Antibodies to a single, conserved epitope in Anopheles APN1 inhibit universal transmission of Plasmodium falciparum and Plasmodium vivax malaria

Abstract

Malaria transmission-blocking vaccines (TBVs) represent a promising approach for the elimination and eradication of this disease. AnAPN1 is a lead TBV candidate that targets a surface antigen on the midgut of the obligate vector of the Plasmodium parasite, the Anopheles mosquito. In this study, we demonstrated that antibodies targeting AnAPN1 block transmission of Plasmodium falciparum and Plasmodium vivax across distantly related anopheline species in countries to which malaria is endemic. Using a biochemical and immunological approach, we determined that the mechanism of action for this phenomenon stems from antibody recognition of a single protective epitope on AnAPN1, which we found to be immunogenic in murine and nonhuman primate models and highly conserved among anophelines. These data indicate that AnAPN1 meets the established target product profile for TBVs and suggest a potential key role for an AnAPN1-based panmalaria TBV in the effort to eradicate malaria.

FIG 1

Rabbit anti-AnAPN1^60–195 antibodies block development of naturally circulating P. falciparum and P. vivax. Membrane-feeding assays conducted with blood collected from gametocytemic volunteers in Yaoundé, Cameroon, demonstrated that rabbit anti-AnAPN1^60–195 IgG (open circles) inhibited P. falciparum oocyst development in An. gambiae Kisumu (A and B) or Ngousso (C and D) strain mosquitoes in a dose-dependent manner compared to results with normal AB serum (filled circles) in all experiments. Rabbit anti-AnAPN1^60–195 IgG similarly inhibited P. vivax oocyst development in An. dirus mosquitoes in Mae Sod (E) or Kanchanaburi (F), Thailand. Mosquito midguts were dissected 7 or 8 days post-blood feeding, and the number of oocysts per midgut was determined for each antibody concentration (0.1 to 1.6 mg/ml) indicated on the x axis for each gametocytemic blood donor, as indicated by the codes below. Horizontal bars represent mean oocyst numbers. For each experiment, each antibody concentration that was significantly effective at reducing oocyst intensity is indicated with an asterisk.

FIG 2

Immunization with rAnAPN1^60–195 elicits a potent, long-lasting antibody response in multiple animal models. Titers of antigen-specific antibody detected by ELISA in the serum (diluted 1:100) of individual BALB/c (A) or Swiss Webster (B) mice immunized with rAnAPN1^60–195-IFA or IFA only or NHPs (C) immunized with rAnAPN1^60–195-Alhydrogel at each time point during the study, as indicated on the x axis. Mean absorbances ± 1 standard deviation at 450 nm from triplicate wells are plotted. Each line represents an individual animal. (D) Native AnAPN1 present in An. gambiae brush border microvillus (AgBBMV) protein lysates (10 μg/lane) is recognized by total IgG (10 μg/ml) purified from BALB/c (lane 1), SWP1 (lane 2), and SWP1+P9 (lane 3) mice and NHP (lane 4) rAnAPN1^60–195 antisera by SDS-PAGE and Western blotting. When probed with peptide 9-specific antibodies (10 μg/ml) purified from SWP1+P9 (SW; lane 5) or NHP (NHP; lane 6) total anti-rAnAPN1^60–195 IgG, the same protein-banding pattern was observed.

FIG 3

Anti-AnAPN1^60–195 antibody recognition of a single, linear B cell epitope confers transmission-blocking efficacy. Standard membrane-feeding assays were performed to assess the transmission-blocking efficacy of BALB/c (A) or Swiss Webster (B) mouse (according to epitope profile, SWP1 or SWP1+P9) or NHP (C) anti-AnAPN1^60–195 antibodies (open circles) against P. falciparum (NF54) in An. gambiae mosquitoes. Enumeration of oocysts per midgut determined 8 days post-blood feeding revealed that oocyst intensity and prevalence were reduced compared to those for the control (IFA) IgG (filled circles) only by anti-AnAPN1^60–195 antibodies recognizing peptide 9. This inhibition was abrogated following depletion of peptide 9-specific antibodies from SW(P1+P9) (B) or NHP (C) anti-AnAPN1^60–195 IgG (blue circles) and recovered when feeding only peptide 9-specific IgG.

FIG 4

Anti-AnAPN1^60–195 antibodies recognize predicted linear B cell and CD4⁺ T cell epitopes. (A) In silico methods utilizing physiochemical properties predict multiple linear B cell epitopes (dashed lines) within AnAPN1^60–195, while data generated by Epitope Identification Suite (Merck Research Labs) predict that several peptides will strongly bind MHC II encoded by a variety of DRB1 alleles (colored lines) represented among Caucasian and East African populations (DRB1*0301, DRB1*0701, and DRB1*1501). Nine peptides capturing these potential immunogenic regions of AnAPN1^60–195 were synthesized for epitope mapping studies (solid black lines). (B to D) rANAPN1^60–195 elicits a strong, long-lasting humoral response in human HLA-DR2 (B), HLA-DR3 (C), or HLA-DR4 (D) transgenic C57BL/6 mice immunized with rAnAPN1^60–195-Alhydrogel (black lines) or Alhydrogel only (control; blue lines), as determined by ELISA. Pooled serum titers (day 70 post-priming immunization) for male (open circles) and female (filled circles) mice are plotted. Optical density (O.D. 450) and reciprocal serum dilutions are plotted. Error bars indicate ±1 standard deviation from results for triplicate wells. (E to G) Two predominant peptides, indicated on the x axis, are recognized by anti-AnAPN1^60–195 antibodies in the serum of BALB/c and Swiss Webster mice (E), rabbits (F), or NHPs (G) by ELISA. Mean ± 1 standard deviation absorbance readings (O.D. 450) from triplicate wells are plotted. ELISA results for an individual animal that is representative of the epitope profiles observed for each host species are shown for AnAPN1^60–195 and preimmune (control) serum.

FIG 5

Anti-peptide 9 antibodies recognize AnAPN1 and inhibit development of P. falciparum. (A and B) Antigen-specific-antibody titers from serum pooled (day 56 post-priming immunization, diluted 1:100) from BALB/c mice immunized with peptide 1 (A) or 9 (B) conjugated to KLH following priming and three boosts, as determined by ELISA. Serum titers for control mice (immunized with KLH) are also plotted. Data represent serum pooled from 5 mice. Optical density (O.D. 450) and reciprocal serum dilutions are plotted. Error bars indicate ±1 standard deviation from triplicate wells. (C) Both anti-peptide 1 (P1) and anti-peptide 9 (P9) antibodies (10 μg/ml) bind rAnAPN1^60–195 (0.5 μg/ml), but only anti-peptide 9 antibodies recognize native midgut AnAPN1 in An. gambiae brush border microvilli (AgBBMV) lysates by Western blotting. Control anti-KLH antibodies did not bind either rAnAPN1^60–195 or AgBBMVs. (D) Total IgG (10 μg/ml) purified from sera of BALB/c mice immunized with peptide 9 (P9) but not peptide 1 (P1) conjugated to KLH (open circles) inhibited P. falciparum oocyst intensity and prevalence in An. gambiae compared to that of control IgG (KLH only; filled circles) in SMFAs. Horizontal bars represent mean oocyst numbers. Data in each panel represent a typical experiment utilizing IgG purified from pooled serum run in triplicate.

FIG 6

Antibodies targeting the highly conserved protective epitope do not inhibit AnAPN1 aminopeptidase activity. (A) Multiple sequence alignment of the An. gambiae (An. gam) AnAPN1^60–195 antigen with putative orthologs in An. funestus (An. fun), An. darlingi (An. dar), and An. albimanus (An. alb), Ae. aegypti (Ae. aeg), and C. quinquefasciatus (Cx. qui) reveals high conservation of AnAPN1^60–195 and the transmission-blocking epitope, peptide 9 (light blue highlights conserved identities among 4 species, dark blue among >4 species. (B and C) Ribbon (B) and space-filling (C) homology models of AnAPN1 based on the crystal structure of human aminopeptidase N suggest that peptide 9 localizes near the binding pocket and catalytic site of AnAPN1. (D) However, mouse and NHP peptide 9-specific antibodies do not inhibit aminopeptidase activity of near-full-length recombinant AnAPN1 (rAnAPN1^60–942) expressed in Drosophila S2 cells compared to results for preimmune IgG in enzymatic assays utilizing an l-leucine p-nitroanilide substrate. The rate of rAnAPN1^60–942 activity was measured as nmol p-nitroaniline (mean ± 1 standard deviation) formed per min at 405 nm. Bestatin and 1,10-phenanthroline were used as controls for inhibition of aminopeptidase and metalloprotease activities, respectively.