming a densely packed palisade at the dentin–pulp interface, from where each cell projects an odontoblastic process into the pre-dentin/dentin matrix. In erupted permanent teeth with completely formed crowns, coronal odontoblasts are organized in a pseudostratified palisade, while they assume a single-layer organization within the root of the tooth. Odontoblasts are post-mitotic cells that share certain features with neurons and myocardiocytes as static cell populations; these cells are normally not replaced during the life of the organism (Terman et al., 2010; Rezzani et al., 2012).
Odontoblasts are essentially dentin-secretory cells that produce pre-dentin, an extracellular matrix formed by type I collagen as the major organic component (about 90%), together with non-collagenous proteins including glycoproteins, proteoglycans, and dentin phosphoproteins (Kawasaki and Weiss, 2008). The intracellular pathway of collagen synthesis and secretion by odontoblasts was characterized in rat incisors in a classic study (Weinstock and Leblond, 1974) that revealed that differentiated odontoblasts form secretory granules that are rapidly discharged at the odontoblastic process by a constitutive secretory pathway, forming the bulk of the pre-dentin. The collagen matrix becomes biomineralized through a complex process by non-collagenous proteins secreted at the mineralization front, including dentin phosphoproteins like dentin matrix protein-1 (DMP-1) and dentin sialophosphoprotein (DSPP), which constitute specific markers for dentin (Goldberg et al., 2011). In human teeth, primary dentin is formed at an appositional rate of about 4 to 8 µm per day, allowing the bulk of the tooth to be created within 2 to 3 yrs of continuous secretion. Since dentin secretion is a continuous unregulated process, coronal odontoblasts are remodeled after tooth eruption into a mature stage, preserving their secretory activity at a reduced rate of about 0.5 µm/day (Couve, 1986; Simon et al., 2009) (Fig. 1).
Schematic representation of the life cycle of human odontoblasts. During tooth development, secretory odontoblasts are the ‘primary dentin’-forming cells. After tooth eruption at the coronal dental pulp, odontoblasts reduce their secretory machinery and acquire a mature stage characterized by the presence of autophagic vacuoles (AV). The progress of aging in mature odontoblasts involves the accumulation of lipofuscin deposits (LF). Three individual age groups are denoted in italics below their corresponding odontoblast condition. GC, Golgi complex; JC, junctional complexes; Ly, lysosome; M, mitochondria; N, nucleus; OP, odontoblastic process; PC, primary cilium; RER, rough endoplasmic reticulum; SG, secretory granules (modified from Couve and Schmachtenberg, 2011).
The term ‘primary dentin’ has been widely accepted for the regular orthodentin formed during tooth development by secretory odontoblasts, as opposed to secondary dentin, referring to a less active dentin apposition mediated by post-eruptive mature odontoblasts. The functional transition between primary and secondary dentinogenesis is accompanied by dramatic changes observed in the odontoblast phenotype (Fig. 1). The highly differentiated secretory machinery of the secretory stage is reduced by autophagic activity, and the remaining cellular organelles are relocated to the base of the cell, giving rise to the mature odontoblast stage (Couve, 1986). In human teeth, the mature odontoblast stage may persist for decades, but eventually aging effects and subtle age-related changes become evident, giving rise to an ‘old odontoblast’ stage, characterized by a reduced cell size and flattened shape (Figs. 2A-2C) (Couve et al., 2012). In addition, changes in cell polarity are evident once primary dentin is completely formed in the coronal region of human permanent teeth (Fig. 1). The conversion of the secretory to the mature stage could be related to differential transcriptional activity, as described for early- and late-secretory odontoblasts from cow incisors (Simon et al., 2009). Furthermore, a high expression of DMP-1 has been observed in early-differentiated odontoblasts forming primary dentin, while a gradual down-regulation of DMP-1 expression is observed in mature odontoblasts forming secondary dentin (Balic and Mina, 2011). Recent studies in transgenic mice also demonstrated that transcription factors like Distal-less 3 (DLX3) play a crucial role in odontoblast differentiation and dentin formation. Odontoblasts lacking DLX3 down-regulate the expression of DSPP and show decreased dentin matrix secretion and altered odontoblast polarization (Choi et al., 2010; Duverger et al., 2012).
Mature human odontoblasts.
The differentiating odontoblast requires trophic signals from its cellular environment, which may be detected by its primary cilium. Expression of a primary cilium is a highly conserved feature of post-mitotic cells. These organelles are permanently present in human odontoblasts and have been described at all stages of their life cycle (Couve, 1986; Magloire et al., 2004). The primary cilium has emerged as a ubiquitous sensory organelle consisting of a 9+0 microtubular axonema and a single non-motile cilium with a membrane expressing signal transduction elements, such as Sonic hedgehog (Shh) and Wingless (Wnt) pathway complexes (Singla and Reiter, 2006). Accordingly, the sensor function of the primary cilium is considered essential for the regulation of tooth germ development. For instance, signaling through the Shh pathway is required during the early differentiation of odontoblasts in mice (Dassule et al., 2000; Ohazama et al., 2009). Moreover, the Wnt/B-catenin signaling pathway plays an important role during odontoblast differentiation and dentin formation (Kim et al., 2011). Furthermore, the alleged role of primary cilia as mechanoreceptors on human odontoblasts suggests their participation in the unresolved sensory transduction processes operating in teeth (Magloire et al., 2004, 2010). In fact, a dual role–as an extracellular antenna sensing trophic factors and as a mechanosensory organelle–has been proposed for the primary cilia of odontoblasts (Thivichon-Prince et al., 2009; Magloire et al., 2010). Using electron microscopy, we have shown that the mature human odontoblast retains its primary cilium for decades and throughout its aging process (Figs. 2D-2F) (Couve, 1986; Couve and Schmachtenberg, 2011). Much remains to be learned about the function of the primary cilium, its associated endocytic membranes, and the ciliary pocket structure as a signaling center in mammalian cells (Benmerah, 2013).
Recent electrophysiological studies demonstrated that human coronal odontoblasts from young permanent teeth are electrically coupled via gap junctions and form a syncytial organization, suggesting a role in the coordination of cellular function among odontoblasts, above all regarding dentin formation (About et al., 2002; Ikeda and Suda, 2013). Gap junctions between human odontoblasts are mainly formed by connexin 43 (Cx43), and reduced expression of Cx43 has been described in the aging human dental pulp, a phenomenon apparently associated with a loss of pulp viability (Muramatsu et al., 2004).
There is an ongoing discussion about the appropriate terminology regarding the odontoblast phenotype along its cellular life (Simon et al., 2009; Larmas and Sándor, 2013). It has been proposed that the mature odontoblast stage should be re-named “odontocyte”, in analogy to osteocytes in bone. While odontoblasts project their odontoblastic processes into dentin, their cell body is never enclosed within the biomineralized dentin matrix, as opposed to osteocytes, which terminate passively embedded in the osteoid. Furthermore, the well-described cytological changes observed in bone cells are absent from human odontoblasts (Simon et al., 2009). We agree with the existence of a confusing nomenclature regarding the dentin-secretory stages of odontoblasts throughout their lifetime, but consider re-naming them inappropriate. In this review, human terminally differentiated odontoblasts are divided into 3 functional phenotypes: secretory, mature, and old (Fig. 1). However, the term “odontoblast” remains valid throughout the life of this amazing cell, reflecting the fact that it retains some secretory and sensory capacity until cell death, when it is replaced by odontoblast-like cells capable of forming reparative dentin (Smith et al., 1995).
Dental Pulp Innervation and Odontoblasts
In recent years, the concept of the dental pulp as a neurosensory organ has emerged, and such a sensory function might have been at the origin of contemporary teeth (Farahani et al., 2011). In human teeth, the odontoblastic layer of the coronal dental pulp is associated with an extensive sensory network (Figs. 3A, 3B) (Bernick, 1967; Sakurai et al., 1999). The coordination of tooth morphogenesis and pioneer trigeminal axon growth within the dental pulp is already a coordinated spatiotemporally controlled event, as reviewed in rodents (Luukko et al., 2005). During tooth development, the establishment of the sensory network within the dental pulp is regulated by the expression of axonal guidance cues (Moe et al., 2012). Some of the molecular signals that regulate development and maturation of sensory nerve fibers within the dental pulp have been described in detail (Fried et al., 2000; Luukko et al., 2005). For instance, during the innervation of human tooth pulp, odontoblasts express semaphorin 7A, one of the multiple guidance cues that lead pioneering axons to innervate the dentin-pulp complex within each tooth cusp (Maurin et al., 2005; Koussoulakou et al., 2009). The major innervated area within the dental pulp is formed by a dense network of nerve fibers in the subodontoblastic region, from where terminal nerve axons project through the odontoblastic layer into the pre-dentin/dentin interface, forming impressive ramifications.
Confocal microscopy of mature odontoblasts.
In response to dental pulp injury like caries or operative procedures, an early sprouting of nerve endings within the affected location has been observed in human and rodent teeth (Sakurai et al., 1999; Byers et al., 2003; Yoshiba et al., 2003). In fact, the sensation of pain through sensory afferents is a protective process, triggering rapid responses of the innate immune system (Hahn and Liewehr, 2007). Accordingly, it has been experimentally demonstrated that denervated teeth have compromised dentin repair responses and show diminished dental pulp survival after injury (Byers and Taylor, 1993).
Axonal endings from the dental pulp require a cellular support to fulfill their sensory functions. Interestingly, the target domains of nerve terminals within the odontoblastic layer change with age (Figs. 3A, 3B), but there is scarce information about an age-dependent remodeling of the dental pulp sensory system. In adult human teeth, a progressive reduction of nerve fibers at the dentin-pulp interface has been described, which may be related to the decrease of tooth sensitivity with age (Bernick, 1967). Aging brings about a reduction in subjective ratings of pain intensity and thermal sensitivity, which could be explained by the progressive obliteration of dentinal tubules and/or a reduction of terminal afferents of the dental pulp (Ikeda and Suda, 2003).
Further evidence for an age-dependent remodeling of the dentin-pulp sensory system came from an interesting recent study in rat molars, revealing that odontoblasts express different patterns of voltage-gated sodium channel isoforms, depending on their developmental stage and location (crown or root) (Byers and Westenbroek, 2011). Indeed, the organizational pattern and functionality of coronal odontoblasts from immature teeth differ widely from those of mature teeth, whose dentin-pulp complex supports a complex system of sensory innervation. However, the axonal cues and signals released by odontoblasts that guide the development of the vast sensory network within the odontoblastic layer and pre-dentin/dentin interface remain to be fully established.
Defensive Capacity of Odontoblasts
In addition to their dentin-secretory activity, odontoblasts play a role in defensive mechanisms and the stimulation of inflammatory responses against pathogen invasion through dentinal tubules. Intercellular junction complexes between odontoblasts create a pre-dentin-odontoblast barrier that constitutes the first line of defense against pathogens. The exposed position of odontoblasts allows them to detect bacterial components early and to trigger a prompt innate immune response, mainly by secreting cytokines that facilitate immune reactions and dentin repair (Hahn and Liewehr, 2007). Human odontoblasts respond to caries-expressing Toll-like receptors (TLRs), which are able to recognize bacterial components and activate cytokine production through an intracellular signaling cascade (Keller et al., 2010; Farges et al., 2011). The expression of TLR2 and TLR4 receptors has been demonstrated in human odontoblasts by immunohistochemistry, suggesting that odontoblasts might differentially recognize Gram-positive and -negative bacteria (Veerayutthwilai et al., 2007). Odontoblasts might be induced to express Toll-like receptor TLR2 in response to lipoteichoic acid, a component of the Gram-positive caries bacteria that frequently invade dentinal tubules (Durand et al., 2006). In parallel, the expression of Toll-like receptor TLR4 has been reported in trigeminal nociceptive neurons of the dental pulp, suggesting that Gram-negative bacterial components like lipopolysaccharides (LPS) could also be directly detected by pulpar nerve fibers, which may underlie the close relation of odontogenic pain with the progress of dentin caries infection (Wadachi and Hargreaves, 2006; Diogenes et al., 2011). The decrease in nociceptive fiber density within the odontoblastic layer and pre-dentin/dentin interface with age may reduce not only tooth sensitivity, but also the immune response against pathogens. The crosstalk between and among odontoblasts, nerve fibers, and immune/inflammatory components throughout the life of a tooth is a hot emerging topic that will provide important new insights in the near future.
The Autophagic-Lysosomal System in Aging Odontoblasts
The autophagic-lysosomal system consists of self-digestive pathways mediated by lysosomes by which cells regulate the turnover of long-lived proteins and organelles to maintain cellular homeostasis (Mizushima et al., 2008). In human teeth, mature odontoblasts develop a well-characterized autophagic-lysosomal system including conspicuous autophagic vacuoles that ensure the turnover and degradation of cellular components (Couve and Schmachtenberg, 2011; Couve et al., 2012) (Figs. 3, 4). Autophagy is active at a basal level in most long-lived cells and is considered a housekeeping process with an anti-aging function (Cuervo et al., 2005). A set of genes evolutionarily conserved from yeast to humans, referred to as autophagy-related genes (ATGs), was shown to orchestrate the different types of autophagic activity (Mizushima et al., 2008; Klionsky et al., 2012). High levels of autophagic activity are critical for the remodeling of cell functionality and as a survival mechanism, during starvation or in response to cellular stress or injury. In contrast, the dysfunction of autophagic processes has been related to several human diseases (Shintani and Klionsky, 2004; Mizushima et al., 2008).
Autophagic-lysosomal system in mature odontoblasts. Schematic overview of autophagy in a human odontoblast. Autophagy of mitochondrial components is initiated by a double membrane sequestration (phagophore). Autophagosomes containing cytoplasmic components fuse with lysosomes, forming autolysosomes. Autolysosomes are degradative compartments mainly characterized as autophagic vacuoles (AV) which receive mitochondrial components. Non-degradable deposits are accumulated in terminal AV containing lipofuscin (LF). Transmission electron microscopic evidence has shown that:
Autophagic activity can be initiated by different signaling pathways, depending on the cellular condition. Pathogen infection appears to induce autophagic activity through a TOR-independent pathway, comprising TLRs in many cases (Shintani and Klionsky, 2004; He and Klionsky, 2009), which are also expressed in human odontoblasts. However, the precise mechanism of autophagy induction by bacterial infections remains to be elucidated. Another factor important for the initiation of autophagy is cellular stress, caused by, among others, hypoxic conditions due to ischemia under orthodontic force application (Caviedes-Bucheli et al., 2011), or by reactive oxygen species (ROS), generated by cellular or metabolic stress (Mizushima et al., 2008). Cellular stress-related autophagy seems to be at least partially regulated by the TOR pathway, which normally senses the metabolic status of the cell (He and Klionsky, 2009).
In human odontoblasts, autophagic activity has been proposed as one of the essential survival mechanisms operating throughout the life of the tooth. With the specific marker microtubule-associated protein light chain 3 (LC3), autophagic activity has been detected during tooth development and in immature odontoblasts, suggesting an early role in the differentiation of these cells (Yang et al., 2013). LC3 is a mammalian orthologue of the autophagy-related protein ATG8 that is synthesized as a precursor protein and converted to its active form (LC3-II) by successive modification through ATGs 4, 3, and 7. As such, it localizes to the membrane of autophagosomes, where it acts as a selective receptor controlling the induction of autophagic activity (He and Klionsky, 2009; Klionsky et al., 2012). Once primary dentinogenesis is completed, odontoblasts reduce their secretory machinery by autophagic activity and limit their dentin-secreting activity to the synthesis of secondary dentin (Couve, 1986). In mature human odontoblasts, morphological and immunohistochemical studies describing LC3 and lysosome-associated membrane protein 2 (LAMP2) co-expression revealed the presence of autophagic vacuoles surrounded by lysosomes and mitochondrial components. A comparative analysis across different age groups revealed a dynamic autophagic machinery that is significantly remodeled during the odontoblastic aging process, suggesting a prominent role in cellular maintenance (Couve and Schmachtenberg, 2011; Couve et al., 2012).
Mitochondria are the main cellular compartments that produce ROS, accumulating damaged components throughout the life of the cell. Autophagy of mitochondria, known as mitophagy, is an important selective pathway for the removal of damaged mitochondria (Ashrafi and Schwarz, 2013). A comparative study in individuals between 15 and 85 yrs of age revealed a progressive decrease in the amount of mitochondrial DNA from human dentin with age (Mornstad et al., 1999). This reduction in mitochondrial DNA is associated with a decreased cell size in aging odontoblasts and the clustering of mitochondrial components, co-localizing with lysosomes in a juxtanuclear position (Fig. 3D). The constitutive turnover of mitochondrial components by autophagy causes the progressive accumulation of lipofuscin deposits within autolysosomes (Terman et al., 2010). Lipofuscin is also called the age pigment and corresponds to waste deposits accumulated within lysosomal compartments, a phenomenon affecting mainly long-lived post-mitotic cells (Terman et al., 2010). Accordingly, in young permanent teeth, odontoblasts show large clusters of lysosomes in close association with mitochondria and only small lipofuscin deposits, while mature teeth display an increased number of autophagic vacuoles per cell, containing heterogeneous and dense autofluorescent deposits (Figs. 3C-3F).
The cellular aging progress affects the functional and physiological activities of odontoblasts, resulting in their diminished ability to respond under injury (Murray et al., 2002). Experimental studies on replicative senescence in serially subcultured human dental pulp cells have shown a decreased expression of odontogenic markers like DMP-1 and DSPP, suggesting that age-related changes lead to reduced dentin apposition and mineralization activity in dental pulp cells (Lee et al., 2013). Moreover, studies on the mechanisms of stress-induced premature senescence (SIPS) in cultured dental pulp cells display increased autophagic activity in senescent compared with young cells, supporting the role of autophagy in cellular maintenance (Li et al., 2012). Likewise, we observed a progressive age-related increase of autophagic vacuoles in human odontoblasts, indicative of higher autophagic activity (Fig. 4) (Couve et al., 2012). However, the accumulation of autophagic vacuoles could also be related to a functional decrease in lysosomal activity (Rezzani et al., 2012). The dynamic changes in the autophagic-lysosomal system during the life cycle of human odontoblasts suggest that a reduction of autophagic activity is associated with a progressive accumulation of lipofuscin deposits (Figs. 1, 3, 4). In post-mitotic cells like neurons, cardiomyocytes, and human odontoblasts, the progressive accumulation of lipofuscin can be considered a hallmark of the aging process (Rezzani et al., 2012). Since cells have a very limited ability to eliminate these undegradable deposits, they may serve as a robust age-marker in permanent post-mitotic cells (Rubinsztein et al., 2011). In fact, old odontoblast phenotypes are characterized by large lipofuscin clustered deposits that comprise a significant portion of the cell volume (Couve et al., 2012). Although the formation of lipofuscin deposits is a common phenomenon during the aging process of post-mitotic cells, these deposits are considered inherently toxic and could affect cell function through inhibition of lysosomal-degradative capacities, leading to even faster lipofuscin accumulation. The incomplete autophagic degradation of cellular debris in lysosomes supports the idea that lipofuscin contributes to the normal aging process in long-lived post-mitotic cells and limits their lifespan (Terman et al., 2010).
With age, there is a progressive reduction in the thickness of the odontoblastic layer, mainly caused by a reduction in cell size (Murray et al., 2002; Couve et al., 2012). Increasing evidence indicates that autophagy could also constitute an alternative mechanism of cell death, named ‘autophagic type II programmed cell death’ (Type II PCD) (Shintani and Klionsky, 2004). As the odontoblast approaches its final days, the autophagic machinery might do its amazing host cell a last favor, providing for an orderly funeral.
Footnotes
This work was supported by FONDECYT 1120513, DIUV 32/2009 (Universidad de Valparaíso) and by the Millennium Institute CINV.
The authors declare no potential conflicts of interest with respect to the authorship and/or publication of this article.