Fig 1.
Schematic/sequence of PfCSP_3D7 and approximate epitopes bound by mAbs used in this study.
Top: color-coded schematic illustrating the N-terminus (N-CSP), repeat region, and C-terminus (C-CSP) of PfCSP_3D7. N-CSP contains a signal peptide (SP), two Plasmodium export element (PEXEL) sites, and the conserved region I (RI). The repeat region is composed of three types of tetrapeptides (1 NPDP, 4 NVDP, and 38 NANP). C-CSP has a linker to the repeat region, an α-thrombospondin type-1 repeat (αTSR) domain that contains two conserved motifs (region III and region II+, RIII and RII+) and several CD4+ helper T cell epitopes (Th2R, Th3R, CS.T3), and a glycosylphosphatidylinositol (GPI) anchor sequence. The binding sites of monoclonal antibodies used in this study are depicted. Bottom: sequence of PfCSP_3D7, color-coded to match the schematic. The sequence of the recombinantly expressed C-CSP construct used in this study is underlined.
Fig 2.
Binding and in vivo protection mediated by C-CSP-specific mAbs.
(A) Heat map of area under the curve (AUC) values of thirteen C-CSP mAbs binding to FL-rCSP, N-CSP, 36mer peptide (NANP)9, and C-CSP by ELISA. VRC01 (anti-gp120 mAb), 5D5 (N-CSP mAb), and mAb10 (NANP-preferring repeat mAb) were respectively included as a negative isotype control, positive control N-CSP mAb, and positive control repeat mAb. Data were averaged from 2–3 independent experiments. (B) Epitope binning of C-CSP mAbs binding to FL-rCSP measured by SPR. All mAbs were tested as both ligands and analytes; several mAbs (L7, L15, L20, R1, mAb10) were excluded due to poor ligand and/or analyte binding to FL-rCSP. Solid lines indicate two-way competition; dotted lines indicate one-way competition. (C) Percentage (%) and median fluorescence intensity (MFI) of Pb-PfCSP-SPZ bound by 20 μg/mL of indicated mAb, measured by flow cytometry. VRC01 and L9 (NVDP-preferring repeat mAb) were included as negative and positive controls. (D) Liver burden (bioluminescence; total flux, photons/sec) in mice 40 hours post-challenge (n = 5/group; line indicates geometric mean) mediated by 300 μg of indicated mAbs administered 2 hours before IV challenge with 2,000 Pb-PfCSP-SPZ. CIS43 (NPDP-preferring repeat mAb) was included as a positive control. Vertical lines separate independent experiments. P-values were determined by comparing mAbs to untreated control using the Kruskal-Wallis test with Dunn’s post-hoc correction. **, p<0.01; ns (not significant), p>0.05.
Fig 3.
C-CSP mAbs differentiate between PfCSP conformations on Pb-PfCSP-SPZ.
(A) Representative flow cytometry plots depicting 20 μg/mL of indicated mAb binding to Pb-PfCSP-SPZ harvested from the salivary glands (SG-SPZ, green) or midguts (MG-SPZ, blue) of mosquitos. (B) Percentage and MFI of SG vs. MG Pb-PfCSP-SPZ bound by 20 μg/mL of indicated mAb, measured by flow cytometry. VRC01 and L9 (NVDP-specific repeat mAb) were included as negative and positive controls.
Fig 4.
PfCSP repeat mAbs potentiate the SPZ binding of C-CSP mAbs.
(A) Experimental schema for measuring the binding of Alexa Fluor 750-labeled mAbs to SG Pb-PfCSP-SPZ or PfSPZ in the presence of unlabeled mAbs. (B) Representative flow cytometry plots depicting 20 μg/mL mAb15-A750 (top panel) or L15-A750 (bottom panel) binding to SG PfSPZ in the presence of 20 μg/mL unlabeled VRC01, 5D5, CIS43, and repeat mAb mix (mAb mix, CIS43-317 in C). Percentages of mAb15+ or L15+SPZ are shown. (C) Percentage and MFI of Pb-PfCSP-SPZ (filled squares) and PfSPZ (open squares) bound by 20 μg/mL mAb15-A750 when co-incubated with 20 μg/mL of specified unlabeled mAb. P-values were determined by comparing PfCSP mAbs to VRC01 using the Kruskal-Wallis test. (D) Dose-dependent improvement in PfSPZ binding of four A750-labeled C-CSP mAbs (20 μg/mL) mediated by increasing concentrations (0.032–100 μg/mL) of unlabeled NANP-preferring mAb10; 100 μg/mL unlabeled VRC01 was set as the baseline (dotted line). Data are representative of two independent experiments. (E) Binding of four A750-labeled C-CSP mAbs (20 μg/mL) to Pb-PfCSP-SPZ (filled symbols) or PfSPZ (open symbols) when co-incubated with unlabeled mAb10 (20 μg/mL), with or without the protease inhibitor E-64 (squares, no E-64; circles, + E-64). P-values were determined by comparing -E-64 to +E-64 for each C-CSP mAb using a two-way ANOVA with Sidak’s post-hoc correction. (C, E): ***, p<0.001; **, p<0.01; *, p<0.05; ns (not significant), p>0.05.
Fig 5.
Repeat mAbs do not potentiate the SPZ neutralization of C-CSP mAbs in vivo.
(A) Liver burden in mice (n = 5/group; line indicates geometric mean) 40 hours post-challenge mediated by indicated mAb combinations (mAb10, 100 μg; C-CSP mAbs or VRC01, 300 μg) administered 2 hours before IV challenge with 2,000 Pb-PfCSP-SPZ. (B) Liver burden (top) and parasitemia (bottom) in mice (n = 5-10/group) respectively 40 hours and 5 days post-challenge mediated by indicated mAb combinations (mAb10, 50 μg in left experiment, 30 μg in right experiment; C-CSP mAbs or VRC01, 300 μg) administered 24 hours before ID challenge with 5,000 Pb-PfCSP-SPZ. (A-B): P-values were determined by comparing C-CSP mAb + mAb10 (orange/purple squares) to mAb10 + VRC01 (purple squares) using either the two-tailed Mann-Whitney test (2 comparisons) or the Kruskal-Wallis test with Dunn’s post-hoc correction (>2 comparisons). Purple dotted lines were set at mAb10 + VRC01; vertical lines separate independent experiments. (C) MFI of Pb-PfCSP-SPZ (filled squares) and PfSPZ (open squares) bound by 2 μg/mL mAb10-A750 when co-incubated with 100 μg/mL of specified unlabeled mAb. P-values were determined by comparing PfCSP mAbs to VRC01 using the Kruskal-Wallis test. (A-C): ***, p<0.001; **, p<0.01; *, p<0.05; ns (not significant), p>0.05.
Fig 6.
Repeat mAb combinations do not cooperatively neutralize SPZ in vivo and antagonize the SPZ binding of other repeat mAbs in vitro.
(A) Liver burden in mice 40 hours post-challenge (n = 10/group) mediated by CIS43, L9, and 317 alone (50 μg) and in combination with each other or isotype control VRC01 (25+25 μg) administered 2 hours before IV challenge with 2,000 Pb-PfCSP-SPZ. P-values were determined by comparing PfCSP mAb combinations to L9 or 317 alone using the two-tailed Mann-Whitney test (solid lines) or PfCSP mAb combinations to their respective PfCSP mAb + VRC01 controls using the Kruskal-Wallis test (dotted lines). (B) MFI of Pb-PfCSP-SPZ bound by 2 μg/mL L9-AF750 when co-incubated with varying concentrations (0.032–100 μg/mL) of unlabeled CIS43, L9, or 317. (C) Liver burden in mice 40 hours post-challenge (n = 5/group) mediated by L9, CIS42, F10, and mAb4 alone (50 μg) and in combination (25+25 μg) administered 2 hours before IV challenge with 2,000 Pb-PfCSP-SPZ. P-values were determined by comparing mAb combinations to L9 alone using the two-tailed Mann-Whitney test. (D) MFI of Pb-PfCSP-SPZ bound by 2 μg/mL L9-AF750 when co-incubated with varying concentrations (0.032–100 μg/mL) of unlabeled CIS42, F10, or mAb4. (E) MFI of Pb-PfCSP-SPZ (filled squares) and PfSPZ (open squares) bound by 2 μg/mL L9-A750 when co-incubated with 100 μg/mL of specified unlabeled mAb. P-values were determined by comparing PfCSP mAbs to VRC01 using the Kruskal-Wallis test. (A, C): lines represent geometric mean. (B, D): dotted lines represent L9-AF750 co-incubated with 100 μg/mL unlabeled VRC01; data are representative of two independent experiments. (A, C, E): ***, p<0.001; **, p<0.01; *, p<0.05; ns (not significant), p>0.05.
Fig 7.
Combining R21 vaccination and passive transfer of PfCSP repeat mAbs provides improved protection against malaria.
(A) MFI of Pb-PfCSP-SPZ bound by 0.2 μg/mL CIS43-, L9-, or 317-AF750 when co-incubated with PBS + 10% FBS (PBS-FBS) or 1:5 diluted serum pooled from either one naïve US adult volunteer, ten naïve mice immunized 3x with 1 μg R21 + adjuvant, fifteen naïve US adults immunized 3x with 9x105 irradiated PfSPZ, and ten Malian children/adults naturally exposed to malaria. For each PfCSP repeat mAb, P-values were determined by comparing each serum type to the PBS-FBS control using a two-way ANOVA with Bonferroni’s post-hoc correction. (B) Endpoint titers of pooled serum from A binding to FL-rCSP measured by ELISA. P-values were determined by comparing serum types to naïve serum using the Kruskal-Wallis test. (C) R21 immunization scheme in normal mice, which received two intramuscular injections of 1 μg R21 + adjuvant at three week intervals prior to passive transfer of PfCSP repeat mAbs two hours before IV challenge with 2,000 Pb-PfCSP-SPZ. (D-F) Liver burden 40 hours after IV challenge with 2,000 Pb-PfCSP-SPZ in mice (n = 5-20/group; 50μg/25μg data in E/F were combined from two experiments, circles and squares) immunized with 1 μg R21 alone; mice administered CIS43, L9, and 317 alone (D, 150 μg; E, 50 μg; F, 25 μg); and mice immunized with 1 μg R21 and administered CIS43, L9, and 317 (D, 150 μg; E, 50 μg; F, 25 μg). P-values were determined by comparing each R21 + mAb combination to the R21 alone or respective mAb alone groups using the Kruskal-Wallis test. (A-B, D-F): ***, p<0.001; **, p<0.01; *, p<0.05; ns (not significant), p>0.05.