Prostaglandins regulate humoral immune responses in Aedes aegypti

Abstract

Prostaglandins (PGs) are immuno-active lipids that mediate the immune response in invertebrates and vertebrates. In insects, PGs play a role on different physiological processes such as reproduction, ion transport and regulation of cellular immunity. However, it is unclear whether PGs play a role in invertebrate's humoral immunity, and, if so, which immune signaling pathways would be modulated by PGs. Here, we show that Aedes aegypti gut microbiota and Gram-negative bacteria challenge induces prostaglandin production sensitive to an irreversible inhibitor of the vertebrate cyclooxygenase, acetylsalicylic acid (ASA). ASA treatment reduced PG synthesis and is associated with decreased expression of components of the Toll and IMD immune pathways, thereby rendering mosquitoes more susceptible to both bacterial and viral infections. We also shown that a cytosolic phospholipase (PLAc), one of the upstream regulators of PG synthesis, is induced by the microbiota in the midgut after blood feeding. The knockdown of the PLAc decreased prostaglandin production and enhanced the replication of Dengue in the midgut. We conclude that in Ae. aegypti, PGs control the amplitude of the immune response to guarantee an efficient pathogen clearance.

Fig 1. E.cloacae (Ec) stimulates PG production in Aag2 cells and Aedes midgut cells.

(A-B) Aag2 cells were incubated with heat-killed bacteria E. cloacae (Gram negative, Ec) in the presence or absence of ASA for 1 hour (A) or 24 hours (B). PG levels in the supernatant of Aag2 cells incubated with heat killed Ec and ASA. (C) PG detection in situ (by EicosaCell assay) in Aag2 cells challenged with heat-killed bacteria. (D) PG levels in midguts of Ae.aegypti females sugar fed (SF) and antibiotic treated (SF + Ab). Statistical analyses were performed using ANOVA followed by Dunnett’s multiple comparison test for figure A and B. In C, statistical analyses were conducted as an unpaired t-test. Error bars represent mean ± SEM. **P≤0.01, ***P≤0.001. Scale Bars = 7μm.

Fig 2. ASA treatment compromises the proper expression of immune genes and the ability to control bacterial and viral infections.

(A) Immune related genes modulated by ASA treatment in the microarray analysis. Green color indicates down-regulated genes and red color is referred to up-regulated ones. Complete detailed expression data can be seen in S1 Dataset. A complete list of the immune genes regulated by ASA treatment can be seen in S2 Dataset. (B) Number of CFU recovered from the supernatant of Aag2 cells that were pre-treated or not with ASA before Ec exposure. (C-D) Viral RNA recovered from the supernatant of Aag2 cells infected with Dengue and Sindbis virus after pretreatment with Aag2 cells (C) Dengue virus RNA recovered after 4 days post infection (D) Sindbis virus RNA recovered 2 days post infection. Error bars represent mean ± SEM. Unpaired t-test, *P≤0.05, **P≤0.01, ***P≤0.001. Each biological replicate corresponds to a well on a plate and at least three independent experiments were performed per assay. Viral RNA amounts were normalized by the number of cells present in the well, which were determined using trypan blue stain.

Fig 3. In the midgut, PG is required for proper induction of immune genes after blood feeding, and its inhibition compromises survival after bacterial infection.

(A-E) Aedes aegypti female mosquitoes were pretreated with a sugar solution supplemented with ASA for three days, before blood feeding. Twenty -four hours after feeding midgut expression of AMPs (A) defensin A, (B) cecropin D, (C) cecropin G, (D) serpin 27A and (E) gambicin was analyzed by qPCR. Statistical analyses were performed using unpaired t-test comparing sucrose and blood groups in each condition. (F) Mortality rate of mosquitoes after feeding with the bacteria Serratia marcescens (5x10⁸ bacteria/mL) with and without ASA sugar pretreatment. (G) Number of infective Dengue units per midgut (PFU) in mosquitoes pre-treated with ASA for 2 days (Dengue New Guinea C strain). (A) to (F) Error bars represent mean ± SEM. (G) Red bars represent the median of each group. (A) to (F) Unpaired t-test, NS (P>0.05), *P≤0.05, **P≤0.01; Conditions were compared with its correspondent SF control. (A) to (E) Pools of 10 midguts were used for each biological replicate, at least 3 biological replicates were used per condition. (F) Mortality rate across 6 independent survival curves. (G) Number of PFU per mosquito midgut, each dot represents one individual mosquito (control = 69 and ASA = 74).

Fig 4. Cytosolic phospholipase A2 (PLA2c) is induced by the microbiota in the midgut after blood feeding.

(A) Schematic representation of canonical PG production pathway. (B) Alignment of the amino acid sequences of six phospholipases of Aedes aegypti. * represents calcium binding domain; # represents catalytic domain. Gene expression of (C) PLA2c and (D) PLA2s in the midgut and fat body in sugar fed (SF) and blood fed mosquitoes (BF), 24 h post feeding. Gene expression of (E) PLA2c and (F) PLA2s in the midgut with or without the presence of the microbiota. Gene expression of (G) PLA2c and (H) PLA2s in the fat body with or without the microbiota presence. Error bars represent mean ± SEM. Unpaired t-test, NS (P>0.05), *P≤0.05, **P≤0.01, ***P≤0.001, ****P≤0.0001; Conditions were compared with its correspondent SF control. Each biological replicate was a pool of 10 midguts, and each experimental group had at least 3 biological replicates.

Fig 5. PG production in the midgut is dependent on PLA2c expression and its impairment increases Dengue viral loads.

Knock-down efficiency in the mosquito midgut injected with (A) dsPLAc dsRNA and (B) dsPLAs dsRNA, in sugar fed (SF) and blood fed (BF) females. (C) PG levels in the midgut after RNAi silencing of PLAc and PLAs. Mosquitoes injected with dsRNA for LacZ were used as control. (D) Dengue virus RNA levels in the midgut relative to mosquito RP-49 expression. Error bars represent mean ± SEM. Unpaired t-test, NS (P>0.05) and *P≤0.05. Each biological replicate was a pool of 10 midguts, and each experimental group had at least 3 biological replicates.