(M1130-08-46) Purification and Heat Inactivation of an Oxidation Prone Protein: Study of Hemoglobin

Purpose: Hemoglobin based oxygen carriers have been under development for battlefield and civilian trauma-associated hemorrhage. However, the class of drugs struggles with limited availability (large volume requirement) as well as special stability constraints (oxidation induced methemoglobin formation)^[1]. Sensitivity of hemoglobin to heat also poses a challenge for the heat based viral inactivation step during the purification step. Here, we have performed process optimization of a single-use centrifugation process to wash (plasma and HCP removal) large volume of bovine whole blood. Moreover, we demonstrate a heat treatment process of deoxyhemoglobin in presence of amino acid based reducing agent to achieve minimum oxidation while clearing common bovine viruses and prions.

Methods: Bovine hemoglobin was processed using single-use fluidized bed centrifugation system and compared side-by-side to a benchtop centrifuge system. Wash quality was assessed by measuring protein content in the permeate and a design of experiments was performed to evaluate critical process parameters. Oxyhemoglobin was extracted from the washed RBCs and subjected to deoxygenation process using a Liqui-Cel membrane and argon to achieve >90% deoxygenation. Deoxygenated hemoglobin was heat treated at 60 °C for 0-10 hours in presence of varying concentrations of reducing agent. Heat treated hemoglobin was then purified & oxygenated using fresh buffer on a 10 kDa TFF system. Catalase, superoxide dismutase, and carbonic anhydrase were analyzed downstream to evaluate HCP clearance. Spike studies and subsequent log reduction in model bovine virus (BVD virus, Retro virus, Parvo virus) were assessed for viral inactivation. Purified hemoglogin purity and oxidation were measured by cIEF as well clinical co-oximeter OSM3. Furthermore, free iron content was measured to evaluate irreversible degradation of the hemoglobin.

Results: Process optimization of the RBC wash by single-use centrifugation demonstrated cycle load, wash buffer diavolumes, speed, and recirculation cycles as CPPs directly impacting the wash quality. The large-scale process was found produce comparable wash quality to benchtop centrifuge. Charge profile of hemoglobin showed main peak at approximately 7.5 while force degraded sample with high methemoglobin showed unique peaks in basic region. Deoxygenation of hemoglobin increased the heat resistance, however, heat treatment experiments showed duration dependent degradation of the hemoglobin with corresponding decrease in main peak purity and increase in impurity peaks in cEIF profile. The rate of degradation of main peak was found to be linear vs heat treatment duration in hours for low concentration of reducing agent. Increasing the reducing agent concentration beyond a threshold, significantly reduced the heat induced degradation (Figure 1). Interestingly, free iron content did not follow linear trend. All model adventitious agents and HCP were found to be cleared with high log reduction.

Conclusion: The large-scale fluidized bed based continuous centrifugation system was optimized to be comparable to benchtop centrifugation method. Deoxygenation of hemoglobin and subsequent reducing agent optimization was critical to control methemoglobin and corresponding degradation. cIEF was found to be invaluable in separating and detecting reversible heat induced degradants in the purified hemoglobin. Hence, a large-scale process and control strategy was established to generate highly purified bovine hemoglobin.

References: Sen Gupta A. Hemoglobin-based Oxygen Carriers: Current State-of-the-art and Novel Molecules. Shock. 2019 Oct;52(1S Suppl 1):70-83

Impact of heat treatment duration on main peak purity at three concentrations of reducing agent