An important part of the answer appears in a report in the latest issue of Science.
The puzzle of enamel formation centers on amelogenin, a protein secreted by cells in gum tissue called ameloblasts. Amelogenin's closest analogue in the human body is collagen, the protein that guides the formation of mineral in bone.
Unlike collagen - which remains an essential part of bone structure, helping it to heal after fractures - amelogenin degrades and disappears during the process of enamel mineral growth, or biomineralization.
Because its transient role makes it hard to study, amelogenin has not been well understood despite investigations stretching over many years, said Janet Moradian-Oldak, a professor in the University of Southern California School of Dentistry and the paper's lead author. USC Postdoctoral research associate Chang Du contributed to the paper.
By its nature, amelogenin cannot form a lasting platform or scaffold for enamel development. The question is: can a protein with a very short life span provide a reliable structure for biomineralization?
The answer is yes, according Oldak and her team, whose discovery was serendipitous.
The researchers, in collaboration with Giuseppe Falini at the Universita di Bologna in Italy, had been attempting to study amelogenin by crystallizing it. Crystallography is a traditional method of exploring molecular structure.
After a year, the researchers were unable to obtain amelogenin crystals. Instead, their efforts produced what looked, under a microscope, like long, fettucine-like fibers. The fibers consisted of chains of amelogenin nanospheres: tiny balls of amelogenin molecules.
Oldak called the fibers "microribbons." She was struck by the similarity in structure between the ribbons and the calcium hydroxy apatite crystals that make up the bulk of enamel.
The apatite crystals in enamel are highly elongated, with a length to width ratio as high as 1000. The microribbons' chains of amelogenin nanospheres were also elongated.
Oldak wondered if the microribbons might be the scaffold for biomineralization she had been looking for. It was a leap that her scientific training told her was excessively optimistic. "I think what you need is a bit of imagination to be able to link these things," she said.
When the researchers mineralized the ribbons and dipped them into calcium phosphate solution, they obtained aligned and organized apatite crystals similar to those found in enamel. The microribbons had functioned perfectly as a scaffold.
The work was done in vitro, but studies of the literature turned up observations of similar structures in vivo, including a report of "beaded rows" of amelogenin nanospheres alongside developing crystals in enamel.
"We demonstrate that amelogenin protein has a strong tendency to assemble in linear arrays of nanospheres, and we propose that this property is a key to its function as a scaffolding protein during the early stage of enamel mineralization," the researchers wrote.
The finding unlocks one mystery of enamel formation and may have long-term applications.
Growing artificial enamel is a decades-old goal among researchers in dental science and, more generally, in the medical-device community. As a filling material, enamel has the potential to outperform less durable substances such as composites and silver-mercury alloys. Medical-device developers are constantly searching for durable natural materials to use in place of titanium and plastic parts.
"The in vitro self-assembly system of Du et al will be a useful guide to the development of biomimetic structures," wrote Arthur Veis, professor of cell and molecular biology at Northwestern University in the perspective companion to the Science paper.
"Others have shown that minerals can develop within protein and synthetic polypeptide gels, but a scaffold was necessary to provide long-range order. In contrast, Du et al show that the self-assembly of the amelogenin nanospheres, and their further assembly into nanosphere arrays, forms its own scaffold that can direct the alignment of the mineral crystallites."
The research was supported by the National Institute of Dental and Craniofacial Research, part of the National Institutes of Health.