Biomimetic Growth and Self-Assembly of Fluorapatite Aggregates by Diffusion into Denatured Collagen Matrices.код для вставкиСкачать
COMMUNICATIONS Angew. Chem. Int. Ed.Engl. 1996,108, No. 22 0 VCH Verlagsgesellschf! mbH, D,-69451 Weinheim,1996 0570-0833/96/3522-2623 $15.W+ .25/0 2623 ~ Biomimetic Growth and Self-Assembly of Fluorapatite Aggregates by Diffusion into Denatured Collagen Matrices** Rudiger Kniep” and Susanne Busch Nature produces inorganic solids with fascinating outer shapes well suited to their function. The mechanisms of formation of these biogenic solids involve the combination of organic and inorganic components. Organic matrices control the mass transfer, define the location and number of nucleation centers, determine the orientation of the (crystal) growth, and/or act as templates for the final morphology of the inorganic solids. In our studies on mimicking the processes involved in bone or tooth formation,[’] we have investigated the biomimetic growth of fluorapatite in gelatin matrices. A simple (static) diffusion cell was employed, like that known, in principle, for solid-state reactionsC2] and that already used in more complex form in dynamic collagen systems.[31The experimental conditionsr4]were chosen such that at constant temperature (25 “C) solutions of calcium and of phosphate/fluoride diffuse into a gelatin plug from opposite ends (the so-called double diffusion techniq~e[~I). In the course of the diffusion, discrete crystallization zones form, which are denoted as Liesegang bands in analogy to the classical work of Liesegang (Liesegang rings) .[61 The time-dependent growth of the fluorapatite aggregates can be particularly well monitored in the middle band on the phosphate side; the largest aggregates were also obtained in this band. The crystallization products in all stages of development correspond to the fluorine-rich boundary region of the solid solution series hydroxyapatite/fluorapatite (Fig. I).”, Although the gelatin141used here as the organic matrix is of rather poorly defined s t r u c t ~ r e , [ ~it. ’ is ~ ~responsible for a specific function in the control of the growth of the fluorapatite aggregates. The use of other organic matrices (e.g., polysaccha400 j501 300 100 50 20 25 30 35 201” 40 45 50 55 Fig. 1. X-ray powder diagram (Cu,,, radiation) of a sphere fraction and calculated diagram of fluorapatite (P6,lm; a = 936.6(3), c = 685.3(3) pm) [S]. Refined lattice constants of the sphere fraction: a = 937.6(3), c = 688.6(2) pm (hydroxyapatite 171: a = 942.4(4), c = 687.9(4) pm). Remarkably, the diffraction peaks of the sphere fraction are comparatively sharp for biomimetic apatite, the X-ray diagram indicates a single phase, and the background is relatively low (small amounts of X-ray amorphous components). N = count rate. [*I [**I Prof. Dr. R. Kniep, Dip1.-Ing. S. Busch Eduard-Zintl-Institut der Technischen Hochschule Hochschulstrasse 10, D-64289 Darmstadt (Germany) Fax: Int. code +(6151)166029 This work was supported by the Fonds der Chemischen Industrie 2624 0 VCH Verlagsgesellschafi mbH, 0-69451 Weinheim, 1996 ~~ rides, agar, and carragenan) under otherwise identical experimental conditionsr4] leads to fundamentally different morphologies of the fluorapatite (irregular intergrowth of short hexagonal prisms or xenomorphic precipitates). When gelatin is used as the organic matrix, “anisotropic” spheres are formed as the end product, the centers of which contain an elongated hexagonal-prismatic seed crystal. For unchanged morphology of the aggregates, the rates of diffusion of the solutions and of formation of the Liesegang bands, and hence the rate of growth of the fluorapatite are determined by the initial pH of the gelatin. Thus, in a gelatin plug that has not been acidified (PH 5.5, length 3 cm), apatite spheres with diameters of 100 pm are first observed after six weeks in the middle Liesegang band on the phosphate side. If the pH of the gelatin is adjusted to 2.5-3.5, apatite spheres with a diameter of 100 pm are formed after only two weeks. The growth of the fluorapatite spheres begins with elongated hexagonal-prismatic seed crystals, which reach a critical length of 5-10 pm. The progressive stages of the self-assembled growth of the needlelike seed crystals to give spherical aggregates are shown in Figure 2, and require approximately one week. First, oriented upgrowths occur on the prism faces at both ends of the seed crystal simultaneously and roughly symmetrically (mirror symmetry perpendicular to the prism axis), leading to the formation of dumbbell-shaped aggregates. The orientation of these upgrowths does not necessarily indicate the existence of strict crystallographic relations; nevertheless, the maximum angle of divergence (between the prism axes of the growing crystals and the prism axis of the seed crystal) appears to be fixed at approximately 45”. The dumbbell region is extended through further growing generations according to the principles of self-similarity;[’’] however, the “bar” of the dumbbell (seed crystal) initially remains visible. This situation can be simulated in a two-dimensional form which mirrors reality very well in a simplified way, as shown in Figure 3.r’21 Starting from a needlelike seed crystal, in the first generation similar needlelike crystals are formed at both ends, which have a maximum angle of 45” to the long axis of the seed crystal. The length of the crystals of each successive generation is shorter than the length of the crystals of the preceding generation by a factor of about 0.68 (estimated mean value from scanning electron microscope (SEM) images, error = ca. f10%). Each “new” end of a crystal generation leads to self-similar multiplication. The result is illustrated in Figure 3, which shows the simulation of five successive self-similar crystal generations at each end of the seed crystal. The fact that the crystals interpenetrate in the simulation is certainly not realistic, but this has no great effect on the morphology of the aggregate. It is more important that the bar of the dumbbell (the seed crystal) remains visible in this growth state and correspondingly in the simulation, that the surface curvature of the dumbbell in the direction of the seed crystal is already indicative of the “closing” of the aggregate, and that the extent of the aggregate, growing from the two ends of the seed crystal, increases by a factor of approximately 2 relative to the length of the seed crystal. The further development of the aggregate-going from the dumbbell to a sphere--can again be seen in Figure 2. Figure 4 shows the SEM image of the surface of a just completed (notched) surface of a sphere. This reveals that also in this last generation needlelike crystals are formed. Their prism cross section is only about 0.05 pm, which means that the closed aggregate has a very large surface area. The (notched) spheres formed in the gelatin matrix continue their growth by adding radial shells of material and forming almost perfect spheres. This is a second (probably more conven- 0570-083319613522-2424 $ 15.00+ .2Sj0 Angew. Chem. I n t . Ed. Engl. 1996,35, No. 22 COMMUNICATIONS Fig. 4. SEM image of the surface of a just closed sphere aggregate. In accordance with the self-similar branching patterns  described in the text and shown in Figure 3, the dumbbell-shaped aggregates close to give notched spheres after the tenth generation. tionally spherolitic) growth mechanism, which will not be described further here. After about four weeks, sphere diameters of up to 400 pm are found. Our observations indicate that the important limiting factor in the growth of the spheres is the gelatin matrix, which, under the given experimental conditions, is stable for only about six weeks. The self-assembled hierarchical growth of the fluorapatite aggregates shows immediate parallels to the topological branching criteria of the macromolecular starburst dendrimer~."~' However, the dimensionality-and Fig. 2. SEM images of progressive stages of the self-assembled (hierarchical) growth of the fluorapatite aggrethus the information transfer-in the gates: from the elongated hexagonal-prismatic seed (top left) through dumbbell shapes to the sphere (bottom aggregates described here extends to right). the macroscopic range. In accordance with their microcrystalline structure, the spherical aggregates are opalescent. A fraction of spheres with diameters of up to 300 pm is shown in Figure 5. On a flat support, the Fig. 5. Optical micrograph of a fluorapatite sphere fraction. The long edge of the figure corresponds to about 4 mm. Fig. 3. SEM image of a dumbbell-shaped fluorapatite aggregate (top), Two-dimensional simulation [I21 of self-similar branching in five successive generations (bottom). The simulation shows branching (splitting) into only four crystals (with divergence angles of 3 0 ) of the new generation, which is a simplified representation. In fact, orders of branching are higher (see also Fig. 2), and are, in principle, only limited by the maximum angle of divergence (ca. +4Y) formed by the long axis of the new crystals and the long axis of the preceding seed crystal. The factor of shortening of preceding generations is discussed in the text. Anww Chem. Int. Ed. E n d . 1996, 35,No. 22 spheres orient themselves preferentially to form chains (electrostatic effects), and can be lifted out in the form of "strings of beads" by using the point Of a Thermoanalyticalinvestigations (thermogravimetry (TG) and differential thermal analysis (DTA)) on spherical aggregates dried at temperature were carried out at temperatures up to a maximum of 1000 "C (heating rate min-l; atmosphere: air).[141The X-ray Powder diagram of the sample after thermal treatment is identical to 0 VCH Verlaxsxesellschafr mbH, 0-69451 OC Weinheim, 1996 0570-0833~U6~3522-2625 $15.00+ .25/0 2625 COMMUNICATIONS the diagram shown in Figure 1 except that the profiles of the diffraction peaks are sharper (refined lattice constants: a = 937.4(1) pm, c = 688.7(1) pm). The mass loss (total ca. 6.5%) takes place in two stages: stage 1 (80-200°C) l.Swt%, and stage 2 (200-860 “C) 5wt %. In accord with DTA experiments on human dentine,the second stage is associated with an exothermic reaction (pyrolysis of the protein fibers). Figure 6 shows the interference image of a just closed fluorapatite aggregate with a diameter of 125 pm, which was obtained with an optical microscope with cross polarizers, as viewed along the optical axis of the hexagonal central seed. The Fig. 6. Image from a polarization microscope (cross polarizers) of a just closed fluorapatite aggregate of diameter 125 pm.Viewed along the hexagonal-prismatic seed. hexagonal cross section of the seed crystal can be seen in the center, its diameter increases approximately twofold during the course of the growth of the spherical aggregate. The interference pattern of the total aggregate is characterized by the “Brewster polarization cross” and the polarization colors within the cross segments (increasing thickness of the spheres from the surface to their centers, and variable inclination between the optical axes of the individual crystallites and the incident polarized light). We conclude by mentioning some possible interpretations of the growth phenomenon described, although detailed investigations have either not been started or are not yet complete.[”- 16, 17]There are two basic possibilities, which, individually or taken together, could be responsible for the selfassembled control of the aggregate growth: The aggregates correspond to organized composite systems in which the growth of generations is controlled by an epitaxially linked composite of fluorapatite and the organic component. The individual crystallites or composite units contain a permanent dipole, so that intrinsic electrical lines of force take over control of the growth of the aggregates. The polarity of collagen, the piezoelectric properties of calcified bone tissue,[’’] and the structural peculiarity of the apatite family, fluctuating between centrosymmetric and acentric distribution of the X species (X = F, C1, OH),[’91support these ideas, also bearing in mind the extremely mild biomimetic growth conditions, which should favor a high degree of order of the atomic arrangements, and the fact that partial exchange of F- by OH- causes an asymmetric displacement of the two sDecies from their normal positions.[201 In summary, a wide interdisciplinary area is emerging, which encompasses chemistry, physics, materials science, and medicine. 2626 0 VCH Verlagsgesellschaft mhH, 0-69451 Weinheim, 1996 Received: April 29, 1996 [Z90771E] German version: Angew. Chem. 1996, 108, 2787-2791 Keywords: apatite gelatin * solids * biomineralization - fluorine compounds * [I] L. T. Kuhn, D. J. Fink, A. H. Heuer in Biomimetic Materials Chemistry (Ed.: S. Mann), VCH, New York, 1996.  H. Schmalzried, T. Frick, Ber. Bunsenges. Phys. Chem. 1995, Y9, 914.  A. L. Boskey, J. Phys. Chem. 1989,93, 1628.  Experimental Procedure: Pig skin gelatin (5.5 g, 300 Bloom, Aldrich) was placed in water (50 mL), acidified with glutamic acid or 2N HCI to pH 2.5-3.5, and then heated with stirring for a few minutes in a water bath. The central tube of the diffusion cell (length 3 cm, diameter 2.5 cm, closed at one end) was filled with the hot gelatin and left at room temperature for 24 h in a vertical position until gelation occurred. The L-shaped tubes (vertical section: 18 cm, horizontal section: 6 cm, standard ground joints) were then attached to the central gelatin tube, filled with the aqueous solutions ( 0 . 1 3 3 ~CaCI, or 0 . 0 8 ~Na,HPO,/ 0 . 0 2 7 ~KF), and closed with ground glass stoppers. Both solutions were previously adjusted to the physiological pH of 7.4 with cc,a,a-tris(hydroxymethyl)methylamine/HCI. The diffusion experiments were carried out at 25 “C (thermostat). At the end of the experiment, the gelatin plug was pressed out of the central tube and cut into sections corresponding to the Liesegang bands. To isolate the solids, these sections were washed several times with hot water and centrifuged.  K. T. Wilke, J. Bohm, Krutullziichtung, Harri Deutsch, Frankfurt/Main, 1988.  R. E. Liesegang, Z . Phys. Chem. 1914, 88, 1.  K. Sudarsanan, R. A. Young, Acta Crystullogr. Sect. B 1978, 34, 1401.  P. E. Mackie, R. A. Young, J. Appl. Crystullogr. 1973, 6, 26.  The Science and Technology of Gelatin (Eds.: A. G. Ward, A. Courts), Academic Press, London, 1977. [lo] W. Babel, Chem. Unserer Zeit, 1996, 30, 86. [Ill B. B. Mandelbrot, Die,fraktale Geometrie der Natur, Birkhauser, Basel, 1991. [I21 “Selbstahnliche Verzweigungen” und “Simulation von elektrischen Feldlinien im Bereich selbstorganisierter polarer Aggregate”: H. Dolhaine, unpublished.  D. A. Tomalia, A. M. Naylor, W. A. Goddard 111, Angew,. Chem. 1990, 102, 119; Angew. Chem. I n t . Ed. Engl. 1990,29, 138.  Liberation of the constitutional water from hydroxyapatite above 1250 ”C: P. W. Arnold, Trans. Faraday. Soc. 1950, 46, 1061. 1151 M. Okazaki, J. Takahashi, H. Kimura, J Osaka Univ.Dent. Srh. 1989,29,1989. [I61 “Kristallfeldberechnungen an Apatiten”, J. Brickmann, S. Hauptmann, unpublished . (171 “Messung von permanenten Dipolen an kleinen Aggregaten”, F. Laeri, unpublished. [I81 S. Mann in Inorganic Materials, (Eds.: D. W. Bruce, D. O’Hare), Wiley, Chichester, 1992. [I91 M. Bauer, W. E. Klee, Z . Kristallogr. 1993, 206, 15.  M. Braun. C. Jana, Chem. Phys. Lett. 1995, 245, 19. Prototypical Reagent for the Synthesis of Substituted Hydrazines** Uno Maeorg, Leif Grehn, and Ulf Ragnarsson” Dedicated to Professor Ivar Ugi on the occasion of his 65th birthday Substituted hydrazines are valuable precursors of heterocycles and components of many pharmaceuticals, agrochemicals, and dyestuffs. Direct alkylation of hydrazine normally gives rise to mixtures of mono-, di-, and trisubstituted products in which all the groups are located at the same nitrogen atoms.[’] This often makes purification troublesome. Although many [*] Dr. U. Ragnarsson, Dr. U. Maeorg, Dr. L. Grehn Department of Biochemistry University of Uppsala, Biomedical Center P.O. Box 576, S-75123 Uppsala (Sweden) Fax: Int code +I8552139 e-mail: firstname.lastname@example.org [**I We thank the Swedish Natural Science Research Council, the National Agency for Higher Education, the National Board for Industrial and Technical Development, and Astra Draco AB for their generous support of our programs. 05?0-0X33/96/3522-2626 $15.00 + ,2510 Angew. Chem. I n t . Ed. E d . 1996.35. N o . ??