| Sign In to gain access to subscriptions and/or personal tools. |
A Novel Role for Twist-1 in Pulp Homeostasis
1 Department of Biomedical Sciences, Texas A&M University Health Science Center, Baylor College of Dentistry, 3302 Gaston Avenue, Dallas, TX 75246, USA; Correspondence: * corresponding author, rd'souza{at}bcd.tamhsc.edu
The molecular mechanisms that maintain the equilibrium of odontoblast progenitor cells in dental pulp are unknown. Here we tested whether homeostasis in dental pulp is modulated by Twist-1, a nuclear protein that partners with Runx2 during osteoblast differentiation. Our analysis of Twist-1(+/–) mice revealed phenotypic changes that involved an earlier onset of dentin matrix formation, increased alkaline phosphatase activity, and pulp stones within the pulp. RT-PCR analyses revealed Twist-1 expression in several adult organs, including pulp. Decreased levels of Twist-1 led to higher levels of type I collagen and Dspp gene expression in perivascular cells associated with the pulp stones. In mice heterozygous for both Twist-1 and Runx2 inactivation, the phenotype of pulp stones appeared completely rescued. These findings suggest that Twist-1 plays a key role in restraining odontoblast differentiation, thus maintaining homeostasis in dental pulp. Furthermore, Twist-1 functions in dental pulp are dependent on its interaction with Runx2.
Key Words: Twist-1 • Runx2 • pulp stones • homeostasis • dental stem cells • odontoblast
During tooth development, the process of odontoblast differentiation begins at the cusp tip, when mesenchymal cells respond to inductive signals from adjacent dental epithelium. The diffusion of signaling molecules from the inner dental epithelium creates a morphogenetic gradient that extends vertically from the peripheral zones of the dental papilla to its core. While the layer of dental mesenchymal cells nearest the interface with epithelium receives the most intense signal and undergoes terminal differentiation into odontoblasts, daughter cells within the subodontoblastic layer of dental pulp retain the competence to differentiate later into functional odontoblast-like cells (Linde and Goldberg, 1993; Ruch et al., 1995). Multiple lines of evidence lend support to the theory that the dental pulp houses subpopulations of cells with various degrees of regenerative potential. In addition to being recruited from the cell-rich zone of dental pulp, the replacement population of odontoblast-like cells is also thought to be derived from perivascular cells, fibroblasts, and an undifferentiated mesenchymal pool (Fitzgerald et al., 1990; Smith and Lesot, 2001). More recently, post-natal stem cells that were isolated from extracted human third molars and deciduous pulps were shown to be capable of forming a dentin-like matrix (Gronthos et al., 2000; Batouli et al., 2003; Miura et al., 2003; Shi and Gronthos, 2003). This provides convincing evidence that dental pulp houses odontoblast-like cell precursors. Despite the growing interest in understanding how the regenerative capacity of pulpal cells can be best garnered for therapeutics, there is little known about the molecular mechanisms that restrain the differentiation of odontoblast precursors in a normal, resting pulp. Such information will advance our knowledge about how a normal resting dental pulp retains its patency and, importantly, how pulpal cells become recruited to the site of injury during reparative dentin formation. These studies evaluated the role of Twist-1, a basic helix-loop-helix containing transcription factor, in regulating the differentiation of odontoblast-like cells, as well as in maintaining the homeostasis of a precursor odontoblast-like cell population in adult pulps. The rationale for our work is derived from observations that Twist-1 functions as a cell-survival factor and an inhibitor of cell differentiation, mineralization, and apoptosis (Bate et al., 1991; Chen and Behringer, 1995; Soo et al., 2002). Interestingly, Twist-1 controls the terminal differentiation of osteoblasts, cells closely akin to odontoblasts, by transiently inhibiting the functions of another transcription factor, Runx2 (Bialek et al., 2004). In these studies, we hypothesized that decreased levels of Twist-1 in dental pulp disrupt its homeostasis and trigger the differentiation of odontoblast-like cell progenitors. We further tested whether alterations in homeostasis, seen in Twist-1-deficient pulps, were related to the relative overabundance of Runx2.
Mutant Animals, Tissue Preparation, and Histochemical Analysis A colony of Twist-1(+/–) mice was established from breeding pairs derived from the original strains as described earlier (Chen and Behringer, 1995). Institutional Review Board approval was obtained for all protocols involving the use of animals (#06010-R1). Genotyping was performed by PCR analysis on tail DNA (Bialek et al., 2004). To assess the genetic interaction between Twist-1 and Runx2 genes, we generated Twist(+/–) x Runx2(+/–) compound heterozygote mice by genetic backcrossing. For histological analysis and in situ hybridization, Twist-1(+/–) and wild-type littermates were killed at different post-natal stages, dissected, and fixed in 4% paraformaldehyde for 24 hrs. Tissues of mice, day 7 and older, were demineralized in 10% EDTA for 8 days. All tissues were processed for paraffin embedding, sectioned at 5 µm, and stained with hematoxylin and eosin (H&E) and Masson’s trichrome. Undemineralized sections were also stained for alkaline phosphatase and calcium-phosphate-rich structures (von Kossa) according to standard methods.
In situ Hybridization
RT-PCR and Real-time PCR
Functional Assays of Primary Pulp Cultures
Phenotypic Analysis Reveals Distinct Changes in Twist-1(+/–) Dentition. An earlier onset of organic matrix and mineral deposition was observed in D1 Twist-1(+/–) maxillary first molars (Figs. 1A, 1B  ). A dentin-like, tubular structure and a zone of predentin, surrounded by a distinct layer of cells stained for alkaline phosphatase activity, were also noticeably enhanced, particularly around blood vessels of D7 Twist(+/–) tissues (Figs. 1C, 1D). Dspp and Collagen type I were also detected by in situ hybridization, where matrix deposits in blood vessels in consecutive sections of D21 Twist(+/–) (Figs. 1E, 1F). As development proceeded, matrix deposits could be seen within the pulpal tissue of D45 Twist-1(+/–) animals (Figs. 1G, 1H). In total, 12 Twist-1(+/+) and 12 Twist-1(+/–) mice were evaluated, with a total of 240 individual teeth (molars and incisors) subject to histologic analysis. The effect of incomplete penetrance, as previously described for Twist-1(+/–) mice (Bourgeois et al., 1998), was confirmed, since the described phenotype was observed in 1 of 3 litters studied.
Twist-1 is Expressed in a Variety of Tissues Post-natally. Since our phenotypic analysis revealed changes in the dentition of Twist-1(+/–) mice, we were interested to see if Twist-1 was expressed in different organs post-natally. To gain an expression analysis, we performed RT-PCR analysis on several organs obtained from Twist-1(+/–) and Twist-1(+/+) animals at D0, 7, 28, and 60. Gene expression could be seen in all tissues at all stages (Fig. 2A ). As expected, Twist-1(+/–) tissues showed reduced expression (Fig. 2A).
Twist-1 Deficiency Leads to a Change in Cell Proliferation and Expression of Marker Genes of Odontoblast Differentiation. Real-time PCR analysis on first molar organs of Twist-1(+/–) and Twist-1(+/+) littermates at D3 showed reduced expression of Twist-1 in heterozygous animals, as expected. Whereas expression of marker genes of odontoblast differentiation was increased in mutant tissues, proliferation rates were slightly decreased, as determined after 3 and 7 days in culture (Fig. 3A ). Genes involved in dentin matrix formation and mineralization were up-regulated, such as  (1)type I collagen (1.4-fold), alkaline phosphatase (1.4-fold), dentin sialophosphoprotein (1.5-fold), and bone sialoprotein (2.1-fold). Only osteocalcin appeared to be expressed in equal quantities in Twist-1(+/–) and Twist-1(+/+) tissues (Fig. 2B ). Determination of alkaline phosphatase activity in primary cell cultures confirmed these results, after 7 days in culture; a twofold increase in Twist-1(+/–) cells was observed (Fig. 3B; See Appendix Fig. A–F for additional views.).
A Runx2 (+/–) Background Rescues Phenotypic Alterations Seen in Twist-1-deficient Dental Pulps. Based on the previously documented genetic interaction of Twist-1 with Runx2, we further evaluated, by hematoxylin and eosin, whether the changes in Twist-1(+/–) dental pulps were due to the relative overabundance of Runx2 protein. In mice heterozygous for Twist-1 and Runx2 inactivation, no abnormal calcifications were visible in dental pulps at all post-natal stages studied, suggesting that concurrent reduction in Runx2 levels rescued the phenotype of pulp stones seen in Twist-1(+/–) dentition (Fig. 4 ).
We investigated the role of Twist-1 in controlling odontoblast-like cell differentiation and in maintaining the homeostasis of a precursor odontoblast-like cell population in dental pulps. Analysis of our data indicated that, in mice, decreased levels of Twist-1 resulted in premature odontoblast-like cell differentiation, as measured by the earlier onset of expression of extracellular matrix gene markers and the formation of dentin matrix. Our results also indicated that Twist-1 deficiency distinctly affects precursor odontoblast-like cells, a population known to exist in adult dental pulps. These cells become more responsive to Runx2 when Twist-1 levels are decreased, leading to the subsequent deposition of pulp stone–like deposits within the pulp core. Taken together, these findings support the hypothesis that Twist-1 is important in the control of the terminal events that lead to odontoblast-like cell differentiation and in maintaining homeostasis in dental pulp. Furthermore, Twist-1 functions in dental pulp are dependent on its interaction with Runx2. Our conclusions are supported by previous molecular and genetic studies that explored the mechanisms that control the differentiation of osteoblasts, a cell population akin to odontoblasts. The inhibitory function of Twist uncovered by these assays, along with its expression in sclerotomes, lateral plate mesoderm, and osteoblast progenitors, raised the possibility that Twist may inhibit Runx2 function during osteoblast differentiation (Bialek et al., 2004). Comparative studies of Twist-1 and Runx2 expression showed that they are co-expressed early in the developing mouse skeleton. However, the expression of bone sialoprotein (Bsp), a marker of differentiated osteoblasts, occurs only after a concomitant decrease in Twist-1 expression, thereby implicating it as a negative regulator of the onset of osteoblast differentiation. Twist-1 overexpression resulted in an inhibition of osteoblast differentiation without an alteration in Runx2 expression. Similarly, Runx2 expression is unaltered in Twist-1 heterozygous mice. The anti-osteogenic function is mediated by the interaction of the Twist box, a novel domain in Twist proteins, with the DNA binding Runt domain of Runx2. Analysis of these data, collectively, indicates that the initiation of osteogenesis must first involve the relief of an inhibitor, Twist-1, so that Runx2 may exert its influence on osteoblast differentiation (Bialek et al., 2004). These studies showed that during skeletogenesis, Twist-1 proteins transiently inhibit osteoblast differentiation through specific interactions with the Runx2 protein (Bialek et al., 2004). Hence, it is the removal of Twist-1 inhibition that leads to Runx2’s transactivation of osteoblast-specific gene expression. Based on the data presented in this report, it is likely that a similar mechanism is responsible for the control of the onset of odontoblast-like cell differentiation, as well as the equilibrium of progenitor odontoblast-like cells in dental pulp. The skeletal abnormalities due to Twist-1 haploinsufficiency that have been observed in mice closely resemble those observed in humans who present with Saethre-Chotzen Syndrome (SCS; Bourgeois et al., 1998). The hallmark feature of SCS is craniosynostosis caused by the haploinsufficiency at the TWIST locus that leads to premature activity of Runx2 and the subsequent bone formation at the sutural margins. Although little information can be found on the effects of Twist-1 deficiency on the dentition, one case report describes abnormal crown and root development in addition to multiple pulpal calcifications (Goho, 1998). Based on our findings of Twist-1 in adult dental pulp and other mature organs, we speculate that the homeostasis of tissues that retain the capacity to mineralize is dependent on the functional antagonism shared between Twist-1 and Runx2 proteins. Lower levels of Twist-1 lead to Runx2 activation of odontoblast-like cells and the formation of ectopic dentin deposits. Two types can be distinguished: deposits displaying a dentin-like, tubular structure, and unstructured deposits in the vicinity of blood vessels. The two different types of matrix deposits might reflect their origin. Whereas undifferentiated mesenchymal cells might reside within the pulpal tissue, a distinct population of progenitor cells is associated with the perivascular niche, as described previously (Shi and Gronthos, 2003). A role for Twist-1 in homeostasis and during regeneration and repair has been shown to be true in various tissues, including muscle and periodontal ligament (Zhao and Hoffman, 2004; Afanador et al., 2005). Furthermore, the use of siTwist showed that Twist acts as a negative regulator of cell differentiation in the periodontal ligament (Komaki et al., 2007). The rescue of the pulp stone phenotype in mice heterozygous for both Runx2 and Twist-1 demonstrates that the functional antagonism shared between these proteins is also important for the homeostasis of dental pulp mesenchyme. Analysis of these data also suggests the possibility that osteoblast and odontoblast differentiation is regulated by similar mechanisms.
The authors acknowledge the support and advice of Dr. Gottfried Schmalz of the University of Regensburg, Germany. This research has been supported by the DAAD, German Academic Exchange Service (to KG), and by the Arthritis Foundation (PB) and NIH grants R01-DE013368 and R01AR045548 (to RDS and GK, respectively).
A supplemental appendix to this article is published electronically only at http://www.dentalresearch.org. Received for publication May 17, 2007. Revision received July 13, 2007. Accepted for publication July 16, 2007.
Journal of Dental Research, Vol. 86, No. 10,
951-955 (2007)
|
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
 |
|
|
 |
|
Sign In to gain access to subscriptions and/or personal tools.
TOP
ABSTRACT
INTRODUCTION
