Human ethics statement

The clinical trials were approved by the FDA, by the ethics review boards from the Faculté de Médecine de Pharmacie et d’OdontoStomatologie (FMPOS), Bamako, Mali, and the US National Institute of Allergy and Infectious Diseases (NIH, Bethesda, MD, USA), as well as the Mali national regulatory authority. Written informed consent was obtained from study participants. The clinical trials are registered in clinicaltrials.gov (NCT02334462 for Pfs230D1-EPA/Alhydrogel, with results published21; NCT02942277 for Pfs230D1-EPA/AS01) and phase I in US healthy adults was published elsewhere22.

Human immunization and samples collection

Malian adults were vaccinated with 40 µg of Pfs230D1-EPA in Alhydrogel® or in AS01 at planned study days 0, 28, 168, and (for Alhydrogel® group) 540. Pfs230D1-EPA/Alhydrogel® 3rd and 4th doses were administered just before the rainy season (period of peak malaria transmission) in Year 1 and Year 2, respectively, of the study. From the Pfs230D1-EPA/Alhydrogel® trial, sera and peripheral blood mononuclear cells (PBMCs) were obtained from eight participants, 14 days after the 4th dose (day 554). From the Pfs230D1-EPA/AS01 trial, collections occurred 7 days after the 3rd dose. PBMCs (5 million cells per sample on average) were prepared for isolation of Pfs230D1-specific single memory B cells.

ELISA

ELISA to assess total anti-Pfs230D1 IgG levels in sera from immunized subjects and evaluation of binding activity of human mAbs was performed as previously reported17. Briefly, Immulon® 4HBX plates were coated with 1ug/well of recombinant Pfs230D1. Plates were incubated overnight at 4 °C and blocked with 320 µL of buffer containing 5% skim milk powder in Tris-buffered saline for 2 h at room temperature. Plates were washed with Tween-TBS. Samples (dilution 1:500 of sera or 100 μg/mL of mAbs) were added to Pfs230D1-coated wells, in triplicate, and incubated for 2 h at RT. Plates were washed, then 100 µL of alkaline phosphatase labeled goat anti-human IgG were added. Plates were incubated for 2 h at room temperature and washed. A colorimetric substrate, p-nitrophenyl phosphate (Sigma, St. Louis, USA) was added and plates were read at absorbances of 450 nm and 550 nm on a multi-well reader (Molecular Devices, San Jose, USA).

Identification and sorting of antigen-specific single memory B cells

Identification and sorting of Pfs230D1M-specific B cells was performed as previously described17. Briefly, recombinant Pfs230D1M was chemically biotinylated using EZ-Link Sulfo-NHS-LC-Biotin (Thermo Fisher Scientific, Waltham, USA). The biotinylated protein was then tetramerized with streptavidin labeled with Phycoerythrin (PE) (Prozyme, Hayward, USA). PBMCs from vaccinees were thawed in 37 °C water bath, and resuspended in complete Roswell Park Memorial Institute (RPMI) 1640 Medium with L-glutamine and 25 mM HEPES (Corning, Corning, NY, USA, 10-041-CV) and washed with phosphate-buffer solution (PBS) (Thermo Fisher Scientific, Waltham, MA, USA, 10010023). One µL of Pfs230D1 tetramers was added to the cells that were then incubated at 4 °C for 20 minutes. Cells were washed with PBS containing 10% fetal bovine serum (FBS) and incubated with 25 µL of anti-PE magnetic beads (Miltenyi Biotech, Bergisch Gladbach, Germany, 130-048-801) for 25 minutes. Four mL of PBS were added to the solution, which was then passed over magnetic LS columns for elution of cell suspension enriched for antigen-specific cells. After enriching with the tetramer, PBMCs were stained with the following surface-conjugated antibodies: CD3 (UCHT1), CD14 (M5E2), CD56 (HCD56) Alexa Fluor 700, CD19 APC-CY7 (HIB19), CD20 PE-CY7 (2H7) and CD27 APC (LG.3A10), purchased from Biolegend (San Diego, USA). The following gating strategy was used: singlet cells were selected for the exclusion of non-B cells using CD3, CD14 and CD56 markers. Lymphocytes were gated for CD19+ CD20+. Pfs230D1M-specific B cells were gated using PE and excluding non-Pfs230D1M cells gated using CF594, the fluorochrome used in the decoy BSA tetramer. Sorting was performed into 96-well plates using a FACSAria™ II instrument (BD Biosciences, San Jose, USA) with blue, red, and violet lasers. Pfs230D1M-specific memory B cells were analyzed according to the fluorescence staining profile described above and sorted directly into 96-well PCR plates using a 100 µM nozzle. After sorting, plates were immediately centrifuged at 1,278 x g for 30 seconds, transported in dry ice and stored at −80 °C.

Sequencing and data processing

Amplification of heavy and light chains and sequencing was performed by iRepertoire Inc. (Huntsville, AL, USA) as previously reported17. Briefly, amplification of BCR heavy and light chains from single sorted cells was performed by iRepertoire Inc. (Huntsville, AL, USA). RT-PCR1 was performed with nested, multiplex primers covering both heavy, kappa, and lambda loci, and including partial Illumina adaptors. Included on the reverse primer was an in-line six nucleotides barcode, which served as a plate identifier so that multiple 96-well plates could be multiplexed in the same sequencing flow cell. After RT-PCR1, the first round PCR1 products were rescued using SPRISelect Beads (Beckman Coulter, Brea, USA). A second PCR was performed with dual-indexed primers that complete the sequencing adaptors introduced during PCR1 and provide plate positional information for the sequenced products. Sequencing was performed using the Illumina MiSeq v2 500-cycle kit with 250 paired-end reads.

Standard membrane feeding assay to assess functional activity of polyclonal antibodies in serum of vaccinees or Pfs230D1 mAbs

Transmission reducing activity (TRA) was measured as the reduction of mean P. falciparum burden in infected mosquitoes fed with intact sera from vaccinees, using the SMFA, as previously reported22. Percentage reduction of P. falciparum oocyst numbers was calculated for immune serum as compared to negative control (naive sera from US donors). Additional negative control (mouse anti P. yoelii P140 at 375 µg/mL) was used to confirm the oocyst reduction activity of the Pfs230D1 mAbs (human IgG1 isotype at 1000 µg/mL in pooled sera from US malaria-naive donors). A mouse anti-Pfs25 mAb (4B7, at 175 µg/mL) was used as a positive control.

Briefly, an in vitro 15-day culture of P. falciparum (NF54 line) containing stage V gametocytes was diluted with washed O + RBCs (Interstate Blood Bank, Memphis, USA) and an AB + serum pool (not heat-inactivated) from US malaria-naive subjects (Interstate Blood Bank, Memphis, USA) to final concentration of 0.07%–0.1% stage V gametocytes and 50% hematocrit. For each individual assay, 200 µL of diluted culture was mixed with 60 µL of mAb diluted in PBS. Samples were then fed to pre-starved (24–30 h) 3–8-day-old Anopheles stephensi (Nijmegen strain) mosquitoes using a Parafilm membrane on a mosquito feeder, kept warm with 40 °C circulating water. After feeding, mosquitoes were kept at 26 °C and 80% humidity conditions to allow parasites to develop. On Day 8 after the feed, mosquito midguts were dissected and stained with 0.05–0.1% mercurochrome solution in water for 20–30 min. Infectivity was measured by counting oocysts in at least 20 mosquitoes per sample. Each sample was tested in at least two independent SMFAs. For SMFA experiments performed to assess the complement-dependent activity of the mAbs, sera were heat-inactivated at 56 °C before the addition of mAbs.

BCR sequence analyses

2393 sequences were assigned Ig genes, out of 2,397 input sequences (99.83%). Gene assignment and other downstream analysis were performed with the Immcantation suite container version 4.1.0, which includes IMGT reference germlines downloaded on 2020-08-12, IgBLAST version 1.16.0, Change-O 1.0.0, Alakazam 1.0.2 and SHazaM 1.0.2. After gene assignment with IgBLAST, 126 unproductive sequences were removed. Data were processed to retain one heavy chain sequence and, at most, two light chain sequences per well. In wells with multiple heavy chain sequences, we retained the one with the highest cdr3_read_count, and required it to account for more than 50% of the total heavy chain cdr3_read_count in the well. In wells with more than two light chain sequences, we retained the two sequences with the highest cdr3_read_count, each accounting for more than 25% of the total light chain cdr3_read_count in the well, and together for >75% of the total light chain cdr3_read_count in the well. Sequences from wells presenting light chain only were removed from the analysis. One well was excluded for possible contamination (subject 2, well A07, Pfs230 Alhydrogel). A total of 1,345 sequences passed these filters (733 heavy, 612 light), and constitute the final set of sequences analyzed (Supplementary Table 1), with an average of 57 sequences per subject for the Alhydrogel group, and 60 sequences for the AS01 group. Most of the sequences (79.26%) from the heavy chain sequences correspond to VH:VL pairs (Supplementary Fig. 4). Most of these analyses consisted in descriptive data and statistical comparison were not performed for all figures.

Sequence selection, mAbs expression, and purification

VH and VL sequences of Pfs230D1-specific single memory B cells were selected for expression based on evidence of clonal expansion. If multiple cells contained identical CDR3 sequences, it was attributed to clonal expansion, and thus evidence of activation. Naturally paired VH/VL sequences were expressed in an IgG1 backbone by LakePharma Inc. Monoclonal antibodies were cloned based on the criteria above, and VH/VL pairs for cloning and expression were obtained from all participants in this study.

Human mAbs were expressed using 0.1 L transient production in HEK293 cells (Thermo Fisher Scientific) and purified using protein A affinity chromatography. Purified proteins were sterilized using a 0.2 μm sterile filter and characterized to confirm 90% purity by capillary electrophoresis-SDS, concentration, and endotoxin levels according to the manufacturer’s procedures.

SMFA to test functional activity of mAbs

SMFA was performed to assess the ability of mAbs to block the development of P. falciparum strain NF54 oocysts in the mosquito midgut, as previously reported41. Each mAb was evaluated once at each test concentration, and n = 20 mosquitoes were examined per sample. In each assay, as a control, normal mouse antibody at 750 µg/mL was included (n = 20 or 40 mosquitoes).

All the mAbs from the AS01 group were tested at 100 µg/mL. The mAbs from the Alhydrogel group were initially screened at 375 µg/mL, and then the mAbs with >65% TRA at 375 µg/mL were further tested at 100 µg/mL. A functional mAb was defined as a mAb with >75% TRA at 100 µg/mL. If a mAb showed <65% TRA at 375 µg/mL (it was not tested at 100 µg/mL by SMFA), the mAb was categorized as “non-functional”.

Statistical analyses

Statistical analyses of antibody genes were performed using Wilcoxon rank sum test. The Benjamini–Hochberg false discovery rate (FDR) was used to correct for multiple hypothesis testing, and the criteria for significance were p < 0.05 and FDR < 0.20. The descriptions of the tests used are contained within the figure legends.