The influence of larval density, food stress, and parasitism on the bionomics of the dengue vector Aedes aegypti (Diptera: Culicidae): implications for integrated vector management

New larval control strategies for integrated vector management of Aedes aegypti are in high demand, including the use of biological control agents. Exposure of Aedes aegypti to parasites, starvation, and overcrowded conditions during larval development reduces the probability of survival to eclosion, can directly affect fitness parameters such as adult size and fecundity, and can affect the size, provisioning, and viability of eggs produced by females. We compared these parameters after exposing larvae to 1) abundant food at low larval densities, 2) food deprivation and high larval density, and 2) infection with the endoparasite Plagiorchis elegans, an entomopathogenic digenean trematode. Female mosquitoes that eclosed from larval conditions of starvation and overcrowding were smaller and laid fewer and smaller eggs than controls. The proportion of females to complete an oviposition cycle was reduced in the P. elegans-infected treatment group. Parasite load was negatively correlated with wing length and egg size. Infection of Ae. aegypti with P. elegans has sublethal effects and may reduce population-level reproductive output, but one-time low-density P. elegans exposure does not have sufficient effect on Ae. aegypti fitness parameters to be considered a viable biocontrol option. The geographic distribution of Aedes aegypti is increasing worldwide (Gubler 2002). Integrated vector management (IVM), including intersectoral evidence-based vector management interventions aimed at preventing disease while decreasing dependence on insecticides, is increasingly popular as an effective and sustainable method of Ae. aegypti control and dengue prevention (Townson et al. 2005). Biological control of mosquito larvae can play an important role in achieving sustainable IVM programs as reduction of larval populations is considered a more effective disease prevention strategy than adult vector control (Rose 2001, Keiser et al. 2005). Understanding the effects of larval rearing conditions, parasitism, and food availability on the bionomics of this vector is key to developing new IVM intervention packages. Parasitic infection can reduce survival and impede or delay development of affected mosquito larvae (Platzer 2007, Zahiri et al. 1997, Petersen et al. 1978). Continuous exposure of a developing Ae. aegypti larvae to infective Plagiorchis elegans xiphidiocercariae significantly decreases the overall biomass of that pre-imago population, significantly decreases the probability of individual larvae to molt to a subsequent instar and/or eclose, and results in a smaller emergent adult mosquito population (Schwab et al. 2003). One-time individual exposure of Ae. aegypti larvae to P. elegans xiphidiocercariae can result in parasitism levels sufficient to kill pre-imago stages and produce impaired and moribund adults (Dempster et al. 1986, Lowenberger and Rau 1994a). Thus, both continuous and one-time exposure of Ae. aegypti to P. elegans have promise as biocontrol strategies. This paper examines the sublethal effects of this parasitism on the reproductive output and other fitness parameters of parasitized mosquitoes that survive to adulthood. Food availability also impacts the probability of larvae emerging as viable adults (Blaustein et al. 2004, Briegel 1990a,b, Lowenberger and Rau 1994b, Nasci 1986, Telang and Wells 2004) and the subsequent properties of adults. Larval conditions of starvation and overcrowding decrease teneral reserves, (Briegel 1990b, Van Handel and Day 1988), adult body size (Begon et al. 1986, Steinwascher 1982), flight potential (Nasci 1986), fecundity (Briegel 1990b), and the allocation of reserves and blood meal nutrients to reproduction (Briegel 1990b, Briegel et al. 2002, Steinwascher 1984, Tsunoda et al. 2010, Zhou et al. 2004). We assessed these dynamics to compare the effects of food availability, larval density, and parasitism on the fitness parameters and reproductive output of female Ae. aegypti. Poor larval rearing conditions, such as overcrowding and starvation, are negatively correlated with adult size, reserves, and fecundity in Ae. aegypti (Briegel 1990b). Endoparasites may cause physical injury, induce costly immune responses, and may sequester specific nutrients during development, all of which may also affect adult size, reserves, and fecundity of adult mosquitoes. Thus, we asked whether parasitic infection with P. elegans during larval stages has similar effects on adult body size and reproductive output as do overcrowded and food-deprived larval rearing conditions for Ae. aegypti mosquitoes. A laboratory colony of Ae. aegypti (LVP strain) was maintained at 27° C and 80% relative humidity with a 14:10 (L:D) light regime (Lowenberger et al. 1999). Adult mosquitoes were fed 10% sucrose solution ad libitum. We reared larvae in four different laboratory-created larval habitats corresponding to a 2×2 factorial design. These were (1) control group (CM), (2) density and food-stressed mosquitoes (DFS), (3) P. elegans-exposed mosquitoes (PEL), and (4) density and food-stressed and P. elegans-exposed group (DFS+PEL). Larvae in the control mosquitoes (CM) were not exposed to density/food stress or parasite burden. These larvae were reared at densities of ∼100/liter of distilled water and fed ∼ 10 ml/week of a fish food and water slurry (15 ml ground Nutrafin™ (Hagen Inc., Montreal, Canada) fish food in 30 ml of distilled water) to produce large adult mosquitoes. Density and food-stressed mosquito larvae (DFS) were exposed only to density/food stress. These were reared at densities of ∼600 larvae/liter of distilled water and fed 1 ml of fish food slurry per week to produce smaller adult mosquitoes (Lowenberger and Rau 1994b). Larvae stressed with P. elegans (PEL) were reared at densities of ∼100/liter of distilled water and fed ∼10 ml/week of a fish food and water slurry. On day three of their development, each PEL mosquito larvae was placed individually in a well of a 96-well plate containing ∼0.5 ml of distilled water and was exposed to five 6–8 h-old P. elegans xiphidiocercariae for 5 h (Lowenberger and Rau 1994b). The parasites were maintained in their snail intermediate hosts, Stagnicola elodes (Say) and Lymnaea stagnalis (L.), in aerated 38 liter aquaria at room temperature with a 16:8 (L:D) light regime, and fed Nutrafin™ (Hagen Inc., Montreal, Canada) fish food and fresh Romaine lettuce ad libitum (Lowenberger and Rau 1993, Zakikhani and Rau 1999). To obtain cecrariae, snails were placed in clear plastic cups containing 20 ml of water prior to scotophase, when cercariae emerge. The number of cercariae per larvae was chosen to allow for the effects of parasitism without excessive numbers that would kill the larvae. The last treatment group was stressed with both density and food stress and the parasite (DFS+PEL). However, due to the extremely high pre-imago mortality rate, individuals from this group were subsequently excluded from the analyses. Mosquitoes from the control and DFS treatment groups were separated by sex as pupae and combined into different mating pairs: CM♀× CM♂, CM♀× DFS♂, DFS♀× CM♂, and DFS♀× DFS♂ to investigate paternal and maternal fitness effects on the F1 generation. For each mating cross, 15 pairs were isolated per replicate and maintained in 355 ml Sweetheart® food cups with screened lids and access to 10% sucrose ad libitum. Females were blood fed to repletion on the arm of K.M.-F. three days after eclosion and provided with an oviposition substrate of a moistened paper towel 72 h after the first blood meal. Oviposition substrates were remoistened every 24 h, then removed from the cup six days after feeding and stored individually in zipper-locked plastic bags until processed. Females were blood fed to repletion a second time and the oviposition protocol was repeated. If females failed to feed 72 h post-eclosion they were offered a blood meal 24 and 48 h later. Unfed females at 120 h post-eclosion were excluded from the study; only females that fed to repletion were followed. Because preliminary CM and DFS data indicated that male size had no significant effects on egg number or egg size (Mitchell-Foster unpublished), CM and PEL mosquitoes were isolated as pupae into mating pairs: CM♀× CM♂ and PEL♀× PEL♂. For each of these mating pair types, 23 pairs were isolated per replicate. Adult maintenance, blood-feeding, oviposition, and parameter measurement protocols were carried out as described above. After the second gonotrophic cycle, a full-body smear of each PEL adult female was examined microscopically to establish the number of developed P. elegans metacercariae within adult tissue and PEL females were subsequently assigned to parasite load classes having zero, one, or two or more metacercariae. At the end of a gonotrophic cycle, the number of eggs laid and average egg size were recorded for each female. Egg size was determined by measuring the longest dimension of the egg using an ocular micrometer in a binocular dissecting microscope. We also recorded the proportion of females that laid eggs in each treatment group. Females that did not complete oviposition were maintained on the same feeding and oviposition substrate protocols and timelines as those who did complete oviposition; their egg counts were recorded as zero. After the second gonotrophic cycle was completed, the females were sacrificed and their winglengths were measured using an ocular micrometer in a binocular dissection microscope. Samples of ten eggs each were taken from each of 15 different females from each treatment group (CM, DFS, and PEL) and placed in covered 60 × 100 plastic Petri plates containing 10 ml of autoclaved double-distilled water with one drop of Nutrafin™ fish food slurry. The Petri plates were kept at 27° C, 80% humidity, and 14:10 L:D regime for 48 h to allow eggs to hatch and larvae to develop. The number of 1st instar larvae that emerged was counted and the proportion of eggs hatched per sample for each treatment was calculated. This experiment was repeated for each of two gonotrophic cycles. Pupae were taken from each treatment group to measure the average levels of protein, carbohydrate, and lipid concentration. It was not possible to gather more than one energy constituent from a single individual, so different individuals were used for each reserve analysis. These reserves were also measured for the eggs laid from female adults from each treatment. To get detectable levels of protein, carbohydrates, and lipids from eggs, it was necessary to combine ten to fifteen eggs in one sample. To determine pupal protein reserves, ten individual pupae from each of the CM, PEL, and DFS treatment groups were homogenized in 200 μl of Triton-X-NaOH in a 1.5 ml microcentrifuge tube and centrifuged at 9600 rpm at room temperature for 1 min. Then, 10 μl of the supernatant was added to 500 μl of Bio-rad Quick Start Bradford Reagent (Bradford 1976) in a clean microcentrifuge tube and incubated at room temperature for 5 min. Samples were mixed by pipette to ensure homogeneity of color and the absorbance was measured at 595 nm using an Eppendorf Biophotometer (Eppendorf, Hamburg, Germany). Each sample to measure egg protein consisted of ten eggs. Samples taken from ten females from each treatment group (CM, DFS, and PEL) were homogenized in 100 μl of Triton-X-NaOH in a 1.5 ml microcentrifuge tube and centrifuged at 9600 rpm at room temperature for 1 min. Twenty microliters of the supernatant were added to 500 μl Bradford reagent and the absorbance measured at 595 nm as described above. The protein content for all samples was estimated by extrapolation to a standard curve made using Bio-Rad bovine gamma-globulin protein standards of 12.5, 25, and 50 μg total protein content. To determine their carbohydrate content, individual pupae (ten from each treatment group) and groups of eggs (15 from each treatment group) were homogenized in 5 ml of anthrone reagent (Van Handel 1985a) in a clean glass culture tube using a glass rod. Samples were placed in a boiling water bath for 17 min and allowed to cool to room temperature. Samples were mixed by inversion to ensure homogeneity of color and the absorbance was measured at 625 nm in a Beckman DU® 640 spectrophotometer. Carbohydrate content for all samples was estimated by extrapolation to a standard curve using glucose standards of 10, 25, and 50 μg total carbohydrate content (Van Handel 1985a). Lipid content was measured in individual pupae (ten from each treatment group) and groups of eggs (15 from each treatment group) that were homogenized in 500 μl of a 1:1 chloroform:methanol mixture in a clean glass culture tube. Samples were evaporated to dryness in a boiling water bath. Then, 200 μl of sulfuric acid (H2SO4) was added to each sample before heating again in a boiling water bath for 10 min. Samples were allowed to cool to room temperature and 5 ml of vanillin reagent (Van Handel 1985b) was added and color allowed to develop for 5 min. Samples were mixed by inversion to ensure homogeneity of color and the absorbance was measured at 525 nm with a Beckman DU® 640 spectrophotometer. If absorbance at 525 nm was above 1.0, samples were read at 490 nm or diluted 5× with fresh vanillin reagent and reread at 525 nm. Lipid content for all samples was estimated by extrapolation to a standard curve using lipid standards of 10, 25, and 50 μg total lipid content (Van Handel 1985b). All statistical analyses were performed with SAS 9.1 statistical software. We determined the effect of density/food stress and parasite stress on wing length, egg number, and egg size using multivariate analysis of variance (MANOVA). The effect of the gonotrophic cycle was included as an independent variable. We performed separate analysis of variance (ANOVA) with contrasts between each pair of treatments on each dependent variable separately. To account for the multiple tests used here, we used Bonferroni-corrected alpha-values of 0.05/3 = 0.0167. We compared the proportion of females laying eggs using Chi-Square contingency analysis. Of those that laid eggs, we also examined egg viability by analysis of variance (ANOVA) and Tukey-Kramer pairwise analyses. We used a series of one-way ANOVAs with contrasts to compare the total protein, carbohydrate, and lipid quantities. Because of the three comparisons made among the pairs of treatment groups here, we used Bonferroni corrected alpha-values of 0.05/3 = 0.0167. No multivariate analysis of variance for the energy constituent analysis was performed because individuals were sacrificed for each energy test and did not represent levels of protein, carbohydrate, and lipid for a single individual. Additionally, we tested for the effect of parasite load on egg number, egg size, and wing length within the PEL treatment group using an ANOVA with Tukey-Kramer pairwise comparisons. There was no significant difference in the wing lengths of F1 offspring (F1,26=4.35; P=0.0775), egg size (F1,26=0.59; P=0.2355), or egg number (F1,26=1.63; P=0.5062) for CM females mated with CM males compared with DFS males. Similarly, there was no significant difference for the wing lengths of F1 offspring (F1,26=2.09; P=0.1439), egg size (F1,26=0.291; P=0.9271), or egg number (F1,26=0.0004; P=0.9962) in DFS females mated with either CM or DFS males. Male size had no effect on either egg size (F1,26=0.1210; P=0.7870), or egg number (F1,26; P=0.5597), indicating that male size has no effect. Because there was no effect of mating pair, PEL and CM females were mated with PEL and CM when control and PEL treatment of the larvae to eclosion and were excluded from the analyses. Of the females that laid eggs, there was a overall effect of among the treatment groups to the difference in the DFS group compared to the other two groups CM PEL CM DFS DFS PEL We no significant in measurement between the numbers or size of eggs produced by the same females subsequent gonotrophic this was not included in the The separate ANOVAs with contrasts that the DFS females were smaller than the CM and PEL females females length ANOVA with CM PEL CM DFS DFS PEL and they laid fewer number ANOVA with CM PEL CM DFS DFS PEL and smaller eggs size ANOVA with CM PEL CM DFS DFS PEL There were no significant between the CM and PEL groups in of wing length, egg number, and egg size. the for egg size in PEL CM is than this is not significant when we corrected for multiple comparisons using the of of 0.0167. of egg number egg size and wing length for female Aedes aegypti different (CM), Density Food (DFS), and Plagiorchis elegans-infected mosquitoes There was an overall effect of treatment on the proportion of females that laid eggs analysis = This was due to a proportion of CM females laying eggs than PEL females = or DFS females = There was no difference in the number of females laying eggs between the PEL or DFS treatments = Of the females that did there were no significant in the proportion of eggs laid by CM and PEL females that hatched = PEL = these were significantly than the proportion of eggs laid by DFS females Tukey-Kramer Parasite load was negatively correlated with wing length and egg size with more parasitized females having wing lengths and smaller eggs The egg number was for females with high parasite but this was not significant after Bonferroni Parasite load had a significant effect on of PEL females laying eggs = of females laid eggs load of females with one P. elegans metacercariae laid eggs load and of females with two or more P. elegans metacercariae laid eggs load of egg number egg size and wing length for Plagiorchis elegans-infected Aedes aegypti mosquitoes Parasite load classes and to individuals with zero, one, and two or more metacercariae by of larval full-body There were no significant in the total carbohydrates, and lipids among the eggs of all treatment groups protein lipids There were significant in the levels of total carbohydrates, and lipids measured in pupae pupae protein lipids with the DFS group having levels than either PEL or CM mosquitoes protein CM PEL CM DFS DFS PEL, CM PEL CM DFS DFS PEL The PEL pupae had total than both the CM and DFS mosquitoes. of protein and lipids for Aedes aegypti larvae different (CM), Density Food (DFS), and Plagiorchis elegans-infected mosquitoes of protein and lipids for Aedes aegypti pupae different (CM), Density Food (DFS), and Plagiorchis elegans-infected mosquitoes We examined the effect of and P. elegans parasite load on the reproductive output of Ae. aegypti mosquitoes to a of the effects of these on the bionomics of potential dengue vector results density/food stress and parasite infection can both result in a reduced reproductive output of Ae. aegypti females at the population length, egg number, and egg size for the different treatment groups that the larval has a significant on adult as by (Briegel 1990b). Adult mosquitoes that emerged from and conditions (DFS) were smaller and laid fewer and smaller eggs that were viable than those from the control (CM) and P. elegans-exposed (PEL) mosquitoes. size and blood meal can affect both egg number and egg size. The effects that result in eggs are not A emerging from a egg laid by a female in a larval develop into a large adult from the of a large egg laid by a large female developing in the same larval (Mitchell-Foster to DFS there were no significant in the wing length, egg the proportion of females that laid eggs, and the proportion of eggs that hatched in CM and PEL treatment There was a between parasite stress and egg size, but overall this was not significant using the Bonferroni when the effect of parasite load was examined in more within the PEL females we that parasite load was negatively correlated with wing length and egg size in more parasitized females. These data there is an effect of parasites on the mosquito and this have more had we the of PEL treatment is as by both et al. and Lowenberger and Rau who used more than the exposure that we We for levels of parasitism to the probability that individuals would survive to eclosion and be to reproductive to of effects of parasitism (Dempster et al. parasitized Ae. aegypti development and levels of larval and pupal mortality and reduced eclosion (Dempster et al. 1986, et al. 2002, Rau et al. et al. 2003). Each parasite that the may an immune by the et al. 2010, et al. and the increasing and parasite load may be (Schwab et al. 2003). with more cercariae exposure is Parasite load also had an effect on the proportion of females to complete of the females with one or more P. elegans metacercariae laid eggs than females. The difference in of females laying eggs may represent a fitness than a females that did not eggs had eggs in their not these females were not to not or were nutrients to not be determined by design. a smaller proportion of parasitized females a gonotrophic cycle than did their reproductive output at the population The significant in the total pupal reserves among the CM, PEL, and DFS treatments may be a result of the size difference between the on the adult as a for individual size, CM and PEL pupae are significantly than DFS this difference in pupal reserves is to be The difference in total carbohydrate reserves between CM and PEL pupae is and is to whether parasitism with P. elegans specific to larval development or in with an of may to low glucose availability for and may in the larvae et al. individuals with for specific that may nutrients for specific The protocols used in this did not allow to than the total levels per group. of P. elegans in the of the are and do carbohydrate and protein reserves from their (Lowenberger et al. Lowenberger and Rau which may not be in of total reserves for of the larvae to eclosion and were excluded from the analyses. We not determine this was the result of between two for specific or whether the larvae in with parasites were to nutrients to and This does not that a one-time exposure of an Ae. aegypti larval population to P. elegans would be an effective of biological control as of an IVM et al. repeated exposure of a larval population to densities of P. elegans cercariae would have a with pre-imago parasite This the dynamics of from snail of cercariae are from their first intermediate snail a of or (Lowenberger and Rau 1994a). However, the mortality for and overcrowded larvae exposed to P. elegans xiphidiocercariae the treatment group) an for biological control for with high larval the effects of parasite load on adult size and pupal carbohydrate reserves using repeated We Van and at and and of the for this was provided from The and the to

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.