An excellent review on improving diffraction that has not been mentioned
yet is: https://doi.org/10.1107/S0907444905032130
As for how often it happens? At my beamline we do see this fairly
infrequently, but often enough that it no longer surprises me. I suppose
that brings some comfort to the user, but not a lot.
Something important to do in these cases is take an exposure with 180
deg of rotation or more on a single image. This is because you might
have a salt crystal and happen to have an orientation where no hkls are
on the Ewald sphere. Doing the wide sweep will make sure any salt
reflections are observed. If you see a beautiful, symmetric pattern of
very bright spots, but none anywhere near the beamstop, then you've got
salt. Mind you some salts have unit cells as long as 10 A or more, but
again the wide sweep gives you an upper limit on the unit cell size and
therefore the unit cell volume and molecular weight. Disappointing to
be sure, but better to know that try to optimize it. So, in a way,
doing a 180-deg shot and still seeing no spots at all is a good sign.
Means you have a protein crystal.
Of course, it is possible your 0,0,1 reflections are there and you just
can't seem them because of the beamstop. But making the beamstop
smaller is probably not going to make you any happier.
As for how a protein crystal can not diffract at all? Yes, it is
difficult to imagine how a crystal lattice can even exist if the "atomic
displacements" are so large as to extinguish even the lowest-order
reflections. Those "displacements" need to be at least as big as a unit
cell or larger to do that. However, for visible light (5000 A
wavelength) the unit cell of a protein crystal is not very big at all,
and movements of a unit cell or more are still not enough to perturb an
optical photon enough to make the crystal start to turn brown from all
the little micro-cracks. So, it is possible to have no order in the
x-ray range and optical clarity in the visible. That said, it is still
hard to imagine how a crystal could _grow_ this way. It is probably
something that happens after the lattice forms. And that is good news.
How does it get that way? Yes, sitting on the bench for 3 weeks might
be why your protein is denaturing in the lattice. Remember, there is
nothing magical about a crystal lattice. Yes, they tend to hold things
in place, but if your protein falls apart after sitting in a tube on the
bench after 3 weeks then it is not surprising that it might do the same
inside a crystal. One of my user labs actually does this as a
pre-crystallization assay: leave it out on the bench for a few weeks,
and then run a gel filtration column again and see if its still good.
If not, then you need to work on stability before setting up trays.
So, the short answer is that optimization is key. It is very rare in
crystallography to get away with not having to do any optimization at
all, and in those situations you should really worry because your
competitor is also having an easy time. It is when things don't go well
that being a critically-thinking scientist is an advantage. And that, I
hope, is also good news.
Good luck!
-James Holton
MAD Scientist
On 8/14/2018 2:58 AM, Careina Edgooms wrote:
I got the most beautiful crystals I have ever seen and they don't
diffract at all. Not poor diffraction, NO diffraction. Anyone know why
this could be and how I can go about fixing it? I had three beautiful
crystals and not one diffracted. I did leave them in the drop for
about 3 weeks before harvesting and in liquid nitrogen for about a
month before diffracting. Could that be a factor? If I regrew more
beautiful crystals and diffracted straight away could that help?
Careina
------------------------------------------------------------------------
To unsubscribe from the CCP4BB list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=CCP4BB&A=1
########################################################################
To unsubscribe from the CCP4BB list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=CCP4BB&A=1
