I have only heard of a few cases of successful Se incorporation into
Cys, and none of them sounded like fun. Sounds like you have gotten a
few suggestions already. There is not much literature on this.
As for alternative solutions to your original question, I can tell you
the success or failure of MAD or SAD phasing depends primarily on the
signal-to-noise ratio of your data and how it compares to your anomalous
signal. A good rule of thumb is that you need I/sigma to be better than
the expected F/delta-F from anomalous scattering. This is the reciprocal
of your "Bijvoet ratio". You don't say if one of your Met residues is
the first one, but if it is you can seldom count on it, so let's say you
only have two. You can get a rough guess of what Bijvoet ratio you can
expect, and what I/sigma you will need using a little web calculator I made:
http://bl831.als.lbl.gov/xtalsize.html
360 residues is roughly 45 kDa, and pessimistically assuming the minimum
4 electrons from SeMet, with 2 sites per 44 kDa I get a Bijvoet ratio of
2%. This isn't all that bad. You need I/sigma > 50 to solve this
structure. Do your crystals typically give you a peak I/sigma of around
30? That is, in the lowest-angle bin? If so, you only need a
multiplicity of (50/30)^2 ~ 3x higher than you usually do to get good
enough signal. This might not be all that hard to do. On the other
hand, if your data collections typically give you I/sigma ~10, then you
need 25x more multiplicity than the dataset that gave you I/sigma=10.
That can also be done, but usually involves either very large crystals
or multiple crystals. If you can grow isomorphous crystals, then you
are golden. Just keep shooting and merging until it solves. You can
even do this with native SAD if you have enough time.
If you have a problem with isomorphism, then your are not alone. Your
best shot with current technologies is to be extra careful with your
harvesting and cryo-cooling protocols:
Warkentin et al. (2006) http://doi.org/10.1107/S0021889806037484
Farley et al. (2014) http://dx.doi.org/10.1107%2FS1399004714012310
Good luck!
-James Holton
MAD Scientist
On 6/21/2017 8:46 AM, Vito Calderone wrote:
I am working on a protein having 360 residues. In its sequence there are 3
Met and 5 free Cys.
I will need MAD to solve the structure since based on the sequence the
closest homologue has 20% identity.I suppose MR would be very unlikely to
work.so I would like to express a selenium derivative to exploit MAD.
Looking in the literature 1 Se-Met every 120 residues seems not to comply
the threshold to get a good anomalous signal. For this reason I would like
to exploit both Met and Cys so I would have 8 seleniums per 360 residues.
Could somenone suggest a reference to a protocol to express the double
mutant protein in NON auxotrophic strains of E. coli which you have
experienced working efficiently?
Thanks
