Try changing the salt concentration, replacing NaCl with a less chaotropic salt such as ammonium sulfate or potassium glutamate, changing the pH, and/or lowering the protein concentration.

Can you give more details? For example, the pI of your protein, the concentration at which your protein started to precipitate, how did you concentrate your protein, etc.

You need to give more details as Son mentioned. Based on your description, You are trying to crystallize a protein and/or its complex with RNA. Is this a new protein or has been crystallized before and you are trying to co-crystallize it with RNA? Few things you can try,

1) 500 mM NaCl appears a  bit too high. I would keep

I totally agree with the above mentioned suggestions. Apart from that you may use Arginine and other additives, which increases solubility of protein.

Thanks for replies. I am working on Viral protein which has two functional domains. Both domains are involved in the Replication of Viral genome.

Theoretical pI of my protein is 8.86 and thus only I was using buffer at pH 7.5. If I decrease salt from 500 mM e.g to 250 mM, it precipitates. Even at this conc of salt 500mM,  it does so. So, I planned to increase glycerol to 20%..

 Conc of protein which I get is approx 2 mg/ml above this it ppts . size of the proetin is ~105 kDa. I use centricon's for concentrating the protein. Also, I keep the proteins as small aliquots of ~ 20 microlitre, Sometimes this happens that when I take the vial out and keep it at Ice, it ppts there also.

Structure of this protein is known in apo form, I am trying to replicate my experiment in similar conditions but in my case it is getting precipitated!!

Have you check the folding (by CD) and oligomeric state (GF or SEC-MALS) of your protein? Even if you are successful to concentrate you protein with high concentration of salt and glycerol, it is not recommended for crystallization!.

You can try to play with Arginine and other additives as suggested by NIlesh Aghera or you can change to other buffers like HEPES or MES etc. But you always have to characterize your protein before crystallization.

If all do not work, you should design different constructs or analogues. You can play with limited proteolysis or think about production of chaperons (Camelid nanobodies).

Good luck!

If the apo form of this protein has been solved then that should be your starting point. 

Are you using the same expression system that has been used before? 

compare the buffer composition that worked for apo form with your buffer.

If possible use the same buffer that worked for apo form.

Mix the protein with RNA just before you set up your crystallization.

When you take out the protein sample from freezer, do not put it on ice, instead quickly thaw it using your hands and it might work.

Good Luck

Venu

Make sure your DTT is fresh.

Also,  many nucleic acid binding proteins require nucleic acids in order to remain soluble. 

Thanks again for prompt replies.

As suggested by Son van, though I have not checked the folding using these biophysical assays, but I have checked the biochemical activity of my protein ( for one domain) . It shows it is active which suggests that it folds in a proper fold.

"High Salt" I agree is not a good option for Protein nucleic acid crystals, (also for protein in many cases), but at low salt conc, my protein seems very unhappy. But I guess high glycerol should be fine, it might slow down the crystallization as vaporization will be slowed down,Is there any other reason that high glycerol is not good for crystallization?

I used Tris buffer (pH 8) before, and currently I am using HEPES(ph 7.5). I can try different additives as suggested by you and Nilesh Aghera and different constructs.

As Venu asked, I am not using same expression system as literature suggests( for apo structure) and my buffer conditions are slightly different. I will try the one used by them.  Also, I will try to work according to the suggestions given by you all. I hope any of your suggestions  work. Thanks again!!

Changing the expression system could potentially affect the nature of protein due to possible changes in post translational modifications, disulphide bridges and folding etc to name a few. So you need to confirm the protein is natively folded and stable to carry crystallization trials. I would suggest again to compare every thing you are doing with the literature of apo protein to get a right direction. 

Good Luck again

Venu

Zwitterions don't interfere with ionic bonds so might want to try a bunch of those (TMAO, beta-alanine etc.). 

Your pI is weird. It may only be stable with a binding partner. When I say weird I mean that most proteins are about 5.5-6 and some are around 10. The ones around 10 are often quite difficult to crystallize. 8.8 is weird but maybe it's normal for viral proteins.

Try a thermal shift assay (https://en.wikipedia.org/wiki/Thermal_shift_assay) - you might find that your protein binds a metal ion or something and that will solve all your problems. A PhD student in a neighbouring group solved all her solubility problems doing that. Also a viral protein. I've attached a publication with details of the conditions we usually check. Try out conditions that increase the melting point of your protein. 

And you might want to consider re-cloning. Chopping a few residues off the N- or C-terminal can make a world of difference. 

Article Protein-ligand interactions investigated by thermal shift as...

Work at room temperature

Hi.. Did you manage to solve your problem? were you able to crystalize your protein?