(493h) Bovine Serum Albumin and Myoglobin Separation on Hap

Separation by chromatography is one of the most relevant processes for purifying biological molecules, such as proteins. Hydroxyapatite (HAp) has been successfully used as adsorbent for protein separation and it is characterized by its dual adsorption sites. The C-site, characterized by the Ca2+ groups of the material, work as an anion exchanger and metal affinity site, which is a strong interaction, and it is crucial for adsorbing the carboxylic groups of proteins. Acidic proteins selectively bind on this site being then eluted using high concentrations of sodium phosphate solutions [1,2]. Conversely, the P-site, characterized by the group PO43- works as a cation exchanger, so the binding is based on electrostatic interactions and is responsible for the adsorption of amino groups of the protein. Consequently, HAp can be considered a mixed-mode medium for protein separations.

Simulated moving bed (SMB) chromatography is a continuous process where counter current operation increases the productivity and reduces eluent consumption. SMB is characterized by having two inlet streams, the feed, the eluent, and two outlet streams, the extract (where the component with higher affinity with the solid phase is obtained), and the raffinate (containing the species that interact less with the solid) [3].

Material and Methods

BSA and Mb are the selected proteins, which are two model proteins with different isoelectric points. BSA carried negative charge at physiologic pH due to its isoeletric point of 4.7. In its turn, Mb isoelectric point is 7.4.

Since HAp was provided in powder, it was necessary to shape it into a granular form. Therefore, a lab extruder was used to shape the granules.

The adsorption equilibrium of BSA and Mb was determined through batch experiments conducted at pH 7, and different phosphate buffer (PBS) concentrations. Subsequently, fixed bed experiments were conducted for both proteins using the same buffer concentrations. Afterwards, the FlexSMB-LSRE unit with six interconnected stainless-steel columns, was used to perform SMB experiments in the configuration of 1-2-2-1 with an equimolar mixture of the two proteins. The main objective is to separate these two proteins, and a buffer step-elution gradient was applied in the system by selecting a different buffer concentration in the feed and eluent streams.

Results and Discussion

Batch and fixed bed experiments were performed for BSA and Mb at pH7 and 0.01 M, 0.05 M, 0.1 M, and 0.4 M of PBS buffer. Adsorption capacity increases with the decrease of PBS concentration for both proteins, Mb and BSA. The adsorbed amounts are considered negligible when using buffer solutions with 0.4 M concentration of PBS. Consequently, this higher concentration of buffer was utilized as the elution medium.

A phenomenological mathematical model, i.e., the general rate model coupled with steric mass action equilibrium, was used and validated against fixed bed dynamic adsorption experiments at different buffer concentrations. Thus, the SMB unit can be designed by employing a buffer gradient mode, with the eluent stream featuring a higher buffer concentration to ensure solid regeneration. Conversely, a lower buffer concentration is used in the feed to enhance adsorption in section III.

To operate the SMB unit, the selection of a suitable operating point was required. Firstly, an appropriate switching time was selected, which was 20 minutes. Then, flow rates in section I and IV were selected, and a separation region was computed, considering a purity requirement of 95% in the extract and raffinate streams.

The SMB experiment was performed, and high-purity products were obtained in extract and raffinate streams. Thus, it was possible to produce the Mb in the extract stream with a purity of 92% and the BSA in the raffinate with a purity of 95%. Regarding the recoveries, they are in the same scale as the purities, being possible to recover 87% of the Mb in the extract and 98% of BSA in the raffinate. Lastly, the productivities obtained were 1.04×10^-3 mol_protein · kg_ads^-1· day^-1, for Mb in the extract and 1.24×10^-3 mol_protein · kg_ads^-1· day^-1 for BSA in the raffinate.

The concentration histories of BSA and Mb on both streams are presented in Figure 1.

Aknowledgements:

This work was supported by national funds through FCT/MCTES (PIDDAC): LSRE-LCM, UIDB/50020/2020 (DOI: 10.54499/UIDB/50020/2020) and UIDP/50020/2020 (DOI: 10.54499/UIDP/50020/2020); and ALiCE, LA/P/0045/2020 (DOI: 10.54499/LA/P/0045/2020). Albertina Rios also acknowledges her Ph.D. research grant awarded by the Foundation of Science and Technology of Portugal (FCT) under SFRH/BD/137891/2018 project.

References

[1] Gorbunoff, M.J., The interaction of proteins with hydroxyapatite: II. Role of acidic and basic groups. Analytical Biochemistry, 1984. 136(2): p. 433-439 https://doi.org/10.1016/0003-2697(84)90240-9.

[2] Cummings, L.J., M.A. Snyder, and K. Brisack, Chapter 24 Protein Chromatography on Hydroxyapatite Columns, in Methods in Enzymology, R.R. Burgess and M.P. Deutscher, Editors. 2009, Academic Press. p. 387-404.

[3] Rodrigues, A.E., et al., Principles of Simulated Moving Bed, in Simulated Moving Bed Technology. 2015. p. 1-30