The interaction between a sexually transferred steroid hormone and a female protein regulates oogenesis in the malaria mosquito Anopheles gambiae

Abstract

Molecular interactions between male and female factors during mating profoundly affect the reproductive behavior and physiology of female insects. In natural populations of the malaria mosquito Anopheles gambiae, blood-fed females direct nutritional resources towards oogenesis only when inseminated. Here we show that the mating-dependent pathway of egg development in these mosquitoes is regulated by the interaction between the steroid hormone 20-hydroxy-ecdysone (20E) transferred by males during copulation and a female Mating-Induced Stimulator of Oogenesis (MISO) protein. RNAi silencing of MISO abolishes the increase in oogenesis caused by mating in blood-fed females, causes a delay in oocyte development, and impairs the function of male-transferred 20E. Co-immunoprecipitation experiments show that MISO and 20E interact in the female reproductive tract. Moreover MISO expression after mating is induced by 20E via the Ecdysone Receptor, demonstrating a close cooperation between the two factors. Male-transferred 20E therefore acts as a mating signal that females translate into an increased investment in egg development via a MISO-dependent pathway. The identification of this male-female reproductive interaction offers novel opportunities for the control of mosquito populations that transmit malaria.

Figure 1. MISO knockdown decreases egg production to virgin levels.

(A) Mated females injected with dsRNA were blood fed and allowed to lay eggs 3 d post-blood-feeding for 4 nights. Control females (dsLacZ) laid on average 82.5 eggs, while dsMISO oviposited a statistically significant lower number of eggs (65.4). The data are representative of three independent replicates. (B) Virgin or mated females injected with dsRNA were blood fed, and eggs developed inside the ovaries were counted 3 d post-blood-feeding without allowing oviposition. Mated dsLacZ produced on average 77.8 eggs, while virgin dsLacZ and mated dsMISO produced a statistically significant lower number of eggs (62.3 and 60.4, respectively). The data are representative of six independent replicates.

Figure 2. MISO silencing alters the expression of the lipid transporter Lipophorin in developing oocytes after blood feeding.

(A and B) Immunofluorescence experiments on ovaries dissected from virgin and mated dsLacZ and mated dsMISO females stained with the lipid-binding reagent Nile-Red (red) at 24 h (A) and 60 h (B) post-blood-feeding. Asterisks in dsMISO and dsLacZ virgin ovaries indicate undeveloped primary follicles. Cell nuclei are labeled with DAPI (blue). Scale bar: 200 µm. (C) qRT-PCR of Lp and Vg from the fat body of virgin and mated dsLacZ and mated dsMISO females 24 h after blood feeding (BF). Expression levels (shown in logarithmic scale) were normalized to the housekeeping gene RpL19. The box-and-whisker diagrams represent five replicates of pools of 6–10 tissues.

Figure 3. MISO silencing affects atrial 20E titers and reduces the activation of 20E-responsive genes after mating.

(A) In vitro ovarian ecdysteroid secretion before and 18 h after a blood meal in virgin and mated dsLacZ and mated dsMISO. Graph shows data from eight individual ovaries. Data are represented as mean ± SEM. Means with the same letter are not significantly different (p>0.05). (B) Changes over time in the geometric mean of the ecdysteroid titer (natural logarithm) of dsLacZ (black solid line with circles) and dsMISO (green dashed line with triangles) females at 5 time points after mating (0.5, 6, 12, 18, and 24 hpm) with the mean trajectories estimated in regression mixed models (dashed and dotted lines). Nine replicates were performed using a pool of three atria each. (C) qRT-PCR of 5 20E-responsive genes (Vg, Lp, EcR, HR3, and USP) in dsMISO and dsLacZ females at different time points (0, 6, 12, 18 h) after mating. The levels of MISO after dsMISO injections are also shown. Four independent replicates were performed using a pool of 5–10 female abdomens. Expression was normalized to the housekeeping gene RpL19. Data are represented as mean ± SEM. One or two asterisks represent p<0.05 and p<0.001, respectively.

Figure 4. MISO is induced by and interacts with 20E in the female atrium.

(A) Western blot under native nondenaturing gel condition using anti-20E and anti-MISO antibodies on atria of virgin or mated (8 hpm) females. Left and right arrowheads indicate 20E and MISO positive bands, respectively. (B) Co-immunoprecipitation of MISO and 20E in atria of virgin or mated (8 hpm) females. The anti-MISO immunoprecipitation (IP) was quantified with anti-20E ELISA. Mated atria showed a mean of 0.68 ng, while in virgin atria no 20E was detected. Three experiments were performed using a pool of 50 atria each, one-third used for the IP (upper panel) and two-thirds for the ELISA (lower panel). Data are represented as mean ± SEM. (C) MISO expression in atria dissected from females previously injected in the hemolymph with three 10-fold dilutions of 20E in ethanol (10% v/v) (starting from 2.5 µg of 20E per mosquito). Ethanol (10% v/v) and water-soluble cholesterol (0.25 µg/mosquito) were used as controls. Expression levels were measured 24 h after injection or mating, and were normalized to RpL19 and then to virgin levels in each replicate. A minimum of three replicates were performed for each condition. Data are represented as mean ± SEM. (D) qRT-PCR of MISO and Vg in females injected with dsEcR, analyzed at 6, 24, and 72 hpm. Expression levels (shown in logarithmic scale as fold changes relative to age-matched virgins) were normalized to the housekeeping gene RpL19. The analysis was performed in four replicates on pools of 5–10 female lower reproductive tracts. Data are represented as mean ± SEM. The asterisk indicates p<0.05.