The JAK-STAT pathway controls Plasmodium vivax load in early stages of Anopheles aquasalis infection

Abstract

Malaria affects 300 million people worldwide every year and 450,000 in Brazil. In coastal areas of Brazil, the main malaria vector is Anopheles aquasalis, and Plasmodium vivax is responsible for the majority of malaria cases in the Americas. Insects possess a powerful immune system to combat infections. Three pathways control the insect immune response: Toll, IMD, and JAK-STAT. Here we analyze the immune role of the A. aquasalis JAK-STAT pathway after P. vivax infection. Three genes, the transcription factor Signal Transducers and Activators of Transcription (STAT), the regulatory Protein Inhibitors of Activated STAT (PIAS) and the Nitric Oxide Synthase enzyme (NOS) were characterized. Expression of STAT and PIAS was higher in males than females and in eggs and first instar larvae when compared to larvae and pupae. RNA levels for STAT and PIAS increased 24 and 36 hours (h) after P. vivax challenge. NOS transcription increased 36 h post infection (hpi) while this protein was already detected in some midgut epithelial cells 24 hpi. Imunocytochemistry experiments using specific antibodies showed that in non-infected insects STAT and PIAS were found mostly in the fat body, while in infected mosquitoes the proteins were found in other body tissues. The knockdown of STAT by RNAi increased the number of oocysts in the midgut of A. aquasalis. This is the first clear evidence for the involvement of a specific immune pathway in the interaction of the Brazilian malaria vector A. aquasalis with P. vivax, delineating a potential target for the future development of disease controlling strategies.

Figure 1. Characterization of the STAT gene.

A: Schematic representation of STAT protein from A. aquasalis (AqSTAT-A), A. gambiae (AgSTAT-A and AgSTAT-B) and A. aegypti (AeSTAT-A) showing the STAT interaction domain (yellow), STAT alpha domain (green), STAT binding domain (blue) and SH2 domain (red). B: Phylogenetic tree for STAT using insect sequences, constructed based on the neighbor-joining method. C: Multiple aminoacid sequence alignment of STAT of insects. Accession numbers of STAT sequences from: A. aquasalis (Aq) – HM851178, A. gambiae (Ag) (STAT-A – ACO05014.1 and STAT-B – CAA09070.1), A. aegypti (Ae) – ABO72629.1, Culex quinquefasciatus (Cq) – XP_001866606.1, Culex tritaeniorhynchus (Ct) – AAQU64663.1, and D. melanogaster (Dm) – NP_996243.1.

Figure 2. Characterization of PIAS gene.

A: Schematic representation of A. aquasalis (Aq), A. gambiae (Ag) and A. aegypti (Ae) PIAS proteins showing the SAP domain (blue) and the MIZ/SP-RING zinc finger domain (red). B: Phylogenetic tree for PIAS of insects constructed based on the neighbor-joining method. C: Multiple aminoacid sequence alignment of PIAS from insects. Accession numbers of PIAS sequences from: A. aquasalis (Aq) – HM851177, A. gambiae (Ag) – XP_001688469.1, A. aegypti (Ae) – XP_001647815.1, D. pseudobscura (Dp) – XP_002138569, and A. mellifera (Am) – XP_623571.

Figure 3. Transcription levels of A. aquasalis STAT determined by RTPCR.

A: immature stages (eggs, larvae (L1–L4) and pupae), sugar-fed males (♂) and females (♀), B: sugar-fed females (dotted line), blood-fed control (BFC) and blood-fed infected females (BFI). h – hours, L1 – first instar larva, L2 – second instar larva, L3 – third instar larva and L4 – fourth instar larva. +–: s.e.m.; * 0.05>p>0.03, ** 0.03>p>0.01, *** p>0.01. The ANOVA test with multiple comparisons of Tukey or Games-Howell was used in A. In B the ANOVA test with multiple comparisons of Tukey or Games-Howell was used in the comparisons between the blood-fed samples analyses and the Kruskal-Wallis test with multiple comparisons of Dunn's in the blood-infected samples analyses. Bonferroni correction was used when necessary in the analyses of the blood-infected samples.

Figure 4. Expression of PIAS in A. aquasalis.

A: Transcription levels of A. aquasalis PIAS in immature stages (eggs, larvae (L1–L4) and pupae), sugar-fed males and females, B: Transcription levels of A. aquasalis PIAS in sugar-fed females (dotted line), and females after blood-feeding and after P. vivax infection, C: Expression of PIAS protein by western blot in A. aquasalis submitted to different feeding regimens (sugar-fed male (♂) and female (♀), blood-fed (control) (BFC) and blood-fed infected (BFI) females) and human blood. h – hours, L1 – first instar larva, L2 – second instar larva, L3 – third instar larva and L4 – fourth instar larva. +–: s.e.m.; * 0.05>p>0.03, ** 0.03>p>0.01, *** p>0.01. The ANOVA test with multiple comparisons of Tukey or Games-Howell was used in A. In B, the ANOVA test with multiple comparisons of Tukey or Games-Howell was used in the comparisons between the blood-fed samples analyses and the Kruskal-Wallis test with multiple comparisons of Dunn's in the blood-infected samples analyses. Bonferroni correction was used when necessary in the analyses of the blood-infected samples.

Figure 5. Characterization of NOS gene.

A: Schematic representation of A. aquasalis NOS protein showing nitric oxide synthase (green), NADPH-dependent FMN reductase (red) and ferrodoxin reductase (red) domains. B: Phylogenetic tree of insects NOS constructed based on the neighbor-joining method. C: Multiple aminoacid sequence alignment of insects NOS. Accession numbers of NOS sequences from: A. aquasalis (Aq) – HM851179, A. gambiae (Ag) – AGAP008255-PA, A. aegypti (Ae) – AAEL009745, A. stephensi (As) – O61608, and D. melanogaster (Dm) – CG6713.

Figure 6. Expression of NOS in A. aquasalis.

A: Transcription of NOS in A. aquasalis following different feeding regimens determined by RTPCR. A: sugar-fed females (dotted line), blood-fed control (BFC) and blood-fed infected (BFI) females. h – hours. *0.05>p>0.03, ** 0.03>p>0.01, *** p>0.01. The ANOVA test with multiple comparisons of Tukey was used in the analyses. B: Light microscopy of a transversally open midgut of A. aquasalis showing the gut epithelium composed by a single cell monolayer. C and D: Immunofluorescence staining of 24 hours BFC and BFI female guts with a universal anti-NOS antibody showing fluorescent epithelial cells (asterisks) positive for the presence of NOS protein.

Figure 7. Detection of STAT protein in different tissues of A. aquasalis.

A, B, C and D: the figures show the expression of the STAT proteins in sugar-fed (SF) males and females. A and B - control figures. E–J: the figures show the expression of the STAT proteins in females submitted to different feeding regimes. E, G and I – 24, 36 and 48 hours (h) blood-fed control (BFC), respectively; G F, H and I - 24, 36 and 48 h blood-fed infected (BFI), respectively. Arrowheads show the fat body, asterisks represent the eggs and arrows represent disperse cells expressing STAT proteins. To - thorax, Ab - abdomen, eg - eggs and Bl - blood.

Figure 8. Expression of PIAS in different tissues of A. aquasalis insects.

A, B, C and D: the figures show the expression of the PIAS proteins in sugar-fed (SF) males and females. A and B - control figures. E–J: the figures show the expression of the PIAS proteins in females submitted to different feeding regimes. E, G and I – 24, 36 and 48 hours (h) blood-fed control (BFC), respectively; G F, H and I - 24, 36 and 48 h blood-fed infected (BFI), respectively. Arrowheads show the fat body, asterisks represent the eggs and arrows represent disperse cells expressing PIAS proteins. To - thorax, Ab - abdomen, eg - eggs and Bl - blood.

Figure 9. Effect of STAT silencing on A. aquasalis susceptibility to P. vivax infection.

A and B - Effect of dsRNA-mediated knockdown of ß-gal (control) and STAT on A. aquasalis STAT expression 1 to 5 days after dsRNA injection evaluated by semi-quantitative PCR. Day zero refers to A. aquasalis sugar-fed females. C- Number of infected insects after dsRNA injections. D and E - Oocyst numbers (D) and visualization (arrows) (E) in midguts of mosquitoes previously injected with dsRNAfor ß-gal (D, E1 and E2) and STAT (D, E3 and E4) three to five days after Plasmodium infection. Three replicates of each experiment were performed. The significance of gene silencing on oocyst load in experimental samples, compared to dsß-gal-treated controls, was determined by Mann-Whitney statistical test with Bonferroni correction.