NERLI Bibiana Beatriz
congresos y reuniones científicas
Response surface methodology for screening and optimization of significant process parameters in trypsin extraction using peg/sodium citrate systems.
PÉREZ, ROCÍO; NERLI, BIBIANA; TUBÍO, GISELA
Conferencia; International Conference on BioPartitioning and Purification.; 2011
The issue of bio-separation has gained importance as a result of growing need for large amounts of enzymes in quick and efficient production bio-processes. The use of aqueous biphasic systems (ATP) to obtain labile molecules such as proteins has responded according to current necessities . The conventional strategy of single factor optimization, by keeping the other factors at constant level, is time consuming and unable of detecting the true optimum. A multifactor optimization strategy allows the fast selection of the most significant parameters using PlakettBurman (P-B) design, the identification of the role of each of the components, the modeling whit regressions and the localization of the true optimum with the help of Response Surface Methodology (RSM) . Continuing with a previous work  the present study deals with the selection and optimization of significant variables, using statistical methodology, in the process of bovine pancreatic trypsin partition in ATP formed by polyethylene glycol (PEG) and sodium citrate (NaCit). The homogenate resultant from bovine pancreas, was subsequently activated to obtain enzymes from their zymogen. The activity of trypsin (TRP) and chymotrypsin were determined with their respective substrate and the total protein concentration by Warburg method. Therefore, partition coefficients (Kr), yield (R) and purification factor (PF) were calculated. The experimental design was performed using Design-Expert 6.0.10 program. To screening a P-B model with six factors at two levels were used: PEG molecular weight (1450-8000), total systems concentration, NaCl concentration (ccNaCl) (without NaCl-8.0 %w/w), pH (5.20-8.20), top/bottom volume ratio (Vt/Vb) (0.1-16.0) and total concentration of homogenate at ATP (HS) (5-15 %w/w). Then the factors selected from the P-B design were optimized in a central composite design using RSM. In all cases the significant factors were obtained by analysis of variance (ANOVA) and tested with Fisher's test measuring the differences statistically significant with a 95% confidence (P ˂ 0.05). PlakettBurman ANOVA results presentation exhibit that ccNaCl, Vt/Vb and HS are significant (p ≤ 0.05) for PFTRP while RTRP was only affected by the Vt/Vb. According to the significance levels obtained by ANOVA, the equation of multiple linear regression to PFTRP showed that the three factors and their combination had significant effects (R2 = 0.97). Finally the tree parameters were found to have the greatest influence on the response and used to determine the optimal partition by RSM. This methodology was effective to test the distribution of homogenate in accordance with the model. Results confirmed that system composed for PEG3350/NaCit with 2.95% w/w NaCl, Vt/Vb 5.5 and 7.5% w/w HP, present a 92% R, 2 PFTRP and a desirability of 0.92. Moreover, from top phase surface graphics can be deduced that an increase of NaCl concentration and Vt/Vb ratio compared to lower HP is a useful tool that can address even more the distribution of TRP. The results show that RSM used is appropriate to model the factors studied which in turn lead to the definition of an optimum condition with the purpose of simultaneous extraction and purification of trypsin from bovine pancreas present undoubtedly a good result for a first step purification. We acknowledge the financial support of CONICET and ANPCyT. ________________  Albertsson, P. A. (1971) Partition of Cell Particles and Macromolecules, 2nd ed. Wiley: NY.  Plackett, R.L. and J.P. Burman (1946) Biometrika. 33(4) 305-325.  Tubio, G., Picó, G.A. and Nerli, B.B. (2009) Journal of Chromatography B. 877(3) 115-120.