The STAT pathway mediates late-phase immunity against Plasmodium in the mosquito Anopheles gambiae

Abstract

The STAT family of transcription factors activates expression of immune system genes in vertebrates. The ancestral STAT gene (AgSTAT-A) appears to have duplicated in the mosquito Anopheles gambiae, giving rise to a second intronless STAT gene (AgSTAT-B), which we show regulates AgSTAT-A expression in adult females. AgSTAT-A participates in the transcriptional activation of nitric oxide synthase (NOS) in response to bacterial and plasmodial infection. Activation of this pathway, however, is not essential for mosquitoes to survive a bacterial challenge. AgSTAT-A silencing reduces the number of early Plasmodium oocysts in the midgut, but nevertheless enhances the overall infection by increasing oocyst survival. Silencing of SOCS, a STAT suppressor, has the opposite effect, reducing Plasmodium infection by increasing NOS expression. Chemical inhibition of mosquito NOS activity after oocyte formation increases oocyte survival. Thus, the AgSTAT-A pathway mediates a late-phase antiplasmodial response that reduces oocyst survival in A. gambiae.

Figure 1. Genomic organization, phylogeny and expression of the two An. gambiae STAT genes

(A)Genomic organization of AgSTAT-A and Ag-STAT-B genes. AgSTAT-A has eight introns, while AgSTAT-B is an “intronless” gene located in a different chromosome. (B) Phylogenetic tree based on the sequence alignment of the deduced amino acid sequence of members of the STAT family from different species. (Hs = Homo sapiens, Dm = Drosophila melanogaster, Ct = Culex tritaeniorhynchus, Ae = Aedes aegypti and Ag = Anopheles gambiae). The sequence alignment used to build the dendrogram is shown in Fig. S2. (C) Relative expression levels of AgSTAT-A and AgSTAT-B mRNA in An. gambiae (G3 strain) at different developmental states. (E = embryo, L = larvae, P = pupae, M = male and F = female). Values were normalized using An. gambiae ribosomal protein S7 mRNA levels as an internal reference. (Mean ± SE)

Figure 2. Effect of AgSTAT A and Ag STAT B silencing on the induction of NOS and SOCS mRNAs in Sua 5.1 cells challenged with heat-killed E. coli and M. luteus.

(A)NOS and SOCS mRNA levels in control (C) and infected (I) cells challenged with a mixture of heat-killed E. coli and M. luteus collected 6 h after post-infection. (B) Silencing of Sua 5.1 cells with AgSTAT-A dsRNA reduces the expression of AgSTAT-A, NOS and SOCS 6 h after bacterial challenge, but does not affect AgSTAT-B mRNA expression. (C) Silencing of Sua 5.1 cells with AgSTAT-B dsRNA reduces the expression of AgSTAT-B, AgSTAT-A, NOS and SOCS mRNAs 6 h after bacterial challenge. Data are shown as Mean ± SE and significant differences (p < 0.05) are indicated by an asterisk.

Figure 3. Effect of AgSTAT-A silencing on the induction of NOS and SOCS mRNAs and on the survival in An. gambiae (G3 strain) adult females challenged with bacteria

(A) Effect of silencing AgSTAT-A on AgSTAT-A and AgSTAT B mRNA expression. In (A–E) data are shown as Mean ± SE and significant differences (p < 0.05) are indicated by an asterisk. (B) Effect of silencing AgSTAT-B with a dsRNA probe from the non-conserved 5′ UTR region on AgSTAT-A and AgSTAT-B mRNA expression. (C) Proposed organization of the STAT signaling cascade in An. gambiae adult females. (D) NOS and SOCS mRNA expression in adult females injected with a mixture of live E. coli and M. luteus. (E) Effect of AgSTAT-A silencing on the induction of NOS and SOCS expression 4 h after a bacterial challenge with a mixture of live E. coli and M. luteus. (F) Effect of AgSTAT-A silencing on survival of adult female mosquitoes challenged by injection of a mixture of live E. coli and M. luteus, Asaia siamensis or Pseudomonas sp. (G) Effect of AgSTAT-A silencing on the survival of adult females fed on sugar solutions containing either E. coli (left panel) or M. luteus (right panel). Survival curves in F–G were compared using the Log Rank survival test.

Figure 4. Effect of AgSTAT-A silencing on NOS, SOCS and TEP1 induction and parasite survival in Plasmodium-infected An. gambiae (G3 strain) females

(A) Effect of AgSTAT-A silencing on the basal mRNA levels of AgSTAT-A, NOS and SOCS in uninfected females (whole body). In (A–C), data are shown as Mean ± SE and significant differences (p < 0.05) are indicated by an asterisk. (B) Expression of STAT-A, NOS, SOCS and TEP1 mRNAs in response to P. berghei infection 24 h PI in carcass samples (midgut and ovaries removed). (C) Effect of AgSTAT-A silencing on AgSTAT-A, NOS, SOCS and TEP1 mRNA levels 24 h after P. berghei infection in carcass samples (midgut and ovaries removed). (D) The number of oocysts (GFP-parasites) present in the midgut was determined directly using fluorescence microscopy. AgSTAT-A silencing increases the number of P. berghei oocysts 7 days PI relative to controls injected with dsLacZ (left panels). AgSOCS silencing has the opposite effect, and decreases oocysts numbers (righ panels). These are representative samples to illustrate the phenotypes. (E) Effect of AgSTAT-A silencing on the number of P. berghei oocysts (left panel) 7 days PI (fluorescent GFP-parasites are shown as green dots) and P. falciparum oocysts (right panel) 8 days PI (oocysts were stained with mercurochrome and are shown as orange dots). The effect of silencing AgSOCS on P. berghei infection 7 days PI is shown in the middle panel. Each dot represents the number of parasites in an individual midgut and the line indicates the median number of oocysts. The distributions in (D) were compared using the Kolmogorov-Smirnov test (KS test).

Figure 5. Effect of AgSTAT-A silencing on P. berghei infection in the An. gambiae refractory (R) strain and on different developmental stages of the parasite in the susceptible G3 strain

(A) Effect of AgSTAT-A silencing on P. berghei infection in R females (L35 strain) 7 days PI. Melanized parasites are indicated by black dots and live ones by green dots. (B) Effect of AgSTAT-A silencing on P. berghei infection in G3 females 2 days PI. (C) Effect of AgSTAT-B silencing on P. berghei infection in G3 females 2 days PI. (D) Direct comparison of the effect of AgSTAT silencing on the number of oocyst present 2 and 8 days PI in G3 female mosquitoes fed on the same mouse. Each dot represents the number of parasites in an individual midgut and the line indicates the median number of oocysts. The distributions in (A–D) were compared using the Kolmogorov-Smirnov test (KS test). (E) Diagram of the transit and development of Plasmodium in the mosquito midgut. The early phase of mosquito anti-plasmodial responses limits the number of ookinetes that survive and transform into oocysts. The STAT pathway defines a distinct late phase in the An. gambiae anti-plasmodial responses, by targeting the oocysts stage and reducing the number of early oocysts that complete maturation.

Figure 6. NOS as an effector of late phase immunity in G3 strain mosquitos

(A) Effect of silencing SOCS, NOS and co-silencing of SOCS and NOS on P. berghei infection 8 days PI. Each dot represents the number of parasites in an individual midgut and the line indicates the median number of oocysts. Distributions were compared using the Kolmogorov-Smirnov test (KS test). (B) Effect of silencing NOS on P. berghei infection 2 days PI. Each dot represents the number of parasites in an individual midgut and the line indicates the median number of oocysts. Distributions were compared using the Kolmogorov-Smirnov test (KS test). (C) Effect of oral administration of the NOS inhibitor L-NAME or inactive D-NAME on P. berghei infection 8 days PI. (D) Effect of oral administration of the NOS inhibitor L-NAME or inactive D-NAME on P. berghei infection 8 days PI in adult females injected with dsSOCS. (E) Immunofluorescence staining of Plasmodium-infected midguts 8 days PI. Oocysts express GFP (green) and NOS is detected with a universal anti-NOS antibody (red). (F) Effect of SOCS silencing on NOS and TEP1 mRNA expression on midgut, hemocytes and carcass tissues collected 4 days PI. Data are shown as Mean ± SE and significant differences (p < 0.05) are indicated by an asterisk.