| Sign In to gain access to subscriptions and/or personal tools. |
Self-oriented Assembly of Nano-apatite Particles: a Subunit Mechanism for Building Biological Mineral CrystalsDivision of Oral Biology, Leeds Dental Institute, University of Leeds, Clarendon Way, Leeds University LS2 9LU, UK; C.Robinson{at}leeds.ac.uk Martin Taubman, Editor
Key Words: apatite • subunits • crystal-initiation • growth The organized precipitation of calcium phosphate and its crystallization into a calcium hydroxyapatite are the central events in the development of skeletal and dental tissues. To bring calcium phosphates out of solution and generate crystals of hydroxyapatite, it is necessary to exceed the solubility product of the precipitating phase in the mineralizing compartment. While the final product is an apatite of some description, the precise nature of this initial phase—and indeed transitional phases—is still a matter of debate. It may be apatitic, but amorphous material has also been detected (Posner, 1969). Such precipitation can be achieved by elevating the local concentration of calcium and/or phosphate ions. Removing water or reducing the dielectric constant of the solution, for example, with organic molecules will have a similar effect. The solubility product itself may also be reduced by the addition of other ions that may be incorporated in some way into the apatite structure; fluoride is an important case in point. Most published images of early biological hydroxyapatite crystals have been obtained by transmission electron microscopy (TEM). However, it occurred to us some years ago that processing for electron microscopy, and indeed most microscopy, involved procedures that would be likely to precipitate calcium phosphate from solution. These include removal of water (dehydration), fixation with organic fixatives, often staining with heavy metals (TEM), and embedding in hydrophobic (low dielectric) media. On this basis, many of the superb images of the earliest hydroxyapatite crystals, seen, for example, in dental enamel, could conceivably be artifact, especially at that point during tissue development where mineral was precipitating from solution. Such treatment could also transform inherently unstable precursor material into more stable apatite. That is not to say that crystals would not eventually appear in that location, but not necessarily at that time or with the observed morphology. The closeness of enamel crystals to the secretory ameloblast could, in these terms, be an illusion. With this in mind, some 20 years ago, together with John Weatherell, then Head of the Department of Oral Biology in Leeds, we gave a great deal of thought to the problem of imaging this stage of crystal development, without, however, the usual processes of dehydration, fixation, staining, and embedding. Discussions were often held on a Friday evening in the lounge bar of the Queens Hotel or the George, a public house close to the infirmary and frequented by generations of medical and dental students. Eventually, a freeze-etching technology was selected. This emerged from the internationally known Proctor Department of Food Science in Leeds, where it was used to study an even more difficult material, ice cream! During this process, the tissue (or ice cream!) was carefully frozen in liquid nitrogen (–198°C) and maintained in this state under vacuum while the sample was fractured with a histological knife. The knife was then repositioned over the fractured surface and its temperature lowered so that ice sublimed from the tissue onto the knife blade. This left a fractured surface of the tissue itself, unencumbered by ice. This fractured surface, still frozen and under vacuum, was then shadowed with gold or aluminum. After being shadowed, the tissue was dissolved away and the metal replica subjected to the high resolution available for TEM investigation.
What emerged in developing enamel, surprisingly at the time, was a series of globular structures (30–50 nm in diameter) of a width similar to that of mature enamel crystals, and arranged in a co-linear fashion but sometimes randomly—possibly cross-sections of the linear arrangements (Fig. 1a
We concluded that the globular structures were enamel subunits composed of matrix-protein and mineral ions, such that, at least initially, apatite crystals did not appear. The globular subunits formed collinear assemblies, in which degradation of the matrix then allowed the mineral ions in each subunit to precipitate. Fusion and crystallization of the co-linear subunits would then occur, perhaps simultaneously, and the elongated crystals characteristic of mature enamel would ultimately emerge. The presence during the secretory stage of specific matrix metalloproteases (Brookes et al., 1998; Hu et al., 2002), which degrade the matrix in a specific manner prior to a later wholesale destruction by a specific serine protease during maturation, adds some support to this. Subsequent growth in crystal thickness and width occurred in the fluid that had replaced the supportive matrix (Robinson et al., 1983, 2003, 2004).
This idea was not particularly well-received—especially by some electron microscopists. Little was done with the concept until atomic force microscopy (AFM) became available to the department, through collaboration with the Department of Physics and Astronomy. Work by previous and then-current PhD students revealed both morphologically and chemically (Fig. 2a
At about this time, data supporting the fabrication of crystals by aggregation of nanoscale subunits were emerging from other areas of research. The formation of zinc sulphide crystals by bacterial metabolism appeared to occur by fusion of subunits units prior to crystal expansion by classic growth. The small size of such units permitted ions to be re-oriented such that one crystal was obtained with little in the way of crystallographic discontinuity (Banfield et al., 2000). Even phase transformations from precursor material to final phase can be dramatically effected by the fusion of nanoscale particles (Gilbert et al., 2003). Some hitherto-existing data on biological apatite may now be explained by a growth mechanism of this sort. Stabilization of precursor amorphous or short-range-order phases would fit in with the earlier concept that apatite crystals were preceded by a transient amorphous phase (Posner, 1969). While the instability of such phases and their low occurrence in mature tissue cast some doubt over this idea, stabilization of such precursor units by protein prior to assembly, followed by degradation of the stabilizing matrix and subsequent crystallization, would provide a reasonable explanation (Höhling et al., 1971). Such a process would provide a workable mechanism for the biological appearance of single crystals of a huge variety of shapes and sizes, not seen in, for example, geological deposits or in laboratory preparations. Subunits of stabilized amorphous precursors could be assembled into any shape or form, either on or in templates, with crystallization then to be achieved by degradation of the stabilizing organic material. In enamel, this is taken to extremes, where the template itself, as well as the stabilizing material, is removed for juxtaposition of the final crystals formed. While the evidence may not be completely incontrovertible, the problem of artifact in the processing of supersaturated solutions for imaging remains unsolved. Subunit assembly of crystals, particularly those of biological origin, is a concept capable of explaining numerous phenomena in biological mineralization, and thus merits serious consideration.
Received for publication February 1, 2007. Revision received March 14, 2007. Accepted for publication March 27, 2007. REFERENCES
Journal of Dental Research, Vol. 86, No. 8,
677-679 (2007) This article has been cited by other articles:
|
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
 |
|
|
 |
|
Sign In to gain access to subscriptions and/or personal tools.
