Book of Abstracts: Albany 2007
June 19-23 2007
Disease-associated Mutations in Proteins: Is Major Conformational Change Required?
In many protein-based genetic diseases the key issue is that the solubility of a normally-soluble protein is substantially lowered -- often by a single point-mutation. A well known example is sickle-cell disease, in which substitution of a glutamic acid by valine in hemoglobin to give sickle-hemoglobin (HbS), results in the formation of a gel phase, which in turn causes the sickling of the red blood cell. The separation of the gel-phase in HbS occurs without a significant change in the conformation of the protein monomer.
We have shown in several cataract-linked mutants of human γD-crystallin (1 and references therein), that the solubility of the proteins in vitro is severely compromised as a result of the mutation. We hypothesize that also in vivo, the cataract forms directly as a result of the lowered solubility of the mutant protein. In fact, there is evidence to support our hypothesis: The R36S mutant of human γD-crystallin spontaneously crystallizes at very low concentration in vitro, and it has been shown that the crystalline deposits are responsible for the cataract (2) in vivo. Protein solubility is also compromised in the P23T mutant, but in this case the lowered solubility is governed effectively by the net hydrophobic protein-protein interactions -- just as in HbS. In yet another cataract-linked mutant, R14C, we have shown that the solubility is lowered largely as a result of the formation of intermolecular disulfide bonds. Interestingly, in all these cases the protein monomer remains essentially intact in structure and conformation in the condensed (i.e., insoluble) protein-phase, whether it is a gel, crystal, fiber, or amorphous solid. This fact has often been overlooked or ignored in the literature, and all protein-derived genetic diseases, in which condensed phases are formed, are invariably categorized as ?protein conformational diseases?. Using high-resolution x-ray crystal structure data on native human γD-crystallin (1.25Å) and its cataract-associated R58H mutant (1.15Å), we showed (3) that even though the equilibrium solubility data clearly indicated a compromised solubility for this mutant, the monomer conformation remained unaltered relative to the native protein. We, therefore, suggest that all these protein-based diseases are best described as "protein condensation diseases," in which the protein separates into an insoluble condensed phase, and many of these phase transformations occur without a concomitant change in the monomer conformation. Mounting evidence in the literature indicates that conformational change in a protein may not be a prerequisite for the formation of pathological condensed phases. This leads us to conclude that depending on the monomer conformation, each mutant might elicit a unique biological response in vivo rather than the typical "unfolded protein response". Based on our work and that of others, we propose that the rational design of drugs to inhibit the formation of condensed phases must take into consideration the precise mechanism by which condensation of the mutant protein occurs.
References and Footnotes
Ajay K. Pande
Department of Chemistry