Programmed autophagy in the fat body of Aedes aegypti is required to maintain egg maturation cycles

Abstract

Autophagy plays a pivotal role by allowing cells to recycle cellular components under conditions of stress, starvation, development and cancer. In this work, we have demonstrated that programmed autophagy in the mosquito fat body plays a critical role in maintaining of developmental switches required for normal progression of gonadotrophic cycles. Mosquitoes must feed on vertebrate blood for their egg development, with each gonadotrophic cycle being tightly coupled to a separate blood meal. As a consequence, some mosquito species are vectors of pathogens that cause devastating diseases in humans and domestic animals, most importantly malaria and Dengue fever. Hence, deciphering mechanisms to control egg developmental cycles is of paramount importance for devising novel approaches for mosquito control. Central to egg development is vitellogenesis, the production of yolk protein precursors in the fat body, the tissue analogous to a vertebrate liver, and their subsequent specific accumulation in developing oocytes. During each egg developmental cycle, the fat body undergoes a developmental program that includes previtellogenic build-up of biosynthetic machinery, intense production of yolk protein precursors, and termination of vitellogenesis. The importance of autophagy for termination of vitellogenesis was confirmed by RNA interference (RNAi) depletions of several autophagic genes (ATGs), which inhibited autophagy and resulted in untimely hyper activation of TOR and prolonged production of the major yolk protein precursor, vitellogenin (Vg). RNAi depletion of the ecdysone receptor (EcR) demonstrated its activating role of autophagy. Depletion of the autophagic genes and of EcR led to inhibition of the competence factor, betaFTZ-F1, which is required for ecdysone-mediated developmental transitions. Moreover, autophagy-incompetent female mosquitoes were unable to complete the second reproductive cycle and exhibited retardation and abnormalities in egg maturation. Thus, our study has revealed a novel function of programmed autophagy in maintaining egg maturation cycles in mosquitoes.

Figure 1. Autophagy is active during the termination phase of vitellogenesis in A. aegypti female mosquitoes.

(A) Autophagy in the fat body of the mosquito was assessed by lysotracker staining at time points 0, 16, 24, 36 and 44 h PBM (A- I, III, V, VII, and IX). Eggs from same batches of mosquitoes are shown. Scale bar (black bar) is 50 µm. (A- II, IV, VI, VIII, and X) where (B) follicle length was plotted versus time PBM. Data shown are mean of 5 individual follicles. (C) Western blot analyses of utilizing antibodies against phosphorylated S6K (S6K-P), native S6K, Vg, Atg8 and actin. Fat bodies were analyzed at time points 0, 16, 24, 36, and 44 h PBM. (D) Transcript analysis of ATG1 and ATG8 in female fat bodies at time points 0, 16, 24, 36, and 44 h PBM. Data shown in (D) are three biological replicates and are illustrated as mean ±SEM.

Figure 2. Immunolocalization of Vg and ATG8 within the mosquito fat body.

(A) Vg and ATG8 were localized by means of fluorescent immunocytochemistry in the fat body at 0, 24, 36, and 44 h PBM. Vg was labeled by monoclonal antibodies against Aedes Vg small subunit (apoprotein), followed by anti-mouse antibodies conjugated to Texas-Red (red), while ATG8 with polyclonal antibodies against Aedes ATG8 followed by anti-rabbit antibodies conjugated to FITC (green). Images were merged. Individual images can be found in Fig. S2. (B) Co-localization of ATG8 and Vg in the fat body at 36 h PBM, where ATG8 and Vg were labeled as in (A). Labeling is shown separately for ATG8 and Vg and merged, where co-localization is shown as yellow. Nuclei are stained with Hoescht 33342.

Figure 3. ATG1 and ATG8 are required for autophagy induction in the A. aegypti female fat body.

(A) ATG8i knockdown specificity was tested using immunoblot analysis. ATG8 RNAi resulted in a depletion of ATG8 protein; while it was present after either Mali or ATG6i treatments. The level of Vg was not affected after any treatment. ATG8 was detected by anti-Drosophila ATG8 antibody and Vg by monoclonal antibodies against Aedes Vg. Fat bodies from female mosquitoes 24 h PBM. (B) Both ATG1 and ATG8 were knocked down in single and double RNAi experiments, where these backgrounds were assessed for lysotracker staining of the fat body at 36 h PBM. Note stronger effect of the double ATG1+ATG8 RNAi depletion than individual RNAi of either this ATGs. Mali did not exhibit any changes in lysotracker staining compared to untreated control (Fig. 1A, VII). Scale bar (white bar) is 50 µm.

Figure 4. Autophagy-incompetent mosquitoes are unable to terminate vitellogenesis in a timely manner.

ATG8 and Vg expression was assessed by means of immunofluorescence in the fat body at 24, 36 and 48 h PBM in MALi and ATG1+8i backgrounds. ATG8 is labeled with polyclonal antibodies against Aedes ATG8 followed by secondary anti-rabbit antibodies conjugated to FITC (green) and Vg with monoclonal antibodies against Aedes Vg small subunit followed by secondary anti-mouse antibodies conjugated to Texas-RED (red). Only merged images are illustrated. Individual images can be found in Figs S6 and S9.

Figure 5. RNA interference depletions of several autophagic genes (ATGs) resulted in untimely activation of TOR and prolonged production of Vg.

(A, B) Western blot analyses utilizing antibodies against phosphorylated S6K (S6K-P), Vg and native S6K. (A) S6K-P phosphorylation and Vg production in fat bodies from MALi, ATG1i, ATG8i, and ATG1+8i (A) and ATG1+6i (B) backgrounds at 24, 36 and 48 h PBM. Native S6K was used as a loading control.

Figure 6. Effect of EcR and ATG1 depletions.

(A) EcR was depleted by RNAi and compared with ATG1i (positive control) and MALi (negative control) for transcript analysis of ATG1, ATG8, EcR, and betaFTZ-F1_A at 24, 36, and 48 h PBM in fat bodies of female mosquitoes. Data shown are two or three biological replicates and presented as mean ±SEM. Data are normalized relative to S7. (B) Fat bodies from EcRi, ATG1i, and MALi were analyzed for lysotracker staining at 36 h PBM. Scale bar (white bar) is 20 µm. (C) Fat bodies from untreated female mosquitoes were analyzed for expression of betaFTZ-F1_A at time points 0, 16, 24, 36, and 44 h PBM.

Figure 7. Autophagy-incompetent mosquitoes failed to induce competence factor betaFTZ-F1_A.

Autophagy-incompetent backgrounds (ATG1i, ATG6i, ATG8i, ATG6+8i, and ATG1+6i) were analyzed for up-regulation of the competence factor betaFTZ-F1_A at 48 h PBM. Data shown are two or three biological replicates and are illustrated as mean ±SEM. An unpaired Student's t test was used for comparison for all ATG-incompetent backgrounds compared with MALi. All comparisons to MALi had significant P values of <0.05.

Figure 8. Second cycle of egg development was severely affected in autophagy-incompetent mosquitoes.

(A) Autophagy-incompetent background ATG6+8i demonstrated severe defects in egg development, illustrating two phenotypes: (i) small or (ii) fewer in number and bigger, compared with the negative control MALi. Scale bar is 1 mm. (B) Follicle lengths for multiple autophagy-incompetent backgrounds (EcRi, ATG1i, ATG1+6i, ATG8i, and ATG1+8i) were smaller than for MALi control. Data shown are individual follicle sizes and are illustrated as mean ±SEM. For EcRi and ATG1i (n = 10). For ATG1+6i (n = 15). For ATG1i, ATG8i, ATG1+8i (n = 12). An unpaired Student's t test was used for comparison. P values<0.0001 are designated with *.