You are using two precipitation procedures that work differently. You are probably denaturing your protein with ammonium sulfate.

It is not a trivial question, indeed.

The conformational state of a protein in solution is an energetic compromise between protein-protein interactions and protein-solvent interactions to favor the most stable conformation(s). The free energy of transfer, ΔGtr, from water to the various solvents, is a good measure of the most probable conformational state in a specific solution. More negative is the value of the free energy difference between the protein in water and the one in solution (ΔGtr), more the solubility of the protein in solution is favored. For example, ΔGtr for the peptide group from water to urea 2 M is about – 70 (cal/mol), while that for the transfer of the peptide group to ethanol is + 1500 (cal/mol), which means that the peptide group is more soluble in urea solution than in ethanol. This explains why globular proteins are more soluble in urea or GdnHCl solutions (because more unfolded, namely, more prone to interact with the solution molecules) in respect to ethanol solutions. This is the quantitative aspect of the protein's behavior in solution, but the interaction mechanism capable of producing these behaviors requires different considerations.

Ethanol is a polar solvent that influences the conformation of proteins through its direct binding, changing significantly with the ethanol concentration and length of exposure and protein type (due to different composition and residues exposure). In general alcohols act as hydrogen bond donors whose binding to the polypeptide chain is stabilized by hydrophobic interactions where properties of alcohol binding sites have been derived from X-ray crystal structures of alcohol-protein complexes. It seems that at first the protein unfolds in ethanol, but it often refolds into rod-like alpha helix structures probably to bury peptide groups inside the compact helical rods, out of interaction with ethanol. Thus, interacting only with exposed non polar side chains. This suggests that alpha helix should be stable in ethanol. In fact, numerous alcohols are alpha-inductors.

Instead, ammonium sulfate is a very soluble inorganic salt that disassociates into ammonium and sulfate ions in aqueous solutions. It stabilizes protein structure and the protein solubility changes according to the salt concentration (that is, according to the ionic strength). At low salt concentrations (between 0 and about 0.5 M) the protein solubility increases, with increasing salt concentration, but when the salt concentration is further increased, the protein solubility decreases. We have the famous “salting in” and “salting out”. But ammonium and sulphate ions are member of the Hofmeister series where the effect of these ions is on the water, at least in part, by ‘making’ or ‘breaking’ bulk water structure, that is, the water molecules that act as cage around the protein. However, models suggest that this effect depend both upon direct ion–macromolecule interactions and interactions with water molecules in the first hydration shell (the water-cage) of the protein. Models also suggest that ions seem to modulate the ability of the various water layers in order to help out the first water layer solvating the protein surface. Therefore, the mechanisms of action of the two solutions are very different, though the behaviors in solution (the thermodynamic aspect) follow the same law.

I hope that these explanations, though a bit rough for necessity, may be useful.

I agree with the excellent comments of Giovanni and to try to summarize it I would like to say that the precipitation of any protein with Ammonium Sulphate or Ethanol is based on the range of different physico-chemical properties of each particular protein such as hydrophobic properties, isoelectric point, glycosylation, molecular mass, and many other characteristics.  

Dear Viktor

I must agree that in a few words you have done an excellent summary of what I have said. However my end was essentially didactic to try to clarify the doubts of our young colleague. Thanks for your appreciation, also because much of modern science forgets fundamental aspects of protein chemistry.

Giovanni

Thank you so much  Dr. Giovanni for sparing so much of your time to explain the things in detail.

Dear Himashu now I'll explain you why.

Biomedical research has had a real leap forward in the 60s, when numerous technologies were gradually transformed into scientific instruments able to make sophisticated quantitative measurements of chemical-physical parameters of living matter molecules, which until then were often very rough or nonexistent. In the field of protein chemistry I speak about machines for the protein sequence, X-ray, dichroism, fluorescence, UV spectroscopy, calorimetry, NMR, stopped-flow associated with various spectroscopies and so on. All this massive experimentally complex and sophisticated work, which had never been developed before then, gave a significant impetus to an ever deeper knowledge of living matter, up to the molecular level and beyond. These data, published in important scientific journals, are not always readily available (generally those before the 80's) on international databases since at that time their digitization was relatively imperfect, being had implemented through scanning processes that lose much of the information and therefore knowledge. In addition, only large publishing companies were able to bear their relative costs. This determines that the much of historical part of the information you requested resides in archives not easely consultable, because the search-system often does not read in the article body, but only title and authors (some time we find also the abstract). I think it's the duty of those who worked at that period to divulge it all. Names like J. Loeb (he was one of the first to speak about Hofmeister's series applied to proteins) but also others as Charles Tanford, Walter Kauzmann, Gregory Weber, Harold Edelhoch, John Edsall and many others, were the first to study the structure of globular proteins in the modern sense, subject today sometimes resumed and deepened by contemporary researchers (for IDPs we are still at the very beginning). However, the correct physico-chemical bases were fixed at that time. This is often ignored by young people who, by operating on sophisticated computer-controlled machines or highly automated analytical methods, do not deepen the fundamentals of what they use, because the machine thinks for them.

Good luck.