Protein folding problem solved?

Rick Venable rvenable at deimos.cber.nih.gov
Sat Jun 10 16:42:38 EST 1995


> >rvenable at deimos.cber.nih.gov (Rick Venable) writes:
> >> Given that the crystal is a somewhat artificial environment,

> In article <3rcbb4$cee at lyra.csx.cam.ac.uk>,
> Simon Brocklehurst (Bioc) <smb18 at mole.bio.cam.ac.uk> wrote:
> >  Sorry, can't let you get away with saying that - too many people 
> >will believe what you're saying.  Crystal structures of proteins are,
> >for the most part, excellent models for the structure in solution.

On 10 Jun 1995 15:30:58 GMT murphy kenneth p added:
> To which one might add that proteins in crystals practically *are* in solution
> since the percentage solvent is so high.

I'll grant that in most cases the interior hydrophobic packing is preserved
for protein structures determined by both NMR and x-ray, there have been one or
two exceptions to this rule, in which the solution structure is very different.
Also, there's a protein which has been crystallized over a wide variety of
conditions (pH, ionic strength, buffer medium, etc.) in which the outer folding
varies quite a bit (but not the hydrophobic core).  The protein in solution
is not packed in a regular lattice, and has greater flexibility; it can refold.

So, while I'm aware that protein crystals are typically 50 percent water, and
tend not to change the packing of hydrophobic regions on solvation, there are
a few situations where differences have been noted.  The crystal structure
is a good model of the solution structure with regard to the core structure,
but surface features can change considerably.  Some research groups working
on insulin were lead astray by differences in some external packing of
secondary structure domains; various residues were incorrectly thought to
be either exposed or buried.

All I'm suggesting is that one shouldn't overinterpret a crystal structure
as *the* solution structure, but it is generally a good model of a low energy
conformation easily accessible from the distribution of conformations sampled
in solution.  Keep in mind that surface features can be fairly mobile, and
that re-folding is rare, but has been shown to occur.

--
Rick Venable                  =====\     |=|    "Eschew Obfuscation"
FDA/CBER Biophysics Lab       |____/     |=|
Bethesda, MD  U.S.A.          |   \  /   |=|  / Not an official statement \
rvenable at deimos.cber.nih.gov       \/    |=|  \  or position of the FDA.  /



More information about the Molmodel mailing list