Immunization with full-length Plasmodium falciparum merozoite surface protein 1 is safe and elicits functional cytophilic antibodies in a randomized first-in-human trial

Abstract

A vaccine remains a priority in the global fight against malaria. Here, we report on a single-center, randomized, double-blind, placebo and adjuvant-controlled, dose escalation phase 1a safety and immunogenicity clinical trial of full-length Plasmodium falciparum merozoite surface protein 1 (MSP1) in combination with GLA-SE adjuvant. Thirty-two healthy volunteers were vaccinated at least three times with MSP1 plus adjuvant, adjuvant alone, or placebo (24:4:4) to evaluate the safety and immunogenicity. MSP1 was safe, well tolerated and immunogenic, with all vaccinees sero-converting independent of the dose. The MSP1-specific IgG and IgM titers persisted above levels found in malaria semi-immune humans for at least 6 months after the last immunization. The antibodies were variant- and strain-transcending and stimulated respiratory activity in granulocytes. Furthermore, full-length MSP1 induced memory T-cells. Our findings encourage challenge studies as the next step to evaluate the efficacy of full-length MSP1 as a vaccine candidate against falciparum malaria (EudraCT 2016-002463-33).

Keywords: Malaria; Protein vaccines.

Conflict of interest statement

Competing interestsH.B., E.B. and M.L. are shareholders of Sumaya GmbH & Co. KG. The other authors declare no competing interests.

© The Author(s) 2020.

Figures

Fig. 1. Immunization scheme and study design.

a Volunteers were immunized with SumayaVAC-1 (full length MSP1 plus GLA-SE as adjuvant, 500 µl volume) on days 0, 29, 57, and optionally after unblinding of the cohort on day 182. Blood samples were taken for serological analysis on the days indicated. The safety follow-up was scheduled 6 months after the last vaccination. b 32 healthy volunteers were recruited in two consecutive cohorts. Sixteen volunteers each were randomly assigned within a cohort. The number of volunteers receiving three or four vaccinations is indicated for each cohort.

Fig. 2. IgG and IgM antibody titers against MSP1-D after immunization with SumayaVac-1.

Blood samples were collected from the vaccines on the days indicated and the a IgG-specific and b IgM-specific MSP1-D antibody titers were determined by ELISA. The data points represent the geometric means ± 95% confidence interval of each cohort. Crossed circles indicate samples from single individuals. Red arrows indicate the days of immunization. Dashed lines indicate reference titers obtained from pooled sera from semi-immune individuals from Nouna, Burkina Faso (orange line, n = 11) and from Kisumu, Kenya (black line, NIBSC code 10/198). Controls comprise the placebo and the GLA-SE vaccinees. The pie chart shows the average percentile distribution of IgG subclasses after three immunizations with SumayaVac-1 (day 85, n = 24).

Fig. 3. IgG responses against MSP1-D sub-fragments and cross-reactivity against MSP1-F.

a Structural organization of MSP1. MSP1 is synthesized as a precursor protein of 196 kDa. During merozoite maturation, MSP1 is cleaved by PfSub1 into four major MSP1 fragments, termed p83, p30, p38, and p42. The cleavage sites are indicated by arrowheads. Conserved, dimorphic and oligomorphic domains of MSP1 are indicated. b The net α-MSP1-D titers of the vaccinees on day 85 are shown for the four major MSP1 fragments. Each data point represents the result (performed in duplicates) from one individual. Black marks indicate the mean value of the corresponding cohort. c Antibody titers against MSP1-D and MSP1-F after three immunizations on day 85.

Fig. 4. Recognition of native MSP1-D by induced antibodies.

a Representative indirect immuno-fluorescence assay (IFA) showing reactivity of serum from an individual immunized 3× with SumayaVac-1 with developing merozoites during schizogony and free mature merozoites. Trophozoites served as a negative control. Trophozoite and schizont samples were fixed for analysis, whereas merozoites were unfixed. Scale bar, 5 µm. b Correlation between the α-merozoite titer and the α-MSP1-D IgG titer for all volunteers vaccinated 3x with SumayaVac-1 (day 85, n = 24). Each data point represents the results from one vaccinee. Titers were determined by quantitative ELISAs, whereby one set of plates was coated with purified full-length MSP1 and the other with purified P. falciparum merozoites. The correlation between two parameters was assessed using Pearson product moment correlation (correlation coefficient, 0.624; p = 0.0011).

Fig. 5. Effect of induced α-MSP1 antibodies on parasite growth inhibition, complement activation, and invasion inhibition.

a Growth inhibition assay (GIA). The P. falciparum strain 3D7 was cultured for one cycle in the presence of total IgGs from volunteers vaccinated 3x with SumayaVac-1 (day 85; n = 14, circles). Two IgG preparations each from MSP1-immunized rabbits (triangles) and semi-immune adults from Burkina Faso (squares) served as positive controls, and pooled IgGs from volunteers collected on day 0 (inverted triangles) as negative controls. The concentrations of purified total IgG antibodies used in the assay are indicated. Inset: GIA activity in the presence of active complement. b Sera from volunteers vaccinated 3x with SumayaVac-1 (day 85; n = 24) fix complement in the presence of purified MSP1, as determined using a C1q-specific ELISA. The readout is the absorbance at a wavelength of 492 nm. A serum pool from semi-immune adults from Burkina Faso (n = 11) served as a positive control and the sera from the placebo and GLA-SE vacinees as negative control. Each data point represents the result (performed in duplicates) from one individual. A box plot analysis is overlaid over the individual data points with the median (black line), mean (red lines), and 25% and 75% quartile ranges being shown. The error bars above and below the box indicate the 90th and 10th percentiles. Filled black circles indicated outliers. Statistical significance was assessed using one way ANOVA. c Direct invasion assay (IIA). Purified merozoites (3D7) were incubated with erythrocytes in the presence of 3 mg ml−1 purified IgG for 30 min before the number of invaded erythrocyte was determined and the parasitemia calculated. IgG preparations from rabbits immunized with apical membrane antigen 1 (AMA-1) and a semi-immune adult from Burkina Faso served as positive controls and an IgG pool from the placebo cohort as negative control. Each data point represents the result (performed in duplicates) from one individual. Box plot analysis as described above.

Fig. 6. Mapping of linear B-cell epitopes across MSP1.

Sera from vaccinees (1:1000), two malaria semi-immune adults from Burkina Faso (1:1000), and two MSP1 immunized rabbits (1:10,000) were applied to a custom-made MSP1 peptide microarray chip consisting of 1706 oligomeric peptides of 15 amino acids with a peptide-to-peptide overlap of 14 amino acids. Each peptide was printed in duplicate on the chip. As secondary antibodies the appropriate DyLight680 conjugated anti-IgG (Fc) antibodies were used. The fluorescence intensity landscapes across MSP1 are shown for each sample. Relevant epitopes have been highlighted. Controls comprise the placebo and the GLA-SE vaccinees. For orientation, the structural organization of MSP1 is shown below the fluorescence profiles. Full details of epitopes and fluorescence intensities are provided in the Supplementary Table 2.

Fig. 7. Association of GIA activity with MSP1 epitopes.

Major epitopes putatively associated with GIA activity as defined by LOD scores > 2.5 (p = 0.001) for at least three consecutive peptides are highlighted. The GIA activity underpinning the correlation are depicted in Fig. 5a, with sera from MSP1-immunized rabbits and semi-immune adults from Burkina Faso showing GIA activity, whereas sera from vaccinees being GIA inactive. Epitopes that map to MSP1 processing sites, are indicated in blue, with the cleavage site being pointed out by an arrow. Note that three processing sites were detected between the p38/p42 junction: the canonical cleavage site (TGEAISV) and two adjacent alternative sites. Another epitope includes a lysine (highlighted in red) that is posttranslationally modified by acetylation. The correlation was determined using a Pearson function followed by a Bonferroni correction and conversion of the resulting p-values into a LOD score (logarithm of odds). The confidence line as defined by p < 0.001 is indicated. For orientation, the structural organization of MSP1 is shown below the graph. Full details of epitopes and LOD scores are provided in the Supplementary Table 2.

Fig. 8. Activation of neutrophils by IgG from volunteers immunized with SumayaVac-1.

P. falciparum merozoites from the a 3D7 strain expressing MSP1-D or the b FCB1 strain expressing MSP1-F and neutrophils were incubated in the presence of purified IgG from volunteers immunized three times with SumayaVac-1 (collected at days 0 and 85). Opsonized merozoites activated neutrophils to produce a respiratory burst, which was quantitatively expressed in terms of an ADRB index. Each data point represents the mean ADRB Index of at least two independent experiments each performed in duplicates from one volunteer. Black marks indicate the geometric mean values of the corresponding cohort. Statistical significance was assessed using paired t-test (between day 0 and day 85) and the Kruskal–Wallis one-way ANOVA on ranks (between different dosing groups). ADRB activities for purified IgG from semi-immune adults from Nouna, Burkina Faso (n = 11), and from Kenya (pooled IgG) are indicated as reference. c Correlation between the net ADRB index (corrected for the background activity at day 0) and the α-MSP1-D titer (at day 85) (correlation coefficient, 0.545; p = 0.006, according to Spearman rank order correlation). Each data point represents the results from one volunteer (n = 24).

Fig. 9. MSP1-specific recall T-cell response.

PBMCs from six volunteers immunized three times with SumayaVac-1 were exposed to purified MSP1-D (5 µg ml−1 for 24 h) and the recall T-cell response as defined by IFN-γ production was measured in the cultured ELISpot assay. a Representative images of ELISpot wells are shown for different vaccinees. b The net mean spot-forming units (SFU) per 1 million PBMCs were calculated and analyzed as a function of time post the first immunization. Each data point represents the mean from three determinations ± SD. The red arrows indicate the days of immunization with SumayaVac-1. Significance was assessed using Kruskal–Wallis one way ANOVA on ranks. c Stimulation of PBMCs by peptides corresponding to HLA-A0201-restricted MSP1 CD8+ T-cell epitopes. PBMCs from the same vaccinees as above were analyzed (same color code as above), in addition to PBMCs from three semi-immune adults from Burkina Faso (B.F.). The following peptides were used: 291, GLHHLITEL; 374, SLLTELQQV; 437, VIYLKPLAGV; 674, KLKEFIPKV.,, Statistical significance was assessed using the two tailed t-test.