Matrix Vol. 12/1992, pp. 251-255 © 1992 by Gustav Fischer Verlag, Stuttgart
Original Papers
Origin of Mineral Crystal Growth in Collagen Fibrils WOLFIE TRAUB, TALMON ARAD and STEPHEN WEINER Department of Structural Biology, Weizmann Institute of Science, Rehovot, Israel.
Abstract Collagen fibrils from young turkey-leg tendons, just beginning to mineralize, were stained with uranyl acetate and examined by electron microscopy. Small needle-like mineral crystals were observed and located, in relation to the collagen banding pattern, as originating at the e band in the gap region and near the surface of the fibrils. These are evidently the sites of crystal nucleation. They lie near binding locations on collagen fibrils of two glycosylated proteins believed to be implicated in the mineralization process, as well as the sites of early crystals in embryonic fowl bones. Key words: apatite crystal, biomineralization, bone, collagen fibril, turkey tendon.
Introduction Bone, dentin and mineralized tendon have, at the molecular level, similar structures derived from the organized growth of carbonated apatite (dahllite) crystals within a matrix of type I collagen fibrils and other organic components (Lowenstam and Weiner, 1989; Glimcher, 1984; Veis and Sabsay, 1983). The crystals are oriented with their c-axes along the fibril directions and are associated primarily with gap regions between collagen molecules (Schmidt, 1936; Hodge and Petruska, 1963). Whereas the mineralized fibrils in bone are organized into quite complex superstructures (Currey, 1984; Reid, 1987) those in turkey tendon are all essentially parallel to each other Oohnson, 1960). Mature turkey tendon has a highly organized structure (Nylen et al., 1960), with the plate-shaped mineral crystals arranged in parallel arrays through the collagen fibrils (Weiner and Traub, 1986; Traub et al., 1989). It has also been found that, within individual lamellae, rat bone has a very similar structural organization of fibrils and crystals, but that there are large variations in orientation between adjacent lamellae (Weiner et al., 1991). .. We have recently been studying the developmental stages 6f crystal growth in young, barely mineralized, turkey tendon (Traub et al., 1992a). The smallest crystals observed were short needles, in bands near the surface of the collagen fibrils. Longer needles, up to the length of the collagen gap regions, were also seen, and, evidently at a later stage, single
crystal belts extending partly or wholly through the fibrils. Finally, in mature tendon, crystal platelets, apparently derived from the cracking of belts, extend partly into the collagen overlap regions. Here we report on the location, in stained collagen fibrils, of the smallest needles, presumably the sites of crystal nucleation.
Materials and Methods Leg tendons were obtained from freshly sacrificed 15week-old domestic turkeys and frozen till used. Adhering tissue was removed and millimeter-sized pieces dissected at, or just ahead of, the mineralization front. These were crushed in liquid nitrogen with a mortar and pestle, and then sonicated in about 0.25 ml of water saturated with respect to apatite, or in distilled water, which gave the same results. A drop of this suspension was placed on a 400-mesh electron microscope grid with a pioloform film supported by a thin layer of carbon and then lyophilized. Subsequently, the grids were treated for 5 sec with 0.5% uranyl acetate, dissolved in 100% ethanol to avoid demineralization of the fibrils (Maitland and Arsenault, 1991). The uranyl acetate caused positive staining of the collagen fibrils, but, as the specimens were not subsequently washed, it also caused some negative contrast. As controls, additional specimens of young turkey tendon were prepared identically, but without uranyl-acetate staining. The speci-
252
W. Traub et al.
a
Fig.l. Electron micrographs of fibrils from 15-week-old turkey tendon stained with 0.5% uranyl acetate in ethanol. (a) shows a tegion of a fibril with very small mineral crystals and (b) shows a part of (a) at somewhat higher magnification; (c) shows a fibril region stained in the same way, but without mineral crystals. The location of one set of collagen a to e bands is shown in (b), as are othere bands on all three photographs. The band periodicity is approximately 67 nm. Many needle-like crystals (dark horizontal lines) can be seen in the gap zones of (a) and (b), but nowhere in the overlap zones, nor in any portion of (c). Some very short crystals (e.g. around A) lie on the e bands; longer ones extend from e to d (right of B), or from e to a (right of C), or right across the gap zone (right of D).
Crystal Growth in Collagen mens were examined in a Philips 400T transmission electron microscope.
Results Fig. 1 shows a collagen fibril, stained with uranyl acetate, with small needle-like mineral crystals parallel to the length of the fibril. The electron micrograph shows a comhination of mainly positive- and some negative-staining effects, with the generally darker zones, containing the crystals, corresponding to the collagen gap regions. The bands can be correlated with the standard collagen type I band pattern (Chapman, 1974), as indicated in Fig. 1 b. The very smallest crystals lie squarely on the e band. Longer crystals also lie on the e band, but extend from it in either or both directions up to the edges of the gap region. There are no signs of any crystals in the overlap regions. Other stained samples taken from near the mineralization front of the same turkey tendon show the same handing pattern with various amounts of mineralization. Some
253
show no crystals at all (Fig. 1 c), others crystals of different lengths, with almost all clearly crossing the e band. In some cases the crystals all essentially fill the gap regions, but still do not extend into the overlap zones. This also serves to distinguish them from the microfilaments observed in some negative-stain preparations of skin collagen, which can be seen to extend along the fibrils through both gap and overlap zones (Veis et aI., 1970). The unstained fibrils show the needle-like crystals rather more clearly (Fig. 2). They do indeed lie in regular bands separated by the 67-nm collagen periodicity and have been observed in various sizes up to the length of the collagen gap region (Fig. 2 c). Stereo pairs of electron micrographs, like those in Fig. 2, show, firstly, that these newly-formed crystals are indeed needle-like rather than edge-on views of crystal plates as in mature fibrils (Weiner and Traub, 1986; Traub et aI., 1989) and, secondly, that the crystals lie near the surface of the fibrils. Apatite-like electron diffraction patterns were obtained from some of these specimens, but of much weaker intensity than those observed from mature mineralized tendon.
a Fig. 2. Electron micrographs of unstained turkey tendon fibrils, showing early stages of mineralization, with bands of needle-like crystals (dark lines) of different lengths separated by the 67-nm collagen period. Crystal lengths in 2a, band c are approximately 10,20 and 30 nm respectively. Many of the crystals in 2 c have grown to the length of the collagen gap zone.
254
W. Traub et aI.
Figs. 1 a und 2 a also show particles, some 20 nm in diameter, which we observed frequently located near bands of crystals on fibrils at the earliest stages of mineralization.
Discussion Our results indicate quite specific locations on the turkey tendon collagen fibrils where crystals start to grow. The e band lies near the middle of the collagen gap region, so apparently crystal growth can proceed relatively unhindered in both directions along the fibril until the gap region is filled. The e band, which extends over about 8 nm of the fibril, has a unique combination of charged collagen amino acid side-chains, as do the a, b, c and d bands (Chapman, 1974; Chapman et aI., 1981). Charged residues in all these bands might concentrate calcium, and possibly phosphate, ions as a prelude to crystal growth, but it is not obvious to us why the one band should be favored for nucleation. A clue regarding the specificity of the e band may be provided by reports of regular binding to collagen fibrils by two non-collagenous proteins. Decorin binds to the surface of collagen fibrils in skin, cornea and tendon preferentially near the centers of the gap regions (Scott and Offord, 1981; Fleischmajer et aI., 1991). Its occurrence primarily in nonmineralizing tissues has led to the suggestion that it may in fact inhibit crystal formation. Phosphophoryn, which has been proposed as a mediator of mineralization in dentin (Stetler-Stevenson and Veis, 1986), has been found attached to the surface of collagen fibrils predominantly near the e band (Traub et aI., 1992 b). The initiation of crystal formation at the e band may therefore result, at least in part, from specific binding on the collagen fibril surface of a protein mediating nucleation. Our investigation of crystal nucleation sites on collagen fibrils has so far been restricted to turkey tendon, but our results are in accord with a remarkable study of developing bone in embryonic fowl, published 35 years ago Oackson, 1957), in which the first crystals were found in apparently the same locations. This result was questioned on technical grounds (Irving, 1973), never repeated and essentially ignored. However, in the light of our findings reported here and the binding of phosphophoryn and decorin near the collagen e band, we believe that the 1957 report does indeed indicate the location of early mineralization in bone. Thus it appears that there is a specific region in the gap zone of type I collagen fibrils where apatite crystal nucleation occurs, probably through the mediation of non-collagenous proteins. Acknowledgements We thank Kibbutz Hatzor for providing turkey legs. This study was supported by US-PHS Grant DE 06954.
References Chapman,J.A.: The staining pattern of collagen fibrils. Connect. Tissue Res. 2: 137-150,1974. Chapman,J.A., Holmes,D.F., Meek,K.M. and Rattew,C.J.: Electron-optical studies of collagen fibril assembly. In: Structural Aspects of Recognition and Assembly in Biological Macromolecules, ed. by Balaban, M., Sussman,J. L., Traub, W. and Yonath, A., ISS, Rehovot, Philadelphia, 1981, pp. 387-401. Currey,J.: The Mechanical Adaptations of Bones. Princeton University Press, Princeton, 1984. Fleischmajer, R., Fisher, L. W., MacDonald, E. D., Jacobs, L., Perlish, J. S. and Termine, J. D.: Decorin interacts with fibrillar collagen of embryonic and adult human skin. ]. Struct. BioI. 106: 82-90,1991. Glimcher, M.J.: Recent studies of the mineral phase in bone and its possible linkage to the organic matrix by protein-bound phosphate bonds. Philos. Trans. Roy. Soc. London Ser. B 304: 479-508, 1984. Hodge, A. J. and Petruska, J. A.: Recent studies with the electron microscope on ordered aggregates of the tropocollagen macromolecule. In: Aspects of Protein Structure, ed. by Ramachandran, G.N., AcademicPress, New York, 1963, pp. 289-300. Irving,J. T.: Theories of mineralization of bone. Clin. Orthop. 97: 225-236,1973. Jackson, S. F.: Fine structure of developing bone in the embryonic fowl. Proc. Roy. Soc. B 146: 270-280, 1957. Johnson, L. c.: Mineralization of turkey leg tendon. I. Histology and histochemistry of mineralization. In: Calcification in Biological Systems, ed. by Sognnaes, R. F., American Association for the Advancement of Science, Washington D. c., 1960, pp.I17-128. Lowenstam, H. A. and Weiner, S.: On Biomineralization. Oxford University Press, New York, Oxford, 1989. Maitland, M. E. and Arsenault, A. L.: A correlation between the distribution of biological apatite and amino acid sequence of type I collagen. Calcif Tissue Int. 48: 341-352, 1991. Nylen, M. U., Scott, D. B. and Mosley, V. M.: Mineralization of turkey leg tendon. II. Collagen-mineral relations revealed by electron and X-ray microscopy. In: Calcification in biological systems, ed. by Sognnaes, R. F., American Association for the Advancement of Science, Washington D.C., 1960, pp.129-142. Reid, S. A.: A study of lamellar organization in juvenile and adult human bone. Anat. Embryol.174: 329-338, 1986. Schmidt, W.J.: Uber der Orientierung der Kristallite im Zahnschmelz. Naturwissenschaften 24: 361, 1936. Scott, J. E. and Orford, C. R.: Dermatan sulphate-rich proteoglycan associates with rat tail-tendon collagen at the d band in the gap region. Biochem.]. 197: 213-216, 1981. Stetler-Stevenson, W. G. and Veis, A.: Type I collagen shows a specific binding affinity for bovine dentin phosphophoryn. Calcif Tissue Int. 38: 135-141,1986. Traub, W., Arad, T. and Weiner,S.: Three-dimensional ordered distribution of crystals in turkey tendon collagen fibers. Proc. Nat. Acad. Sci. USA 86: 9822-9826, 1989. Traub, W., Arad, T. and Weiner, S.: Growth of mineral crystals in turkey tendon collagen fibers. Conn. Tissue Res., In Press, 1992a. Traub, W., Jodaikin, A., Arad, T., Veis, A. and Sabsay, B.: Dentin phosphophoryn binding to collagen fibrils. Matrix, In Press, 1992b. Veis, A. and Sabsay, B.: Bone and tooth formation. Insights into
Crystal Growth in Collagen mineralization strategies. In: Biomineralization and Biological Metal Accumulation. ed. by Westbroek, P. and de jong, E. W., D. Reidel Publishing, Dordrecht, 1983, pp. 273 - 284. Veis,A., Bhamagar, R.S., Shuttleworth, CA. and Mussell, S.: The solubilization of mature polymeric collagen fibrils by lypotropic relaxation. Biochim. Biophys. Acta 200: 97 -112, 1970. Weiner,S. and Traub, W.: Organization of hydroxyapatite crystals within collagen fibrils. FEBS Lett. 206: 262 - 266, 1986.
255
Weiner, S., Arad, T. and Traub, W.: The structure of bone lamellae. FEBS Lett. 285: 49-54, 1991. Dr. Wolfie Traub, Department of Structural Biology, Weizmann Institute of Science, Rehovot, Israel.
Original Papers
Origin of Mineral Crystal Growth in Collagen Fibrils WOLFIE TRAUB, TALMON ARAD and STEPHEN WEINER Department of Structural Biology, Weizmann Institute of Science, Rehovot, Israel.
Abstract Collagen fibrils from young turkey-leg tendons, just beginning to mineralize, were stained with uranyl acetate and examined by electron microscopy. Small needle-like mineral crystals were observed and located, in relation to the collagen banding pattern, as originating at the e band in the gap region and near the surface of the fibrils. These are evidently the sites of crystal nucleation. They lie near binding locations on collagen fibrils of two glycosylated proteins believed to be implicated in the mineralization process, as well as the sites of early crystals in embryonic fowl bones. Key words: apatite crystal, biomineralization, bone, collagen fibril, turkey tendon.
Introduction Bone, dentin and mineralized tendon have, at the molecular level, similar structures derived from the organized growth of carbonated apatite (dahllite) crystals within a matrix of type I collagen fibrils and other organic components (Lowenstam and Weiner, 1989; Glimcher, 1984; Veis and Sabsay, 1983). The crystals are oriented with their c-axes along the fibril directions and are associated primarily with gap regions between collagen molecules (Schmidt, 1936; Hodge and Petruska, 1963). Whereas the mineralized fibrils in bone are organized into quite complex superstructures (Currey, 1984; Reid, 1987) those in turkey tendon are all essentially parallel to each other Oohnson, 1960). Mature turkey tendon has a highly organized structure (Nylen et al., 1960), with the plate-shaped mineral crystals arranged in parallel arrays through the collagen fibrils (Weiner and Traub, 1986; Traub et al., 1989). It has also been found that, within individual lamellae, rat bone has a very similar structural organization of fibrils and crystals, but that there are large variations in orientation between adjacent lamellae (Weiner et al., 1991). .. We have recently been studying the developmental stages 6f crystal growth in young, barely mineralized, turkey tendon (Traub et al., 1992a). The smallest crystals observed were short needles, in bands near the surface of the collagen fibrils. Longer needles, up to the length of the collagen gap regions, were also seen, and, evidently at a later stage, single
crystal belts extending partly or wholly through the fibrils. Finally, in mature tendon, crystal platelets, apparently derived from the cracking of belts, extend partly into the collagen overlap regions. Here we report on the location, in stained collagen fibrils, of the smallest needles, presumably the sites of crystal nucleation.
Materials and Methods Leg tendons were obtained from freshly sacrificed 15week-old domestic turkeys and frozen till used. Adhering tissue was removed and millimeter-sized pieces dissected at, or just ahead of, the mineralization front. These were crushed in liquid nitrogen with a mortar and pestle, and then sonicated in about 0.25 ml of water saturated with respect to apatite, or in distilled water, which gave the same results. A drop of this suspension was placed on a 400-mesh electron microscope grid with a pioloform film supported by a thin layer of carbon and then lyophilized. Subsequently, the grids were treated for 5 sec with 0.5% uranyl acetate, dissolved in 100% ethanol to avoid demineralization of the fibrils (Maitland and Arsenault, 1991). The uranyl acetate caused positive staining of the collagen fibrils, but, as the specimens were not subsequently washed, it also caused some negative contrast. As controls, additional specimens of young turkey tendon were prepared identically, but without uranyl-acetate staining. The speci-
252
W. Traub et al.
a
Fig.l. Electron micrographs of fibrils from 15-week-old turkey tendon stained with 0.5% uranyl acetate in ethanol. (a) shows a tegion of a fibril with very small mineral crystals and (b) shows a part of (a) at somewhat higher magnification; (c) shows a fibril region stained in the same way, but without mineral crystals. The location of one set of collagen a to e bands is shown in (b), as are othere bands on all three photographs. The band periodicity is approximately 67 nm. Many needle-like crystals (dark horizontal lines) can be seen in the gap zones of (a) and (b), but nowhere in the overlap zones, nor in any portion of (c). Some very short crystals (e.g. around A) lie on the e bands; longer ones extend from e to d (right of B), or from e to a (right of C), or right across the gap zone (right of D).
Crystal Growth in Collagen mens were examined in a Philips 400T transmission electron microscope.
Results Fig. 1 shows a collagen fibril, stained with uranyl acetate, with small needle-like mineral crystals parallel to the length of the fibril. The electron micrograph shows a comhination of mainly positive- and some negative-staining effects, with the generally darker zones, containing the crystals, corresponding to the collagen gap regions. The bands can be correlated with the standard collagen type I band pattern (Chapman, 1974), as indicated in Fig. 1 b. The very smallest crystals lie squarely on the e band. Longer crystals also lie on the e band, but extend from it in either or both directions up to the edges of the gap region. There are no signs of any crystals in the overlap regions. Other stained samples taken from near the mineralization front of the same turkey tendon show the same handing pattern with various amounts of mineralization. Some
253
show no crystals at all (Fig. 1 c), others crystals of different lengths, with almost all clearly crossing the e band. In some cases the crystals all essentially fill the gap regions, but still do not extend into the overlap zones. This also serves to distinguish them from the microfilaments observed in some negative-stain preparations of skin collagen, which can be seen to extend along the fibrils through both gap and overlap zones (Veis et aI., 1970). The unstained fibrils show the needle-like crystals rather more clearly (Fig. 2). They do indeed lie in regular bands separated by the 67-nm collagen periodicity and have been observed in various sizes up to the length of the collagen gap region (Fig. 2 c). Stereo pairs of electron micrographs, like those in Fig. 2, show, firstly, that these newly-formed crystals are indeed needle-like rather than edge-on views of crystal plates as in mature fibrils (Weiner and Traub, 1986; Traub et aI., 1989) and, secondly, that the crystals lie near the surface of the fibrils. Apatite-like electron diffraction patterns were obtained from some of these specimens, but of much weaker intensity than those observed from mature mineralized tendon.
a Fig. 2. Electron micrographs of unstained turkey tendon fibrils, showing early stages of mineralization, with bands of needle-like crystals (dark lines) of different lengths separated by the 67-nm collagen period. Crystal lengths in 2a, band c are approximately 10,20 and 30 nm respectively. Many of the crystals in 2 c have grown to the length of the collagen gap zone.
254
W. Traub et aI.
Figs. 1 a und 2 a also show particles, some 20 nm in diameter, which we observed frequently located near bands of crystals on fibrils at the earliest stages of mineralization.
Discussion Our results indicate quite specific locations on the turkey tendon collagen fibrils where crystals start to grow. The e band lies near the middle of the collagen gap region, so apparently crystal growth can proceed relatively unhindered in both directions along the fibril until the gap region is filled. The e band, which extends over about 8 nm of the fibril, has a unique combination of charged collagen amino acid side-chains, as do the a, b, c and d bands (Chapman, 1974; Chapman et aI., 1981). Charged residues in all these bands might concentrate calcium, and possibly phosphate, ions as a prelude to crystal growth, but it is not obvious to us why the one band should be favored for nucleation. A clue regarding the specificity of the e band may be provided by reports of regular binding to collagen fibrils by two non-collagenous proteins. Decorin binds to the surface of collagen fibrils in skin, cornea and tendon preferentially near the centers of the gap regions (Scott and Offord, 1981; Fleischmajer et aI., 1991). Its occurrence primarily in nonmineralizing tissues has led to the suggestion that it may in fact inhibit crystal formation. Phosphophoryn, which has been proposed as a mediator of mineralization in dentin (Stetler-Stevenson and Veis, 1986), has been found attached to the surface of collagen fibrils predominantly near the e band (Traub et aI., 1992 b). The initiation of crystal formation at the e band may therefore result, at least in part, from specific binding on the collagen fibril surface of a protein mediating nucleation. Our investigation of crystal nucleation sites on collagen fibrils has so far been restricted to turkey tendon, but our results are in accord with a remarkable study of developing bone in embryonic fowl, published 35 years ago Oackson, 1957), in which the first crystals were found in apparently the same locations. This result was questioned on technical grounds (Irving, 1973), never repeated and essentially ignored. However, in the light of our findings reported here and the binding of phosphophoryn and decorin near the collagen e band, we believe that the 1957 report does indeed indicate the location of early mineralization in bone. Thus it appears that there is a specific region in the gap zone of type I collagen fibrils where apatite crystal nucleation occurs, probably through the mediation of non-collagenous proteins. Acknowledgements We thank Kibbutz Hatzor for providing turkey legs. This study was supported by US-PHS Grant DE 06954.
References Chapman,J.A.: The staining pattern of collagen fibrils. Connect. Tissue Res. 2: 137-150,1974. Chapman,J.A., Holmes,D.F., Meek,K.M. and Rattew,C.J.: Electron-optical studies of collagen fibril assembly. In: Structural Aspects of Recognition and Assembly in Biological Macromolecules, ed. by Balaban, M., Sussman,J. L., Traub, W. and Yonath, A., ISS, Rehovot, Philadelphia, 1981, pp. 387-401. Currey,J.: The Mechanical Adaptations of Bones. Princeton University Press, Princeton, 1984. Fleischmajer, R., Fisher, L. W., MacDonald, E. D., Jacobs, L., Perlish, J. S. and Termine, J. D.: Decorin interacts with fibrillar collagen of embryonic and adult human skin. ]. Struct. BioI. 106: 82-90,1991. Glimcher, M.J.: Recent studies of the mineral phase in bone and its possible linkage to the organic matrix by protein-bound phosphate bonds. Philos. Trans. Roy. Soc. London Ser. B 304: 479-508, 1984. Hodge, A. J. and Petruska, J. A.: Recent studies with the electron microscope on ordered aggregates of the tropocollagen macromolecule. In: Aspects of Protein Structure, ed. by Ramachandran, G.N., AcademicPress, New York, 1963, pp. 289-300. Irving,J. T.: Theories of mineralization of bone. Clin. Orthop. 97: 225-236,1973. Jackson, S. F.: Fine structure of developing bone in the embryonic fowl. Proc. Roy. Soc. B 146: 270-280, 1957. Johnson, L. c.: Mineralization of turkey leg tendon. I. Histology and histochemistry of mineralization. In: Calcification in Biological Systems, ed. by Sognnaes, R. F., American Association for the Advancement of Science, Washington D. c., 1960, pp.I17-128. Lowenstam, H. A. and Weiner, S.: On Biomineralization. Oxford University Press, New York, Oxford, 1989. Maitland, M. E. and Arsenault, A. L.: A correlation between the distribution of biological apatite and amino acid sequence of type I collagen. Calcif Tissue Int. 48: 341-352, 1991. Nylen, M. U., Scott, D. B. and Mosley, V. M.: Mineralization of turkey leg tendon. II. Collagen-mineral relations revealed by electron and X-ray microscopy. In: Calcification in biological systems, ed. by Sognnaes, R. F., American Association for the Advancement of Science, Washington D.C., 1960, pp.129-142. Reid, S. A.: A study of lamellar organization in juvenile and adult human bone. Anat. Embryol.174: 329-338, 1986. Schmidt, W.J.: Uber der Orientierung der Kristallite im Zahnschmelz. Naturwissenschaften 24: 361, 1936. Scott, J. E. and Orford, C. R.: Dermatan sulphate-rich proteoglycan associates with rat tail-tendon collagen at the d band in the gap region. Biochem.]. 197: 213-216, 1981. Stetler-Stevenson, W. G. and Veis, A.: Type I collagen shows a specific binding affinity for bovine dentin phosphophoryn. Calcif Tissue Int. 38: 135-141,1986. Traub, W., Arad, T. and Weiner,S.: Three-dimensional ordered distribution of crystals in turkey tendon collagen fibers. Proc. Nat. Acad. Sci. USA 86: 9822-9826, 1989. Traub, W., Arad, T. and Weiner, S.: Growth of mineral crystals in turkey tendon collagen fibers. Conn. Tissue Res., In Press, 1992a. Traub, W., Jodaikin, A., Arad, T., Veis, A. and Sabsay, B.: Dentin phosphophoryn binding to collagen fibrils. Matrix, In Press, 1992b. Veis, A. and Sabsay, B.: Bone and tooth formation. Insights into
Crystal Growth in Collagen mineralization strategies. In: Biomineralization and Biological Metal Accumulation. ed. by Westbroek, P. and de jong, E. W., D. Reidel Publishing, Dordrecht, 1983, pp. 273 - 284. Veis,A., Bhamagar, R.S., Shuttleworth, CA. and Mussell, S.: The solubilization of mature polymeric collagen fibrils by lypotropic relaxation. Biochim. Biophys. Acta 200: 97 -112, 1970. Weiner,S. and Traub, W.: Organization of hydroxyapatite crystals within collagen fibrils. FEBS Lett. 206: 262 - 266, 1986.
255
Weiner, S., Arad, T. and Traub, W.: The structure of bone lamellae. FEBS Lett. 285: 49-54, 1991. Dr. Wolfie Traub, Department of Structural Biology, Weizmann Institute of Science, Rehovot, Israel.
