PURPOSE: The potential to estimate protein solubilities, with limited protein, by excluded-volume interactions was evaluated using polyethylene glycols (PEG) and recombinant bovine Somatotropin (rbSt). METHODS: Solutions of rbSt were prepared at concentrations significantly below saturation solubility. Subsequently, varying amounts of PEG were added to force protein precipitation. Following centrifugation, the protein concentration in the supernatant was assayed by reversed-phase HPLC, where a logarithmic relationship between solubility and % PEG was observed. RESULTS: An apparent protein solubility in the absence of PEG was determined by extrapolation and compared well with values measured by conventional approaches. Slopes of log solubility versus % PEG curves were consistent with excluded-volume principles and depended on the molecular weight of the PEG used. Furthermore, the precipitation process proved to be reversible, allowing for recovery of intact protein. Solubility-pH profiles obtained in the presence of PEG greatly reduced the quantities of protein needed and compared favorably with profiles in the absence of PEG. CONCLUSIONS: Thus, it appears feasible and practical, with certain limitations, to obtain solubility estimates of proteins by volume-exclusion methods with limited supplies of protein.
PURPOSE: The potential to estimate protein solubilities, with limited protein, by excluded-volume interactions was evaluated using polyethylene glycols (PEG) and recombinant bovineSomatotropin (rbSt). METHODS: Solutions of rbSt were prepared at concentrations significantly below saturation solubility. Subsequently, varying amounts of PEG were added to force protein precipitation. Following centrifugation, the protein concentration in the supernatant was assayed by reversed-phase HPLC, where a logarithmic relationship between solubility and % PEG was observed. RESULTS: An apparent protein solubility in the absence of PEG was determined by extrapolation and compared well with values measured by conventional approaches. Slopes of log solubility versus % PEG curves were consistent with excluded-volume principles and depended on the molecular weight of the PEG used. Furthermore, the precipitation process proved to be reversible, allowing for recovery of intact protein. Solubility-pH profiles obtained in the presence of PEG greatly reduced the quantities of protein needed and compared favorably with profiles in the absence of PEG. CONCLUSIONS: Thus, it appears feasible and practical, with certain limitations, to obtain solubility estimates of proteins by volume-exclusion methods with limited supplies of protein.