Type-I collagen production by human odontoblast-like cells in explants cultured on cyanoacrylate films

Cell Tissue Res (1986) 244: 133-140 Cell and Tissue Research © Springer-Verlag 1986 Type-I collagen production by human odontoblast-like cells in explants cultured on cyanoacrylate films Electron-immunolocalization of fibronectin at cell/film interface H. Magloire 1, A. Callé 1, D.J. Hartmann 2, A. Joffre 1, B. Serre 3, J.A. Grimaud“, and F. Schué3 1 Laboratoire d’Histophysiologie et de Pathologie des Tissus Dentaires. U.A. CNRS 819; Faculté d’Odontologie, Lyon and C.M.E.A.B.G.; Faculté des Sciences, Villeurbanne, France; 2 Centre de Radioanalyse, U.A. CNRS 819, Institut Pasteur, Lyon, France; 3 Laboratoire de Chimie Macromoléculaire, Université des Sciences et des Techniques, Montpellier, France; 4 Laboratoire de Pathologie Cellulaire du Foie, U.A. CNRS 819, Institut Pasteur, Lyon, France Summary. Odontoblast-like cells derived from human tooth pulps were maintained in explant culture and grown either on glass coverslips only (used as control) or on glass covers- lips coated with cyanoacrylate films. Ultrastructural and cyto-morphometric evidence showed that cells exposed to cyanoacrylate, in contrast to controls, display a significant decrease of rough endoplasmic reticulum and mitochon- dria. In addition, immunofluorescent staining and radioim- munoassays for type-I collagen suggested disturbances in production for the exposed cells. The use of anti-fibronectin antibodies with electron—microscopic immunoperoxidase- labelling demonstrated that the adherence of cells to cyano- acrylate can involve both adhesion plaques and fibronectin. These results therefore suggest that there were no apparent differences in the adhesion interaction of cells between glass and cyanoacrylate substrates. Key words: Collagen — Odontoblast — Cyanoacrylate — Fi- bronectin — Human The treatment of dental carious lesions, i.e., the treatment of the adjacent pulpal inflammation, involves a variety of operative procedures and dental restorative materials. Thus, a careful removal of the injured dentine is followed by endo- dontic therapy if the exposure of the pulpal connective tis- sue is evident. If an exposure is not evident, the tooth should be restored with materials having a low degree of toxicity and which allow the pulp to heal without further damage (Taintor et al. 1981). Thus, alpha-alkyl-cyanoacrylates (methyl, ethyl, butyl, hexyl, octyl, decyl) were widely tested for intra-oral surgery as well as adhesives or pulp capping in restorative dentistry (for review see: Bhaskar et al. 1972; Mc Graw and Caffesse 1978; Hashida et al. 1981). Tissue response to higher homologues of methyl-2-cya- noacrylate (the first adhesive introduced but the most toxic) frequently comprises a foreign body reaction, indicating that alpha cyanoacrylates could be clinically valuable. In vitro application of isobutyl—cyanoacrylate to mouse fibro- blasts (de Renzis and Aleo 1970) or cells grown from sali- vary gland explants (Lamey et al. 1981) have demonstrated the absence of signs of toxicity, ethyl-alpha—cyanoacrylate Send offprint requests to : Prof. H. Magloire, Faculté d’Odontolo- gie, rue G. Paradin, 69372 Lyon Cedcx 2, France even enhancing the cell adhesion to glass. In addition, equal growth occurred when cultured on glass or cyanoacrylate substrates. However, no information is available concerning the ability of cultured odontoblast-like cells grown on n-butyl or n-hexyl alpha cyanoacrylate films to elaborate a collage- nous matrix. In normal conditions, when such cells are grown directly on glass coverslips, they mostly produce type I collagen (Magloire et al. 1981) and exhibit numerous fea- tures similar to those found in young secretory odontoblasts in vivo (Magloire and Dumont 1976; Magloire and Joffre 1979), the only cells responsible for the synthesis of dentinal collagen (mainly consisting of type-I collagen fibres (Scott et al. 1976; Magloire et al. 1983b). This paper describes the ultrastructure of odontoblast-like cells cultured on cya- noacrylate films and correlates production of type-I colla- gen by immunofiuorescence on the cell layer using specific anti-type-I collagen antibodies and competition radioimmu- noassay for type-I collagen released into the culture medi- um. In addition, the presence of fibronectin at the cell/ cyanoacrylate interface was investigated because of the in- volvement of this major glycoprotein in processes of adhe- sion to biological or artificial surfaces (Y amada and Olden 1978; Grinnell 1978; Hedman et al. 1978; Grinnell and Bennett 1981; Kleinman et al. 1981; Ruoslahti et al. 1981). Preliminary reports have been described previously (Mag- loire et al. 1983a). Materials and methods Monomer preparation The monomer was prepared by reacting paraformaldehyde and n-alkyl-cyanoacetate in a mixture of methanol and dig- Iyme using piperidine as catalyst according to the method described by Leonard et al. (1966). As the monomer poly- merized instantaneously as it formed, due to the presence of water and piperidine, leading to low molecular weight chains, a thermal depolymerization treatment at a tempera- ture of 150° C to 190° C was necessary to obtain the monomer. Water was removed from the reaction mixture before the depolymerization step by azeotropic distillation with benzene. The depolymerization was conducted in vacu- um using sulphur dioxide as an inhibitor to prevent anionic polymerization. Moreover, phosphorus pentoxide and hy- 134 droquinone were introduced into the receiver flask to re- move water and to prevent free-radical polymerization re- spectively. The pure monomer was obtained after redistilla- tion under a reduced pressure of nitrogen using propane sultone in the receiver flask as anionic polymerization inhib- itor. It was stored frozen under dry nitrogen before use. Tissue culture Explant cultures, obtained from children’s permanent tooth germs removed for orthodontic reasons, were grown on glass coverslips for control cultures or on glass coverslips coated with n-butyl or n-hexyl alpha cyanoacrylate pre- pared as described above. The preparations were sterilized with ultraviolet light, suspended in Eagle basal medium (BME, Biomérieux, France) supplemented with 10% foetal calf serum (KC Biological, Lenexa, Kansas, USA), ascorbic acid, penicillin and streptomycin as described previously (Magloire and Dumont 1976). The cultures (cyanoacrylate- exposed and control) were incubated for two weeks at 37° C, the medium being renewed every 3-4 days. Growth was considered to have occurred if cellular outgrowth from the explant was seen under the inverted microscope. Electron microscopy Each culture was fixed in a solution (pH 7.4) containing 2% glutaraldehyde and 0.15 M sodium cacodylate for five minutes at room temperature. After three rinses in sodium cacodylate (0.2 M; pH 7.4), the cultures were post-fixed in 2% aqueous osmium tetroxide, dehydrated in a graded se- ries of ethanol concentrations and embedded in Epon. The samples, sectioned perpendicular to their plane of growth were stained with uranyl acetate and lead citrate before examination with a Philips 300 electron microscope. Cytomorphometric procedures For the analysis of surface areas of mitochondria and rough endoplasmic reticulum (rER) from cyanoacrylate-exposed (20 blocks) or control cells grown on glass (20 blocks), ap- proximately 300 ultrathin sections were systematically sur- veyed in an electron microscope. Every area containing cells in close contact with the cyanoacrylate film or in an equiva- lent situation for control specimens was photographed at x 20000 and enlarged photographically to x 40000. From a pool of several hundred electron micrographs, ten cells from each group (cyanoacrylate—exposed or control cul- tures) were randomly selected to cytomorphometric analysis (Weibel 1979). Each selected electron micrograph was cov- ered by a transparency on which the cell and nucleus periph- ery, the perimeter of rough endoplasmic reticulum (rER) and mitochondria (mit.) could be marked. Then, a quanti- met 900 (Cambridge Instruments) video digital image ana- lyser connected to a computer programmed for surface area measurements expressed as umz, was used. For each se- lected cell, the surface area of rough endoplasmic reticulum (SrER) or mitochondria (S mit.) relative to the cytoplasm cell surface was estimated by: _ S rER or S mit. — S cell — S nucleus These data were expressed as meani-standard deviation of the mean (SD). Differences were assessed by Student’s t—test; p