rlrchr oral Eioi. Vol. 37. So. 8, pp. 603-61 I. 1992 Printed in Greai Britain. All rights reserved
Copyright
c
CKPl3-9969 9: 53.00 + 0.00 1992 Pergamn Press Lrd
HISTOGENETIC ASPECTS OF THE COMPOSITION AND STRUCTURE OF HUMAN ECTOPIC ENAMEL, STUDIED BY SCANNING ELECTRON MICROSCOPY D. GASPER% Faculty of Medicine. Department of Stomatology, 61000 Ljubljana, Hrvatski trg 6, Slovenia (Accepred 8 .Ipril 1992)
Summary-This study was made on 12 enamel projections and I2 enamel pearls on human permanent molars to compare their structure and composition, which might elucidate possible histogenetic differences between these two forms of ectopic enamel. The enamel projections contained approx. 2 (weight)% less Ca and P than control occlusal and mid-coronal enamel. Their mineralization was less homogeneous and probably defective. Their enamel structure was markedly irregular in the prismatic as well as the aprismatic layer, but the composition and structure were very similar to those of the control enamel close to cervical border. The enamel of the pearls largely corresponded in composition and structure to the control occlusal and mid-coronal enamel. The enamel of large pearls also resembled coronal enamel in its compositional and structural variability in the occlusal-cervical direction. These findings suggest that amelogenesis in enamel projections is a continuation of amelogenesis in the cervical region. In contrast, amelogenesis in enamel pearls follows the same pattern as that in a dental crown, from the occlusal region to the cervical enamel border. Therefore. the enamel pearl may well be regarded as an attempt at new tooth formation. Key words: enamel projection, enamel pearl, amelogcnesis, cervical enamel, structure and composition of tooth enamel.
INTRODUCTION
The term ectopic enamel refers mainly to the occurrence of enamel on the tooth root. This phenomenon has clinical as well as wider biological -notably anthropological and hereditary-implications, in that the prevalence of the ectopic enamel varies with race and there are indications for a constitutional predisposition to it (Risnes, 1974a, b; Brabant and Kettelbrant, 1975; Bohne, Pouezat and Kerebel, 1989). Ectopic enamel includes both enamel projections of different length, and enamel pearls. An enamel projection is a protuberance of enamel on the surface of dentine (Risnes, 1974b; Takahashi, 1987), which contains all the layers of normal enamel, but shows an irregular structure, suggesting a degenerative form of enamel (Takahashi, 1987). They are either isolated from or connected with the crown enamel and are directed toward or even reach into the bifurcation; most are situated on the second permanent mandibular molar (Masters and Hoskins, 1964; Schriider, 1983; Sutalo, Ciglar and Bacic, 1989). Enamel projections probably arise from a local activity of ameloblasts, derived from Hertwig’s epithelial root sheath, which have remained adherent to the dentine surface (Bernick and Levy, 1968; Pindborg, 1970; Risnes, 1974a). According to Bower (1983), enamel in the bifurcation area may be formed by ameloblasts of the enamel organ that produced the crown enamel. The finding of the amelogenin transcripts within cells overlying the root surface may partially explain the occurrence of ectopic enamel; perhaps such cells continue to express the genetic
repertoire required for enamel biomineralization (Luo, Slavkin and Snead, 1991). Enamel pearls (pearls), on the other hand, are hemispherical or hemi-oval structures occurring most frequently on the approximal surfaces of the maxillary third molar (Risnes, 1974b; Brabant and Kettelbrant, 1975; Funaki, 1977). The great majority of enamel pearls have a dentine core, of a size proportional to that of the pearl (Moscow, 1971, Risnes, 1974b; GaSperSiE, 1985; Takahashi, 1987; Moscow and Canut, 1990). Very large enamel pearls also contain pulpal tissue (Cavanha, 1965). Other typical features include prominent interglobular dentine and a cone-shaped enamel-dentine junction, which is characteristic for dental cusps; many enamel pearls are partially (Takiguchi and Funaki, 1977; GaSperSiE, 1985; Kerebel er al., 1986), and some completely, buried in dentine (Cavanha, 1965; Kaugers, 1983). Their enamel structure is similar to that of a dental crown but shows localized defects (Takiguchi and Funaki, 1977; Kerebel et al., 1986; Bohne er al., 1989; Risnes, 1989). Because of its dentine core and the occasional presence of dental pulp, an enamel pearl resembles a supernumerary cusp or tooth (Malassez and Gallipe, 1908; Risnes, 1974b; Brabant and Kettelbrant, 1975), which may arise from a local bulging of the odontoblastic layer (Fleishmann, 1922). This could favour a prolonged contact between the epithelial root sheath and the developing dentine surface, thereby triggering the induction sequence leading to enamel production (Kollar, 1981). The aim of my study was to assess the histological and compositional differences between two types of
603
604
D.
GASPERG
ectopic enamel in relation to the different sites of origin and generative tissues.
MATERIALS
AND METHODS
The material comprised 20 extracted human permanent molars, collected at the Department of Stomatology. Faculty of Medicine, Ljubljana, Slovenia. Dates of and reasons for extraction were unkno\vn and earlier relevant history was not available. The teeth contained 12 enamel projections and 12 enamel pearls. The diameter of the pearls ranged from 0.4 to 6.1 mm; all had dentine cores and one also contained a dental pulp. One molar had two pearls. and one had two pearls and one enamel projection; three molars had one enamel projection and one enamel pear1 each, and the rest had either one enamel projection or one enamel pearl each. Four teeth bearing enamel projections were sectioned longitudinally and the rest transversely using a watercooled diamond wheel; all the teeth bearing pearls were sectioned transversely except one, which was sectioned longitudinally. In 14 cases, crown enamel of the same teeth served for comparison: six of these samples were of enamel extending from the cusp to the cervical enamel border, two were only cervical enamel. and six included mid-coronal and occlusal enamel of the carrier tooth. The samples were embedded in a two-component epoxy resin (Araldite M, Hy 956, Ciba-Geigy, Switzerland), polished with diamond paste (Struers, Koebenhavn, Denmark), rinsed under running tap water. cleaned ultrasonically (Bandelin Sonorex GT 120, Germany) in distilled water for a few minutes, dehydrated in a graded series of ethanol, dried with compressed warm air and coated with carbon. Samples so prepared were then analysed under a scanning electron microscope (Jeol JSM 35) with wave length-dispersive spectrometers, operated at 25 kV accelerating voltage, using currents of IO-* A (or IO-!’ A for scanning electron micrographs), with a tilt angle of 35’. A Ca,(PO,), crystal (Josef Stefan Institute, Slovenia), prepared in the same way as the samples. was used as a standard. In each sample, point-to-point microanalysis was done along four to six lines from the enamel surface to the dentine. From the values obtained the mean Ca and P concentrations were calculated. After microanalysis the samples were again polished, ultrasonically cleaned, dried with compressed warm air and etched for 20 s with 4.65 M phosphoric acid (Ivoclar, Vivadent Schaan. Liechtenstein). Immediately afterwards the etched sample surface was rinsed with air-water spray, then with distilled water. After dehydration in graded ethanol, the samples were dried with compressed. warm air and coated with carbon. Scanning electron micrographs were then obtained. The samples were further ground and prepared for repeated microanalysis and scanning electron micrography by the procedure described above. One enamel projection and one enamel pearl were fractured to obtain micrographs from the unetched fracture surface. These unetched fracture surfaces were prepared for scanning electron micrography by the same procedure as described above.
RESULTS
Enamel composition The results of microanalysis of the enamel are presented in Table I. The average Ca concentration in the control occlusal enamel was 32.2% (range: 37.4-38.8%). in the control mid-coronal enamel 39.1% (range: 38.9%-39.3%) and in the control cervical enamel 36.6% (range: 3-l&38.1%). The P concentration exhibited minor differences between the control enamels; it varied from 17.6 to 20.0%. the average concentration being 19.3Oo in the occlusal enamel and 18.8% in the mid-coronal and in the cervical enamel. The Ca!P ratio by weight varied between I.94 and 2.02 (i.e. 1.50-1.57 molar ratio) in the control occlusal enamel, between 1.99 and 2. I6 (1.55-1.67 molar ratio) in the control mid-coronal enamel, and between 1.86 and 2.05 (1.44-1.59 molar ratio) in the control cervical enamel. In the enamel projections the Ca concentration was similar to the control cervical enamel. It varied between 34.3 and 38.7%. the average being 36.5%. The P concentration was lower than in the control enamels and varied between 15.0 and 19. I%, the average being 17.5%. The Ca/P ratio by weight was between 1.96 and 2.29 (1.52-l .77 molar ratio). The pearls exhibited a similar enamel mineral content to the control occlusal and mid-coronal enamel. Ca concentrations varied between 36.9 and 39.l%, the average being the same as in the control occlusal enamel, and P concentrations varied between 17.8 and 19.5%, the average being I8.7”0. The Ca,P ratio by weight was between 1.92 and 2.15 (1.39-1.67 molar ratio). Enamel originating from different pearls of the same tooth (samples I3 and 14) shotsed only very little difference in mineral content. The difference in the mineral content between the cervical and the occlusal enamel of the large pearl (sample 12) was similar to that between the control cervical and occlusal/mid-coronal enamel. The occlusal enamel of the large pearl contained 37.8% Ca and 18.1%I P, the Ca:P ratio by weight being 2.09 (1.62 molar ratio), while its cervical enamel contained 35.5% Ca and 19.7% P, corresponding to a Ca/P ratio by weight of 1.80 (1.39 molar ratio). Samples having an enamel projection as well as an enamel pearl (samples 7, 8, 9 and 13) exhibited lovver Ca and P concentrations in the enamel projections (average Ca concentration: 36.4%; average P concentration: 17.5%) than in the enamel of the pearls (average Ca concentration: 38.3%; average P concentration: 15.7%).
The control enamel had characteristic structural attributes of dental enamel and is illustrated alongside the photomicrographs of the ectopic enamel in Figs l-25. Of the eight samples studied, five showed irregular prism arrangement in the enamel close to the cervical enamel border. In six cases, cervical enamel showed surface aprismatic enamel, its thickness ranging from IO to 30 pm. Striae of Retzius were observed in 10, Hunter-Schreger bands in 12 and cross-striation in six samples of control enamel. Compared with control occlusal or control midcoronal enamel, the structure of enamel projections (Figs I and 2) was largely irregular regarding the
*SD.
enamel
Values are means
Enamel pearl
Mid-coronal
Sample
Mid-coronal enamel Enamel projection Enamel pearl
Sample
Occlusal enamel Cervical enamel Enamel projcclion
Sample
Ca
37.9 + 0.4
39.3 + 0.5
Ca
35.8 f. 1.6 -
Ca
38.7 jr 0.5 36.6 + 0.6 36.6 + 1.5
--_-~-
Table I. Ca and P concentrations
6
I
I1
_I_
(weight
19.5 & 0.2
19.7 + 0.6
-____-P
17.6 k 0.4
P
20.0 + 0.2 19.7 _+ 0.2 17.3 + I.0
P
%) in control
P
Occlusal enamel Cervical enamel Pearl enamelocclusal Pearl enamelcervical
__~ P
17.6 + 0.2 IX.1 kO.1 19.7 * 1.4
37.8 rt 0.7 35.5 * 0.4
19.0 + 0.3
12 _ __
36.4 + 0.4 37.8 * 0.2
-
Ca
37.8 + 0.4 36.7 f 0.2 34.9 * 0.3
34.8 2 0.2
37.4 + 0.2
~______ Ca
19.1 + 0.4 19. I f 0.3
-
3x.7 + 0.5 38.0 L- 0.4
P
-
enamel,
______ Ca
mid-coronal
19.2 + 0.8 18.4 f 0.4 16.9 + 0.6
control
Ca
I
2
enamel,
38.8 k 1.0 36.2 _+ 0.4 36.8 + 0.7
Cd
occlusal
x
3
control
Mid-coronal enamel Enamel projection Enamel pearl I Enamel pearl 2
17.3 * 0.3 18.9 + 0.7
P
3x.x + 0.3 39.1 + 0.2
17.8 _t 1.6 38.3 * 3.0
36.3 + 3.2
19.3 + 2.5
Ca
39.1 + 3.6
18.0 f 0.1
P
-
I3
36.9 f 1.3
39.0 f 0.5 15.0 + 0.3 18.1 k0.l
-.
18.5 + 2.3
3g.9 + 0.3
Ca
34.3 f. 1.4 38.5 + 0.4
-
Ca
37.2 + 0.2 36.9 + 0.3
Ca
19.9 * 0.4 18.5 + 0.4
38.1 +O.i 37.9 f 0.4
Ca
projections
P
-
-
9
P
of the enamel
Ca
4
and in the enamels
__.._~
enamel
18.8 f 0.3 17.9+0.1 17.1 + 0.4
P
cervical
I4
IO
5
18.4 + 0.5
IX.5 & 0.4
-
P
19.2 k 0.7
lg.6 + 0.4
P
: a v1 r; FJ s 2
2
j:
8 3
;
r
m g 2.
19.2 + 0.3 IX.1 +0.2
P
pearls
-
and enamel
606
D.
GASPER%
shape of enamel prisms, so that sometimes it was difficult to ascertain the separate identity of prisms and that of the interprismatic region (Fig. 3). Similar irregularities were observed in the control enamel close to the cervical enamel border (Fig. 4). Figure 4 is a microphotograph of a longitudinal section through the tooth cervix; very probably a transverse section would reveal similarly shaped enamel prisms. The basic structural elements (prism/interprism) characteristic of the majority of tooth enamel (Fig. 5) were apparent only in small areas of enamel projections; and even here the prisms often showed minor defects (Fig. 6). All the projections had a surface aprismatic enamel, which could comprise from onehalf up to two-thirds of the enamel layer (Fig. 7); its thickness ranged from 20 to IOO~m. The intervals between the substriae in the aprismatic enamel of projections were more variable and occasionally smaller (0.5-4 pm) than in the control enamel (Figs 8 and 9). Cross-striation with a periodicity of 3-4 itm was evident in one enamel projection, HunterSchreger bands were present in a very small area in another but striae of Ret&s were not apparent in any of the enamel projections studied. The enamel pearls (Fig. 10) showed a predominantly normal enamel structure, similar to that displayed by control occlusal and control mid-coronal enamel (Figs 11 and 12). However, structural differences between a pearl and a projection were distinctly apparent (Figs 12 and 13) and were also obvious on the unetched fracture surfaces (Figs 14 and 15). In
the ectopic enamel, areas with indistinctly outlined prisms could be observed; these areas in pearls were smaller than in projections (Figs 16 and 17). Four enamel pearls had striae of Retzius, resembling those seen in control enamel (Figs 19 and 20). All the pearls showed Hunter-Schreger bands; in one-half of the samples they were similar or identical to those in the control enamel (Figs 21 and 22). A surface zone of aprismatic enamel, resembling that in the control enamel, was observed in IO pearls: it was IO-SO pm thick and had a regularly spaced striation with a periodicity of 3-5 jtm. Cross-striations ivith a periodicity of 3-4 jt m were observed in two enamel pearls. The cervical enamel of the large pearl exhibited structural irregularities. resembling those in control cervical enamel (Figs 23 and 24). The shape of the prisms became mcreasingly irregular totvards the cervical enamel border of the large enamel pearl (Fig. 25). DISCUSSION
The concentrations of Ca and P in the control enamel agree with the expected values for normal enamel as determined by chemical analyses (Brudevold and Soremark, 1967; Driessens, 1982; Schroder. 1982) and by microprobe analyses (Frank, Capitant and Goni, 1966; Frazier, 1967; Besic et al., 1970; Wei, 1970). The amount of P was approximately the same throughout the control enamel and the amount of Ca was approx. 2 (weight)% lower in the control cervical
Plate 1 Figs l-6 are from the same sample. Fig. 1. Low-power micrograph of transversely sectioned enamel projection (arrowhead). D, root dentine. x IO Fig. 2. Low-power micrograph of part of longitudinally sectioned tooth crown. D, dentine; S, enamel; o, occlusal; m, mid-coronal; c, cervical. x 10 Fig. 3. Higher-power micrograph of enamel projection: very irregular enamel prisms. x 2000 Fig. 4. Irregular prismatic structure of control cervical enamel (S), close to cervical enamel border. D, dentine; A, Araldite. x 2000 Fig. 5. Normal appearance of longitudinally sectioned enamel prisms in control occlusal enamel. x 2000 Fig. 6. Defects in longitudinally sectioned enamel prisms (*) of enamel projection.
x 2000
Figs 7-9 are from the same sample. Fig. 7. Thick surface layer of aprismatic enamel (*) in enamel projection. S, enamel; D, dentine; A, Araldite. x 500 Fig. 8. Higher magnification of aprismatic enamel in the enamel projection shown in Fig. 7 reveals irregularly spaced striation (arrows). x 3000 Fig. 9. Typical appearance of surface aprismatic enamel layer (*) in control enamel. Intervals between substriae are equal as or greater than in Fig. 8. x 500 Figs IO-13 are from the same sample Fig. 10. Low-power micrograph of transversely sectioned enamel projection (arrowhead) and enamel pearl (two arrowheads). D, root dentine. x 10 Fig. 11. Typical shape of transversely sectioned enamel prisms in control occlusal enamel. x 2000 Fig. 12. Enamel prisms in enamel pearl resemble those seen in control occlusal enamel. x 2000 Fig. 13. Very irregular enamel prisms in enamel projection. x 2000 Fig. 14. Micrograph of unetched enamel surface of transversely fractured enamel projection: simplified, nearly amorphous enamel structure. S. enamel; D, dentine. x 1000 Fig. 15. Micrograph of unetched enamel surface of fractured enamel pearl: typical prismatic enamel. x 1000
Ectopic enamel composition
Plate
I
and structure
Plate 2
609
Ectopic enamel composition and structure enamel than in the control occlusal and mid-coronal enamel. This Ca concentration gradient is in agreement with published data for mineral content. 1966b: Wetherell, Robinson and (Gwinnett. Hallsworth. 1974; Theuns er al., 1983). The low Ca;P ratios by weight (1 .S6 and 1.91) can be explained by the presence of Ca-deficient crystals, e.g. Mg whitlockite (Wiiltgens et al., 1981b; Driessens et al.. 1985). Compared to control cervical enamel, the mineralization in the enamel projections is lower still. The Ca content was 2.2 (u-eight)% and the P content 1.6 (weight)% lower than in control occlusal and midcoronal enamel. The high Ca/P ratio in some samples could indicate a higher content of CaCO, (Trautz, 1967) or Na- and CO,-containing apatite (Driessens, 1982). On the other hand, in the pearl enamel the Ca and P content, as well as the Ca/P ratio were very similar to the values obtained for the control occlusal and control mid-corona1 enamel. These findings confirm the conclusions of Takiguchi and Funaki (1977), Kerebel et al. (1986) and Bohne et al. (1989), that the enamel mineralization in the pearl is similar to that of the carrier tooth. Additionally, it was noted that large pearls had another feature in common with the tooth enamel, namelv variability of mineralization in an occlusal
Copyright
c
CKPl3-9969 9: 53.00 + 0.00 1992 Pergamn Press Lrd
HISTOGENETIC ASPECTS OF THE COMPOSITION AND STRUCTURE OF HUMAN ECTOPIC ENAMEL, STUDIED BY SCANNING ELECTRON MICROSCOPY D. GASPER% Faculty of Medicine. Department of Stomatology, 61000 Ljubljana, Hrvatski trg 6, Slovenia (Accepred 8 .Ipril 1992)
Summary-This study was made on 12 enamel projections and I2 enamel pearls on human permanent molars to compare their structure and composition, which might elucidate possible histogenetic differences between these two forms of ectopic enamel. The enamel projections contained approx. 2 (weight)% less Ca and P than control occlusal and mid-coronal enamel. Their mineralization was less homogeneous and probably defective. Their enamel structure was markedly irregular in the prismatic as well as the aprismatic layer, but the composition and structure were very similar to those of the control enamel close to cervical border. The enamel of the pearls largely corresponded in composition and structure to the control occlusal and mid-coronal enamel. The enamel of large pearls also resembled coronal enamel in its compositional and structural variability in the occlusal-cervical direction. These findings suggest that amelogenesis in enamel projections is a continuation of amelogenesis in the cervical region. In contrast, amelogenesis in enamel pearls follows the same pattern as that in a dental crown, from the occlusal region to the cervical enamel border. Therefore. the enamel pearl may well be regarded as an attempt at new tooth formation. Key words: enamel projection, enamel pearl, amelogcnesis, cervical enamel, structure and composition of tooth enamel.
INTRODUCTION
The term ectopic enamel refers mainly to the occurrence of enamel on the tooth root. This phenomenon has clinical as well as wider biological -notably anthropological and hereditary-implications, in that the prevalence of the ectopic enamel varies with race and there are indications for a constitutional predisposition to it (Risnes, 1974a, b; Brabant and Kettelbrant, 1975; Bohne, Pouezat and Kerebel, 1989). Ectopic enamel includes both enamel projections of different length, and enamel pearls. An enamel projection is a protuberance of enamel on the surface of dentine (Risnes, 1974b; Takahashi, 1987), which contains all the layers of normal enamel, but shows an irregular structure, suggesting a degenerative form of enamel (Takahashi, 1987). They are either isolated from or connected with the crown enamel and are directed toward or even reach into the bifurcation; most are situated on the second permanent mandibular molar (Masters and Hoskins, 1964; Schriider, 1983; Sutalo, Ciglar and Bacic, 1989). Enamel projections probably arise from a local activity of ameloblasts, derived from Hertwig’s epithelial root sheath, which have remained adherent to the dentine surface (Bernick and Levy, 1968; Pindborg, 1970; Risnes, 1974a). According to Bower (1983), enamel in the bifurcation area may be formed by ameloblasts of the enamel organ that produced the crown enamel. The finding of the amelogenin transcripts within cells overlying the root surface may partially explain the occurrence of ectopic enamel; perhaps such cells continue to express the genetic
repertoire required for enamel biomineralization (Luo, Slavkin and Snead, 1991). Enamel pearls (pearls), on the other hand, are hemispherical or hemi-oval structures occurring most frequently on the approximal surfaces of the maxillary third molar (Risnes, 1974b; Brabant and Kettelbrant, 1975; Funaki, 1977). The great majority of enamel pearls have a dentine core, of a size proportional to that of the pearl (Moscow, 1971, Risnes, 1974b; GaSperSiE, 1985; Takahashi, 1987; Moscow and Canut, 1990). Very large enamel pearls also contain pulpal tissue (Cavanha, 1965). Other typical features include prominent interglobular dentine and a cone-shaped enamel-dentine junction, which is characteristic for dental cusps; many enamel pearls are partially (Takiguchi and Funaki, 1977; GaSperSiE, 1985; Kerebel er al., 1986), and some completely, buried in dentine (Cavanha, 1965; Kaugers, 1983). Their enamel structure is similar to that of a dental crown but shows localized defects (Takiguchi and Funaki, 1977; Kerebel et al., 1986; Bohne er al., 1989; Risnes, 1989). Because of its dentine core and the occasional presence of dental pulp, an enamel pearl resembles a supernumerary cusp or tooth (Malassez and Gallipe, 1908; Risnes, 1974b; Brabant and Kettelbrant, 1975), which may arise from a local bulging of the odontoblastic layer (Fleishmann, 1922). This could favour a prolonged contact between the epithelial root sheath and the developing dentine surface, thereby triggering the induction sequence leading to enamel production (Kollar, 1981). The aim of my study was to assess the histological and compositional differences between two types of
603
604
D.
GASPERG
ectopic enamel in relation to the different sites of origin and generative tissues.
MATERIALS
AND METHODS
The material comprised 20 extracted human permanent molars, collected at the Department of Stomatology. Faculty of Medicine, Ljubljana, Slovenia. Dates of and reasons for extraction were unkno\vn and earlier relevant history was not available. The teeth contained 12 enamel projections and 12 enamel pearls. The diameter of the pearls ranged from 0.4 to 6.1 mm; all had dentine cores and one also contained a dental pulp. One molar had two pearls. and one had two pearls and one enamel projection; three molars had one enamel projection and one enamel pear1 each, and the rest had either one enamel projection or one enamel pearl each. Four teeth bearing enamel projections were sectioned longitudinally and the rest transversely using a watercooled diamond wheel; all the teeth bearing pearls were sectioned transversely except one, which was sectioned longitudinally. In 14 cases, crown enamel of the same teeth served for comparison: six of these samples were of enamel extending from the cusp to the cervical enamel border, two were only cervical enamel. and six included mid-coronal and occlusal enamel of the carrier tooth. The samples were embedded in a two-component epoxy resin (Araldite M, Hy 956, Ciba-Geigy, Switzerland), polished with diamond paste (Struers, Koebenhavn, Denmark), rinsed under running tap water. cleaned ultrasonically (Bandelin Sonorex GT 120, Germany) in distilled water for a few minutes, dehydrated in a graded series of ethanol, dried with compressed warm air and coated with carbon. Samples so prepared were then analysed under a scanning electron microscope (Jeol JSM 35) with wave length-dispersive spectrometers, operated at 25 kV accelerating voltage, using currents of IO-* A (or IO-!’ A for scanning electron micrographs), with a tilt angle of 35’. A Ca,(PO,), crystal (Josef Stefan Institute, Slovenia), prepared in the same way as the samples. was used as a standard. In each sample, point-to-point microanalysis was done along four to six lines from the enamel surface to the dentine. From the values obtained the mean Ca and P concentrations were calculated. After microanalysis the samples were again polished, ultrasonically cleaned, dried with compressed warm air and etched for 20 s with 4.65 M phosphoric acid (Ivoclar, Vivadent Schaan. Liechtenstein). Immediately afterwards the etched sample surface was rinsed with air-water spray, then with distilled water. After dehydration in graded ethanol, the samples were dried with compressed. warm air and coated with carbon. Scanning electron micrographs were then obtained. The samples were further ground and prepared for repeated microanalysis and scanning electron micrography by the procedure described above. One enamel projection and one enamel pearl were fractured to obtain micrographs from the unetched fracture surface. These unetched fracture surfaces were prepared for scanning electron micrography by the same procedure as described above.
RESULTS
Enamel composition The results of microanalysis of the enamel are presented in Table I. The average Ca concentration in the control occlusal enamel was 32.2% (range: 37.4-38.8%). in the control mid-coronal enamel 39.1% (range: 38.9%-39.3%) and in the control cervical enamel 36.6% (range: 3-l&38.1%). The P concentration exhibited minor differences between the control enamels; it varied from 17.6 to 20.0%. the average concentration being 19.3Oo in the occlusal enamel and 18.8% in the mid-coronal and in the cervical enamel. The Ca!P ratio by weight varied between I.94 and 2.02 (i.e. 1.50-1.57 molar ratio) in the control occlusal enamel, between 1.99 and 2. I6 (1.55-1.67 molar ratio) in the control mid-coronal enamel, and between 1.86 and 2.05 (1.44-1.59 molar ratio) in the control cervical enamel. In the enamel projections the Ca concentration was similar to the control cervical enamel. It varied between 34.3 and 38.7%. the average being 36.5%. The P concentration was lower than in the control enamels and varied between 15.0 and 19. I%, the average being 17.5%. The Ca/P ratio by weight was between 1.96 and 2.29 (1.52-l .77 molar ratio). The pearls exhibited a similar enamel mineral content to the control occlusal and mid-coronal enamel. Ca concentrations varied between 36.9 and 39.l%, the average being the same as in the control occlusal enamel, and P concentrations varied between 17.8 and 19.5%, the average being I8.7”0. The Ca,P ratio by weight was between 1.92 and 2.15 (1.39-1.67 molar ratio). Enamel originating from different pearls of the same tooth (samples I3 and 14) shotsed only very little difference in mineral content. The difference in the mineral content between the cervical and the occlusal enamel of the large pearl (sample 12) was similar to that between the control cervical and occlusal/mid-coronal enamel. The occlusal enamel of the large pearl contained 37.8% Ca and 18.1%I P, the Ca:P ratio by weight being 2.09 (1.62 molar ratio), while its cervical enamel contained 35.5% Ca and 19.7% P, corresponding to a Ca/P ratio by weight of 1.80 (1.39 molar ratio). Samples having an enamel projection as well as an enamel pearl (samples 7, 8, 9 and 13) exhibited lovver Ca and P concentrations in the enamel projections (average Ca concentration: 36.4%; average P concentration: 17.5%) than in the enamel of the pearls (average Ca concentration: 38.3%; average P concentration: 15.7%).
The control enamel had characteristic structural attributes of dental enamel and is illustrated alongside the photomicrographs of the ectopic enamel in Figs l-25. Of the eight samples studied, five showed irregular prism arrangement in the enamel close to the cervical enamel border. In six cases, cervical enamel showed surface aprismatic enamel, its thickness ranging from IO to 30 pm. Striae of Retzius were observed in 10, Hunter-Schreger bands in 12 and cross-striation in six samples of control enamel. Compared with control occlusal or control midcoronal enamel, the structure of enamel projections (Figs I and 2) was largely irregular regarding the
*SD.
enamel
Values are means
Enamel pearl
Mid-coronal
Sample
Mid-coronal enamel Enamel projection Enamel pearl
Sample
Occlusal enamel Cervical enamel Enamel projcclion
Sample
Ca
37.9 + 0.4
39.3 + 0.5
Ca
35.8 f. 1.6 -
Ca
38.7 jr 0.5 36.6 + 0.6 36.6 + 1.5
--_-~-
Table I. Ca and P concentrations
6
I
I1
_I_
(weight
19.5 & 0.2
19.7 + 0.6
-____-P
17.6 k 0.4
P
20.0 + 0.2 19.7 _+ 0.2 17.3 + I.0
P
%) in control
P
Occlusal enamel Cervical enamel Pearl enamelocclusal Pearl enamelcervical
__~ P
17.6 + 0.2 IX.1 kO.1 19.7 * 1.4
37.8 rt 0.7 35.5 * 0.4
19.0 + 0.3
12 _ __
36.4 + 0.4 37.8 * 0.2
-
Ca
37.8 + 0.4 36.7 f 0.2 34.9 * 0.3
34.8 2 0.2
37.4 + 0.2
~______ Ca
19.1 + 0.4 19. I f 0.3
-
3x.7 + 0.5 38.0 L- 0.4
P
-
enamel,
______ Ca
mid-coronal
19.2 + 0.8 18.4 f 0.4 16.9 + 0.6
control
Ca
I
2
enamel,
38.8 k 1.0 36.2 _+ 0.4 36.8 + 0.7
Cd
occlusal
x
3
control
Mid-coronal enamel Enamel projection Enamel pearl I Enamel pearl 2
17.3 * 0.3 18.9 + 0.7
P
3x.x + 0.3 39.1 + 0.2
17.8 _t 1.6 38.3 * 3.0
36.3 + 3.2
19.3 + 2.5
Ca
39.1 + 3.6
18.0 f 0.1
P
-
I3
36.9 f 1.3
39.0 f 0.5 15.0 + 0.3 18.1 k0.l
-.
18.5 + 2.3
3g.9 + 0.3
Ca
34.3 f. 1.4 38.5 + 0.4
-
Ca
37.2 + 0.2 36.9 + 0.3
Ca
19.9 * 0.4 18.5 + 0.4
38.1 +O.i 37.9 f 0.4
Ca
projections
P
-
-
9
P
of the enamel
Ca
4
and in the enamels
__.._~
enamel
18.8 f 0.3 17.9+0.1 17.1 + 0.4
P
cervical
I4
IO
5
18.4 + 0.5
IX.5 & 0.4
-
P
19.2 k 0.7
lg.6 + 0.4
P
: a v1 r; FJ s 2
2
j:
8 3
;
r
m g 2.
19.2 + 0.3 IX.1 +0.2
P
pearls
-
and enamel
606
D.
GASPER%
shape of enamel prisms, so that sometimes it was difficult to ascertain the separate identity of prisms and that of the interprismatic region (Fig. 3). Similar irregularities were observed in the control enamel close to the cervical enamel border (Fig. 4). Figure 4 is a microphotograph of a longitudinal section through the tooth cervix; very probably a transverse section would reveal similarly shaped enamel prisms. The basic structural elements (prism/interprism) characteristic of the majority of tooth enamel (Fig. 5) were apparent only in small areas of enamel projections; and even here the prisms often showed minor defects (Fig. 6). All the projections had a surface aprismatic enamel, which could comprise from onehalf up to two-thirds of the enamel layer (Fig. 7); its thickness ranged from 20 to IOO~m. The intervals between the substriae in the aprismatic enamel of projections were more variable and occasionally smaller (0.5-4 pm) than in the control enamel (Figs 8 and 9). Cross-striation with a periodicity of 3-4 itm was evident in one enamel projection, HunterSchreger bands were present in a very small area in another but striae of Ret&s were not apparent in any of the enamel projections studied. The enamel pearls (Fig. 10) showed a predominantly normal enamel structure, similar to that displayed by control occlusal and control mid-coronal enamel (Figs 11 and 12). However, structural differences between a pearl and a projection were distinctly apparent (Figs 12 and 13) and were also obvious on the unetched fracture surfaces (Figs 14 and 15). In
the ectopic enamel, areas with indistinctly outlined prisms could be observed; these areas in pearls were smaller than in projections (Figs 16 and 17). Four enamel pearls had striae of Retzius, resembling those seen in control enamel (Figs 19 and 20). All the pearls showed Hunter-Schreger bands; in one-half of the samples they were similar or identical to those in the control enamel (Figs 21 and 22). A surface zone of aprismatic enamel, resembling that in the control enamel, was observed in IO pearls: it was IO-SO pm thick and had a regularly spaced striation with a periodicity of 3-5 jtm. Cross-striations ivith a periodicity of 3-4 jt m were observed in two enamel pearls. The cervical enamel of the large pearl exhibited structural irregularities. resembling those in control cervical enamel (Figs 23 and 24). The shape of the prisms became mcreasingly irregular totvards the cervical enamel border of the large enamel pearl (Fig. 25). DISCUSSION
The concentrations of Ca and P in the control enamel agree with the expected values for normal enamel as determined by chemical analyses (Brudevold and Soremark, 1967; Driessens, 1982; Schroder. 1982) and by microprobe analyses (Frank, Capitant and Goni, 1966; Frazier, 1967; Besic et al., 1970; Wei, 1970). The amount of P was approximately the same throughout the control enamel and the amount of Ca was approx. 2 (weight)% lower in the control cervical
Plate 1 Figs l-6 are from the same sample. Fig. 1. Low-power micrograph of transversely sectioned enamel projection (arrowhead). D, root dentine. x IO Fig. 2. Low-power micrograph of part of longitudinally sectioned tooth crown. D, dentine; S, enamel; o, occlusal; m, mid-coronal; c, cervical. x 10 Fig. 3. Higher-power micrograph of enamel projection: very irregular enamel prisms. x 2000 Fig. 4. Irregular prismatic structure of control cervical enamel (S), close to cervical enamel border. D, dentine; A, Araldite. x 2000 Fig. 5. Normal appearance of longitudinally sectioned enamel prisms in control occlusal enamel. x 2000 Fig. 6. Defects in longitudinally sectioned enamel prisms (*) of enamel projection.
x 2000
Figs 7-9 are from the same sample. Fig. 7. Thick surface layer of aprismatic enamel (*) in enamel projection. S, enamel; D, dentine; A, Araldite. x 500 Fig. 8. Higher magnification of aprismatic enamel in the enamel projection shown in Fig. 7 reveals irregularly spaced striation (arrows). x 3000 Fig. 9. Typical appearance of surface aprismatic enamel layer (*) in control enamel. Intervals between substriae are equal as or greater than in Fig. 8. x 500 Figs IO-13 are from the same sample Fig. 10. Low-power micrograph of transversely sectioned enamel projection (arrowhead) and enamel pearl (two arrowheads). D, root dentine. x 10 Fig. 11. Typical shape of transversely sectioned enamel prisms in control occlusal enamel. x 2000 Fig. 12. Enamel prisms in enamel pearl resemble those seen in control occlusal enamel. x 2000 Fig. 13. Very irregular enamel prisms in enamel projection. x 2000 Fig. 14. Micrograph of unetched enamel surface of transversely fractured enamel projection: simplified, nearly amorphous enamel structure. S. enamel; D, dentine. x 1000 Fig. 15. Micrograph of unetched enamel surface of fractured enamel pearl: typical prismatic enamel. x 1000
Ectopic enamel composition
Plate
I
and structure
Plate 2
609
Ectopic enamel composition and structure enamel than in the control occlusal and mid-coronal enamel. This Ca concentration gradient is in agreement with published data for mineral content. 1966b: Wetherell, Robinson and (Gwinnett. Hallsworth. 1974; Theuns er al., 1983). The low Ca;P ratios by weight (1 .S6 and 1.91) can be explained by the presence of Ca-deficient crystals, e.g. Mg whitlockite (Wiiltgens et al., 1981b; Driessens et al.. 1985). Compared to control cervical enamel, the mineralization in the enamel projections is lower still. The Ca content was 2.2 (u-eight)% and the P content 1.6 (weight)% lower than in control occlusal and midcoronal enamel. The high Ca/P ratio in some samples could indicate a higher content of CaCO, (Trautz, 1967) or Na- and CO,-containing apatite (Driessens, 1982). On the other hand, in the pearl enamel the Ca and P content, as well as the Ca/P ratio were very similar to the values obtained for the control occlusal and control mid-corona1 enamel. These findings confirm the conclusions of Takiguchi and Funaki (1977), Kerebel et al. (1986) and Bohne et al. (1989), that the enamel mineralization in the pearl is similar to that of the carrier tooth. Additionally, it was noted that large pearls had another feature in common with the tooth enamel, namelv variability of mineralization in an occlusal
