| Sign In to gain access to subscriptions and/or personal tools. |
Human Hertwig’s Epithelial Root Sheath Cells Play Crucial Roles in Cementum Formation
1 Center for Craniofacial Molecular Biology, University of Southern California School of Dentistry, 2250 Alcazar Street, CSA 103, Los Angeles, CA 90033, USA; Correspondence: * corresponding author, songtaos{at}usc.edu
Hertwig’s epithelial root sheath (HERS) cells are a unique population of epithelial cells in the periodontal ligament compartment. To date, their functional role has not been fully elucidated. Our hypothesis was that HERS cells may be involved in regulating differentiation of periodontal ligament stem cells (PDLSCs) and forming cementum in vivo. In this study, we found that HERS cells may be capable of promoting PDLSC differentiation and undergoing epithelial-mesenchymal transition in vitro. Immunohistochemical staining, Western blot analysis, a transwell co-culture system, and in vivo transplantation were used to characterize the interplay between HERS cells and PDLSCs, as well as the epithelial-mesenchymal transition (EMT) of HERS cells. TGFβ1 was capable of inducing the epithelial-mesenchymal transition of HERS cells through activating the PI3K/AKT pathway. Furthermore, HERS cells were able to form cementum-like tissue when transplanted into immunocompromised mice. Abbreviations: bone marrow mesenchymal stem cell, BMMSC; bone sialoprotein, BSP; hydroxyapatite/tricalcium phosphate, HA/TCP; Hertwig’s epithelial root sheath, HERS; osteocalcin, OCN; periodontal ligament, PDL; periodontal ligament stem cell, PDLSC; phosphatidylinositol 3-kinase, PI3K.
Key Words: Hertwig’s epithelial root sheath cell • periodontal ligament stem cell • epithelial-mesenchymal transition • cementum formation
Tooth growth and development are dynamic and complex processes in which reciprocal interactions between mesenchymal and epithelial cells play a pivotal role in controlling tooth and periodontal tissue formation. After crown formation, the inner and outer enamel epithelial cells form a bi-layered epithelial sheath, named Hertwig’s epithelial root sheath (HERS). The fate of HERS cells is, eventually, to form the epithelial rests of Malassez (Wentz et al., 1950), and some of them undergo apoptosis (Wentz et al., 1950; Kaneko et al., 1999). It has been speculated that HERS cells play an important role in root development, such as that affecting cementum repair (Spouge, 1980); however, whether HERS cells interplay with other types of periodontal ligament (PDL) cells, such as periodontal ligament stem cells (PDLSCs) (Seo et al., 2004), to control cementum formation remains unclear (Ten Cate, 1996; Zeichner-David et al., 2003). Epithelial-mesenchymal transition is a fundamental process whereby cells undergo a developmental switch from a polarized epithelial phenotype to a highly motile mesenchymal phenotype. Epithelial-mesenchymal transition has emerged as a central process during embryonic development, wound healing, fibrosis, and tumor metastasis (Thiery, 2002). During normal developmental processes, epithelial-mesenchymal transition plays an important role in many events at various developmental stages, such as gastrulation, neurulation, neural crest cell migration, sclerotome formation, and cardiac cushion mesenchymal development (Solursh et al., 1979; Weston, 1982; Sanders and Prasad, 1989; Boyer et al., 1999; Kang and Svoboda, 2005). Although epithelial-mesenchymal transition of HERS cells has been suggested (Thomas, 1995; Luan et al., 2006), the functional role of HERS cell epithelial-mesenchymal transition is not clear. Our hypothesis is that HERS cells play pivotal roles in cementum formation through the interplay with PDLSCs, as well as the epithelial-mesenchymal-transition process.
Participants and Cell Culture Normal human impacted third molars (n = 18) were collected from 12 adults (from 18–22 yrs old) at the Dental Clinic of the National Institute of Dental and Craniofacial Research (NIDCR), under the approved guidelines set by the NIH Office of Human Subjects Research. Collection of human PDLSCs and HERS cells for research has been approved by the IRBs of the NIDCR/NIH and the University of Southern California. PDL tissues were gently separated from the root surface of multiple numbers of third molars and separated into 2 portions: one for PDLSC isolation, and another for HERS cell isolation. For PDLSC isolation, PDL tissues were digested in a solution of 3 mg/mL collagenase type I (Worthington Biochemicals Corp., Freehold, NJ, USA) and 4 mg/mL dispase (Roche Diagnostic/Boehringer Mannheim Corp., Indianapolis, IN, USA) for 1 hr at 37°C. Single-cell suspensions were obtained by passing through a 70-µm strainer (Falcon, BD Labware, Franklin Lakes, NJ, USA). PDLSCs were isolated and cultured as previously described (Seo et al., 2004). For HERS cell isolation, pooled PDL tissues were digested in a solution of 3 mg/mL collagenase type I and 4 mg/mL dispase for 1 hr at 37°C to release single-cell suspensions that were seeded into 6-well culture plate (Costar, Cambridge, MA, USA) with 3 mL of defined Keratinocyte-SFM (GIBCO/ Invitrogen, Carlsbad, CA, USA). SFM medium was changed after 48 hrs from initial seeding, and then changed every 3 days. HERS cells usually require 6–8 wks to reach confluence, and they were passaged once for experimental use (at passage #1). Human bone marrow mesenchymal stem cells (BMMSCs) were purchased from a commercially available resource (AllCells LLC, Berkeley, CA, USA), and cultured as previously described (Batouli et al., 2003). Cemento/osteogenic differentiation was induced and detected as previously described (Seo et al., 2005). For co-culture experiments, after ex vivo serum-free expansion, HERS cells were loaded into upper chambers (Transwell, 0.4 µm pore size, Costar), and PDLSCs were loaded into lower chambers in osteo-inductive medium containing alpha-MEM supplemented with 10% FBS, L-glutamine (2 mmol/L), dexamethasone sodium phosphate (DEX; 10–8 mol/L), L-ascorbic acid phosphate (100 mmol/L), and monopotassium phosphate (KH2PO4, 1.8 mmol/L) for 28 days. Recombinant human TGFβ1 (R&D systems, Minneapolis, MN, USA) was added to the medium at 10 ng/mL to induce epithelial-mesenchymal transition of HERS cells. To inhibit the phosphatidylinositol 3-kinase (PI3K) pathway, we added LY294002 (Alexis, San Diego, CA, USA) to the medium at 20 µM.
Antibodies
Transplantation
Immunohistochemistry
Western Blot
Fluorescence-activated Cell Sorting (FACS)
HERS Cells Promote PDLSC Differentiation. Since PDLSCs are responsible for the formation of cementum and Sharpey’s fibers, and reside in the same compartment as HERS cells, our initial assumption was that HERS cells might influence the stem-cell property of PDLSCs. To test this assumption, we developed a culture system to expand primary human HERS cells (Fig. 1A  ). To avoid PDLSC contamination in the HERS cell culture, we examined whether PDLSCs could survive and divide in the non-serum culture condition used for HERS cell culture. As expected, PDLSCs stopped proliferating and underwent senescence within 4 wks (data not shown). These human HERS cells were capable of forming mineralized matrix when cultured under osteo-inductive conditions akin to those for immortalized rodent HERS cells (Fig. 1B). To examine whether HERS cells affect the differentiation of PDLSCs, we used a co-culture system and demonstrated that HERS cells promoted cemento/osteogenic differentiation of PDLSCs, but not BMMSCs, as assessed by mineralized nodule formation (Figs. 1D–1I). Interestingly, we also found that HERS cells reduced the number of STRO-1-positive cells in cultured PDLSCs, but not in BMMSCs (Fig. 1J).
Immune-phenotype of Human HERS Cells Next, we characterized the phenotype of HERS cells by immunohistochemical staining of decalcified human teeth. We found that human HERS cells expressed multiple types of cell markers, including epithelial molecules (amelogenin, cytokeratin, and E-cadherin), cemento/osteogenic molecules (BSP and OCN), mesenchymal molecule (vimentin), and epithelial-mesenchymal transition-associated molecules (β-catenin and N-cadherin) (Fig. 2A ). It appears that both HERS cell clusters and individual HERS cells can co-express different markers, such as cytokeratin/N-cadherin and E-cadherin/osteocalcin (Fig. 2B). In addition, HERS cells expressed TGFβ-related molecules TGFβ1, TGFβ receptor I, and TGFβ receptor II (Fig. 2A).
Epithelial-Mesenchymal Transition of Human HERS Cells Given the fact that human HERS cells expressed numerous epithelial-mesenchymal transition-associated molecules, it could therefore be postulated that human HERS cells undergo epithelial-mesenchymal transition. Since the TGFβ1 signaling pathway has been identified as one of the most important mechanisms that contribute to the epithelial-mesenchymal-transition process in a variety of cell types, here we assessed whether TGFβ1 plays a functional role in the epithelial-mesenchymal transition, if any, of HERS cells. Interestingly, TGFβ1 treatment triggered a partial morphological alteration of HERS cells, from typical epithelial cells to mesenchymal-like cells (Figs. 3A–3D ). Furthermore, we found that TGFβ1 treatment altered the protein expression levels typically seen in epithelial-mesenchymal transition, which include down-regulated expression of E-cadherin and β-catenin and up-regulated expression of N-cadherin and vimentin (Figs. 3E, 3F). To examine whether the AKT pathway is involved in TGFβ1-mediated epithelial-mesenchymal transition of HERS cells, as seen in primary epithelial cells, we used the PI3K-specific inhibitor, LY294002, at 20 µM to pre-treat HERS cells for 3 hrs. Then, HERS cells were stimulated with TGFβ1 for 1 hr (for phospho-AKT and AKT) or for 48 hrs (for vimentin and E-cadherin). As expected, LY294002 blocked TGFβ1-induced phosphorylation of AKT and vimentin expression (Figs. 4A, 4B), but it did not seem to influence E-cadherin expression (Fig. 4B).
The next question was whether epithelial-mesenchymal transition of HERS cells plays a functional role in cementum formation. To answer this question, we first treated primary human HERS cells with TGFβ1 for 48 hrs and then transplanted them. Eight weeks after transplantation, a very small amount of cementum-like structure had been formed on the surface of the HA/TCP, along with connective tissue (Fig. 4C  ). The cementum-like structure contained human-specific mitochondria-positive cells (Figs. 4D, 4E), along with BSP- and ALP-positive cementum-forming cells (Figs. 4F–4H). To confirm that HERS cells directly form cementum, which is not due to the contamination of PDLSCs, we examined the survivability of PDLSCs in the non-serum culture condition that was used for HERS cell culture, and found that PDLSCs cannot divide and undergo senescence (data not shown).
It has been demonstrated that human PDL contains a population of multipotent post-natal stem cells that can be expanded ex vivo, providing a unique reservoir of stem cells (Seo et al., 2004). It is interesting to know whether HERS cells, which reside in the same compartment with PDLSCs, interplay with PDLSCs. In this study, we found that HERS cells regulate cemento/osteogenic differentiation of PDLSCs, but not BMMSCs, perhaps through a mechanism associated with the expression of STRO-1, a surface molecule of mesenchymal stem cells (Gronthos et al., 1994; Shi and Gronthos, 2003). Analysis of these data suggests that HERS cells may provide factor(s) for the specific regulation of cemento/osteogenic differentiation of PDLSCs in vitro. Following the fenestration and breakdown of the HERS, the cells move away from the root surface and reform as the cellular aggregates of the epithelial rests of Malassez, which may be involved in cementum maintenance (Spouge, 1980) and HERS cell apoptosis (Kaneko et al., 1999; Cerri and Katchburian, 2005). These HERS cells express the epithelial markers cytokeratin and amelogenin, and the mesenchymal cell markers vimentin, OCN, and BSP. Therefore, it is reasonable to believe that HERS cells are a heterogenic cell population, and that a variety of signaling pathways may contribute to the functional differentiation of HERS cells. Our experimental evidence suggests that human HERS cells may not only undergo epithelial-mesenchymal transition under the induction of TGFβ1, but also develop into cementum-forming cells. It is critical to identify that HERS cells directly form cementum, which is not due to the contamination of PDLSCs. To clarify this issue, we specified that PDLSCs cannot survive and divide in the serum-free culture condition that was used for HERS cell culture. Furthermore, analysis of data from our study and previous studies identified that HERS cells are capable of forming mineralized matrix and expressing an array of osteogenic genes when cultured under osteo-inductive conditions (Ten Cate, 1996; Zeichner-David et al., 2003), implying their in vivo mineralized tissue regeneration potential. Although TGFβ1-treated HERS cells formed cementum-like tissue when transplanted into immunocompromised mice, the amount of cementum-like tissue formed by HERS cells was not comparable with the large amount of cementum-like tissue generated by PDLSCs, suggesting that only a subset of HERS cells carries the capacity of forming cementum-like tissue in vivo. Occasionally, we found tiny amounts of cementum-like structures in non-TGFβ-treated HERS cell transplants; this suggests that TGFβ may not be the sole inducer of epithelial-mesenchymal transition in HERS cells. There has been controversy over whether HERS cells undergo an epithelial-mesenchymal transition and become functional cementoblasts (Thomas, 1995; Luan et al., 2006). In this study, we found that HERS cells expressed enamel proteins, cytokeratin, vimentin, E-cadherin, and cemento/osteo-associated proteins, supporting the idea that HERS cells undergo epithelial-mesenchymal transition to become functional cementoblasts (Thomas, 1995; Kaneko et al., 1999; Obara et al., 1999; Zeichner-David et al., 2003). In contrast, analysis of recent experimental data implies that epithelial-mesenchymal transition may not occur in immortalized mouse HERS cells (Zeichner-David et al., 2003). Here, our experimental evidence suggests that human HERS cells may undergo epithelial-mesenchymal transition in vitro, which is accelerated by the TGFβ1 treatment. TGFβ1 was identified as an epithelial-mesenchymal-transition inducer by activating PI3K/AKT signaling in non-transformed murine mammalian cells (Bakin et al., 2000). Moreover, TGFβ1 treatment resulted in a N-cadherin distribution at the cell membrane (Bhowmick et al., 2001), and a cell morphology transition from cuboid form to a spindle-like elongated shape. Both molecular and cellular changes were inhibited by the PI3K inhibitor, LY294002, and by a dominant-negative (kinase inactive) AKT mutant. In contrast, the Smad signaling pathway did not appear to contribute to TGFβ1-triggered epithelial-mesenchymal transition, indicating that the TGFβ1-mediated epithelial-mesenchymal transition is independent of Smad signaling (Bhowmick et al., 2001). The mechanisms of PI3K/AKT activation by TGFβ remain to be fully elucidated (Zavadil and Bottinger, 2005). Here, we confirmed that TGFβ1 can trigger the epithelial-mesenchymal transition of HERS cells through the PI3K/AKT pathway in human primary epithelial cells. In this study, we provided experimental evidence to demonstrate that human HERS cells are capable of controlling PDLSC differentiation, and that HERS cells experience epithelial-mesenchymal transition to give rise to cementum-forming cells. However, to date, the underlying mechanism by which HERS cells regulate PDLSC differentiation is unknown. Also, it is not clear which specified portion of HERS cells undergo epithelial-mesenchymal transition, and which portion of HERS cells resists epithelial-mesenchymal transition, largely due to the difficulty of isolating subpopulations of HERS cells.
The authors thank Drs. Larry Fisher, Wu Li, and Stan Gronthos for providing antibodies for this study. This work was supported by the University of Southern California School of Dentistry, by National Institutes of Health Grant ROIDE17449 (to S.S.), and by the Division of Intramural Research, the National Institute of Dental and Craniofacial Research, the National Institutes of Health, US Department of Health and Human Services. Received for publication July 21, 2006. Revision received January 24, 2007. Accepted for publication March 5, 2007.
Journal of Dental Research, Vol. 86, No. 7,
594-599 (2007)
|
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
 |
|
|
 |
|
Sign In to gain access to subscriptions and/or personal tools.
TOP
ABSTRACT
INTRODUCTION
