Interactions between epithelial and fibroblast-like cells in cultures derived from monkey periodontal ligament

Isolation and characterization of epithelial and myogenic cells by "fishing" for the morphologically distinct cell types in rat primary periodontal ligament cultures

The periodontal ligament (PDL) contains various cell populations and plays a central role in the maintenance, repair, and regeneration of the periodontium, i.e., tooth-supporting structures. Because primary cells isolated from PDL tissue are heterogeneous, the establishment of an effective isolation method for cells of interest is desired. In the present study, two morphologically distinct cell types were identified in confluent primary cultures derived from rat PDL. To isolate these cell populations, a small piece of filter paper soaked with trypsin-EDTA was placed directly onto the target cell population, enabling the cells to detach from the culture dish. The filter papers were then transferred into fresh culture dishes to establish outgrowth cultures; these two steps constitute the "cell fishing" method. The "fished" cell types were propagated and subcultured for further analyses. In morphological evaluation, immunocytochemical analyses, and reverse transcription-polymerase chain reaction, the isolated cells exhibited a polygonal appearance or a mono- or multinucleated appearance, with a high cytoplasm-to-nucleus ratio, leading to their being characterized as epithelial or myogenic cell populations, respectively. Surprisingly, a notable proportion of the multinuclear cells in the primary and subsequent isolated cultures demonstrated dramatic, spontaneous contractions, a feature typical of skeletal muscle cells. Finally, the isolated cell populations maintained a normal karyotype with a diploid chromosomal number. These results demonstrated that physiological epithelial and skeletal muscle cells can be obtained from primary PDL cultures without artificial induction using growth factors or chemicals, and can be propagated as individual lineage-committed cell populations; the populations consisted of differentiated and progenitor cells that maintained chromosomal stability. This simple, classical culture procedure provides new insights into the biological properties of PDL cells, which are potentially important for the differentiation of tissue or somatic stem cells and for the development of future cell-based therapies for dental and muscular diseases.

Expression of type IV collagen and laminin at the interface between epithelial cells and fibroblasts from human periodontal ligament

The present study was undertaken to examine whether synthesis of type IV collagen and laminin around the epithelial rests of Malassez (ERM) requires direct contact between cells from ERM and periodontal ligament fibroblasts. Human periodontal ligament (HPDL) explants produced outgrowths containing both ERM cells and fibroblasts when cultured in a modified serum-free medium. The interface between ERM cells and fibroblasts was examined using phase-contrast microscopy (PCM) and scanning electron microscopy (SEM). Expression of type IV collagen and laminin was studied by immunohistochemistry and in situ hybridization. It was observed that ERM cells grew underneath fibroblasts or attached to them. At the interface, type IV collagen and laminin and their respective mRNAs were abundant in both ERM cells and fibroblasts, while these proteins and mRNAs showed little if any staining in cells further away from the interface. Hence, these findings indicate that synthesis of type IV collagen and laminin is induced by direct interaction between ERM cells and periodontal ligament fibroblasts.

Tony Melcher's contributions to the regeneration of the periodontium

The impact of Tony Melcher's scientific work on periodontal regeneration is discussed in light of his interests in cellular domains and bone healing, the physiology of the periodontal ligament, and his development of organ and cell culture methods. Collectively, his scientific accomplishments have pointed us in new directions for approaching periodontal regeneration and have helped to create new research opportunities, many of which were previously unrecognized. Tony's scientific ideas are insightful, numerous, and compelling; his ability to make these ideas come alive and relevant to clinical problems has been a major influence in shaping our current thinking on the biology of the periodontium and on periodontal regeneration.

Production of matrix-degrading enzymes and inhibition of osteoclast-like cell differentiation by fibroblast-like cells from the periodontal ligament of human primary teeth

Clinically, the most apparent difference between the primary and permanent dentitions is the physiologic loss of the primary tooth by root resorption. Root resorption is associated with loss of integrity of the periodontal ligament (PDL), followed by recruitment of resorptive cells that remove root structure. We therefore cultured primary dentition PDL fibroblasts (PPDL cells) to investigate in vitro their production of matrix metalloproteinases (MMPs) and tissue inhibitors of MMP (TIMPs), and the effects of soluble factors produced by these cells on osteoclast-like cell differentiation. These studies demonstrate that PPDL cells in vitro have a heterogeneous morphology, and they constitutively synthesize 92-kDa gelatinase, 72-kDa gelatinase, and 53/57-kDa procollagenase as well as TIMP-1, -2, and a third inhibitor of matrix metalloproteinase, as determined by substrate gel zymography and immunoblot analysis. Compared with PDL cells from the permanent dentition, PPDL cells generally produced a greater amount of collagenase but similar amounts of the gelatinases and inhibitors. PPDL cells were treated with pro-inflammatory cytokines to determine their effect on the expression of matrix-degrading enzymes and inhibitors. Interleukin-1alpha and tumor necrosis factor-alpha enhanced the constitutive expression of proteinases but not that of inhibitors in PPDL cells. Conditioned media from PPDL cell lines inhibited the differentiation of osteoclast-like cells in mouse bone marrow cultures. These findings indicate that PPDL cells may modulate the cascade of root resorption both by their regulated production of proteinases and inhibitors and by synthesis of unknown soluble factor(s) that may regulate osteoclast development.

Origins and functions of cells essential for periodontal repair: the role of fibroblasts in tissue homeostasis

OBJECTIVE

A review is undertaken of rodent model systems and cell culture studies that address the role of periodontal fibroblasts in tissue homeostasis in both normal function and after wound healing.

RATIONALE

Fibroblasts are the predominant cells of the periodontal ligament (PL) and of healthy gingiva and have important roles in the development, function and regeneration of the tooth support apparatus.

REVIEW

In normally functioning periodontal tissues cell turnover involves generation of new cells by proliferation which in turn is balanced by apoptopic cell death. Consequently PL fibroblasts comprise a renewal cell system in steady-state. PL cell progenitors can generate multiple types of more differentiated, specialized cells including large numbers of fibroblastic cells and more limited numbers of osteogenic or cementogenic cells. However PL fibroblasts constitutively block osteogenesis and thereby maintain the PL width. Proliferating progenitor cell populations of the PL are enriched in locations adjacent to blood vessels and in contiguous endosteal spaces from where they migrate to the body of the PL. Large increases of cell formation and cell differentiation occur after wounding but surprisingly, the cells that repopulate the PL adjacent to the root surface are largely post-mitotic. As PL cell populations comprise multiple lineages, it is likely that after wounding, the separate phenotypes repopulating the wound site will be selected by environmental factors. Further, the specific repopulating lineages will strongly influence the form and function of the nascent tissue. To illustrate the specificity of fibroblast functions, examples of migratory and contractile fibroblast phenotypes are provided which exhibit constitutively different levels of gelsolin and alpha-smooth muscle actin respectively, cytoskeletal proteins which are markers for these cell types.

CONCLUSION

Fibroblasts contribute to PL homeostasis by their abilities to remodel tissues, to repopulate wounds, to influence the metabolism of other cell types and to create a new fibrous attachment.

Morphological changes in cultured human periodontal ligament cells exposed to dental materials

The cytotoxicity of a range of dental restorative cements was assessed by continuous observation of cultures with inverted microscopy and by light microscopic study of fixed preparations, using an in vitro model with cultured test cells derived from human periodontal ligament. The sequential morphological changes observed over a seven day period showed different degrees of cell loss and patterns of injury in response to different restorative materials, reflecting primarily either nuclear or cytoplasmic damage. Attempts at recovery were frequently identified as the culture period was extended and were characterized by recolonization of denuded areas of the culture well. It was concluded that differing dental cements damage cells through a variety of mechanisms and that the test cells exhibit differing degrees of susceptibility to injury. Assays based on short-term cultures may overestimate cytotoxicity by not allowing for cell recovery from reversible injury or repopulation of monolayers by proliferation of resistant cells.

Maintenance of cell-type-specific cytoskeletal character in epithelial cells out of epithelial context: cytokeratins and other cytoskeletal proteins in the rests of Malassez of the periodontal ligament

We have determined the patterns of synthesis of cytokeratins and other epithelial marker proteins in the "rests of Malassez" of the periodontium of rabbits and humans, by immunofluorescence microscopy of cryosections prepared from fixed and decalcified rabbit teeth with attached ligament or from manually isolated human periodontal ligaments. Proteins of the major cell structures characterizing epithelial differentiation are present in Malassez cells: a complex set of cytokeratins as well as desmosomal, hemidesmosomal and basal lamina proteins. In addition, we have shown these cytoskeletal and extracellular matrix structures by electron microscopy. The cytokeratin complement of Malassez cells was found to be highly complex, as 8 of the total of 20 known epithelial cytokeratins were detected (nos. 5, 7, 8, 14, 15, 17, 18, 19). This pattern, together with the presence of the desmosomal cadherins Dsg2 and Dsc2 and the cytoplasmic desmosome plaque-associated protein plakophilin 1, indicates that the cells of the rests of Malassez are derived from the basal cell layer of a stratified squamous epithelium rather than from simple epithelial or neuroendocrine epithelial cells. Our observations show that Malassez cells retain the major characteristics of epithelial cells throughout their differentiation from the root sheath epithelium into the rests of Malassez, even though the surface location and the polar tissue architecture that typify epithelial are lost during this process. From this study we further conclude that the specific cytoskeletal complement of the Malassez cells represents an intrinsic gene expression program that neither depends on nor causes the formation of a stratified epithelium. We also compare the specific cytoskeletal features of Malassez cells with those of other persisting epithelial residues and discuss the potential value of these findings in relation to the histogenesis and diagnostic classification of dental and periodontal cysts and tumors.

Ultrastructural changes in cultured human periodontal ligament cells exposed to dental materials

The cytopathic effects of a range of dental restorative cements were examined by electron microscopy using an in vitro model with cultured test cells derived from human periodontal ligament. Monolayers were fixed and processed in situ after 2, 24, and 48 h exposure. Many cells showed lysis after two hours exposure consistent with immediate acute injury. Cultures subsequently recovered, reflecting different susceptibilities of cells to injury, and at later stages showed distinct patterns of cell damage in response to different restorative materials. These were related primarily to either cytoplasmic or nuclear damage and to changes resembling apoptosis.

Cultural and morphological characteristics of human periodontal ligament cells in vitro

Primary cell lines from explants of human periodontal ligament were established in vitro to assess their suitability as test cells in a biocompatibility assay. They were maintained over nine passages. The constituent fibroblast-like cells (F-cells) were monitored throughout by light and electron microscopy. Although F-cells from several different donors displayed stable and consistent growth characteristics, after the ninth subculture changes in the fine structure suggested that some cells were undergoing differentiation or senescence. It is therefore recommended that cells from earlier passages be used in in vitro biocompatibility assays.

Epithelial cell-fibroblast interactions: modulation of extracellular matrix proteins in cultured oral cells

A model system involving co-cultures of human gingival or periodontal ligament fibroblasts with mouse epithelial root sheath cells or human gingival epithelial cells was used to study epithelial cell-fibroblast interactions. Double-labeled immunofluorescence and microfluorometry were used to investigate the expression of extracellular matrix molecules of collagen type I (collagen I), type III (collagen III) and fibronectin in fibroblasts. When fibroblasts from either source were cultured alone, the fluorescence for collagen I and fibronectin ranged from strongly positive to almost negative. Collagen III staining was relatively weak compared with that of collagen I. After 2-3 days of co-culture, gingival fibroblasts and ligament fibroblasts adjacent to the mouse sheath cells exhibited enhanced intracellular fluorescence for collagen I and fibronectin. Very little change was observed for collagen III. Gingival fibroblasts cultured with gingival epithelial cells showed increased fluorescence for collagen I but decreased fluorescence for fibronectin. In contrast, the fluorescence intensity for both collagen I and fibronectin in ligament fibroblasts were reduced after 3 days of co-culture with gingival epithelial cells. Ultrastructural changes in fibroblasts co-cultured with mouse root sheath cells included increased Golgi cisternae and vesicles and an increased abundance of rough endoplasmic reticulum, polyribosomes, secretory vesicles and pinocytotic vesicles. Thus, the expression of extracellular matrix proteins and the metabolic activity of fibroblasts can be modulated by oral epithelial cells.

Tooth movement

This article reviews the evolution of concepts regarding the biological foundation of force-induced tooth movement. Nineteenth century hypotheses proposed two mechanisms: application of pressure and tension to the periodontal ligament (PDL), and bending of the alveolar bone. Histologic investigations in the early and middle years of the 20th century revealed that both phenomena actually occur concomitantly, and that cells, as well as extracellular components of the PDL and alveolar bone, participate in the response to applied mechanical forces, which ultimately results in remodeling activities. Experiments with isolated cells in culture demonstrated that shape distortion might lead to cellular activation, either by opening plasma membrane ion channels, or by crystallizing cytoskeletal filaments. Mechanical distortion of collagenous matrices, mineralized or non-mineralized, may, on the other hand, evoke the development of bioelectric phenomena (stress-generated potentials and streaming potentials) that are capable of stimulating cells by altering the electric charge on their membrane or their fluid envelope. In intact animals, mechanical perturbations on the order of about 1 min/d are apparently sufficient to cause profound osteogenic responses, perhaps due to matrix proteoglycan-related "strain memory". Enzymatically isolated human PDL cells respond biochemically to mechanical and chemical signals. The latter include endocrines, autocrines, and paracrines. Histochemical and immunohistochemical studies showed that during the early places of tooth movement, PDL fluids are shifted, and cells and matrix are distorted. Vasoactive neurotransmitters are released from periodontal nerve terminals, causing leukocytes to migrate out of adjacent capillaries. Cytokines and growth factors are secreted by these cells, stimulating PDL cells and alveolar bone lining cells to remodel their related matrices. This remodeling activity facilitates movement of teeth into areas in which bone had been resorbed. This emerging information suggests that in the living mammal, many cell types are involved in the biological response to applied mechanical stress to teeth, and thereby to bone. Essentially, cells of the nervous, immune, and endocrine systems become involved in the activation and response of PDL and alveolar bone cells to applied stresses. This fact implies that research in the area of the biological response to force application to teeth should be sufficiently broad to include explorations of possible associations between physical, cellular, and molecular phenomena. The goals of this investigative field should continue to expound on fundamental principles, particularly on extrapolating new findings to the clinical environment, where millions of patients are subjected annually to applications of mechanical forces to their teeth for long periods of time in an effort to improve their position in the oral cavity.(ABSTRACT TRUNCATED AT 400 WORDS)

Estrogen receptors and growth response in cultured human periodontal ligament cells

The periodontal ligament (PDL) is a connective tissue involved in the remodeling process associated with tooth development and positioning. PDL cells grown in culture were analyzed for the capacity to specifically bind steroid hormones and for growth response to estradiol-17 beta. Using [3H]estradiol-17 beta as the ligand, PDL cells in first passage cultures exhibited a specific estrogen binding capacity of 881 fmol/mg cell protein. With [3H]dexamethasone as a ligand, the binding capacity of the glucocorticoid receptor was 143 fmol/mg protein. With [3H]R5020 as a ligand, the progestin receptor exhibited a binding capacity of 5 pmol/mg protein. Scatchard analysis of estradiol binding at 37 degrees revealed a dissociation constant of 2.7 X 10(-9) M, representative of the estrogen receptor. The addition of estradiol-17 beta at concentrations of 10(-9) and 10(-8) M to culture media induced a dose-dependent decrease in growth (DNA content) to 62% and 38% control values, respectively. The addition of the antiestrogens tamoxifen and 4-hydroxytamoxifen at concentrations of 10(-7) and 10(-6) M similarly depressed cell growth. These results show that PDL cells contain high affinity receptors for several steroid hormones and further that these cells are targets for the action of estrogens.

Immunocytochemical and biochemical characterization of the connective tissue component of fibrous papular lesions of oral mucosa

In an attempt to define the fibroblastic components of fibrous papular lesions of oral mucosa, 21 cases representing these lesions were selected for study. Paraffin or frozen sections of the lesions were stained, using the immunoperoxidase technique, with antibodies to: alpha-l-antitrypsin, alpha-l-antichymotrypsin, lysozyme, fibronectin, actin, myosin, and for Clq binding. Three of these cases were biochemically analyzed for collagen Types I, III, and V. This study demonstrated the presence of cytochemical markers in fibrous papules that were similar to those observed for the compartment of circumvascular fibroblasts in control normal mucosa. Analysis of the collagens present indicated that in addition to histologic similarities, the gene products of these extracellular matrices were similar to those reported for angiofibromas and Shagreen patches in tuberous sclerosis. The cells in these lesions appear to be distinct from myofibroblasts and the high affinity complement binding fibroblasts characterized by Bordin et al.(1).

Isolation and growth of human periodontal ligament cells in vitro

The periodontal ligaments from single human teeth were dissected free of the root surface and plated into culture dishes. In each case two populations of cells were obtained: fibroblasts and epithelial cells. These two populations were separated and grown in culture. The epithelial nature of the epithelioid-appearing cells was confirmed by staining for keratin, exclusively an epithelial cell protein.

Cholera toxin and dibutyryl cyclic-AMP stimulate the growth of epithelial cells derived from epithelial rests from porcine periodontal ligament

The epithelial cells (E-cells) grew best at high (greater than 5 per cent) concentrations of fetal bovine serum (FBS). Growth could be obtained at low concentrations of dialysed FBS (DFBS) if the medium (alpha-MEM) was modified so that the levels of Ca2+ and K+ were reduced to 0.1 and 1.0 mM, respectively (beta-MEM). The addition of 0.5 per cent DFBS to the beta-MEM did not initiate good growth but was sufficiently supportive to enable the effects of various growth promoters to be tested. Cholera toxin and dibutyryl cyclic-3'5'-adenosine monophosphate (Bt2cAMP), which are known to increase intracellular cAMP levels, at concentrations of 1 ng/ml and 0.5 mM, respectively increased cell number. Cholera toxin caused the E-cells to be more flattened when viewed by phase-contrast; this appeared to be due to spread of the cells. No difference in cell-size distributions obtained between the trypsinized E-cells grown in the presence or absence of cholera toxin was observed. Epithelial proliferation that occurs in dental cysts could result from a rise in intracellular cAMP levels in epithelial cell rests.

Evidence that healthy human gingiva contains functionally heterogeneous fibroblast subpopulations

Six mass cultures of human fibroblasts derived from a single biopsy of a normal gingival papilla tip were studied with regard to their protein, collagen and glycosaminoglycan synthesis in vitro, using incorporation of radiolabelled substances. The proliferation rates, replicative life-spans and cell-size distributions of these mass cultures were determined. There were significant differences among the 6 cultures and these differences persisted throughout numerous cell replications in vitro. It is suggested that functional heterogeneity exists among phenotypically stable fibroblast subpopulations or subpopulation mixtures from normal tissue. The concept of participation of fibroblast subpopulations in disease pathogenesis is supported by these preliminary findings.

Dome formation by oral epithelia in vitro

Multicellular, cystic structures, termed domes, have been described previously in epithelia cultured from various tissues that have a known transport or secretory function in vivo and in vitro. We report for the first time dome formation in cells cultured from "covering" and "rest" epithelia of oral tissues: porcine gingival and alveolar mucosa epithelium and epithelial rests of Malassez. As demonstrated by light- and electron microscopy, the morphology of the domes varied with the location of their lumen and the number of cells or cell layers involved in their structure. Sequential observations using phase contrast microscopy and time lapse cinematography of living cultures showed that the domes were dynamic structures with expansion-collapse cycles of between 30 min and 17 h duration. Dome formation in oral epithelia was stimulated by dibutyryl cyclic AMP (dbcAMP, 10(-3) to 10(-6) M) and abolished by ouabain (10(-10) M), an inhibitor of sodium transport. The morphological features and the dynamic nature of domes found in oral epithelia, and their dbcAMP and ouabain responsiveness are similar to those demonstrated previously in several other epithelia that have a known transport function in vivo and in vitro. Such fluid transport is not thought to be a property of oral epithelia in vivo. Our data, however, suggest a similar function of these epithelia cultured in vitro, and perhaps in pathological cyst formation in vivo.

Sulfated galactosaminoglycans of bovine periodontal ligament. Evidence for the presence of two major types of hybrids but no chondroitin sulfate

Sulfated galactosaminoglycans of mature bovine periodontal ligament were separated into four fractions by ethanol precipitation. Fractions I and II were dermatan sulfates with high contents of L-iduronate, but only small amounts of this hexuronic acid were present in fractions III and IV. Effects of digestion with testicular hyaluronidase or a periodate-alkali treatment showed that most if not all of the glycans in fractions I, II and III were hybrid chains containing both L-iduronate and D-glucuronate. The composition of fraction IV was less certain, but the chains strongly resembled fraction III hybrids in electrophoretic characteristics, not chondroitin sulfate. The total amount of the D-glucuronate-rich fractions III and IV in the ligament was similar to that of I plus II. In contrast, almost all of the sulfated galactosaminoglycans of mature skin were rich in L-iduronate. The more varied composition of the ligament glycosaminoglycans may be related to the mixed population of cells in this tissue.

Inability of mesoderm cells to locomote on the modified free surface of epithelial cell sheets in vitro

Previous work has shown that when chick embryo mesoderm tissue is seeded onto the free, dorsal surface of established sheets of embryonic epithelial endoblast, the former penetrates the latter and spreads on the underlying artificial substratum. In this work, the surface charge on the epithelial sheet has been altered, prior to seeding the mesoderm, to ascertain whether such a change could alter the behavior of the mesoderm with respect to the free surface of the epithelium. Charge alteration was accomplished using the polycations, poly-L-lysine. Surface charge characteristics were examined ultrastructurally using cationized and anionized ferritin. Results showed that although surface charge changes were detectable, there was no difference in the behavior of the mesoderm with respect to the endoblast. Neuraminidase did not detectably affect the epithelial surface charge. These results are consistent with the view that changes in substratum surface charge are not necessarily correlated with changes in adhesiveness.

Composition of human periodontal ligament cells in tissue culture

The cells in the PDL of monkeys have been demonstrated to have an unusual growth behavior in vitro since fibroblast-like cells have been reported to grow on top of epithelial-like cells. In this study growth of epithelial-like cells along with fibroblast-like cells was seen in human PDL from three out of 15 permanent teeth when explantation technique was used whereas only fibroblast-like cells were seen after enzymatical degradation. Occasionally fibroblast-like cells were seen growing on top of epithelial-like cells.

Isolation and identification of epithelial-like cells in culture by a collagenase-separation technique

An operational criterion for the identification and isolation of epithelial-like (E) cells, based on their ability to cover and protect a collagen gel from the action of collagenase, has been developed. The E cells isolated by this collagenase-separation technique (CST) exhibited the ultrastructural features, including desmosomes and abundant tonofilaments, that are considered characteristic of this cell type. Unlike confluent cultures of fibroblast-like (F) cells, E cells were not found to have large external transformation-sensitive (LETS) protein on their surface membranes. The CST provides a nondestructive and efficient means of identifying and isolating E cells from mixed populations.