Using phase transitions to investigate the effect of salts on protein interactions

We have investigated liquid-liquid and solid-liquid phase separation of aqueous solutions of lysozyme. We have determined experimentally how the phase transition temperatures depend on protein concentration and the ionic composition of the solution. For a wide range of solution conditions, we find that the cloud-point temperature ${\mathit{T}}_{\mathrm{cloud}}$---which signals the onset of liquid-liquid phase separation---is 15--45 \ifmmode^\circ\else\textdegree\fi{}C below the crystallization temperature ${\mathit{T}}_{\mathrm{xtal}}$. This indicates that liquid-liquid phase separation occurs in a highly metastable solution. When a series of chloride, bromide, and sulfate salts are added to lysozyme, we find that ${\mathit{T}}_{\mathrm{}\mathrm{cloud}}$ varies by as much as 60 \ifmmode^\circ\else\textdegree\fi{}C over the salt concentration range of 0.2M to 1.5M. The precise change in ${\mathit{T}}_{\mathrm{cloud}}$ depends sensitively on the identities of both the cation and the anion of the added salt. The effect of salts on ${\mathit{T}}_{\mathrm{xtal}}$ is very similar to their effect on ${\mathit{T}}_{\mathrm{cloud}}$. The Derjaguin-Landau-Verwey-Overbeek theory for the interaction energy between charged spheres cannot account for our observations and indicates that hydration forces play an important role in protein interactions. \textcopyright{} 1996 The American Physical Society.