Accelerating the clinical development of protein-based vaccines for malaria by efficient purification using a four amino acid C-terminal 'C-tag'

Abstract

Development of bespoke biomanufacturing processes remains a critical bottleneck for translational studies, in particular when modest quantities of a novel product are required for proof-of-concept Phase I/II clinical trials. In these instances the ability to develop a biomanufacturing process quickly and relatively cheaply, without risk to product quality or safety, provides a great advantage by allowing new antigens or concepts in immunogen design to more rapidly enter human testing. These challenges with production and purification are particularly apparent when developing recombinant protein-based vaccines for difficult parasitic diseases, with Plasmodium falciparum malaria being a prime example. To that end, we have previously reported the expression of a novel protein vaccine for malaria using the ExpreS²Drosophila melanogaster Schneider 2 stable cell line system, however, a very low overall process yield (typically <5% recovery of hexa-histidine-tagged protein) meant the initial purification strategy was not suitable for scale-up and clinical biomanufacture of such a vaccine. Here we describe a newly available affinity purification method that was ideally suited to purification of the same protein which encodes the P. falciparum reticulocyte-binding protein homolog 5 - currently the leading antigen for assessment in next generation vaccines aiming to prevent red blood cell invasion by the blood-stage parasite. This purification system makes use of a C-terminal tag known as 'C-tag', composed of the four amino acids, glutamic acid - proline - glutamic acid - alanine (E-P-E-A), which is selectively purified on a CaptureSelect™ affinity resin coupled to a camelid single chain antibody, called NbSyn2. The C-terminal fusion of this short C-tag to P. falciparum reticulocyte-binding protein homolog 5 achieved >85% recovery and >70% purity in a single step purification directly from clarified, concentrated Schneider 2 cell supernatant under mild conditions. Biochemical and immunological analysis showed that the C-tagged and hexa-histidine-tagged P. falciparum reticulocyte-binding protein homolog 5 proteins are comparable. The C-tag technology has the potential to form the basis of a current good manufacturing practice-compliant platform, which could greatly improve the speed and ease with which novel protein-based products progress to clinical testing.

Keywords: Biomanufacture; Blood-stage; Malaria; Plasmodium falciparum; Protein purification; RH5; Vaccine.

Graphical abstract

Fig. 1

Plasmodium falciparum reticulocyte-binding protein homolog 5 vaccine constructs showing a schematic of P. falciparum reticulocyte-binding protein homolog 5 proteins. Both constructs encoded from the N-terminus: a BiP insect signal peptide (green) followed by the ectodomain of P. falciparum reticulocyte-binding protein homolog 5 (amino acids 26–526) (blue) followed by a C-terminal tag – either hexa-histidine or the glutamic acid – proline – glutamic acid – alanine C-tag. The proteins were based on the P. falciparum 7G8 strain sequence which has tyrosine (Y) at position 203 (and not cysteine as in the 3D7 clone reference genome sequence) (yellow circle). The other cysteine residues in P. falciparum reticulocyte-binding protein homolog 5 are indicated by small black boxes (C224, C317, C329, C345 and C351). Threonine (T) to alanine (A) substitutions to remove N-linked glycan sequons are indicated by red asterisks. The predicted molecular weight for each protein is indicated.

Fig. 2

Purification of Plasmodium falciparum reticulocyte-binding protein homolog 5 proteins. (A) Supernatant samples from batch cultures were run on SDS–PAGE under reducing conditions and western blotting performed with polyclonal anti-P. falciparum reticulocyte-binding protein homolog 5 rabbit serum. C, glutamic acid – proline – glutamic acid – alanine C-tagged protein; H, hexa-histidine-tagged protein. (B) Purity assessment of hexa-histidine-tagged P. falciparum reticulocyte-binding protein homolog 5 TALON column eluate (labelled E); hexa-histidine-tagged P. falciparum reticulocyte-binding protein homolog 5 Size Exclusion Chromatography eluate; C-tagged P. falciparum reticulocyte-binding protein homolog 5 C-tag eluate (labelled E); and C-tagged P. falciparum reticulocyte-binding protein homolog 5 Size Exclusion Chromatography eluate by SDS–PAGE under reducing conditions. (C) UV 280 nm absorbance chromatogram of hexa-histidine-tagged and (D) C-tagged P. falciparum reticulocyte-binding protein homolog 5 polishing SEC steps.

Fig. 3

Characterization of Plasmodium falciparum reticulocyte-binding protein homolog 5-C-tag (glutamic acid – proline – glutamic acid – alanine) protein. Surface plasmon resonance analysis of the interaction and affinity (K_D) of (A) C-terminal hexa-histidine-tagged and (B) C-tagged P. falciparum reticulocyte-binding protein homolog 5 protein with basigin. RU, Response Units. (C) Capture ELISA using a panel of P. falciparum reticulocyte-binding protein homolog 5-specific monoclonal antibodies (2AC7, QA5, 6BF10, RB3, 8BB10, 4BA7, 9AD4, QA1). Both P. falciparum reticulocyte-binding protein homolog 5 proteins were tested for binding using a dilution series ranging from 800 ng/mL to 12.5 ng/mL. Each sample was tested in triplicate for each concentration. Bars show the median plus range. Abs, absorbance.

Fig. 4

Immunological analysis of Plasmodium falciparum reticulocyte-binding protein homolog 5-C-tag protein vaccine. Growth inhibition activity against (A) the P. falciparum 7G8 laboratory-adapted parasite line and (B) P. falciparum 3D7 clone parasites versus total IgG concentration, with lines connecting data for each group of rabbits (n = 4/group) either vaccinated with C-terminal hexa-histidine-tagged, or glutamic acid – proline – glutamic acid – alanine C-tagged P. falciparum reticulocyte-binding protein homolog 5 protein. Individual data points are shown plus the line connecting the mean responses. Each GIA value is the mean of triplicate wells tested in the experiment. All GIA experiments were performed twice, with one representative result shown. Dotted line indicates 50% growth inhibition. (C) Total IgG concentrations (mg/mL) that gave 50% growth inhibition (EC₅₀) in the assay of growth inhibition activity. Points show the mean result for each rabbit tested in duplicate in two independent experiments. Where individual rabbits did not achieve 50% GIA at the highest tested IgG concentration, i.e. EC₅₀ > 10 mg/mL (indicated by the dotted line), these are plotted arbitrarily as 12 mg/mL. Median lines are shown. (D) Anti-P. falciparum reticulocyte-binding protein homolog 5 ELISA results are shown quantified in terms of μg/mL. Individual and median results are shown for each group. (E) Dose–response curve fitted to all 7G8 growth inhibition activity versus anti-P. falciparum reticulocyte-binding protein homolog 5 antigen-specific antibody concentration data (all IgG dilutions for each rabbit are shown). Dashed horizontal line indicates 50% growth inhibition activity. Non-linear least squares regression line is shown; r² = 0.96 for hexa-histidine and 0.69 for C-tag, n = 24 for both. (F) Same analysis as (E) against 3D7 parasites; r² = 0.87 for hexa-histidine and 0.82 for C-tag, n = 24 for both.

Fig. 5

Assessment of ELISA responses against C-terminal tags. Rabbit sera were diluted 1:100 (and 1:300) and tested by ELISA against a panel of recombinant proteins and peptides for responses against the C-terminal purification tags. Sera were tested pre- and post-immunisation with Plasmodium falciparum reticulocyte-binding protein homolog 5 C-terminal hexa-histidine tag, C-terminal EPEA-tag (PfRH5-C-tag) and C-terminus of P. falciparum reticulocyte-binding protein homolog 5 fused to rat CD4 domains 3 and 4 followed by hexa-histidine tag (PfRH5-CD4d3 + 4-His6) formulated in Freund’s adjuvant. Control reagents against the hexa-histidine sequence (anti-hexa-histidine) and the EPEA C-tag (anti-C-tag) were included. Raw O.D. at 405 nm (OD₄₀₅) data are shown plus medians for each group. (A, C) P. falciparum sexual-stage malaria antigen Pfs25 with a hexa-histidine tag (Pfs25-His6 protein); (B, D) α-synuclein protein; (E) 15 mer peptide and (F) 8 mer peptide corresponding to the C-terminal sequence of α-synuclein; (G) 8 mer peptide corresponding to the C-terminal sequence of PfRH5-C-tag; and (H) 4 mer EPEA (C-tag) peptide.