Statistical thermodynamic models for the precipitation of globular proteins using nonionic polymers
Abstract
The use of non-ionic polymers as precipitating agents for the separation of enzymes (globular proteins) is becoming increasingly important because of its ability to preserve the activity of the enzymes (globular proteins). A statistical thermodynamic model for the prediction of solubilities as well as precipitation curves of globular proteins using non-ionic polymers is proposed. Unlike the existing models for protein precipitation, which are mainly based on volume exclusion, this model accounts for random coil conformation of the polymers as well as protein-polymer, polymer-solvent, hydrophobic and electrostatic interactions. This model employs the simplifying assumption of negligible amount of polymers in the precipitate phase. The protein-polymer interaction parameter $\chi\sb{s}$ was obtained by fitting the model to experimental data for one molecular weight of polymers and was found to be 0.132, 0.115, and 0.122 for the systems of Ovalbumin-Dextran, Lysozyme-Dextran, and Human Serum Albumin-Polyethylene Glycol, respectively. The model predictions agreed well with the experimental data for these systems for different molecular weights of polymers. Protein solubility was found to be very sensitive to protein-polymer interactions and increase with more favorable protein-polymer interactions (larger $\chi\sb{s}$), less favorable polymer-solvent interactions (larger Flory-Huggins parameter $\chi$), smaller size of protein molecules and lower molecular weights of polymers. In order to account for protein-protein interactions, a statistical mechanical formulation was employed to predict the solubility and phase diagram of globular proteins in non-ionic polymer solutions. This statistical mechanical formulation accounted for the presence of polymers in the solid phase. The free energy of globular proteins was calculated using a second order perturbation with the reference system of hard spheres. The solubility was found to be lower for higher polymer concentrations, larger protein size, higher molecular weights of polymers, weaker protein-polymer interactions, poorer solvent, higher ionic strengths and for pH values closer to the isoelectric point. The protein-polymer interaction parameter $\chi\sb{s}$, obtained by fitting the model to the experimental data, was found to be around zero for HSA-PEG system. The model predictions agreed fairly well with the experimental data for PEG of relatively higher molecular weights.
Degree
Ph.D.
Advisors
Narsimhan, Purdue University.
Subject Area
Agricultural engineering
Off-Campus Purdue Users:
To access this dissertation, please log in to our
proxy server.