Experimental determination and thermodynamic modeling of the osmotic pressure

ALVES, KELLY; NERLI, BIBIANA; PESSOA FILHO, PEDRO DE ALCANTARA

Buzios

Conferencia; 22nd International Conference on Chemical Thermodynamics (ICCT); 2012

Organizing Committee for The 22nd International Conference on Chemical Thermodynamics

The osmotic pressure is one of the key properties for the understanding of the behavior of proteins in aqueous solutions. It allows a direct assessment of the non-ideality of such solutions; moreover, the value of the osmotic second virial coefficient can be related to the possibility of obtaining crystals, instead of amorphous solids, in precipitation and crystallization operations. Nevertheless, experimental information on the osmotic pressure of protein solutions is scarce in open literature. In this work, a study on the behavior of bovine serum albumin in aqueous solutions containing co-solvents is presented. This protein was considered not only due to its commercial value, but also due to its scientific importance, related to its wide use as a model protein in downstream processing research. Aqueous solutions of bovine serum albumin containing as co-solvents either polyethylene glycol (of different molar masses) or sodium sulfate were investigated. Experiments were conducted at pH 4.0, 5.0 and 6.0 (i.e., the vicinity of the protein isoelectric point); protein mass fractions up to 0.05 were investigated. Experimental investigations comprise osmotic pressure measurements (conducted in a membrane osmometer), fluorescence measurements and circular dichroism measurements. Protein samples were previously desalted in ultra-centrifugation tubes with membranes of cut-off size of 5 kDa. The osmotic pressure was measured against protein-free reference solutions containing the same co-solvent at the same concentrations. For the thermodynamic modeling of experimental osmotic pressure data, two approaches were considered: the osmotic virial equation truncated at the second coefficient, and the coupling of an adhesive hard sphere term (B. Barboy, J. Chem. Phys., 61, 3194-3196, 1974) with a zero-order perturbation term. The results can be summarized as follows: ? The presence of small amounts of salts and low-molecular mass contaminants in the protein sample may have a large influence on the measured osmotic pressure. Failure of conducting this previous step may lead to falsely high measured osmotic pressures.The presence of small amounts of salts and low-molecular mass contaminants in the protein sample may have a large influence on the measured osmotic pressure. Failure of conducting this previous step may lead to falsely high measured osmotic pressures. ? The osmotic pressure showed a definite behavior concerning the presence of cosolvents: they lower the osmotic pressure for the same protein concentration; the higher is the co-solvent concentration, the lower is the osmotic pressure. This behavior results in more negative osmotic second virial coefficients, which are related to a stronger attraction between protein molecules and to the decrease of protein solubility that occurs by the addition of these substances (?salting-out?).The osmotic pressure showed a definite behavior concerning the presence of cosolvents: they lower the osmotic pressure for the same protein concentration; the higher is the co-solvent concentration, the lower is the osmotic pressure. This behavior results in more negative osmotic second virial coefficients, which are related to a stronger attraction between protein molecules and to the decrease of protein solubility that occurs by the addition of these substances (?salting-out?). ? The spectra of fluorescence showed that there is no significant change in the accessibility of the tryptophan residues in the protein molecule. The spectra of circular dichroism also showed that the protein secondary structure (á-helixesThe spectra of fluorescence showed that there is no significant change in the accessibility of the tryptophan residues in the protein molecule. The spectra of circular dichroism also showed that the protein secondary structure (á-helixesá-helixes and â-sheets) does not undergo significant changes. These results show that these co-solvents do not alter the protein structure, and hence the hypotheses behind the interpretation of the experimental data hold. The use of the proposed equation resulted in a better agreement with the experimental data than the virial equation. As both equations have the same number of adjustable parameters, this better performance is related to the fact that the hard-sphere term used (adhesive) represents the behavior of dilute solutions more adequately than the ideal one (implicit in the virial expansion). (implicit in the virial expansion). â-sheets) does not undergo significant changes. These results show that these co-solvents do not alter the protein structure, and hence the hypotheses behind the interpretation of the experimental data hold. The use of the proposed equation resulted in a better agreement with the experimental data than the virial equation. As both equations have the same number of adjustable parameters, this better performance is related to the fact that the hard-sphere term used (adhesive) represents the behavior of dilute solutions more adequately than the ideal one (implicit in the virial expansion). (implicit in the virial expansion).