A human monoclonal antibody blocks malaria transmission and defines a highly conserved neutralizing epitope on gametes

Abstract

Malaria elimination requires tools that interrupt parasite transmission. Here, we characterize B cell receptor responses among Malian adults vaccinated against the first domain of the cysteine-rich 230 kDa gamete surface protein Pfs230, a key protein in sexual stage development of P. falciparum parasites. Among nine Pfs230 human monoclonal antibodies (mAbs) that we generated, one potently blocks transmission to mosquitoes in a complement-dependent manner and reacts to the gamete surface; the other eight show only low or no blocking activity. The structure of the transmission-blocking mAb in complex with vaccine antigen reveals a large discontinuous conformational epitope, specific to domain 1 of Pfs230 and comprising six structural elements in the protein. The epitope is conserved, suggesting the transmission-blocking mAb is broadly functional. This study provides a rational basis to improve malaria vaccines and develop therapeutic antibodies for malaria elimination.

Conflict of interest statement

M.B.S., W.P., X.H., B.B., and M.E. declare competing financial interests as all are employees of iRepertoire Inc., and J.H. is co-founder and CEO. J.D.G. is an employee of Alchemab Therapeutics Limited.

Figures

Fig. 1. Human recombinant mAbs were generated from Pfs230D1-specific single memory B cells of Malian adults vaccinated with the Pfs230D1-EPA/Alhydrogel® TBV.

a VH and VL genes corresponding to each mAb. LMIV230-01 and LMIV230-02 sequences originate from the IGHV1-69 heavy chain gene but utilize different kappa chain genes. Complete V gene usage determined in Pfs230-specific memory B cells is described in Supplementary Fig. 3e, f. b Epitope binning of human anti-Pfs230D1 scFvs. The primary binding scFv is listed on the left and the competing scFv is listed on the top. Reported scores are a percentage of total binding of that antibody in the absence of a competitor scFv. Values greater that 50% display low amounts of competition, while values lower than 50% exhibit greater competition. Any experiment with >100% binding was given a score of 100, while negative values were given a score of 0. Potential epitope bins are grouped and labeled above the table. c Functional activity of each mAb, assessed by Standard Membrane Feeding Assay (SMFA) and measured as the % reduction (versus control mAb) in the number of P. falciparum NF54 oocysts in midguts of Anopheles mosquitoes (“TRA”). d LMIV230-01 and LMIV230-02 mAbs bound similarly to Pfs230D1 and (e) show high affinity to recombinant Pfs230D1 (Supplementary Fig. 5, Supplementary Table 2). f LMIV230-01 reduces P. falciparum NF54 oocyst numbers by 91.7% at 1000 µg/mL, 86.7% at 500 µg/mL, 87.7% at 250 µg/mL, and 80.3% at 60 µg/mL, while LMIV230-02 displays only modest activity with 58.7% reduction at the maximum concentration of 1000 µg/mL, in SMFA. At least three biological replicates were performed for each concentration of mAb, in the presence of intact serum containing human complement. N ≥ 20 mosquitos per assay. Average oocyst numbers in the control mosquitoes (fed with mouse IgG1 mAb targeting P. yoelii P140 protein) for each experiment were: exp. 1 = 29.73; exp.2 = 7.18; exp. 3 = 57.86; exp. 4 = 36.41; exp. 5 = 51.71, exp. 6 = 4.55; exp. 7 = 62.35; exp. 8 = 20.50, exp.9 = 8.71, exp 10 = 18.05, exp. 11 = 5.86. Negative oocyst reduction values were set to zero. Human isotype IgG1 and US human serum pool were used as additional negative controls (Supplementary Fig. 6b). Values are shown as mean ± s.e.m. g LMIV230-01 and LMIV230-02 bind to non-reduced (NR) protein extract of P. falciparum NF54 gametes purified on Nycodenz after 2 h in exflagellation medium. h LMIV230-01 binds to gametes at 7.5 µg/mL while LMIV230-02 does not bind at 7.5 µg/mL. or 30 µg/mL. Both mAbs were labeled with Alexa Fluor 488. Scale bars: 5 µM. This experiment was performed in triplicate. Source data are provided as a Source Data file.

Fig. 2. LMIV230-01 binds to multiple parasite stages and its activity is complement-dependent.

a LMIV230-01 binds to permeabilized gametocytes, gametes, and zygotes and does not bind to ookinetes. Parasites were fixed and permeabilized, and 7.5 µg/mL of antibody was used to stain the different parasite stages. Scale bars: 5 µM. This experiment was performed in duplicate. b In vitro parasite lysis by LMIV230-01 is complement-dependent. Samples were tested in two independent assays, using two different parasite cultures. Data for LMIV230-02 mAb are shown in Supplementary Fig. 8a. c Functional activity of LMIV230-01 is also complement-dependent in vivo (SMFA with mosquitoes). Data from three independent SMFA assays. N ≥ 20 mosquitos per assay. Data for LMIV230-02 mAb are shown in Supplementary Fig. 8b. Oocyst averages in the control mosquitoes (fed with IgG1 targeting P. yoelli P140) for each of the experiments were: exp. 1 = 4.55; exp. 2 = 20.50, exp. 3 = 5.86. Data obtained from mosquitoes fed with LMIV230-01 at 1000 µg/mL with intact sera were also used to generate Fig. 1f. Values are shown as mean ± s.e.m. One-Way ANOVA and Tukey’s multiple comparisons test were used to compare the different groups. d Live imaging of P. falciparum NF54 female gametes incubated with LMIV230-01 in the presence of intact serum from a healthy donor revealed surface-deposited MAC (membrane attack complex) using anti-C5b-9 + C5b-8 antibody (magenta color). MAC deposition was not detected in the presence of heat-inactivated (HI) serum. Scale bars: 5 µM. This experiment was performed in triplicate. Source data are provided as a Source Data file.

Fig. 3. Structural definition of LMIV230–01 scFv epitope in Pfs230D1.

a Domain organization of Pfs230D1. A-Type (gray) and B-Type (white) 6-Cys domains are shown in boxes. Numbers inside the boxes indicate the number of conserved cysteines. Domain numbers are labeled in Roman numerals. -vvvv- indicates a highly repetitive region that is cleaved from the mature protein. b Structure of Pfs230D1. N-terminal loop in violet, beta-sandwich in green, beta strands in orange, helices in red, loops in gray and disulfide bonds in blue sticks. c Overall structure and epitope for the Pfs230D1–LMIV230–01 scFv complex. Pfs230D1 in gray; LMIV230–01 scFv heavy chain in blue; LMIV230–01 scFv light chain in violet; epitope in red. d Orthogonal detailed view of the Pfs230D1 epitope for LMIV230–01 scFv. Pfs230D1 in gray ribbon; residues contacted by LMIV230–01 scFv heavy chain, light chain and both are highlighted in blue, violet, and orange, respectively. e Surface representation of the epitope on Pfs230D1. The orientation and the color scheme are the same as in (b). f Polymorphisms in Pfs230 mapped onto the surface of Pfs230D1. Orientation as in (c). The LMIV230–01 scFv binding epitope in blue, polymorphic residues are in green, polymorphic residues within the epitope are in black. The frequency of the polymorphisms is in parentheses.

Fig. 4. LMIV230-01 binds to heterologous P. falciparum strains and antisera from Pfs230D1 vaccinees vary widely in levels of antibody that compete with LMIV230-01 for binding.

a LMIV230-01 bound to gametes of St. Lucia parasite strain and b of an isolate obtained from a Malian adult and adapted to culture. c Murine anti-48/45 mAb confirms formation of gametes by Malian isolate and its signal colocalizes with LMIV230-01. “Merged” refers to combination of green and red channels. d Membrane attack complex forms on gametes of St. Lucia strain and (e) of a Malian isolate incubated with LMIV230-01 in the presence of intact but not heat-inactivated serum. All imaging experiments shown in this figure were performed in duplicate. Scale bars for all images in this panel: 5 µM. f Cartoon schematizing LMIV230-01 competition ELISA assay. g Distribution of serum antibody levels that compete with LMIV230-01 for binding to Pfs230D1 in 36 subjects who received Pfs230D1-EPA vaccine. Values displayed represent mean from three independent experiments. h Relationship of LMIV230-01-competing antibody levels to total Pfs230D1 antibody titers, or (i) to serum functional activity (TBA, transmission-blocking activity) measured by SMFA. Source data are provided as a Source Data file.