Thanks for the quick and instructive replies. This is my original question:



Could any one recommend some heavy atoms used for crystals grown in 0.1M tri-sodium citrate pH 5.6, 1M Lithium sulfate and 0.5M ammonium sulfate? I read from Hampton user guide of heavy atom kit that "high salt concentrations are not the ideal medium for heavy atom reactions with macromolecules", so is there any type of heavy atom we should avoid using?



And we plan to try the following methods after considering all the helpful suggestions:



1)	Try normal heavy atom soaking methods first with Hg, Pt, tungstate, and etc.

2)	To change the crystals, in steps, to another “heavy atom tolerable” mother liquor for soaking. In our case, we wanna replace the original salt with LiAc or sodium malonate, and the original buffer with MES. Gentle crosslinking could be used to stabilize the crystal in the new environment. 

3)	Try halide soaking using KI or KBr.



The original replies I received are as following. Thanks again for all the kind helps!



-------------------------



Tommi Kajander:



hi, if they diffract well maybe you can try 1 M KI or KBr quick soaks for phasing. ..others will  have better idea of heavy metals. you can probably adjust the pH a bit upwards if you want to try Pt/Hg/Au/Ag. at this pH or at aroudn 6 Pt might be better than Hg. try KPtCl4, 1-10 mM. Also test first that the crystals are stable in the solution when taken out of the their drop. if not increase the precipitant concentratin to find "stabilizing"conditions so you can e.g. do o/n soaks. also if you  can change the buffer from citrate to something else (which isnt a metal chelator) that probably would help. that would be the first thing to worry about. and perhaps reactions with ammonium.



you could try Pb, Sm, U, ..etc hard (carboxylate) ligand prefering metals or things like tungstate. but 0.1M citrate will be a problem probably.



-----------------------



Jan Abendroth:



reminds me of an old project of mine a number of years back. I tried to replace ammonium sulphate by Li-sulphate to avoid possible heavy metal ligation by ammonia. Li-sulphate at 2M turned out to be a cryo salt. I solved the structure by Se-Met MAD. However in hindsight, I think to remember that HgCl2 yielded in a derivative.



------------------------



Ronnie Berntsson:



I can recommend using a Halide soak (bromide or iodine). You could try to "exchange" part your salt with sodium bromide. Always use a bad looking crystal for the first soak though to see that it doesn't get damaged by the soak.



See for further reference:

Novel approach to phasing proteins: derivatization by short cryo-soaking with halides

Z. Dauter, M. Dauter and K. R. Rajashankar

Acta Cryst. (2000). D56, 232-237



----------------------



Jeffrey Kieft:



How critical is it that the crystals remain in the sulfate salts?  I ask because we recently had success with an RNA crystal that was very intractable to heavy atom soaks because of very high lithium sulfate growth conditions, and the fact that cryo-protection was best done by increasing the lithium sulfate to near saturation.  At that high sulfate concentration, many of our favorite heavy atoms (mostly divalent and trivalent cations since the crystals were RNA) are insoluble, and the high Li+ concentration competed off the heavy atoms.



What I did was move the crystals, in steps, from the 2M lithium sulfate into high concentration of lithium Acetate.  What this did was increase the solubility of the heavy atom cations, which still maintaining high ionic strength and offering cryoprotection.  Ultimately, what worked was 3M lithium acetate + 100 mM cobalt hexammine or iridium hexammine (yes, 100 mM trivalent!).  We got a nice derivative, and in fact the diffraction limit of the crystal increased by almost 0.5 Angstroms!   



This was with an RNA crystal, so your problems may be different, but perhaps something analogous might work, or at least you can get some ideas from our experience.



In short, if solubility of the heavy atoms is an issue, you may be able to move the crystals into an alternate high salt condition that will give better solubility, and then crank the heavy atom concentration up to compete with the salt.





Tommi Kajander’s reply to this post:



we were also once able to change 1.7 M Amm sulphate to about equal ionic strength of citrate, in a few concentration steps of mixtures (for other reasons, it was supposedly an inhibitor..).you probably can switch to some friendlier salt... plus you could combine that with gentle crosslinking

through vapour diffusion first, if needed etc. (although try without first.)



---------------



Artem Evdokimov:



If you have enough crystals to experiment with, just do a normal series of heavy atom derivatives, starting with good old favorites such as mercury and platinum. As an added touch of class you can try tungstate - it's like sulphate, but has fat W so it can be used probably just like a halide soak...



------------------



George Lountos:



You could try transferring your crystal into a solution in which you replace lithium sulfate and ammonium sulfate with sodium malonate. There is a paper describing this in the following reference. This can work as a really good substitute for the properties of Lithium sulfate or ammonium sulfate and also can be a very good cryoprotectant at higher concentrations. The big plus is that is it does not interfere with many heavy atoms.



Malonate: a versatile cryoprotectant and stabilizing solution for salt-grown macromolecular crystals



T. Holyoak, T. D. Fenn, M. A. Wilson, A. G. Moulin, D. Ringe and G. A. Petsko



Acta Cryst. (2003). D59, 2356-2358



------------------