Forkhead transcription factors regulate mosquito reproduction

Abstract

Forkhead-box (Fox) genes encode a family of transcription factors defined by a 'winged helix' DNA-binding domain. In this study we aimed to identify Fox factors that are expressed within the fat body of the yellow fever mosquito Aedes aegypti, and determine whether any of these are involved in the regulation of mosquito yolk protein gene expression. The Ae. aegypti genome contains 18 loci that encode putative Fox factors. Our stringent cladistic analysis has profound implications for the use of Fox genes as phylogenetic markers. Twelve Ae. aegypti Fox genes are expressed within various tissues of adult females, six of which are expressed within the fat body. All six Fox genes expressed in the fat body displayed dynamic expression profiles following a blood meal. We knocked down the 'fat body Foxes' through RNAi to determine whether these 'knockdowns' hindered amino acid-induced vitellogenin gene expression. We also determined the effect of these knockdowns on the number of eggs deposited following a blood meal. Knockdown of FoxN1, FoxN2, FoxL, and FoxO, had a negative effect on amino acid-induced vitellogenin gene expression and resulted in significantly fewer eggs laid. Our analysis stresses the importance of Fox transcription factors in regulating mosquito reproduction.

Fig. 1

Aedes aegypti Fox genes. A) Designated name. B) Number of AAs (#AA). C) Schematic location of protein domains: Fox, winged helix DNA-binding domain; FHA, forkhead-associated domain; Zn, zinc finger domain. D) Scheme of genomic organization. Exons are marked with boxes, total gene length is indicated above.

Fig. 2

Phylogenetic analysis of the Aedes aegypti Fox genes. A) Alignment of the DNA binding domain from Ae. aegypti Fox factors. Alpha helices, H; beta sheets, S; wings, W. Identical residues in at least 9 of 18 sequences are shaded black, similar residues are shaded grey. B) RAxML best-known likelihood tree (-Ln -9340.947535) for the MAFFT generated alignment of amino acid data and 500 replicates. Bootstrap values for 500 replicates are shown, with those in large font signifying subfamily support.

Fig. 2

Phylogenetic analysis of the Aedes aegypti Fox genes. A) Alignment of the DNA binding domain from Ae. aegypti Fox factors. Alpha helices, H; beta sheets, S; wings, W. Identical residues in at least 9 of 18 sequences are shaded black, similar residues are shaded grey. B) RAxML best-known likelihood tree (-Ln -9340.947535) for the MAFFT generated alignment of amino acid data and 500 replicates. Bootstrap values for 500 replicates are shown, with those in large font signifying subfamily support.

Fig. 3

Expression of Aedes aegypti Fox genes in different tissues of previtellogenic female mosquitoes (PV) and 24 h after a blood meal (PBM). Total RNA of head, hd; thorax, tx; fat body, fb; midgut, gt; malpighian tubules, mt; and ovaries, ov was isolated, treated with DNase I to remove DNA contamination and subsequently used as template for RT-PCR. Below each gel image is a synopsis of RT-PCR results for three independent repeats: +, transcripts detectable in all samples, -, transcript absent in all samples; (-), transcripts absent in two of three samples. Amplification of actin cDNA was used to confirm cDNA integrity (last line).

Fig. 4

Fox gene expression in the fat body of Aedes aegypti without (PV) and 6, 16, 22, 24, 36, 48, and 65 h PBM. Relative Fox mRNA levels were determined by qPCR. Three groups of twenty fat bodies were used per time point, samples from three biological replicates were analyzed, and their mean separated by Tukey-Kramer HSD (p ≤ 0.05). Means with the same letter are not significantly different.

Fig. 5

RNAi-mediated knockdown of specific Fox factors affects Vg gene expression in mosquito in vitro fat body culture. A) RNAi-mediated knockdown of Fox transcripts is highly effective. Mosquitoes were injected with control dsRNA (MAL) or dsRNA for specific Fox factors. Three days post injection fat bodies were isolated and subjected to fat body culture (see below). RNA from fat bodies incubated in AA-containing medium was isolated and used as template for RT-PCR, using gene specific primers for each Fox factor. Samples marked with ‘-‘ are control RNAs isolated from MAL-injected mosquitoes. Actin primers were added to the PCR reaction to confirm the integrity of the cDNA. B) RNAi-mediated knockdown of specific Fox factors decreases Vg gene expression. As above, mosquitoes were treated with dsRNAs and allowed to recover for three days. Fat bodies were then isolated and incubated in fat body culture medium with and without balanced mixture of AAs. Vg mRNA levels were determined by qPCR as described above. Data represent means ± S.E. of triplicate samples. C) Actin gene expression is not altered by RNAi-mediated knockdown of specific Fox factors. Actin mRNA levels in the samples used in B) were determined by qPCR. Data represent means ± S.E. of triplicate samples.

Fig. 6

RNAi-mediated knockdown of specific Fox factors decreases egg deposition. FoxL-, FoxN1-, FoxN2-, and FoxO-knockdown mosquitoes lay significantly fewer eggs than control mosquitoes (MAL). 20 females were analyzed per group. Similar results were obtained in three independent experiments. We used the nonparametric Mann-Whitney (Wilcoxon) test to determine significant differences in mean egg numbers for knockdown mosquito groups compared with the MAL-injected control group. Groups with statistical significant difference in mean egg numbers are marked with a ‘*’.