Specific phytochemicals in floral nectar up‐regulate genes involved in longevity regulation and xenobiotic metabolism, extending mosquito life span

Abstract During nectar feeding, mosquitoes ingest a plethora of phytochemicals present in nectar. The ecological and physiological impacts of these ingested phytochemicals on the disease vectors are poorly understood. In this study, we evaluated the effects of three nectar phytochemicals‐‐ caffeine, p‐coumaric acid, and quercetin‐‐on longevity, fecundity, and sugar‐feeding behavior of the Asian tiger mosquito (Aedes albopictus). Adult females of Ae. albopictus were provided continuous access to 10% sucrose supplemented with one of the three phytochemicals and their fecundity, longevity, and the amount of sucrose consumed determined. Transcriptome response of Ae. albopictus females to p‐coumaric acid and quercetin was also evaluated. Dietary quercetin and p‐coumaric acid enhanced the longevity of female Ae. albopictus, while caffeine resulted in reduced sugar consumption and enhanced fecundity of gravid females. RNA‐seq analyses identified 237 genes that were differentially expressed (DE) in mosquitoes consuming p‐coumaric acid or quercetin relative to mosquitoes consuming an unamended sucrose solution diet. Among the DE genes, several encoding antioxidant enzymes, cytochrome P450s, and heat shock proteins were upregulated, whereas histones were downregulated. Overall, our findings show that consuming certain nectar phytochemicals can enhance adult longevity of female Asian tiger mosquitoes, apparently by differentially regulating the expression level of genes involved in longevity and xenobiotic metabolism; this has potential impacts not only on life span but also on vectorial capacity and insecticide resistance.

exchange for pollination services (Ollerton et al., 2011). Insects, primarily hymenopterans, lepidopterans, and dipterans, are major plant pollinators and hence key beneficiaries of plant nectar as a food resource (Ollerton et al., 2011;Peach & Gries, 2020). Mosquitoes are widely assumed to be nectar thieves, and, when compared to other insects, they may be less effective in pollen transfer, but there is abundant evidence that they are important and even essential pollinators for a diversity of flowering plants (Peach & Gries, 2016. In addition to nutrients such as sugars, amino acids, and vitamins that are required by nectar feeders for growth and development, floral nectar contains secondary metabolites (phytochemicals), albeit in smaller concentrations than in foliage and other plant tissues.
These phytochemicals include phenolics, terpenoids, coumarins, and alkaloids, among many other structural types (Adler, 2000;Nicolson & Thornburg, 2007). Although phytochemicals function primarily in defense against herbivores and microorganisms (Bennett & Wallsgrove, 1994), nectar feeders experience both beneficial and detrimental effects from these compounds. Because phytochemicals vary geographically and are taxonomically idiosyncratically distributed, nectar chemistry from the perspective of pollinating insects varies in time and space (Nicolson & Thornburg, 2007).
For most species of mosquitoes, nectar is an essential dietary requirement for adults of both sexes (Foster, 1995). Male mosquitoes feed solely on plant sugars while females of many mosquito species also require a blood meal to complete the gonotrophic cycle.
An energy-rich diet from nectar is known to influence mosquito survival, fecundity, host-seeking behavior, blood feeding, and capacity to transmit pathogens (Foster, 1995;Stone & Foster, 2013). Floral and extra-floral nectars and honeydew comprise the main sources of sugar meals for mosquitoes, although other sources can include fruit juices, plant sap, plant exudates, or even foliage (Foster, 1995;Gary & Foster, 2004;Peach & Gries, 2020). The choice of nectar sources by nectar feeders including mosquitoes is likely determined through a combination of the nutritional quality of the nectar, the visual and olfactory attractiveness, and accessibility of the host plant (Manda et al., 2007;Müller et al., 2011;Nicolson & Thornburg, 2007;Nikbakhtzadeh et al., 2014;Nyasembe et al., 2012).
Studies exploring mosquito-nectar interactions have focused mainly on Anopheles mosquitoes and highlighted the ecological significance of nectar from different plant species. Different nectar sources are known to influence mosquito survival, fecundity, hostseeking behavior, biting rate (Gary & Foster, 2001Impoinvil et al., 2004;Manda et al., 2007;Nikbakhtzadeh et al., 2016), and vectorial capacity (Stone et al., 2012;Ebrahimi et al., 2018), as well as the infection rate and intensity of Plasmodium falciparum in Anopheles gambiae, an observation that led to a suggestion that the mechanism underlying these differences may be due to variation in the phytochemical content of the nectar (Hien et al., 2016).
A growing number of studies have examined the impacts of secondary metabolites on mosquitoes (Johnson & Riehle, 2015;Nunes et al., 2016;Nyasembe et al., 2015). For instance, sucrose diets containing the polyphenols genistein, resveratrol, and quercetin extended the adult life span and reduced the proliferation of gut microbiota in female Ae. aegypti (Nunes et al., 2016). However, a comprehensive understanding of the impacts of nectar phytochemicals on various aspects of mosquito life-history traits, including, sugar-feeding behavior, vector competence, metabolism, and immunity, has remained elusive.
In this study, we used the Asian tiger mosquito Ae. albopictus ( Figure 1) to conduct laboratory assays aimed at determining the impacts of nectar phytochemicals on mosquito behavior and physiology.
Aedes albopictus is an invasive species with high ecological plasticity that has become established in temperate regions of Europe and America (Swanson et al., 2000;Johnson et al., 2017). This containerbreeding species inhabits peri-urban and rural areas and is a vector of epidemiologically important human arboviruses, including dengue and chikungunya, and is also capable of transmitting a wide array of F I G U R E 1 Asian tiger mosquito, Aedes albopictus, resting on a leaf. Photo courtesy of Susan Ellis (www.bugwo od.org) viruses under laboratory conditions (Paupy et al., 2009). Specifically, we examined the effects of the alkaloid caffeine, the phenolic acid p-coumaric acid, and the flavonol quercetin on longevity, fecundity, and sugar-feeding behavior of the mosquitoes. Additionally, using next-generation sequencing, we undertook a whole-transcriptome analysis of female Ae. albopictus consuming sucrose diets supplemented with p-coumaric acid or quercetin to characterize their transcriptional profile. The specific phytochemicals were selected because they are found in nectar, honey, and pollen of many plant species and have been demonstrated to influence the sugar-feeding behavior and to enhance memory (caffeine) of adult honey bees and bumble bees (Singaravelan et al., 2005;Wright et al., 2013), in addition to extending the life span (p-coumaric acid and quercetin) of adult worker honey bees (Liao et al., 2017a) and adult females of the yellow fever mosquito Ae. aegypti by quercetin (Nunes et al., 2016).
Our findings highlight a particularly pronounced effect of certain phytochemicals on the longevity of mosquitoes, which has important ramifications for understanding how mosquito fitness and vectorial capacity can be influenced by the presence of nectar sources in mosquito habitats. Additionally, results from our whole-transcriptome analysis after consumption of nectar phytochemicals suggest further implications for insecticide resistance and mosquito-pathogen interactions.

| Mosquitoes for bioassays
All of the bioassays were conducted at the Medical Entomology Laboratory, Illinois Natural History Survey (INHS), University of Illinois at Urbana-Champaign, using eggs of Ae. albopictus from a colony that was originally collected from Jacksonville, Florida. The adult mosquitoes were generated from a colony reared at 28°C, 80% relative humidity under a 16:8 hr photoperiod (light:dark cycle). Larvae were reared on lactalbumin:yeast (1:1) diet (Sigma-Aldrich, St Louis, USA) in batches of approximately 100 larvae in 22.8 × 30.5 × 7.5 cm enamel pans.
LLC., St. Louis, MO, USA, and used to prepare experimental solutions at concentrations within the natural range documented in nectar, honey, and pollen Martos, Ferreres, Yao, et al., 2000;Serra Bonvehi et al., 2001;Wright et al., 2013;Kaškonienė et al., 2015;Mao et al., 2015;Cheung et al., 2019) of diverse plant species. Both quercetin and p-coumaric acid were dissolved in dimethyl sulfoxide (DMSO; D128, Fisher Scientific International, Inc., Pittsburgh, PA, USA), whereas caffeine was dissolved in deionized (DI) water to make stock solutions. The dietary phytochemicals were prepared fresh from the stock solutions immediately before use. Anthrone reagent  used in the cold-anthrone test applied for the sugar-feeding assays was purchased from Sigma-Aldrich Co. LLC., St. Louis, MO, USA.

| Longevity assays
Newly emerged adult female mosquitoes (1-3 days) were placed in paperboard cages (11 cm height × 9.5 cm diameter) in batches of 25 and fed ad libitum on diets of 10% sucrose containing either caffeine at 50, 100, or 200 ppm; p-coumaric acid at 50, 100, or 200 ppm; or quercetin at 100, 200, or 400 ppm. The control group received 10% sucrose dissolved in deionized water and a solvent control with DMSO. We tested a wide range of concentrations of phytochemicals to determine whether any effects are concentration-dependent.
Each of the 11 treatment combinations was replicated four times, with 25 mosquitoes per replicate. Dead individuals were counted and removed from the cages daily.

| Sugar-feeding behavior
To determine the amount of sucrose consumed by female Ae. albopictus provided with 10% sucrose solution containing caffeine, p-coumaric acid, or quercetin, visual quantification of ingested sucrose was carried out using the cold-anthrone test (Haramis & Foster, 1983). Newly emerged female mosquitoes were placed in paperboard cages (11 cm height × 9.5 cm diameter) in batches of 25 and supplied with 10% sucrose solution for 24 hr, starved for 24 hr, and then provided with the diets described in the longevity assays for 3 hr using the highest concentration of each phytochemical. Each treatment was replicated four times, with 25 mosquitoes per replicate, yielding 500 experimental units. After 3 hr of feeding, individual mosquitoes were placed in a 1.5-ml centrifuge tube and frozen at −80°C for quantification of the amount of sucrose consumed with anthrone. Anthrone solution (2 mg/ml) was prepared by dissolving 200 mg of anthrone reagent in 100 ml of 70% sulfuric acid.
The mosquitoes were thawed and moistened with 1:1 chloroformmethanol solution (2 drops per mosquito) for 20 min to remove cuticular wax. The mosquitoes were then crushed gently with a glass rod and 1 ml of anthrone solution added to each test tube. The test tubes were vortexed and then incubated at 26°C in a water bath for 1 hr. The test tubes were agitated on a vortex mixer halfway through the hour and again at the end of the hour.
Standard sucrose solutions were prepared from 9 twofold serial dilutions of sucrose solutions corresponding to 1,2,4,8,16,32,64,128,256, and 512 µg/µl from an initial stock sucrose solution made by dissolving 51.2 g of sucrose in 100 ml of deionized water, and the resulting dilutions were mixed with anthrone solution. To quantify the amount of sucrose consumed by the mosquitoes, the color strength of each of the experimental tubes (mosquitoes fed with sucrose solutions containing the individual phytochemicals) was compared visually with the color strength of the standards prepared with known amounts of sucrose and designated according to the standard it most closely resembled.

| Fecundity assays
Newly emerged female mosquitoes were maintained in the presence of males in a 1:1 sex ratio in paperboard cages (11 cm height × 9.5 cm diameter) and fed on diets of 10% sucrose containing: 200 ppm caffeine, 200 ppm p-coumaric acid, 400 ppm quercetin, DI water (experimental control), or DMSO (solvent control) for 7 days. The females were then starved for 24 hr and thereafter exposed to a single blood meal of citrate-buffered bovine blood (Hemostat Laboratories, Inc., Dixon, CA) using an artificial membrane feeder system (Hemotek Ltd., Blackburn, UK). Bloodengorged female mosquitoes were immediately isolated in individual containers supplied with 10% sucrose solution containing a dietary phytochemical matching the prestarvation phytochemical to which they were initially exposed, and an oviposition cup. For each treatment, three trials comprising 35 gravid females per trial were conducted. Oviposition was monitored after 3 days, and eggs laid by individual mosquitoes were counted after 5 days. Mosquito body mass was used as a covariate because it is known to affect fecundity (Steinwascher, 1984) and was assessed by measuring the dry weight of the individual mosquitoes after egg-laying. To measure the dry weight of the females, the mosquito specimens were dried at 40°C for 2 days and weighed on a Mettler M-5 balance with a precision of ±0.005 mg.

| Statistical analysis
For the longevity assays, the survival curves for the treatments were

| Mosquito sugar-feeding for RNA sequencing
Newly emerged female Ae. albopictus were placed in paperboard cages as described earlier in batches of 25 females/cage and fed ad libitum on treatments of 10% sucrose solution containing either 400 ppm quercetin or 200 ppm p-coumaric acid. The control group received a 10% sucrose solution. The females were fed for 60 days; this time-point was selected based on the results of the longevity assays. Each of the three treatments was replicated five times. After 60 days, individual mosquitoes from the three treatments were placed in 1.5-ml centrifuge tubes, immediately flash-frozen in liquid nitrogen, and stored in a −80°C freezer until RNA extraction.

| RNA extraction, library construction, and RNA sequencing
One individual adult mosquito was randomly sampled from each of the five replicates per treatment and whole-body RNA extracted using the NucleoSpin RNA® kit (Takara, Japan) according to the

| Quality control and mapping
The quality of the generated reads was assessed with FastQC (Andrews, 2010), and low-quality and adaptor sequences were trimmed using trimmomatic (Bolger et al., 2014). The trimmed reads were mapped to the Ae. albopictus reference genome from the National Center for Biotechnology Information (NCBI), (https:// www.ncbi.nlm.nih.gov/assem bly/GCA_00651 6635.1) using STAR 2.6.0c (Dobin et al., 2013).

| Analysis of differentially expressed (DE) genes
The resulting mapping files in bam format from the previous step were sorted and used to estimate transcript abundance with RSEM (Li & Dewey, 2011). These values were normalized via transcript per million (TPM). In addition, the transcript abundances were crossnormalized using the Trimmed Means of M-values (TMM) method (Li & Dewey, 2011). The TPM values were used to calculate differential gene expression between the three treatments with the R Bioconductor package EdgeR (Robinson et al., 2009)

| Quantitative RT-PCR validation of RNA-Seq data
To validate the results of differential gene expression detected with RNA-Seq, ten candidate genes were selected from the most significantly differentially expressed genes to quantify their relative expression levels in the three treatments by quantitative real-time polymerase chain reaction (qRT-PCR). RNA was extracted from three  (Table S1). The relative expression level of each gene was calculated by the 2 −ΔΔCt method (Schmittgen & Livak, 2008). Relative expression values were assessed with t tests. The Pearson correlation coefficient was calculated between fold changes in transcript accumulation levels for p-coumaric acid and quercetin treatments, as obtained by qRT-PCR and RNA-Seq, respectively.

| Survival
Survival times of female Ae. albopictus provided with dietary caffeine, p-coumaric acid, or quercetin were influenced by the specific phytochemical consumed (χ 2 = 243.06, df = 10, p < .001). Overall, mosquitoes receiving sucrose diets containing p-coumaric acid and quercetin at all concentrations survived the longest compared with those consuming caffeine and control diets. The mosquitoes survived the longest on sucrose diets containing either 100 and 200 ppm p-coumaric acid (median = 74 days), followed by 100 ppm quercetin, 50 ppm p-coumaric acid, 200 and 400 ppm quercetin, controls, and caffeine in decreasing order (median = 72, 70, 69, 60, 58, 50, 58, and 51 days, respectively) ( Figure 2 and Table 1). There were significant differences in survival times between mosquitoes consuming dietary p-coumaric acid, quercetin, and the controls at all concentrations (Table S2). Generally, the survival times of mosquitoes consuming sucrose diets containing caffeine were comparable to survival times on control diets (Figure 2). There was no significant difference in survival time between the control and the 50-ppm F I G U R E 2 Kaplan-Meier survival plots of female Ae. albopictus consuming sucrose diets supplemented with phytochemicals (CAF) caffeine, PC (p-coumaric acid) and QC (quercetin), treatment control (control-DI), and solvent control (control-DMSO). The concentrations are in parts per million (ppm). The survival curve comparisons with log-rank (Mantel-Cox) test revealed significant differences in survival times (p < .001) caffeine diet, but higher concentrations (100 ppm and 200 ppm) of caffeine significantly reduced female Ae. albopictus survival time ( Figure 2 and Table S2).

| Sugar-feeding behavior
There was a significant difference between treatments in the amount of sucrose consumed by female Ae. albopictus (χ 2 = 155.942, df = 4, p < .001). Specifically, mosquitoes receiving a sucrose diet contain-

| Fecundity
After controlling for mosquito body weight, there was a significant   (Figure 4).

| Effects of dietary p-coumaric acid and quercetin on global gene expression
The sequencing resulted in over 2 billion 150 bp paired-end reads across 15 libraries (Table S3). On average per library, about 90% of reads were mapped to the reference Ae. albopictus genome, of which about 62% mapped uniquely, and the rest mapped to multiple locations. The biological replicates of control as well as p-coumaric acid and quercetin treatments clustered closely, indicating that our sequencing data were qualified for identification of differentially expressed genes ( Figures S1 and S2). Of the 29,586 predicted gene-coding transcripts and isoform models (hereafter referred to as "genes") in the reference genome, we recorded 237 that were significantly differentially expressed (DE) across pairwise treatment comparisons, with a cut-off of log 2 fold change of 1.5 (>2.8-fold change) and FDR-corrected pvalue <.05 ( Figure 5 and Table S4). Compared with quercetin diet, consumption of dietary p-coumaric acid by female Ae. albopictus changed the expression of more genes ( Figure 5 and Table S4). In comparison with the control diet, a total of 96 genes were upregulated in female mosquitoes consuming dietary p-coumaric acid and 19 genes in those consuming dietary quercetin. Additionally, 104 genes were downregulated with sucrose diets containing p-coumaric acid and 11 genes with sucrose diets supplemented with quercetin ( Figure 6).
Genes downregulated with p-coumaric acid consumption included multiple genes encoding histone proteins (Table 3).
Of special interest are genes that were differentially expressed in both p-coumaric acid and quercetin treatments, given that the phenotypic effect of enhanced longevity was observed in both groups. There were eight genes upregulated in both treatments, including trans-1,2- Significant GO term enrichment was found in genes upregulated with p-coumaric acid (FDR p-value <.05) ( TA B L E 2 Top 10 up-and downregulated genes in female Ae. albopictus consuming sucrose diets containing p-coumaric acid. A cutoff of fold change ratio of ≥2 and p-value <.05 was applied groups being amino acid metabolism, cell growth and death, immune system, and nervous system. Human disease subgroups were again present with 36 genes in nine subgroups (Table S5).
There were 19 reference pathways affected by genes upregulated with quercetin, with major representation in the xenobiotic biodegradation by the cytochrome P450 group. Only nine reference pathways were affected by genes downregulated with quercetin, most of which were within the human diseases category. Generally, the DE genes in both treatments affected key pathways such as longevity regulation ( Figure S3), xenobiotic metabolism, cell death and growth, senescence, and human diseases (Table S5).
To validate the results obtained with the RNA-Seq, the expression levels of several genes that showed significant differential expression in both p-coumaric acid and quercetin treatments were quantified with qRT-PCR. The qRT-PCR results showed consistency in differential gene expression with RNA-Seq data, and the correlation between the two methods was highly significant, with R 2 = 0.84 and R 2 = 0.93 for p-coumaric acid and quercetin treatments, respectively (Figures 7 and 8).

| D ISCUSS I ON
In this study, we conducted laboratory assays and RNA sequencing to evaluate the ecological and physiological impacts of the nectar phytochemicals caffeine, p-coumaric acid, and quercetin on adult female Ae. albopictus. Overall, our results revealed that the consumption of sucrose supplemented with certain phytochemicals enhanced longevity and fecundity, deterred sugar-feeding, and changed the expression of genes involved in longevity regulation and xenobiotic metabolism, among others in female adult Ae. albopictus.
Dietary p-coumaric acid and quercetin enhanced the longevity of the female mosquitoes, consistent with previous findings involving Ae. aegypti (Nunes et al., 2016) (for quercetin), honey bees (Bernklau et al., 2019;Liao et al., 2017a), and multiple model organisms, includ- ing Drosophila melanogaster and Caenorhabditis elegans (Kampkötter et al., 2008;Sunthonkun et al., 2019). Dietary p-coumaric acid extended the life span of adult worker honey bees by 14.1% (Liao et al., 2017a) whereas quercetin enhanced the longevity of adult female Ae. aegypti by 30% (Nunes et al., 2016) and that of C. elegans by 15% (Kampkötter et al., 2008). The presence of these two phytochemicals in the diet did not affect the amount of sugar intake, suggesting that the extension of life span was not related to differential intake of sugars. These phytochemicals slightly affected fecundity, suggesting that the positive effect on life span was not due to a trade-off between these two life-history traits.
With the pronounced effects of enhanced mosquito longevity by    (Lithgow & Walker, 2002;Morrow et al., 2004;Tower, 2011), and we found here that the heat shock factor-binding protein 1-like gene (XM-029853002.1) was upregulated in female mosquitoes consuming dietary p-coumaric acid and quercetin.
In both p-coumaric acid and quercetin treatments, histone proteins were downregulated in the female mosquitoes. Changes in histones at the gene and protein level have been linked to cell death and senescence pathways in aging organisms. Histones are part of chromatin-based processes in the nucleus, and they are major regulators of cellular and organismal aging. For instance, both loss of histones and change in expression levels are linked to aging in a mouse model and human fetal brain (Song & Johnson, 2018). The H2A subunit was overexpressed in senescent human fibroblasts, as well as in aging mice (Contrepois et al., 2017), in contrast with our finding, where this histone protein was downregulated with both p-coumaric acid and quercetin, indicating its potential contribution in extending female Ae. albopictus life span.
The contribution of cytochrome P450s to the metabolism of natural products and synthetic xenobiotics in insects has been extensively evaluated (Feyereisen, 2006(Feyereisen, , 2011. In our study, three   (Højland et al., 2014;Poupardin et al., 2010). Another including CYP12F7, are associated with pyrethroid resistance (Bariami et al., 2012;Faucon et al., 2015). The role of quercetin and p-coumaric acid present in honey and pollen of angiosperms in enhancing pesticide tolerance and upregulation of P450 genes that metabolize pyrethroids and natural toxins has been documented previously in honey bees (Johnson et al., 2012;Liao et al., 2017a;Mao et al., 2009Mao et al., , 2015. The XM_029852718.1 transcript encoding dihydrodiol dehydrogenase, an oxidoreductase enzyme involved in metabolism of xenobiotics by P450s (Schomburg et al., 1993), and the gene coding for cytochrome c oxidase subunit 6B1-like (XM_029878939.1), an enzyme involved in cellular respiration and also contributes to antiaging effects in in vitro studies (Kim et al., 2015), were upregulated in both treatments. The acetylcholinesterase-like gene (XM_019709882.2) was also upregulated in mosquitoes consuming dietary quercetin. This specialized carboxylic hydrolase is found at neuromuscular junctions where it terminates synaptic transmission, preventing continuous nerve firings at nerve endings (Lionetto et al., 2013) and is inhibited by organophosphates and carbamate pesticides (Hobbiger, 1961) Mehboob et al., 2017;Naydenova et al., 1999) and resistance to multiple insecticides in Anopheles gambiae (Vontas et al., 2005), D. melanogaster (Pedra et al., 2004), diamondback moth (Plutella xylostella) (Li et al., 2018) and African cotton leafworm (Spodoptera littoralis) Bozzolan et al., 2014). Additionally, the gene coding for a glutathione-S-transferase (GST1) (XM_019694071.2), belonging to a family of phase II detoxification enzymes, was upregulated in both treatments. GSTs are generally strongly implicated in resistance to multiple insecticides in mosquitoes (Hemingway et al., 2004). The activity of GST1 in the breakdown of xenobiotics and metabolites produced during cellular division and morphogenesis has been documented previously in D. melanogaster (Tu & Akgül, 2005). Apart from breakdown of toxic products, glutathione-dependent enzymes are also involved in regulation of oxidative stress through ROS reduction, which is linked to enhanced longevity (Hayes & McLellan, 1999), indicating the potential contribution of GST1 in life span extension in female mosquitoes.
We also found that the homologues of glutathione-S-transferases-1, superoxide dismutase [Mn] mitochondria-like, and the heat shock factor-binding protein 1-like genes (upregulated in the female mosquitoes consuming p-coumaric acid and quercetin) are involved in the  (Boz, 2015). D. melanogaster (Itoyama et al., 1998;Suh et al., 2017). By contrast, caffeine increased the life span of C. elegans at 15°C and 20°C in a temperature-dependent lifespan extension study (Sutphin et al., 2012).
Aedes albopictus consumed less sucrose from solutions that contained caffeine, suggesting that it can act as a feeding deterrent. As a neuromodulator, caffeine exhibits both attractant and deterrent properties to foraging pollinators in a concentrationdependent manner. High concentrations of caffeine (150 ppm and 200 ppm) comparable to levels used in our study repelled adult worker honey bees (Mustard et al., 2012;Singaravelan et al., 2005;Wright et al., 2013), possibly due to its bitter taste. The feeding deterrent property of caffeine in our study could explain the reduced life span in mosquitoes consuming a sucrose diet containing caffeine. A similar concentration of caffeine (200 ppm) supplied in sucrose diet, however, improved the memory of honey bee foragers, thereby enhancing flower visitation (Si et al., 2005;Wright et al., 2013). In two other separate studies, caffeinated nectar enhanced its quality, attracting more honey bees and leading to more efficient pollination (Couvillon et al., 2015;Thomson et al., 2015).
Apart from concentration, the availability of alternative nectar sources also alters the deterrence of these compounds (Gegear et al., 2007;Stevenson et al., 2017).
In summary, this is the first study to evaluate the physiological impact of nectar phytochemicals on female adult mosquitoes at the molecular level. We demonstrated that p-coumaric acid, and quercetin, present in nectars of many plant species, enhanced longevity, and altered the expression of genes involved in longevity regulation and xenobiotic metabolism in Ae. albopictus. We suggest that the lifespan extension capacity of p-coumaric acid and quercetin is likely linked to the regulation of gene expression of life span-related genes and xenobiotic metabolism. The phenolic acid p-coumaric acid exerted a stronger effect and affected a wider range of genes than quercetin, possibly due to its ability to conjugate with small molecules that include sugars, which likely enhances its biological effects (Pei et al., 2016).
Our findings provide insights into the direct implications of nectar phytochemicals in the diet of adult female mosquitoes both at the organismal and molecular level by altering their sugar-feeding behavior, fecundity, longevity, and gene expression. Our findings highlight that the role of phytochemicals needs to be considered when assessing how different nectar sources in the environment influence mosquito fitness, vectorial capacity, and, potentially, insecticide resistance. Future studies are needed to examine the effects of a wider range of nectar phytochemicals on the ecology of mosquito vectors, expanding the focus to encompass a greater diversity of vector species and the effect on vectorial capacity to establish whether enhancement of longevity by nectar feeding is a widespread feature of mosquito biology. Gene knockout studies are also needed to ascertain the phenotypic effects of differentially expressed genes affecting the longevity of mosquitoes.

ACK N OWLED G M ENTS
We thank Chang-Hyun Kim

CO N FLI C T O F I NTE R E S T
The authors have no conflict of interest.

DATA AVA I L A B I L I T Y S TAT E M E N T
All data from the study are either within the manuscript and in a supplemental file or a public repository at the NCBI Sequence Read Archive under the accession number PRJNA680162 (sequence reads).