Proteins are polymers (or "chains") of amino acids, usually from dozens to hundreds of amino acids long. If DNA is the code for life, then proteins are life: they provide structure (collagen in the skin, for example), mobility (in muscle fibers), and as enzymes, they keep the engines of life going. Remarkably, as Christian Anfinsen showed many decades ago, the unique sequence of amino acids of each protein encodes its three-dimensional structure. Although chaperonins (other proteins!) have evolved to help misfolded or large unwieldy proteins escape from incorrectly folded "traps", proteins are not meticulously assembled - they can do the assembly job by-and-large by themselves.

How proteins fold and misfold is of great medical interest, as many diseases originate from proteins not doing their job correctly, often because of a mutation (a single amino acid that has been substituted by a different one because of an error in the DNA coding for the protein). Young-onset Alzheimer's disease and sickle cell anemia are but two of thousands of examples of such protein diseases. For this reason, how proteins fold - or fail to fold - is a subject of great current interest. The forces that hold proteins together or structure them, such as hydrophobicity, hydrogen bonding, van der Waals contacts, or salt bridges, are less simple and less directional than chemical bonds, and have complicated the development of protein folding models and experiments to test them.

Some of these diseases involve the aggregation of many proteins into sheet-like structures (beta sheets), first proposed by Linus Pauling in the 1930s. It is usually thought that these sheets require nucleation (a certain minimum size must form before the continue to grow spontaneously). Whether nucleation is needed or not, we have shown in a paper in J. Am. Chem. Soc. that pieces of this sheet-like structure exist in unfolded proteins at elevated temperature (say a 102 degree fever!), predisposing even the individual protein towards the formation of beta sheets.

Many years ago, Cyrus Levinthal stated that proteins cannot fold by randomly trying out all possible conformations, or shapes. It would simply take too long. In fact, proteins turn out to be extremely efficient at folding. We have engineered proteins in our lab that can fold in a matter of microseconds. While protein folding in textbooks has been treated as an ordinary chemical reaction, with unfolded and folded states separated by a free energy barrier, it appears that some proteins can fold downhill without such barriers. It appears that the mutations necessary to improve folding and eliminate the barrier are often detrimental to the function of the protein. Evolution has to simultaneously optimize both a protein's function and its ability to fold, and a compromise may be necessary between these two aspects of a protein.

In our experiments, we unfold wild-type and genetically engineered proteins by cooling to low temperatures (cold denaturation). Folding is then initiated by laser-jumping the temperature of the aqueous protein sample back up in a few nanoseconds. Alternatively, temperature-jump initiated unfolding may also be studied by taking folded proteins and jumping them up. Fluorescence lifetime detection tells us about the environment near amino acids such as tryptophan, the time evolution of fluorescence wavelength tells us about solvation, and the time evolution of infrared spectra tells us about the formation of hydrogen bonds and secondary structure. More details can be found in the clickable highlights of some recent publications shown below, or in the papers in the reference section of this website.