Calcified Tissue Research
Calcif. Tiss. Res. 24, 191-197 (1977)
9 by Springer-Verlag 1977
Ultrastructural Study of Amelogenesis Imperfecta Bertrand Kerebel and G u y Daculsi Centre de Recherche, Facult~ de ChirurgieDentaire, Place Alexis Ricordeau, F-44 042 Nantes, France
Summary. A n ultrastructural study of teeth with amelogenesis imperfecta revelaed various aspects of microcavities in the enamel surface, which ranged from isolated imprints of ameloblasts corresponding to the mildest lesions at the end of amelogenesis, to pits caused by the death of 20 to 30 ameloblasts at the beginning of amelogenesis. Abnormalities in the shape of the prisms can be observed. Further, crystals are distributed randomly within a prism or at the junction of 2 contiguous prisms while intercrystaUine spaces are widened, indicating in various places the lack of a preferred orientation of the crystals. In amelogenesis imperfecta, two different crystalline periods are found: 1 of about 250 A, the other of about 5 0 0 / ~ and over. The fact that amorphous areas are found among the crystals of enamel m a y be related to different stages of crystallization. However, it was not possible to find any lattice defect. Key
words:
cification - -
Amelogenesis imperfecta - - HypocalHypoplasia - - Electron microscopy.
Introduction
Amelogenesis imperfecta represents a group of structural anomalies of enamel originating from some malfunction of the enamel organ. Two stages in the development of normal enamel (the formative stage and the maturation stage) are usually mentioned. Accordingly, there are recognized 2 basic types of amelogenesis imperfecta; (1) enamel hypoplasia, in which there is defective formation of matrix, and (2) enamel hypocalcification (hypomineralization), in which there is defective mineralization of the formed matrix (Shafer et al., 1974). Send offprint requests to B. Kerebel at the above address
There have been m a n y studies of amelogenesis imperfecta at the light microscope level, whereas electron microscopic studies are few (Kerebel and Ribay, 1971; Vahl, 1971; Sauk et al., 1972a, b; Kerebel and Daculsi, 1976a). Materials and Methods
Eighteen temporary and 14 permanent teeth with amelogenesis imperfecta were obtained from 3 members of the same family. According to the usual classificationthese teeth belong to the first type of amelogenesisimperfecta, that is to say, enamel hypoplasia. Only the caries-free teeth were kept. These teeth, of a marked yellowish colour and presenting hypoplasia, were fixed in 4% neutral HCHO. Selected portions of vestibular enamel surface and subsurface were prepared. Some of them were submitted to etching over 0.1 N HC1 for 0.5 min. The rest remained undecalcified.All portions were post-fixed in osmium tetroxide solution and embeddedin methacrylate. The embedding was started in vacuo for 1 h, and then continued for 72 h in order to ensure suitable penetration. The portions were sectioned with a diamond knife. The observations were made with a JEOL 100 B TEM operating at 100 kV. Cut portions of enamel surface and fractured portions of subsurface enamel were placed in sodium hypochlorite for 5 min to remove organic material, then thoroughly washed in distilledwater for 1 h, and finally put into an ultrasonic cleaner for 3 min. The specimens were then dehydrated in 100~ methanol, and coated twice in an electron-conductive mixture of carbon-gold to be examined with a JEOL JSM 2SEM. The diameters of 1300 crystals observed with transmission electron microscopy were measured; our measurements were performed on crystal groups having the same section plane, the section plane being identified with high resolution electron microscopy when at least 2 lattice planes intersect. The values of lattice spacings were calculated from an average of about 20 lattice planes. Interplanar angles ~ were calculated from: Cos
~ =
(hh' + kk' + 1/2hk' + 1/2h'k) a.2 + ll'c.2 x/(h 2 +'k 2 + hk)a.2 + 12C.2 v/(h'2 + kt2 + h'k')a .2 + 1'2c.2 a.2 = b.2 =
4 - -
3a2
C.2 =
I --
c2
192
B. Kerebel and G. Daculsi: Ultrastructural Study of Amelogenesis Imperfecta
a and c being parameters of the equivalent lattice of phosphocalcic hydroxyapatite and h, k, 1, h', k', 1' being Miller indices.
Results
With scanning electron microscopy, the e n a m e l surface either presents s o m e microcavities with s m o o t h a n d
h o m o g e n e o u s areas o v e r h a n g i n g in places, or shows strongly m a r k e d imprints of prisms (Fig. 1). I n the s m o o t h e n a m e l areas, shallow, well-delimited cavities, distributed r a n d o m l y , are f o u n d (Fig. 2). R o u n d e d pits with diameters of a b o u t 100 # m m a y also be seen in the e n a m e l surface, together with p o r o u s
Fig. 1. SEM image of enamel surface showing an hypoplastic area and several microcavities, x280 Fig. 2. SEM image of a smooth enamel area presenting "punched out" cavities, • 330 Fig. 3. SEM image showing rounded pits and porous areas, x65 Fig. 4. SEM image of the hypoplastic area visualized in Figure 1. A honeycomb structure corresponding to enamel prisms with dissociated crystals within can be observed, x 2800' Fig. 5. SEM image of hypoplastic areas filled with a granular disorganized material, • 1100 Fig. 6. gEM image of a globular material distributed randomly between the prisms, •
B. Kerebel and G. Daculsi: Ultrastructural Study of Amelogenesis Imperfecta areas (Fig. 3). In places, portions o f enamel surface appear to be fracturing away. The hypoplastic areas show honeycomb structure corresponding to the 5 /2m diameter enamel prisms, presenting a granular texture within (Fig. 4). These granules are presumably groups of dissociated crystal-
Fig. 7. SEM image of prisms presenting a grooved structure perpendicular to their long axis, overhanging an hypoplastic area. Interprismatic gaps widened. Several other prisms coalesce, • 3000 Fig. 8. SEM image of the prisms. The boundaries cannot be distinguished. A fan-shaped distribution of the crystals can be seen, x 6000 Fig. 9. TEM image of the altered morphology of the prisms. The interprismatic gaps are widened, though in some places artifacts are obvious, x4200 Fig. 10. TEM image of crystals distributed randomly within a prism and in its peripheral region, x 24,000
193
lites. In the enamel surface, some hypoplastic areas, with diameters of about 50 /tm, are filled with a granular material, probably constituted by a packing of disoriented prisms (Fig. 5). The fractured portions of enamel show a globular material distributed randomly between incompletely
194
B. Kerebel and G. Daculsi: Ultrastructural Study of Amelogenesis Imperfecta
formed "foliated" prisms (Fig. 6). Overhanging an hypoplastic area, prisms with a grooved structure perpen-
dicular to their long axis occur, the interprismatic gaps are widened, while some other prisms coalesce (Fig. 7).
13
1 .IO~A
2
3
4
5
6
7
8
9
10
11
12
13
14
Fig. 11. TEM image of crystals distributed randomly within a prism, while intercrystalline spacings are much widened, x 38,000 Fig. 12. TEM image of crystals of different shapes, some needle-shaped, the others platelike. Some crystals are strongly distorted, x 38,000 Fig. 13. Diagram of the distribution of 1300 crystals.
B. Kerebel and G. Daculsi: Ultrastructural Study of Amelogenesis Imperfecta
In places, the boundaries of the prisms cannot be distinguished because of the fan-shaped distribution of the crystals (Fig. 8). With transmission electron microscopy, a general view of the prisms reveals alterations in their morphology,
Fig. 14. TEM image of amorphous areas among crystals, x 75,000 Fig. 15. TEM image of crystal alterations after HC1 etching, x 150,000 Fig. 16. TEM image of crystal alterations after HC1 etching, x 300,000 Fig. 17. TEM image of an apatite crystal sectioned according to d(120), x 1,900,000 Fig. 18. TEM image of an apatite crystal sectioned according to d(121), • 1,900,000
195
and some widened interprismatic gaps, up to about 0.5 /an in places (Fig. 9). Often, crystals appear to be distributed randomly in the peripheral regions of the prisms (Fig. 10). Within the prism itself, crystals are disordered and intercrystalline spaces are much widened (Fig. l 1). The presence of two types of
196
B. Kerebeland G. Daculsi:UltrastructuralStudyof AmelogenesisImperfecta
crystals, different in size and shape, should be noted, some needle-shaped, the others plate-like, a few being strongly distorted (Fig. 12). The values of crystal diameters are given in a diagram (Fig. 13). Some amorphous aggregates, giving no typical electron diffraction diagram of apatite group, may sometimes be found among disordered crystals (Fig. 14). The HC1 etching reveals crystal defects, consisting of clear areas of different sizes, scattered throughout the crystals (Figs. 15-16). The high resolution study shows periodic lattice patterns related to several families of lattice planes. The interplanar spacings observed are 8.2 _A and 6.9 A; the interplanar angle is 44 ~ between the 8.2 ,~ and the 3.8 A planes, while the interplanar angle is 72 ~ between the 8.2 _Aand the 5.3 A planes.
Discussion
It has been stated that sound enamel surface presents several 6.5 am diameter pits, usually some 5 am deep (Boyde, 1967). The scanning electron microscopic study of our samples revealed various aspects of hypoplasia in the enamel surface. In the enamel surface presenting a normal macroscopic image, some isolated imprints of ameloblasts may be related to the mildest lesions at the end of amelogenesis, while the fact that pits with diameters of about 100 am are found is evidence of the untimely death of about 20 to 30 ameloblasts at the beginning of amelogenesis. These distinctions have not been pointed out in previous studies (Kerebel and Ribay, 1971; Vahl, I971; Sauk et al., 1972a, b; Kerebel and Dalcusi, 1976a). Porosities and pits observed in the enamel are an unlikely result of resorption occurring before tooth eruption, as suggested by Witkop and Rao (1971): they are quite different from resorption lacunae which can be observed in the enamel of embedded teeth. It appears to be hypoplasia, as already described by Boyde (1970). The fact that enamel areas presenting an homogeneous image overhang a honeycomb structured subsurface is indicative of a crumbling enamel, subjected to fractures. Some microcavities, with diameters of about 60 am, appear to be remnants of the honeycomb structure of prisms which have been submitted to salivary and traumatic attacks (those similar to trauma-induced microcavities described by Poole and Silverstone, 1969). Observation with electron microscopy is the only means of emphasizing alterations in the shape of the prisms. The grooved structure of the prisms, perpendicular to their long axis, observed with scanning
electron microscopy, may be related to the segmentation of prisms observed at the light microscope level, and may be indicative of an hypomineralization. The coalescence of prisms in amelogenesis imperfecta has not been described before. The globular material associated with enamel prisms is presumably due to defective formation of enamel. It should be noted that the usual classification of amelogenesis imperfecta, though convenient, remains too stereotyped, being based upon early ideas about amelogenesis, according to which the formation stage and the maturation stage were distinctly separated. In fact, it is quite clear that the two stages closely overlap and that hypoplasia may occur together with various degrees of hypomineralization, as follows from our observations above. The fact that crystals are disordered has been reported recently (Kerebel and Daculsi, 1976a). The orientation of the crystals within 2 contiguous prisms is quite different from what may be found in sound enamel prisms: in amelogenesis imperfecta, there is not always a preferred orientation of the crystals. It seems that there are different stages and different states of crystallization, as revealed by the amorphous areas found among the enamel crystals. As regards the widened interprismatic gaps observed by transmission electron microscopy, it must be kept in mind that sectioning and methacrylate may produce artifacts (Fig. 9). However, our measurements of interprismatic gaps in normal mature human enamel embedded in methacrylate give an average of 0.1 am. On scanning electron microscope images, too, these widened interprismatic spaces have been observed. Measurements of lengths are not significant, since the orientation within the section is not always suitable, whereas the diameters of the crystals remain unchanged whatever the orientation within the section plane may be. That is why only the measurements of diameters have been recorded. In amelogenesis imperfecta, 2 different crystal periods are found, one of 250 /k, the other of about 500/~ and over, the "average diameter" of the crystallites being 370 .A for sound enamel (Jongebloed et al., 1975). The effects of "beam damage" with high resolution microscopy of sound enamel have often been referred to (Hirai and Fearnhead, 1972; Langdon et al., 1973). The damaged areas appear as electrolucent spots occurring progressively in the course of the observation and due to the electron beam. In amelogenesis imperfecta, on the contrary, this phenomena occurs immediately when the observation begins. On the other hand, after HC1 etching, these electrolucent spots increase in number and size, which suggests that they existed before the electron bombardment: they might be indicative of fragile areas within the crystals. Likewise,
B. Kerebel and G. Daculsi: Ultrastructural Study of Amelogenesis Imperfecta it has been shown that the central holes in the crystallites of acid-treated sound h u m a n enamel are not due to b e a m - d a m a g e (Jongebloed et al., 1975). A s regards the measurements o f lattice planes, all our values have been corrected on the basis o f interplanar spacing d(100). The interplanar spacings recorded, 3 . 8 / ~ , 8.2 A , 5.3 A , 6.9 .A, correspond respectively to plane families (111), (100), (011), (001), or their equivalents. Since several lattice planes were visualized on electron micrographs (Figs. 17-18), we m a y assert that the orientation within the section is d(120) as regards Figure 17, and d(121) as regards Figure 18. However, there was no evidence o f any lattice defects of the crystals in the samples o f amelogenesis imperfecta submitted to high resolution microscopy, whereas these lattice defects are always found in fluorosis (Kerebel and Daculsi, 1976b).
Acknowledgements. We wish to thank Mrs. Cottrel-Gengoux for her technical assistance. This investigation was supported both by C.N.R.S. Research Grant No. 031371 and by INSEKM Research Grant No. 7710074.
References Boyde, A.: The development of enamel structure. Proc. Royal Soc. Med. 60, 923-928 (1967) Boyde, A.: The surface of the enamel in human hypoplastic teeth. Archs Oral Biol. 15, 897-898 (1970) Hirai, G., Fearnhead, R.W.: Lattice defects in biological and synthetic apatites. Proc. of the 6th Conference on X-ray Optics and Microanalysis, pp. 863-872, Univ. of Tokyo Press 1972 Jongebloed, W.L., Molenaar, I., Arends, J.: Morphology and size-
197
distribution of sound and acid-treated enamel crystallites. Calcif. Tiss. Res. 19, 109-123 (1975) Kerebel, B., Daculsi, G.: Amelogenesis Imperfecta; &ude structurale, ultrastructurale et cristallographique. J. Biol. Buc. 4, 4360 (1976a) Kerebet, B., Daeulsi, G.: Ultrastructural and cristallographic study of some biological apatites. 54th general session International Association for Dental Research, Marburg, Germ. Fed. Rep., March 12 and 13, 1976b Kerebel, B., Ribay, B.: Amtlogtn~se imparfaite; 6tude au microscope photonique et au microscope ~lectronique ~t balayage. Actualitts Odonto-Stomatologiques 93, 37-56 (1971) Langdon, D., Dykes, E., Fearnhead, R.W.: Defects, diffusion and dissolution in biological and synthetic apatite. Colloques internationaux CNRS 230, Physicochimie et cristallographie des apatites d'intbr& biologique, 381-388 (1973) Poole, D.F.G., Silverstone, L.M.: Observations with the scanning electron microscope on trauma-induced microcavities in human enamel. Archs Oral Biol. 14, !323-1329 (1969) Sank, J.J., Cotton, W.R., Lyon, H.W., Witkop, C.J., Jr.: Electron optic analysis of hypomineralized Amelogenesis Imperfecta in man. Archs. oral Biol. 17, 771-779 (1972a) Sank, J.J., Lyon, H.W., Witkop, C.J., Jr.: Electron optic microanalysis of two gene products in enamel of female heterozygous for X-linked hypomaturation Amelogenesis Imperfecta. Amer. J. Hum. Gen. 24, 267-276 (1972b) Shafer, W.G., Hine, M.K., Levy, B.M.: A textbook of oral pathology. Philadelphia: Saunders, 1974 Vahl, J.: Gesunder und pathologish ver~inderterZahnschmelz. Eine mikromorphologische und biokristaUographische Strukturanalyse. (Sammlung Meusser) Leipzig: J. Ambrosius Barth 1971 Witkop, C.J., Rao, S.: Inherited defects in tooth structure. In: Bergsma, D., Birth defects, original article series. Part XI, Orofacial Structures, vol. 7, 153-184. New York: Williams and Wiikins Co. 1971
Received March 1 / Accepted November 22, 1976
Calcif. Tiss. Res. 24, 191-197 (1977)
9 by Springer-Verlag 1977
Ultrastructural Study of Amelogenesis Imperfecta Bertrand Kerebel and G u y Daculsi Centre de Recherche, Facult~ de ChirurgieDentaire, Place Alexis Ricordeau, F-44 042 Nantes, France
Summary. A n ultrastructural study of teeth with amelogenesis imperfecta revelaed various aspects of microcavities in the enamel surface, which ranged from isolated imprints of ameloblasts corresponding to the mildest lesions at the end of amelogenesis, to pits caused by the death of 20 to 30 ameloblasts at the beginning of amelogenesis. Abnormalities in the shape of the prisms can be observed. Further, crystals are distributed randomly within a prism or at the junction of 2 contiguous prisms while intercrystaUine spaces are widened, indicating in various places the lack of a preferred orientation of the crystals. In amelogenesis imperfecta, two different crystalline periods are found: 1 of about 250 A, the other of about 5 0 0 / ~ and over. The fact that amorphous areas are found among the crystals of enamel m a y be related to different stages of crystallization. However, it was not possible to find any lattice defect. Key
words:
cification - -
Amelogenesis imperfecta - - HypocalHypoplasia - - Electron microscopy.
Introduction
Amelogenesis imperfecta represents a group of structural anomalies of enamel originating from some malfunction of the enamel organ. Two stages in the development of normal enamel (the formative stage and the maturation stage) are usually mentioned. Accordingly, there are recognized 2 basic types of amelogenesis imperfecta; (1) enamel hypoplasia, in which there is defective formation of matrix, and (2) enamel hypocalcification (hypomineralization), in which there is defective mineralization of the formed matrix (Shafer et al., 1974). Send offprint requests to B. Kerebel at the above address
There have been m a n y studies of amelogenesis imperfecta at the light microscope level, whereas electron microscopic studies are few (Kerebel and Ribay, 1971; Vahl, 1971; Sauk et al., 1972a, b; Kerebel and Daculsi, 1976a). Materials and Methods
Eighteen temporary and 14 permanent teeth with amelogenesis imperfecta were obtained from 3 members of the same family. According to the usual classificationthese teeth belong to the first type of amelogenesisimperfecta, that is to say, enamel hypoplasia. Only the caries-free teeth were kept. These teeth, of a marked yellowish colour and presenting hypoplasia, were fixed in 4% neutral HCHO. Selected portions of vestibular enamel surface and subsurface were prepared. Some of them were submitted to etching over 0.1 N HC1 for 0.5 min. The rest remained undecalcified.All portions were post-fixed in osmium tetroxide solution and embeddedin methacrylate. The embedding was started in vacuo for 1 h, and then continued for 72 h in order to ensure suitable penetration. The portions were sectioned with a diamond knife. The observations were made with a JEOL 100 B TEM operating at 100 kV. Cut portions of enamel surface and fractured portions of subsurface enamel were placed in sodium hypochlorite for 5 min to remove organic material, then thoroughly washed in distilledwater for 1 h, and finally put into an ultrasonic cleaner for 3 min. The specimens were then dehydrated in 100~ methanol, and coated twice in an electron-conductive mixture of carbon-gold to be examined with a JEOL JSM 2SEM. The diameters of 1300 crystals observed with transmission electron microscopy were measured; our measurements were performed on crystal groups having the same section plane, the section plane being identified with high resolution electron microscopy when at least 2 lattice planes intersect. The values of lattice spacings were calculated from an average of about 20 lattice planes. Interplanar angles ~ were calculated from: Cos
~ =
(hh' + kk' + 1/2hk' + 1/2h'k) a.2 + ll'c.2 x/(h 2 +'k 2 + hk)a.2 + 12C.2 v/(h'2 + kt2 + h'k')a .2 + 1'2c.2 a.2 = b.2 =
4 - -
3a2
C.2 =
I --
c2
192
B. Kerebel and G. Daculsi: Ultrastructural Study of Amelogenesis Imperfecta
a and c being parameters of the equivalent lattice of phosphocalcic hydroxyapatite and h, k, 1, h', k', 1' being Miller indices.
Results
With scanning electron microscopy, the e n a m e l surface either presents s o m e microcavities with s m o o t h a n d
h o m o g e n e o u s areas o v e r h a n g i n g in places, or shows strongly m a r k e d imprints of prisms (Fig. 1). I n the s m o o t h e n a m e l areas, shallow, well-delimited cavities, distributed r a n d o m l y , are f o u n d (Fig. 2). R o u n d e d pits with diameters of a b o u t 100 # m m a y also be seen in the e n a m e l surface, together with p o r o u s
Fig. 1. SEM image of enamel surface showing an hypoplastic area and several microcavities, x280 Fig. 2. SEM image of a smooth enamel area presenting "punched out" cavities, • 330 Fig. 3. SEM image showing rounded pits and porous areas, x65 Fig. 4. SEM image of the hypoplastic area visualized in Figure 1. A honeycomb structure corresponding to enamel prisms with dissociated crystals within can be observed, x 2800' Fig. 5. SEM image of hypoplastic areas filled with a granular disorganized material, • 1100 Fig. 6. gEM image of a globular material distributed randomly between the prisms, •
B. Kerebel and G. Daculsi: Ultrastructural Study of Amelogenesis Imperfecta areas (Fig. 3). In places, portions o f enamel surface appear to be fracturing away. The hypoplastic areas show honeycomb structure corresponding to the 5 /2m diameter enamel prisms, presenting a granular texture within (Fig. 4). These granules are presumably groups of dissociated crystal-
Fig. 7. SEM image of prisms presenting a grooved structure perpendicular to their long axis, overhanging an hypoplastic area. Interprismatic gaps widened. Several other prisms coalesce, • 3000 Fig. 8. SEM image of the prisms. The boundaries cannot be distinguished. A fan-shaped distribution of the crystals can be seen, x 6000 Fig. 9. TEM image of the altered morphology of the prisms. The interprismatic gaps are widened, though in some places artifacts are obvious, x4200 Fig. 10. TEM image of crystals distributed randomly within a prism and in its peripheral region, x 24,000
193
lites. In the enamel surface, some hypoplastic areas, with diameters of about 50 /tm, are filled with a granular material, probably constituted by a packing of disoriented prisms (Fig. 5). The fractured portions of enamel show a globular material distributed randomly between incompletely
194
B. Kerebel and G. Daculsi: Ultrastructural Study of Amelogenesis Imperfecta
formed "foliated" prisms (Fig. 6). Overhanging an hypoplastic area, prisms with a grooved structure perpen-
dicular to their long axis occur, the interprismatic gaps are widened, while some other prisms coalesce (Fig. 7).
13
1 .IO~A
2
3
4
5
6
7
8
9
10
11
12
13
14
Fig. 11. TEM image of crystals distributed randomly within a prism, while intercrystalline spacings are much widened, x 38,000 Fig. 12. TEM image of crystals of different shapes, some needle-shaped, the others platelike. Some crystals are strongly distorted, x 38,000 Fig. 13. Diagram of the distribution of 1300 crystals.
B. Kerebel and G. Daculsi: Ultrastructural Study of Amelogenesis Imperfecta
In places, the boundaries of the prisms cannot be distinguished because of the fan-shaped distribution of the crystals (Fig. 8). With transmission electron microscopy, a general view of the prisms reveals alterations in their morphology,
Fig. 14. TEM image of amorphous areas among crystals, x 75,000 Fig. 15. TEM image of crystal alterations after HC1 etching, x 150,000 Fig. 16. TEM image of crystal alterations after HC1 etching, x 300,000 Fig. 17. TEM image of an apatite crystal sectioned according to d(120), x 1,900,000 Fig. 18. TEM image of an apatite crystal sectioned according to d(121), • 1,900,000
195
and some widened interprismatic gaps, up to about 0.5 /an in places (Fig. 9). Often, crystals appear to be distributed randomly in the peripheral regions of the prisms (Fig. 10). Within the prism itself, crystals are disordered and intercrystalline spaces are much widened (Fig. l 1). The presence of two types of
196
B. Kerebeland G. Daculsi:UltrastructuralStudyof AmelogenesisImperfecta
crystals, different in size and shape, should be noted, some needle-shaped, the others plate-like, a few being strongly distorted (Fig. 12). The values of crystal diameters are given in a diagram (Fig. 13). Some amorphous aggregates, giving no typical electron diffraction diagram of apatite group, may sometimes be found among disordered crystals (Fig. 14). The HC1 etching reveals crystal defects, consisting of clear areas of different sizes, scattered throughout the crystals (Figs. 15-16). The high resolution study shows periodic lattice patterns related to several families of lattice planes. The interplanar spacings observed are 8.2 _A and 6.9 A; the interplanar angle is 44 ~ between the 8.2 ,~ and the 3.8 A planes, while the interplanar angle is 72 ~ between the 8.2 _Aand the 5.3 A planes.
Discussion
It has been stated that sound enamel surface presents several 6.5 am diameter pits, usually some 5 am deep (Boyde, 1967). The scanning electron microscopic study of our samples revealed various aspects of hypoplasia in the enamel surface. In the enamel surface presenting a normal macroscopic image, some isolated imprints of ameloblasts may be related to the mildest lesions at the end of amelogenesis, while the fact that pits with diameters of about 100 am are found is evidence of the untimely death of about 20 to 30 ameloblasts at the beginning of amelogenesis. These distinctions have not been pointed out in previous studies (Kerebel and Ribay, 1971; Vahl, I971; Sauk et al., 1972a, b; Kerebel and Dalcusi, 1976a). Porosities and pits observed in the enamel are an unlikely result of resorption occurring before tooth eruption, as suggested by Witkop and Rao (1971): they are quite different from resorption lacunae which can be observed in the enamel of embedded teeth. It appears to be hypoplasia, as already described by Boyde (1970). The fact that enamel areas presenting an homogeneous image overhang a honeycomb structured subsurface is indicative of a crumbling enamel, subjected to fractures. Some microcavities, with diameters of about 60 am, appear to be remnants of the honeycomb structure of prisms which have been submitted to salivary and traumatic attacks (those similar to trauma-induced microcavities described by Poole and Silverstone, 1969). Observation with electron microscopy is the only means of emphasizing alterations in the shape of the prisms. The grooved structure of the prisms, perpendicular to their long axis, observed with scanning
electron microscopy, may be related to the segmentation of prisms observed at the light microscope level, and may be indicative of an hypomineralization. The coalescence of prisms in amelogenesis imperfecta has not been described before. The globular material associated with enamel prisms is presumably due to defective formation of enamel. It should be noted that the usual classification of amelogenesis imperfecta, though convenient, remains too stereotyped, being based upon early ideas about amelogenesis, according to which the formation stage and the maturation stage were distinctly separated. In fact, it is quite clear that the two stages closely overlap and that hypoplasia may occur together with various degrees of hypomineralization, as follows from our observations above. The fact that crystals are disordered has been reported recently (Kerebel and Daculsi, 1976a). The orientation of the crystals within 2 contiguous prisms is quite different from what may be found in sound enamel prisms: in amelogenesis imperfecta, there is not always a preferred orientation of the crystals. It seems that there are different stages and different states of crystallization, as revealed by the amorphous areas found among the enamel crystals. As regards the widened interprismatic gaps observed by transmission electron microscopy, it must be kept in mind that sectioning and methacrylate may produce artifacts (Fig. 9). However, our measurements of interprismatic gaps in normal mature human enamel embedded in methacrylate give an average of 0.1 am. On scanning electron microscope images, too, these widened interprismatic spaces have been observed. Measurements of lengths are not significant, since the orientation within the section is not always suitable, whereas the diameters of the crystals remain unchanged whatever the orientation within the section plane may be. That is why only the measurements of diameters have been recorded. In amelogenesis imperfecta, 2 different crystal periods are found, one of 250 /k, the other of about 500/~ and over, the "average diameter" of the crystallites being 370 .A for sound enamel (Jongebloed et al., 1975). The effects of "beam damage" with high resolution microscopy of sound enamel have often been referred to (Hirai and Fearnhead, 1972; Langdon et al., 1973). The damaged areas appear as electrolucent spots occurring progressively in the course of the observation and due to the electron beam. In amelogenesis imperfecta, on the contrary, this phenomena occurs immediately when the observation begins. On the other hand, after HC1 etching, these electrolucent spots increase in number and size, which suggests that they existed before the electron bombardment: they might be indicative of fragile areas within the crystals. Likewise,
B. Kerebel and G. Daculsi: Ultrastructural Study of Amelogenesis Imperfecta it has been shown that the central holes in the crystallites of acid-treated sound h u m a n enamel are not due to b e a m - d a m a g e (Jongebloed et al., 1975). A s regards the measurements o f lattice planes, all our values have been corrected on the basis o f interplanar spacing d(100). The interplanar spacings recorded, 3 . 8 / ~ , 8.2 A , 5.3 A , 6.9 .A, correspond respectively to plane families (111), (100), (011), (001), or their equivalents. Since several lattice planes were visualized on electron micrographs (Figs. 17-18), we m a y assert that the orientation within the section is d(120) as regards Figure 17, and d(121) as regards Figure 18. However, there was no evidence o f any lattice defects of the crystals in the samples o f amelogenesis imperfecta submitted to high resolution microscopy, whereas these lattice defects are always found in fluorosis (Kerebel and Daculsi, 1976b).
Acknowledgements. We wish to thank Mrs. Cottrel-Gengoux for her technical assistance. This investigation was supported both by C.N.R.S. Research Grant No. 031371 and by INSEKM Research Grant No. 7710074.
References Boyde, A.: The development of enamel structure. Proc. Royal Soc. Med. 60, 923-928 (1967) Boyde, A.: The surface of the enamel in human hypoplastic teeth. Archs Oral Biol. 15, 897-898 (1970) Hirai, G., Fearnhead, R.W.: Lattice defects in biological and synthetic apatites. Proc. of the 6th Conference on X-ray Optics and Microanalysis, pp. 863-872, Univ. of Tokyo Press 1972 Jongebloed, W.L., Molenaar, I., Arends, J.: Morphology and size-
197
distribution of sound and acid-treated enamel crystallites. Calcif. Tiss. Res. 19, 109-123 (1975) Kerebel, B., Daculsi, G.: Amelogenesis Imperfecta; &ude structurale, ultrastructurale et cristallographique. J. Biol. Buc. 4, 4360 (1976a) Kerebet, B., Daeulsi, G.: Ultrastructural and cristallographic study of some biological apatites. 54th general session International Association for Dental Research, Marburg, Germ. Fed. Rep., March 12 and 13, 1976b Kerebel, B., Ribay, B.: Amtlogtn~se imparfaite; 6tude au microscope photonique et au microscope ~lectronique ~t balayage. Actualitts Odonto-Stomatologiques 93, 37-56 (1971) Langdon, D., Dykes, E., Fearnhead, R.W.: Defects, diffusion and dissolution in biological and synthetic apatite. Colloques internationaux CNRS 230, Physicochimie et cristallographie des apatites d'intbr& biologique, 381-388 (1973) Poole, D.F.G., Silverstone, L.M.: Observations with the scanning electron microscope on trauma-induced microcavities in human enamel. Archs Oral Biol. 14, !323-1329 (1969) Sank, J.J., Cotton, W.R., Lyon, H.W., Witkop, C.J., Jr.: Electron optic analysis of hypomineralized Amelogenesis Imperfecta in man. Archs. oral Biol. 17, 771-779 (1972a) Sank, J.J., Lyon, H.W., Witkop, C.J., Jr.: Electron optic microanalysis of two gene products in enamel of female heterozygous for X-linked hypomaturation Amelogenesis Imperfecta. Amer. J. Hum. Gen. 24, 267-276 (1972b) Shafer, W.G., Hine, M.K., Levy, B.M.: A textbook of oral pathology. Philadelphia: Saunders, 1974 Vahl, J.: Gesunder und pathologish ver~inderterZahnschmelz. Eine mikromorphologische und biokristaUographische Strukturanalyse. (Sammlung Meusser) Leipzig: J. Ambrosius Barth 1971 Witkop, C.J., Rao, S.: Inherited defects in tooth structure. In: Bergsma, D., Birth defects, original article series. Part XI, Orofacial Structures, vol. 7, 153-184. New York: Williams and Wiikins Co. 1971
Received March 1 / Accepted November 22, 1976
