The Indian strain of mosquito An. stephensi was reared and maintained in the insectary at a temperature of 28 ± 2°C, with a relative humidity of ∼80% with a 12-h light–dark cycle, as mentioned previously (Sharma et al., 2015; Thomas et al., 2016). Adult mosquitoes were fed daily on sterile sugar solution (10%) using a cotton swab throughout the experiment. All protocols for rearing and mosquito infection to Plasmodium vivax were approved by the Institute Ethics Committee (ECR/NIMR/EC/2012/41).

Experimentally required tissues were dissected and pooled from the cold anesthetized adult female mosquitoes under different physiological conditions. To examine the tissue-specific expression of target genes, selected tissues such as hemocyte, spermatheca, and salivary gland and male reproductive tissues were dissected from 3- to 4-day-old naïve sugar-fed mosquitoes. Further, to check the effect of starvation on the mosquito survival and trehalase expression, we collected the midgut sample at different time points (24, 48, and 72 h) post starvation. To examine the molecular response under nutritional and reproductive physiological conditions, we collected ∼25 post blood-fed mosquito ovaries (4, 24, 48, and 72 h) and fat body (8, 16, 24, and 48 h). Midguts were dissected from 25 mosquitoes fed on rabbit blood and collected at different time points (3, 24, 48, 72 h, 10, and 14 days). For the collection of midgut infected with P. vivax sporozoites, 3–4-days-old An. stephensi mosquitoes were fed on the blood of P. vivax-infected patients (∼2% gametocytaemia) through a pre-optimized artificial membrane feeding assay as described earlier (Sharma, 2020). The confirmation of the P. vivax infection was done by staining the midgut with 5% mercurochrome to visualize the oocysts after 4 days of infection (DPI), as described earlier (Sharma, 2020). After confirmation, ∼20–25 mosquito tissues, i.e., midgut and salivary glands, were dissected at selected time points. Total RNA from the salivary gland, midgut, and other tissues was isolated using the standard Trizol method as described previously.

Approximately 1 μg total RNA was utilized for the synthesis of first-strand cDNA synthesis using a mixture of oligo-dT and random hexamer primers and Superscript II reverse transcriptase as per the described protocol (Verso cDNA synthesis Kit, Cat#AB-1453/A, EU, Lithuania; Sharma et al., 2015; Thomas et al., 2016). For differential gene expression analysis, routine RT-PCR and agarose gel electrophoresis protocols were used. Approximately 25 ng of cDNA was used in 10-μL real-time PCR reactions. The relative abundance of the gene of interest was assessed using the SYBR Green qPCR master mix (Thermo Fisher Scientific), using the Bio-Rad CFX96 PCR machine. PCR cycle parameters involved an initial denaturation at 95°C for 15 min, 40 cycles of 10 s at 95°C, 15 s at 52°C, and 22 s at 72°C. After the final extension step, the melting curve was derived and examined for quality control. Each experiment was performed in three independent biological replicates. The relative quantification results were normalized with an internal control (actin) and analyzed by the 2^–ΔΔCt method and statistical analysis was performed using Origin 8.1 software. Differences between test samples and their respective controls were evaluated by the Student’s “t”-test.

For the knockdown of As-TreH, dsRNA primers carrying T7 overhang were synthesized as listed in Supplementary Table 2. The amplified PCR products were examined by agarose gel electrophoresis, purified (Thermo Fisher Scientific Gene JET PCR Purification Kit #K0701), quantified, and subjected to double-stranded RNA synthesis using Transcript Aid T7 high-yield transcription kit (Cat# K044, Ambion, United States), while the dsrLacZ gene of bacterial origin was used as a control. Approximately, ∼69 nl (3 μg/μL) of purified dsRNA product was injected into the thorax of a cold anesthetized 1–2-day-old female mosquito using a nano-injector (Drummond Scientific, CA, United States) as described earlier (Kumari et al., 2020). The silencing of the respective gene was confirmed by quantitative RT-PCR after 3–4 days of dsRNA injection.

In the starvation assay, 2–3-day-old (∼230 adult female) mosquitoes were securely emerged in a cotton cage and kept in the insectary at the temperature of 28 ± 2°C and relative humidity of ∼80% with a 12 h light–dark cycle. The control mosquito group was kept on sugar meal (sugar+water), while the starved mosquito group was deprived of both sugar and water until the end of the experiment. All the test assays were performed in a non-crowded condition, and the number of living mosquitoes was tracked until all had died. In the survival time assay, the live mosquitoes were moved into new cages every alternate day, and the numbers of dead mosquitoes were counted daily basis. Gehan-Breslow-Wilcoxon test was used for statistical analysis of data using Prism 8.0 software (GraphPad Software Inc, CA).

Phylogenetic trees were prepared with selected trehalase protein sequences by the maximum likelihood (ML) method in MEGA X program as described previously (Kumar et al., 2018). We aligned all selected insects and other putative trehalase protein sequences using the ClustalW algorithm, where the reliability of the branching was tested by 1,000 bootstrap value of the replicates. The processed phylogenetic tree was examined based on clusters and nodes formed.

Our recent RNA-Seq study demonstrates that early (3–6-days) P. vivax infection significantly alters the metabolic machinery of the mosquitoes midgut (Tevatiya et al., 2019). Identification of a unique transcript, encoding the trehalase homolog protein, from the P. vivax-infected gut-RNAseq library, prompted further investigation of its possible role in mosquito physiology. The initial BLASTX analysis of the selected 1,839 bp long TreH transcript showed 91.61% identity with the TreH homolog of Anopheles gambiae (AGAP012053-PA), having a conserved trehalase domain (Figure 1A). The AsTreH transcript showed 53.2% identity with the TreH homolog of Tribolium castaneum (Coleoptera) and 52% identity with Apis florea (Hymenoptera) showed a less conserved region. Additionally, six copies of the trehalase gene in Tribolium castaneum (Coleoptera) and three copies of the trehalase gene in Apis florea (Hymenoptera) compared to only two copies in the Diptera reflect functional diversification of the trehalase enzyme during the course of evolution. For example, the existence of several copies of the trehalase gene in Coleoptera may be described as an adaptation to a trehalose-rich diet of insectivorous, detritivorous, and mycophagous species, as trehalose sugar is present in high concentrations in fungi (Nardelli et al., 2019).

A homology search against An. stephensi database further verified that the identified As-TreH is a 1,839 bp full-length (ASTEI06114-RA), a single-copy gene, which encodes a protein of 555 amino acids. Multiple-sequence alignment with other trehalase members, originating from Hymenoptera and Diptera orders, showed the highest identity and conservation with Anopheles mosquitoes (Figures 1B,C). Phylogenetic analysis showed the formation of two independent clades, where each clade defines a separate lineage for trehalase, which is further divided into specific subclades.

To monitor the developmental regulation of trehalase, we performed real-time PCR-based transcriptional profiling. Though As-TreH showed constitutive expression throughout the aquatic developmental stages (Figure 2A), we observed a higher abundance during egg (active embryogenesis) and pupae stages (p < 0.01716) than other larval stages. We also observed that the relative expression of trehalase was higher (p < 0.01476) in adult males than in female mosquitoes (Figure 2A). Tissue-specific transcriptional profiling also showed a dominant expression of trehalase in the midgut (p < 0.000291), spermatheca (p < 0.033) of the adult female, and testis of adult male mosquitoes (Figure 2B). Together, we hypothesize that As-TreH may function in the maintenance of gut-nutritional homeostasis and reproductive physiology.

To test the above hypothesis, first, we evaluated the molecular correlation between sugar/water deprivation and mosquito survival. A 42-h-long deprivation of sugar and water not only caused significant mortality (p < 0.000239) but also elevated the expression of trehalase in surviving mosquitoes (Supplementary Table 1). However, rapid downregulation in response to re-supply of sugar indicated that As-TreH may have a role in the maintenance of nutritional physiology (Figure 3A). Next, to test the possible role in reproductive physiology, we examined the spatial and temporal expression of As-TreH in the midgut, fat body, and ovary of the blood-fed adult female mosquitoes. As-TreH showed a gradual enrichment until 24 h in the fat body (p < 6.03E-06) and 48 h in the ovary (p < 0.032802) (Figures 3B,C), indicating a unique role in egg development in the mosquito. Finally, to establish a functional correlation, we performed dsRNA-mediated gene silencing in virgin 3–4-day-old male as well as female mosquitoes. However, after silencing we did not observe any significant behavioral changes either in virgin or in mated mosquitoes (data not shown); surprisingly, we noticed a 50% reduction (p < 0.03288) in the egg-laying of the As-TreH silenced blood-fed female mosquitoes (Figures 3D,E), indicating that trehalase plays a key role in the regulation of mosquito reproductive physiology.

Previous studies have shown that the depletion of hemolymph trehalose sugar in Plasmodium-infected mosquitoes is likely due to rapid consumption by the fast-developing parasite, to meet their energy requirements (Mack and Vanderberg, 1978; Mack et al., 1979). Being a disaccharide, trehalose could not be utilized directly by the parasite, and also we retrieved a hexose transporter transcript of Plasmodium from RNAseq results; we opined that trehalase may have a role to convert trehalose to a sugar molecule. To test this, we tracked the trehalase expression during the development of oocyst in the gut, and sporozoite invaded salivary glands of P. vivax-infected female mosquitoes. A multifold enriched expression during early to late-stage oocysts in the gut as well as salivary glands (Figures 4A,B) suggests that trehalase may significantly contribute to hydrolyzes of the trehalose to provide glucose for the rapid proliferation of parasites and generation of new biomass.