Glycosaminoglycans in Embryonic Mouse Teeth and the Dissociated Dental Constituents

Differentiation

Differentiation (1983) 23: 234242

Springer-Verlag 1983

Glycosaminoglycans in Embryonic Mouse Teeth and the Dissociated Dental Constituents Eduardo C. Lau and Jean V. Ruch Institut de Biologie Midicale, Faculti de Mtdecine, Universitk Louis Pasteur, F-67085, Strasbourg, France

Abstract. The nature, amounts, and distribution of glycosaminoglycans (GAG) before and during odontoblast terminal differentiation were studied. GAG have been isolated from intact mouse tooth germs and from dissociated dental epithelia and dental papillae after labeling with ['HI glucosamine or 3sSO:- as precursor. The kinds and relative amounts of 3H-labeled GAG were analyzed by chromatography on a DEAE-cellulose column and cellulose thin-layer sheets. The amounts of individual GAG relative to total GAG were determined from the elution profiles, whereas their nature was identified by the selective removal of chromatographic peaks after enzymatic or chemical degradation. We found hyaluronate and probably a minute quantity of heparan sulfate in the dental epithelium, while hyaluronate, heparan sulfate, and chondroitin sulfate were the main types of GAG in the dental papilla. The chondroitin sulfate recovered was further fractionated by cellulose thinlayer chromatography into two isomers, namely chondroitin-4-sulfate (the major component) and chondroitin-6sulfate. Changes in the elution profile from DEAE-cellulose chromatography of tooth GAG extracted from different developmental stages suggest that modifications of GAG occur during odontogenesis. Alcian blue staining localized large amounts of hyaluronate and sulfated GAG along the epithelio-mesenchymal junction. Tissue specificity and changing patterns of GAG were demonstrated during odontogenesis.

Introduction Glycosaminoglycans (GAG) are groups of related heterosaccharides usually composed of two types of alternating monosaccharide units, namely hexosamine and hexuronate [25]. Recently, the importance of GAG has been stressed [2-4, 8, 15, 26, 27-28, 35, 39, 44-46]. GAG stimulate cell aggregation and cell adhesion [27,46], control cell proliferation [8] and cell differentiation [26, 291, and may play a role during morphogenesis [2,4]. Tooth morphogenesis and differentiation also require cell-matrix interactions [39, 451. Current theory supports the notion that the dental basement membrane (BM) - basal lamina and associated material - may play an important role in mediating the interactions between the mesenchymal and epithelial cells, which lead to the differentiation of odontoblasts [15, 35, 42, 441. Since GAG, collagens, and glycoproteins are the major macromolecular components of the dental BM, an under-

standing of the mechanisms of the matrix-mediated cell interactions is not possible without a knowledge of the nature, amounts, distribution, and turnover of these molecules. Though GAG have been widely investigated in a variety of tissues and organisms, as far as we know, no biochemical study has been made on the developing mouse tooth germs. In the work reported in this paper the method of isolating GAG from mouse tooth germs is presented, and results based on biochemical analysis of the types and relative amounts of individual types of GAG in the dental epithelia and in the dental papillae are given. Tissue specificity and stage dependent patterns of GAG are documented. The localization and changing patterns of GAG during the developmental steps of odontogenesis were also demonstrated by histochemical staining.

Methods Collection and Preparation of Tissues Lower first molars from 16, 18-, and 19-day-old laboratory raised Swiss mouse embryos were dissected free. Day 0 of gestation corresponds to the day when the vaginal plug was evident. At day 16, the embryonic teeth contain only preodontoblasts. At day 18, the first polarizing odontoblasts exist. The dental papillae and the dental epithelia were isolated either after trypsin or after EDTA treatment [35]. Some of the day 16 and day 18 tooth germs were cultured in vitro for two days prior to labeling or staining, they are denoted as day 16+2 and day 18+2 teeth, respectively. Trypsin Dissociation The teeth were incubated for 90 min at 4" C in 1% trypsin (Difco 1:250) in Hanks' balanced salt solution (HBSS). After washing in HBSS: fetal calf serum (1 : 1) (v/v), the dental epithelia and the dental papillae were mechanically dissected. In these conditions the BM was always hydrolyzed. ED TA Dissociation The teeth were incubated at room temperature for 5 min in 10 mM EDTA (Merck) in Ca2+ free phosphate buffered saline (PBS), pH 7.3. The tooth constituents were then mechanically dissociated in PBS. The dental epithelium isolated in this way is devoid of the BM, which remains associated with the dental papilla [35].

235

Isotopic Labeling of the GAG

Intact teeth or dissociated dental constituents were pooled separately into groups of 20 and placed in 1 ml Eagle's minimum essential medium containing 10% fetal calf serum and 25 pCi aml-' of D4L3H] glucosamine hydrochloride (CEA, France; specific activity 17 Ci.mmol-') or 100 pCi. ml-' of [35s] sulfate (Amersham, England; activity of 25 Ci-mg-'). Incubation was carried out at 37" C in a controlled atmosphere of 5% COz in air for a period of either 6 h or 12 h. In some cases the teeth or dental constituents were grown for two days on top of a plasma coagulum [34]before labeling. At the end of the incubation period, the tissues were washed in prewarmed HBSS and either dehydrated immediately in acetone for one day at 4" C or stored in acetone at -90" C until used. Isolation of GAG

The labeled material was delipidized in chloroform: methanol (1 :1) (v/v) for one day at 4" C and then dried under reduced pressure. The tissues were rehydrated by boiling in water for 15 min in the presence of 60 pg of total GAG carrier - 15 pg of each of hyaluronate (HA) (Sigma), chondroitin-4-sulfate (C4S) (Sigma), chondroitin-6-sulfate (C6S) (Sigma), and heparan sulfate (HS) (a gift from the Upjohn Co., Kalamazoo). They were then homogenized with a Potter homogenizer and lyophilized to dryness. The lyophilisate was dissolved in 1 ml of 1 mM calcium acetate/O.l mM NazS0,/5 mM glucosamine/lO mM MgCl,/lO mM Tris buffer, pH 8.5 [13] and then exhaustively digested by the endopeptidase thermolysin (EC 3.4.24.4, Sigma Protease type X) at 58" C for one day (6 pg thermolysin per pg of tissue protein was used; the protein contents of tissues and culture media were determined by the method of Lowry [24]). The thermolysin used was tested for the absence of GAG-lytic activity. Enzyme and residual proteins were precipitated with 10% (w/v) TCA. The mixture was blended vigorously and then chilled in an ice-bath for 90 min. Coagulated proteins were precipitated by centrifuging at 25000 g for 40 rnin at 4" C. The pellet was washed with 1 ml of ice cold 5% TCA before discarding. The washing was pooled with the supernatant, extracted with 1 vol of chloroform: methanol (2: 1) (v/v) by agitating vigorously for 3 min, and then extracted with 2 vol of ether. The aqueous layer was dialyzed against 3 mM unlabeled glucosamine for two days at 4" C, and lyopilized to dryness. The lyophilisate was dissolved in 50 pl of 2 mM NaZSO4/0.6M potassium acetate buffer, pH 5.1 and then precipitated with 4.5 vol of ethanol at -20" C for two days. The GAG were recovered by centrifuging at 12000 g for 20 min at 4" C. The pellet was dissolved in 50 pl of 2 mM glucosamine/2 mM NazS04/0.6 M potassium acetate buffer, pH 5.1. The procedures of ethanol precipitation were then repeated. The pellet was finally washed with 85% ethanol, dried, and stored at - 20" C. Gradient Elution of GAG from a DEAE-Cellulose Column

The total GAG were separated by anion-exchange chromatography on DEAE-cellulose which fractionated indiviudal GAG according to their polyanion properties. The method was adapted from that used by Gordon and Bernfield [13]. A DEAE-cellulose (Whatman DE52, preswollen microgranular anion exchanger) column of 5.5 x 20 mm (i.e., a

Pasteur pipette filled to a height of 2cm) was used. The buffer used for equilibration of the column was 1 mM CaClJ1 mM MgC1,/5O mM NaC1/5O mM Tris buffer, pH 7.2. The labeled GAG were loaded with 1.5 mg HA (Sigma), 20 pg HS (Upjohn Co., Kalamazoo), and 0.5 mg chondroitin sulfate (CS) (Sigma) carriers. Elution was carried out at room temperature with 40ml of the same buffer, in a linear gradient of 0.05-0.6 M NaCl at a rate of about 13 m1.h-I. The eluant was collected in fractions of about 450 pl and then quantitated by liquid scintillation spectrometry in 5 ml of Instagel (Packard). The elution positions of HA and CS were established by measuring the ODz18 of the eluant by means of the Gilson Spectrochrom M. The identification of the peaks was established by changes in the elution profile of the chromatogram after differential degradation of the GAG by Streptomyces hyaluronidase (S. Hase) (EC 4.2.99.1 ; Calbiochem or Seikagaku Kogyo), chondroitin ABC lyase (CHaseABC) (EC 4.2.2.4. ; Seikagaku), chondroitin AC-I1 lyase (CHaseACII) (EC 4.2.2.5.; Seikagaku), and nitrous acid. Enzymatic Digestion of GAG

Twenty pg of GAG were digested with 2TRU (turbidity reducing unit [32]) of S. Hase in 100 pl of 0.15 M NaCl/ 0.02% sodium azide/30 mM sodium acetate buffer, pH 5.0, at 37" C for 24 h. HA was the only GAG degraded by S. Hase. The enzyme was finally inactivated by boiling for 5 min. GAG (0.5mg) were digested with 0.1 U of CHaseACII or CHaseABC in 50 pl of enriched Tris buffer [40] in the presence of 10 mM MgCl, and 0.04% NaF (NaF inhibits the activity of the contaminating chondroitinases). For CHaseABC and CHaseACII, the optimum pHs for digesting a mixture of GAG were 8.0 and 6.4 respectively [481. Selective Deamination of Heparan Suljhte and Heparin

The procedures used were a modification of the method of Cifonelli and King [5, 61 as described by Fisher and Schraer [ll]. N-butyl nitrite was freshly synthesized by the action of nitrous acid on n-butanol [31]. Labeled GAG were mixed with 20 pg each of HA, CS, and HS in 1 ml of water and the reaction was started by the addition of 0.5 ml of 1 N HC1 and 0.5 ml of n-butyl nitrite: ethanol (1:4) (v/v). The reaction proceeded at 27" C for 2 h in an open vessel with continuous gentle shaking and the reaction was terminated by the action of 50 mg of solid sodium bicarbonate. Nitrous acid was removed by repeatedly adding dry methanol and evaporating under reduced pressure. Chromatographic Purification of GAG

For analytical purposes, the HA and CS separated by DEAEcellulose chromatography were recovered by podling the fractions under the chromatographic peaks at 0.22 M and 0.44 M of NaCl respectively. The samples were desalted either by extensive dialyzing against water or by filtering on a column (2.5 x 70 cm)of Sephadex G 25 (Pharmacia). The HA fraction could be eluted free from contaminating glycopeptides (GP) by washing the column with 0.093 M NaCl prior to the elution. The contaminating HA in the CS preparation was eliminated by specific degradation with S. Hase and the resulting degradation products were then separated from the CS either by extensive dialyzing against water or by gel filtration on the Sephadex G25

236

column. The HA and CS recovered were characterized by treatment with CHaseACII and separating their degradation products on the cellulose thin-layer chromatographic sheets. Identiji'cation of Disaccharides by Chromatography on Cellulose Thin-layer Sheets In the absence of contaminating chondrosulfatases, the two isomers of CS, namely C4S and C6S, were degraded by CHaseACII to the unsaturated disaccharides dDi-4S and dDi-6S respectively, which could then be separated on the PEI-cellulose F thin-layer sheets (20 x 20 cm, Merck). The reference substances used were the unsaturated disaccharides dDi-OS (a nonsulfated unsaturated disaccharide from chondroitin), dDi-4S (a 4-sulfated unsaturated disaccharide from chondroitin-4-sulfate), and dDi-6S (a 6-sulfated unsaturated disaccharide from chondroitin-6-sulfate) purchased from Seikagaku Kogyo Co. (Tokyo); and dDi-OHA (a nonsulfated unsaturated disaccharide from HA) prepared by digestion of HA (Sigma, type I) with chondroitin ABC lyase (CHaseABC) (Seikagaku Kogyo Co.). Samples were prepared by procedures modified from the method of Wassermann et al. [47l. [3H] GAG extracted from the tooth germs, together with GAG carriers (240 pg each of C4S and C6S) were digested exhaustively with 0.1 U of CHaseACII at 37" C for 24 h. (Under such conditions, HA was almost completely degraded and CS was totally digested, data not shown). Six volumes of absolute ethanol were then added, chilled at -20" C for one day, and centrifuged at 45000 g at 4" C for 1 h. The supernatant was evaporated to dryness under reduced pressure. The samples (about 20000 cpm each) were dissolved in a few microliters of water and 20pg of dDi-OS were added. The samples were then spotted on the thin-layer sheet. The sheet was desalted overnight in n-butanol :ethanol :water (25 :32 :16) by volume [40]and developed for 24 h in n-butanol :acetic acid: 2 N ammonia (2 :3 :1) by volume [l], by ascending chromatography. A maximum separation of the disaccharides was obtained after 24 h of developing and the sheet was photographed under UV illumination. A disaccharide isomer of 6-10 pg gave a clearly visible spot under UV illumination. Fluorography was done by impregnating the sheet with ENHANCE Spray (New England Nuclear). The X-ray film used was XAR-5 X-ray film (Kodak). The exposure time at - 90"C varied from 15 days to three months.

acid, pH 2.5. The AB-PAS method of Mowry [30] was also employed for the staining of GAG. Histologic sections were stained for 2 h in 1% AB 8GX/3% acetic acid, pH 2.5, then oxidized for 10 min in 1% periodic acid and finally counterstained in SchiFs reagent (Gurr) for 10 min. According to Pearse [37l, GAG (both HA and sulfated GAG) are not periodate-reactive. HA and sialomucins are Alcianophilic since they are rich in acidic groups and thus stained blue. The appearance of red staining after the periodate reaction may result from staining of some carbohydrate protein complexes. Other substances which are nonAlcianophilic and periodate-unreactive remain colorless. GAG were also stained by the AB-critical electrolyte concentration (AB-CEC) method [43]. Histologic sections were stained at room temperature in 60 ml of 0.1% AB 8GX/ 50 m M sodium acetate buffer, pH 5.1/0-1.0 M MgCl,, in an upright Coplin jar for 17 h. Results Biochemical Data In agreement with Gordon and Bernfield [13], we found that chromatography of GAG on DEAE-cellulose gave repeatable results. The results of analysis of each specimen were based on at least two consistent observations on GAG obtained from two independent extractions. GAG Extracted from Intact Teeth

Day 16+ 2 and day 18 + 2 teeth were labeled with 25 pCi . ml-' [3H] glucosamine for 12 h. Four [3H]-labeled peaks

Histochemical Methods For histochemical work 16-, 18-, and 19-day-old teeth were used. The 16 or 18 day teeth were grown on top of semisolid coagulum for two days prior to fixation, while the day 19 teeth were fixed immedaitely after dissection. Teeth were fixed with 0.5% cetylpyridinium chloride (CPC) (Sigma) in Carnoy's solution for 45 min at room temperature. CPCcontaining fixatives extract negligible amounts of GAG from tissues and accurately preserve GAG within tissue sections [9]. Sulfated GAG in tissues were demonstrated by Alcian blue (AB) staining at pH 1.0 according to the procedures of Lev and Spicer [17]. Histologic sections were stained for 30 min in 1% AB 8GX (Gurr)/O.l N HC1. HA was demonstrated in the teeth by Alcian blue staining at pH 2.5 according to the procedures of Mowry [29]. Histologic sections were stained for 30 min in 1% AB 8GX/3% acetic

Fig. 1 A-D. Fractionation of total GAG by DEAE-cellulose chromatography according to the gradient elution mcthod. The peaks (1-4) correspond to GP, HA, HS, and CS respectively. A GAG isolated from the day 16 2 teeth, labeled for 12 h with ['HI glucosamine. B GAG isolated from the day 18+2 teeth, labeled for 12 h with [3H] glucosamine. C GAG isolated from the day 18 tooth germs, labeled for 6 h with ['HI glucosamine. D GAG isolated from day 18+2 teeth, labeled for 12 h with [35S] sulfate

+

237

Fraction number Fig. 2A-D. DEAE-cellulose chromatography of the residual GAG after enzymatic or nitrous acid treatment. The day 18+2 teeth were labeled with [3H] glucosamine. GAG were then isolated and analyzed by DEAE-cellulose chromatography according to the gradient elution method. Peaks (1-4) correspond to GP, HA, HS, and CS respectively. A [3H] GAG were treated with 2 TRU of S. Hase. B ['HI GAG, together with GAG carrier, were treated with 0.1 U of CHaseABC. The digestion mixture was then loaded on the column together with GAG carriers. C [3H]GAG, together with GAG carrier, were treated with 0.1 U of CHaseACII. The digestion mixture was loaded on the column with GAG carriers. D [31GAG, 3 together with 20 pg each of HA, C4S, and HS, were treated with nitrous acid. After neutralizing, the residues were loaded on the column together with GAG carriers (20pg each of HA, HS,C4S, and C6S)

with maximum elution at 0.15 M, 0.22 M, 0.30 M, and 0.44 M NaCl were obtained by chromatography of the total GAG extracted (Fig. 1A-C). To identify the chromatographic peaks, the total GAG prepared from the tooth germs were treated with S. Hase, CHaseACII, CHaseABC, or nitrous acid. The peak at 0.22 M NaCl was completely removed by S. Hase treatment, mostly removed by either CHaseABC or CHaseACII treatment, and thus identified as HA. The peak at 0.30 M NaCl was completely removed by nitrous acid treatment, unchanged by either CHaseABC or S. Hase treatment, only slightly degraded by CHaseACII treatment, and thus identified as HS. The peak at 0.44 M NaCl was unchanged by either S. Hase or nitrous acid treatment, but completely removed by either CHaseABC or CHaseACII treatment, and thus identified as CS (Fig. 2A-D). A comparison of the elution profiles of GAG extracted from the day 16 with the day 18 tooth germs indicated that some modification have occurred (Fig. 1, compare A

Fig. 3. Cellulose thin-layer chromatography of the CHaseACII digestion products of GAG. The plate was developed by ascending chromatography for 24 h. The film was exposed at -90" C for 110 days. Lane A, the digestion product of GAG recovered from the 0.22M NaCl peak. Lane B, the digestion products of GAG recovered from the 0.44 M NaCl peak. Note an unidentified spot near the starting point. Lane C-D, the digestion products of total GAG isolated from the day 16+2 tooth germs, labeled for 6 h and 12 h respectively. Lane E-F, the digestion products of total GAG isolated from the day 18+2 tooth g e m , labeled for 6 h and 12 h respectively. The positions of the reference substances (standard disaccharides dDi-0.9, dDi-4S, and dDi-6S from Seikagaku Kogyo Co., and dDi-OHA prepared by CHdseABC digestion of HA) are indicated. They were run on either side of the thin-layer sheet and identified as dark spots under UV illumination. SP represents the starting point. The vertical arrow indicates the direction of mobility. (Compare the intensity of spots occumng within the same lane)

with B). On the [35S]chromatogramonly the last two peaks appeared (Fig. 1D). The [3H]-peaks at 0.15 M and 0.22 M of NaCI, which were shown to correspond to GP and HA respectively, were unsulfated and therefore unlabeled. The total GAG extracted from the tooth germs and the GAG recovered from the chromatographic peaks were further characterized by thin-layer chromatography. CHaseACII digestion of HA, C4S, and C6S produced ADi-OHA, ADi4S, and ADi-6S respectively as the end-products which could be separated as spots on the thin-layer sheet (Fig. 3, Table 1). The degradation products of the total GAG isolated from the whole tooth germ gave a strong spot of ADi-OHA, a weaker spot of ADi-4S, and a still weaker spot of ADi-6S on the fluorogram (Fig. 3C-F). This indicated that HA is the predominant type of GAG in the tooth germ and that C4S and C6S were the major and minor isomers of the CS respectively. The digestion product of GAG recovered from the 0.22 M NaCl peak was shown to be ADi-OHA (Fig. 3A), since it appeared as a single spot migrating at the same speed as the disaccharide of highest mobility by chromatographing the CHaseACII digestion products of total GAG extracted from intact tooth germs (Fig. 3C-F). The digestion products of GAG recovered from the 0.44M NaCl peak appeared as two spots corresponding to the reference standards ADi-4s and ADi6s (Fig. 3B). These chromatograms also revealed a small

238

Fig. 4A-I. Fractionation of total GAG by chromatography on DEAE-cellulose. Teeth were labeled with 25 pCi.ml-' ['HI glucosamine for 6 h. Peaks (1-4) correspond to GP, HA, HS, and CS respectively. A GAG isolated from dental epithelia, which were dissociated from ['H]-labelcd 18-day-old tooth germs by EDTA treatment. B GAG isolated from dental papillae, which were dissociated from ['HI-labeled 18day-old intact tooth germs by EDTA treatment. C GAG isolated from the dental epithelia, which were dissociated from ['HI-labeled 18-day-old tooth germs by trypsin treatment. D GAG isolated from trypsin-dissociated dental papillae, which were dissociated from [3H]-labeled 18-day-old tooth germs by trypsin treatmcnt. E GAG released in the trypsin incubation solution after dissociation of the [3H]-labeled 18-day-old tooth germs. F GAG isolated from dental epithelia, which were dissociated from the 18day-old tooth germs by trypsin treatment prior to labeling with ['HI glucosamine. G GAG isolated from the labeling medium of dental epithelia, which had previously been dissociated from the 18-day-old tooth germs by trypsin treatmcnt. H GAG isolated from dental papillae, which were dissociated from the 18day-old tooth germs prior to labeling with ['HI glucosamine. 1 GAG isolated from the labeling medium of the dental papillae, which had previously been dissociated from the 18-day-old tooth germs by trypsin treatment

Fraction number Table 1. The R,-valucs and the relative mobilities with respect to ADi-OS unsaturated disaccharides (Seikagaku) on the cellulose F thin-layer sheet (Merck). The mobilities were measured after about 24 h of developing in the 2 N ammonia solvent system of Mason, as cited by Adams and Muir [l]. ADi-OHA resulting from CHaseABC digestion of HA (from human umbilical cord) gave a broad spot with relative mobility greater than 1

Unsaturated disaccharide

R,-value

Relative mobility

ADi-OS ADi-4S ADi-6S

0.485 k0.034 0.319f0.033 0.191 &0.019

1.00 (by definition) 0.657 f0.023 0.394k 0.016

fraction of material closer to the starting point (Fig. 3B). This material is probably oligosaccharides resistant to CHaseACII degradation. GAG Extracted from Dental Epithelia and Dental Papillae after Labeling of Intact Teeth

Intact teeth were labeled with 25 pCi.ml- I [jH] glucosamine for 6 h prior to dissociation with EDTA or trypsin. The elution profiles of GAG extracted from EDTA- or trypsin-isolated dental epithelia - both of which are devoid

of a BM - are shown in Fig. 4 (A and C respectively). They show only a peak of [jH] HA and traces of HS. The chromatograms of GAG extracted from EDTA- or trypsin-isolated dental papillae are shown in Fig. 4 (B and D respectively). The trypsin-isolated dental papillae are devoid of the BM, the EDTA-isolated dental papillae remain covered by intact BM. In both cases four peaks corresponding to GP, HA, HS, and CS exist. The EDTAisolated dental papillae show higher amounts of CS and HS relative to total GAG than the trypsin-isolated dental papillae. No [jH] GAG were found in the EDTA incubation solution, while much [3H] GAG were found in the trypsin incubation solution. These [jH] GAG were extracted and analyzed. The elution profile was not very different from that of the GAG extracted from intact teeth (compare Fig. 4E with Fig. 1C). (The absence of GAG-degrading activity in the Difco trypsin used was verified). GAG Extracted after Labeling of Isolated Dental Epithelia and Dental Papillae

GAG were extracted from dental tissues as well as from the labeling media. 3H-labels were incorporated into the same types of GAG in the isolated dental constituents as in the dental epithelia and dental papillae of the intact teeth, but the amounts of individual types of GAG relative to

239 Table 2. Patterns of staining of the day 16 + 2 and day 18 + 2 molars by AB methods: AB-pH 2.5, AB-pH 1.0, ABpH 2.S-PAS, ABpH 5.7-0.2 M MgCI,. HA and sialomucins were stained by AB at pH 2.5. Sulfated GAG were stained by AB at pH 1.0. By the ABpH 2.5-PAS procedures, HA was stained blue and sulfated GAG were stained either red or bluish-purple. The symbols used to denote different colors of the AB-PAS staining are: B: Pure blue, B>R: Blue with a trace of red, BR: Blue mixed with red, with the blue predominating (including all blue and red combinations, except B > R and R > B), R > B: Red with a trace of blue, R: Pure red. At 0.2 M concentration of MgCl,, pH 5.7, AB stains mainly the total GAG. The symbols used to denote the strength of staining are: + + +very strong; + +strong; +moderate; weak ; -negative Tissue components

Method of staining

Stages of development Day 16+2

pH pH pH pH

Stratum reticulum

pH 2.5 pH 1.0 pH 2.5 - PAS pH 5.7 - 0.2 M MgCI,

Predentine

Basement mcmbrdne

Preodontoblastic layer

pH 2.5 pH 1.0 pH 2.5 - PAS pH 5.7 - 0.2 M MgC1, pH 2.5 pH 1.0 pH 2.5 - PAS pH 5.7 - 0.2 M MgCI, pH 2.5 pH 1.0 pH 2.5 - PAS pH 5.7 - 0.2 M MgCI, pH 2.5 pH 1.0 pH 2.5 - PAS pH 5.7 - 0.2 M MgCI,

Odontoblastic pH 2.5 layer pH 1.0 pH 2.5 - PAS pH 5.7 - 0.2 M MgCI, Lower dental papilla

Day 19

2.5 1.0 2.5 - PAS 5.7 - 0.2 M MgCI,

Outer dental epithelium

Inner dental epithelium

Day 18+2

pH 2.5 pH 1.0 pH 2.5 - PAS pH 5.7 - 0.2 M MgCI,

follows. The outer and inner dental epithelia were unstained (Fig. 5). The stratum reticulum contained HA and traces of sulfated GAG, of which the concentration decreased during odontoblastic differentiation. The BM and predentine were intensely stained. At the beginning of odontoblastic differentiation there were more sulfated GAG and HA in the early predentine (Fig. 5B, E, and H), but at later stages it appeared rich in both sulfated GAG and HA (Fig. 5C, F, and I). The BM was found to be rich in both sulfated GAG and HA before, during, and after the odontoblastic differentiation (Fig. 5 A-I). The preodontoblastic layer had a constant amount of HA (Fig. SD-I) and was rich in sulfated GAG before odontoblastic differentiation began (Fig. 5 A-C). The odontoblastic layer was practically unstained (Fig. SB-C, E-F, H, and I). The lower dental papilla was rich in GAG (both HA and sulfated GAG) but there was no obvious change during any stage of odontogenesis.

Discussion

++ f B>R

+

+t

+

B

B>R

+ +

+ ++ R +++ + + ++ ++ BR BR +++ +++ + + ++ f ++ + + -

*

f

+++

+++

BR

+++ + +++ B +++ + f -

f

BR

++ ++ ++ ++ B>R B ++ ++

+++ +++ B>R ++

total GAG were different (Fig. 4, compare F with C and H with D). More 3H-labels were incorporated into CS of isolated dental papillae than that of the dental papillae of intact teeth (Fig. 4, compare H with D). Histochemical Localization of GAG in the Tooth Germs

The observations from AB staining according to ABpH 2.5, AB-pH 1.0, ABpH 2.5-PAS, and AB-CEC methods are summarized in Table2. Interpreted according to the binding specificity of the dye, the main results are as

The composition and relative amounts of individual GAG in human, bovine, porcine, rabbit, and rat dental papillae, either from adult permanent teeth or from continuously growing teeth, have been analyzed by several authors [7, 10, 14, 18-23, 33, 351. The results were, however, quite different from one another: both HA and C4S have been isolated from human, bovine, rabbit, porcine, and rat dental papillae; HS has been isolated from bovine, rabbit, and rat dental papillae; and C6S was detected in human, bovine, porcine, and rat dental papillae; dermdtan sulfate (DS) was reported by some authors to be the major component of human dental papilla [lo, 22,411 and the minor component of bovine, rabbit, and porcine dental papillae [18, 20, 411. The presence of keratan sulfate (KS) was reported in human, rabbit, porcine, and rat dental papillae [lo, 14, 19-20, 221. The present results demonstrate that the embryonic mouse molars synthesize HA as a major GAG component as well as HS, C4S, and C6S as minor components. The change in the elution profile of DEAE-cellulose chromatography might indicate quantitative and/or qualitative modifications. Quantitative modifications may include changes in the percentage amounts of individual types of GAG and qualitative modifications may include changes in the charge density and/or heterogeneity of GAG. The histochemical evidence strongly supports the possibility that changing patterns of GAG occur during odontogenesis. The biochemical and histochemical analysis demonstrate further that in the intact tooth the dental epithelium contains mainly HA and traces of HS, while the dental papilla contains mainly HA, less HS and C4S, as well as traces of C6S. The analysis of GAG extracted from isolated dental epithelia and dental papillae provides us with some interesting information: it appears that a higher amount of 3Hlabels, relative to those incorporated into total GAG, is incorporated into the isolated dental epithelium than into the dental epithelium in situ. This apparent stimulation might be explained: (a) by the absence of epithelial-mesenchymal interaction-dependent control mechanisms of GAG metabolism, which have been suggested for the teeth [34] and for salivary gland [35], (b) by increased synthesis accompanying the BM reconstitution [36], (c) by a stimulating effect of depletion produced by trypsin. Such an effect has

Fig. 5. AB staining of day 16. day 18, and day 19 molars. The predentine and basement membrane are intensively stained. A Day 16+2 tooth germ stained by AB at pH 1.0. B Day 18+2 tooth germ stained by AB at pH 1.0. C Day 19 tooth germ stained by AB at pH 1.0. D Day 16+2 tooth germ stained by AB at pH 2.5. E Day 18+2 tooth germ stained by AB at pH 2.5. F Day 19 tooth germ stained by AB at pH 2.5. G Day 16+2 tooth germ stained by AB at pH 5.7 in the presence of 0.2 M MgCI,. H Day 18+2 tooth germ stained by AB at pH 5.7 in the presence of 0.2 M MgCI,. I Day 19 tooth germ stained by AB at pH 5.7 in the presencc of 0.2 M MgCI,. Outer dental epithelium (ODE); stratum reticulum (SR);inner dental epithelium (IDE); predentine ( P D ) ; bascrnent membrane ( B M ) ; preodontoblasts (PO); odontoblasts (0); dental papilla (DP). All figures 122 x

241

been documented for the chick embryonic cartilage [12]. The isolated dental papillae showed a 2-3 fold increase in the amount for CS relative to total GAG. The absence of epithelial-mesenchymal interactions and/or the trypsinproduced depletion might also explain these ‘stimulations’. The elution profiles of the GAG extracted from EDTAor trypsin- isolated dental papillae (after labeling of the intact teeth) are not identical. The relative amounts of CS and HS are higher in EDTA-isolated dental papillae. After trypsin treatment the BM is hydrolyzed, while after EDTA treatment the BM remains attached to the dental papillae [16]. The trypsin solution (after incubation of labeled teeth) contains HA, HS, and CS, which originate mainly, but probably not exclusively, from the BM. Thus, it is highly probable that HA, HS, and CS are integral components of the BM. Interaction between the dental BM and the preodontoblasts may trigger the terminal differentiation of odontoblasts [39, 451. A possible role of sulfated GAG and/or glycoproteins was suggested by the inhibition of odontoblastic differentiation by vitamin A, diazo-0x0-norleucine, and tunicamycin [45]. The fibrous lattice of the dental BM is a three dimensional structure resulting from interactions between collagens, glycoproteins, proteoglycans, and GAG. In our opinion, the dental BM should be considered as an integrated, changing system, interacting by a transmembranous mechanism with the cytoskeleton of adjacent cells [39]. In order to know the precise role of GAG (and other components) rather than speculative correlations, functional assays in addition to physico-chemical studies must be performed. Acknowledgements. Our thanks are due to Dr. Paul W. O’Connell

of the Upjohn C., Kalamazoo for the gift of heparan sulfate, Nourdine Amlaiky for the synthesis of n-butyl nitrite, Prof. Harold C. Slavkin and Dr. G. Rebel for useful discussions, Mrs. V. Karcher and M. Olive for help with the preparation of mouse teeth, and Marie Assmann for preparing graphical illustrations. The excellent technical assistance of Ms. A. Stlubli in the histochemical work is gratefully acknowledged. This work was supported by DGRST grant no. 81/901/070.

References

8. Cohn RH, Cassiman JJ, Bernfield MR (1976) Relationship of transformation, cell density, and growth control to the cellular distribution of newly synthesized glycosaminoglycan. J Cell Biol71: 28&294 9. Derby MA (1978) Analysis of glycosaminoglycans within the extracellular environments encountered by migrating neural crest cells. Develop Biol66: 321-336 10. Embery G (1976) Glycosaminoglycans of human dental pulp. J Biol Buccale 4 :229-236 11. Fisher LW, Schraer H (1980) The glycosaminoglycans of estrogen-induced medullary bone in Japanese quail. Arch Biochem Biophys 205 :396403 12. Fitton-Jackson S (1970) Environmental control of macromolecular synthesis in cartilage and bone: morphogenetic response to hyaluronidase. Proc R SOCLond (Biol) 175:405453 13. Gordon JR, Bernfield MR (1980) The basal lamina of the postnatal mammary epithelium contains glycosaminoglycans in a precise ultrastructural organization. Develop Biol74: 118-135 14. Jones IL, Leaver AG (1974) Glycosaminoglycans of human dentine. Calcified Tissue Res 16 :37-44 15. Karcher-Djuricic V, Osman M, Meyer JM, Stiiubli A, Ruch JV (1978) Basement membrane reconstitution and cytodifferentiation of odontoblasts in isochronal and heterochronal re-associations of enamel organs and pulps. J Biol Buccale 6:257-265 16. h o t H, Osman M, Ruch JV (1981) Immunofluorescent localization of collagens, fibronectin, and laminin during terminal differentiation of odontoblasts. Develop Biol82:371-381 17. Lev R, Spicer SS (1964) Specific staining of sulfate groups with Alcian blue at low pH. J Histochem Cytochem 12:309 18. Linde A (1 970) Glycosaminoglycans (mucopolysaccharides) of the procine dental pulp. Arch Oral Bioll5 :103S1046 19. Linde A (1972) Glycosaminoglycans of the rat incisor pulp. Biochim Biophys Acta 279 :446455 20. Linde A (1973) Glycosaminoglycans of the odontoblasts- predentine layer in dentinogenically active porcine teeth. Calcified Tissue Res 12: 281-301 21. Linde A (1973) Glycosaminoglycan turnover and synthesis in the rat incisor pulp. Scand J Dental Res 81 :145-154 22. Linde A (1973) A study of the dental pulp glycosaminoglycans from permanent human teeth and rat and rabbit incisors. Arch Oral Biol 18:49-59 23. Linde A (1973) Glycosaminoglycans of the dental pulp. A biochemical study. Scand J Dental Res 81 :177-201 24. Lowry OH, Rosebrough NJ, Farr AH, Randall RJ (1951) Protein measurement with the Fohn phenol reagent. J Biol Chem 193 :265-275 25. Manasek FJ (1975) The extracellular matrix: a dynamic component of the developing embryo. Current Topics Develop Biol

1. Adams ME, Muir H (1980) An automated version of the periodate-thiobarbituric acid assay for analysis of 4-43 unsaturated

10:35-120 26. Markwald RR, Fitzharris TP, Smith WNA (1975) Structural

uronic acids and its application to the assay of hyaluronic acid and chondroitin sulfates. Anal Biochem 109:426-431 2. Banerjee SD, Cohn RH, Bernfield MR (1977) Basal lamina of embryonic salivary epithelia. Production by the epithelium and role in maintaining lobular morphology. J Biol Chem

analysis of endocardia1 cytodifferentiation. Develop Biol

73 :44-63 3. Bernfield MR, Banerjee SD (1978) The basal lamina in epitheli-

al-mesenchymal morphogenetic interactions. In: Kefalides N A (ed) Biology and chemistry of basement membranes. Academic Press, New York, p 137-148 4.Bernfield MR, Cohn RH, Banerjee SD (1973) Glycosaminoglycans and epithelial organ formation. Am Zoo1 13: 1067-1083 5. Cifonelli JA, King J (1972) The distribution of 2-acetamido-2deoxy-D-glucose residues in mammalian heparins. Carbohydrate Res 21: 17S186 6. Cifonelli JA, King J (1973) Structural studies on heparins with unusually high N-acetylglucosamine contents. Biochim Biophys Acta 320: 331-340 7. Clark RO, Smith JG Jr, Davidson EA (1965) Hexosamine and acid glycosaminoglycan in human teeth. Biochim Biophys A d a 101:267-272

42: 160-180 27. Morris JE (1979) Steric exclusion of cells; a mechanism of GAG-induced cell aggregation. Exptl Cell Res 120: 141-153 28. Morris JE, Hopwood JJ, Dorfman A (1977) Biosynthesis of

glycosaminoglycans in Lhe devcloping rctina. Devclop Biol 58: 31 S327 29. Mowry RW (1960) Revised method producing improved col-

ouration of acidic polysaccharides with Alcian blue 8GX supplied currently. J Histochem Cytochem 8:323-324 30. Mowry RW (1963) The special value of methods that color both acidic and vicinal hydroxyl groups in the histochemical study of mucins. With revised directions for the colloidal iron stain, the use of Alcian blue G8X and their combinations with the periodic acid-Schiff reaction. Ann NY Acad Sci 106 :402-423 31. Noyes WA (1943) n-Butyl nitrite. In: Blatt AH (ed)Organic synthesis collective, vol 11. John Willy and Sons, New York, p 108 32. Ohya T, Kaneko Y (1970) Novel hyaluronidase from streptomyces. Biochim Biophys Acta 198: 607-609

242 33. Orlowski WA (1974) Analysis of collagen, glycoproteins and

42. Saxdn L, Ekblom P, Thesleff I (1980) Mechanisms of morpho-

acid mucopolysaccharides in the bovine and porcine dental pulp. Arch Oral Biol 19:255-258 34. Osman M, Ruch JV (1981) 3H-glucosamine and 3H-proline radioautography of embryonic mouse dental basement membrane. J Craniofacial Genet Develop Biol 1 :95-108 35. Osman M, Ruch JV (1981) Behavior of odontoblasts of basal lamina of trypsin or EDTA isolated mouse dental papillae in short-term culture. J Dental Res 60:1015-1027 36. Osman M, Ruch JV (1981) Reconstitution of the basement membrane from the inner and outer dental epithelia of trypsinisolated mouse molar enamel organs. J Biol Buccale 9: 129-139 37. Pearse AGE (1968) Histochemistry - Theoretical and applied, vol 1 (3rd ed). J. & A. Churchill Ltd., London, p 312, p 674 38. Quintarelli C, Vocaturo A, Roden L, Bellocci L, Vassallo LM (1978) Role of hyaluronic acid in the in vivo aggregation of cartilage proteoglycans. Connective Tissue Res 5: 237-248 39. Ruch JV, Lesot H, Karcher-Djuricic V. Meyer JM, Olive M (1982) Facts and hypotheses concerning the control of odontoblast differentiation. Differentiation 21 :7-12 40. Saito H, Yamagata T, Suzuki S (1968) Enzymatic methods for the determination of small quantities of isomeric chondroitin sulfates. J Biol Chem 243: 1536-1542 41. Sakamoto N, Okamoto H, Okuda K (1979) Qualitativc and quantitative analysis of bovine, rabbit and human dental pulp glycosaminoglycans. J Dental Res 58 :646-655

genetic cell interactions. In : Johnson MH (ed) Development in mammals, vol4. ElsevierINorth-Holland Biomedical Press, Amsterdam, p 161-202 43. Scott JE, Dorling J (1965) Differential staining of acid glycosaminoglycans (mucopolysaccharides) by Alcian blue in salt solutions. Histochemie 5 :221-233 44. Thesleff I (1978) Role of the basement membrane in odontoblast differentiation. J Biol Buccale 6:241-249 45. Thesleff I, Hurmerinta K (1981) Tissue interactions in tooth development. Differentiation 18: 75-88 46. Underhill C, Dorfman A (1978) The role of hyaluronic acid in intercellular adhesion of cultured mouse cells. Exptl Cell R e 117~155-165 47. Wassermann L, Ber A, Allalouf D (1977) Use of thin-layer chromatography in the separation of disaccharides resulting from digestion of chondroitin sulphates with chondroitinases. J Chromatography 136 :342-347 48. Yamagata T, Saito H, Habuchi 0, Suzuki S (1968) Purification and properties of bacterial chondroitinases and chondrosulfatases. J Biol Chem 243 :1523-1 535

Received August 1982 / Accepted in revised form November 1982