Mosquito salivary apyrase regulates blood meal hemostasis and facilitates malaria parasite transmission

Abstract

The evolution of hematophagy involves a series of adaptations that allow blood-feeding insects to access and consume blood efficiently while managing and circumventing the host's hemostatic and immune responses. Mosquito, and other insects, utilize salivary proteins to regulate these responses at the bite site during and after blood feeding. We investigated the function of Anopheles gambiae salivary apyrase (AgApyrase) in regulating hemostasis in the mosquito blood meal and in Plasmodium transmission. Our results demonstrate that salivary apyrase, a known inhibitor of platelet aggregation, interacts with and activates tissue plasminogen activator, facilitating the conversion of plasminogen to plasmin, a human protease that degrades fibrin and facilitates Plasmodium transmission. We show that mosquitoes ingest a substantial amount of apyrase during blood feeding, which reduces coagulation in the blood meal by enhancing fibrin degradation and inhibiting platelet aggregation. AgApyrase significantly enhanced Plasmodium infection in the mosquito midgut, whereas AgApyrase immunization inhibited Plasmodium mosquito infection and sporozoite transmission. This study highlights a pivotal role for mosquito salivary apyrase for regulation of hemostasis in the mosquito blood meal and for Plasmodium transmission to mosquitoes and to the mammalian host, underscoring the potential for strategies to prevent malaria transmission.

Fig. 1. An. gambiae salivary protein AGAP011026 activates tPA.

a Diagram summarizing clotting and fibrinolysis. PTb: pro-thrombin, Tb: thrombin, PLG: plasminogen, Pm: plasmin, PAI: plasminogen activator inhibitor, t-PA: tissue plasminogen activator, α2AP: α−2 antiplasmin. b Workflow diagram to identify the mosquito salivary protein activating tPA. Figure 1b was created with BioRender.com released under a Creative Commons Attribution-NonCommercial-NoDerivs license. c, d Fluorogenic assay for single-chain tPA (sc-tPA) activation reveals that mosquito salivary gland extracts (SG) (c) and saliva (d) activate sc-tPA. Heating the extract at 65 °C or 100 °C prevents tPA activation. n = 2, each performed in duplicates. One-way ANOVA with Šídák’s multiple comparisons test, *P = 0.0214, ****P < 0.0001. e Fractions from An. gambiae salivary gland extracts obtained by size-exclusion chromatography were tested for tPA activation. Representative experiment. n = 1. f Mosquito salivary proteins shortlisted as potential tPA activators. g An. gambiae salivary 5’ nucleotidase ecto (Ag5’NTE) identified as the saliva tPA activator. n = 3. One-way ANOVA with Šídák’s multiple comparisons test, *P = 0.0226, **P = 0.0042. Error bars ± S.E.M. Source Data are provided as a Source Data file.

Fig. 2. The salivary tPA activator AGAP011026 is a mosquito salivary apyrase.

a 5’ nucleotidase ecto activity was measured by the release of inorganic phosphate from AMP using the malachite green assay. n = 2, each performed in duplicates. Two-tailed unpaired t-test, ****P < 0.0001. b AGAP011026 is an apyrase. Inorganic phosphate release from ATP or ADP was measured using Fiske Subbarow reagent. n = 2, each performed in duplicate ns: not significant. Two-tailed unpaired t-test, ns: not significant. c Time course of ADP hydrolysis by AgApyrase. Data shown from one representative experiment (n = 3) performed in duplicates. d Apyrase activity of AGAP011026 was measured using malachite green assay in presence of either 5 mM CaCl2 and/or 5 mM MgCl2 with or without addition of 5 mM EDTA. Representative experiment performed in triplicates. Two-tailed unpaired t-test, ****P < 0.0001, **P = 0.0075 for Mg vs Mg+EDTA, and 0.0020 for Ca vs Ca + Mg + EDTA. e rAgApyrase inhibits the platelet aggregation induced by ADP (orange) in contrast to the control buffer with ADP (blue) which aggregated all the available platelets, measured by light transmittance. Representative experiment. f Interaction of tPA with Ag5’NTE using ELISA overlays. Wells were coated with rAgApyrase and then incubated with or without sc-tPA. Anti-tPA antibodies were used to detect sc-tPA binding. A well was coated with an unrelated protein AGAP007393 as negative control. One-way ANOVA with Dunnet’s multiple comparison test, ****P < 0.0001; *P < 0.0109. g Biolayer interferometry experiments demonstrate the in vitro interaction between biotinylated AgApyrase immobilized on SA biosensors and 250-1000 nM sc-tPA. Data and the 1:1 fit curves are shown in blue and black, respectively. Representative experiment. h Model of the AlphaFold2 and OmegaFold predicted interaction of AgApyrase with human tPA in two orientations, with Apyrase in red (with N and C lobes labeled) and tPA in blue. The buried surface area for the Apyrase-tPA complex final model is 1467 A2. The interaction between the two molecules is mediated by one of tPA the protruding loops (1006-HEALSP-1011, indicated with a green box) in the tPA serine protease domain, which is recognizable and marked by a red box. i Western blot analysis of sc-tPA after incubation with rAgApyrase. The top arrow indicates sc-tPA ( ~ 65 kDa) and the bottom arrow indicates tc-tPA ( ~ 32 kDa) positive controls. No extra cleaved bands or an increase in sc-tPA cleavage are seen in any of the sc-tPA samples incubated with rAgApyrase. M: molecular marker. Error bars indicate SD in (a–d) and (f) indicate standard deviation; and SEM in panel c. Source Data are provided as a Source Data file.

Fig. 3. Salivary AgApyrase is ingested during blood feeding and enhances fibrin degradation in the blood bolus.

a Immunohistochemistry performed on blood fed mosquito midgut shows ingestion of salivary AgApyrase in the mosquito midgut (dark brown patches on the right panel) by staining with anti-apyrase antibodies. Blood fed mosquito midgut with no primary antibody did not show any signal. b Supplementation of the blood meal with rAgApyrase increases D-dimer formation. Mosquitoes were fed on mice before or after intravenous injection of rAgApyrase, and midguts were dissected 30 min post feeding to measure D-dimer formation. Each dot represents a pool of 5 (left) or 10 (right) midguts. N = 4, two replicates each. Two-tailed unpaired t-test, **P = 0.0021 for 5 midguts and 0.0012 for 10 midguts. c Immunohistochemistry was performed on mosquito blood boluses before and after supplementation with rAgApyrase and stained with an anti-fibrinogen antibody (labels fibrinogen and fibrin). d Quantification of the percentage area (left) and the mean signal (right) of fibrinogen (Fbg)/fibrin (Fn) per midgut (additional representative IHC images shown in Fig. S4). Each dot represents an individual midgut. Data from two independent experiments, n = 22 and 26 midguts for control and rAgApyrase, respectively. Two-tailed Mann Whitney test, ***P = 0.0001. e Scanning electron microscopy (SEM) of blood boluses from An. gambiae female mosquitoes fed on mice before or after intravenous injection of rAgApyrase. Note the well organize fibers formed in the blood bolus before the supplementation with rAgApyrase (white arrows) as compared to the sponge-like structure formed after rAgApyrase supplementation (yellow arrows). Yellow dotted pattern shows the region magnified in adjacent panels. f SEM of blood boluses from mosquitoes fed on rAgApyrase immunized mice. Control mosquitoes were fed on mice treated with adjuvant. Midguts were dissected 30 min post feeding. Note the fiber organization and thickness (red arrows) observed in boluses from mosquitoes fed on rAgApyrase immunized mice as compared to the normal fibers (white arrows) from mosquitoes fed on adjuvant treated mice. Representative images (e, f) from a single experiment at two time points (30 min and 4 h post infection). Source Data are provided as a Source Data file.

Fig. 4. Salivary AgApyrase inhibits platelet activation and aggregation in the blood bolus.

a Immunohistochemistry was performed on mosquito blood boluses before and after supplementation with rAgApyrase and stained with P-selectin (a marker for platelet activation). b Quantification of the percentage area and the mean signal of P-selectin staining per midgut (additional representative IHC images shown in Fig. S9). Each dot represents an individual midgut. Data pooled from two independent experiments, n = 19 and 30 midguts for control and rAgApyrase, respectively. Mann Whitney test, ***P = 0.0001. c, d SEM showing platelet aggregation in blood boluses before and after supplementation with rAgApyrase (c) or from mosquitoes fed on adjuvant or rAgApyrase immunized mice (d). White arrows indicate the aggregation of platelets. Red asterisks show fragmentation of platelets into vesicular bodies reminiscent of hypercoagulation. Representative images from duplicate experiments at two time points (30 min and 4 h post infection). Source Data are provided as a Source Data file.

Fig. 5. AgApyrase inhibits platelet-mediated NET formation.

a, b Neutrophils were incubated with platelets in the presence or absence of rAgApyrase. NETs were quantified by immunofluorescence microscopy using an anti-MPO (myeloperoxidase) antibody (a). DNA was stained with DAPI. Yellow arrows point released neutrophil DNA stained with MPO. The percentage of NETs (b) was calculated as an average of 5–10 fields (400X) normalized to total number of neutrophils. Results expressed as mean % ± SEM. n = 4. One-way ANOVA with Šídák’s multiple comparisons test, ***P = 0.0006 for control vs 5 µM and 0.001 for 10 µM. c, d AgApyrase reduces neutrophil ROS production. Whole cell (c) or mitochondrial (d) ROS production in neutrophils activated or not with phorbol myristate acetate (PMA) and incubated in the presence or absence of rAgApyrase. One-way ANOVA with Friedman test. **P = 0.003. n = 6. Error bars indicate SEM. e Immunohistochemistry was performed to quantify neutrophil elastase as a marker for NETosis. f Quantification of the percentage area and the mean signal of neutrophil elastase staining per midgut (additional representative IHC images shown in Fig. S10). Data pooled from two independent experiments. Each dot represents an individual midgut. n = 22 and 35 midguts for control and rAgApyrase, respectively. Mann Whitney test, ****P = 0.0001. Source Data are provided as a Source Data file.

Fig. 6. Effect of AgApyrase on P. berghei transmission.

a AgApyrase facilitates P. berghei infection of mosquito midguts. Oocyst numbers from midguts of An. gambiae mosquitoes that fed on a P. berghei infected mouse before or after the intravenous injection of native or heat-denatured rAgApyrase. Data pooled from three individual experiments shown in Fig. S10, and groups were compared with two-tailed t-test followed by Two-tailed Mann-Whitney comparison test. n = 168, 158, 105 and 117 midguts for before and after in native and denatured, respectively. Red lines indicate median. ****P < 0.0001; *P = 0.0456. b, c AgApyrase immunization inhibits P. berghei midgut infection. Oocyst numbers (b) and infection prevalence (c) were determined in the midguts of An. gambiae mosquitoes fed on P. berghei infected BALB/c mice previously immunized with rAgApyrase using Magic Mouse adjuvant or with adjuvant alone as control. Each dot represents median oocyst number (b) or prevalence (c) from mosquitoes fed on one mouse. n = 10 mice per group. Groups were compared with two-tailed t-test followed by two-tailed Mann-Whitney comparison test. Red lines indicate median. ****P < 0.0001. The oocysts numbers from mosquitoes feeding on each individual mouse are shown on Fig. S11. d, e AgApyrase immunization inhibits sporozoite transmission. BALB/c mice immunized with rAgApyrase in Magic Mouse adjuvant or adjuvant alone (control) were challenged with the bite of five An. stephensi mosquitoes infected with P. berghei sporozoites expressing the luciferase gene. Sporozoite infectivity was determined by measuring luciferase activity in the mouse liver 42 h post challenge. Luminescence signal in the mice livers is shown in panel (d) and the quantification in panel (e). Data pooled from two independent experiments and groups were compared with two-tailed t-test followed by two-tailed Mann–Whitney comparison test. n = 10 and 14 mice for control and AgApyr-imm, respectively. Red lines indicate median. ***P = 0.0002. f, g Similar experiment as in (d, e), but mice were challenged with 5000 sporozoites injected intravenously. Data from a single experiment, n = 5 mice per group. Two-tailed Mann-Whitney comparison test. Source Data are provided as a Source Data file. Figure 6d and f were created with BioRender.com released under a Creative Commons Attribution-NonCommercial-NoDerivs license.

Fig. 7. Model for the role of AgApyrase in the mosquito midgut.

a AgApyrase acts as a classical apyrase with phosphatase activity by hydrolyzing ADP to release AMP and phosphate and therefore, prevents ADP-mediated platelet aggregation. AgApyrase activates tissue plasminogen activator (tPA) which in turn activates plasminogen to plasmin. Plasmin enhances the degradation of fibrin. b Fibrin polymerization is detected in the mosquito midgut within minutes of an infectious blood meal ingestion. The salivary apyrase ingested during blood feeding, enhances fibrin degradation, and inhibits platelet aggregation thus facilitating the migration of Plasmodium gametes in the blood bolus and promoting parasite infection. c Inhibition of apyrase with anti-apyrase antibodies results in the formation of a denser fibrin network and increased platelet aggregation, which interferes with Plasmodium gamete migration and parasite infectivity.