We had a zinc-finger containing protein that we were soaking with different heavy atom compounds.  It turns out KAu(CN)2 provided the best diffraction of several soaks.  We found out it was because the gold had replaced the zinc ion and was coordinated by the Cys and His's.  The lab nickednamed the protein Goldfinger!  Although we didn't have the problem with disulfides, perhaps a similar gold soak might help if you tried to crystallize the protein in the presence of TCEP as well.

HTH

----- Original Message ----- From: "Sue Roberts" <[log in to unmask]>
To: <[log in to unmask]>
Sent: Thursday, September 25, 2008 4:35 PM
Subject: losing zinc during crystallization

Hello Everyone

I've been trying to crystallize a zinc-containing enzyme for what
seems to me to be an eternity. The protein contains stoichiometric
zinc  (1 zinc/ protein monomer) when isolated and the zinc is required
for activity.  Each crystal we've obtained has lost the zinc and
contains a disulfide bond between two cysteine residues that should be
zinc ligands (based on structures of similar proteins).

We've tried crystalizing in the presence of reducing agents,
crystallizing with substrate analogs, and supplementing the
crystallization drops with zinc with no success (and combinations of
these approaches).  We've obtained a variety of crystals and
determined structures, but none contain any zinc.

Attempts to insert zinc into the crystal (zinc + reducting agent or
zinc alone) have not been successful.

Does anyone have any tricks to suggest that might help?

Thanks in advance.

Sue

Dr. Sue A. Roberts
Biochemistry & Molecular Biophysics
University of Arizona
520 621 8171
[log in to unmask]