Assuming your protein is normally soluble when in the native state, the situation you are in is that the protein has not folded properly when overexpressed. The insoluble protein dissolved in 4M urea, but formed insoluble aggregates when the urea was removed. The best solution to your problem would be to find a way to overexpress the protein in soluble form, possibly by changing expression vectors. If that does not work, you will have to find a way to refold the protein.

For refolding, I would recommend increasing the urea concentration to 8M, or use 6 M guanidine HCl, to make sure the protein is completely unfolded and aggregation is minimized. Then, try using a protein refolding kit to identify the most promising conditions for refolding it. There are additives that help the protein to refold, such as arginine. Also, aggregation during refolding is reduced by lowering the protein concentration.

Since you had success binding your unfolded protein to the HisTrap column, I'd try to reduce the urea concentration SLOWLY while the protein is bound to the column. The idea is that the protein has time to refold without contact with other proteins, and hence without a chance to precipitate (this is similar to the "infinite dilution" hypothesis of GroEL/GroES chaperonin mechanism). You may need to suitably buffer the S-S/-SH system so that necessary disulphides can form, but others do not.

Another option would be to control hydrophobic interactions with a mild detergent like Tween or Triton, but it is always a lot of trouble to get rid of detergents once they are in the sample.

Of course the outcomes and feasibility of solubilizing your protein depends upon what kind of protein you are dealing with, but I believe as I have seen with the protein I study, that when your protein has precipitated, that is not the end. You can still solubilize precipitated protein.

I will give you an example. I know that my protein crystallizes well in 50 mM Citrate Buffer (pH = 5.55) and 100 mM NaCl and a certain amount of polyethene gylcol. So had the bright idea of exchanging out my protein that I finally purified in 20 mM Tris-HCl (pH = 8) and 400 mM NaCl. But doing so precipitated my protein. Of course when you crystallize a protein you want it to be soluble but the crystallization landscape is nextdoor to the precipitation zone.

If I exchange out the 400 mM NaCl to 50 mM NaCl in the same Tris-buffer (pH = 8), my protein eventually precipitates heavily. My protein can be resolubilized by spiking my protein with a couple 5 to 20 µL injections of 4 M NaCl. I note the initial volume of protein solution, how much salt was injected to figure out the concentration of NaCl that my protein is soluble. Then you stop. Sometimes I need more than 400 mM NaCl to resolubilize my protein, maybe 1 M NaCl but I will explain how that is not a problem.

If I exchange out the Tris-buffer for Citrate buffer such that the concentration of Citrate buffer is greater than the concentration of Tris-buffer then the pH dominates to that of acidic citrate and my protein eventually precipitates. That is a problem if I am too slow to setup crystallization conditions. If I spike my protein with a couple 5 to 20 µL injections of 1 M Tris-buffer (pH = 8) then my protein resolubilizes.

If I exchange out the original Tris-buffer and lower the salt concentration, spiking my protein with a couple injections of 4 M NaCl doesn't save my protein. If I also spike with a couple injections of 1 M Tris-HCl (pH = 8), my heavy precipitated protein goes into solution easily. Keep in mind that precipitated protein is reversible, at least in my observations.

Once you figure out the salt concentration and the concentration of buffer that maintains a proper pH by spiking your protein solution, you can easily re-concentrate your protein to get enough protein for analysis in the correct buffer.

From your observations, make the proper buffer, 0.20 micron filter it.

Put your spikeed protein solution that you resolubilized in a 20 mL Centrifugal concentrator. If the molecular weight of your protein is quite greater than 10,000 Daltons, then a 10,000 Dalton Molecular Weight Cutoff (MWCO) concentrator will totally work. I recommend the Vivaspin 20 mL centrifugal concentrators. Some other centrifugal concentrators are total garbage products that don't even concentrate your protein. I have totally lost good protein to some Bioworld jokes of a centrifugal concentrator.

Anyway, put your spiked protein into the centrifugal concentrator, and centrifuge it at 4,000 rpm at 4 centrigrade for however much time it takes. The buffer and stuff will flow-through and your protein will remain at the top of the concentrator. If your protein sticks to the filter you can resuspend it with a pipette. Waste the flow-through and keep adding the proper buffer, and your protein will eventually totally be in the desired buffer concentration and the concentration of your protein should be sufficient for your experiments.

Using these centrifugal concentrators is like at least 5 times faster than dialysis. If arginine is useful to you as proposed, 100 mM Arginine should be great.

Dear

some bellow point may help you

1- Perform dialysis in 4C

2-Add glycerol to dialysis buffer (final concentration 20%)

3- Use a cold buffer for dialysis

4- You can also perform "on column" refolding to remove urea but you should put your column on ice powder