Parameters affecting the efficiency of affinity-based reversed micellar extraction and separation (ARMES) in glycoprotein purification.

Abstract

Affinity-based reversed micellar extraction and separation (ARMES) is an effective method for purifying both low and high molecular weight glycoproteins via liquid-liquid extraction. A range of extraction conditions were examined to gain insight into the mechanism of ARMES. Concanavalin A (Con A) was used as the model affinity ligand to bind soybean peroxidase (SBP) and beta-galactosidase as model glycoproteins. Factorial design was used to investigate the effect of various system variables on the extraction of SBP via ARMES. A quadratic model described the systems well, resulting in a standard deviation of 7% between calculated and experimental extraction efficiencies. Sensitivity analysis suggested that the key criteria in ARMES were the NaCl concentration and pH of the aqueous feed phase. Extraction of both glycoproteins decreased above pH 7 but fell to zero only at pH values significantly above the pI of the model glycoproteins and the Con A affinity ligand. It is proposed that the complex of the affinity lectin with the glycoprotein results in a sufficiently hydrophobic species that can be extracted into a reversed micellar organic phase even at pH's far above the pI's of the individual proteins that comprise the complex. This finding has practical considerations for the use of ARMES in the resolution and purification of protein glycoforms.