Dear Rongjin,
In addition to what the others said, I would do the following:
1) check your apo-structure to make sure the putative ligand binding is
not blocked by a crystal contact and also check that there is a channel
leading to the binding site. You may not get hits in your
cocrystallization experiments because the ligands interfere with a
crystal contact. If the binding site is blocked, you need a completely
different crystal form. Best would then be to do a de novo screening in
the presence of the most potent ligand.
2) get independent evidence that your ligand does bind. This can be an
(enzyme) assay, NMR, biacore, ITC etc. If you are lucky, you may be able
to also check whether compounds of your crystallization buffer or DMSO
are interfering with binding.
3) Take the most soluble and most potent compound from (2) and soak in
the highest concentration you can achieve, which may be as high as
0.5-1.0 M. First try to dissolve the compound in water and only if that
does not work, try ethanol or DMSO.
4) Try to soak crystals grown under very different conditions. We
recently had a case where we could not soak-in compounds in the presence
of PEG, but high-salt was no problem. I know cases where the reverse was
true, where high salt was preventing binding.
Good luck!
Herman
________________________________
From: CCP4 bulletin board [mailto:ccp...@jiscmail.ac.uk] On
Behalf Of Rongjin Guan
Sent: Thursday, November 12, 2009 10:57 PM
To: CCP4BB@JISCMAIL.AC.UK
Subject: [ccp4bb] small molecule soaking screening
Hi All
Sorry that this is a non-ccp4 question, but I hope I can get
some good
suggestions from the community.
We have protein crystals under various conditons and want to
soak them
with different potential inhibitors. Most of inhibitors have
very small
molecular weights (200-300), so it become a problem how to
detect if the
small compounds have been soaked into the crystals or not.
(co-crystallization experiments yielded no hits so far, though
the free
form is easy to be crysatllized under many conditions)
We pay $500/day for local X-ray facility access, so we wonder if
there are
some more efficient ways that allow us to know if the small
compounds
soaked in or not, without collecting a whole data set for MR.
We are also thinking if we can mix several compounds together
for soaking,
to reduce the combinations of soaking experiments with various
compounds
and crystals from various conditions. Is this practical, if some
of them have
pretty similar binding affinities to the protein?
All comments/suggestions are welcome.
Thank you
Rongjin Guan
