- Email: [email protected]
Synthesis of collagen type I, type I trimer and type III by embryonic mouse dental epithelial and mesenchymal cells in vitro
206
Biochimica et Biophysiea Acta, 656 (1981) 206-212
Elsevier/North-HollandBiomedicalPress BBA 99971 SYNTHESIS OF COLLAGEN TYPE I, TYPE I TRIMER AND TYPE III BY EMBRYONIC MOUSE DENTAL EPITHELIAL AND MESENCHYMAL CELLS IN VITRO H. LESOT, V. KARCHER-DJURICICand J.V. RUCH lnstitut de Biologie M~dicale, Facult~ de M~dicine, 11, rue Humann, 67085 Strasbourg Cedex (France)
(Received June 9th, 1981)
Key words: Collagen synthesis; Tissue culture; (Mouse dental cell)
Epithelial and mesenchymal dental cells were grown in primary monolayer culture and the ability of both cell types to synthesize interstitial collagens was investigated. Pepsin-solubilized collagens were analyzed by CMcellulose chromatography and both cell types were found to synthesize collagen type I, type III and type I trimer. The collagen phenotype of mesenchymal cells (type I: 82.4%, type III: 8.5%, type I trimer: 9.1%) was different from that of epithelial cells (type I: 71.8%, type III: 9.5%, type I trimer: 18.7%). The radioactivity incorporated into collagen molecules by mesenehymal cells was 34-times greater than the radioactivity incorporated by epithefial cells. This result agreed with previous observations obtained from tissue culture experiments (Lesot, H. and Ruth, J.V. (1979) Biol. Cell. 34, 23-37) which indicated a low synthesis of interstitial collagens by isolated dental epithelia when compared to isolated dental mesenchymes.
Introduction
The composition of the dental extracellular matrix has been investigated using biochemical and immunological methods. At a stage preceding the mineralization of predentin, this matrix was shown to contain glycosaminoglycans [1-3], non-collagenous glycoproteins and collagens. Fibronectin was localized within the mesenchymal component (the dental pulp) and associated with the basement membrane at the epithelio-mesenchymal junction [3-5]. Laminin was present exclusively in the basement membrane at the epithelio-mesenchymal junction and in the basement membranes of blood vessels in the dental pulp [3,5]. Functional odontoblasts have been shown to produce a phosphoprotein involved in the further mineralization of dentin [6,7]. The synthesis of the collagenous components of this matrix has also been extensively analyzed at different stages of tooth development. At embryonic stages, tooth germs synthesized collagen type I, type I trimer, type III and type IV (3,4,8-10]. All these collagen types have also been detected by immunochemistry or by indi-
rect immunofluorescence at the epithelio-mesenchyreal junction [3-5,11,12]. Epithelio-mesenchymal interactions play an important role in the control of tooth morphogenesis and cell differentiation [1317]. Components of the extracellular matrix have been implicated as possible mediators of these interactions [18-21]. For all these reasons, it was important to determine the tissue origin of the collagens present at the epithelio-mesenchymal junction. A partial answer to this question has been obtained by tissue culture experiments. Trelstad and Slavkin [8] previously found that both isolated enamel organs and isolated dental papillae synthesized collagen type I. Based on the presence of an excess of al chains (cq/a2 ~ 3), these authors concluded that the enamel organ also synthesized collagen type IV [8]. In similar experimental conditions, Lesot and Ruch [9] observed that the enamel organ synthesized collagen type I and that the dental papilla produced collagen type I and type III. The synthesis of collagen type IV and type I trimer was not investigated. At a stage preceding the terminal differentiation of odontoblasts, the enamel organ can be disso-
0005-2787/81/0000-0000/$02.50 © 1981 Elsevier/North-HollandBiomedicalPress
207 ciated free from any contaminating mesenchymal cells by trypsin treatment. In these conditions, the enamel organ synthesized only very small amounts of collagen when compared to an equivalent number of dental papillae [9]. To complete these results, we compared qualitatively and quantitatively the interstitial collagens synthesized by epithelial and mesenchymal dental cells in primary monolayer culture. Material and Methods
Cell culture. About 1600 first mandibular molars were dissected from 18-day-old Swiss mouse embryos (vaginal plug = day 0). The teeth were incubated for 90 min at +4°C in 1% trypsin (Difco 1 : 2 5 0 ) in Hanks' balanced salt solution. After washing in Hanks' solution containing 50% fetal calf serum, the tooth components (enamel organs and dental papillae) were mechanically dissociated. Histological controls demonstrated the absence of contamination of either tissue by the other. Enamel organs as well as dental papillae were cut into small fragments which were cultured for 4 days in Ham F 10 synthetic medium supplemented with 20% fetal calf serum. The culture flasks were incubated at 37°C in a humidified incubator under 5% CO2/95% air atmosphere. Radioactive labeling. After 4 days in culture, the cells were incubated for 24 h in Eagle's minimum essential medium (Dulbecco's modification)supplemented with 20% fetal calf serum, 50 gg/ml ~-aminopropionitrile, 100 /~g/ml ascorbic acid and 292 /.tg/ ml L-glutamine. For radioactive labeling of collagens, the cells were then incubated for 6 h in this medium devoid of serum and containing 10 gCi/ml L-[4,53H]proline (30 Ci/mmol) and 10 /.tCi/ml [2-3H]glycine (50 Ci/mmol). To measure the DNA synthesis, cells were incubated for 1 h in this medium containing 5/.tCi/ml [6-3H]thymidine (28 Ci/mmol). Histology. For histological controls, cells were fixed in Carnoy and stained with methylene blue. Collagen extraction and purification. After radioactive labeling, the culture medium was collected and centrifuged for 10 min at 800 Xg. The cell layer was harvested and collagen was extracted from the culture medium and from the cell layer separately by pepsin treatment according to Miller [22]. All extraction and purification procedures were performed at 4°C. In order to inactivate pepsin, the solution was
titrated to pH 8 with 1 M NaOH and was precipitated by dialysis against 0.02 M N%I-IPO4 (pH 8.5) containing 15% (w/v) KC1, in the presence of carrier rat skin type I collagen. The precipitated collagen was collected by centrifugation (25 000 × g, 30 rain), redissolved in 0.1 M acetic acid and dialyzed against the same solution. This fraction containing collagen was then dialyzed against the CM-cellulose buffer, heat denatured (48°C for 30 min) and clarified by centrifugation. The supernatant was analyzed and the material present in the pellet was solubilized by reduction and alkylation. The material soluble in 0.02 M Na2HPO4 (pH 8.5) containing 15% KC1 was precipitated in the presence of carrier type II collagen by 25% saturation with (NI-Ia)2SO4, redissolved in 1.0 M NaC1/0.05 M TrisHC1 (pH 7.5) and subsequently dialyzed against 0.01 M Na2HPO4. The precipitated collagen was collected by centrifugation and furtlaer analyzed. Reduction and alkylation. Samples were dissolved in 8.0 M urea/0.05 M Tris-HC1 (pH 7.5) and reduced with 2 mM dithiothreitol at room temperature under nitrogen for 2 h. The solution was made up to 4 mM iodoacetate. After 2 h, alkylation was stopped by adding excess 2-mercaptoethanol and dialysis against CM-cellulose buffer. CM-cellulose chromatography. After heat denaturation and clarification, a chains were separated by chromatography on a 0.5 ×4.5 cm CM-cellulose column according to Bornstein and Piez [23]. The column was preequilibrated with 0.04 M sodium acerate/4.0 M urea (pH 4.8) and eluted with a superimposed gradient from 0 to 0.1 M NaC1 over a total volume of 50 ml. Equilibration, loading and elution of the column were performed at 42°C at a flow rate of 16 ml/h. Absorbance at 230 nm was monitored continuously and 1-ml fractions were collected. Aliquots of the fractions were mixed with 10 vol. Instagel and the radioactivity was measured in a liquidscintillation counter. DNA analysis. DNA was extracted and purified according to Howell [24] and was quantified by measuring the absorbance at 260 nm and by the diphenylamine method as described by Burton [25]. To estimate the DNA synthesis, the amount of [3H]thymidine incorporated by the cells was determined after 5 days in culture. An aliquot of the DNA frac-
208 tion was mixed with ! 0 vol. Instagel and the radioactivity was counted.
Results Cells were cultured in Ham's F 10 synthetic medium supplemented with 20% fetal calf serum: these conditions were found to be optimal for the epithelial cells (Fig. la) and identical conditions were used for mesenchymal cells (Fig. 1b). After 5 days in culture, neither epithelial cells nor mesenchymal cells had reached the stationary phase. Both cell types actively incorporated [aH]thymidine;nevertheless this incorporation was relatively greater in the mesenchymal cells (Table I). After labeling of the cells for 6 h with [aH]glycine and [aH]proline, the culture medium was centrifuged to remove any cells present in this fraction. The cell layer was not washed with acetic acid but scrapped and collected as a second fraction. The presence of collagen was then investigated and interstitial collagens were analyzed qualitatively and quantitatively by CM-cellulose chromatography. These analyses were performed separately for each cell type and each fraction. A collagenous fraction, solubilized by pepsin treatment was precipitated first by 15% KC1 at neutral pH in the presence o f carrier type I collagen. The precipitated material consisted of a fraction soluble in the CM-cellulose start buffer and a fraction requiring reduction and alkylation before it was soluble in this buffer. The fraction immediately soluble in the CMcellulose start buffer was chromatographed on a CMcellulose column. The elution pattern showed a radioactive peak of non-collagenous material not retained
Fig. 1. Primary monolayer cultures of dental epithelial (a) and mesenchymal (b) cells. Cells were stained with methylene blue. Magnification = ×800.
TABLE I DNA CONTENT AND [3H]THYMIDINE INCORPORATION IN PRIMARY MONOLAYER CULTURES OF DENTAL MESENCHYMALAND EPITHELIAL CELLS Mesenchymal cells
Epithelial ceils
DNA a
0~g/cell culture flask) [ 3HI Thymidine b (cpm/cell culture flask) [ 3H ] Thymidine (cpm/t~g DNA)
4.5
2.5
118 300
28 400
26 300
11 400
a Determined by measuring the absorbance at 260 nm and by the diphenylamine method after extraction and purification. b Determined by measuring the radioactivity in the DNA fraction after extraction and purification as described in Material and Methods.
on the column (Figs. 2a, d and 3a, d). A second and a third peak of radioactivity eluted respectively with carrier al(I) and a2 chains (Figs. 2a, d and 3a, d). This fraction corresponded to type I collagen * The fraction soluble in the CM-cellulose start buffer after reduction and alkylation was analyzed in the same way. After CM-cellulose chromatography, a peak of non-collagenous material eluted first from the column (Figs. 2b, e and 3b, e) and a second peak of radioactivity eluted between ~12 and a2 components of marker type I collagen (Figs. 2b, e and 3b, e): this fraction corresponded to al(III) chains*. The differential solubility of these two collagen types allowed a very good separation which facilitated further quantitative analysis. Radioactive material remained soluble in the presence of 15% KC1 at neutral pH and was then precipitated in the presence of carrier type II collagen,
* The identification of the different peaks of radioactivity obtained after CM-cellulose chromatography was performed by means of electrophoretic separation of native chains and of their related CNBr-peptides. These data have been previously published in detail for collagen type I and type III [9] and type I trimer [10], extracted from cultured embryonic mouse tooth germs, isolated enamel organs and dental papillae.
209 ~x.lJ)
® ~1(11
(~)
::1.
i
(J !
.~,,~
,%.-
T
_
J
1!i®
|
?
o
s
=E
0. U
\i e~...i, _
i ~,.
0
.I; V' \~. f-~2
,~(ll
®
3.
?
o
0O
c~1(I) ~r-~x2
®
(~)
3.
i'
2
c~I(I)
24
G
S
3.
m:k
i
°1 i"'Fl ®
G
O
0
0
c(1~-t
®
',2\ l
l
l
l
l
o(1(11)
f- ~I{II)
24
l
=L :1.
•
o "~. 16
8
G
0
10
~
2O
30
40
O
50 6O-
FRACTIONS
o 10 I~
! 30
,k__
40 SO I ~ FRACTIONS
!
®
i ¢:1
IL (J
E
/\/
o
/
\ %
FRACTIONS
FRACTIONS
Figs. 2 and 3. CM-cellulose chromatography of all-labeled collagens synthesized by dental mesenchymal (Fig. 2, left-hand figures) and epitheliel (Fig. 3, right-hand f~ures) cells. Collagens were extracted by pepsin treatment of the cell layer (a, b, c) and of culture medium (d, e, f). The solubilized material was precipitated in the presence of carrier type I collagen with 0.02 M Na~HPO4 (pH 8.5) containing 15% KCI, collected, heat denatured and clarified by eentrifugation. The supernatant (a, d) and another fraction, solubilized by reduction and alkylation of the pellet Co, e) were ehromatographed. A collagenous fraction, soluble after precipitation with 0.02 M Na2HPO4 (pH 8.5) containing 15% KC1, was successively precipitated in the presence of carrier type II collagen with 25% (NH4)2SO4 and with 0.01 M Na2HPO4. The precipitated material was redissolved in 0.5 M acetic acid, dialyzed against 0.1 M acetic acid and then against the CM-eellulose buffer. The collagen was heat denatured, clarified and ehromatographed (c, f). The column (0.5 X 4.5 era) was equilibrated with 0.04 M sodium citrate/4.0 M urea (pH 4.8) and eluted with superimposed gradient of 0-0.1 M NaC1in a volume of 50 ml. G indicates the start of the gradient. The positions of al (I), c~2 and • 1(II) chains of carrier collagens are indicated. successively by 25% (NI-I4)2SO4 and by 0.01 M Na2HPO4. The radioactivity precipitated in these conditions was analyzed by chromatography on a CM-cellulose column. The first peak of radioactivity which eluted corresponded to non-collagenous matedal (Figs. 2c, f and 3c, f). A second peak o f radioactivity co-eluted with ott(II) chains of carrier type II collagen (Figs. 2c, f and 3c, f): this fraction consisted of collagen type I trimer * * See footnote, previous page.
These qualitative results demonstrated that dental mesenchymal cells as well as dental epithelial cells synthesized collagen type I, type III and type I trimer. Collagen type II was not detected. The quantitative results showed that the mesenchymal cells synthesized 34.tirnes more collagen than did the epithelial cells (Table II). The data presented in Table II further indicated that the collagen phenotype was different for epithelial and mesenchymal cells: particularly, the proportion of collagen type I trimer was twice as great in epithelial cells (18.7%) as
210 TABLE II PRODUCTION OF INTERSTITIAL COLLAGENS BY MOUSE DENTAL MESENCHYMAL AND EPITHELIAL CELLS IN PRIMARY MONOLAYER CULTURE Cells were labeled for 6 h with 10/aCi/ml [3H]glycine and 10 ~Ci/ml [3H]proline. The different collagen types were purified by CM-cellulose chromatography. The values were obtained by totalling the radioactivity under appropriate peaks. Total collagen represents the total amount of collagen type I, type III and type I trimer. Values are expressed as cpm/ug DNA. Figures in parentheses are percentages. Epithelial cells
Mesenchymal cells Total
Cell layer
Culture medium
Total
Cell layer
Culture medium
Total collagen
419 550 (100.0)
278 250 (66.3)
141 300 (33.7)
12 400 (100.0)
6 050 (48.8)
6 350 (51.2)
Collagen type I
345 800 (82.4 ) 35 600 (8.5) 38 150 (9.1)
224 000 (53.4) 35 350 (8.4) 18 900 (4.5)
121 800 (29.0) 250 (0.1) 19 250 (4.6)
8 900 (71.8) 1 180 (9.5) 2 320 (18.7)
4 100 (33.1 ) 1 130 (9.1) 820 (6.6)
4 800 (38.7) 50 (0.4) 1 500 (12.1)
Collagen type III Collagen type I trimer
in the mesenchymal cells (9.l%). Collagen type I was the most abundant collagen type representing 71.8% of the radioactivity incorporated into collagens by epithelial cells and 82.4% by mesenchymal cells (Table II). Another point concerned the distribution of the interstitial collagens between the culture medium and the cell layer (Table II). After 6 h of labeling, all interstitial collagen types were found in higher proportion in the culture medium of epithelial cells than in the culture medium of mesenchymal cells (Table II). Collagen type III was almost absent from the culture medium of both cell types (Table II).
Discussion The present data show the ability of both isolated dental epithelial and mesenchymal cells grown in primary monolayer culture to synthesize all interstitial collagens except collagen type II. After 6 h of labeling, only 34% of the radioactivity incorporated into collagen molecules by mesenchymal cells was present in the culture medium; the proportion of collagen synthesized by epithelial cells and released into the culture medium was higher (51%). These relatively low amounts of collagen present in the
culture medium of dental cells might be attributed to the short period of labeling (6 h) as this has been observed for cloned cells from rat liver parenchyma [26]. Collagen type IV was not identified and this could be explained by (a) the low amount, (b) the relative insolubility or (c) the susceptibility to pepsin treatment of this collagen type. Dental mesenchymal cells produce collagen type I (82.4%), type I trimer (9.1%) and type III (8.5%). The collagen phenotype of dental mesenchymal cells in primary monolayer culture is different from that of the same cells in situ. At a stage preceding the odontoblast terminal differentiation, cultured tooth germs synthesized collagen type I (64.8%), type I trimer (14.6%) and type III (20.6%) [10]. Such changes in the phenotype of cells, related to culture conditions, have been described previously [27-29] and could result from several factors such as (a) physiologically specific cell interactions, (b) interactions between cultured cells and molecules present in the culture medium or the serum [29], (c) modification in the cell density [30] or (d) selection of cell types during the culture. It is noteworthy that cells from the dental pulp, cultured in monolayer and reassociated with enamel organs, maintained their specific inductive potentiali-
211 ties and remained able to control the histogenesis of the enamel organ [31,32]. This control remained possible even when the dental mesenchymal cells had been cultured for 15 days [31]. Gotoh et al. [33] analyzed the procoUagens synthesized by cultured odontogenic cells: after 25 passages, these cells were found to produce 60% procollagen type III and 40% procollagen type I. Concerning collagen types I and III, the discrepancy between our data and that of Gotoh et al. resulted from the fact that these authors quantitated the relative amounts of procollagens while we measured the total radioactivity incorporated into both collagen and procollagen molecules. Furthermore, it has previously been shown that freshly dissociated chick tendon fibroblasts in suspension cultures did not secrete detectable amounts of collagen type III [34], although collagen type III was secreted into the culture medium when tendon fibroblasts were cultured in monolayer [35]. Studies by indirect immunofluoresence showed that the proportion of fibroblasts synthesizing collagen type III increased as a function of the number of passages [35]. The results presented here demonstrated that the dental epithelial ceils also synthesized collagen type I (71.8%), type I trimer (18.7%) and type III (9.5%). The use of specific antibodies and observations made by indirect immunofluorescence showed that no collagen (type I, type II, type III or type IV) could be detected within the enamel organ itself. However, collagen type I, type III and type IV were present at the epithelio-mesenchymal junction [3-5,11 ]. The cell culture experiments do no exclude a possible participation of the cells of the enamel organ in the secretion of the interstitial collagens localized at the epithelio-mesenchymal junction in tooth germ. However, these results have to be interpreted carefully, as the phenotype of epithelial cells may alter during transfer from the environment in vivo to conditions in vitro. The amount of radioactivity incorporated into collagen molecules by epithelial cells was from 16- to 39-times (according to the collagen type) less than the radioactivity incorporated by mesenchymal cells. The low amount of collagen produced by dental epithelial cells correlates well with previous results obtained by tissue culture experiments: in these conditions, isolated enamel organs secreted 10times less collagen than the dental papilla [9]. After
immunohistochemical or immunofluorescent staining with anti-procoUagen type I [12] or anti-procollagen type III [4,5,11,12] antibodies, the enamel organ remained negative. A definitive answer to the real participation of the inner and outer dental epithelial cells in the secretion of the interstitial collagens localized at the epithelio-mesenchymal junction will only be obtained by the use of antibodies specific for procollagen type I and procollagen type III and the observation of the enamel organ by electron microscopy.
Acknowledgement This work was supported by an INSERM grant.
References 1 Linde, A. (1973) Calc. Tissue Res. 12, 281-294 2 Hjerpe, A. and Engfeldt, B. (1976) Calc. Tissue Res. 22, 173-182 3 Thesleff, I., Barrach, H.J., Foidart, J.M., Vaheri, A., Pratt, R.M. and Martin, G.R. (1981) Develop. Biol. 81, 182-192 4 Thesleff, I., Stenman, S., Vaheri, A. and Timpl, R. (1979) Develop. Biol. 70, 116-126 5 Lesot, H., Osman, M. and Ruth, J.V. (1981) Develop. Biol. 82, 371-381 6 Veis, A. and Schlueter, R.J. (1964) Biochemistry 3, 1650-1657 7 Butler, W.T., Finch, J,E. and Desteno, C.V. (1972) Biochim. Biophys. Acta 257, 167-171 8 Trelstad, R.L. and Slavkin, H.C. (1974) Biochem. Biophys. Res. Commun. 59,443-449 9 Lesot, H. and Ruth, J.V. (1979) Biol. Cell, 34, 23-37 10 Lesot, H. (1981) Eur. J. Biochem. 116,541-546 11 Lesot, H., Von der Mark, K. and Ruch, J.V. (1978) C.R. Acad. Sci. Paris 286,765-768 12 Cournil, I., Leblond, C.P., Pomponio, J., Hand, A.R., Sederlof, L. and Martin, G.R. (1979) J. Histochem. Cytochem. 27, 1059-1069 13 Koch. W.E. (1967) J. Exp. Zool. 165,155-170 14 Kollar, E.J. and Baird, G.R. (1970) J. Embryol. Exp. Morphol. 24,173-186 15 Ruth, J.V. and Karcher-Djuricic, V. (1975) in Extraceilular Matrix Influences on Gene Expression (Slavkin, H.C. and Greulich, R.C., eds.), pp. 549-554, Academic Press, New York 16 Slavkin, H.C. and Bringas, P. (1976) Develop. Biol. 50, 428-442 17 Thesleff, I. and Hurmerinta, K. (1981) Differentiation 18, 75-88 18 Karcher-Djuricie, V., Osman, M., Meyer, J.M., Staubli, A. and Ruch, J.V. (1978) J. Biol. Buccale 6,257-265
212 19 Thesleff, I., (1978) J. Biol. Buccale 6,241-249 20 Ruch, J.V., Fabre, M., Kareher-Djudcic, V. and Staubli, A. (1974) Differentiation 2, 211-220 21 Slavkin, H.C. (1978) J. Biol. Buccale 6, 189-204 22 Miller, E.J. (1972) Biochemistry 11, 4903-4909 23 Bornstein, P. and Piez, K.A. (1966) Biochemistry 5, 3460-3473 24 Howell, S.H. (1973) in Molecular Techniques and Approaches in Developmental Biology (Chrispeels, M.J., ed.), pp. 117-139, John Wiley & Sons, New York 25 Burton, K. (1956) Biochem. J. 62, 315-323 26 Hata, R.I., Ninomiya, Y., Nagai, Y. and Tsukada, Y. (1980) Biochemistry 19, 169-176 27 Cheung, H.S., Harvey, W., Benya, P.D. and Nimni, M.E. (1976) Biochem. Biophys. Res. Commun. 68, 13711378
28 Benya, P.D., Padilla, S.R. and Nimni, M.E. (1978) Cell 15, 1313-1321 29 Conrad, G.W., Dessau, W. and Von der Mark, K. (1980) J. Cell. Biol. 84, 501-512 30 Abe, S., Steinman, B.U., Wahl, L.M. and Martin, G.R. (1979) Nature (Lond,) 279,442-444 31 KoUar, E.J. (1972) in Developmental Aspects of Oral Biology (Slavkin, H.C. and Bavetta, L.A., eds.), pp. 125149, Academic Press, New York 32 Osman, M., Karcher-Djuricic, V. and Ruch. J.V. (1977) Arch. Anat. Microsc. Morphol. Exp. 66,109-118 33 Gotoh, Y., Saito, S. and Sato, A. (1979) Biochim. Biophys. Acta 587, 253-262 34 Uitto, J,, Lichtenstein, J.R. and Bauer, E.A. (1976) Biochemistry 15, 4935-4942 35 Herrmann, H., Dessau, W., Fessler, L.I. and Von der Mark, K. (1980) Eur. J. Biochem. 105, 63-74
Biochimica et Biophysiea Acta, 656 (1981) 206-212
Elsevier/North-HollandBiomedicalPress BBA 99971 SYNTHESIS OF COLLAGEN TYPE I, TYPE I TRIMER AND TYPE III BY EMBRYONIC MOUSE DENTAL EPITHELIAL AND MESENCHYMAL CELLS IN VITRO H. LESOT, V. KARCHER-DJURICICand J.V. RUCH lnstitut de Biologie M~dicale, Facult~ de M~dicine, 11, rue Humann, 67085 Strasbourg Cedex (France)
(Received June 9th, 1981)
Key words: Collagen synthesis; Tissue culture; (Mouse dental cell)
Epithelial and mesenchymal dental cells were grown in primary monolayer culture and the ability of both cell types to synthesize interstitial collagens was investigated. Pepsin-solubilized collagens were analyzed by CMcellulose chromatography and both cell types were found to synthesize collagen type I, type III and type I trimer. The collagen phenotype of mesenchymal cells (type I: 82.4%, type III: 8.5%, type I trimer: 9.1%) was different from that of epithelial cells (type I: 71.8%, type III: 9.5%, type I trimer: 18.7%). The radioactivity incorporated into collagen molecules by mesenehymal cells was 34-times greater than the radioactivity incorporated by epithefial cells. This result agreed with previous observations obtained from tissue culture experiments (Lesot, H. and Ruth, J.V. (1979) Biol. Cell. 34, 23-37) which indicated a low synthesis of interstitial collagens by isolated dental epithelia when compared to isolated dental mesenchymes.
Introduction
The composition of the dental extracellular matrix has been investigated using biochemical and immunological methods. At a stage preceding the mineralization of predentin, this matrix was shown to contain glycosaminoglycans [1-3], non-collagenous glycoproteins and collagens. Fibronectin was localized within the mesenchymal component (the dental pulp) and associated with the basement membrane at the epithelio-mesenchymal junction [3-5]. Laminin was present exclusively in the basement membrane at the epithelio-mesenchymal junction and in the basement membranes of blood vessels in the dental pulp [3,5]. Functional odontoblasts have been shown to produce a phosphoprotein involved in the further mineralization of dentin [6,7]. The synthesis of the collagenous components of this matrix has also been extensively analyzed at different stages of tooth development. At embryonic stages, tooth germs synthesized collagen type I, type I trimer, type III and type IV (3,4,8-10]. All these collagen types have also been detected by immunochemistry or by indi-
rect immunofluorescence at the epithelio-mesenchyreal junction [3-5,11,12]. Epithelio-mesenchymal interactions play an important role in the control of tooth morphogenesis and cell differentiation [1317]. Components of the extracellular matrix have been implicated as possible mediators of these interactions [18-21]. For all these reasons, it was important to determine the tissue origin of the collagens present at the epithelio-mesenchymal junction. A partial answer to this question has been obtained by tissue culture experiments. Trelstad and Slavkin [8] previously found that both isolated enamel organs and isolated dental papillae synthesized collagen type I. Based on the presence of an excess of al chains (cq/a2 ~ 3), these authors concluded that the enamel organ also synthesized collagen type IV [8]. In similar experimental conditions, Lesot and Ruch [9] observed that the enamel organ synthesized collagen type I and that the dental papilla produced collagen type I and type III. The synthesis of collagen type IV and type I trimer was not investigated. At a stage preceding the terminal differentiation of odontoblasts, the enamel organ can be disso-
0005-2787/81/0000-0000/$02.50 © 1981 Elsevier/North-HollandBiomedicalPress
207 ciated free from any contaminating mesenchymal cells by trypsin treatment. In these conditions, the enamel organ synthesized only very small amounts of collagen when compared to an equivalent number of dental papillae [9]. To complete these results, we compared qualitatively and quantitatively the interstitial collagens synthesized by epithelial and mesenchymal dental cells in primary monolayer culture. Material and Methods
Cell culture. About 1600 first mandibular molars were dissected from 18-day-old Swiss mouse embryos (vaginal plug = day 0). The teeth were incubated for 90 min at +4°C in 1% trypsin (Difco 1 : 2 5 0 ) in Hanks' balanced salt solution. After washing in Hanks' solution containing 50% fetal calf serum, the tooth components (enamel organs and dental papillae) were mechanically dissociated. Histological controls demonstrated the absence of contamination of either tissue by the other. Enamel organs as well as dental papillae were cut into small fragments which were cultured for 4 days in Ham F 10 synthetic medium supplemented with 20% fetal calf serum. The culture flasks were incubated at 37°C in a humidified incubator under 5% CO2/95% air atmosphere. Radioactive labeling. After 4 days in culture, the cells were incubated for 24 h in Eagle's minimum essential medium (Dulbecco's modification)supplemented with 20% fetal calf serum, 50 gg/ml ~-aminopropionitrile, 100 /~g/ml ascorbic acid and 292 /.tg/ ml L-glutamine. For radioactive labeling of collagens, the cells were then incubated for 6 h in this medium devoid of serum and containing 10 gCi/ml L-[4,53H]proline (30 Ci/mmol) and 10 /.tCi/ml [2-3H]glycine (50 Ci/mmol). To measure the DNA synthesis, cells were incubated for 1 h in this medium containing 5/.tCi/ml [6-3H]thymidine (28 Ci/mmol). Histology. For histological controls, cells were fixed in Carnoy and stained with methylene blue. Collagen extraction and purification. After radioactive labeling, the culture medium was collected and centrifuged for 10 min at 800 Xg. The cell layer was harvested and collagen was extracted from the culture medium and from the cell layer separately by pepsin treatment according to Miller [22]. All extraction and purification procedures were performed at 4°C. In order to inactivate pepsin, the solution was
titrated to pH 8 with 1 M NaOH and was precipitated by dialysis against 0.02 M N%I-IPO4 (pH 8.5) containing 15% (w/v) KC1, in the presence of carrier rat skin type I collagen. The precipitated collagen was collected by centrifugation (25 000 × g, 30 rain), redissolved in 0.1 M acetic acid and dialyzed against the same solution. This fraction containing collagen was then dialyzed against the CM-cellulose buffer, heat denatured (48°C for 30 min) and clarified by centrifugation. The supernatant was analyzed and the material present in the pellet was solubilized by reduction and alkylation. The material soluble in 0.02 M Na2HPO4 (pH 8.5) containing 15% KC1 was precipitated in the presence of carrier type II collagen by 25% saturation with (NI-Ia)2SO4, redissolved in 1.0 M NaC1/0.05 M TrisHC1 (pH 7.5) and subsequently dialyzed against 0.01 M Na2HPO4. The precipitated collagen was collected by centrifugation and furtlaer analyzed. Reduction and alkylation. Samples were dissolved in 8.0 M urea/0.05 M Tris-HC1 (pH 7.5) and reduced with 2 mM dithiothreitol at room temperature under nitrogen for 2 h. The solution was made up to 4 mM iodoacetate. After 2 h, alkylation was stopped by adding excess 2-mercaptoethanol and dialysis against CM-cellulose buffer. CM-cellulose chromatography. After heat denaturation and clarification, a chains were separated by chromatography on a 0.5 ×4.5 cm CM-cellulose column according to Bornstein and Piez [23]. The column was preequilibrated with 0.04 M sodium acerate/4.0 M urea (pH 4.8) and eluted with a superimposed gradient from 0 to 0.1 M NaC1 over a total volume of 50 ml. Equilibration, loading and elution of the column were performed at 42°C at a flow rate of 16 ml/h. Absorbance at 230 nm was monitored continuously and 1-ml fractions were collected. Aliquots of the fractions were mixed with 10 vol. Instagel and the radioactivity was measured in a liquidscintillation counter. DNA analysis. DNA was extracted and purified according to Howell [24] and was quantified by measuring the absorbance at 260 nm and by the diphenylamine method as described by Burton [25]. To estimate the DNA synthesis, the amount of [3H]thymidine incorporated by the cells was determined after 5 days in culture. An aliquot of the DNA frac-
208 tion was mixed with ! 0 vol. Instagel and the radioactivity was counted.
Results Cells were cultured in Ham's F 10 synthetic medium supplemented with 20% fetal calf serum: these conditions were found to be optimal for the epithelial cells (Fig. la) and identical conditions were used for mesenchymal cells (Fig. 1b). After 5 days in culture, neither epithelial cells nor mesenchymal cells had reached the stationary phase. Both cell types actively incorporated [aH]thymidine;nevertheless this incorporation was relatively greater in the mesenchymal cells (Table I). After labeling of the cells for 6 h with [aH]glycine and [aH]proline, the culture medium was centrifuged to remove any cells present in this fraction. The cell layer was not washed with acetic acid but scrapped and collected as a second fraction. The presence of collagen was then investigated and interstitial collagens were analyzed qualitatively and quantitatively by CM-cellulose chromatography. These analyses were performed separately for each cell type and each fraction. A collagenous fraction, solubilized by pepsin treatment was precipitated first by 15% KC1 at neutral pH in the presence o f carrier type I collagen. The precipitated material consisted of a fraction soluble in the CM-cellulose start buffer and a fraction requiring reduction and alkylation before it was soluble in this buffer. The fraction immediately soluble in the CMcellulose start buffer was chromatographed on a CMcellulose column. The elution pattern showed a radioactive peak of non-collagenous material not retained
Fig. 1. Primary monolayer cultures of dental epithelial (a) and mesenchymal (b) cells. Cells were stained with methylene blue. Magnification = ×800.
TABLE I DNA CONTENT AND [3H]THYMIDINE INCORPORATION IN PRIMARY MONOLAYER CULTURES OF DENTAL MESENCHYMALAND EPITHELIAL CELLS Mesenchymal cells
Epithelial ceils
DNA a
0~g/cell culture flask) [ 3HI Thymidine b (cpm/cell culture flask) [ 3H ] Thymidine (cpm/t~g DNA)
4.5
2.5
118 300
28 400
26 300
11 400
a Determined by measuring the absorbance at 260 nm and by the diphenylamine method after extraction and purification. b Determined by measuring the radioactivity in the DNA fraction after extraction and purification as described in Material and Methods.
on the column (Figs. 2a, d and 3a, d). A second and a third peak of radioactivity eluted respectively with carrier al(I) and a2 chains (Figs. 2a, d and 3a, d). This fraction corresponded to type I collagen * The fraction soluble in the CM-cellulose start buffer after reduction and alkylation was analyzed in the same way. After CM-cellulose chromatography, a peak of non-collagenous material eluted first from the column (Figs. 2b, e and 3b, e) and a second peak of radioactivity eluted between ~12 and a2 components of marker type I collagen (Figs. 2b, e and 3b, e): this fraction corresponded to al(III) chains*. The differential solubility of these two collagen types allowed a very good separation which facilitated further quantitative analysis. Radioactive material remained soluble in the presence of 15% KC1 at neutral pH and was then precipitated in the presence of carrier type II collagen,
* The identification of the different peaks of radioactivity obtained after CM-cellulose chromatography was performed by means of electrophoretic separation of native chains and of their related CNBr-peptides. These data have been previously published in detail for collagen type I and type III [9] and type I trimer [10], extracted from cultured embryonic mouse tooth germs, isolated enamel organs and dental papillae.
209 ~x.lJ)
® ~1(11
(~)
::1.
i
(J !
.~,,~
,%.-
T
_
J
1!i®
|
?
o
s
=E
0. U
\i e~...i, _
i ~,.
0
.I; V' \~. f-~2
,~(ll
®
3.
?
o
0O
c~1(I) ~r-~x2
®
(~)
3.
i'
2
c~I(I)
24
G
S
3.
m:k
i
°1 i"'Fl ®
G
O
0
0
c(1~-t
®
',2\ l
l
l
l
l
o(1(11)
f- ~I{II)
24
l
=L :1.
•
o "~. 16
8
G
0
10
~
2O
30
40
O
50 6O-
FRACTIONS
o 10 I~
! 30
,k__
40 SO I ~ FRACTIONS
!
®
i ¢:1
IL (J
E
/\/
o
/
\ %
FRACTIONS
FRACTIONS
Figs. 2 and 3. CM-cellulose chromatography of all-labeled collagens synthesized by dental mesenchymal (Fig. 2, left-hand figures) and epitheliel (Fig. 3, right-hand f~ures) cells. Collagens were extracted by pepsin treatment of the cell layer (a, b, c) and of culture medium (d, e, f). The solubilized material was precipitated in the presence of carrier type I collagen with 0.02 M Na~HPO4 (pH 8.5) containing 15% KCI, collected, heat denatured and clarified by eentrifugation. The supernatant (a, d) and another fraction, solubilized by reduction and alkylation of the pellet Co, e) were ehromatographed. A collagenous fraction, soluble after precipitation with 0.02 M Na2HPO4 (pH 8.5) containing 15% KC1, was successively precipitated in the presence of carrier type II collagen with 25% (NH4)2SO4 and with 0.01 M Na2HPO4. The precipitated material was redissolved in 0.5 M acetic acid, dialyzed against 0.1 M acetic acid and then against the CM-eellulose buffer. The collagen was heat denatured, clarified and ehromatographed (c, f). The column (0.5 X 4.5 era) was equilibrated with 0.04 M sodium citrate/4.0 M urea (pH 4.8) and eluted with superimposed gradient of 0-0.1 M NaC1in a volume of 50 ml. G indicates the start of the gradient. The positions of al (I), c~2 and • 1(II) chains of carrier collagens are indicated. successively by 25% (NI-I4)2SO4 and by 0.01 M Na2HPO4. The radioactivity precipitated in these conditions was analyzed by chromatography on a CM-cellulose column. The first peak of radioactivity which eluted corresponded to non-collagenous matedal (Figs. 2c, f and 3c, f). A second peak o f radioactivity co-eluted with ott(II) chains of carrier type II collagen (Figs. 2c, f and 3c, f): this fraction consisted of collagen type I trimer * * See footnote, previous page.
These qualitative results demonstrated that dental mesenchymal cells as well as dental epithelial cells synthesized collagen type I, type III and type I trimer. Collagen type II was not detected. The quantitative results showed that the mesenchymal cells synthesized 34.tirnes more collagen than did the epithelial cells (Table II). The data presented in Table II further indicated that the collagen phenotype was different for epithelial and mesenchymal cells: particularly, the proportion of collagen type I trimer was twice as great in epithelial cells (18.7%) as
210 TABLE II PRODUCTION OF INTERSTITIAL COLLAGENS BY MOUSE DENTAL MESENCHYMAL AND EPITHELIAL CELLS IN PRIMARY MONOLAYER CULTURE Cells were labeled for 6 h with 10/aCi/ml [3H]glycine and 10 ~Ci/ml [3H]proline. The different collagen types were purified by CM-cellulose chromatography. The values were obtained by totalling the radioactivity under appropriate peaks. Total collagen represents the total amount of collagen type I, type III and type I trimer. Values are expressed as cpm/ug DNA. Figures in parentheses are percentages. Epithelial cells
Mesenchymal cells Total
Cell layer
Culture medium
Total
Cell layer
Culture medium
Total collagen
419 550 (100.0)
278 250 (66.3)
141 300 (33.7)
12 400 (100.0)
6 050 (48.8)
6 350 (51.2)
Collagen type I
345 800 (82.4 ) 35 600 (8.5) 38 150 (9.1)
224 000 (53.4) 35 350 (8.4) 18 900 (4.5)
121 800 (29.0) 250 (0.1) 19 250 (4.6)
8 900 (71.8) 1 180 (9.5) 2 320 (18.7)
4 100 (33.1 ) 1 130 (9.1) 820 (6.6)
4 800 (38.7) 50 (0.4) 1 500 (12.1)
Collagen type III Collagen type I trimer
in the mesenchymal cells (9.l%). Collagen type I was the most abundant collagen type representing 71.8% of the radioactivity incorporated into collagens by epithelial cells and 82.4% by mesenchymal cells (Table II). Another point concerned the distribution of the interstitial collagens between the culture medium and the cell layer (Table II). After 6 h of labeling, all interstitial collagen types were found in higher proportion in the culture medium of epithelial cells than in the culture medium of mesenchymal cells (Table II). Collagen type III was almost absent from the culture medium of both cell types (Table II).
Discussion The present data show the ability of both isolated dental epithelial and mesenchymal cells grown in primary monolayer culture to synthesize all interstitial collagens except collagen type II. After 6 h of labeling, only 34% of the radioactivity incorporated into collagen molecules by mesenchymal cells was present in the culture medium; the proportion of collagen synthesized by epithelial cells and released into the culture medium was higher (51%). These relatively low amounts of collagen present in the
culture medium of dental cells might be attributed to the short period of labeling (6 h) as this has been observed for cloned cells from rat liver parenchyma [26]. Collagen type IV was not identified and this could be explained by (a) the low amount, (b) the relative insolubility or (c) the susceptibility to pepsin treatment of this collagen type. Dental mesenchymal cells produce collagen type I (82.4%), type I trimer (9.1%) and type III (8.5%). The collagen phenotype of dental mesenchymal cells in primary monolayer culture is different from that of the same cells in situ. At a stage preceding the odontoblast terminal differentiation, cultured tooth germs synthesized collagen type I (64.8%), type I trimer (14.6%) and type III (20.6%) [10]. Such changes in the phenotype of cells, related to culture conditions, have been described previously [27-29] and could result from several factors such as (a) physiologically specific cell interactions, (b) interactions between cultured cells and molecules present in the culture medium or the serum [29], (c) modification in the cell density [30] or (d) selection of cell types during the culture. It is noteworthy that cells from the dental pulp, cultured in monolayer and reassociated with enamel organs, maintained their specific inductive potentiali-
211 ties and remained able to control the histogenesis of the enamel organ [31,32]. This control remained possible even when the dental mesenchymal cells had been cultured for 15 days [31]. Gotoh et al. [33] analyzed the procoUagens synthesized by cultured odontogenic cells: after 25 passages, these cells were found to produce 60% procollagen type III and 40% procollagen type I. Concerning collagen types I and III, the discrepancy between our data and that of Gotoh et al. resulted from the fact that these authors quantitated the relative amounts of procollagens while we measured the total radioactivity incorporated into both collagen and procollagen molecules. Furthermore, it has previously been shown that freshly dissociated chick tendon fibroblasts in suspension cultures did not secrete detectable amounts of collagen type III [34], although collagen type III was secreted into the culture medium when tendon fibroblasts were cultured in monolayer [35]. Studies by indirect immunofluoresence showed that the proportion of fibroblasts synthesizing collagen type III increased as a function of the number of passages [35]. The results presented here demonstrated that the dental epithelial ceils also synthesized collagen type I (71.8%), type I trimer (18.7%) and type III (9.5%). The use of specific antibodies and observations made by indirect immunofluorescence showed that no collagen (type I, type II, type III or type IV) could be detected within the enamel organ itself. However, collagen type I, type III and type IV were present at the epithelio-mesenchymal junction [3-5,11 ]. The cell culture experiments do no exclude a possible participation of the cells of the enamel organ in the secretion of the interstitial collagens localized at the epithelio-mesenchymal junction in tooth germ. However, these results have to be interpreted carefully, as the phenotype of epithelial cells may alter during transfer from the environment in vivo to conditions in vitro. The amount of radioactivity incorporated into collagen molecules by epithelial cells was from 16- to 39-times (according to the collagen type) less than the radioactivity incorporated by mesenchymal cells. The low amount of collagen produced by dental epithelial cells correlates well with previous results obtained by tissue culture experiments: in these conditions, isolated enamel organs secreted 10times less collagen than the dental papilla [9]. After
immunohistochemical or immunofluorescent staining with anti-procoUagen type I [12] or anti-procollagen type III [4,5,11,12] antibodies, the enamel organ remained negative. A definitive answer to the real participation of the inner and outer dental epithelial cells in the secretion of the interstitial collagens localized at the epithelio-mesenchymal junction will only be obtained by the use of antibodies specific for procollagen type I and procollagen type III and the observation of the enamel organ by electron microscopy.
Acknowledgement This work was supported by an INSERM grant.
References 1 Linde, A. (1973) Calc. Tissue Res. 12, 281-294 2 Hjerpe, A. and Engfeldt, B. (1976) Calc. Tissue Res. 22, 173-182 3 Thesleff, I., Barrach, H.J., Foidart, J.M., Vaheri, A., Pratt, R.M. and Martin, G.R. (1981) Develop. Biol. 81, 182-192 4 Thesleff, I., Stenman, S., Vaheri, A. and Timpl, R. (1979) Develop. Biol. 70, 116-126 5 Lesot, H., Osman, M. and Ruth, J.V. (1981) Develop. Biol. 82, 371-381 6 Veis, A. and Schlueter, R.J. (1964) Biochemistry 3, 1650-1657 7 Butler, W.T., Finch, J,E. and Desteno, C.V. (1972) Biochim. Biophys. Acta 257, 167-171 8 Trelstad, R.L. and Slavkin, H.C. (1974) Biochem. Biophys. Res. Commun. 59,443-449 9 Lesot, H. and Ruth, J.V. (1979) Biol. Cell, 34, 23-37 10 Lesot, H. (1981) Eur. J. Biochem. 116,541-546 11 Lesot, H., Von der Mark, K. and Ruch, J.V. (1978) C.R. Acad. Sci. Paris 286,765-768 12 Cournil, I., Leblond, C.P., Pomponio, J., Hand, A.R., Sederlof, L. and Martin, G.R. (1979) J. Histochem. Cytochem. 27, 1059-1069 13 Koch. W.E. (1967) J. Exp. Zool. 165,155-170 14 Kollar, E.J. and Baird, G.R. (1970) J. Embryol. Exp. Morphol. 24,173-186 15 Ruth, J.V. and Karcher-Djuricic, V. (1975) in Extraceilular Matrix Influences on Gene Expression (Slavkin, H.C. and Greulich, R.C., eds.), pp. 549-554, Academic Press, New York 16 Slavkin, H.C. and Bringas, P. (1976) Develop. Biol. 50, 428-442 17 Thesleff, I. and Hurmerinta, K. (1981) Differentiation 18, 75-88 18 Karcher-Djuricie, V., Osman, M., Meyer, J.M., Staubli, A. and Ruch, J.V. (1978) J. Biol. Buccale 6,257-265
212 19 Thesleff, I., (1978) J. Biol. Buccale 6,241-249 20 Ruch, J.V., Fabre, M., Kareher-Djudcic, V. and Staubli, A. (1974) Differentiation 2, 211-220 21 Slavkin, H.C. (1978) J. Biol. Buccale 6, 189-204 22 Miller, E.J. (1972) Biochemistry 11, 4903-4909 23 Bornstein, P. and Piez, K.A. (1966) Biochemistry 5, 3460-3473 24 Howell, S.H. (1973) in Molecular Techniques and Approaches in Developmental Biology (Chrispeels, M.J., ed.), pp. 117-139, John Wiley & Sons, New York 25 Burton, K. (1956) Biochem. J. 62, 315-323 26 Hata, R.I., Ninomiya, Y., Nagai, Y. and Tsukada, Y. (1980) Biochemistry 19, 169-176 27 Cheung, H.S., Harvey, W., Benya, P.D. and Nimni, M.E. (1976) Biochem. Biophys. Res. Commun. 68, 13711378
28 Benya, P.D., Padilla, S.R. and Nimni, M.E. (1978) Cell 15, 1313-1321 29 Conrad, G.W., Dessau, W. and Von der Mark, K. (1980) J. Cell. Biol. 84, 501-512 30 Abe, S., Steinman, B.U., Wahl, L.M. and Martin, G.R. (1979) Nature (Lond,) 279,442-444 31 KoUar, E.J. (1972) in Developmental Aspects of Oral Biology (Slavkin, H.C. and Bavetta, L.A., eds.), pp. 125149, Academic Press, New York 32 Osman, M., Karcher-Djuricic, V. and Ruch. J.V. (1977) Arch. Anat. Microsc. Morphol. Exp. 66,109-118 33 Gotoh, Y., Saito, S. and Sato, A. (1979) Biochim. Biophys. Acta 587, 253-262 34 Uitto, J,, Lichtenstein, J.R. and Bauer, E.A. (1976) Biochemistry 15, 4935-4942 35 Herrmann, H., Dessau, W., Fessler, L.I. and Von der Mark, K. (1980) Eur. J. Biochem. 105, 63-74