In vitro study of the mechanism of oriented growth of tooth enamel apatite

Mayumi Iijima and Yutaka Moriwaki

Dental Materials and Technology
Asahi University, School of Dentistry,
1851-1 Hozumi, Hozumi-cho, Motosu-gun, Gifu 501-02, Japan

Introduction

In the early stage of tooth enamel formation, enamel crystals grow as extremely thin ribbons1-3), and are highly oriented along the c-axis3). In the mechanism of tooth enamel crystal formation, it is still inconclusive what regulates the initial lengthwise and oriented growth :Some previous studies suggested that fibrillar3,4), lamellar5), or tubular6-8) structure in the enamel matrix assist the orientation of the crystals. The recent studies9-11) hypothesize that enamel proteins, such as amelogenins, are related to the oriented growth of enamel crystals. It is also suggested that streaming of organic molecules12,13), a secretory force produced by ameloblasts14) , some additional orienting factor15) causes the orientational enamel growth. These hypotheses, however, do not exclude other possibilities.
We have been studying the mechanism of the lengthwise and oriented growth of tooth enamel crystals on the basis of a hypothesis : (1) ribbon-like enamel crystals grow initially as octacalcium phosphate (Ca8H2(PO4)6・5H2O, OCP)16,17) and (2) ionic diffusion through the layer of ameloblasts promotes the one directional growth and the orientation of OCP crystals18). This paper introduces our previous in vitro studies on (1) the mechanism of the lengthwise and oriented growth of tooth enamel apatite19-22) and (2) the roles of F- ion on tooth enamel-like apatite formation23-25).

Materials and Methods

In our model system, a synthetic cation selective membrane was used to control the ionic
diffusion. Several sets of large (500ml) and small (50ml) Pyrex glass bottles were used. A 50ml glass bottle was used as a Ca2+ ion diffusion vessel and was put into a 500ml bottle, which contained 300ml of PO4 solution. A cation selective membrane ( Asahi Glass Co., CMV, about 3cm2) was fixed to the Ca-bottle by silicon rubber and a plastic screw stopper. The reactions were carried out using Ca(CH3COO)2・H2O (30mM) and NH4H2PO4+(NH4)2HPO4 (3.6, 7.2, 14.4mM) at various pH levels ranging from 6.0 to 7.4 and at 37 ℃ for 3days. The effect of F- ion was studied in 7.2mM PO4 at pH6.5. Various amounts of NaF (0.1-2 ppm F ) were added to the PO4 solution. Details of the experiments are described elsewhere22,23).
After the reaction, the membrane was removed from solution and rinsed with distilled water for a few seconds, and then dried in the air. The product was analyzed using a micro-beam X-ray
diffraction camera ( with a collimator of 100μm diameter), an X-ray diffracto-meter ( XRD ), a scanning electron microscope ( SEM ) and a high resolution transmission electron microscope (HR TEM).

Results

OCP crystals grew onto the membrane with the c-axis perpendicular to the membrane on the whole (Fig.1a). Figure 1b represents the XRD patterns of the powdered and as grown product on the membrane. Since crystals grew with orientation, as grown product exhibited strong 002 and 004 reflections of OCP, whereas, powdered product exhibited a strong 100 reflection of OCP at 1.867nm. In case of 14.4mM and at pH6.0, both OCP and dicalcium phosphate dihydrate (DCPD) were obtained. In case of 7.2mM and 3.6mM, precipitation did not take place at pH6.0 and at pH6.5, respectively. The lower the pH became, the higher PO4 concentration was necessary to yield the precipitation on the membrane.
Table 1 represents the morphology of OCP crystals grown at various PO4 concentrations and pHs. Figures 2 (a )- (d) are the SEM photographs of OCP crystals grown in the 7.2mM PO4 solution at various pHs. When the pH was higher than 7.0, flake-like crystals grew on the membrane (Fig.2a). The lengthwise growth (R in Table1) took place at pH lower than 6.8. As the pH decreased, the growth along the c-axis of OCP increased (Fig.2b-2d). The growth in the c-axis direction of OCP increased with a decrease in the pH and the PO4 concentration. However, the growth depended more strongly on the pH than on the concentration of PO4.
Fluoride ions affected the overall property of the crystals grown on the membrane. Long and thin ribbon-like crystals grew without and in the presence of 0.1ppm F (Fig.3a, 3b). The XRD photographs of these crystals showed OCP reflections with orientation (Fig.4a). The crystals obtained in the presence of 1ppm F had thin and ribbon-like morphology (Fig.3c) but the XRD photograph was apatite-like (Fig.4b). The morphology and size of crystals changed drastically between 1 and 2 ppm F. Small needle-like crystals were obtained in the presence of 2 ppm F (Fig.3d).
In the presence of 0.1 and 1ppm F, epitaxial overgrowth of apatite on the (100) of OCP took place and apatite/OCP/apatite sandwich structure was formed (Fig.5). These crystals have long and thin plate-like morphology and embed an OCP lamellar in the center of the crystals. In the lattice images of these crystals, the (100) fringes of OCP and the (100) fringes of apatite were parallel to each other, indicating that apatite grew on the (100) of thin plate-like OCP crystals epitaxially. Although ribbon-like crystals grown in the presence of 0.1ppm F appeared superficially homogeneous (Fig.3b) and they were identified to be OCP by the XRD method, a small amount of apatite was observed on the marginal part of the plate-like OCP crystals (Fig.5a). In the presence of 1ppm F, the OCP region was smaller than that of apatite (Fig.5b). The width of OCP lamellar in the a*-axis direction was one to several unit cells thick. The main part of the crystals were apatite, resulting in the XRD pattern of apatite (Fig.4b). Some crystals embedded a central plane instead of the distinct OCP lamellar. Figure 6 is a cross-section of the crystal with a central plane. The thickness of the plane was about 0.8-1.2nm. The central plane in the present study is probably a variation of the central OCP crystal, as it was formed in the same reaction chamber. The feature is close to that of tooth enamel which shows a so-called central dark line. Table 2 summarizes the property of the crystals obtained under various amounts of F- ions.

Discussion

Concerning to the mechanism of the oriented growth of enamel crystals, some of the studies on the enamel matrix suggested that the enamel proteins assist the oriented growth9-11). The present study demonstrated that the oriented growth took place on the membrane, even though there was no organic matrix which supports the oriented growth. The ribbon-like crystals which grew on the membrane were morphologically resemble to the crystals in the early stage of enamel crystal formation. Morphology of crystal sometimes can suggest the process or the condition under which crystal growth took place. In the usual precipitation reaction in solution without the membrane, OCP grew into rectangular plates even at pH6.5 (unpublished data, Iijima ). The length in the c-axis direction was much shorter than that of OCP grown in the membrane system at the same pH. The one-directional supply of Ca2+ ions through the membrane should cause the oriented and lengthwise growth of OCP. During the enamel formation, Ca2+ ions are supplied from the layer of ameloblasts26-30). Such one-directional ionic flow might, in part, contribute to the oriented and lengthwise growth of tooth enamel crystals.
Although the F concentration in the enamel fluid is very low (4-7x10-7M31)), F- ion seems to be included in the process of enamel apatite formation, because a remarkably small amount of F- ion promotes apatite formation32,33), increases the crystallinity of apatite34), induces and accelerates the hydrolysis of OCP to apatite35,36). The OCP/apatite ratio in the crystals changed between 0.1 and 1ppm F, in other words, F- ions regulated the predominant growing phase, i.e., OCP or apatite. F- ions might change the surface energy of the crystals, the driving force of crystallization and hydrolysis. The thickness of the central OCP lamellar would be small when the stability of OCP in solution is small.
A cross-sectioned tooth enamel crystal of mammals with 0.01-0.03%wt F often exhibits a central line. The origin of the central line is still inconclusive, though it reflects the initial stage of the crystallization of tooth enamel. Apatite with the central layer has been synthesized in CO3-containing solution, and moreover, the reaction related to OCP37,38). The cross-section of such crystals exhibited a central line whose feature was similar to that of tooth enamel crystal with the " central dark line". In the present study, it is probable that the central layer thinner than 1.87nm was a variation of the central OCP lamellar, because they were formed in the same reaction. Apatite should precipitate on the OCP crystal immediately after one unit cell thick OCP was formed.
In conclusion, when Ca2+ ions were supplied one directionally through the membrane, OCP
crystals grew with orientation and attained a ribbon-like morphology in slightly acidic solution. In the presence of 1ppm F, apatite crystals embedded a layer in the center, which should relate to OCP, exhibiting a central line in the cross- section, whose feature was similar to that of tooth enamel crystal with "the central dark line". The results strongly support the idea that initially formed thin plate-like OCP crystals acted as a template for subsequent overgrowth of apatite in the process of tooth enamel formation37-39).


References

1) Fearnhead,R.W. : Nature 189 : 509-510, 1960.
2) Ronnholm,E. : J. Ultrastruct. Res. 6 : 249-303, 1962.
3) Nylen, M.U., Eanes, E.D. and Omnell, K.A. : J.Cell Biol. 18 : 109-123, 1963.
4) Frank, R.M. : Arch. Oral Biol. 4 : 29, 1961.
5) Ronnholm,E. : J. Ultrastruct. Res. 6 : 368, 1962.
6) Decker,J.D. : J. Ultrastruct. Res. 44 : 58, 1973.
7) Jessen,H. : Arch. Oral Biol. 13 : 351, 1968.
8) Travis,D.F. and Glimcher,M.J. : J. Cell Biol. 23 : 447-497, 1964.
9) Robinson,C. and Kirkham,J. : Tooth Enamel IV. pp.261-265, Elsevier, Amsterdam, 1984.
10) Slavkin,H.C., Bessem,C., Bringas,P., Zeichner,D.M., Nanci,A. and Snead, M. L. : Differenciation 37 : 26, 1988.
11) Diekwisch,T., Fincham,A.G. and Slavkin H.C. :in Abstract of the 5th International Conference
on Tooth Morphogenesis and Differenciation. No.36. 1994.
12) Eastoe,J.E. : Arch. Oral Biol. 8 : 633, 1963.
13) Fearnhead,R.W.: in Tooth Enamel II, Wright, Bristol, 1965.
14) Osborn,J.W. : Calcif. Tissue Res. 5 : 115-132, 1970.
15) Boyde,A. : in Tooth Enamel II, Wright, Bristol, 1965.
16) Brown,W.E., Smith,J.P., Lehr, J.R. and Frazier, A.W. : Nature, 196 : 1048-1055, 1962.
17) Brown,W.E. : in Tooth Enamel II, pp.11-14, John Wright & Sons, Bristol, 1965.
18) Moriwaki,Y., Doi,Y., Kani,T., Aoba,T., Takahashi,J. and Okazaki,M. : Mechanism of Tooth enamel Formation. pp.239-256, Quintescence, Tokyo, 1983.
19) Hata,M., Moriwaki,Y., Doi,Y., Goto, T., Wakamatsu, N., Kamemizu, H. and Yoshida,S. Tooth Enamel IV. pp.37-41, Elsevier, Amsterdam, 1984.
20) Hata,M., Moriwaki,Y., Doi,Y., Goto, T., Wakamatsu, N., and Kamemizu, H.: J.Japn. Assoc. Crystal Growth 12 : 91-99, 1985.
21) Iijima,M. and Moriwaki,Y. : J. Crystal Growth 96 : 59-64, 1989.
22) Iijima,M. and Moriwaki, Y. : J. Crystal Growth 112 : 571-579, 1991.
23) Iijima,M., Tohda,H., and Moriwaki, Y. : J. Crystal Growth 116 : 319-326, 1992.
24) Iijima,M., Tohda,H., Suzuki,H., Yanagisawa,T., and Moriwaki, Y. : Calcif. Tissue Int. 50 : 357-361, 1992.
25) Iijima,M., Tohda,H., Suzuki,H., Yanagisawa,T., and Moriwaki, Y. : Jpn. J.Oral Biol. 33 : 495-499, 1991.
26) Munhoz,C.O. and Leblond,C.P. : Calcif. Tissue Res. 15 : 221, 1974.
27) Takano,Y., Crenshaw, M. and Reith, E.J. : Calcif. Tissue Int. 34 : 211, 1982.
28) Bawden, J.W. : Anat. Record. 224 : 226-233, 1989.
29) Kawamoto,T. and Schimidzu,M. : Calcif. Tissue Int. 46 : 406-414, 1990.
30) Kawamoto,T. and Shimidzu,M. : Jpn. J. Oral Biol. 36 : 365-382, 1994.
31) Aoba,T. and Moreno,E.C. : Calcif. Tissue Int. 41 : 86-94, 1987.
32) Taves,D.R. and Neuman,W.F. : Arch Biochem. Biophys. 108 : 390-397, 1964.
33) Eanes,E.D. : J.Dent.Res. 59 : 144-150, 1980.
34) Frazier,P.D., Little,M.F. and Casciani, F.S. : Arch. Oral Biol. 12 : 35-42, 1967.
35) Newesely,H. : Adv. Oral Biol. 4 : 11-42, 1970.
36) Aoba,T., Ishida,T., Yagi, T., Hasegawa, K. and Moriwaki,Y. : Jpn. J. Oral Biol. 17 ; 1-7, 1975.
37) Nelson,D.G.A., Barry, J.C. , Scields, C.P. , Glena,R. and Featherstone,J.D. B.: J. Colloid Interface Sci. 130 : 467, 1989.
38) Nelson,D.G.A. and Barry, J.C. : Anat.Rec. 224 : 265-276, 1989.
39) Miake,Y., Shimoda,S., Fukae,M. and Aoba,T. : Calcif. Tissue Int. 53 : 249- 256, 1993.


Table 1 The morphology of OCP grown on the membrane at various PO4 concentrations and pHs
at 37℃22).

    pH
PO4 (mM)  6.0  6.3  6.5  6.8  7.0  7.4
 14.4    Rs   R   R   SP   F   F
7.2    -    Rs Rs LP SP F  
3.6    -    -   - Rs SP F

F : flake-like ; SP : short plate-like ; LP : long plate-like ; R : ribbon-like crystals ; s : spherical
aggregate of crystals



Table 2 Characterization of the products grown on the membrane in 7.2mM PO4 solution in the presence of various amount of F at pH6.5 and 37℃25).

F concentration
(ppm)         0       0.1         1        2

Product      OCP  lamellar mixed  lamellar mixed    Apatite
                crystals of OCP  crystals of OCP
                 and apatite    and apatite

 SEM     Ribbon-like   Ribbon-like  Ribbon-like   Needle-like

 XRD      OCP       OCP     Apatite       Apatite

HRTEM      OCP    OCP >> Apatite  OCP << Apatite     Apatite


SEM : Scanning electron microscopy ; XRD : Micro-beam X-ray Diffraction ;
HRTEM : High resolution transmission electron microscopy



Figure captions

{ 現在付図のインストール準備中 }


Fig.1
(a) SEM photograph of the OCP crystals grown on the membrane at pH6.5. Reactionwas carried out in 14.4mM PO4 solution at 37 ℃ for 3days. (b) XRD patterns of powdered (lower) and as grown (upper) product on the membrane.

Fig.2
The morphology of OCP grown on the membrane in a 7.2mM PO4 solution at 37℃ for3 days22). The pHs of solutions were (a) pH7.4, (b) pH7.0, (c) pH6.5 and (d) pH6.3. Scale line is 5μm for (a), (b) and (c) and 50μm for (d).

Fig.3
SEM photographs of the crystals grown (a) without F, and in the presence of (b)0.1ppm F, (c) 1ppm F and (d) 2ppm F at pH6.5 and 37℃25).

Fig.4
Micro-beam XRD photographs of the products grown (a) without F and (b) in the resence of 1ppm F.

Fig.5
High resolution TEM photographs of crystals grown in the presence of (a) 0.1ppm and (b) 1ppm F in 7.2mM PO4 solution with the pH of 6.5 and 37 ℃.

Fig.6
A lattice image of a cross-sectioned crystal with a central plane25). The central plane is parallel to the (100) plane of apatite. Note that it looks like a centralline of tooth enamel crystal.