How do proteins fold and why?

Published on October 1, 2007 Reviewed on April 12, 2022   44 min

A selection of talks on Cell Biology

Please wait while the transcript is being prepared...
How do proteins fold and why do they fold in that way? These questions have engaged the experimentalists and theoreticians with gathering intensity since about 1960, when Chris Anfinsen first showed that proteins are able to fold to their native structure all by themselves, without outside help.
Here's a slide of Anfinsen's classical experiment. He unfolded the protein ribonuclease A in chemical denaturating, that is with concentrated urea. Carefully reduced all four of the covalent disulfide bonds that stabilize the native structure. Actually, that was the hard part. Then he put the protein back into native conditions: no denaturing, normally oxygenated buffer and watched ribonuclease spontaneously reform its normal disulfides, refold to end and recover its native enzymatic activity, pretty exciting. In this case, however, the folding reaction went on a timescale of hours, limited not by the difficulty of spontaneous refolding, but by the slow rate for reforming the native disulfides. Later on we'll see that proteins fold an awful lot faster than that. On this basis, Anfinsen framed his famous Thermodynamic hypothesis, that protein folding, just like any other spontaneous chemical process, simply proceeds energetically downhill to its lowest free energy form, the native state.
Soon after, Cyrus Levinthal suggested the opposite view, that the native state is kinetically rather than thermodynamically determined. Levinthal noted that an unstructured polypeptide chain can take up an extremely large number of alternative conformations, so I've written the numbers down here. The number of conformations that a polypeptide chain can take up in principle is larger than the number of atoms in the universe. It would obviously take a very long time for it to search randomly through all of those possibilities for its lowest energy native state, but as it turns out proteins can fold in milliseconds. Levinthal argued that some kind of built in distinct pathway, predetermined pathway, must exist to conduct the chain to its target native form, which therefore may not be the lowest free energy state. We now know that in some sense that is true. Many native proteins can be spontaneously recruited into the even more stable amyloid structure. The protein form that produces such devastating conditions as Alzheimer's, Parkinson's, Huntington's disease in humans, other diseases in animals, mad cow, many others. Well, okay, but experimentalists have now repeated the Anfinsen experiment with literally hundreds of proteins, which can be unfolded and then will spontaneously refold to their native normal, native structure all by themselves without outside help. So, it's now universally agreed that the folding process is a thermodynamically downhill process and that it can go spontaneously without outside help, just as Anfinsen suggested. Another point, although protein folding can go very fast, it isn't as easy as it looks. Errors are made with spontaneous frequency. We now are aware that biology does provide a large cadre of ancillary systems, helper proteins that function to correct for errors in the folding process.