- Email: [email protected]
Using human neural crest-derived progenitor cells to investigate osteogenesis: An in vitro study
Matrix Biology 29 (2010) 219–227
Contents lists available at ScienceDirect
Matrix Biology j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / m a t b i o
Using human neural crest-derived progenitor cells to investigate osteogenesis: An in vitro study Özer Degistirici a,⁎,1, Florian Grabellus b, Stephan Irsen a, Kurt Werner Schmid b, Michael Thie a,⁎,2 a b
Center of Advanced European Studies and Research (caesar), Bonn, Germany Institute of Pathology and Neuropathology, University Hospital of Essen, University of Duisburg-Essen, Germany
a r t i c l e
i n f o
Article history: Received 7 July 2009 Received in revised form 14 December 2009 Accepted 14 December 2009 Keywords: Human tooth Progenitor cells Neural crest Differentiation In vitro osteogenesis Bone matrix
a b s t r a c t Human tooth contains a distinct population of neural crest-derived progenitor cells (dNC-PCs) which are known to give rise to specialized daughter cells of an osteogenic lineage. We hypothesised that dNC-PCs could develop into neural crest-derived bone in a self-propagating and extracorporal culture system. Thus, we examined the three-dimensional structure obtained from osteogenic-stimulated dNC-PCs by morphological, biochemical and spectroscopic methods. After the onset of stimulation, cells formed a multilayer with outer cells covering the surface and inner cells secreting a hyaline matrix. With prolonged culture, multilayers contracted and formed a three-dimensional construct which subsequently converted to a calcified mass. Differentiation of progenitor cells was associated with apoptosis. Cell types which survived were smooth muscle actin-positive cells and bone-like cells. The expression of osteoblastic markers and the secretion of a collagenous matrix indicate that the bone cells had acquired their functional phenotype. Furthermore, these cells produced and secreted membrane-bound vesicles into the newly forming matrix. Consequently, an early biomineralized extracellular matrix was found with calcium phosphate deposits being associated with the newly formed collagen matrix framework. The molar calcium–phosphorus-ratio of the mineralized collagen indicated that amorphous calcium phosphate was present within this matrix. The data suggest that stimulated cultures of dNC-PCs are able to recapitulate some processes of the early phase of osteogenesis. © 2009 Elsevier B.V. All rights reserved.
1. Introduction A variety of different stem cells have been identified in many animal and human tissues (Young et al., 1999). Adult stem or progenitor cells are being widely investigated with a focus on their ability to differentiate into a broad array of cell types potentially useful as cell material for purposes of tissue engineering and regenerative medicine (Kassem, 2006; Aejaz et al., 2007; Atala, 2007; Redi et al., 2007; Bajada et al., 2008). The developing field of craniofacial tissue engineering promises the regeneration of dental, oral, and craniofacial structures (Mao et al., 2006). Recent evidence
⁎ Corresponding authors. Degistirici is to be contacted at Universitätsklinikum Düsseldorf, Klinik für Kinder-Onkologie, -Hämatologie und-Immunologie, Geb. 13.42, Moorenstrasse 5, D-40225 Düsseldorf, Germany. Tel.: + 49 211 811 6277; fax: + 49 211 811 6436. Thie, Institut für Anatomie und Klinische Morphologie, Private Universität Witten/Herdecke gGmbH, Alfred-Herrhausen-Straße 50, D-58448 Witten, Germany. Tel.: + 49 2302 926 721; fax: + 49 2302 926 739. E-mail addresses: [email protected] (Ö. Degistirici), [email protected] (M. Thie). 1 Current address: Department of Paediatric Oncology, Haematology and Immunology, Düsseldorf University Hospital, Düsseldorf, Germany. 2 Current address: Institute of Anatomy and Clinical Morphology, Faculty of Medicine, University Witten/Herdecke, Witten, Germany. 0945-053X/$ – see front matter © 2009 Elsevier B.V. All rights reserved. doi:10.1016/j.matbio.2009.12.005
suggests that the connective tissue of the human tooth contains a particular population of neural crest-derived progenitor cells (termed dNC-PCs) which give rise to a wide range of specialized daughter cells when outside their dental signal network (Degistirici et al., 2008; Schoenebeck et al., 2009). In attempting to use dNC-PCs for tissue engineering we showed that it was possible to induce bone formation in a nude mouse model by combining dNC-PCs with bovine bonederived granulates, which act as a scaffolding component. Under these conditions, the only differentiated stem cells observed were those of an osteogenic lineage (Degistirici et al., 2008). To take full advantage of dNC-PCs, however, it will be important to understand how these cells differentiate into bone-specific cell types and what environmental conditions are stimulating these cells to become functional bone. In craniofacial tissues, the bone cell lineages originate from cranial neural crest cells that have committed to the osteogenic cell lineage becoming osteoprogenitor cells, such as preosteoblasts, osteoblasts, and osteocytes (Le Douarin et al., 1994; Dupin et al., 2006; Chai and Maxson, 2006). It turns out that the molecular mechanisms inducing osteogenesis in cranial neural crest cells, which produce the facial and jaw skeleton, are distinct from those operating in mesodermal cells, which produce most of the skeleton (Helms and Schneider, 2003; Tucker and Lumsden, 2004; Abzhanov et al., 2007; Han et al., 2007; Xu et al., 2007; Deng et al., 2008). Therefore, our focus is on the specific
220
Ö. Degistirici et al. / Matrix Biology 29 (2010) 219–227
biology of the process of dNC-PCs developing into bone-like tissue. Aspects considered are the differentiation of dNC-PCs from stem or progenitor cells to cells of the osteogenic lineage and the cellular changes that accompany their differentiation. Our current notion is that the extracellular matrix components and their structural architecture trigger this differentiation process, thus we investigated the potential of dNC-PCs for osteogenesis and bone development when cultured in osteogenic growth medium and maintained within a self-established extracorporal microenvironment. Here, we report that dNC-PC cultures responded to osteogenic stimulation by progressing through a series of structural and biochemical changes that culminated in the formation of a simply ordered bone-like matrix. Although we are left with several questions, these data give new insights into the molecular processes of human bone formation by stem or progenitor cells derived from the neural crest. 2. Materials and methods 2.1. Cell isolation and expansion culture Surgically removed impacted third molars of young adults were used to prepare dental neural crest-derived progenitor cells (dNCPCs) (Degistirici et al., 2008; Schoenebeck et al., 2009). Written consent was obtained from all parents of the participating patients. Briefly, apical pad-like tissue of the developing tooth was cut into small pieces, enzymatically pretreated with collagenase/dispase solution (Sigma, Munich, Germany) and microexplants seeded in tissue culture flasks (BD Falcon, Germany). Subsequently, microexplants were maintained in Dulbecco's modified Eagle's mediumlow glucose (DMEM-LG) (Cambrex Bio Science, Verviers, Belgium) supplemented with 2 mM glutamine and 1% penicillin/streptomycin (Sigma) and 10% fetal calf serum (FCS) (Biochrom, Berlin, Germany). Upon reaching 70–80% confluence, outgrowing cells were detached by trypsinization (trypsin/EDTA) (Sigma) and replated at a density of 5 × 10³ cells/cm². The resulting cells were further incubated under standard culture conditions for expansion. For experiments, dNC-PCs were used at passages 3 to 4. 2.2. Osteogenic culture For osteogenic induction, cultured dNC-PCs were initially seeded at a density of 8 × 103 cells/cm2 in 24-well plates (BD Falcon, Germany) and cultured in growth medium containing DMEM-LG, 10% FCS and 1% penicillin/streptomycin. When reaching subconfluency, the growth medium was replaced with differentiation medium, DMEM-LG supplemented with 100 nM dexamethasone, 50 μM ascorbic acid-2 phosphate, and 10 mM ß-glycerol phosphate (all from Sigma). The medium was changed twice a week. For a threedimensional culture, cells were seeded at a density of 5 × 103 cells/cm2 in 25-cm² culture flasks and incubated in osteogenic medium as described above. Formation of a three-dimensional structure was brought about by prolonged culture. Samples were analyzed at the indicated time points. 2.3. Assay of alkaline phosphatase activity Alkaline phosphatase activity was determined in cell lysates using p-nitrophenyl phosphate as a substrate (Sigma). The procedure was carried out as recommended by the manufacturer. Briefly, cells growing on 24-well plates were washed with phosphate-buffered saline (PBS) and incubated with 1% Triton X-100 (Sigma). The resulting lysate was then incubated with substrate and the released p-nitrophenol was measured at 405 nm (PerkinElmer, Waltham, MA). Alkaline phosphatase activity was expressed as μM p-nitrophenol/l/min/sample.
2.4. Quantification of calcium The total calcium of the samples was measured by the o-cresolphthalein complexone method using the commercial Calcium Assay-CA590 kit (Randox Laboratories, Co Antrim, United Kingdom). The procedure was carried out according to the manufacturer's protocol. Briefly, cells growing on 24-well plates were washed with PBS and extracted in 200 μl of 0.5 N hydrochloric acid for 5 min. Samples were then vigorously shaken for 4–16 h at 4 °C. Calcium determination was done in 96-well plates with 10 μl of test solution and 90 μl of substrate solution mixed up with 100 μl of distilled water. The amount of deposited calcium was determined at 570 nm using a spectrofluorometer (PerkinElmer) and was expressed as μg/sample. 2.5. TUNEL assay In situ DNA fragmentation was established using the terminal desoxyribonucleotide transferase (TdT)-mediated dUTP nick-endlabelling technique (TUNEL) in paraffin-embedded sections. We used the ApoTag™ plus peroxidase in situ apoptosis detection kit (KIT S7101, Intergen, USA). The staining procedures were performed according to the manufacturer's recommendations. Slides of colorectal cancer (CRC) treated in the same way served as positive control. Negative controls were carried out with CRC slides without exposure to TdT enzyme. 2.6. Histology and immunohistochemistry The cultured cells were harvested, paraffin-embedded and sectioned following standard protocols. Briefly, samples were fixed in 4% formaldehyde in PBS, washed with PBS and dehydrated with ethanol. After dehydration, samples were washed with toluene, infiltrated with molten paraffin and embedded and sectioned. Sections of 5 μm were prepared for a panel of immunohistochemical examinations. Dewaxed and rehydrated sections were incubated with hydrogen peroxide to block endogenous peroxidase, antigen retrieval was performed in a hot water bath. The following monoclonal primary antibodies were used: anti-smooth-muscle actin (dilution 1:500; Dako, Hamburg, Germany), anti-osteocalcin (1:500; Acris-Antibodies, Herford, Germany), and anti-collagen type II (1:500; MP-Biomedicals, Illkirch, France). For primary antibody detection the Zytomed Polymer Kit DAB (dilution 1:10; Transduction Laboratories, San Diego, CA, U.S. A) was used. All immunohistochemical stains were performed with an automated staining device (Dako Autostainer, Glostrup, Denmark). Omission of the primary antibodies served as negative controls. Selected sections were stained with hematoxylin and eosin. To identify the formation of mineralization, monolayers were stained with von Kossa. To detect alkaline phosphatase, monolayers were stained with naphtol and fast red violet. 2.7. Electron microscopy For transmission electron microscopy, samples were fixed in 2.5% glutaraldehyde in 0.1 M cacodylate buffer, pH 7.2, for 2 h at room temperature, washed in cacodylate buffer, post-fixed with 1% osmium tetroxide in cacodylate buffer, dehydrated with ethanol and propylene oxide and embedded in epoxy resin. Ultrathin sections were mounted on copper grids, double-stained with uranyl acetate and lead citrate and examined with a Zeiss TEM 902A. For scanning electron microscopy, samples were fixed in 2.5% glutaraldehyde in 0.1 M cacodylate buffer, pH 7.2, for 2 h at room temperature. After repeated washings in cacodylate buffer, samples were dehydrated with ethanol and dried using hexamethyldisilazane (Polysciences, Eppelheim, Germany) as the drying medium. Then samples were sputtered with a conductive layer of gold and imaged
Ö. Degistirici et al. / Matrix Biology 29 (2010) 219–227
with a Leo SEM Supra 55. For some analysis, samples were imaged with a LEO XB1540 cross beam.
2.8. EELS/EDS analysis Electron energy loss spectroscopy (EELS) was performed using a Zeiss TEM Libra 200 (CRISP edition) with imaging in column energyfilter and detector for energy dispersive X-rays (EDS). The experiments were performed in scanning (STEM) mode with an energy resolution of 0.5 eV. Experiments were taken at the L-edge of calcium (350 eV) and the L-edge of oxygen (536 eV). Energy dispersive X-ray analysis (EDS) was performed in STEM mode at a convergence angle of 25 mrad. The spot size of the scanning probe was approximately 1 nm. For quantitative analysis rectangular regions of interest with approximately 50 nm edge length were selected and analyzed in an energy range between 0 and 20 keV. The spot diameter for line scans was in the range of 1 nm.
3. Results 3.1. Preparation of progenitor cells Dental neural crest-derived progenitor cells (dNC-PCs) were prepared from the pad-like tissue of impacted third molars of young adults using the microexplant technique. When subcultured as monolayers in standard growth medium, dNC-PCs showed the distinct phenotype of multipotent progenitor cells as described previously (Degistirici et al., 2008; Schoenebeck et al., 2009). To induce cells of osteogenic lineage for experimentation, dNC-PCs were initially preincubated for 4–7 days with growth medium in the presence of 10% FCS. Then, dexamethasone, ascorbic acid-2 phosphate and ß-glycerol phosphate were added to the pre-incubation medium, and culture systems were further cultivated. dNC-PCs grown without osteogenic factors were used as control.
3.2. Formation of a three-dimensional culture system In general, dNC-PCs responded to osteogenic stimulation through a succession of stages of developing structural complexity (Fig. 1). The first change was a significant increase in monolayer height (Fig. 1A, B) that was due to an increase in cell numbers as well as an increase in the proportion of extracellular matrix (Fig. 1A). After 14 days of stimulation, cells had formed a multilayered structure with the development of several distinct nodules scattered within the culture system (Fig. 1B). Two types of cellular layering could be observed: an outer one and an inner one. Flat cells covering the surface represented the outer cells, cuboidal cells trapped within the layered structure represented the inner cells. Inner cells secreted a hyaline matrix with which they surrounded themselves. As cell growth and extracellular matrix production continued, the overall shape of the culture system changed (Fig. 1C–F). At around 28 days of stimulation, multilayered cultures detached from the plastic, folded up and upon contraction, formed a free-floating spherical cell–matrix mass (Fig. 1C; D). Outer cells grew on the surface of the mass, thus forming a circumferential sheath of living cells (Fig. 1E, F). At the same time as contraction, the spherical mass converted to a calcified structure and highly mineralized areas began to form. Histological sections of the non-decalcified mass showed the presence of mineralized extracellular matrix (Fig. 1E, F). Inner cells produced matrix and then mineralized it whereas outer cells did not. Outer cells persisted on the surface of the mass (Fig. 2A), thus enveloping the calcifying core (Fig. 2B). In contrast, dNC-PCs cultured without osteogenic factors continued as non-calcified monolayers remaining attached to the surface of the plastic dish (data not shown).
221
3.3. Diversification of cell phenotype Different phenotypes of dNC-PCs in different parts of the selfforming culture were observed (Fig. 3). Specialized types of cells were identified by various markers, such as smooth muscle actin, which identifies a distinct subset of contractile cells (Fig. 3A, B). Typically, smooth muscle actin-positive cells were found within the sheath of living cells on the surface of the culture system where they were localized between cells that did not react with the antibody. A disparity between inner cells/inner part and outer cells/outer part as a result of apoptosis was observed (Fig. 3C, D). If culture systems were analyzed at an early stage, i.e. 28 days after stimulation, only a few cells undergoing apoptosis were detected among the outer cells and none among inner cells. Conversely, after 42 days of stimulation, a large proportion of the inner cells were dying by apoptosis while outer cells were not and thus were saved. Differences also existed when looking at cells specialized for secretion of extracellular matrix proteins, for example, osteoblast-specific osteocalcin (Fig. 3E, F). If culture systems were analyzed at an early stage, i.e. 28 days after stimulation, only a few dispersed cells in the outer layer were positive for an antibody against osteocalcin. In contrast, 42 days after stimulation, a large number of osteocalcin-positive cells were found in the core region of the mass while the outer cells did not react. Type II collagen antibodies were used as a typical marker for a chondrogenic extracellular matrix. When analyzing the culture systems by immunohistochemistry, no reaction of antibodies was observed.
3.4. Ultrastructural features Electron micrographs of dNC-PC cell types and their surrounding matrices which formed 28 days after stimulation are shown in Fig. 4. Outer part cells which grew as a circumferential sheath of cells lining the spherical mass appeared mostly spindle-shaped and exhibited a fibroblast-like phenotype (Fig. 4A). The nucleus was elongated and showed distinctive invaginations. The cytoplasm was filled with elements of the endoplasmic reticulum, Golgi complexes and mitochondria. Small and large vesicles were present. A basal lamina and depositions of extracellular matrix were not observed, instead cells were in close contact with each other. Inner part cells appeared more spherical and exhibited a synthetic phenotype characterized by a prominent rough endoplasmic reticulum (Fig. 4B). Cells grew a large number of cytoplasmic extensions and were surrounded by depositions of extracellular matrix. Scanning electron micrographs of inner part cells displayed numerous membrane-bound vesicles budding off from the free cell surface (Fig. 4C–E). When observed by transmission electron microscopy vesicles were seen to be filled with an amorphous osmiophilic substance (Fig. 4D). The extracellular space between cells was endowed with a meshwork of characteristic fibrelike structures (Fig. 4F). Some of the ordered structures appeared as banded fibrils, resembling thin (40–50 nm in diameter) cable-like constructs.
3.5. Quantification of initial calcification and alkaline phosphatase activity When matrix molecules were deposited by osteogenic-stimulated dNC-PCs, a conversion of soft matrix into hard matrix was observed (Fig. 1; Fig. 5). After 28 days of stimulation, multilayered cultures of dNC-PCs were shown to contain substantial deposits of calcium salts indicating the biological mineralization process of the extracellular matrix. A quantitative analysis of total calcium showed that stimulated cultures exhibited higher values than controls (Fig. 5A, B). Alkaline phosphatase activity increased continuously in osteogenic-stimulated cultures as well as in controls (Fig. 5C, D).
222
Ö. Degistirici et al. / Matrix Biology 29 (2010) 219–227
Fig. 1. Histological sections showing steps in the process of organization of dNC-PC cultures. When growing on plastic, cells make contact to the surface, proliferate and secrete an extracellular matrix with which the inner cells surround themselves (A). Thus, a multilayer with several distinct nodules scattered within the culture is formed (B). As the culture grows, the number of cells and the quantity of extracellular matrix become much greater. Next, the multilayer detaches from the plastic surface, folds up and forms a roundish mass (C). With prolonged culture time, masses get more compact (D). The matrix calcifies soon after the cultures have formed masses. Note the smooth cells that still line the mass (E, F). Medium: growth medium; plastic: plastic surface; calcification: a compact portion of calcified mass; asterisk: the surrounding layer of cells; rhomb: the inner sheets of cells; A + B: same magnification; C + D: same magnification.
3.6. Spectroscopic analysis of biological mineralization After a fine meshwork of fibrils, of varying diameter, had formed in the extracellular space of osteogenic-stimulated cultures (Fig. 6A), composites were characterized by EELS spectroscopy and EDS analysis (Fig. 6B–E). EELS spectroscopy revealed the intensity of the calcium L2,3 edge of several scans, proving calcification (Fig. 6B). Fig. 6C shows two EELS spectra of calcium-rich and calcium-free areas of the sample which confirmed the presence of calcium was restricted to the fibrillar meshwork, where the intensity of the calcium signal was reasonably constant. For further identification of this phase, EDS analysis at numerous positions was performed (Fig. 6D) and used for quantification (Fig. 6E). The calcium–phosphorus-ratio in this phase varied between 1.1 and 1.3. The mean ratio of 13 selected regions was
1.11 +/− 0.06 indicating the presence of amorphous calcium phosphate (Fig. 6E). 4. Discussion The starting point for this study were our recent reports that identified dental neural crest-derived progenitor cells (dNC-PCs) as good candidates to regenerate neural crest-derived tissues within the craniofacial region (Degistirici et al., 2008; Schoenebeck et al., 2009). As dNC-PCs could be easily differentiated into osteogenic lineages we speculated whether these cells may be useful to engineer facial and jaw skeleton. Indeed, the growing realization that skeletogenesis in the head is a unique and separable process from that occurring elsewhere in the body makes this a crucial new approach to take
Ö. Degistirici et al. / Matrix Biology 29 (2010) 219–227
223
Fig. 2. Scanning electron micrographs of fractures of dNC-PC masses. Note the many folds, i.e. sheets of cell layers, which are visible within the calcified mass (A). Note also the surrounding layer of smooth cells (asterisk) enwrapping the calcified mass (B). Boxed area: drawing of a sheet of cells; asterisk: the surrounding layer of cells; core: calcified mass.
Fig. 3. Localization of cells positive for smooth muscle actin (SMA) (A, B), apoptotic nuclei (TUNEL) (C, D), and osteocalcin (OCN) (E, F) in young (A, C, E) and old (B, D, F) stages of dNC-PC cultures. SMA-positive cells are always found in the outer part of the culture, i.e. the surrounding layer. In contrast, TUNEL- and OCN-positive cells are found in the inner part. Note the strong increase in the number of TUNEL- and OCN-positive cells with prolonged culture time (D, F).
224
Ö. Degistirici et al. / Matrix Biology 29 (2010) 219–227
Fig. 4. Electron micrographs of cell types and their matrix seen in dNC-PC masses. (A) Transmission electron micrograph of a section of the outer part of the structure, showing single cells in their normal context. Cells exhibit an elongated mesenchymal phenotype and are in close contact to each other. Note lack of matrix deposits. (B) Photograph of a section of the inner part of the structure. Cells have several cellular processes and are filled with an extensive rough endoplasmic reticulum, indicating an active phenotype. Cells seem responsible for secretion of the matrix surrounding them. (C, E) Scanning electron micrographs of active cells of the inner portion of the spheroid. Cell surfaces and cell processes are covered by numerous vesicles (globular structures). As shown in (E), spaces between cells are filled with extracellular matrix collapsed during the dehydration process for specimen preparation. (D, E) Transmission electron micrographs of calcified/calcifying vesicles, containing amorphous deposits. Note the loosely packed collagen fibrils which surround the cells. The characteristic features of collagen fibril ultrastructure are shown in (F). n: nucleus; v: vesicle; ECM: extracellular matrix; arrows: regular banding of fibrils.
(Helms and Schneider, 2003). In this report, we provide further evidence, that dNC-PCs are able to produce an osteogenic matrix via a self-propagating extracorporal culture system. The construct is composed of an outer cell layer surrounding a compact mass, which itself consists of bone cells embedded in an organic collagenous framework containing calcium phosphate deposits. As the biomineralized matrix consists of amorphous calcium phosphate, i.e. the precursor phase of calcium-deficient hydroxyapatite or hydroxyapatite found in mature bone (Weiner, 2006, 2008), this kind of cranial neural crest cell-produced bone may be of transient (early) nature. Our analysis of dNC-PC cultures autonomously differentiating in vitro revealed that cultures are developing along a pathway that should eventually lead to mature bone formation. The expression of osteoblastic markers like alkaline phosphatase and osteocalcin and the secretion of a mineralizing matrix indicates that cells had acquired a functional bone-forming phenotype. It should be noted however, that cultures of osteogenic-stimulated dNC-PCs did not develop into fully organized and anatomically structured bone. In particular, sitespecific formation of blood vessels that would support the metabolic needs of mature tissue was not observed. The lack of formation of
capillary-like structures may be due to a number of factors, not least, the molecular pathway of endothelial differentiation of dNC-PCs may not have been activated under the given conditions. Future research will focus on various strategies to generate osteogenic–vasculogenic constructs (for references see: Mikos et al. (2006)). One solution will include the supplementation with pro-angiogenic factors reported in the literature to induce the phenotype of endothelial cells from multipotent stem or progenitor cells. If the dNC-PC based single cell approach for tissue vascularization fails, the alternative will be to test mineralizing co-culturing systems with autologous endothelial progenitor cells. For example, the formation of microvessel-like structures was achieved in co-culture systems with cells relevant for bone tissue engineering, namely, human primary osteoblasts and human outgrowth endothelial cells (Fuchs et al., 2007). Similarly, specific cell–cell interactions between developing dNC-PCs and endothelial cells could eventually make the developing construct form proper bone. Osteogenic stimulation of dNC-PC cultures resulted in rapid formation of a cell multilayer consisting of two cell types. We observed flat cells covering the surface (outer cells) as well as cuboidal cells
Ö. Degistirici et al. / Matrix Biology 29 (2010) 219–227
225
Fig. 5. Quantification and localization of calcium (A, B) and of alkaline phosphatase activity (C, D) in dNC-PC cultures. The content of calcium was clearly upregulated in stimulated cells after 28 days when compared to controls (A). Accordingly, areas of strong mineralization in treated cultures became visible (stained by von Kossa) (B). Alkaline phosphatase activity was equally detectable in both stimulated cultures and controls throughout the experiment (C, D).
trapped within the layer (inner cells). As cells were viable, the diffusion of nutrients and the removal of waste products and cell metabolites may be sufficient for the given static culture system and the chosen cell density. Over time, the osteogenic construct was further formed dynamically by both cell populations depending on the state of their cellular differentiation. With prolonged culture time, the multilayer folded up, contracted and formed a free-floating spherical cell–matrix structure. The outer cells seemed to be essential to the maintenance of the construct, persisting as a sheath of living cells on its surface. These cells expressed alpha smooth muscle actin which is described in several stem cells (Cai et al., 2001; Kinner et al., 2002; Yamada et al., 2005) and myofibroblasts during tissue regeneration after injury (Chaponnier and Gabbiani, 2004; van Beurden et al., 2005). Indeed, smooth muscle actin-positive cells are described to be involved in the process of tissue regeneration of craniofacial bone (van Beurden et al., 2005). In the developing tooth tissue, alpha smooth muscle actin-positive cells are localized in the periodontal tissue near the alveolar bone. These cells may differentiate into osteoblasts and contribute to alveolar bone formation during wound repair (Hosoya et al., 2006). Furthermore, during the process of tissue regeneration after tooth replantation or transplantation, undifferentiated alpha smooth muscle actin-positive cells appeared in the damaged region and differentiated into osteoblast-like cells whereas no such cells were found around the newly formed dentin-like tissue (Zhao et al., 2007). It seems probable that the outer cells exhibit a similar function and thus may play an important role in the formation of the bone-like matrix under investigation. There was evidence that most of the inner cells underwent apoptosis. Although the mechanisms involved in this apoptotic
scenario are not yet known, apoptosis may facilitate the process of construct ossification (Huitema and Vaandrager, 2007). Whether this feature of dNC-PC differentiation is unique or shows certain parallels to chondrocyte hypertrophy and cartilage mineralization during the process of endochondral ossification has to be shown (Ballock and O'Keefe, 2003). Osteogenic-stimulated cultures of dNC-PCs manufactured and deposited an extracellular matrix representing the dynamic substratum to which the cells attached themselves and grew. It should be noted however, that biochemical studies on collagens, glycoproteins, glycosaminoglycans and other organic components have not yet been done. Here, we have undertaken a structural study of early constructs in the hope of obtaining information on the organization of the extracellular matrix. Light and electron microscope findings exhibited both calcified and uncalcified regions simultaneously. The material of the uncalcified regions showed irregularly arranged collagen fibrils as well as a fine filamentous background. Although not proved yet, the fibrillar collagen is thought to represent type I collagen, the most abundant protein of the bone matrix and contributing to its physical properties (Knott and Bailey, 1998; Burr, 2002). As our next step in understanding the bone-specific differentiation of dNC-PCs, the set of collagen genes expressed in cultured cells will be studied in detail. Most interestingly, the cells which were localized within this substance produced and secreted membrane-bound vesicles into the newly forming matrix. When examined by transmission electron microscopy, these vesicles looked like matrix vesicles which were first described by Ghadially et al. (1965), Anderson (1967) and Bonucci (1967). Matrix vesicles are known to be generated during biomineralization of growth plate cartilage, newly formed bone, tendon and the
226
Ö. Degistirici et al. / Matrix Biology 29 (2010) 219–227
Fig. 6. Spectroscopic analysis of the mineralized extracellular matrix in dNC-PC masses. (A) Bright field electron micrograph of calcium containing regions of the sample; the inset shows a detailed view (2.3-fold). (B) High angle annular dark field micrograph of the same region; an EELS spectrum imaging line scan was performed along the green line. The overlaid false color image reveals the intensity of the calcium L2,3 edge. (C) Two EELS spectra comparing a calcium containing (red) and a calcium-free region (green) of the sample; arrows (red, green) mark points in (B) where spectra were recorded. (D) Two EDS spectra of selected regions; numerous spectra were used for quantification of the Ca/P ratio (E). m1: measurement 1; m13: measurement 13.
dentine of teeth (Anderson et al., 2005). Therefore, it seems reasonable to conclude that both secretion of collagen and release of matrix vesicles from the surface of osteoblast-like cells may promote bone formation of the growing construct. However, unlike biomineralization of the cartilaginous skeleton, the way in which these components interact in mineralization of dNC-PC constructs is still far from being clear. Since apoptosis occurred within the construct (see above) it is not possible to be certain, for example, whether the release of soluble molecules from apoptotic cells and their local concentration (Anderson, 1995, 2003) may be involved in the process too. The relationship between collagen and vesicles may result in calcification of the extracellular matrix, converting the compacted mass to a calcified structure. To determine how mineralization of constructs occurs, requires an understanding of the mineral phase of
the matrix. Here, we described early state mineralized fibres. The results are in good accordance with other examples of biomineralization (Arnold et al., 2001). Since the crystallinity of samples was very low however, electron diffraction could not be used to identify the mineral phase per se (Grynpas et al., 1984). Even so, by combining spatial and chemical analyses, calcium was exclusively detected on or between the fibrillar meshwork. The intensity of the calcium signal was constant, leading to the assumption that crystallisation occurred at a constant rate and only one dominating calcium apatite phase was formed. The molar calcium–phosphorus-ratio in this phase was found to vary between 1.1 and 1.3 with a mean value of 1.11 +/− 0.06. This is in accordance with published data of amorphous calcium apatite (Dorozhkin and Epple, 2002). Thus, the mineral composition of the extracellular matrix points to a transient inorganic phase in bone
Ö. Degistirici et al. / Matrix Biology 29 (2010) 219–227
mineralization. Indeed, it is widely accepted that biomineralization proceeds through a series of intermediates, beginning with amorphous precipitates and hydrated crystallites and proceeding to the stable hydroxyapatite (Weiner et al., 2005; Weiner, 2006; Crane et al., 2006). If indeed amorphous calcium phosphate is the initial phase in bone formation, the osteogenic-stimulated cultures of dNC-PCs may represent an early stage of forming bone tissue. Confirmation of such a possibility might come from examination of cultures with more sophisticated methods (Crane et al., 2006), thus giving more insight into bone formation by stem or progenitor cells derived from the cranial neural crest. 5. Conclusions We report the spontaneous formation of a three-dimensional structure from monolayer cultured dNC-PCs when exposed to osteogenic stimulation in vitro. This cranial neural crest-derived cell culture system showed formation of bone-like matrix. We speculate that dNC-PCs secrete the extracellular matrix molecules that influence the further development of daughter cells, which require specific signals in order to differentiate. Thus, distinct stem or progenitor cells and their three-dimensional extracellular matrix stimulate each other leading to a self-propagating culture, which may resemble developing human craniofacial bone. By experimentally dissecting and studying the sequence of tissue formation, the (cranial-specific) osteogenic pathway will be further understood. Acknowledgments The skilful assistance of C. Sippel (Bonn), A. Sehrbrock (Bonn), G. Ladwig (Essen) and N. Cramer (Essen) is acknowledged with great gratitude. The authors would like to thank Dr. Naomi McGregor for her help in editing the manuscript, and Dr. Corinna Bernsdorff for help with processing images. Special thanks to Prof. H.P. Jennissen for comments about microparticles. This work was funded by Stiftung Center of Advanced European Studies and Research (caesar), Bonn. KWS and MT are acting joint senior authors concerning this work. References Abzhanov, A., Rodda, S.J., McMahon, A.P., Tabin, C.J., 2007. Regulation of skeletogenic differentiation in cranial dermal bone. Development 134, 3133–3144. Aejaz, H.M., Aleem, A.K., Parveen, N., Khaja, M.N., Narusu, M.L., Habibullah, C.M., 2007. Stem cell therapy—present status. Transplant. Proc. 39, 694–699. Anderson, H.C., 1967. Electron microscopic studies of induced cartilage development and calcification. J. Cell Biol. 35, 81–101. Anderson, H.C., 1995. Molecular biology of matrix vesicles. Clin. Orthop. Relat. Res. 266–280. Anderson, H.C., 2003. Matrix vesicles and calcification. Curr. Rheumatol. Rep. 5, 222–226. Anderson, H.C., Garimella, R., Tague, S.E., 2005. The role of matrix vesicles in growth plate development and biomineralization. Front Biosci. 10, 822–837. Arnold, S., Plate, U., Wiesmann, H.P., Stratmann, U., Kohl, H., Hohling, H.J., 2001. Quantitative analyses of the biomineralization of different hard tissues. J. Microsc. 202, 488–494. Atala, A., 2007. Engineering tissues, organs and cells. J. Tissue Eng. Regen. Med. 1, 83–96. Bajada, S., Mazakova, I., Richardson, J.B., Ashammakhi, N., 2008. Updates on stem cells and their applications in regenerative medicine. J. Tissue Eng. Regen. Med. 2, 169–183. Ballock, R.T., O'Keefe, R.J., 2003. Physiology and pathophysiology of the growth plate. Birth Defects Res. C. Embryo. Today 69, 123–143. Bonucci, E., 1967. Fine structure of early cartilage calcification. J. Ultrastruct. Res. 20, 33–50. Burr, D.B., 2002. The contribution of the organic matrix to bone's material properties. Bone 31, 8–11. Cai, D., Marty-Roix, R., Hsu, H.P., Spector, M., 2001. Lapine and canine bone marrow stromal cells contain smooth muscle actin and contract a collagen–glycosaminoglycan matrix. Tissue Eng. 7, 829–841.
227
Chai, Y., Maxson Jr., R.E., 2006. Recent advances in craniofacial morphogenesis. Dev. Dyn. 235, 2353–2375. Chaponnier, C., Gabbiani, G., 2004. Pathological situations characterized by altered actin isoform expression. J. Pathol. 204, 386–395. Crane, N.J., Popescu, V., Morris, M.D., Steenhuis, P., Ignelzi Jr., M.A., 2006. Raman spectroscopic evidence for octacalcium phosphate and other transient mineral species deposited during intramembranous mineralization. Bone 39, 434–442. Degistirici, O., Jaquiery, C., Schonebeck, B., Siemonsmeier, J., Gotz, W., Martin, I., Thie, M., 2008. Defining properties of neural crest-derived progenitor cells from the apex of human developing tooth. Tissue Eng. Part A 14, 317–330. Deng, Z.L., Sharff, K.A., Tang, N., Song, W.X., Luo, J., Luo, X., Chen, J., Bennett, E., Reid, R., Manning, D., Xue, A., Montag, A.G., Luu, H.H., Haydon, R.C., He, T.C., 2008. Regulation of osteogenic differentiation during skeletal development. Front Biosci. 13, 2001–2021. Dorozhkin, S.V., Epple, M., 2002. Biological and medical significance of calcium phosphates. Angew. Chem. Int. Ed. Engl. 41, 3130–3146. Dupin, E., Creuzet, S., Le Douarin, N.M., 2006. The contribution of the neural crest to the vertebrate body. Adv. Exp. Med. Biol. 589, 96–119. Fuchs, S., Hofmann, A., Kirkpatrick, C.J., 2007. Microvessel-like structures from outgrowth endothelial cells from human peripheral blood in 2-dimensional and 3-dimensional co-cultures with osteoblastic lineage cells. Tissue Eng. 13, 2577–2588. Ghadially, F.N., Meachim, G., Collins, D.H., 1965. Extra-cellular lipid in the matrix of human articular cartilage. Ann. Rheum. Dis. 24, 136–146. Grynpas, M.D., Bonar, L.C., Glimcher, M.J., 1984. Failure to detect an amorphous calcium-phosphate solid phase in bone mineral: a radial distribution function study. Calcif. Tissue Int. 36, 291–301. Han, J., Ishii, M., Bringas Jr., P., Maas, R.L., Maxson Jr., R.E., Chai, Y., 2007. Concerted action of Msx1 and Msx2 in regulating cranial neural crest cell differentiation during frontal bone development. Mech. Dev. 124, 729–745. Helms, J.A., Schneider, R.A., 2003. Cranial skeletal biology. Nature 423, 326–331. Hosoya, A., Nakamura, H., Ninomiya, T., Yoshiba, K., Yoshiba, N., Nakaya, H., Wakitani, S., Yamada, H., Kasahara, E., Ozawa, H., 2006. Immunohistochemical localization of alpha— smooth muscle actin during rat molar tooth development. J. Histochem. Cytochem. 54, 1371–1378. Huitema, L.F., Vaandrager, A.B., 2007. What triggers cell-mediated mineralization? Front Biosci. 12, 2631–2645. Kassem, M., 2006. Stem cells: potential therapy for age-related diseases. Ann. N. Y. Acad. Sci. 1067, 436–442. Kinner, B., Zaleskas, J.M., Spector, M., 2002. Regulation of smooth muscle actin expression and contraction in adult human mesenchymal stem cells. Exp. Cell Res. 278, 72–83. Knott, L., Bailey, A.J., 1998. Collagen cross-links in mineralizing tissues: a review of their chemistry, function, and clinical relevance. Bone 22, 181–187. Le Douarin, N.M., Dupin, E., Ziller, C., 1994. Genetic and epigenetic control in neural crest development. Curr. Opin. Genet. Dev. 4, 685–695. Mao, J.J., Giannobile, W.V., Helms, J.A., Hollister, S.J., Krebsbach, P.H., Longaker, M.T., Shi, S., 2006. Craniofacial tissue engineering by stem cells. J. Dent. Res. 85, 966–979. Mikos, A.G., et al., 2006. Engineering complex tissues. Tissue Eng. 12, 3307–3339. Redi, C.A., Monti, M., Merico, V., Neri, T., Zanoni, M., Zuccotti, M., Garagna, S., 2007. Stem cells. Endocr. Dev. 11, 145–151. Schoenebeck, B., Hartschen, H.J., Schindel, M., Degistirici, O., Siemonsmeier, J., Goetz, W., Thie, M., 2009. Molecular characterization of human impacted third molars: diversification of compartments. Cells Tissues. Organs 189, 356–370. Tucker, A.S., Lumsden, A., 2004. Neural crest cells provide species-specific patterning information in the developing branchial skeleton. Evol. Dev. 6, 32–40. van Beurden, H.E., Von den Hoff, J.W., Torensma, R., Maltha, J.C., Kuijpers-Jagtman, A.M., 2005. Myofibroblasts in palatal wound healing: prospects for the reduction of wound contraction after cleft palate repair. J. Dent. Res. 84, 871–880. Weiner, S., 2006. Transient precursor strategy in mineral formation of bone. Bone 39, 431–433. Weiner, S., 2008. Biomineralization: a structural perspective. J. Struct. Biol. 163, 229–234. Weiner, S., Sagi, I., Addadi, L., 2005. Structural biology. Choosing the crystallization path less traveled. Science 309, 1027–1028. Xu, Y., Malladi, P., Zhou, D., Longaker, M.T., 2007. Molecular and cellular characterization of mouse calvarial osteoblasts derived from neural crest and paraxial mesoderm. Plast. Reconstr. Surg. 120, 1783–1795. Yamada, M., Kurihara, H., Kinoshita, K., Sakai, T., 2005. Temporal expression of alphasmooth muscle actin and drebrin in septal interstitial cells during alveolar maturation. J. Histochem. Cytochem. 53, 735–744. Young, H.E., Steele, T.A., Bray, R.A., Detmer, K., Blake, L.W., Lucas, P.W., Black Jr., A.C., 1999. Human pluripotent and progenitor cells display cell surface cluster differentiation markers CD10, CD13, CD56, and MHC class-I. Proc. Soc. Exp. Biol. Med. 221, 63–71. Zhao, C., Zhao, C., Hosoya, A., Kurita, H., Hu, T., Hiraga, T., Ninomiya, T., Yoshiba, K., Yoshiba, N., Takahashi, M., Kurashina, K., Ozawa, H., Nakamura, H., 2007. Immunohistochemical study of hard tissue formation in the rat pulp cavity after tooth replantation. Arch. Oral Biol. 52, 945–953.
Contents lists available at ScienceDirect
Matrix Biology j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / m a t b i o
Using human neural crest-derived progenitor cells to investigate osteogenesis: An in vitro study Özer Degistirici a,⁎,1, Florian Grabellus b, Stephan Irsen a, Kurt Werner Schmid b, Michael Thie a,⁎,2 a b
Center of Advanced European Studies and Research (caesar), Bonn, Germany Institute of Pathology and Neuropathology, University Hospital of Essen, University of Duisburg-Essen, Germany
a r t i c l e
i n f o
Article history: Received 7 July 2009 Received in revised form 14 December 2009 Accepted 14 December 2009 Keywords: Human tooth Progenitor cells Neural crest Differentiation In vitro osteogenesis Bone matrix
a b s t r a c t Human tooth contains a distinct population of neural crest-derived progenitor cells (dNC-PCs) which are known to give rise to specialized daughter cells of an osteogenic lineage. We hypothesised that dNC-PCs could develop into neural crest-derived bone in a self-propagating and extracorporal culture system. Thus, we examined the three-dimensional structure obtained from osteogenic-stimulated dNC-PCs by morphological, biochemical and spectroscopic methods. After the onset of stimulation, cells formed a multilayer with outer cells covering the surface and inner cells secreting a hyaline matrix. With prolonged culture, multilayers contracted and formed a three-dimensional construct which subsequently converted to a calcified mass. Differentiation of progenitor cells was associated with apoptosis. Cell types which survived were smooth muscle actin-positive cells and bone-like cells. The expression of osteoblastic markers and the secretion of a collagenous matrix indicate that the bone cells had acquired their functional phenotype. Furthermore, these cells produced and secreted membrane-bound vesicles into the newly forming matrix. Consequently, an early biomineralized extracellular matrix was found with calcium phosphate deposits being associated with the newly formed collagen matrix framework. The molar calcium–phosphorus-ratio of the mineralized collagen indicated that amorphous calcium phosphate was present within this matrix. The data suggest that stimulated cultures of dNC-PCs are able to recapitulate some processes of the early phase of osteogenesis. © 2009 Elsevier B.V. All rights reserved.
1. Introduction A variety of different stem cells have been identified in many animal and human tissues (Young et al., 1999). Adult stem or progenitor cells are being widely investigated with a focus on their ability to differentiate into a broad array of cell types potentially useful as cell material for purposes of tissue engineering and regenerative medicine (Kassem, 2006; Aejaz et al., 2007; Atala, 2007; Redi et al., 2007; Bajada et al., 2008). The developing field of craniofacial tissue engineering promises the regeneration of dental, oral, and craniofacial structures (Mao et al., 2006). Recent evidence
⁎ Corresponding authors. Degistirici is to be contacted at Universitätsklinikum Düsseldorf, Klinik für Kinder-Onkologie, -Hämatologie und-Immunologie, Geb. 13.42, Moorenstrasse 5, D-40225 Düsseldorf, Germany. Tel.: + 49 211 811 6277; fax: + 49 211 811 6436. Thie, Institut für Anatomie und Klinische Morphologie, Private Universität Witten/Herdecke gGmbH, Alfred-Herrhausen-Straße 50, D-58448 Witten, Germany. Tel.: + 49 2302 926 721; fax: + 49 2302 926 739. E-mail addresses: [email protected] (Ö. Degistirici), [email protected] (M. Thie). 1 Current address: Department of Paediatric Oncology, Haematology and Immunology, Düsseldorf University Hospital, Düsseldorf, Germany. 2 Current address: Institute of Anatomy and Clinical Morphology, Faculty of Medicine, University Witten/Herdecke, Witten, Germany. 0945-053X/$ – see front matter © 2009 Elsevier B.V. All rights reserved. doi:10.1016/j.matbio.2009.12.005
suggests that the connective tissue of the human tooth contains a particular population of neural crest-derived progenitor cells (termed dNC-PCs) which give rise to a wide range of specialized daughter cells when outside their dental signal network (Degistirici et al., 2008; Schoenebeck et al., 2009). In attempting to use dNC-PCs for tissue engineering we showed that it was possible to induce bone formation in a nude mouse model by combining dNC-PCs with bovine bonederived granulates, which act as a scaffolding component. Under these conditions, the only differentiated stem cells observed were those of an osteogenic lineage (Degistirici et al., 2008). To take full advantage of dNC-PCs, however, it will be important to understand how these cells differentiate into bone-specific cell types and what environmental conditions are stimulating these cells to become functional bone. In craniofacial tissues, the bone cell lineages originate from cranial neural crest cells that have committed to the osteogenic cell lineage becoming osteoprogenitor cells, such as preosteoblasts, osteoblasts, and osteocytes (Le Douarin et al., 1994; Dupin et al., 2006; Chai and Maxson, 2006). It turns out that the molecular mechanisms inducing osteogenesis in cranial neural crest cells, which produce the facial and jaw skeleton, are distinct from those operating in mesodermal cells, which produce most of the skeleton (Helms and Schneider, 2003; Tucker and Lumsden, 2004; Abzhanov et al., 2007; Han et al., 2007; Xu et al., 2007; Deng et al., 2008). Therefore, our focus is on the specific
220
Ö. Degistirici et al. / Matrix Biology 29 (2010) 219–227
biology of the process of dNC-PCs developing into bone-like tissue. Aspects considered are the differentiation of dNC-PCs from stem or progenitor cells to cells of the osteogenic lineage and the cellular changes that accompany their differentiation. Our current notion is that the extracellular matrix components and their structural architecture trigger this differentiation process, thus we investigated the potential of dNC-PCs for osteogenesis and bone development when cultured in osteogenic growth medium and maintained within a self-established extracorporal microenvironment. Here, we report that dNC-PC cultures responded to osteogenic stimulation by progressing through a series of structural and biochemical changes that culminated in the formation of a simply ordered bone-like matrix. Although we are left with several questions, these data give new insights into the molecular processes of human bone formation by stem or progenitor cells derived from the neural crest. 2. Materials and methods 2.1. Cell isolation and expansion culture Surgically removed impacted third molars of young adults were used to prepare dental neural crest-derived progenitor cells (dNCPCs) (Degistirici et al., 2008; Schoenebeck et al., 2009). Written consent was obtained from all parents of the participating patients. Briefly, apical pad-like tissue of the developing tooth was cut into small pieces, enzymatically pretreated with collagenase/dispase solution (Sigma, Munich, Germany) and microexplants seeded in tissue culture flasks (BD Falcon, Germany). Subsequently, microexplants were maintained in Dulbecco's modified Eagle's mediumlow glucose (DMEM-LG) (Cambrex Bio Science, Verviers, Belgium) supplemented with 2 mM glutamine and 1% penicillin/streptomycin (Sigma) and 10% fetal calf serum (FCS) (Biochrom, Berlin, Germany). Upon reaching 70–80% confluence, outgrowing cells were detached by trypsinization (trypsin/EDTA) (Sigma) and replated at a density of 5 × 10³ cells/cm². The resulting cells were further incubated under standard culture conditions for expansion. For experiments, dNC-PCs were used at passages 3 to 4. 2.2. Osteogenic culture For osteogenic induction, cultured dNC-PCs were initially seeded at a density of 8 × 103 cells/cm2 in 24-well plates (BD Falcon, Germany) and cultured in growth medium containing DMEM-LG, 10% FCS and 1% penicillin/streptomycin. When reaching subconfluency, the growth medium was replaced with differentiation medium, DMEM-LG supplemented with 100 nM dexamethasone, 50 μM ascorbic acid-2 phosphate, and 10 mM ß-glycerol phosphate (all from Sigma). The medium was changed twice a week. For a threedimensional culture, cells were seeded at a density of 5 × 103 cells/cm2 in 25-cm² culture flasks and incubated in osteogenic medium as described above. Formation of a three-dimensional structure was brought about by prolonged culture. Samples were analyzed at the indicated time points. 2.3. Assay of alkaline phosphatase activity Alkaline phosphatase activity was determined in cell lysates using p-nitrophenyl phosphate as a substrate (Sigma). The procedure was carried out as recommended by the manufacturer. Briefly, cells growing on 24-well plates were washed with phosphate-buffered saline (PBS) and incubated with 1% Triton X-100 (Sigma). The resulting lysate was then incubated with substrate and the released p-nitrophenol was measured at 405 nm (PerkinElmer, Waltham, MA). Alkaline phosphatase activity was expressed as μM p-nitrophenol/l/min/sample.
2.4. Quantification of calcium The total calcium of the samples was measured by the o-cresolphthalein complexone method using the commercial Calcium Assay-CA590 kit (Randox Laboratories, Co Antrim, United Kingdom). The procedure was carried out according to the manufacturer's protocol. Briefly, cells growing on 24-well plates were washed with PBS and extracted in 200 μl of 0.5 N hydrochloric acid for 5 min. Samples were then vigorously shaken for 4–16 h at 4 °C. Calcium determination was done in 96-well plates with 10 μl of test solution and 90 μl of substrate solution mixed up with 100 μl of distilled water. The amount of deposited calcium was determined at 570 nm using a spectrofluorometer (PerkinElmer) and was expressed as μg/sample. 2.5. TUNEL assay In situ DNA fragmentation was established using the terminal desoxyribonucleotide transferase (TdT)-mediated dUTP nick-endlabelling technique (TUNEL) in paraffin-embedded sections. We used the ApoTag™ plus peroxidase in situ apoptosis detection kit (KIT S7101, Intergen, USA). The staining procedures were performed according to the manufacturer's recommendations. Slides of colorectal cancer (CRC) treated in the same way served as positive control. Negative controls were carried out with CRC slides without exposure to TdT enzyme. 2.6. Histology and immunohistochemistry The cultured cells were harvested, paraffin-embedded and sectioned following standard protocols. Briefly, samples were fixed in 4% formaldehyde in PBS, washed with PBS and dehydrated with ethanol. After dehydration, samples were washed with toluene, infiltrated with molten paraffin and embedded and sectioned. Sections of 5 μm were prepared for a panel of immunohistochemical examinations. Dewaxed and rehydrated sections were incubated with hydrogen peroxide to block endogenous peroxidase, antigen retrieval was performed in a hot water bath. The following monoclonal primary antibodies were used: anti-smooth-muscle actin (dilution 1:500; Dako, Hamburg, Germany), anti-osteocalcin (1:500; Acris-Antibodies, Herford, Germany), and anti-collagen type II (1:500; MP-Biomedicals, Illkirch, France). For primary antibody detection the Zytomed Polymer Kit DAB (dilution 1:10; Transduction Laboratories, San Diego, CA, U.S. A) was used. All immunohistochemical stains were performed with an automated staining device (Dako Autostainer, Glostrup, Denmark). Omission of the primary antibodies served as negative controls. Selected sections were stained with hematoxylin and eosin. To identify the formation of mineralization, monolayers were stained with von Kossa. To detect alkaline phosphatase, monolayers were stained with naphtol and fast red violet. 2.7. Electron microscopy For transmission electron microscopy, samples were fixed in 2.5% glutaraldehyde in 0.1 M cacodylate buffer, pH 7.2, for 2 h at room temperature, washed in cacodylate buffer, post-fixed with 1% osmium tetroxide in cacodylate buffer, dehydrated with ethanol and propylene oxide and embedded in epoxy resin. Ultrathin sections were mounted on copper grids, double-stained with uranyl acetate and lead citrate and examined with a Zeiss TEM 902A. For scanning electron microscopy, samples were fixed in 2.5% glutaraldehyde in 0.1 M cacodylate buffer, pH 7.2, for 2 h at room temperature. After repeated washings in cacodylate buffer, samples were dehydrated with ethanol and dried using hexamethyldisilazane (Polysciences, Eppelheim, Germany) as the drying medium. Then samples were sputtered with a conductive layer of gold and imaged
Ö. Degistirici et al. / Matrix Biology 29 (2010) 219–227
with a Leo SEM Supra 55. For some analysis, samples were imaged with a LEO XB1540 cross beam.
2.8. EELS/EDS analysis Electron energy loss spectroscopy (EELS) was performed using a Zeiss TEM Libra 200 (CRISP edition) with imaging in column energyfilter and detector for energy dispersive X-rays (EDS). The experiments were performed in scanning (STEM) mode with an energy resolution of 0.5 eV. Experiments were taken at the L-edge of calcium (350 eV) and the L-edge of oxygen (536 eV). Energy dispersive X-ray analysis (EDS) was performed in STEM mode at a convergence angle of 25 mrad. The spot size of the scanning probe was approximately 1 nm. For quantitative analysis rectangular regions of interest with approximately 50 nm edge length were selected and analyzed in an energy range between 0 and 20 keV. The spot diameter for line scans was in the range of 1 nm.
3. Results 3.1. Preparation of progenitor cells Dental neural crest-derived progenitor cells (dNC-PCs) were prepared from the pad-like tissue of impacted third molars of young adults using the microexplant technique. When subcultured as monolayers in standard growth medium, dNC-PCs showed the distinct phenotype of multipotent progenitor cells as described previously (Degistirici et al., 2008; Schoenebeck et al., 2009). To induce cells of osteogenic lineage for experimentation, dNC-PCs were initially preincubated for 4–7 days with growth medium in the presence of 10% FCS. Then, dexamethasone, ascorbic acid-2 phosphate and ß-glycerol phosphate were added to the pre-incubation medium, and culture systems were further cultivated. dNC-PCs grown without osteogenic factors were used as control.
3.2. Formation of a three-dimensional culture system In general, dNC-PCs responded to osteogenic stimulation through a succession of stages of developing structural complexity (Fig. 1). The first change was a significant increase in monolayer height (Fig. 1A, B) that was due to an increase in cell numbers as well as an increase in the proportion of extracellular matrix (Fig. 1A). After 14 days of stimulation, cells had formed a multilayered structure with the development of several distinct nodules scattered within the culture system (Fig. 1B). Two types of cellular layering could be observed: an outer one and an inner one. Flat cells covering the surface represented the outer cells, cuboidal cells trapped within the layered structure represented the inner cells. Inner cells secreted a hyaline matrix with which they surrounded themselves. As cell growth and extracellular matrix production continued, the overall shape of the culture system changed (Fig. 1C–F). At around 28 days of stimulation, multilayered cultures detached from the plastic, folded up and upon contraction, formed a free-floating spherical cell–matrix mass (Fig. 1C; D). Outer cells grew on the surface of the mass, thus forming a circumferential sheath of living cells (Fig. 1E, F). At the same time as contraction, the spherical mass converted to a calcified structure and highly mineralized areas began to form. Histological sections of the non-decalcified mass showed the presence of mineralized extracellular matrix (Fig. 1E, F). Inner cells produced matrix and then mineralized it whereas outer cells did not. Outer cells persisted on the surface of the mass (Fig. 2A), thus enveloping the calcifying core (Fig. 2B). In contrast, dNC-PCs cultured without osteogenic factors continued as non-calcified monolayers remaining attached to the surface of the plastic dish (data not shown).
221
3.3. Diversification of cell phenotype Different phenotypes of dNC-PCs in different parts of the selfforming culture were observed (Fig. 3). Specialized types of cells were identified by various markers, such as smooth muscle actin, which identifies a distinct subset of contractile cells (Fig. 3A, B). Typically, smooth muscle actin-positive cells were found within the sheath of living cells on the surface of the culture system where they were localized between cells that did not react with the antibody. A disparity between inner cells/inner part and outer cells/outer part as a result of apoptosis was observed (Fig. 3C, D). If culture systems were analyzed at an early stage, i.e. 28 days after stimulation, only a few cells undergoing apoptosis were detected among the outer cells and none among inner cells. Conversely, after 42 days of stimulation, a large proportion of the inner cells were dying by apoptosis while outer cells were not and thus were saved. Differences also existed when looking at cells specialized for secretion of extracellular matrix proteins, for example, osteoblast-specific osteocalcin (Fig. 3E, F). If culture systems were analyzed at an early stage, i.e. 28 days after stimulation, only a few dispersed cells in the outer layer were positive for an antibody against osteocalcin. In contrast, 42 days after stimulation, a large number of osteocalcin-positive cells were found in the core region of the mass while the outer cells did not react. Type II collagen antibodies were used as a typical marker for a chondrogenic extracellular matrix. When analyzing the culture systems by immunohistochemistry, no reaction of antibodies was observed.
3.4. Ultrastructural features Electron micrographs of dNC-PC cell types and their surrounding matrices which formed 28 days after stimulation are shown in Fig. 4. Outer part cells which grew as a circumferential sheath of cells lining the spherical mass appeared mostly spindle-shaped and exhibited a fibroblast-like phenotype (Fig. 4A). The nucleus was elongated and showed distinctive invaginations. The cytoplasm was filled with elements of the endoplasmic reticulum, Golgi complexes and mitochondria. Small and large vesicles were present. A basal lamina and depositions of extracellular matrix were not observed, instead cells were in close contact with each other. Inner part cells appeared more spherical and exhibited a synthetic phenotype characterized by a prominent rough endoplasmic reticulum (Fig. 4B). Cells grew a large number of cytoplasmic extensions and were surrounded by depositions of extracellular matrix. Scanning electron micrographs of inner part cells displayed numerous membrane-bound vesicles budding off from the free cell surface (Fig. 4C–E). When observed by transmission electron microscopy vesicles were seen to be filled with an amorphous osmiophilic substance (Fig. 4D). The extracellular space between cells was endowed with a meshwork of characteristic fibrelike structures (Fig. 4F). Some of the ordered structures appeared as banded fibrils, resembling thin (40–50 nm in diameter) cable-like constructs.
3.5. Quantification of initial calcification and alkaline phosphatase activity When matrix molecules were deposited by osteogenic-stimulated dNC-PCs, a conversion of soft matrix into hard matrix was observed (Fig. 1; Fig. 5). After 28 days of stimulation, multilayered cultures of dNC-PCs were shown to contain substantial deposits of calcium salts indicating the biological mineralization process of the extracellular matrix. A quantitative analysis of total calcium showed that stimulated cultures exhibited higher values than controls (Fig. 5A, B). Alkaline phosphatase activity increased continuously in osteogenic-stimulated cultures as well as in controls (Fig. 5C, D).
222
Ö. Degistirici et al. / Matrix Biology 29 (2010) 219–227
Fig. 1. Histological sections showing steps in the process of organization of dNC-PC cultures. When growing on plastic, cells make contact to the surface, proliferate and secrete an extracellular matrix with which the inner cells surround themselves (A). Thus, a multilayer with several distinct nodules scattered within the culture is formed (B). As the culture grows, the number of cells and the quantity of extracellular matrix become much greater. Next, the multilayer detaches from the plastic surface, folds up and forms a roundish mass (C). With prolonged culture time, masses get more compact (D). The matrix calcifies soon after the cultures have formed masses. Note the smooth cells that still line the mass (E, F). Medium: growth medium; plastic: plastic surface; calcification: a compact portion of calcified mass; asterisk: the surrounding layer of cells; rhomb: the inner sheets of cells; A + B: same magnification; C + D: same magnification.
3.6. Spectroscopic analysis of biological mineralization After a fine meshwork of fibrils, of varying diameter, had formed in the extracellular space of osteogenic-stimulated cultures (Fig. 6A), composites were characterized by EELS spectroscopy and EDS analysis (Fig. 6B–E). EELS spectroscopy revealed the intensity of the calcium L2,3 edge of several scans, proving calcification (Fig. 6B). Fig. 6C shows two EELS spectra of calcium-rich and calcium-free areas of the sample which confirmed the presence of calcium was restricted to the fibrillar meshwork, where the intensity of the calcium signal was reasonably constant. For further identification of this phase, EDS analysis at numerous positions was performed (Fig. 6D) and used for quantification (Fig. 6E). The calcium–phosphorus-ratio in this phase varied between 1.1 and 1.3. The mean ratio of 13 selected regions was
1.11 +/− 0.06 indicating the presence of amorphous calcium phosphate (Fig. 6E). 4. Discussion The starting point for this study were our recent reports that identified dental neural crest-derived progenitor cells (dNC-PCs) as good candidates to regenerate neural crest-derived tissues within the craniofacial region (Degistirici et al., 2008; Schoenebeck et al., 2009). As dNC-PCs could be easily differentiated into osteogenic lineages we speculated whether these cells may be useful to engineer facial and jaw skeleton. Indeed, the growing realization that skeletogenesis in the head is a unique and separable process from that occurring elsewhere in the body makes this a crucial new approach to take
Ö. Degistirici et al. / Matrix Biology 29 (2010) 219–227
223
Fig. 2. Scanning electron micrographs of fractures of dNC-PC masses. Note the many folds, i.e. sheets of cell layers, which are visible within the calcified mass (A). Note also the surrounding layer of smooth cells (asterisk) enwrapping the calcified mass (B). Boxed area: drawing of a sheet of cells; asterisk: the surrounding layer of cells; core: calcified mass.
Fig. 3. Localization of cells positive for smooth muscle actin (SMA) (A, B), apoptotic nuclei (TUNEL) (C, D), and osteocalcin (OCN) (E, F) in young (A, C, E) and old (B, D, F) stages of dNC-PC cultures. SMA-positive cells are always found in the outer part of the culture, i.e. the surrounding layer. In contrast, TUNEL- and OCN-positive cells are found in the inner part. Note the strong increase in the number of TUNEL- and OCN-positive cells with prolonged culture time (D, F).
224
Ö. Degistirici et al. / Matrix Biology 29 (2010) 219–227
Fig. 4. Electron micrographs of cell types and their matrix seen in dNC-PC masses. (A) Transmission electron micrograph of a section of the outer part of the structure, showing single cells in their normal context. Cells exhibit an elongated mesenchymal phenotype and are in close contact to each other. Note lack of matrix deposits. (B) Photograph of a section of the inner part of the structure. Cells have several cellular processes and are filled with an extensive rough endoplasmic reticulum, indicating an active phenotype. Cells seem responsible for secretion of the matrix surrounding them. (C, E) Scanning electron micrographs of active cells of the inner portion of the spheroid. Cell surfaces and cell processes are covered by numerous vesicles (globular structures). As shown in (E), spaces between cells are filled with extracellular matrix collapsed during the dehydration process for specimen preparation. (D, E) Transmission electron micrographs of calcified/calcifying vesicles, containing amorphous deposits. Note the loosely packed collagen fibrils which surround the cells. The characteristic features of collagen fibril ultrastructure are shown in (F). n: nucleus; v: vesicle; ECM: extracellular matrix; arrows: regular banding of fibrils.
(Helms and Schneider, 2003). In this report, we provide further evidence, that dNC-PCs are able to produce an osteogenic matrix via a self-propagating extracorporal culture system. The construct is composed of an outer cell layer surrounding a compact mass, which itself consists of bone cells embedded in an organic collagenous framework containing calcium phosphate deposits. As the biomineralized matrix consists of amorphous calcium phosphate, i.e. the precursor phase of calcium-deficient hydroxyapatite or hydroxyapatite found in mature bone (Weiner, 2006, 2008), this kind of cranial neural crest cell-produced bone may be of transient (early) nature. Our analysis of dNC-PC cultures autonomously differentiating in vitro revealed that cultures are developing along a pathway that should eventually lead to mature bone formation. The expression of osteoblastic markers like alkaline phosphatase and osteocalcin and the secretion of a mineralizing matrix indicates that cells had acquired a functional bone-forming phenotype. It should be noted however, that cultures of osteogenic-stimulated dNC-PCs did not develop into fully organized and anatomically structured bone. In particular, sitespecific formation of blood vessels that would support the metabolic needs of mature tissue was not observed. The lack of formation of
capillary-like structures may be due to a number of factors, not least, the molecular pathway of endothelial differentiation of dNC-PCs may not have been activated under the given conditions. Future research will focus on various strategies to generate osteogenic–vasculogenic constructs (for references see: Mikos et al. (2006)). One solution will include the supplementation with pro-angiogenic factors reported in the literature to induce the phenotype of endothelial cells from multipotent stem or progenitor cells. If the dNC-PC based single cell approach for tissue vascularization fails, the alternative will be to test mineralizing co-culturing systems with autologous endothelial progenitor cells. For example, the formation of microvessel-like structures was achieved in co-culture systems with cells relevant for bone tissue engineering, namely, human primary osteoblasts and human outgrowth endothelial cells (Fuchs et al., 2007). Similarly, specific cell–cell interactions between developing dNC-PCs and endothelial cells could eventually make the developing construct form proper bone. Osteogenic stimulation of dNC-PC cultures resulted in rapid formation of a cell multilayer consisting of two cell types. We observed flat cells covering the surface (outer cells) as well as cuboidal cells
Ö. Degistirici et al. / Matrix Biology 29 (2010) 219–227
225
Fig. 5. Quantification and localization of calcium (A, B) and of alkaline phosphatase activity (C, D) in dNC-PC cultures. The content of calcium was clearly upregulated in stimulated cells after 28 days when compared to controls (A). Accordingly, areas of strong mineralization in treated cultures became visible (stained by von Kossa) (B). Alkaline phosphatase activity was equally detectable in both stimulated cultures and controls throughout the experiment (C, D).
trapped within the layer (inner cells). As cells were viable, the diffusion of nutrients and the removal of waste products and cell metabolites may be sufficient for the given static culture system and the chosen cell density. Over time, the osteogenic construct was further formed dynamically by both cell populations depending on the state of their cellular differentiation. With prolonged culture time, the multilayer folded up, contracted and formed a free-floating spherical cell–matrix structure. The outer cells seemed to be essential to the maintenance of the construct, persisting as a sheath of living cells on its surface. These cells expressed alpha smooth muscle actin which is described in several stem cells (Cai et al., 2001; Kinner et al., 2002; Yamada et al., 2005) and myofibroblasts during tissue regeneration after injury (Chaponnier and Gabbiani, 2004; van Beurden et al., 2005). Indeed, smooth muscle actin-positive cells are described to be involved in the process of tissue regeneration of craniofacial bone (van Beurden et al., 2005). In the developing tooth tissue, alpha smooth muscle actin-positive cells are localized in the periodontal tissue near the alveolar bone. These cells may differentiate into osteoblasts and contribute to alveolar bone formation during wound repair (Hosoya et al., 2006). Furthermore, during the process of tissue regeneration after tooth replantation or transplantation, undifferentiated alpha smooth muscle actin-positive cells appeared in the damaged region and differentiated into osteoblast-like cells whereas no such cells were found around the newly formed dentin-like tissue (Zhao et al., 2007). It seems probable that the outer cells exhibit a similar function and thus may play an important role in the formation of the bone-like matrix under investigation. There was evidence that most of the inner cells underwent apoptosis. Although the mechanisms involved in this apoptotic
scenario are not yet known, apoptosis may facilitate the process of construct ossification (Huitema and Vaandrager, 2007). Whether this feature of dNC-PC differentiation is unique or shows certain parallels to chondrocyte hypertrophy and cartilage mineralization during the process of endochondral ossification has to be shown (Ballock and O'Keefe, 2003). Osteogenic-stimulated cultures of dNC-PCs manufactured and deposited an extracellular matrix representing the dynamic substratum to which the cells attached themselves and grew. It should be noted however, that biochemical studies on collagens, glycoproteins, glycosaminoglycans and other organic components have not yet been done. Here, we have undertaken a structural study of early constructs in the hope of obtaining information on the organization of the extracellular matrix. Light and electron microscope findings exhibited both calcified and uncalcified regions simultaneously. The material of the uncalcified regions showed irregularly arranged collagen fibrils as well as a fine filamentous background. Although not proved yet, the fibrillar collagen is thought to represent type I collagen, the most abundant protein of the bone matrix and contributing to its physical properties (Knott and Bailey, 1998; Burr, 2002). As our next step in understanding the bone-specific differentiation of dNC-PCs, the set of collagen genes expressed in cultured cells will be studied in detail. Most interestingly, the cells which were localized within this substance produced and secreted membrane-bound vesicles into the newly forming matrix. When examined by transmission electron microscopy, these vesicles looked like matrix vesicles which were first described by Ghadially et al. (1965), Anderson (1967) and Bonucci (1967). Matrix vesicles are known to be generated during biomineralization of growth plate cartilage, newly formed bone, tendon and the
226
Ö. Degistirici et al. / Matrix Biology 29 (2010) 219–227
Fig. 6. Spectroscopic analysis of the mineralized extracellular matrix in dNC-PC masses. (A) Bright field electron micrograph of calcium containing regions of the sample; the inset shows a detailed view (2.3-fold). (B) High angle annular dark field micrograph of the same region; an EELS spectrum imaging line scan was performed along the green line. The overlaid false color image reveals the intensity of the calcium L2,3 edge. (C) Two EELS spectra comparing a calcium containing (red) and a calcium-free region (green) of the sample; arrows (red, green) mark points in (B) where spectra were recorded. (D) Two EDS spectra of selected regions; numerous spectra were used for quantification of the Ca/P ratio (E). m1: measurement 1; m13: measurement 13.
dentine of teeth (Anderson et al., 2005). Therefore, it seems reasonable to conclude that both secretion of collagen and release of matrix vesicles from the surface of osteoblast-like cells may promote bone formation of the growing construct. However, unlike biomineralization of the cartilaginous skeleton, the way in which these components interact in mineralization of dNC-PC constructs is still far from being clear. Since apoptosis occurred within the construct (see above) it is not possible to be certain, for example, whether the release of soluble molecules from apoptotic cells and their local concentration (Anderson, 1995, 2003) may be involved in the process too. The relationship between collagen and vesicles may result in calcification of the extracellular matrix, converting the compacted mass to a calcified structure. To determine how mineralization of constructs occurs, requires an understanding of the mineral phase of
the matrix. Here, we described early state mineralized fibres. The results are in good accordance with other examples of biomineralization (Arnold et al., 2001). Since the crystallinity of samples was very low however, electron diffraction could not be used to identify the mineral phase per se (Grynpas et al., 1984). Even so, by combining spatial and chemical analyses, calcium was exclusively detected on or between the fibrillar meshwork. The intensity of the calcium signal was constant, leading to the assumption that crystallisation occurred at a constant rate and only one dominating calcium apatite phase was formed. The molar calcium–phosphorus-ratio in this phase was found to vary between 1.1 and 1.3 with a mean value of 1.11 +/− 0.06. This is in accordance with published data of amorphous calcium apatite (Dorozhkin and Epple, 2002). Thus, the mineral composition of the extracellular matrix points to a transient inorganic phase in bone
Ö. Degistirici et al. / Matrix Biology 29 (2010) 219–227
mineralization. Indeed, it is widely accepted that biomineralization proceeds through a series of intermediates, beginning with amorphous precipitates and hydrated crystallites and proceeding to the stable hydroxyapatite (Weiner et al., 2005; Weiner, 2006; Crane et al., 2006). If indeed amorphous calcium phosphate is the initial phase in bone formation, the osteogenic-stimulated cultures of dNC-PCs may represent an early stage of forming bone tissue. Confirmation of such a possibility might come from examination of cultures with more sophisticated methods (Crane et al., 2006), thus giving more insight into bone formation by stem or progenitor cells derived from the cranial neural crest. 5. Conclusions We report the spontaneous formation of a three-dimensional structure from monolayer cultured dNC-PCs when exposed to osteogenic stimulation in vitro. This cranial neural crest-derived cell culture system showed formation of bone-like matrix. We speculate that dNC-PCs secrete the extracellular matrix molecules that influence the further development of daughter cells, which require specific signals in order to differentiate. Thus, distinct stem or progenitor cells and their three-dimensional extracellular matrix stimulate each other leading to a self-propagating culture, which may resemble developing human craniofacial bone. By experimentally dissecting and studying the sequence of tissue formation, the (cranial-specific) osteogenic pathway will be further understood. Acknowledgments The skilful assistance of C. Sippel (Bonn), A. Sehrbrock (Bonn), G. Ladwig (Essen) and N. Cramer (Essen) is acknowledged with great gratitude. The authors would like to thank Dr. Naomi McGregor for her help in editing the manuscript, and Dr. Corinna Bernsdorff for help with processing images. Special thanks to Prof. H.P. Jennissen for comments about microparticles. This work was funded by Stiftung Center of Advanced European Studies and Research (caesar), Bonn. KWS and MT are acting joint senior authors concerning this work. References Abzhanov, A., Rodda, S.J., McMahon, A.P., Tabin, C.J., 2007. Regulation of skeletogenic differentiation in cranial dermal bone. Development 134, 3133–3144. Aejaz, H.M., Aleem, A.K., Parveen, N., Khaja, M.N., Narusu, M.L., Habibullah, C.M., 2007. Stem cell therapy—present status. Transplant. Proc. 39, 694–699. Anderson, H.C., 1967. Electron microscopic studies of induced cartilage development and calcification. J. Cell Biol. 35, 81–101. Anderson, H.C., 1995. Molecular biology of matrix vesicles. Clin. Orthop. Relat. Res. 266–280. Anderson, H.C., 2003. Matrix vesicles and calcification. Curr. Rheumatol. Rep. 5, 222–226. Anderson, H.C., Garimella, R., Tague, S.E., 2005. The role of matrix vesicles in growth plate development and biomineralization. Front Biosci. 10, 822–837. Arnold, S., Plate, U., Wiesmann, H.P., Stratmann, U., Kohl, H., Hohling, H.J., 2001. Quantitative analyses of the biomineralization of different hard tissues. J. Microsc. 202, 488–494. Atala, A., 2007. Engineering tissues, organs and cells. J. Tissue Eng. Regen. Med. 1, 83–96. Bajada, S., Mazakova, I., Richardson, J.B., Ashammakhi, N., 2008. Updates on stem cells and their applications in regenerative medicine. J. Tissue Eng. Regen. Med. 2, 169–183. Ballock, R.T., O'Keefe, R.J., 2003. Physiology and pathophysiology of the growth plate. Birth Defects Res. C. Embryo. Today 69, 123–143. Bonucci, E., 1967. Fine structure of early cartilage calcification. J. Ultrastruct. Res. 20, 33–50. Burr, D.B., 2002. The contribution of the organic matrix to bone's material properties. Bone 31, 8–11. Cai, D., Marty-Roix, R., Hsu, H.P., Spector, M., 2001. Lapine and canine bone marrow stromal cells contain smooth muscle actin and contract a collagen–glycosaminoglycan matrix. Tissue Eng. 7, 829–841.
227
Chai, Y., Maxson Jr., R.E., 2006. Recent advances in craniofacial morphogenesis. Dev. Dyn. 235, 2353–2375. Chaponnier, C., Gabbiani, G., 2004. Pathological situations characterized by altered actin isoform expression. J. Pathol. 204, 386–395. Crane, N.J., Popescu, V., Morris, M.D., Steenhuis, P., Ignelzi Jr., M.A., 2006. Raman spectroscopic evidence for octacalcium phosphate and other transient mineral species deposited during intramembranous mineralization. Bone 39, 434–442. Degistirici, O., Jaquiery, C., Schonebeck, B., Siemonsmeier, J., Gotz, W., Martin, I., Thie, M., 2008. Defining properties of neural crest-derived progenitor cells from the apex of human developing tooth. Tissue Eng. Part A 14, 317–330. Deng, Z.L., Sharff, K.A., Tang, N., Song, W.X., Luo, J., Luo, X., Chen, J., Bennett, E., Reid, R., Manning, D., Xue, A., Montag, A.G., Luu, H.H., Haydon, R.C., He, T.C., 2008. Regulation of osteogenic differentiation during skeletal development. Front Biosci. 13, 2001–2021. Dorozhkin, S.V., Epple, M., 2002. Biological and medical significance of calcium phosphates. Angew. Chem. Int. Ed. Engl. 41, 3130–3146. Dupin, E., Creuzet, S., Le Douarin, N.M., 2006. The contribution of the neural crest to the vertebrate body. Adv. Exp. Med. Biol. 589, 96–119. Fuchs, S., Hofmann, A., Kirkpatrick, C.J., 2007. Microvessel-like structures from outgrowth endothelial cells from human peripheral blood in 2-dimensional and 3-dimensional co-cultures with osteoblastic lineage cells. Tissue Eng. 13, 2577–2588. Ghadially, F.N., Meachim, G., Collins, D.H., 1965. Extra-cellular lipid in the matrix of human articular cartilage. Ann. Rheum. Dis. 24, 136–146. Grynpas, M.D., Bonar, L.C., Glimcher, M.J., 1984. Failure to detect an amorphous calcium-phosphate solid phase in bone mineral: a radial distribution function study. Calcif. Tissue Int. 36, 291–301. Han, J., Ishii, M., Bringas Jr., P., Maas, R.L., Maxson Jr., R.E., Chai, Y., 2007. Concerted action of Msx1 and Msx2 in regulating cranial neural crest cell differentiation during frontal bone development. Mech. Dev. 124, 729–745. Helms, J.A., Schneider, R.A., 2003. Cranial skeletal biology. Nature 423, 326–331. Hosoya, A., Nakamura, H., Ninomiya, T., Yoshiba, K., Yoshiba, N., Nakaya, H., Wakitani, S., Yamada, H., Kasahara, E., Ozawa, H., 2006. Immunohistochemical localization of alpha— smooth muscle actin during rat molar tooth development. J. Histochem. Cytochem. 54, 1371–1378. Huitema, L.F., Vaandrager, A.B., 2007. What triggers cell-mediated mineralization? Front Biosci. 12, 2631–2645. Kassem, M., 2006. Stem cells: potential therapy for age-related diseases. Ann. N. Y. Acad. Sci. 1067, 436–442. Kinner, B., Zaleskas, J.M., Spector, M., 2002. Regulation of smooth muscle actin expression and contraction in adult human mesenchymal stem cells. Exp. Cell Res. 278, 72–83. Knott, L., Bailey, A.J., 1998. Collagen cross-links in mineralizing tissues: a review of their chemistry, function, and clinical relevance. Bone 22, 181–187. Le Douarin, N.M., Dupin, E., Ziller, C., 1994. Genetic and epigenetic control in neural crest development. Curr. Opin. Genet. Dev. 4, 685–695. Mao, J.J., Giannobile, W.V., Helms, J.A., Hollister, S.J., Krebsbach, P.H., Longaker, M.T., Shi, S., 2006. Craniofacial tissue engineering by stem cells. J. Dent. Res. 85, 966–979. Mikos, A.G., et al., 2006. Engineering complex tissues. Tissue Eng. 12, 3307–3339. Redi, C.A., Monti, M., Merico, V., Neri, T., Zanoni, M., Zuccotti, M., Garagna, S., 2007. Stem cells. Endocr. Dev. 11, 145–151. Schoenebeck, B., Hartschen, H.J., Schindel, M., Degistirici, O., Siemonsmeier, J., Goetz, W., Thie, M., 2009. Molecular characterization of human impacted third molars: diversification of compartments. Cells Tissues. Organs 189, 356–370. Tucker, A.S., Lumsden, A., 2004. Neural crest cells provide species-specific patterning information in the developing branchial skeleton. Evol. Dev. 6, 32–40. van Beurden, H.E., Von den Hoff, J.W., Torensma, R., Maltha, J.C., Kuijpers-Jagtman, A.M., 2005. Myofibroblasts in palatal wound healing: prospects for the reduction of wound contraction after cleft palate repair. J. Dent. Res. 84, 871–880. Weiner, S., 2006. Transient precursor strategy in mineral formation of bone. Bone 39, 431–433. Weiner, S., 2008. Biomineralization: a structural perspective. J. Struct. Biol. 163, 229–234. Weiner, S., Sagi, I., Addadi, L., 2005. Structural biology. Choosing the crystallization path less traveled. Science 309, 1027–1028. Xu, Y., Malladi, P., Zhou, D., Longaker, M.T., 2007. Molecular and cellular characterization of mouse calvarial osteoblasts derived from neural crest and paraxial mesoderm. Plast. Reconstr. Surg. 120, 1783–1795. Yamada, M., Kurihara, H., Kinoshita, K., Sakai, T., 2005. Temporal expression of alphasmooth muscle actin and drebrin in septal interstitial cells during alveolar maturation. J. Histochem. Cytochem. 53, 735–744. Young, H.E., Steele, T.A., Bray, R.A., Detmer, K., Blake, L.W., Lucas, P.W., Black Jr., A.C., 1999. Human pluripotent and progenitor cells display cell surface cluster differentiation markers CD10, CD13, CD56, and MHC class-I. Proc. Soc. Exp. Biol. Med. 221, 63–71. Zhao, C., Zhao, C., Hosoya, A., Kurita, H., Hu, T., Hiraga, T., Ninomiya, T., Yoshiba, K., Yoshiba, N., Takahashi, M., Kurashina, K., Ozawa, H., Nakamura, H., 2007. Immunohistochemical study of hard tissue formation in the rat pulp cavity after tooth replantation. Arch. Oral Biol. 52, 945–953.