Chitobiase plays a role in digestion of the cuticle during a molt. Ecdysone is an insect hormone important for initiating the molting process when larvae are transitioning between instars, and is structurally similar to estrone. Bisphenol A , a xenoestrogen, has the ability to bind and express not only the estrogenbinding proteins in mammals, but also to the ecdysone-binding protein in Chironomus riparius . Exposure of the non-biting midge, C. riparius, to estrogenic compounds has been shown to cause mouth deformities, decreased fecundity, and increased developmental time if administered over multiple generations. Culex quinquefasciatusare aquatic arthropods in their larval in stars, and terrestrial as adults. From their first to fourth instar, most mosquito larvae feed on detritus and thereby recycle nutrients back into the environment . Once they reach the fourth instar, larvae cease feeding and prepare to molt into a non-feeding pupal stage. Mosquitoes, like many insects, rely on endosymbionts to grow and develop. For example, bacteria in the genus Buchnera are commonly endosymbionts of aphids and provide the aphid with essential amino acids . Similarly, bacterial symbionts in the genus Asaia have been shown to be crucial in the development of the mosquito, Anopheles stephensi . Not surprisingly, mosquitoes treated with antibiotics to eliminate bacteria took significantly longer to develop than untreated control larvae. However mosquitoes “rescued” by a subsequent introduction of the bacteria following antibiotic exposure showed no difference in development . Currently, there is little information available on the effects of PPCPs at environmentally relevant concentrations on aquatic invertebrates, on the effects of PPCPs on the holobiome of insects, or on the effects of PPCPs on efficacy of Bacillus thuringiensis subsp. israelensis ,plastic plant pot a bacterial insecticide commonly used to control larvae of mosquitoes. Joint exposure to a pollutant and a microbial larvicide can be used to detect sublethal physiological stress .

Further, since 1997, 1000s of hectares of floodwater mosquito breeding sites have been treated with Bti . Thus Bti is likely to coexist with PPCPs in aqueous environments. Therefore, we used a series of bio-assays to evaluate the effects of PPCPs on development and mortality of the mosquito, Culex quinquefasciatus, which is a lower trophic level arthropod found worldwide and native to the Southern U.S.. This species is a vector of encephalitides including West Nile Virus and the nematode, Wuchereria bancrofti, which causes filariasis in the tropics and subtropics. Thus, any potential effects would have interesting implications from both ecological and medical perspectives.Culex quinquefasciatus mosquito egg rafts were obtained from a parental colony maintained at the University of California, Riverside using the procedures described by Wirth et al.. Rafts were maintained in shallow porcelain pans containing 3 L water or water and one of the PPCP treatments. Following emergence, larvae were kept in an incubator at 28°C, approximately 70% RH, and a light: dark cycle of 16:8. Second in stars were transferred individually by disposable pipette to a bio-assay container consisting of a 29.mL plastic cup with clear plastic lid containing 15 mL of CGSW. Each mosquito larva was given 67.0 µL of diet on day 1 and 33.5 µL of diet every other day thereafter. Diet was prepared as in Sorensen et al.; briefly a 3:1 mixture of ground mouse chow: brewers yeast was rehydrated by 50 mL of CGSW for 4 g of dry mixture. Bio-assay containers were treated with stock solutions to ensure environmentally relevant concentrations of PPCPs and/or the correct concentration of Bti before larvae were transferred. Volume was checked periodically throughout the experiments with no noticeable difference. This methodology was used for all experiments.To determine the chronic LC50 of Bti for C. quinquefasciatus, mosquito larvae were treated with one of seven concentrations and an untreated control. Initially, doses of Bti were chosen by dividing a 24-h acute LC50 , determined by Mogren et al., by ten. Based on the resulting mortalities, an eight-dose concentration range was developed to determine the chronic LC50 covering the time span from second instar through adult eclosion. Second in stars were transferred from pans to bio-assay cups and given an 8-h acclimation period. If a larva died before the end of the acclimation period, it was replaced. Sixty individuals were used for each replicate, with three replicates per treatment. This replication was used throughout all experiments.

After treatment, larvae were maintained in incubators as previously described. Larvae were monitored daily until all larvae died or eclosed. The resulting chronic LC50 of Bti on C. quinquefasciatus was then used as a standard Bti concentration in all subsequent treatments with Bti and/or PPCPs and Bti.In order to determine the effect of PPCPs on growth and development of C. quinquefasciatus, larvae were reared in CGSW treated with each of the five PPCP treatments, or an untreated control. Mosquito larvae were reared and transferred to their respective bio-assay containers as stated previously. Larvae were monitored daily for growth, developmental stage, mortality, and number of days to pupation. The endosymbiont microbial community of mosquitoes reared under the various PPCP regimes was sequenced and quantified. Mosquitoes were reared in pans as described previously. Three subsets of ten mosquitoes were collected when mosquitoes reached the second, third, and fourth instar. Mosquitoes were then twice washed with 95% ethanol to remove any external microorganisms. After washing, larvae were transferred to a sterile 2 mL micro-centrifuge tube with 95% ethanol and frozen at -60 ± 3°C in an ultra cold freezer until DNA extraction. DNA was extracted using a Qiagen DNeasy® Blood and Tissue Kit following the manufacturers protocols with the following amendments. Mosquitoes were crushed by micropestles in a sterile 2 mL micro-centrifuge tube and 20 µL of Buffer ATL. After thorough pulverization, 160 µL of the Buffer ATL and the 20 µL of proteinase K was added. Nucleic acid concentration was quantified using a Nanodrop ND- 2000c Spectrophotometer . A commercial sequencing facility performed Roche 454 bacteria barcoded amplicon pyrosequencing .After PCR, all amplicons were mixed in equal concentrations and purified using Agencourt Ampure beads . Samples were sequenced with Roche 454 FLX titanium instruments and reagents following the manufacturer’s guidelines.Statistical analyses were performed in R . The chronic LC50 was calculated using logistic regression, and a logit link function. Lethal concentrations at 95% confidence intervals were calculated using the dose.p function of the MASS package. Growth and development were examined using a generalized linear model with a Poisson probability distribution. Individual treatments were examined using linear contrasts with the untreated control.

In susceptibility assays, overall significance in mortality was determined using ANOVA; individual significances were determined using a binomial generalized linear model. Bacterial community data from bTEFAP® was examined with respect to instar and treatment using “permutational MANOVA” . PERMANOVA is analogous to MANOVA but is robust to non-normality that is commonly associated with count data. Microbial community data from pyrosequencing was further examined with principal component analysis performed in the FactoMineR package.When the microbiome of third and fourth instar larvae was examined with PCA, there were 17 dimensions with an Eigenvalue greater than one. However, the first two dimensions explained 23% of the total variation . When examined across the first and second principal components , bacterial communities treated with acetaminophen and caffeine and the controls cluster together,plastic planter pot suggesting they are similar. The micro-biomes of the mosquitoes in the mixture or antibiotic only treatments are similar to each other. The hormone treated mosquitoes are distinct from all other treatment groups. There were 30 bacterial families with non-zero contributions to one of the first two principal components. Twenty of these families account for at least 96% of the PPCP-treated mosquitoes’ bacterial community and cluster in three distinct groups . Eight bacterial families each contribute greater than 0.01% to the overall bacterial community in all treatment groups . The family Enterobacteriaceae is mostly described by the first dimension and is associated with the control, acetaminophen, and caffeine treatments . Rickettsiaceae is the most abundant bacterial family in all of the treatment groups except for the hormonetreated mosquitoes, where it is second most abundant. Wolbachia pipientis accounts for >99% of this family . Microbacteriaceae is the most represented bacterial family in the treatment groups and, like Rickettsiaceae, has > 9% abundance in all treatment groups . However, the Microbacteriaceae species vary among treatments, but Rickettsiaceae bacteria are consistently represented by Wolbachia pipientis . While the eight families presented in Table 2.1 account for at least 96% of the bacterial community, the total counts are reduced by 66% in the antibiotic treatment and reduced by 33% the mixture treatments. Thus, while Wolbachia pipientis has a relatively similar number of counts in all distinct treatments, this species accounts for 86.7% of all bacteria in the antibiotic-treated and 69.1% of bacteria in the mosquitoes exposed to mixtures of PPCP.Previous studies reporting LC50 values for C. quinquefasciatus exposed to B. thuringiensis subsp. israelensishave only reported acute values . The dosedependent toxicity of Bti to C. quinquefasciatus documented in our study was used to calculate the first chronic LC50 of Bti on C. quinquefasciatus. This value of 10.20 ng/mL is much lower than previously reported acute LC50s determined in 24 h tests. For example an acute LC50 value was reported at 140.0 ng/mL by Mogren et al.. Thus, although acute assays are faster to conduct, and can be compared against previously published reports, they overestimate the amounts of Bti needed to kill 50 percent of mosquito populations that are exposed over their entire life spans. Bechmann 88 noted that in lifetable experiments some toxicants, especially pesticides, can drive a population to extinction even at concentrations well below an acute LC50. However, effective mosquito control typically requires suppression of late instar larvae even with only an acute exposure and a relatively high dose of Bti is be needed to achieve that goal. Mosquitoes treated with PPCPs at environmentally relevant concentrations displayed increased developmental time in the acetaminophen and mixture treatments. As these two treatments were not significantly different, it is possible the effects on mixture treated mosquitoes could be from acetaminophen alone.

The majority of these two treatments pupated 1-2 days after the control group. The cause of this delay is difficult to determine and may be due to individual or joint actions of many factors including an effect on the nutrients during rearing, an effect of these specific PPCPs on the mosquito physiology, or some general stress. Stress has been shown to influence insect physiology and behavior 89,90. Regardless of the cause, increased developmental time would increase larval exposure to PPCPs and potentially to predation. All mosquitoes treated with Bti exhibited an increase in developmental time of 1-2 days, including those in the acetaminophen and mixture treatments. Since Bti is a larvicide that damages the gut lumen in certain mosquitoes 91, a reduced nutrient absorption could lead to increased developmental time. Longer developmental times caused by PPCPs would increase exposure to contaminants and exposure time to Bti. Increased mortality would almost certainly occur. Larvae in the acetaminophen, antibiotic, and mixture treatments were more susceptible to Bti than the larvae exposed to Bti alone. While the increased time to exposure described above may have played a role in the increased mortality, the pyrosequencing results indicated significant changes in the bacterial community of the mosquito. If some of the affected bacterial communities, especially those eliminated, have detoxifying abilities for Bti-toxin this could also play a role. However, more research in the area is warranted. Previous studies show the inability of Anopheles mosquitoes to fend off the malaria parasite Plasmodium falciparum following treatment with the antibiotic gentamycin. Contrary to our findings, Broderick et al. showed the gypsy moth had a reduced susceptibility to Bacillus thuringiensis after treatment with antibiotics, which was removed after reintroducing the Enterobacter sp. NAB3. As the mosquitoes in our study retained their Enterobacteria, the Bti was still activated and the mosquitoes were still susceptible.In our study, larvae exposed to hormones contained substantially different bacterial communities as compared to controls, suggesting that at least some hormones likely play a role in altering bacterial communities.