Characterization of the mechanisms of action and mosquito larva midgut response to a yeast- encapsulated orange oil larvicide

Patrick H. Kelly Uniformed Services University: Uniformed Services University of the Health Sciences Alexandra V. Yingling University of New Mexico Health Sciences Center Anwar Ahmed Uniformed Services University: Uniformed Services University of the Health Sciences Ivy Hurwitz University of New Mexico Health Sciences Center Marcelo Ramalho-Ortigao (  mrortigao@gmail.com ) Uniformed Services University of the Health Sciences https://orcid.org/0000-0002-1490-4946


Introduction
Mosquito-borne diseases continue to represent a major threat to global health and to the economic development of many countries worldwide. Methods to combat these diseases require strategic planning as effective control of mosquito populations are labor intensive, di cult to sustain, and fraught with issues of vector resistance. Primary approaches to mosquito abatement includes adulticide spraying with synthetic pyrethroids and organophosphates, larval source management with bacterial larvicides such as Bacillus thuringiensis (Bti), or a combination of both. Control of Aedes aegypti, a major vector of dengue, Chikungunya, and Zika has posed a tremendous challenge for vector-control programs. The endophilic behavior of this species in urban areas has allowed for low mosquito populations to sustain disease outbreaks. Initial successes achieved with insecticide control Ae. aegypti have been severely hampered by the emergence of resistance. Recently, methods that apply either transgenic technology via release of insect carrying a dominant lethal (RIDL)1,2 or mosquitoes carrying the endosymbiont Wolbachia to promote cytoplasmic incompatibility of the offspring3 have been used to reduce mosquito populations.
However, these methods of control are still in their infancy, and are still hampered by regulatory barriers, cost, negative public perception, and may require several years of investigation to fully assess their perceived potential.
Larvicides have remained an effective means of larval source management. Bti, for instance, is highly effective in reducing the risk of mosquito-borne diseases as they can be easily administered in urban and rural standing water. Bacterial larvicides are also considered to be more environmentally-friendly than insecticide spraying, with fewer or no toxic effects on bene cial insects (e.g. bees, butter ies) as well as other non-target species. As such, this class of larvicides are widely accepted by the general public.
Multiple novel approaches for mosquito control are being evaluated, including the use of natural essential oils (EOs)4-8. EOs are volatile, aromatic oils that have long been used for many purposes, including medicinal, bactericidal, insecticidal, agricultural, and pharmaceutical applications. Although EOs possess robust mosquito larvicidal traits, they are hydrophobic in the aquatic environment and susceptible to breakdown not only from ultraviolet light but also to temperature and oxidation, highlighting a need to develop an e cient yet effective delivery system to utilize their potency against mosquito larvae.
Saccharomyces cerevisiae (Baker's yeast) can serve as a suitable and effective biodegradable microencapsulation container of several compounds and particles including fungicides, drug treatments, and EOs9, 10. The encapsulation process renders the yeast cells nonviable, but increases water solubility, bioavailability, long-term viability and stability as a lyophilized material, and delivery e ciency as a mosquito larvicide. Further, yeast is readily consumed and digested in the gut by mosquito larvae which possess specialized intestinal enzymes that rapidly break down and digests the yeast β-1,3-glucan cell membranes11, which releases the essential oil within the gut, subsequently killing the larvae.
We previously demonstrated that yeast-encapsulated orange oil (YEOO) is an effective larvicide against multiple larval life stages of Ae. aegypti and Culex quinquefasciatus12. Moreover, we also showed that the YEOO LC50 and LC90 after 24 h exposure in third larvae instar (L3) Ae. aegypti larvae are 12 and 28 ppm, respectively, which is nearly half the required dose necessary when using non-encapsulated orange oil13. In this study, we describe the changes in gene expression pro les and midgut epithelial morphology induced by the ingestion of YEOO in Ae. aegypti larvae. Innate response gene expression pro les have been extensively investigated in adult mosquitoes14-17, however little is known regarding innate responses of mosquito larvae following exposure to larvicides. The killing or toxic effects of larvicides on mosquito larvae presumably lead to changes in gene expression pro les, activation of apoptotic pathways, epithelial regeneration, and the activation of detoxi cation mechanisms linked with P450 gene expression. Our results point to a mosquito larval killing mechanism that is characterized by a nonspeci c and unregulated midgut epithelial damage and apoptosis. The combination of the stability of the orange oil provide by the yeast-encapsulated formulation with the irreversible effects on the mosquito larvae without affecting non-target organisms indicate the potential commercial usage of YEOO or other yeast-encapsulated oils as effective mosquito larvae control approaches.
Orange oil, water, and yeast were combined in a ba ed ask and agitated for 24 h at 40˚C. The resulting mixture was then centrifuged, the supernatant discarded, and the remaining precipitate was subsequently washed three times to remove excess oil and lyophilized prior to storage. The dried larvicide was reconstituted in water between 4.4 -5% oil in solution prior to bioassay application.

Larval strains and bioassays
Larvae used in our studies were obtained from mosquito colonies maintained at Uniformed Services University of the Health Sciences (USUHS). Ae. aegypti (Liverpool strain -AAE-L and Washington, D.C. strain -AAE-DC strains) and Anopheles gambiae (G3 strain -AGA) were tested in our bioassays and the AAE-L strain was used in all other experiments. Larvae were kept in DI water at 28°C and were fed with sh food ad libitum. Larvae bioassays were performed according to the standard WHOPES protocol18. Brie y, 25 third instar (L3) larvae were placed into cups with 100 ml of DI water. After a 30 min acclimation, different quantities of YEOO larvicide were added to each cup to reach concentrations ranging from 2.5 mg/L to 60 mg/L, and lethality was assessed after 24 hours. In the experiments described here, concentrations of 5, 10, and 30 mg/L of YEOO correspond to LC20, LC50, and LC90, respectively, and were based on our previous results 12 using L3 stage AAE-L. The effects of YEOO on AAE-L larval swimming behavior and lethality can be seen in Supplemental Videos 1-3.
Gene Expression Analyses. Individual live whole larvae or guts and carcasses dissected from AAE-L L3 were collected following 1, 4, 8, and 24 h exposure to 10 or 30 mg/L (equivalent to LC50 or LC90, respectively) of YEOO. Samples were placed into 30 µl of RNAlater (Thermo Scienti c, Waltham, MA), and frozen at -80°C until RNA extraction. Genes associated with apoptosis (IAP1 and IAP2), autophagy (ATG1, ATG6, and ATG8) and innate immune response genes (Vein, Pirk, Serpin-1, Serpin-2, IMP2, CASPS7 and CASPS8) were assessed in two independent experiments using RNA samples (in triplicate) obtained from pools of three L3 dissected guts or carcasses for each of the four time points. Here, gut samples were composed of foregut and midgut whereas carcasses included the hindgut, the Malpighian tubules and all remaining larvae-matched body parts such as head, thorax, and abdomen. In contrast, for cytochrome P450 gene expression analysis, one whole L3 individual was used per sample (up to eight samples) for each of the three independent experiments (i.e., three replicates) for both 5 mg/L (LC20) and 10 mg/L (LC50) YEOO concentrations tested after 4 h exposure. Regarding P450, we assessed genes representative of families CYP6 (CYP6M11, CYP6N12, and CYP6Z8) and CYP9 (CYP9J10 and CYP9M9). Individual whole larvae were used to determine the systemic insecticide resistance response to YEOO to determine individual gene response variation. In addition, for the P450 analysis speci cally, we were not interested in ascertaining spatial expression pro les.
RNA extraction and cDNA Synthesis. AAE-L total RNA was isolated using the RNeasy tissue kit (Qiagen, Hilden, Germany) followed by DNase treatment using TURBO DNA-free kit (Invitrogen, Carlsbad, CA). Each RNA sample was quanti ed on NanoDrop (Thermo Scienti c, Waltham, MA) and assessed on 1% agarose-5% formaldehyde in 1x MOPS. RNA samples were stored at -80°C until use. First-strand cDNAs were obtained using Superscript III reverse transcription kit (Invitrogen) following manufacture protocols with oligo dT12-20 primers and 200 ng of each RNA.
Quantitative Real-Time PCR. mRNA levels were quanti ed with PowerUp SYBR Green Supermix (Applied Biosystems, Foster City, CA) using 10 ng of cDNA with a 50°C for 2 min and 95°C for 2 min hot start initialization, followed by 40 cycles of 60°C for 5 sec and 95°C for 15 sec using a ABI7500 FAST (Applied Biosystems, Foster City, CA). Relative fold changes were assessed using the ∆∆Ct method19 and calibrated against the expression of the housekeeping gene Actin6 (Supplemental Table 1) in L3 larvae control samples fed on blank yeast. Primer sequences for each gene measured are listed in Supplemental Table 1.
Confocal Microscopy. Confocal microscopy was used to assess the distribution of YEOO throughout the larva alimentary canal. Whole guts from AAE-L L3 larvae were dissected after exposure to either 30 mg/L YEOO or blank yeast control (30 mg/L) in PBS and xed for 15-30 min at room temperature with Zamboni's Fixative20 (4% paraformaldehyde). For experiments focused on the permeability of the type 2 peritrophic matrix (PM2), 20 L3 larvae were exposed to blank yeast control (30 mg/L) or 30 mg/L YEOO in 10 ml DI water at 28°C supplemented with FITC-dextran (Cell Signaling Technology, Danvers, MA) (Molecular weights of 150, 500, or 2,000 kDa) at 0.5 mg/ml for 4 h. Fixed guts were washed 3x in PBS, stained for 5 min with 10 µg/ml of DAPI (Invitrogen, Carlsbad, CA) and washed 3x in PBS before imaging on a ZEISS 710 Two Photon (Carl Zeiss AG, Hoberkochen, Germany) confocal microscope. The permeability of larval PM2 was qualitatively assessed by presence/absence of uorescein signal in the caeca of dissected guts/larvae21. For microscopy, entire alimentary canals were used for viewing clarity. TUNEL Analysis. TUNEL was used to assess apoptosis via DNA fragmentation. AAE-L L3 larvae were exposed to either YEOO (10 mg/L) or yeast blank for 24 h. Guts dissected from insects were xed for 15-30 min at room temperature with Zamboni's Fixative, washed three times (5 min each) in PBS, twice (2 min each) in PBS containing 0.3% Triton X-100 (PBST), and stained for TUNEL analysis (Roche, Basel, Switzerland) following manufacturer's protocol. Samples were counterstained with 10 µg/mL of DAPI (Invitrogen, Carlsbad, CA), mounted onto slides in Vectashield→ (Vector Laboratories, Burlingame, CA), and subsequently imaged on a ZEISS 710 Two Photon (Carl Zeiss AG, Hoberkochen, Germany) confocal microscope.
Transmission Electron Microscopy. TEM was used to assess cellular damage caused by YEOO ingestion and to verify the structural integrity of YEOO. AAE-L larvae were exposed to either blank yeast control (30 mg/L) or 30 mg/L YEOO. Following 4 h exposure, guts were dissected and xed overnight at room temperature in freshly prepared 2% formaldehyde and 2% EM grade glutaraldehyde (Electron Microscopy Sciences, Hat eld, PA) in 0.1M cacodylate buffer, pH 7.2. Fixed guts were washed three times in cacodylate buffer (without aldehydes) and incubated in 2% OsO4 in 0.1M cacodylate buffer for 1 h, After another round of three washes 10 min each in cacodylate buffer, guts were dehydrated in a graduated series of ethanol (10 min each in 30%, 50%, 70%, and 95% ethanol and 2 x 10 min in 100% ethanol), in ltrated in a graduated series of Spurr's epoxy resin (Electron Microscopy Sciences), and then polymerized at 70°C for 11 h. Polymerized blocks were sectioned in a Leica UC6 ultramicrotome. Thin sections of guts cut transversely and longitudinally were collected on 3 mm copper grids. Grids were poststained in a Leica EM AC20 grid stainer and then examined on a JEOL JEM-1011 transmission electron microscope (JEOL USA, Peabody, MA). Images were collected on an Advanced Microscopy Techniques (AMT Corp., Woburn, MA) digital camera.
Statistical Analysis. Larvae bioassay data to determine LC50 and LC90 YEOO effective concentrations were determine with Nonlinear or Probit regression analyses22. Gene expression data of dissected guts and carcasses were subjected to two-way analysis of variance (ANOVA) with repeated measures, with Dunnett's multiple comparisons tests and p-values evaluated at α = 0.05, 0.01, and 0.001. In whole larvae P450 gene expression analyses, data were subjected to one-way ANOVA, and Tukey's Post-Hoc tests with p-values evaluated at α = 0.05, 0.01, and 0.001. Statistics were carried out in SAS 9.4 (Cary, NC) or GraphPad Prism8 (San Diego, CA).

Bioassays
In accordance with our prior results12, YEOO was effective against larvae of both Ae. aegypti strains (AAE-L and AAE-DC), as well as against larvae of An. gambiae G3 (AGA). For AAE-DC and AGA, the LC50 was calculated at 11.1 mg/L and 10.3 mg/L, and the LC90 at 17.9 mg/L and 28.1 mg/L, respectively ( Figure 1 and Table 1). For AAE-L larvae, the LC50 and LC90 were recalibrated from our original results12 to 9.4 mg/L and 27.8 mg/L, respectively. These concentrations are consistent with our previous results against AAE-L L3 larvae12, and also when assessed using Probit analyses22 (shown in the Supplemental Figure 1 and the Supplemental Table 1). In addition, the sublethal dose of LC20 was calculated at 5 mg/L. Gene expression pro les We investigated the expression of genes associated with cellular processes such as apoptosis and autophagy as well as innate immune response genes in AAE-L L3 larvae after exposure to YEOO in live, motile larvae (no moribund or dead larvae were included for gene expression analyses). Both LC50 and LC90 concentrations led to signi cant differences in gene expression pro les in apoptosis-related genes, including inhibitors of apoptosis and effector caspases.
In larvae exposed to LC50 concentrations, the inhibitor of apoptosis gene IAP1 was upregulated more than 7-fold in the carcass 24 h post exposure compared to yeast controls (Figure 2A, p-value = 0.0489). Additionally, at 24 h, the effector caspase CASPS8 was signi cantly upregulated 12-fold in the gut ( Figure  2B, p-value = 0.0301). The autophagy gene ATG1 was downregulated 2.5-fold ( Figure 2C, p-value = 0.0403) following a 1 h exposure to YEOO at LC50 concentration with expression levels returning to those comparable to the yeast control at 4h and 24 h. Lastly, the midgut epithelial cell regenerator gene Vein was upregulated more than17-fold in the carcass at 24 h post exposure, but did not reach signi cance ( Figure 2D, p-value = 0.059).
Larvae exposed to LC90 concentrations displayed opposing effects for IAP1 and IAP2. For IAP1, a 26-fold induction in the carcass at 8 h ( Figure 3A, p-value = 0.0001) was observed, whereas for IAP2 we detected a signi cant reduction in the carcass at 1 h ( Figure 3B, p-value = 0.0389) and gut at 8 h for ( Figure 3B For other transcripts measured after LC50 or LC90 YEOO exposure, including the IMD negative regulator Pirk, Serpin-1, Serpin-2, and IMP-2 (genes associated with melanization cascade), the autophagy gene ATG6, and the effector caspase CASPS7, though a trend suggestive of upregulation was noted, such increases were not statistically signi cant (LC50 Supplemental Figure 2, and LC90 Supplemental Figure  3). Levels of induction or repression, standard error, and p-values measured for all genes assessed after exposure to LC50 and LC90 concentrations are shown in Supplemental Table 2 and Supplemental Table  3, respectively.
Cellular and physiological damage in midgut after YEOO exposure.
Transmission electron microscopy (TEM) was utilized to verify the structural integrity of YEOO. TEM revealed nearly all S. cerevisiae cells were loaded with EOs ( Figure 5D-F) compared to yeast blank controls ( Figure 5A-C). As expected, no cell division was observed in YEOO as the encapsulation process effectively kills the yeast cells. Bud scars which are visible in this population likely occurred before EO loading.
Ingestion of YEOO by AAE-L larvae led to signi cant morphological and physiological cytotoxic effects, as observed via TEM. Compared to control larvae (Figure 6), there was substantial damage to midgut epithelial cells within 30 mins of exposure (Figure 7), followed by signi cant damage systemically throughout the midgut epithelia after 4 hours (Figure 8). Microvilli were severely damaged/shortened by 30 min exposure ( Figure 7A-C) or destroyed in larvae exposed for 4 h ( Figure 8A-B). Further, we noted an increase in cellular vacuolization particularly along the basal lamina ( Figure 7D, Figure 8C and 8D), and alterations in the mitochondrial shape and cristae ( Figure 7D, Figure 8C and 8D). Interestingly, unlike unexposed larvae which displayed visibly intact yeast control cells within the peritrophic space, no intact YEOO cells were visible in YEOO exposed larvae regardless of the exposure duration. We tested the hypothesis that the breakdown of YEOO in larval guts might be due to pH as no difference in the levels of digestive proteases has been observed (Dr. Fernando Genta, personal communication). As the mosquito larva gut pH ranges from 10 to 7.5 from the anterior to posterior regions23, YEOO larvicide was incubated with in HEPES with incremental increases of pH from pH 7 to pH 10 for 30 min, and no changes in the YEOO stability were observed. Lastly, both living and moribund AAE-L larvae were subjected to TUNEL analysis after 24 h exposure of LC50 concentrations. Guts of moribund larvae exhibited increased DNA fragmentation when compared to live larvae. (Supplemental Figure 4).Although the structural integrity of the larval peritrophic matrix 2 (PM2) in AAE-L larvae did not appear to be affected by YEOO exposure up to 4 h, we assessed if its permeability or porosity were altered. Under normal physiological conditions Ae. aegypti larva PM2 is not permeable to 2000kDa FITC-labeled dextran particles21. Following simultaneous feeding of FITC-labeled dextran at 2000 kDa with YEOO (at LC90 concentration) in 20 AAE-L L3 larvae for 4 h, FITC-related uorescence was neither detected in the caeca of control larvae as expected ( Figure 9A) nor in the larvae fed with YEOO ( Figure 9B), suggestive of a lack effect of the YEOO on AAE-L larva PM2 permeability. Simultaneous feeding of FITC-Dextran particles with YEOO was our only option to test permeability as the larvae refused to feed on the particles following YEOO exposure alone.

Discussion
Our work demonstrates the potent effects of a novel, innovative mosquito larvicide on the physiology and innate immune response gene expression pro le in L3 Ae. aegypti larvae.
Ingestion of the YEOO larvicide, resulted in unregulated cytotoxic events to the larva midgut epithelial cells as well as an upregulation of apoptosis-related genes. Unlike traditional single target insecticides, YEOO displays a multi-modal action affecting multiple cellular pathways. Thus, we predict YEOO activity is independent of speci c midgut receptors and killing mechanisms are based upon toxicity to midgut cells. Multi-target larvicides, such as Bti, or multi-modal larvicides like YEOO are thought to be less likely to incur in resistance8-10. To date, resistance to EOs has not been reported.
EOs and their derivatives have been investigated in vitro against neglected tropical parasites and arboviruses for decades (Reviewed in Luna et al.) 24. EOs also have been utilized as contact irritants against adult and larval stages of various insect pests, and have been shown to cause testicular apoptosis and morphological damage to the fat body and midgut epithelium of Spodoptera frugiperda (lepidopteran) larvae25,26. Other natural compounds, such as squamocin and extract from the Amazonian plant Derris urucu, have been tested against Ae. aegypti larvae on which both cytotoxic and gene expression pro les effects were assessed27,28. Encapsulated essential oils offers signi cant advantage over free EOs. For instance, when sequestered into yeast cells, EOs are effectively protected against photolytic degradation, allowing for long term stability. The lyophilized YEOO can be stored for over a year without any discernible differences between stored vs. freshly generated YEOO, as determined by TEM analysis. Moreover, whereas the hydrophobicity of free EOs makes them unlikely candidates for commercial use, the encapsulation approach using S. cerevisiae provides an optimal delivery mechanism that facilitates dissemination throughout aquatic environments, and is readily ingested by mosquito larvae. The degradation of the yeast cell wall results in the release of the sequestered EO within the gut of the insect and allows for targeted delivery of the larvicide.
Though the gene expression pro les only provide indirect evidence, the increase expression of the midgut epithelial cell regenerator Vein at 4 h with concomitant upregulation of CASPS8 at 8 h post YEOO exposure in L3 AAE-L larvae is suggestive of an attempt by the larvae to reconstitute the midgut epithelia damaged by the larvicide, combined with a last-ditch effort to clear dead and dying cells unable to overcome the YEOO non-speci c, broad-acting mode of action (MoA). Moreover, the modulation observed in the gene expression levels regarding the inhibitor of apoptosis IAP1 and IAP2 may be associated with their regulatory abilities on initiator or effector caspases29. For instance, IAP1 interacts and regulates the expression of both initiator caspases (i.e. Dredd) and effector caspases (CASPS7 and CASPS8)30,31. Both CASPS7 and CASPS8 carry out proteolysis and disintegration of proteins during cell death29-31. Although the Ae. aegypti apoptosis pathway is not entirely understood, studies in Drosophila melanogaster have demonstrated that IAP2 not only regulates apoptotic-associated genes, but also negatively regulates the innate immunity (IMD) pathway32 in a fashion similar to the negative regulator Pirk33.
The cytotoxic effects of YEOO are broad, with drastic alteration in cell morphology at both the cellular and subcellular levels in the insect midgut, as demonstrated by the TEM analyses. While YEOO appeared to have produced no speci c alterations of the larva PM2, there was substantial damage to midgut epithelia, including loss of microvilli, disordered cellular junctions, and increased cytoplasmic vacuolization in laminar bodies. In the cytosol, increased vacuolization and abnormal morphological alterations of the mitochondria and cristae also were observed.
Apoptotic signaling is linked with mitochondrial release of cytochrome c and other cofactors to activate effector caspases34. The presence of circular-shaped mitochondria observed in our TEM images of midgut cells following YEOO exposure is likely due to the cellular cytotoxicity affecting mitochondrial proteins that coordinate cytochrome c release and promote the apoptosis expression cascade suggested by the differential expression of IAP and CASP, as seen in AAE-L larvae.
With regards to autophagy being mediated by YEOO, no differences were detected for either ATG1 or ATG6 when we compared YEOO vs. control yeast ingestion by the mosquito larvae. In addition, no differential expression was detected in V-ATPase between YEOO-exposed (LC50) and unexposed L3 larvae. It has been previously reported that the expression of ATG1, ATG6, and ATG8 is correlated with the expression of V-ATPase following exposure of Ae. aegypti larva to the plant-derived fatty-acid compound squamocin28. Further, in Drosophila melanogaster, autophagic cell death in the midgut is accompanied by markers of apoptosis such as DNA fragmentation35. Similarly, in moribund AAE-L larvae exposed to YEOO for 24 h, the typical pattern of DNA degradation present during apoptosis was observed in our TUNEL analysis. Although the relationship between the autophagic and apoptotic pathways is not yet clear, following the ingestion of YEOO by AAE-L larvae, increased levels of cytoplasmic vacuolization and cell death are observed within 30 min (as depicted in our TEM images). However, despite our current lack of evidence, autophagy as an inducible response to the ability of the mosquito larvae to tolerate the nonspeci c toxic and apoptotic effects associated with YEOO has not yet been discarded.
In our investigations we also noted the disappearance of the YEOO within the gut of the larvae. To exclude the possibility that the breakdown of the yeast cell wall within the larva gut is simply due to pH variations within the mosquito larva gut as previously shown23,36, we tested YEOO in increasing concentrations of pH solutions. As no changes in the YEOO stability were detected in a range of pHs from pH7 to pH10, it suggests that other mechanism(s) in the gut are likely responsible for the YEOO breakdown, presumably digestive proteases11. It also remains to be determined if such YEOO breakdown is due to the continued action of digestive proteases, if indeed these are involved, or whether it involves a combination of factors that create a domino effect regarding the breakdown of the YEOO cell walls and the release of the orange oil.
Insecticide resistance poses a serious threat to the control of mosquito-borne diseases. Traditional vector management programs either increase insecticide applications to kill resistant populations or must switch to another pesticide to achieve control. These approaches have not generated any signi cant advantages for mosquito control but have instead contributed to the increase of resistance in mosquito vectors. Continuous exposure to organophosphates and pyrethroids37-39 have resulted in the overproduction of cytochrome P450 enzymes37,40,41 that are necessary for detoxi cation of the various pesticides in Ae. aegypti. As is found in nature, it is conceivable that distinct mosquito strains or species possess unique detoxi cation or innate gene regulatory mechanisms to cope to various pesticide classes, including organophosphates, pyrethroids, and natural products like YEOO. In response to YEOO, the differential gene expression between CYP6 family and CYP9 cytochrome P450 family genes suggest that different families of genes may be regulated according speci cally to each respective chemical exposure challenge (for example receptor-based pesticides like pyrethroids versus non-receptor based pesticides such as the YEOO larvicide). Additionally, cytochrome P450s are consistent markers of resistance but do not encompass the entire spectrum of underlying resistance mechanisms. Therein lies the possibility that other mechanisms of resistance exist that have yet to be identi ed against YEOOs.
Larviciding approaches such as Bti, which is effective against Ae. aegypti larvae, are often still out of reach for many affected communities due to frequent need for reapplications and elevated cost. Thus, gaps exist both in the availability of safe, stable, cost-effective, and e cacious alternatives in mosquito control approaches. YEOO mosquito larvicide is easy to produce and is stable after long-term storage 12 .
Moreover, YEOO is affordable as it primarily relies on local resources (essential oils). It is not yet known what effects if any YEOO has on Ae. aegypti larval development into adults, including effects on the physiology, fecundity, and fertility of the adult mosquito, especially after sublethal YEOO exposure (i.e. LC20 concentrations). Experiments focused on understanding these effects in larvae in relation to hormesis are currently underway42.
In conclusion, YEOO is highly effective against Ae. aegypti larvae through mechanisms involving acute midgut cell damage and apoptotic pathways leading to larval death. Because of its multi-target action, resistance and tolerance are unlikely to develop against YEOO. Nevertheless, many aspects of how YEOO act remain uncharacterized. Primary components of EOs have been demonstrated to exert their larvicidal effects through at least three different mechanisms: neurotoxicity43,44, growth inhibition, and interruption of metabolic pathways45-47. Additional work will uncover potential tolerance or resistance mechanisms associated with the non-receptor based larvicides. Identifying its effectiveness against other mosquito species as well as their speci c killing mechanisms will provide critical information towards the development of YEOO as a safe, biologically-friendly larvicide for wide spread applications to control mosquito populations.

Declarations
Ethics and Approval to Participate -All persons include in the study were cleared to participate by their respective institutions.
Consent -All authors have given consent for publication.
Availability of data and material -Data and material used in the studies are available Competing interests -Authors declare no competing interests Funding -The studies described were funded through contract #200-2017-93140 with the Center for Disease Control and Prevention. The funder had no role in study design, data collection and analysis, decision to publish or preparation of the manuscript. PHK is a Henry Jackson Foundation for the Advancement of Military Medicine contractor.
Authors' contributions -PHK was responsible for mosquito rearing, performed the assays, data analyses, and help craft the manuscript; AVY produced and puri ed the YEOO; AA assisted with the statistical analyzes; IH helped with the study design and helped with crafting the manuscript; MRO coordinated all aspects of the study including design and data analyses and the crafting of the manuscript.    Cytochrome P450 mRNA expression in AAE-L after YEOO LC20 or LC 50 doses. Expression of cytochrome P450 gene families CYP6 (CYP6M11 and CYP6N12) and CYP9 (CYP9J10) were assessed in three replicates containing 3-8 individual whole larvae exposed to YEOO LC 20 or LC 50 concentrations. 1-Way ANOVA, Tukey's Post-Hoc test. *p<0.05, **p<0.01, ***p<0.001.