Production, Quality Control, Stability and Pharmacotoxicity of a Malaria Vaccine Comprising Three Highly Similar PfAMA1 Protein Molecules to Overcome Antigenic Variation

Abstract

Plasmodium falciparum apical membrane antigen 1 (PfAMA1) is a leading asexual blood stage vaccine candidate for malaria. In preparation for clinical trials, three Diversity Covering (DiCo) PfAMA1 ectodomain proteins, designed to overcome the intrinsic polymorphism that is present in PfAMA1, were produced under Good Manufacturing Practice (GMP) in Pichia pastoris. Using identical methodology, the 3 strains were cultivated in 70-L scale fed-batch fermentations and PfAMA1-DiCos were purified by two chromatography steps, an ultrafiltration/diafiltration procedure and size exclusion chromatography, resulting in highly pure (>95%) PfAMA1-DiCo1, PfAMA1 DiCo2 and PfAMA1 DiCo3, with final yields of 1.8, 1.9 and 1.3 gram, respectively. N-terminal determinations showed that approximately 50% of each of the proteins lost 12 residues from their N-terminus, in accordance with SDS-PAGE (2 main bands) and MS-data. Under reducing conditions a site of limited proteolytic cleavage within a disulphide bonded region became evident. The three proteins quantitatively bound to the mAb 4G2 that recognizes a conformational epitope, suggesting proper folding of the proteins. The lyophilized Drug Product (1:1:1 mixture of PfAMA1-DiCo1, DiCo2, DiCo3) fulfilled all pre-set release criteria (appearance, dissolution rate, identity, purity, protein content, moisture content, sub-visible particles, immuno-potency (after reconstitution with adjuvant), abnormal toxicity, sterility and endotoxin), was stable in accelerated and real-time stability studies at -20°C for over 24 months. When formulated with adjuvants selected for clinical phase I evaluation, the Drug Product did not show adverse effect in a repeated-dose toxicity study in rabbits. The Drug Product has entered a phase Ia/Ib clinical trial.

Fig 1. Fermentation profiles for the GMP runs for PfAMA1-DiCo Drug Substances.

Upper panel. Base consumption and pH. Base consumption (NH₄OH addition) is related to growth and metabolic activity, since NH₄⁺ is the single source of nitrogen. Addition of base compensates the acidification during aerobic growth of microbes in mineral media. A pH setpoint of pH = 6.0 was used throughout the fermentations. Lower panel. Biomass and dissolved oxygen. The offline determined OD₆₀₀ was taken as a measure for biomass (1 OD₆₀₀ ~ 1 g/L w/w). The curves show the typical batch, glycerol fed-batch, methanol adaptation and methanol induction phases of Mut^s Pichia strain fermentations. Black lines; PfAMA1 DiCo1; blue lines, PfAMA1 DiCo2; red lines, PfAMA1 DiCo3.

Fig 2. SDS-PAGE, Western Blotting and SEC profiles of the Drug substances PfAMA1 DiCo1, PfAMA1 DiCo2 and PfAMA1 DiCo3.

Panel A: Reducing SDS-PAGE, silver-stained. Panel B: Non-reducing SDS-PAGE, silver-stained. Panel C: Non-reducing SDS-PAGE and Western-Blot, immune-staining using the reduction-sensitive monoclonal anti-PfAMA1 antibody 4G2. Panel D: SE-Chromatography of each of the PfAMA1 Drug Substances. PfAMA1 DiCo1 (black), PfAMA1 DiCo2 (blue) and PfAMA1 DiCo3 (red) was carried out on a Superdex200 10/300 GL prepacked column (GE Healthcare), using an Åkta Purifier Basic (GE Healthcare). PBS (140 mM NaCl, 10 mM sodium phosphate, 3 mM KCl, pH 7.4, Calbiochem) was used as running buffer at 0.5 mL.min^-1.

Fig 3. Signature peptide MS analysis of pre-GMP produced PfAMA1 DiCo Drug substance intermediates.

Panel A. Overloaded SDS-Gel of HIC eluate fractions with excised bands in marked boxes. Panel B. Examples of signature peptides identified in the 21 excised protein bands for PfAMA1-DiCo preparations. Key to preparations: PfAMA1 DiCo1 (Green), PfAMA1 DiCo2 (Orange) and PfAMA1 DiCo3 (Blue). In yellow the signature amino acids in these peptides, allowing for the differentiation between individual PfAMA1 DiCo proteins. Numbering of the amino acids is as in [15].

Fig 4. Quantitative binding of Drug Substances and Drug Product to mAb 4G2.

Panel A: PfAMA1 DiCo1 Drug Substance (black dotted line) and DiCo1 Drug Substance incubated with a 2.5 fold excess of 4G2 (black solid line). Panel B: PfAMA1 DiCo2 Drug Substance (blue dotted line) and DiCo2 Drug Substance incubated with a 1.5 fold excess of 4G2 (blue solid line). Panel C: PfAMA1 DiCo3 Drug Substance (red dotted line) and DiCo3 Drug Substance incubated with a 1.5 fold excess of 4G2 (red solid line). Panel D. PfAMA1 DiCo1:DiCo2:DiCo3 Drug Substance mixture (1:1:1 w/w/w) (green dotted line) incubated with a 1.5 fold excess of 4G2 (green solid line). PfAMA1 DiCos and mAb 4G2 were mixed and incubated for 1 h at room temperature before injection.

Fig 5. RP-HPLC analysis of individual and mixed PfAMA1-DiCo1, PfAMA1-DiCo2 and PfAMA1-DiCo3 samples.

Upper panel: Injection of the individual PfAMA1 DiCo proteins to establish the chromatographic profile of each of these proteins. PfAMA1 DiCo1 in black, PfAMA1 DiCo2, in blue and PfAMA1 DiCo3 in red. Lower panel: Chromatogram of the drug product containing the three PfAMA1 DiCo proteins at equimolar ratios, in green. For chromatographic details see the Methods Section.

Fig 6. Seroconversion as a function of dose.

Groups of 11 mice were immunized with indicated amounts of Drug Product formulated in either Rehydragel (Panels A, C and E) or GLA-SE (Panels B, D and F). On day 28 they were immunized again. Serum was collected on day 42 and ELISAs were performed using either PfAMA1 DiCo1 (Panels A,B), PfAMA1 DiCo2 (Panels C,D) or PfAMA1 DiCo3 (Panels E,F) as coating antigen. Individual mice within a treatment group are indicated by symbols, where the same symbol within a treatment groups represents the same animal for PfAMA1 DiCo1, PfAMA1 DiCo2 and PfAMA1 DiCo3, respectively.

Fig 7. Potency of the Drug Product as a function of storage time.

Panel A: Immunopotency of the drug product as a function of real-time storage time at -20°C for the high dose (1 μg Drug Product), formulated in Alhydrogel. The fold increase (seroconversion) is shown. Panel B: Immunopotency of the drug product as a function of real-time storage time at -20°C for the low dose (0.1 μg Drug Product), formulated in Alhydrogel. The fold increase (seroconversion) is shown. Panel C: Immunopotency of the formulated PfAMA1-DiCo vaccine (formulated with Alhydrogel) after storage for 24 hours at 5°C. Both dosages are shown. Panel D: Immunopotency of the PfAMA1-DiCo vaccine (formulated with Alhydrogel) after storage for 24 hours at 25°C. Both dosages are shown. Panel E: Immunopotency of the PfAMA1-DiCo Drug Product after storage for seven days at 30°C (before formulation). Immunopotency was measured as described in the Material and Methods section. Boxes indicate medians (middle) and quartile ranges (bottom and top).

Fig 8. Short time stability of the Drug Product in the presence of adjuvants.

Drug Product (DP) was formulated with either Alhydrogel (alum) or GLA-SE. Formulae were broken either directly after formulation (0 hours) or after 24 hours of storage at 4°C or at 25°C. Samples were analysed with SDS-PAGE, both reduced/silver staining (panel A) and non-reduced/silver staining (panel B) and by (non-reduced) western-blot with the monoclonal antibody 4G2 (panel C).