Alkaline Phosphatase Induces the Deposition of Calcified Layers in Relation to Dentin: An in vitro Study to Mimic the Formation of Afibrillar Acellular Cementum W. BEERTSEN' and T. VAN DEN BOS" 2 Experimental Oral Biology Group, 'Department of Periodontology, Academic Centre for Dentistry Amsterdam (ACTA), Louwesweg 1, 1066 EA Amsterdam, The Netherlands; and 2Department of Cell Biology and Histology, Faculty of Medicine, University of Amsterdam
An attempt was made to test the hypothesis that alkaline phosphatase, an enzyme which is abundant in periodontal ligament, plays a role in the formation of acellular root cementum. Thin slices of bovine dentin were incubated in Iscove Modified Dulbecco's Medium supplemented with 10% normal rabbit serum and 10 mmol/L P-glycerophosphate (P-GP) or folded into pericardial explants. Intestinal bovine alkaline phosphatase (APase), covalently linked to agarose beads, was added to the cultures. In the presence of the enzyme, the dentin slices were covered with thin layers of mineralized material. Such layers were not observed in cultures not provided with APase-beads or 3-GP. They also did not form in relation to demineralized dentin. The layers of calcified material appeared to consist of crystallites embedded in a granular matrix of moderate electron density, which often exhibited the presence of incremental lines and resembled the matrix of afibrillar acellular cementum formed under in vivo conditions. When pericardial explants were interposed between the enzyme-containing beads and the dentin, mineral deposition in relation to the dentin was retarded. This finding lends support to the view that soft connective tissues interfere with the free diffusion of phosphate. J Dent Res 70(3):176-181, March, 1991
Introduction. It has been reported (Beertsen and Everts, 1990) that in healing rat periodontium, layers resembling acellular cementum are not
only deposited along the hard dental tissues, but may also be laid down along the inner wall of the alveolar bone. Since deposition of these layers seemed to be associated with cells having the phenotype of periodontal ligament fibroblasts, it was argued that it is these cells that are primarily responsible for acellular cementum formation. Periodontal ligament cells are said to be rich in alkaline phosphatase activity (e.g., Mafrkova and Kindlova, 1972; Halackova et al., 1980; Yamashita et al., 1987; Oshima et al., 1988; Piche et al., 1989; Nojima et al., 1990). The enzyme is thought to release phosphate from organic phosphate esters (Robison, 1923; Beertsen and Van den Bos, 1989) and, in doing so, may play a role in the deposition of calcified layers along exposed calcified surfaces (Beertsen and Everts, 1990). As a test of this hypothesis, the present study was undertaken. Thin slices of bovine dentin were incubated in culture medium or folded into pericardial explants, which themselves are very poor in alkaline phosphatase activity. Alkaline phosphatase and the monophosphate ester P3-glycerophosphate were added to the cultures, and the dentin slices were examined for the deposition of cementum-like layers. Received for publication October 26, 1990 Accepted for publication December 14, 1990
Materials and methods. Materials. -Iscove Modified Dulbecco's Medium (IMDM), normal rabbit serum (NRS), Fungizone® (amphotericin), streptomycin, and penicillin were purchased from Gibco (Gibco Labs., Grand Island, NY). L( + )-ascorbic acid and 3-glycerophosphate disodium salt (3-GP) were from Merck (Darmstadt, Germany), and multi-well culture dishes from Costar (Costar, Cambridge, MA). p-nitrophenylphosphate (pNPP), L-levamisole, and calf intestinal alkaline phosphatase (APase), covalently linked to agarose beads, were from Sigma Chemical Co. (St. Louis, MO). [45Ca]C12 (50.2 mCi/mg) was from New England Nuclear (NEN Chemicals, Germany). Preparation of bovine dentin slices. -Bovine permanent incisors were collected at the local slaughterhouse immediately after the animals were killed (age, one to three years) and frozen at - 80'C until used. After being defrosted, the gingiva and periodontal ligament were removed, and each tooth was cut with a diamond disk parallel to its longitudinal axis from the apex to the cervical area-under constant irrigation with tap water-and split with a chisel. The roots were then cleaned and freed from pulp and cementum. They were washed with ice-cold phosphate-buffered saline in the presence of proteinase inhibitors (Van den Bos and Beertsen, 1987), and the outer dentin (containing the mantle dentin layer) was removed with a diamond disk under cooling with tap water. Demineralized dentin slices (DDS) were made as reported previously (Beertsen and Van den Bos, 1989). So that nondemineralized dentin slices (DS) could be prepared, sections of 100-pm thickness were taken from the circumpulpal dentin by means of a Microslice 2 saw (Metal Research United, Cambridge, UK) equipped with a diamond blade. The slices were washed overnight in 70% ethanol, then air-dried. Culture method. -Pericardia were dissected from NZW rabbits (one to two weeks old) and cut into pieces of about four square millimeters. DS or DDS were inserted into the pericardial fragments and incubated in 0.3 mL of medium for periods of one day to three weeks, at 370C, in a humidified atmosphere containing 5% CO2 and 95% air. The culture medium consisted of IMDM supplemented with 10% heat-inactivated NRS, 100 pug/L ascorbic acid, 10 mmol/L 3-GP, and antibiotics (amphotericin, 2.5 jLg/mL; streptomycin, 100 ag/mL; penicillin, 100 U/mL). It contained 1.6 + 0.1 mmol/L total calcium. APase beads were folded into the tissue fragments, together with the DS or DDS, or were added to the culture medium (30 mU enzyme per
well).
This
activity corresponded
Downloaded from jdr.sagepub.com at NORTH DAKOTA STATE UNIV LIB on June 19, 2015 For personal use only. No other uses without permission.
176
with that in
periosteal explants that were shown to induce the remineralization of DDS with which they were co-cultured (Beertsen and Van den Bos, 1989). Some media were supplemented with 10 mmol/L 1-GP, and both types of media were changed every two or three days. In parallel experiments, pNPP (10 mmol/L) was used instead of f-GP so that the concentration of phosphate could be determined spectrophotometrically as a function of time. [From a previous study, it appeared that the rate of hydrolysis of
FORMATION OF CALCIFIED LAYERS
Vol. 70 No. 3
1
177
2
pNPP by periosteal APase is of the same order as that of r-GP (Beertsen and Van den Bos, 1989).] Pi released by the enzyme increased gradually until a maximum concentration of 3-4 mmol/L was reached (within two days). In several experiments, solitary DS or DDS were incubated without being folded into pericardium. Light and electron microscopy. -Co-cultures were fixed in a solution of 4% paraformaldehyde and 1% glutaraldehyde in 0.1 mol/L Na-cacodylate buffer (pH 7.4) for 24 h, and some were then decalcified for four days in 0.1 mol/L EDTA and 2.5% glutaraldehyde in a solution of the same buffer. The specimens were post-fixed in 1% OS04 in cacodylate buffer, pH 7.4, for one h at 4°C, dehydrated through a graded series of ethanols, and embedded in LX-112 epoxy resin (Ladd). Sections of 2-im thickness were stained with methylene blue or according to the Von Kossa method. Ultrathin sections were cut with a diamond knife, stained with uranyl acetate and lead citrate, and examined in a Zeiss EM lOC electron microscope.
For comparison of mineralized layers formed in vitro with afibrillar acellular cementum in vivo, ultrathin sections from mouse incisors used in previous
experiments were re-exam-
ined. For details, see Beertsen et al. (1983). Biochemical determinations. -APase activity in periodontal ligament was determined as follows. Ligament was collected from the middle one-third of each root from human third molars, bovine teeth, and rabbit and rat incisors. Also, pericardial and periosteal (calvaria) tissues from rabbits were tested. Samples of 2-10 mg wet weight were minced and extracted in 0.2 mL of glycine buffer, pH 10.5 (0.1 mol/L glycine, 1 mmol/L MgCl2, 0.1 mmol/L ZnCl2), containing 0.1% Triton X-100, for one h at 4°C. Then, the samples were supplemented with
Fig. 1-Mineral crystallites deposited in demineralized dentin co-cultured with pericardium for three days. APase-beads were also folded within the tissue explant. Fig. 2-Mineral (electron-dense area, E) deposited in a de(eineralized dentin slice co-cultured with pericardium (PC) for seven days. APase-beads were added to the medium outside the co-culture.
2 mL glycine buffer (37°C). After ten min, pNPP was added as a substrate (final concentration, 6 mmol/L), and the optical density was monitored at 405 nm in a Varian 220 spectrophotometer. Control samples were treated with 1 mmol/L levamisole, an inhibitor of the enzyme. Enzyme activity (U) was expressed as ,umol pNP released per min (at 37°C and pH 10.5). The DNA content of the tissues was determined according to the method described by Janakidevi et al. (1988). Hydroxyproline was determined colorimetrically by the method of Stegeman and Stalder (1967) as modified by Guis et al. (1973). [45Ca]-uptake experiments. -For the quantitative study of mineral deposition, pericardium-DDS co-cultures were incubated in the presence of radiolabeled calcium (1 FLCi [45Ca]C12 TABLE 1 ALKALINE PHOSPHATASE ACTIVITY IN VARIOUS TISSUES AS MEASURED AT 37'C, pH 10.5 APase/hypro APase/DNA Periodontal ligament human 0.82 ± 0.10 (4) 1.18 ± 0.16 (4) bovine 0.51 + 0.26 (4) 0.37 + 0.06 (4) rat 1.63 ± 0.53 (8) 0.33 ± 0.16 (8) rabbit 5.78 ± 0.59 (4) 0.75 ± 0.13 (4) Periosteum (rabbit) 0.71 ± 0.23 (4) 0.21 ± 0.09 (4) Pericardium (rabbit) 0.005 + 0.002 (4) 0.03 ± 0.01 (4) Enzyme activity is expressed as mU per pLg hydroxyproline or per tg DNA. Data are presented as mean + standard deviation. Between parentheses is the number of observations. Controls were assayed in the presence of 1 mmol/L levamisole. They contained less than 0.01 mU APase per jxg hypro or pug DNA (not included in the Table).
Downloaded from jdr.sagepub.com at NORTH DAKOTA STATE UNIV LIB on June 19, 2015 For personal use only. No other uses without permission.
178
J Dent Res March 1991
BEERTSEN & VAN DEN BOS
X-ray diffraction. -X-ray diffraction of mineral newly deposited into demineralized dentin slices was carried out by means of a Philips diffractometer with Co Ka radiation at 40 kV and 40 mA. The counting time was 20 s/0.02' 2 theta. For this purpose, calcified DDS (in total, 10 mg) was cut into small fragments and transferred to a vaseline-coated Si crystal.
Results.
Fig. 3-Dentin slice folded into pericardium (PC) and cultured in the presence of APase-beads (added to the medium outside the co-culture). The specimen was demineralized with EDTA prior to being embedded. Note that the DS is lined by a narrow layer of newly deposited material
(between arrowheads).
Fig. 4-Dentin slice incubated in medium supplemented with APasebeads. Note the presence of a layer along the dentin staining intensely with methylene blue (arrows).
Fig. 5-Layer of calcified material deposited along dentin slice. Note that the crystallites are oriented in a more or less random fashion. D = dentin.
per well) for three days in the presence of 10% NRS, APasebeads, and 3-GP. The DDS were then separated from the soft connective tissue, and both DDS and pericardium decalcified in 0.5 mL of 1 mol/L HCl for one h at 37TC. Samples of 300 pL were added to Optifluor® scintillation cocktail (Packard Instruments, Inc.) and counted in a Packard Tricarb 4530 scintillation counter.
Alkaline phosphatase in periodontal ligament. -As shown in Table 1, the periodontal ligament was rich in APase. Although there was some variability among species, the activity of the enzyme in periodontal ligament was considerably higher than in rabbit periosteum. Rabbit pericardium was almost free of APase. In all cases, the enzyme could be inhibited with levamisole (1 mmol/L). Demineralized bovine dentin slices. -When demineralized dentin slices (DDS) and APase-beads, folded together into pericardial fragments, were incubated in medium supplemented with 10 mmol/L 1-GP, mineral crystallites were deposited in the DDS, as revealed by electron microscopy (Fig. 1) and [45Ca]-uptake measurements (Table 2). Although a certain amount of radiolabeled calcium was also deposited in the pericardium (probably associated with cellular debris), the radioactivity per pLg hydroxyproline in the DDS was far higher than that in the pericardium. In the absence of APase-beads (or 3-GP, data not shown), no mineral was deposited, either in the DDS or in the pericardium. Levamisole did not inhibit APase activity bound to the agarose beads (the enzyme was of intestinal origin). X-ray diffraction revealed that the mineral deposited into the DDS was of an apatitic nature. Next to a prominent sharp peak corresponding to a d-spacing of 0.282 nm (211 plane), peaks at d-spacings of 0.414 nm (200 plane) and 0.343 nm (002 plane) were obtained in the diffractograms. The peaks corresponding to planes (112) and (300) were poorly discernible, probably due to the presence of collagen. Peak broadening (e.g., plane 002) indicated that the apatite was of a non-ideal crystalline form. When APase-beads were added to the medium outside the DDS-pericardium co-cultures, instead of being folded into the tissue, very little [45Ca] was incorporated into the DDS at the shorter time intervals (Table 2). Prolonged incubation (one to two weeks), however, did result in the appearance of mineral deposits in the dentin matrix (Fig. 2). When solitary DDS (not folded into pericardium) were incubated in the presence of APase-beads and 1-GP, mineral was deposited within the dentin matrix, as revealed by electron microscopy and [45Ca]-uptake measurements. After a threeday incubation period, the amount of radiolabeled calcium per DDS was as high as that in the DDS folded into the pericardial explants together with the enzyme-containing beads (Table 2). After prolonged incubation, thin layers of mineralized material were seen along the periphery of the remineralized dentin slices. Mineralized dentin slices. -When untreated dentin slices (DS) were folded into pericardial explants and incubated in the presence of APase-beads and 1-GP, thin layers of calcified material were deposited along their periphery (Fig. 3). These layers followed the outline of the dentin and the course of the dentinal tubules and had reached a thickness of about 1-2 pLm within one week. Sometimes a complete closure of the dentinal tubules had been established. In cultures not provided with APasebeads or f-GP, calcified layers were never seen along the dentin slices. When APase-beads were added to the medium instead of being folded into the pericardium, the formation of calcified
Downloaded from jdr.sagepub.com at NORTH DAKOTA STATE UNIV LIB on June 19, 2015 For personal use only. No other uses without permission.
FORMATION OF CALCIFIED LAYERS
Vol. 70 No. 3
layers was retarded, but not prevented. They proved to be Von Kossa-positive and stained intensely with methylene blue (Fig.
4).
At the ultrastructural level, the layers deposited along the dentin had a relatively smooth outline and contained mineral crystallizes oriented in a more or less random fashion (Fig. 5). Following demineralization, a matrix of moderate electron density was seen (Fig. 6a) which, at higher magnification, had a granular and sometimes filamentous appearance (Fig. 6b) and resembled afibrillar acellular cementum deposited in vivo (Fig. 7; Listgarten, 1975; Beertsen et al., 1983). As with cementum, the calcified material deposited in vitro often showed the presence of incremental lines (Fig. 8). In the pericardium bordering the DS, fibroblasts were usually oriented parallel to the calcified surface, thus forming a capsule-like structure. Collagen fibrils were sometimes embedded in the newly formed calcified layers (not shown). However, functionally-oriented collagen fibers, as in periodontal ligament, were not observed. Solitary dentin slices not folded into pericardial explants showed the presence of layers of mineralized material similar to those observed along DS co-cultured with the soft connective tissue. Here too, mineral crystallites were deposited into a matrix of granular material.
Discussion. The mechanism underlying the formation of acellular root cementum is not understood. Recently, it was hypothesized that cementum formation relies in part on synthesis and expression of alkaline phosphatase (APase) by periodontal ligament cells (Beertsen and Everts, 1990). We believe that the present in vivo and in vitro findings lend support to our hypothesis. We have observed that the periodontal ligament is rich in APase activity, richer than periosteum when expressed on a per DNA or hydroxyproline basis. This is in line with observations reported by Yamashita et al. (1987), who found that APase activity in periodontal ligament is relatively high as compared with that in tissues such as kidney, liver, small intestine, alveolar bone, and dental pulp. Using intestinal APase, we were able to demonstrate that layers of mineralized material were deposited along calcified dentin slices when incubated in media supplemented with the organic monophosphate ester f-GP. From a morphological point of view, these layers closely resembled afibrillar acellular root cementum formed in vivo. Both contained a matrix of moderate electron density that had a granular or somewhat filamentous appearance and often exhibited the presence of incremental lines. Of course, this does not mean that the layers formed in vitro were identical to afibrillar root cementum. After all, we did not use biochemical or immunological probes to verify
their similarity in composition. The calcified layers observed in the present investigation also resembled the afibrillar acellular cementum-like layers found by Melcher et al. (1986, 1987) in cultures of bone-derived cells. According to these workers, cells derived from bone can express in vitro the phenotype for the cementum. Their observations, however, can perhaps also be explained by assuming that bone-derived cells are rich in endogenous APase activity. It could be this activity (together with the f-glycerophosphate they added to the culture medium) which brought about the formation of cementum-like layers in a fashion similar to that in the present study. Additional support for our view that APase plays a significant role in cementogenesis comes from a naturally-occurring condition, i.e., hypophosphatasia, an inherited disorder of os-
179
teogenesis and cementogenesis, characterized by a defect in the synthesis and expression of the liver/bone/kidney (L/B/K) isoenzyme of APase (reviewed by Whyte, 1989). Almost no cementum formation is observed in children suffering from this disorder, as a result of which teeth are lost prematurely (Beumer et al., 1973). Since our results and those of others (Yamashita et al., 1987; Oshima et al., 1988) have shown that APase in periodontal ligament is likely to represent the LIB/K isoenzyme (it can be inhibited by levamisole and L-homoarginine), the enzyme in this tissue is probably also affected by the genetic defect. However, no direct evidence has appeared in the literature that periodontal APase is affected in hypophosphatasia. If APase expression by periodontal ligament cells is important, what regulates the thickness of the cementum layer, and why does not the entire periodontal ligament become calcified? Although these questions cannot adequately be dealt with at present, it is conceivable that the collagenous fiber framework of the periodontal ligament contains inhibitors that prevent its mineralization. Alternatively, the calcified root may act as a homogeneous nucleator (Nancollas et al., 1989), as a result of which calcium phosphate would preferentially precipitate onto the root instead of the collagenous fiber framework of the periodontal ligament. As shown previously, acellular cementum thickness increases with age at a daily apposition rate of about 3-20 nm (Beertsen et al., 1983; Dastmalchi et al., 1990). With respect to dentin remineralization, the present study has shown that mineral crystallites are preferentially deposited in relation to the dentinal collagen and not in the collagenous fiber framework of the pericardium. Perhaps this was due to the presence of inhibitor molecules in the pericardium. Another explanation might be that the dentinal collagen, although thoroughly extracted, still contained traces of collagen-bound phosphoproteins, substances known to promote calcification (Linde and Lussi, 1989). From where did the organic material in the calcified layers along the dentin slices originate? Was it produced by pericardial cells? Probably not. It must be emphasized that calcified layers formed not only in relation to dentin wrapped into pericardium. Layers of similar morphology also formed in relation to dentin slices directly immersed in serum-containing culture medium supplemented with APase and f-GP. This suggests that the matrix of moderate electron density was not made by cells, but represented medium- or serum-derived components co-precipitated with the mineral. Whether during the formation of afibrillar acellular cementum in vivo the matrix is also laid down by co-precipitation with the mineral is not known. It is interesting, however, that compounds like hydroxyethylidene-1,1-bisphosphonate (that TABLE 2
[45Ca]-UPTAKE IN DDS AND PERICARDIUM DDS A. Co-cultures APase inside PC APase outside PC Controls B. Solitary DDS
dpm/,ug hypro Pericardium
± 12 (7) 4 ± 5 (8) 0.3 ± 0.2 (8) 48 ± 21 (8)
41
(A) [45Ca]-uptake in DDS and pericardium (PC) following a three-day co-culture period in the presence of APase-beads (inside the co-cultures or added to the medium outside) and 10 mmol/L ,B-GP. Control cultures did not receive APase-beads. Data are expressed as dpm (thousands) per pug hydroxyproline ± standard deviation. Between parentheses is the number of observations. (B) [45Ca]-uptake in solitary DDS, not co-cultured with pericardium. Same culture conditions as in A.
Downloaded from jdr.sagepub.com at NORTH DAKOTA STATE UNIV LIB on June 19, 2015 For personal use only. No other uses without permission.
JS10 DENTAL2729 03-04-91 10-14-14
7 ± 5 (7) 1 ± 1 (8) 0.3 ± 0.2 (8)
180
J Dent Res March 1991
BEERTSEN & VAN DEN BOS
Fig. 6-(a) (left) APase-induced layer (L) at the dentin-pericardium interface after one week of culture. Note that the layer follows the outline of the dentin and is of more or less uniform thickness. The specimen was demineralized with EDTA prior to being embedded. At higher magnification (b) (right), the layer appeared to consist of granular material of moderate electron density, resembling that of afibrillar acellular root cementum (compare with Fig. 7b). D = dentin; F = fibroblast.
interfere effectively with mineralization processes) not only inhibit the deposition of mineral in acellular cementum of rodent teeth but also prevent the deposition of its matrix (Beertsen et al., 1985; Wesselink and Beertsen, 1989). This would suggest that the intrinsic matrix components and the mineral are laid down as an integrated unit. With respect to cellular cementum, which is rich in intrinsic collagen (Schroeder, 1986), matrix formation and deposition are not inhibited by the bisphosphonate (Beertsen et al., 1985; Wesselink and Beertsen, 1989), suggesting that with this type of cementum, the deposition of matrix is not strictly coupled to that of the mineral. Finally, the present study has shown that mineralization was significantly delayed when pericardial tissue was interposed between the phosphate source (APase-beads) and the dentin slices. This lends further support to the view (Hunter et al., 1986; Beertsen and Van den Bos, 1989) that tissue macromolecules may influence mineralization by restricting ionic diffusion.
Acknowledgments. We are indebted to Anneke Niehof for excellent technical assistance, to Dr. S. van der Gaast (NIOZ, Texel, The Neth-
errands) for his help with the x-ray diffraction, and to Vincent Everts for helpful comments on the manuscript. REFERENCES BEERTSEN, W. and EVERTS, V. (1990): Formation of Acellular Root Cementum in Relation to Dental and Non-dental Hard Tissues in the Rat, J Dent Res 69:1669-1673. BEERTSEN, W.; EVERTS, V.; and HOEBEN, K. (1983): Loss of Connective Tissue Attachment in the Marginal Periodontium of the Mouse Following Blockage of Eruption: Electron Microscopic Observations, J Periodont Res 18:276-291. BEERTSEN, W.; NIEHOF, A.; and EVERTS, V. (1985): Effects of 1-Hydroxyethylidene-1,1-Bisphosphonate (HEBP) on the Formation of Dentin and the Periodontal Attachment Apparatus in the Mouse, Am JAnat 174:83-103. BEERTSEN, W. and VAN DEN BOS, T. (1989): Calcification of Dentinal Collagen by Cultured Rabbit Periosteum: the Role of Alkaline Phosphatase, Matrix 9:159-171. BEUMER, J.; TROWBRIDGE, H.O.; SILVERMAN, S.; and EISENBERG, E. (1973): Childhood Hypophosphatasia and the Premature Loss of Teeth, Oral Swg Oral Med Oral Pathol 35:631-640. DASTMALCHI, R.; POLSON, A.; BOUWSMA, O.; and PROSKIN, H. (1990): Ceementum Thickness and Mesial Drift, J Clin Periodontol 17:709-713.
Fig. 7-(a) (left) Afibrillar acellular root cementum (AAC) formed in relation to a mouse incisor that had been prevented from erupting for a period of six months. The specimen was demineralized with EDTA prior to being embedded. For details, see Beertsen et al. (1983). Note the presence of incremental lines. (b) (right) At higher magnification, the cementum appeared to consist of granular material of moderate electron density. D = dentin.
Downloaded from jdr.sagepub.com at NORTH DAKOTA STATE UNIV LIB on June 19, 2015 For personal use only. No other uses without permission.
Vol. 70 No. 3
FORMATION OF CALCIFIED LAYERS
GUIS, M.B.; SLOOTWEG, R.N.; and TONINO, G.J.M. (1973): A Biochemical Study of Collagen in the Periodontal Ligament from Erupting and Non-erupting Bovine Incisors, Arch Oral Biol 18:253263. HALACKOVA, Z.; OUDRAN, L.; and KUKLETOVA, M. (1980): Localization of Some Enzymes in the Periodontium of the Rat Molar, Acta Histochem 67:173-179. HUNTER, G.K.; NYBURG, S.C.; and PRITZKER, K.P.H. (1986): Hydroxyapatite Formation in Collagen, Gelatin, and Agarose Gels, Collagen Rel Res 6:229-238. JANAKIDEVI, K.; MURRAY, C.D.; SELL, C.; and HELD, P. (1988): Fluorometric Analysis of DNA in Cell Cultures, Anal Biochem 172:78-81. LINDE, A. and LUSSI, A. (1989): Mineral Induction by Polyanionic Dentin and Bone Proteins at Physiological Ionic Conditions, Connect Tissue Res 21:197-203. LISTGARTEN, M.A. (1975): Afibrillar Dental Cementum in the Rat and Hamster, J Periodont Res 10:158-167. MARIKOVA, Z. and KINDLOVA, M. (1972): Alkaline Phosphatase in the Periodontium of Rat Molars, J Dent Res 51:1502. MELCHER, A.H.; CHEONG, T.; COX, J.; NEMETH, E.; and SHIGA, A. (1986): Synthesis of Cementum-like Tissue in vitro by Cells Cultured from Bone: a Light and Electron Microscope Study, J Periodont Res 21:592-612. MELCHER, A.H.; McCULLOCH, C.A.G.; CHEONG, T.; NEMETH, E.; and SHIGA, A. (1987): Cells from Bone Synthesize Cementum-like and Bone-like Tissue in vitro and May Migrate into Periodontal Ligament in vivo, J Periodont Res 22:246-247. NANCOLLAS, G.H.; LoRE, M.; PEREZ, L.; RICHARDSON, C.; and ZAWACKI, S.J. (1989): Mineral Phases of Calcium Phosphate, Anat Rec 224:234-241. NOJIMA, N.; KOBAYASHI, M.; SHIONOME, M.; TAKAHASHI, N.; SUGA, T.; and HASEGAWA, K. (1990): Fibroblastic Cells Derived from Bovine Periodontal Ligaments Have the Phenotypes of Osteoblasts, J Periodont Res 25:179-185. OSHIMA, M.; KUWATA, F.; OTSUKA, K.; and SUZUKI, K. (1988): Acid and Alkaline Phosphatase Activities of Cultured Human Periodontal Ligament Cells. In: Recent Advances in Clinical Periodontology, Proceedings of the Third Meeting of the Intemational Academy of Periodontology, J. Ishikawa, H. Kawasaki, K. Ikeda, and K. Hasegawa, Eds., Amsterdam: Excerpta Medica, pp. 573-577. PICHE, J.E.; CARNES, D.E.; and GRAVES, D.T. (1989): Initial Characterization of Cells Derived from Human Periodontia, JDent Res 68:761-767. ROBISON, R. (1923): The Possible Significance of Hexosephosphoric Esters in Ossification, Biochem J 17:286-293. SCHROEDER, H. (1986): The Periodontium, Berlin: Springer Verlag, pp. 23-127. STEGEMAN, H. and STALDER, K.H. (1967): Determination of Hydroxyproline, Clin Chim Acta 18:267-273. VAN DEN BOS, T. and BEERTSEN, W. (1987): Effects of 1- Hy-
181
Fig. 8-Incremental lines (arrows) in APase-induced calcified layers deposited along dentin (D). The DS was co-cultured with pericardium for a period of 14 days. F = fibroblast.
droxyethylidene-1,1-Bisphosphonate (HEBP) on the Synthesis of Dentin Matrix Proteins in the Mouse, Collagen Rel Res 7:135-147. WESSELINK, P.R. and BEERTSEN, W. (1989): The Influence of 1- Hydroxyethylidene-1,1-Bisphosphonate (HEBP) on Dental Root Resorption in the Mouse, Calcif Tissue Int 45:104-110. WHYTE, M.P. (1989): Hypophosphatasia. In: The Metabolic Basis of Inherited Disease, C.R. Scriver, A.L. Beaudet, W.S. Sly, and D. Valle, Eds., New York: McGraw-Hill, pp. 2843-2856. YAMASHITA, Y.; SATO, M.; and NOGUCHI, T. (1987): Alkaline Phosphatase in the Periodontal Ligament of the Rabbit and Macaque Monkey, Arch Oral Biol 32:677-678.
Downloaded from jdr.sagepub.com at NORTH DAKOTA STATE UNIV LIB on June 19, 2015 For personal use only. No other uses without permission.
An attempt was made to test the hypothesis that alkaline phosphatase, an enzyme which is abundant in periodontal ligament, plays a role in the formation of acellular root cementum. Thin slices of bovine dentin were incubated in Iscove Modified Dulbecco's Medium supplemented with 10% normal rabbit serum and 10 mmol/L P-glycerophosphate (P-GP) or folded into pericardial explants. Intestinal bovine alkaline phosphatase (APase), covalently linked to agarose beads, was added to the cultures. In the presence of the enzyme, the dentin slices were covered with thin layers of mineralized material. Such layers were not observed in cultures not provided with APase-beads or 3-GP. They also did not form in relation to demineralized dentin. The layers of calcified material appeared to consist of crystallites embedded in a granular matrix of moderate electron density, which often exhibited the presence of incremental lines and resembled the matrix of afibrillar acellular cementum formed under in vivo conditions. When pericardial explants were interposed between the enzyme-containing beads and the dentin, mineral deposition in relation to the dentin was retarded. This finding lends support to the view that soft connective tissues interfere with the free diffusion of phosphate. J Dent Res 70(3):176-181, March, 1991
Introduction. It has been reported (Beertsen and Everts, 1990) that in healing rat periodontium, layers resembling acellular cementum are not
only deposited along the hard dental tissues, but may also be laid down along the inner wall of the alveolar bone. Since deposition of these layers seemed to be associated with cells having the phenotype of periodontal ligament fibroblasts, it was argued that it is these cells that are primarily responsible for acellular cementum formation. Periodontal ligament cells are said to be rich in alkaline phosphatase activity (e.g., Mafrkova and Kindlova, 1972; Halackova et al., 1980; Yamashita et al., 1987; Oshima et al., 1988; Piche et al., 1989; Nojima et al., 1990). The enzyme is thought to release phosphate from organic phosphate esters (Robison, 1923; Beertsen and Van den Bos, 1989) and, in doing so, may play a role in the deposition of calcified layers along exposed calcified surfaces (Beertsen and Everts, 1990). As a test of this hypothesis, the present study was undertaken. Thin slices of bovine dentin were incubated in culture medium or folded into pericardial explants, which themselves are very poor in alkaline phosphatase activity. Alkaline phosphatase and the monophosphate ester P3-glycerophosphate were added to the cultures, and the dentin slices were examined for the deposition of cementum-like layers. Received for publication October 26, 1990 Accepted for publication December 14, 1990
Materials and methods. Materials. -Iscove Modified Dulbecco's Medium (IMDM), normal rabbit serum (NRS), Fungizone® (amphotericin), streptomycin, and penicillin were purchased from Gibco (Gibco Labs., Grand Island, NY). L( + )-ascorbic acid and 3-glycerophosphate disodium salt (3-GP) were from Merck (Darmstadt, Germany), and multi-well culture dishes from Costar (Costar, Cambridge, MA). p-nitrophenylphosphate (pNPP), L-levamisole, and calf intestinal alkaline phosphatase (APase), covalently linked to agarose beads, were from Sigma Chemical Co. (St. Louis, MO). [45Ca]C12 (50.2 mCi/mg) was from New England Nuclear (NEN Chemicals, Germany). Preparation of bovine dentin slices. -Bovine permanent incisors were collected at the local slaughterhouse immediately after the animals were killed (age, one to three years) and frozen at - 80'C until used. After being defrosted, the gingiva and periodontal ligament were removed, and each tooth was cut with a diamond disk parallel to its longitudinal axis from the apex to the cervical area-under constant irrigation with tap water-and split with a chisel. The roots were then cleaned and freed from pulp and cementum. They were washed with ice-cold phosphate-buffered saline in the presence of proteinase inhibitors (Van den Bos and Beertsen, 1987), and the outer dentin (containing the mantle dentin layer) was removed with a diamond disk under cooling with tap water. Demineralized dentin slices (DDS) were made as reported previously (Beertsen and Van den Bos, 1989). So that nondemineralized dentin slices (DS) could be prepared, sections of 100-pm thickness were taken from the circumpulpal dentin by means of a Microslice 2 saw (Metal Research United, Cambridge, UK) equipped with a diamond blade. The slices were washed overnight in 70% ethanol, then air-dried. Culture method. -Pericardia were dissected from NZW rabbits (one to two weeks old) and cut into pieces of about four square millimeters. DS or DDS were inserted into the pericardial fragments and incubated in 0.3 mL of medium for periods of one day to three weeks, at 370C, in a humidified atmosphere containing 5% CO2 and 95% air. The culture medium consisted of IMDM supplemented with 10% heat-inactivated NRS, 100 pug/L ascorbic acid, 10 mmol/L 3-GP, and antibiotics (amphotericin, 2.5 jLg/mL; streptomycin, 100 ag/mL; penicillin, 100 U/mL). It contained 1.6 + 0.1 mmol/L total calcium. APase beads were folded into the tissue fragments, together with the DS or DDS, or were added to the culture medium (30 mU enzyme per
well).
This
activity corresponded
Downloaded from jdr.sagepub.com at NORTH DAKOTA STATE UNIV LIB on June 19, 2015 For personal use only. No other uses without permission.
176
with that in
periosteal explants that were shown to induce the remineralization of DDS with which they were co-cultured (Beertsen and Van den Bos, 1989). Some media were supplemented with 10 mmol/L 1-GP, and both types of media were changed every two or three days. In parallel experiments, pNPP (10 mmol/L) was used instead of f-GP so that the concentration of phosphate could be determined spectrophotometrically as a function of time. [From a previous study, it appeared that the rate of hydrolysis of
FORMATION OF CALCIFIED LAYERS
Vol. 70 No. 3
1
177
2
pNPP by periosteal APase is of the same order as that of r-GP (Beertsen and Van den Bos, 1989).] Pi released by the enzyme increased gradually until a maximum concentration of 3-4 mmol/L was reached (within two days). In several experiments, solitary DS or DDS were incubated without being folded into pericardium. Light and electron microscopy. -Co-cultures were fixed in a solution of 4% paraformaldehyde and 1% glutaraldehyde in 0.1 mol/L Na-cacodylate buffer (pH 7.4) for 24 h, and some were then decalcified for four days in 0.1 mol/L EDTA and 2.5% glutaraldehyde in a solution of the same buffer. The specimens were post-fixed in 1% OS04 in cacodylate buffer, pH 7.4, for one h at 4°C, dehydrated through a graded series of ethanols, and embedded in LX-112 epoxy resin (Ladd). Sections of 2-im thickness were stained with methylene blue or according to the Von Kossa method. Ultrathin sections were cut with a diamond knife, stained with uranyl acetate and lead citrate, and examined in a Zeiss EM lOC electron microscope.
For comparison of mineralized layers formed in vitro with afibrillar acellular cementum in vivo, ultrathin sections from mouse incisors used in previous
experiments were re-exam-
ined. For details, see Beertsen et al. (1983). Biochemical determinations. -APase activity in periodontal ligament was determined as follows. Ligament was collected from the middle one-third of each root from human third molars, bovine teeth, and rabbit and rat incisors. Also, pericardial and periosteal (calvaria) tissues from rabbits were tested. Samples of 2-10 mg wet weight were minced and extracted in 0.2 mL of glycine buffer, pH 10.5 (0.1 mol/L glycine, 1 mmol/L MgCl2, 0.1 mmol/L ZnCl2), containing 0.1% Triton X-100, for one h at 4°C. Then, the samples were supplemented with
Fig. 1-Mineral crystallites deposited in demineralized dentin co-cultured with pericardium for three days. APase-beads were also folded within the tissue explant. Fig. 2-Mineral (electron-dense area, E) deposited in a de(eineralized dentin slice co-cultured with pericardium (PC) for seven days. APase-beads were added to the medium outside the co-culture.
2 mL glycine buffer (37°C). After ten min, pNPP was added as a substrate (final concentration, 6 mmol/L), and the optical density was monitored at 405 nm in a Varian 220 spectrophotometer. Control samples were treated with 1 mmol/L levamisole, an inhibitor of the enzyme. Enzyme activity (U) was expressed as ,umol pNP released per min (at 37°C and pH 10.5). The DNA content of the tissues was determined according to the method described by Janakidevi et al. (1988). Hydroxyproline was determined colorimetrically by the method of Stegeman and Stalder (1967) as modified by Guis et al. (1973). [45Ca]-uptake experiments. -For the quantitative study of mineral deposition, pericardium-DDS co-cultures were incubated in the presence of radiolabeled calcium (1 FLCi [45Ca]C12 TABLE 1 ALKALINE PHOSPHATASE ACTIVITY IN VARIOUS TISSUES AS MEASURED AT 37'C, pH 10.5 APase/hypro APase/DNA Periodontal ligament human 0.82 ± 0.10 (4) 1.18 ± 0.16 (4) bovine 0.51 + 0.26 (4) 0.37 + 0.06 (4) rat 1.63 ± 0.53 (8) 0.33 ± 0.16 (8) rabbit 5.78 ± 0.59 (4) 0.75 ± 0.13 (4) Periosteum (rabbit) 0.71 ± 0.23 (4) 0.21 ± 0.09 (4) Pericardium (rabbit) 0.005 + 0.002 (4) 0.03 ± 0.01 (4) Enzyme activity is expressed as mU per pLg hydroxyproline or per tg DNA. Data are presented as mean + standard deviation. Between parentheses is the number of observations. Controls were assayed in the presence of 1 mmol/L levamisole. They contained less than 0.01 mU APase per jxg hypro or pug DNA (not included in the Table).
Downloaded from jdr.sagepub.com at NORTH DAKOTA STATE UNIV LIB on June 19, 2015 For personal use only. No other uses without permission.
178
J Dent Res March 1991
BEERTSEN & VAN DEN BOS
X-ray diffraction. -X-ray diffraction of mineral newly deposited into demineralized dentin slices was carried out by means of a Philips diffractometer with Co Ka radiation at 40 kV and 40 mA. The counting time was 20 s/0.02' 2 theta. For this purpose, calcified DDS (in total, 10 mg) was cut into small fragments and transferred to a vaseline-coated Si crystal.
Results.
Fig. 3-Dentin slice folded into pericardium (PC) and cultured in the presence of APase-beads (added to the medium outside the co-culture). The specimen was demineralized with EDTA prior to being embedded. Note that the DS is lined by a narrow layer of newly deposited material
(between arrowheads).
Fig. 4-Dentin slice incubated in medium supplemented with APasebeads. Note the presence of a layer along the dentin staining intensely with methylene blue (arrows).
Fig. 5-Layer of calcified material deposited along dentin slice. Note that the crystallites are oriented in a more or less random fashion. D = dentin.
per well) for three days in the presence of 10% NRS, APasebeads, and 3-GP. The DDS were then separated from the soft connective tissue, and both DDS and pericardium decalcified in 0.5 mL of 1 mol/L HCl for one h at 37TC. Samples of 300 pL were added to Optifluor® scintillation cocktail (Packard Instruments, Inc.) and counted in a Packard Tricarb 4530 scintillation counter.
Alkaline phosphatase in periodontal ligament. -As shown in Table 1, the periodontal ligament was rich in APase. Although there was some variability among species, the activity of the enzyme in periodontal ligament was considerably higher than in rabbit periosteum. Rabbit pericardium was almost free of APase. In all cases, the enzyme could be inhibited with levamisole (1 mmol/L). Demineralized bovine dentin slices. -When demineralized dentin slices (DDS) and APase-beads, folded together into pericardial fragments, were incubated in medium supplemented with 10 mmol/L 1-GP, mineral crystallites were deposited in the DDS, as revealed by electron microscopy (Fig. 1) and [45Ca]-uptake measurements (Table 2). Although a certain amount of radiolabeled calcium was also deposited in the pericardium (probably associated with cellular debris), the radioactivity per pLg hydroxyproline in the DDS was far higher than that in the pericardium. In the absence of APase-beads (or 3-GP, data not shown), no mineral was deposited, either in the DDS or in the pericardium. Levamisole did not inhibit APase activity bound to the agarose beads (the enzyme was of intestinal origin). X-ray diffraction revealed that the mineral deposited into the DDS was of an apatitic nature. Next to a prominent sharp peak corresponding to a d-spacing of 0.282 nm (211 plane), peaks at d-spacings of 0.414 nm (200 plane) and 0.343 nm (002 plane) were obtained in the diffractograms. The peaks corresponding to planes (112) and (300) were poorly discernible, probably due to the presence of collagen. Peak broadening (e.g., plane 002) indicated that the apatite was of a non-ideal crystalline form. When APase-beads were added to the medium outside the DDS-pericardium co-cultures, instead of being folded into the tissue, very little [45Ca] was incorporated into the DDS at the shorter time intervals (Table 2). Prolonged incubation (one to two weeks), however, did result in the appearance of mineral deposits in the dentin matrix (Fig. 2). When solitary DDS (not folded into pericardium) were incubated in the presence of APase-beads and 1-GP, mineral was deposited within the dentin matrix, as revealed by electron microscopy and [45Ca]-uptake measurements. After a threeday incubation period, the amount of radiolabeled calcium per DDS was as high as that in the DDS folded into the pericardial explants together with the enzyme-containing beads (Table 2). After prolonged incubation, thin layers of mineralized material were seen along the periphery of the remineralized dentin slices. Mineralized dentin slices. -When untreated dentin slices (DS) were folded into pericardial explants and incubated in the presence of APase-beads and 1-GP, thin layers of calcified material were deposited along their periphery (Fig. 3). These layers followed the outline of the dentin and the course of the dentinal tubules and had reached a thickness of about 1-2 pLm within one week. Sometimes a complete closure of the dentinal tubules had been established. In cultures not provided with APasebeads or f-GP, calcified layers were never seen along the dentin slices. When APase-beads were added to the medium instead of being folded into the pericardium, the formation of calcified
Downloaded from jdr.sagepub.com at NORTH DAKOTA STATE UNIV LIB on June 19, 2015 For personal use only. No other uses without permission.
FORMATION OF CALCIFIED LAYERS
Vol. 70 No. 3
layers was retarded, but not prevented. They proved to be Von Kossa-positive and stained intensely with methylene blue (Fig.
4).
At the ultrastructural level, the layers deposited along the dentin had a relatively smooth outline and contained mineral crystallizes oriented in a more or less random fashion (Fig. 5). Following demineralization, a matrix of moderate electron density was seen (Fig. 6a) which, at higher magnification, had a granular and sometimes filamentous appearance (Fig. 6b) and resembled afibrillar acellular cementum deposited in vivo (Fig. 7; Listgarten, 1975; Beertsen et al., 1983). As with cementum, the calcified material deposited in vitro often showed the presence of incremental lines (Fig. 8). In the pericardium bordering the DS, fibroblasts were usually oriented parallel to the calcified surface, thus forming a capsule-like structure. Collagen fibrils were sometimes embedded in the newly formed calcified layers (not shown). However, functionally-oriented collagen fibers, as in periodontal ligament, were not observed. Solitary dentin slices not folded into pericardial explants showed the presence of layers of mineralized material similar to those observed along DS co-cultured with the soft connective tissue. Here too, mineral crystallites were deposited into a matrix of granular material.
Discussion. The mechanism underlying the formation of acellular root cementum is not understood. Recently, it was hypothesized that cementum formation relies in part on synthesis and expression of alkaline phosphatase (APase) by periodontal ligament cells (Beertsen and Everts, 1990). We believe that the present in vivo and in vitro findings lend support to our hypothesis. We have observed that the periodontal ligament is rich in APase activity, richer than periosteum when expressed on a per DNA or hydroxyproline basis. This is in line with observations reported by Yamashita et al. (1987), who found that APase activity in periodontal ligament is relatively high as compared with that in tissues such as kidney, liver, small intestine, alveolar bone, and dental pulp. Using intestinal APase, we were able to demonstrate that layers of mineralized material were deposited along calcified dentin slices when incubated in media supplemented with the organic monophosphate ester f-GP. From a morphological point of view, these layers closely resembled afibrillar acellular root cementum formed in vivo. Both contained a matrix of moderate electron density that had a granular or somewhat filamentous appearance and often exhibited the presence of incremental lines. Of course, this does not mean that the layers formed in vitro were identical to afibrillar root cementum. After all, we did not use biochemical or immunological probes to verify
their similarity in composition. The calcified layers observed in the present investigation also resembled the afibrillar acellular cementum-like layers found by Melcher et al. (1986, 1987) in cultures of bone-derived cells. According to these workers, cells derived from bone can express in vitro the phenotype for the cementum. Their observations, however, can perhaps also be explained by assuming that bone-derived cells are rich in endogenous APase activity. It could be this activity (together with the f-glycerophosphate they added to the culture medium) which brought about the formation of cementum-like layers in a fashion similar to that in the present study. Additional support for our view that APase plays a significant role in cementogenesis comes from a naturally-occurring condition, i.e., hypophosphatasia, an inherited disorder of os-
179
teogenesis and cementogenesis, characterized by a defect in the synthesis and expression of the liver/bone/kidney (L/B/K) isoenzyme of APase (reviewed by Whyte, 1989). Almost no cementum formation is observed in children suffering from this disorder, as a result of which teeth are lost prematurely (Beumer et al., 1973). Since our results and those of others (Yamashita et al., 1987; Oshima et al., 1988) have shown that APase in periodontal ligament is likely to represent the LIB/K isoenzyme (it can be inhibited by levamisole and L-homoarginine), the enzyme in this tissue is probably also affected by the genetic defect. However, no direct evidence has appeared in the literature that periodontal APase is affected in hypophosphatasia. If APase expression by periodontal ligament cells is important, what regulates the thickness of the cementum layer, and why does not the entire periodontal ligament become calcified? Although these questions cannot adequately be dealt with at present, it is conceivable that the collagenous fiber framework of the periodontal ligament contains inhibitors that prevent its mineralization. Alternatively, the calcified root may act as a homogeneous nucleator (Nancollas et al., 1989), as a result of which calcium phosphate would preferentially precipitate onto the root instead of the collagenous fiber framework of the periodontal ligament. As shown previously, acellular cementum thickness increases with age at a daily apposition rate of about 3-20 nm (Beertsen et al., 1983; Dastmalchi et al., 1990). With respect to dentin remineralization, the present study has shown that mineral crystallites are preferentially deposited in relation to the dentinal collagen and not in the collagenous fiber framework of the pericardium. Perhaps this was due to the presence of inhibitor molecules in the pericardium. Another explanation might be that the dentinal collagen, although thoroughly extracted, still contained traces of collagen-bound phosphoproteins, substances known to promote calcification (Linde and Lussi, 1989). From where did the organic material in the calcified layers along the dentin slices originate? Was it produced by pericardial cells? Probably not. It must be emphasized that calcified layers formed not only in relation to dentin wrapped into pericardium. Layers of similar morphology also formed in relation to dentin slices directly immersed in serum-containing culture medium supplemented with APase and f-GP. This suggests that the matrix of moderate electron density was not made by cells, but represented medium- or serum-derived components co-precipitated with the mineral. Whether during the formation of afibrillar acellular cementum in vivo the matrix is also laid down by co-precipitation with the mineral is not known. It is interesting, however, that compounds like hydroxyethylidene-1,1-bisphosphonate (that TABLE 2
[45Ca]-UPTAKE IN DDS AND PERICARDIUM DDS A. Co-cultures APase inside PC APase outside PC Controls B. Solitary DDS
dpm/,ug hypro Pericardium
± 12 (7) 4 ± 5 (8) 0.3 ± 0.2 (8) 48 ± 21 (8)
41
(A) [45Ca]-uptake in DDS and pericardium (PC) following a three-day co-culture period in the presence of APase-beads (inside the co-cultures or added to the medium outside) and 10 mmol/L ,B-GP. Control cultures did not receive APase-beads. Data are expressed as dpm (thousands) per pug hydroxyproline ± standard deviation. Between parentheses is the number of observations. (B) [45Ca]-uptake in solitary DDS, not co-cultured with pericardium. Same culture conditions as in A.
Downloaded from jdr.sagepub.com at NORTH DAKOTA STATE UNIV LIB on June 19, 2015 For personal use only. No other uses without permission.
JS10 DENTAL2729 03-04-91 10-14-14
7 ± 5 (7) 1 ± 1 (8) 0.3 ± 0.2 (8)
180
J Dent Res March 1991
BEERTSEN & VAN DEN BOS
Fig. 6-(a) (left) APase-induced layer (L) at the dentin-pericardium interface after one week of culture. Note that the layer follows the outline of the dentin and is of more or less uniform thickness. The specimen was demineralized with EDTA prior to being embedded. At higher magnification (b) (right), the layer appeared to consist of granular material of moderate electron density, resembling that of afibrillar acellular root cementum (compare with Fig. 7b). D = dentin; F = fibroblast.
interfere effectively with mineralization processes) not only inhibit the deposition of mineral in acellular cementum of rodent teeth but also prevent the deposition of its matrix (Beertsen et al., 1985; Wesselink and Beertsen, 1989). This would suggest that the intrinsic matrix components and the mineral are laid down as an integrated unit. With respect to cellular cementum, which is rich in intrinsic collagen (Schroeder, 1986), matrix formation and deposition are not inhibited by the bisphosphonate (Beertsen et al., 1985; Wesselink and Beertsen, 1989), suggesting that with this type of cementum, the deposition of matrix is not strictly coupled to that of the mineral. Finally, the present study has shown that mineralization was significantly delayed when pericardial tissue was interposed between the phosphate source (APase-beads) and the dentin slices. This lends further support to the view (Hunter et al., 1986; Beertsen and Van den Bos, 1989) that tissue macromolecules may influence mineralization by restricting ionic diffusion.
Acknowledgments. We are indebted to Anneke Niehof for excellent technical assistance, to Dr. S. van der Gaast (NIOZ, Texel, The Neth-
errands) for his help with the x-ray diffraction, and to Vincent Everts for helpful comments on the manuscript. REFERENCES BEERTSEN, W. and EVERTS, V. (1990): Formation of Acellular Root Cementum in Relation to Dental and Non-dental Hard Tissues in the Rat, J Dent Res 69:1669-1673. BEERTSEN, W.; EVERTS, V.; and HOEBEN, K. (1983): Loss of Connective Tissue Attachment in the Marginal Periodontium of the Mouse Following Blockage of Eruption: Electron Microscopic Observations, J Periodont Res 18:276-291. BEERTSEN, W.; NIEHOF, A.; and EVERTS, V. (1985): Effects of 1-Hydroxyethylidene-1,1-Bisphosphonate (HEBP) on the Formation of Dentin and the Periodontal Attachment Apparatus in the Mouse, Am JAnat 174:83-103. BEERTSEN, W. and VAN DEN BOS, T. (1989): Calcification of Dentinal Collagen by Cultured Rabbit Periosteum: the Role of Alkaline Phosphatase, Matrix 9:159-171. BEUMER, J.; TROWBRIDGE, H.O.; SILVERMAN, S.; and EISENBERG, E. (1973): Childhood Hypophosphatasia and the Premature Loss of Teeth, Oral Swg Oral Med Oral Pathol 35:631-640. DASTMALCHI, R.; POLSON, A.; BOUWSMA, O.; and PROSKIN, H. (1990): Ceementum Thickness and Mesial Drift, J Clin Periodontol 17:709-713.
Fig. 7-(a) (left) Afibrillar acellular root cementum (AAC) formed in relation to a mouse incisor that had been prevented from erupting for a period of six months. The specimen was demineralized with EDTA prior to being embedded. For details, see Beertsen et al. (1983). Note the presence of incremental lines. (b) (right) At higher magnification, the cementum appeared to consist of granular material of moderate electron density. D = dentin.
Downloaded from jdr.sagepub.com at NORTH DAKOTA STATE UNIV LIB on June 19, 2015 For personal use only. No other uses without permission.
Vol. 70 No. 3
FORMATION OF CALCIFIED LAYERS
GUIS, M.B.; SLOOTWEG, R.N.; and TONINO, G.J.M. (1973): A Biochemical Study of Collagen in the Periodontal Ligament from Erupting and Non-erupting Bovine Incisors, Arch Oral Biol 18:253263. HALACKOVA, Z.; OUDRAN, L.; and KUKLETOVA, M. (1980): Localization of Some Enzymes in the Periodontium of the Rat Molar, Acta Histochem 67:173-179. HUNTER, G.K.; NYBURG, S.C.; and PRITZKER, K.P.H. (1986): Hydroxyapatite Formation in Collagen, Gelatin, and Agarose Gels, Collagen Rel Res 6:229-238. JANAKIDEVI, K.; MURRAY, C.D.; SELL, C.; and HELD, P. (1988): Fluorometric Analysis of DNA in Cell Cultures, Anal Biochem 172:78-81. LINDE, A. and LUSSI, A. (1989): Mineral Induction by Polyanionic Dentin and Bone Proteins at Physiological Ionic Conditions, Connect Tissue Res 21:197-203. LISTGARTEN, M.A. (1975): Afibrillar Dental Cementum in the Rat and Hamster, J Periodont Res 10:158-167. MARIKOVA, Z. and KINDLOVA, M. (1972): Alkaline Phosphatase in the Periodontium of Rat Molars, J Dent Res 51:1502. MELCHER, A.H.; CHEONG, T.; COX, J.; NEMETH, E.; and SHIGA, A. (1986): Synthesis of Cementum-like Tissue in vitro by Cells Cultured from Bone: a Light and Electron Microscope Study, J Periodont Res 21:592-612. MELCHER, A.H.; McCULLOCH, C.A.G.; CHEONG, T.; NEMETH, E.; and SHIGA, A. (1987): Cells from Bone Synthesize Cementum-like and Bone-like Tissue in vitro and May Migrate into Periodontal Ligament in vivo, J Periodont Res 22:246-247. NANCOLLAS, G.H.; LoRE, M.; PEREZ, L.; RICHARDSON, C.; and ZAWACKI, S.J. (1989): Mineral Phases of Calcium Phosphate, Anat Rec 224:234-241. NOJIMA, N.; KOBAYASHI, M.; SHIONOME, M.; TAKAHASHI, N.; SUGA, T.; and HASEGAWA, K. (1990): Fibroblastic Cells Derived from Bovine Periodontal Ligaments Have the Phenotypes of Osteoblasts, J Periodont Res 25:179-185. OSHIMA, M.; KUWATA, F.; OTSUKA, K.; and SUZUKI, K. (1988): Acid and Alkaline Phosphatase Activities of Cultured Human Periodontal Ligament Cells. In: Recent Advances in Clinical Periodontology, Proceedings of the Third Meeting of the Intemational Academy of Periodontology, J. Ishikawa, H. Kawasaki, K. Ikeda, and K. Hasegawa, Eds., Amsterdam: Excerpta Medica, pp. 573-577. PICHE, J.E.; CARNES, D.E.; and GRAVES, D.T. (1989): Initial Characterization of Cells Derived from Human Periodontia, JDent Res 68:761-767. ROBISON, R. (1923): The Possible Significance of Hexosephosphoric Esters in Ossification, Biochem J 17:286-293. SCHROEDER, H. (1986): The Periodontium, Berlin: Springer Verlag, pp. 23-127. STEGEMAN, H. and STALDER, K.H. (1967): Determination of Hydroxyproline, Clin Chim Acta 18:267-273. VAN DEN BOS, T. and BEERTSEN, W. (1987): Effects of 1- Hy-
181
Fig. 8-Incremental lines (arrows) in APase-induced calcified layers deposited along dentin (D). The DS was co-cultured with pericardium for a period of 14 days. F = fibroblast.
droxyethylidene-1,1-Bisphosphonate (HEBP) on the Synthesis of Dentin Matrix Proteins in the Mouse, Collagen Rel Res 7:135-147. WESSELINK, P.R. and BEERTSEN, W. (1989): The Influence of 1- Hydroxyethylidene-1,1-Bisphosphonate (HEBP) on Dental Root Resorption in the Mouse, Calcif Tissue Int 45:104-110. WHYTE, M.P. (1989): Hypophosphatasia. In: The Metabolic Basis of Inherited Disease, C.R. Scriver, A.L. Beaudet, W.S. Sly, and D. Valle, Eds., New York: McGraw-Hill, pp. 2843-2856. YAMASHITA, Y.; SATO, M.; and NOGUCHI, T. (1987): Alkaline Phosphatase in the Periodontal Ligament of the Rabbit and Macaque Monkey, Arch Oral Biol 32:677-678.
Downloaded from jdr.sagepub.com at NORTH DAKOTA STATE UNIV LIB on June 19, 2015 For personal use only. No other uses without permission.
