Gene copy number and function of the APL1 immune factor changed during Anopheles evolution

Abstract

Background: The recent reference genome assembly and annotation of the Asian malaria vector Anopheles stephensi detected only one gene encoding the leucine-rich repeat immune factor APL1, while in the Anopheles gambiae and sibling Anopheles coluzzii, APL1 factors are encoded by a family of three paralogs. The phylogeny and biological function of the unique APL1 gene in An. stephensi have not yet been specifically examined.

Methods: The APL1 locus was manually annotated to confirm the computationally predicted single APL1 gene in An. stephensi. APL1 evolution within Anopheles was explored by phylogenomic analysis. The single or paralogous APL1 genes were silenced in An. stephensi and An. coluzzii, respectively, followed by mosquito survival analysis, experimental infection with Plasmodium and expression analysis.

Results: APL1 is present as a single ancestral gene in most Anopheles including An. stephensi but has expanded to three paralogs in an African lineage that includes only the Anopheles gambiae species complex and Anopheles christyi. Silencing of the unique APL1 copy in An. stephensi results in significant mosquito mortality. Elevated mortality of APL1-depleted An. stephensi is rescued by antibiotic treatment, suggesting that pathology due to bacteria is the cause of mortality, and indicating that the unique APL1 gene is essential for host survival. Successful Plasmodium development in An. stephensi depends upon APL1 activity for protection from high host mortality due to bacteria. In contrast, silencing of all three APL1 paralogs in An. coluzzii does not result in elevated mortality, either with or without Plasmodium infection. Expression of the single An. stephensi APL1 gene is regulated by both the Imd and Toll immune pathways, while the two signaling pathways regulate different APL1 paralogs in the expanded APL1 locus.

Conclusions: APL1 underwent loss and gain of functions concomitant with expansion from a single ancestral gene to three paralogs in one lineage of African Anopheles. We infer that activity of the unique APL1 gene promotes longevity in An. stephensi by conferring protection from or tolerance to an effect of bacterial pathology. The evolution of an expanded APL1 gene family could be a factor contributing to the exceptional levels of malaria transmission mediated by human-feeding members of the An. gambiae species complex in Africa.

Keywords: Gene essentiality; Gene family; Gene neofunctionalization; Insect immunity; Mosquito.

Fig. 1

The APL1 gene underwent an expansion in an African Anopheles lineage. Anopheles phylogenetic tree indicates the number of APL1 gene paralogs present in the genome of 18 Anopheles species. Geographical locations of species and the number of APL1 genes in each species are indicated in columns, “Location” and “# of APL1 genes”, respectively. Anopheles species worldwide, including An. funestus in Africa, carry a single APL1 gene, which is the ancestral state. An exclusively African lineage displays an increased number of APL1 paralogs, including the Gambiae species complex and An. christyi (expanded APL1 lineage indicated by shaded box). The five sequenced species of the An. gambiae complex clearly carry three APL1 paralogs, while An. christyi carries more than one and possibly three, but the genome assembly is poor, thus indicated as > 1 APL1 gene. Phylogeny modified from [35]

Fig. 2

Depletion of APL1 leads to mosquito mortality in Anopheles stephensi. Survival curves of An. stephensi depleted for APL1 activity by dsAPL1 treatment (red lines) as compared to dsGFP-treated controls (green lines) under different experimental conditions. a Sugar-fed mosquitoes. b Mosquitoes fed an uninfected normal blood meal. c Mosquitoes fed a Plasmodium yoelii-infected blood meal. d Mosquitoes treated with antibiotics and fed a P. yoelii-infected blood meal. Replicate experiments are distinguished by line type (plain, dashed or dotted, respectively). X-axis indicates time after the start of recording. A Cox proportional hazards regression model was fitted to the data using treatment and replicate as predictor terms. The P-value associated with the dsRNA treatment term of the Cox model is shown on each panel. Panel a Wald statistic = 4.195, df = 1, P = 2.75e⁻⁵; Panel b Wald statistic = 3.648, df = 1, P = 0.0003; Panel c Wald statistic = 8.376, df = 1, P < 2e⁻¹⁶; Panel d Wald statistic = 1.1518, df = 1, P = 0.129

Fig. 3

Simultaneous depletion of all three APL1 paralogs in Anopheles coluzzii does not cause mosquito morality. a Survival curves of An. coluzzii depleted for APL1 activity by dsAPL1 treatment (red lines) as compared to dsGFP controls (green lines), for sugar-fed mosquitoes. b Survival curves for mosquitoes fed a Plasmodium yoelii-infected blood meal. Survival curves from replicates are distinguished by line type (plain, dashed or dotted, respectively). X-axis indicates time after the start of recording, not mosquito age (see Methods). A Cox proportional hazards regression model was fitted to the data using treatment and replicate as predictor terms. The P-value associated with the dsRNA treatment term of the Cox model is shown on each panel. Panel a Wald statistic 0.95, df = 1, P = 0.34; Panel b Wald statistic = 1.589, df = 1, P = 0.112

Fig. 4

Anopheles stephensi APL1 protection from Plasmodium yoelii infection is secondary to its antibacterial function. a P. yoelii oocyst infection intensity in An. stephensi mosquitoes treated with dsAPL1 or control dsGFP, both without antibiotic treatment. Oocyst intensity is the oocyst count in mosquitoes with ≥ 1 oocyst, to avoid confounding with infection prevalence. Oocyst infection prevalence, the proportion of mosquitoes carrying ≥ 1 oocyst, is indicated as a percentage below sample sizes. Number of biological replicates is indicated below plots. Combined P-value: χ² = 22.3529, df = 4, P = -0.0002 (Replicate 1, W = 30.5, P = 0.0075; Replicate 2, W = 226.5, P = 0.002). b As in a, but mosquitoes were subject to antibiotic treatment before Plasmodium exposure. Combined P-value, χ² = 21.85, df = 6, P = 0.001 (Replicate 1, W = 1144.5, P = 0.009; Replicate 2, W = 463.5, P = 0.043; Replicate 3, W = 40, P = 0.05549)

Fig. 5

Transcription of Anopheles stephensi APL1 is regulated by both Toll and Imd immune signaling pathways. Regulation of expression of the unique An. stephensi APL1 gene was queried by silencing the negative regulator of Toll, Cactus (a and b) or the positive regulator of Imd, Rel2 (c and d). a Cactus expression is efficiently repressed by treatment with dsRNA targeting Cactus (dsCactus). Graph indicates fold change of Cactus expression by dsCactus treatment as compared to dsGFP controls. b APL1 expression is augmented by silencing of Cactus, which constitutively activates the Toll pathway. Graph indicates fold change of APL1 gene expression in An. stephensi depleted for Cactus by dsCactus treatment, relative to dsGFP treated controls. c Rel2 expression is efficiently suppressed by treatment with dsRNA targeting Rel2 (dsRel2). Graph indicates fold change of Rel2 expression by dsRel2 treatment as compared to dsGFP controls. d APL1 expression is diminished by silencing of Rel2, which inhibits Imd pathway activity. Graph indicates fold change of APL1 gene expression in An. stephensi depleted for Rel2 by dsRel2 treatment, relative to dsGFP treated controls. Transcript abundance is measured by quantitative RT-PCR in two biological replicates as indicated