When dentists and their patients think of yeast and the mouth, the first thing that generally comes to mind is Candida albicans- the culprit that leads to burning, pasty-white, foul-tasting thrush. But in the laboratory of CCMB researchers, Dr. Malcolm Snead and Dr. Michael Paine, yeast and teeth have found a way to live in a more harmonious coexistence. In their lab, Snead and Paine use yeast to help them understand how tooth enamel is formed.

But why use a single-cell organism like yeast to study a complicated, mineralized tissue like enamel? Well, like most biological workings of the cell, the formation of enamel is difficult, in fact impossible, to observe with the naked eye. To overcome this barrier, scientists have constantly struggled to find tools that allow the inner activities of cells to be examined. One such endeavor led to the development of the yeast two-hybrid experimental model. This system allows scientists to gather information about the nature of interaction between proteins both inside and outside a cell. In studies conducted by both Snead and Paine, the yeast two-hybrid system has enabled an exploration of relationships between the numerous proteins that contribute to tooth enamel formation.

The scientists who originally devised the yeast two-hybrid system took advantage of a biochemical event that occurs in bacteria, the conversion of a special sugar by an enzyme called b-galactosidase into a dense blue precipitate. The protein that regulates the production of the sugar-converting enzyme is called the GAL4 complex. GAL4 has two components, a DNA binding domain and an activation domain. While these domains are contiguous in their natural form, biologists can separate them physically by cutting the DNA that encodes the GAL4 protein. Without a functional GAL4 complex the cell won’t make b-galactosidase and won’t convert the sugar substrate into a blue precipitate. The beauty of the yeast two-hybrid system is that the two domains can be cloned and used in recombinant DNA techniques to investigate cell processes.

Of course, the ultimate goal is to capitalize on this observable, biochemical phenomena to demonstrate whether proteins or specific parts of known proteins interact. To accomplish this, scientists synthesize hybrid proteins. One hybrid protein is designed to tether the genetic coding information of the GAL4 DNA binding domain to the coding information of the protein of interest. The other hybrid protein is engineered to connect the genetic information encoding the GAL4 activation domain to the genetic information encoding a second protein of interest. These recombinant DNA procedures result in the generation of two new proteins with hybrid features of the GAL4 system and the potentially interacting proteins under investigation. 

The model also utilizes other aspects of recombinant DNA technology to enhance the opportunity to detect important protein interactions. For example, the hybrid DNA constructs also contain nutrient selection markers, such as a gene that encodes for an important amino acid’s synthesis. These elaborate DNA constructs are then put into yeast so that the yeast will express the new gene and produce the hybrid protein. The yeast, in turn, are plated on media that lacks certain amino acids so that only those yeast containing the recombinant DNA will live. 

If the hybrid proteins derived from the hybrid DNA constructs physically interact within the yeast, the GAL4 DNA binding domain and the GAL4 activation domain will be brought into approximation. When the two domains are approximated the ability to activate gene expression is restored again permitting the production of b-galactosidase and thus a mechanism for converting the sugar substrate into something blue. If the [yeast] colony is white then there’s no activity, no interaction. If they’re blue it tells you that the two proteins of interest have an interaction, explained Paine. 

While Paine was a post-doctoral fellow in Snead’s research group, he used the yeast two-hybrid system to determine whether enamel matrix proteins such as amelogenin, ameloblastin and enamelin interacted with each other or with themselves. Each of these proteins is present in the enamel matrix and play an important role in the orderly formation of enamel’s hydroxyapatite crystals or rods. Because amelogenin is over 90% of the enamel matrix, we want to know how it dictates the architecture of the rods, said Paine. 

Biochemical data generated by other scientists at the CCMB first suggested that the amelogenin protein self-assembled. What Paine and Snead demonstrated via the use of the two-hybrid system was that amelogenin physically interacts with itself through specific protein regions or domains. Our whole thrust was to find the domains that were important in this assembly, said Paine. Snead and Paine’s strategy was to delete different regions of the DNA sequence from the amelogenin gene and determine the effect on protein-protein interactions. The resultant amelogenin proteins were shortened versions of the original. We were simply eliminating the bulk of the gene so that we could look at the part of the protein that would have functional activity or binding activity, said Paine. What they discovered was that specific areas at the ends of the amelogenin protein (the carboxyl and amino termini) were critical for self-assembly.

The yeast two-hybrid system does have limitations. The nature of the hybrid proteins themselves is problematic, Paine explained, since it is possible that important functional domains could be blocked or otherwise compromised. Native 3-D protein configurations may be influenced as well. What a scientist may determine to be a significant interaction between two proteins with the yeast two-hybrid system may not translate to a significant relationship in a real organism with real teeth. Thus, claims that particular proteins interact must withstand further scientific scrutiny. 

Snead and Paine have fruitful collaborations with CCMB biochemists, Dr. Alan Fincham and Dr. Janet Oldak. Fincham and Oldak have used Snead and Paine’s various amelogenin deletion constructs to make recombinant amelogenin proteins. Using atomic force microscopy, the biochemists have observed amelogenin assemble in the presence or absence of critical interacting domains. Under normal conditions, amelogenin molecules come together and form sphere-like structures called nanospheres. You can explain the different nanosphere assemblies by the disruption of these amelogenin assembly domains. If you disrupt the amino end you get different nanosphere assembly to when you disrupt the carboxyl end of the protein, explained Paine. 

The most convincing test of biological significance comes from testing these various amelogenin constructs in animals. To this end, Paine and members from Snead’s research group have taken the amelogenin deletion constructs and inserted them into the mice. The enamel in these transgenic mice appears to be modified and has been observed with both scanning and transmission electron microscopy (SEM and TEM). What is normally a tight latticework of interwoven hydroxyapatite crystals looks almost as if it has been passed over a flame. As such the borders between the lattice layers appears blurred. It’s as if there’s fusion amongst crystallites, said Paine, and it’s more pronounced in the amino terminus deletions than in the carboxyl terminus deletions.

Paine suggests that the results from the yeast two-hybrid system haven’t changed theories about the process of biomineralization as much as it has strengthened the proposed role proteins like amelogenin play in the process. The way amelogenin assembles, the way it’s degraded, how it forms around the crystallites, how it dictates the direction in which the crystals can grow, all of these things emphasize that amelogenin isn’t a passive player. commented Paine.

A selection of recent publications:

  • Moradian-Oldak J, Paine ML, Lei YP, Fincham AG, Snead ML. Self-assembly properties of recombinant engineered amelogenin proteins analyzed by dynamic light scattering and atomic force microscopy. J Struct Bio.2000 Jul;131(1):27-37.

  • Paine CT, Paine ML, Luo W, Okamoto CT, Lyngstadaas SP, Snead ML. A tuftelin-interacting protein (TIP39) localizes to the apical secretory pole of mouse ameloblasts. J Biol Chem. 2000 Jul 21; 275(29): 22284-92.

  • Paine ML, Krebsbach PH, Chen LS, Paine CT, Yamada Y, Deutsch D, Snead ML. Protein-to-protein interactions: criteria defining the assembly of the enamel organic matrix. J Dent Res. 1998 Mar; 77(3): 496-502.

  • Paine ML, Snead ML. Protein interactions during assembly of the enamel organic extracellular matrix. J Bone Miner Res. 1997 Feb; 12(2): 221-7.

  • Paine ML, Deutsch D, Snead ML. Carboxyl-region of tuftelin mediates self-assembly. Connect Tissue Res. 1996; 35(1-4): 157-61.

 

 

 

 

HOME | Overview | Founder | Director | Faculty | Administration Staff | Laboratory Staff | Publications
 Research Grants | CBY | T32 | Seminar Schedule | Resources & Facilities | Web Resources

Last Updated: 04/19/07