Jacob,
The main reason why it is not common practice to saturate every crystal
with every heavy metal under the sun is radiation damage. X-ray
absorption increases very rapidly with atomic number (third power), so
on the order of 100 mM of "heavy atom" is usually enough to cut your
crystal's useful life in half. It doesn't matter if the heavy atom is
"in the lattice" or not. At 12 keV the photoelectron cascade crashes
through about 3 microns worth of organic matter before finally coming to
rest (Cole, 1969).
The program RADDOSE (Murray et al. 2004, 2005, Paithankar et al 2009,
2010) was written to calculate expected crystal lifetime given your
buffer concentrations, etc. For example, with a 50% solvent crystal,
250 mM iodide in the solvent channels is "dose doubling" (life
halving). That is, the number of photons you will get scattered into
spots before the crystal is dead will be half that which you would get
with no iodine. This is because the iodine-soaked crystal is absorbing
twice as much energy per incident photon, but scattering at pretty much
the same rate with or without the iodine. 500 mM iodine will cut the
"useful life" to 1/3 of what it would be with no iodine, and 1 M will
cut it to 1/5th. (250/([iodine]+250)). For Cs, 200 mM is the
dose-doubling buffer concentration. For Rb it is 1 M. For Cl, it is
2.4 M, and for fluoride there is no concentration that doubles the dose
because water absorbs more than fouoride. This is why I recommend using
"low-Z" buffers in crystallography whenever possible.
I have listed dose-doubling concentrations of a few common elements on
the last page of this document:
http://bl831.als.lbl.gov/damage_rates.pdf
and a fairly comprehensive "look up table" is here:
http://bl831.als.lbl.gov/~jamesh/pickup/dose_doubler_solc50.txt
The dose doubling concentration does depend on the wavelength. The
"worst" one at 1 A is tungsten (57 mM), but this is not all that
different from tantalum (62 mM) or mercury (74 mM), or gold (76 mM), or
uranium (105 mM). So, by and large ~100 mM concentration of anything
"heavy" will double the "dose per scattered photon". Note that this is
the concentration of the atoms, not molecules. The dose-doubling
concentration of Ta6Br12 clusters is 9.5 mM at 1 A.
As for the original poster's question, 20 mM iodide disrupting your
crystallization is actually a good sign that it's binding, but the bad
news is that it appears your crystals don't like that. Nevertheless it
could be worth a shot, you can estimate your expected Bijvoet ratio
assuming one site per molecule using this little web app I wrote:
http://bl831.als.lbl.gov/xtalsize.html
In general, a Bijvoet ratio of 3% or so is needed to solve a structure
(the current world record is 0.5% and lots of multiplicity). The above
web page will also tell you how many crystals you need if you type in
their size in all three dimensions. but this estimate assumes that you
don't have high concentrations of heavy metals in your solution! So, if
it says you can get away with one crystal but you know your have a
dose-doubling concentration of something, then you're going to need to
average data from two crystals, etc.
Good luck!
-James Holton
MAD Scientist
On 5/3/2012 8:29 AM, Jacob Keller wrote:
I have wondered for a long time now why it is not standard practice
for all crystallization protein stocks to contain either Br- or I-
ions instead of Cl-, even for cationic buffers like TRIS, which could
be titrated with HBr or HI to get in the 10+ mM range. Also, one could
use Cs or Rb for the cations (and titrate anionic buffers with the
respective hydroxides). What's there to lose? The gain is obviously
the possible anomalous signal (always helpful), and one might pick up
additional interesting and possibly physiologically-relevant halide or
alkali metal sites. Seems that structural genomics people might
standardize this into the pipeline as well, and thereby potentially
cut out SeMet protein production in many if not most cases.
JPK
On Thu, May 3, 2012 at 9:05 AM, Jan Abendroth <jan.abendr...@gmail.com
<mailto:jan.abendr...@gmail.com>> wrote:
Hi Rajesh,
it can be a bit all over the place:
For quick soaks, we typically use 500mM-1000mM. A good starting
point might be to simply replace the NaCl concentration in your
protein buffer. By some serendipity we also managed to solve a
structure by I/S SAD after a 1mM NaI soak. One iodide had found
its way into a nice binding pocket.
For co-crystallization, mostly 200mM should be fine.
Another approach could be to supplement your cryo buffer with
iodide, replacing NaCl. NaI is highly soluble in ethylene glycol.
Also see here: http://www.ncbi.nlm.nih.gov/pubmed/21359836
Good luck!
Cheers,
Jan
--
Jan Abendroth
Emerald Bio
Seattle / Bainbridge Island WA, USA
home: Jan.Abendroth_at_gmail.com <http://Jan.Abendroth_at_gmail.com>
work: JAbendroth_at_embios.com
http://www.emeraldbiostructures.com
On May 3, 2012, at 6:51 AM, Rajesh Kumar wrote:
Dear All,
I have very thin crystals but diffracting. I was not able to
handle them easily for iodide soak. I always lost the crystals
during manipulation and other big crystals obtained after
seeding doesn't even give any diffraction. I tried for
co-crystallizing with NaI. The crystals appear only up to 20 mM
in 1:2 (3ul drop of 1 ul protein and 2ul reservoir).
Is this concentration of iodide is enough for SAD data ( if it
had good incorporation) ?
I appreciate your help.
Thanks
Rajesh
--
*******************************************
Jacob Pearson Keller
Northwestern University
Medical Scientist Training Program
email: j-kell...@northwestern.edu <mailto:j-kell...@northwestern.edu>
*******************************************
