Aedes aegypti is the main vector of dengue, a disease that is increasing its geographical range as well as incidence rates. Despite its public health importance, the effect of dengue virus (DENV) on some mosquito traits remains unknown. Here, we investigated the impact of DENV-2 infection on the feeding behavior, survival, oviposition success and fecundity of Ae. aegypti females.
After orally-challenging Ae. aegypti females with a DENV-2 strain using a membrane feeder, we monitored the feeding behavior, survival, oviposition success and fecundity throughout the mosquito lifespan. We observed an age-dependent cost of DENV infection on mosquito feeding behavior and fecundity. Infected individuals took more time to ingest blood from anesthetized mice in the 2nd and 3rd weeks post-infection, and also longer overall blood-feeding times in the 3rd week post-infection, when females were around 20 days old. Often, infected Ae. aegypti females did not lay eggs and when they were laid, smaller number of eggs were laid compared to uninfected controls. A reduction in the number of eggs laid per female was evident starting on the 3rd week post-infection. DENV-2 negatively affected mosquito lifespan, since overall the longevity of infected females was halved compared to that of the uninfected control group.
The DENV-2 strain tested significantly affected Ae. aegypti traits directly correlated with vectorial capacity or mosquito population density, such as feeding behavior, survival, fecundity and oviposition success. Infected mosquitoes spent more time ingesting blood, had reduced lifespan, laid eggs less frequently, and when they did lay eggs, the clutches were smaller than uninfected mosquitoes.
Citation: Sylvestre G, Gandini M, Maciel-de-Freitas R (2013) Age-Dependent Effects of Oral Infection with Dengue Virus on Aedes aegypti (Diptera: Culicidae) Feeding Behavior, Survival, Oviposition Success and Fecundity. PLoS ONE 8(3): e59933. doi:10.1371/journal.pone.0059933
Editor: Lark L. Coffey, Blood Systems Research Institute, United States of America
Received: November 28, 2012; Accepted: February 20, 2013; Published: March 29, 2013
Copyright: © 2013 Sylvestre et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Funding: The authors have no support or funding to report.
Competing interests: The authors have declared that no competing interests exist.
Dengue has become one of the most important vector-borne diseases in the recent decades because of its increasing incidence, high morbidity and economic impact. The World Health Organization (WHO) estimates that 50 million cases of dengue occur per year, with more than 2.5 billion people living at risk of infection in endemic areas. The geographic extension of dengue is primarily confined to the tropical areas of Southeast Asia, the Pacific and the Americas, but regions in Africa and the eastern Mediterranean are increasingly affected .
Dengue virus (DENV) belongs to the family Flaviviridae and is transmitted between humans through the bite of the mosquito Aedes aegypti, its only known natural vector in the Americas. This mosquito species is commonly found in high numbers in urban areas, living close to human dwellings, where females usually blood-feed on humans and 3–4 days later lay eggs in man-made containers –. Thus, reducing or eliminating larval habitat located in and around human habitations has been advocated as an important component of sustainable vector control programs –.
It is believed that the intensity of dengue transmission is largely influenced by the parameters governing Ae. aegypti vectorial capacity such as the population density of the mosquito, its biting rate, its daily survival rate, its susceptibility to DENV infection and the extrinsic incubation period of the virus , . A thorough understanding of DENV epidemiology therefore depends upon a detailed understanding of how these life history traits affect Ae. aegypti vectorial competence. Furthermore, little is known about how the mosquito–dengue interaction itself influences these epidemiologically relevant parameters. Like other arboviruses, dengue virus is able to invade the mosquito’s brain, which may modify the mosquito’s physiology, metabolism and behavior, thereby changing vectorial capacity and the pattern of disease transmission , . In this context, estimates of the mortality, longevity and biting rate of dengue-infected Ae. aegypti still need further characterization –. Most mathematical models assume that there is little or no effect of dengue virus infection on these parameters or if there is, it is not well understood . Recently, some ecological aspects of the interaction between DENV and Ae. aegypti have been explored. For instance, it was observed that mosquitoes challenged with DENV had lower survival rates and longevity when compared to those uninfected mosquitoes . Furthermore, DENV had an age-dependent effect on mosquito fecundity, since infected females laid fewer eggs per clutch than uninfected controls in the third and subsequent oviposition cycles . Ae. aegypti females infected intra-thoracically with DENV-2 also had an increase of up to 50% in their locomotor activity when compared to uninfected control mosquitoes .
Dengue transmission is heavily dependent on mosquito biting behavior. As with other disease vectors, Ae. aegypti females are attracted by visual and chemical cues such as carbon dioxide, air movement and heat, as well as body odors from human skin such as lactic acid and ammonia . Once a suitable host is located, the mosquito must obtain blood as quickly as possible to avoid any host defensive behavior . Yet, the potential influence that DENV may play on Ae. aegypti biting behavior remains poorly known. For instance, only two studies have considered the biting rate of DENV-infected Ae. aegypti and conflicting results were reported , . Using mosquitoes from long-established laboratory colonies, Putnam and Scott  found no evidence that DENV-2 influences the feeding behavior of the mosquito (i.e. the biting rate). On the other hand, Platt et al.  observed that the time required for blood-feeding and the time spent during probing were longer in dengue-infected mosquitoes than in uninfected individuals. Furthermore, the aforementioned papers performed intra-thoracic inoculation to infect mosquitoes, which is an invasive and non-natural mode of transmission that has important consequences on the dynamics of virus-vector interactions .
In this study, we compared the effect of oral dengue infection on some life-history traits of Ae. aegypti such as the mortality, longevity, age-specific fecundity, oviposition success and biting behavior of two mosquito populations.
Two Ae. aegypti populations were tested, since previous experiments revealed the effects of DENV-2 infection varied between mosquito populations . The field population was derived from the F1-generation of mosquitoes collected at Fiocruz campus, Rio de Janeiro, using 80 ovitraps filled with hay infusion and distributed in an area of approximately 2.32 hectares, in close proximity to a typical Brazilian slum, which historically has a high incidence of dengue. The laboratory colony used was the Paea strain, which is highly susceptible to oral dengue infection . The colony was initiated with mosquitoes caught in French Polynesia in 1994. Based on the observation of 15–20 generations per year in our laboratory, the Paea strain has been maintained for approximately 250–300 generations .
In all assays, larval populations were reared and maintained under identical laboratory conditions. Larvae were reared on yeast extract and raised in plastic basins at 25±3°C. After emergence, adults were allowed to mate and maintained in 45 cm3 cages kept at 27±2°C and 75±5% relative humidity and in approximately 12–12 h light-dark photoperiod. They were fed ad libitum with cotton soaked in a 10% sucrose solution for up to approximately 36 h before being offered an infectious bloodmeal to the Ae. aegypti females.
Dengue virus type 2 strain 16681  with 6 passages on C6/36 cells was used for virus stock preparation as described elsewhere . Briefly, Aedes albopictus cell clone C6/36 (CRL-1660, ATCC) was maintained at 28°C in Dubelcco’s modified Eagle Medium (Gibco, Life Technologies) with sodium bicarbonate and supplemented with 5% fetal bovine serum (FBS), 1% penicillin-streptomycin-glutamine (Gibco), 0.5% non-essential amino acids (Gibco) and 10% tryptose phosphate (Sigma). C6/36 cell monolayers were infected with DENV-2 and cell culture supernatants were harvested 8 days later. A purified DENV-2 stock was obtained by ultracentrifugation at 100,000 g for 1 h and set to a final volume 20 times smaller than initial . Titration was performed in C6/36 cells using a standard TCID50 (50% tissue culture infective dose) assay as described elsewhere . Uninfected flasks were maintained, also purified and used as negative control (MOCK).
DENV-2 Vector Competence Assay
Before performing the experiments investigating the effect of DENV-2 on Ae. aegypti life-history parameters, we carried out controlled preliminary assays to assess the vector competence of the two Aedes aegypti populations to our DENV-2 strain. For this, 4–5 days after emergence, 60 inseminated females were placed into small cylindrical plastic cages, with no access to sugar. Approximately 36 h later, they were offered a DENV-2 infectious blood meal, which consisted of one milliliter of viral stock added to 2 ml of washed rabbit erythrocytes. The blood-meal was heated to 37°C and given to the mosquitoes in an artificial membrane feeding apparatus . The mosquitoes were allowed to feed for 25 min on infectious blood that contained a viral titre of 2×108 TCID50. Fourteen days post-infection (dpi), the heads of the mosquitoes were examined by indirect immunofluorescence to detect DENV.
Due to the satisfactory susceptibility of Ae. aegypti mosquitoes to the DENV-2 strain used in these experiments (data in the Results section), we assumed all mosquitoes from both populations that received the same viral stock were infected.
Oral Infection of Mosquitoes with DENV-2
For the experiments investigating the effect of DENV-2 on Ae. aegypti life-history parameters, for each Ae. aegypti population, female mosquitoes were split into two groups, one of which received an infectious blood meal as described above, while the other received a non-infectious blood meal, in which the viral supernatant was replaced with 1 ml of culture medium. The same procedure and apparatus were otherwise used to blood-feed both groups of mosquitoes.
Those females that were visually fully engorged were isolated in labeled cylindrical plastic vials (6.5 cm height, 2.5 cm diameter) containing at the bottom moistened cotton, overlaid with filter paper, as a substrate for oviposition, and closed on the top with mosquito netting.
Aedes aegypti Feeding Behavior
In order to estimate the potential changes in Ae. aegypti feeding behavior, we offered an anesthetized mouse to the mosquitoes once a week, for a period of three weeks. At each blood feeding event, (7, 14 and 21 dpi), we measured five different aspects of mosquito blood-feeding.
The period of time elapsed between contact of the vial with the mouse until the mosquito introduced its proboscis into the body of the host. Herein, we aimed to assess the host-seeking behavior and evaluate whether DENV- infected individuals were more efficient in locating the host and initiating the feeding process than uninfected mosquitoes.
Number of probings.
The number of times a female inserted its proboscis into the host during blood feeding. This parameter is perhaps the most epidemiologically relevant, because an increase in probing may lead to a direct increase in virus transmission.
The time elapsed between the beginning until the end of probing. This time ended when we observed the abdomen beginning to expand due to blood ingestion. Here, we intended to evaluate whether DENV-infected individuals initiate ingestion of blood more rapidly than uninfected mosquitoes.
The time elapsed from the beginning of abdominal expansion to the end of engorgement, which was characterized by proboscis withdrawal from the host body. Here, we intended to evaluate whether infected individuals would feed to repletion more rapidly than the control group, thus favoring virus transmission by reducing the time vector is exposed to host defensive behavior.
Mosquito Fecundity and Survival
Three days after each blood meal, the filter papers were removed and checked for eggs, which were counted, and a new filter paper was added as oviposition substrate to the vials. Survival was checked daily at 09∶00 am and 04∶00 pm. When a dead mosquito was observed, it was removed from the plastic tube, and both wing lengths were measured as the distance from the axillary incision to the apical margin excluding the fringe .
As a test for whether DENV-2 altered any of the five tested components of Ae. aegypti blood feeding behavior, we used survival analysis to censor the time at which mosquitoes started feeding, probing and ingesting blood as the censoring event. We had a total of four curves, which were derived from the two populations (field/lab) and one treatment (infected/control mosquitoes). Initially, we performed a Log-Rank test to compare the four curves. If this first test revealed statistically significant differences between the four curves, we then ran a paired Log-Rank with Bonferroni correction to identify which curves differed from the others. Finally, the number of probes between control and infected groups in weeks 1 to 3 after DENV infection were compared using a Mann-Whitney U test.
Aedes aegypti longevity (which was defined by the day each mosquito died) was non-normally distributed, but the square-root of longevity satisfied the assumption of normality (Shapiro-Wilk W = 0.996, P = 0.081). We analyzed the effect of treatment (control or infected), mosquito population (field or lab) and looked for a correlation between wing length and mosquito longevity by using an ANOVA. We included in our initial model all main effects and interactions, and then deleted the highest order interaction (three way) when it was not significant, and retested the reduced model.
We performed a log-rank test to compare on a two-sample basis the survival distribution of Ae. aegypti females from control/treatments as well as field/lab populations. Herein, we define survival rate as the number of individuals still alive as a function of time. If log-rank detected that at least one of the four curves was statistically different, paired comparisons with Bonferroni corrections were performed.
Fecundity was analyzed by considering the first four clutches of eggs laid, as only a small number of females laid eggs when they were more than five weeks old, precluding adequate numbers for analysis. We analyzed two aspects of fecundity. First, we analyzed the oviposition success, i.e. likelihood that a mosquito laid at least one egg (at a given clutch) with a logistic analysis that included treatment, population, wing length and clutch-number (i.e. age), again backwards-eliminating the higher order interactions if they were non-significant. Second, we analyzed the number of eggs per clutch from those mosquitoes that laid at least one egg, using a repeated measures analysis. We square-root transformed the number of eggs to satisfy the assumptions of normality. We included clutch-number as the variable repeatedly measured and estimated the effects of treatment, population and looked for a correlation to wing length. All analyses were carried out with the statistical software JMP 9 (http://www.jmp.com/).
Vector Competence Assay
From a total of 49 mosquitoes that received an infective blood meal, 45 of them survived up to 14 dpi. Of these, 42 (93.3%) had DENV in the head, suggesting a satisfactory susceptibility of our mosquitoes to the DENV-2 strain used.
Effect of DENV-2 on Ae. aegypti Life-history Traits
We used a total of 173 Ae. aegypti females in this study, 123 mosquitoes were infected with DENV-2 and 50 were controls, i.e. received only uninfected blood meals. Of the 173 mosquitoes, 96 (55%) were from the field population and 77 (45%) were from the Paea strain.
Overall, DENV-2 produced some age-dependent alterations on Aedes aegypti feeding behavior. Host-seeking behavior, i.e. efficiency of locating the mouse and beginning the blood feeding process was not influenced by DENV-2 infection at any of the post-infection time points evaluated (Table 1). Probing time was also unaffected by DENV-2 infection, indicating that infected individuals do not to start blood-feeding faster, as originally hypothesized. However, although the ingestion time for infected and control individuals was statistically similar in the 1st week, in the 2nd and 3rd weeks infected mosquitoes spent more time ingesting blood compared to uninfected counterparts (Second week: χ2 = 8.38, df = 3, P = 0.031; Third week: χ2 = 10.25, df = 3, P = 0.017). Finally, on the third week after infection, the overall feeding time of DENV-2 infected females was longer than that observed for controls (Table 1).
The average number of probes per female was not significantly different between infected and uninfected mosquitoes (First week: U = 1323.5, P = 0.657; Second week: U = 923, P = 0.305; Third week: U = 640.5, P = 0.719).
Overall, for both mosquito populations combined, uninfected Ae. aegypti females lived longer (median = 18 days) than the DENV-infected group (median = 11 days). Infected mosquitoes exhibited higher mortality than controls (χ2 = 22.7, P<0.001), especially on the first ten days after DENV-2 infection (Figure 1). The two mosquito populations reacted differently after being challenged with DENV-2. Ae. aegypti females from Paea strain had shorter survival when infected (χ2 = 16.5, P<0.001), whereas DENV-2 infection had no influence on the survivorship of mosquitoes from the field population (χ2 = 3.4, P = 0.063). When uninfected, mosquitoes from the Paea strain had a similar survival rate to that observed for field population females. When infected, the Paea mosquitoes survived less than the field population (χ2 = 5.44, P = 0.02). When considering one single point on the mortality curve we observed that 75.2% and 34.3% of the mosquitoes from infected and control groups, respectively, were still alive on the 14th day after infection, when the virus can be expected to have completed its extrinsic incubation period in the constant temperature of 27±2°C of our insectary.
DENV-infected females demonstrated significantly higher mortality than controls (Table 2). Remarkably, mosquito wing size and population, as well as the interactions among tested variables were not statistically informative in the determination of mosquito mortality (Table 2).
The first part of our analysis of fecundity considered whether or not a female laid eggs at all. Similar proportions, 86% of infected and 78.8% of control Ae. aegypti females laid at least one egg during any of the four oviposition cycles evaluated here. Overall, egg-laying success was strongly affected by mosquito age, (χ2 = 15.03, d.f. = 3, P = 0.0016) decreasing with female age from 78.04% success rate at the first clutch to 70.84% at the fourth egg clutch (Table 3). There was a significant interaction between mosquito age and treatment, with oviposition success dropping more rapidly with age in infected mosquitoes, especially for the third and fourth egg clutches (Figure 2A). Another important interaction was observed between mosquito age and population, with egg-laying success dropping faster with age in mosquitoes from the field population rather that in those from the Paea strain.
(A) Effect of DENV infection on the oviposition success of Ae. aegypti females during four oviposition cycles and (B) the mean number of eggs laid per female per oviposition cycle.
The second part of the analysis considered the number of eggs from females that laid at least one egg (Table 4). Overall, we observed a tendency of a decreasing number of eggs with time irrespective of mosquito treatment and population, from a mean of 104.78 in the first clutch to 86.57 in the fourth clutch (Figure 2B). The number of eggs laid by infected females dropped more rapidly than it did in the control group after the third oviposition cycle and irrespective of mosquito population (Figure 2B, Table 4).
We designed experiments to study the potential impact of DENV infection on some Ae. aegypti life-history traits by orally challenging female mosquitoes. We showed that Ae. aegypti females presented significant negative effects in longevity, fecundity and oviposition success once infected with DENV-2. Furthermore, DENV-infected mosquitoes required more time to ingest blood and complete a blood meal. This phenomenon observed two and three weeks after infection, when mosquitoes were around 14 and 20 days old. By using the F1 generation of a field-collected Ae. aegypti population that was orally infected rather than intrathoracically challenged and a dengue strain with low passage history we created a close simulation of natural transmission.
The potential effect of DENV infection on Ae. aegypti feeding behavior has received minimal attention, with only a few studies addressing this subject despite its epidemiological relevance. Two studies considered the biting rate of DENV-infected Ae. aegypti, but presented opposing conclusions. Using mosquitoes from long-established laboratory colonies, a DENV-2 with a long but unknown passage history and infecting mosquitoes by intra-thoracic inoculation, (i.e., a non-natural system) Putnam and Scott  found no evidence that DENV-2 influences the mosquito’s feeding behavior (i.e., biting rate). Using F1 to F3 generations of field-collected Ae. aegypti intra-thoracically inoculated with a low passage DENV-3 strain, it was demonstrated that the time required for blood-feeding and the time spent during probing was longer in dengue-infected mosquitoes than in uninfected individuals .
Our results showed age-dependent alterations in the feeding behavior of infected individuals in comparison with their uninfected counterparts, especially by increasing the time required to ingest blood to repletion and on the total feeding time. If a mosquito takes longer to complete a blood meal due to DENV infection, the chances of being killed by host defensive behavior are increased, thus potentially reducing parasite transmission rate would not be maximized . The finding of age-dependent modifications might be one of the potential outcomes of DENV infection considering the dynamics and tropism of DENV in the Ae. aegypti body, which was elegantly shown elsewhere . At 7 dpi, less that 50% of mosquitoes from F3 to F6 generation of the Chetumal strain had DENV in their head tissues, and between 51 to 75% of them had DENV in their salivary glands. At 14 and 21 dpi, up to 95% of Ae. aegypti females had DENV in their head tissues and salivary glands . Our data suggest that somehow the dissemination of DENV through the mosquito over time may be correlated with the age-dependent effects on the life-history traits we observed.
The effect of DENV infection on Ae. aegypti longevity is a crucial parameter to determine the vectorial capacity of a mosquito population. In order to support dengue transmission, a mosquito must survive at least 10–14 days. Despite its public health importance, little is known about the Ae. aegypti–DENV interactions, particularly the longevity of DENV-infected mosquitoes. One study reported that DENV-2 reduced mosquito longevity and survival rate of infected individuals . These authors speculated that the high mortality of the infected group was due to an intense immune activation. While eliminating the viral infection, the immune response may create a biological burden that consequently increases mortality, as has been described for other vector-pathogen systems , . The cellular and humoral immunity responses of Ae. aegypti mosquitoes to arboviruses such as DENV may use RNA intereference, a major component of the mosquito innate immune response, to modulate infection by producing molecules that inhibit virus replication –. Although there is no prior evidence that Ae. aegypti mosquitoes experience a fitness cost due to DENV-specific immune reactions, other vector-pathogen systems have reported similar findings –.
The effects of DENV infection on Ae. aegypti were more evident on fecundity rather than on the feeding behavior and survival since infected mosquitoes were qualitatively (exhibited reduced oviposition success) and quantitatively (laid fewer eggs per female) affected. Joshi et al.  observed that the fertility and fecundity in vertically DENV-infected Ae. aegypti batches were lower than in control individuals. Our results considered horizontally (orally) infected mosquitoes and showed that the effect of DENV on fecundity varied over the mosquito lifespan. In the first two clutches, infected and uninfected mosquitoes laid similar numbers of eggs. On the third clutch, infected females had lower fecundity than the control treatment, corroborating the negative impact of DENV on the number of eggs laid per female per oviposition cycle . Ae. aegypti females laid the eggs of their 2nd and 3rd clutches around 11 and 18 days after infection, respectively. It is possible that an immune response elicited by DENV infection created a fitness cost that was revealed by a reduction in fecundity. Furthermore, around 15 days after infection, several mosquito tissues such as the midgut, nervous system and salivary glands are severely infected by dengue virus . At around 15 dpi the DENV may be found in females ovaries, which may have a direct impact on the oviposition success as well as the number of eggs laid per female per gonotrophic cycle , .
The biological relevance of the reduction of oviposition success and fecundity in DENV-infected individuals might be limited for several reasons. First, Ae. aegypti daily survival rates in the field suggest that only a small number of mosquitoes would survive to reach the 3rd oviposition cycle, which is where we observed the strongest reduction in fecundity . Additionally, because the incidence of DENV infection in natural mosquito population is extremely low, any differential effect on fecundity that DENV infection might impose on female mosquito is likely negligible. Parasite-induced fecundity reduction in other vector/parasite interactions has already been observed for the Ae. aegypti-DENV system , , as well as in Leishmania-infected sandflies , Dirofilaria-infected Ae. trivittatus mosquitoes  and Onchocerca-infected blackflies .
Overall, the outcome of infection and the potential effects of DENV on mosquito life-history traits must consider at least four interconnected points: (a) virus preparation (titer, serotype and passage history), (b) mosquito population, (c) mode of inoculation and (d) mosquito-virus genome interactions.
Infection of Ae. aegypti with DENV is closely correlated with the amount of virus ingested, and typically, requires as much as 106 or more infectious units per orally administered milliliter to be infective , . Likewise, mosquito survival and longevity also seem to be affected by virus titer, since higher mortality was observed with increasing viral RNA copies . Previous studies examined different virus titers, however these studies used intrathoracic inoculation, a non-natural mode of transmission in which more viral particles are delivered directly to the insect hemocoele through bypassing the midgut infection and/or escape barriers , , . Thus, it is possible that the increase observed in (i) the total time required for feeding  and (ii) the locomotor activity of DENV-infected females  was overestimated due to the mode of DENV transmission to mosquitoes.
Other important factors that can modify the outcome of a dengue infection are the origin and history of mosquito population as well as the passage history of virus strain. The infection rate for dengue virus increased from 52.0% in the F2 generation to 87.3% for the F20 generation of Ae. aegypti formosus collected in Africa . Mosquitoes from the Paea strain had lower survival when infected with DENV-2 in comparison with the field population, a pattern opposite to what was previously observed . Mosquito populations may interact differently with different DENV strains, with the outcome of infection being governed to a large extent by genotype-by-genotype (G×G) interactions between mosquito and virus in genetically diverse natural populations , . The general definition of G×G interactions allows the characterization of phenotypes from macroscopic traits such as lifespan , . Thus, it seems reasonable to speculate that our experimental design, which used (i) sympatric vector-virus populations, with (ii) mosquitoes recently collected from the field, (iii) DENV with low passage history provided (IV) offered to Ae. aegypti by artificial membrane feeders would be the most appropriate form of studying mosquito-virus interactions. Herein, we used a low-passage DENV to orally infect the F1 generation of field-collected mosquitoes, a system that best models and reflects natural transmission .
We have described significant new information regarding the negative impact of DENV-2 infection on Ae. aegypti biology. Overall, we observed that infected mosquitoes need more time to complete a blood meal, showed reduced survivorship, showed reduced egg laying incidence and reduced egg number. Potentially, the fitness cost of DENV infection observed herein might be of limited relevance to natural mosquito population dynamics by at least three reasons. First, the observed reduction on mosquito survivorship, oviposition success and fecundity was highly associated with aging, i.e. mosquitoes must survive at least around 20 days. Before that, fitness cost of DENV infection looks less intense. Second, Ae. aegypti mosquitoes has limited daily survival. For instance, mark-release-recapture experiments estimated the probability of daily survival of Ae. aegypti females as around 0.80–0.85 . Therefore, the number of mosquitoes that would live up to the day when the fitness cost is more intensive would be small. Finally, the number of naturally DENV-infected mosquitoes on natural populations is extremely low. Thus, overall, the cost of DENV infection may have limited influence on natural mosquito population dynamic.
We thank Luciano Moreira, Steve Juliano, Luke A. Baton and Ricardo Lourenço de Oliveira for reading earlier versions of the manuscript, and to Bianca Costa Pimenta and Roberto Peres for technical assistance.
Conceived and designed the experiments: RMF GS. Performed the experiments: MG RMF. Analyzed the data: GS RMF. Contributed reagents/materials/analysis tools: MG RMF. Wrote the paper: GS MG RMF.
- 1. Halstead SB (2008) Dengue-virus mosquito interactions. Annu Rev Entomol 53: 273–291.
- 2. Edman JD, Strickman D, Kittayapong P, Scott TW (1992) Female Aedes aegypti (Diptera: Culicidae) in Thailand rarely feed on sugar. J Med Entomol 29: 1035–1038.
- 3. Braks MAH, Honorio NA, Lourenço-de-Oliveira R, Juliano AS, Lounibos LP (2003) Convergent habitat segregation of Aedes aegypti and Aedes albopictus (Diptera: Culicidae) in southeastern Brazil and Florida. J Med Entomol 40: 785–794.
- 4. Maciel-de-Freitas, Marques WA, Peres RC, Cunha SP, Lourenço-de-Oliveira R (2007) Variation in Aedes aegypti (Diptera: Culicidae) container productivity in a slum and a suburban district in Rio de Janeiro during dry and wet seasons. Mem Inst Oswaldo Cruz 102: 489–496.
- 5. Vezzani D, Carbajo A (2008) Aedes aegypti, Aedes albopictus, and dengue in Argentina: current knowledge and future directions. Mem Inst Oswaldo Cruz 103: 66–74.
- 6. PAHO – Pan-American Health Organization (1994) Dengue and dengue hemorrhagic fever in the Americas: Guidelines for prevention and control, PAHO Scientific Publication n.548, Washington.
- 7. PNCD-Programa Nacional de Controle da Dengue (2002) Ministério da Saúde, Secretaria de Vigilância em Saúde, 34 p.
- 8. Maciel-de-Freitas R, Lourenço-de-Oliveira R (2011) Does targeting key-containers effectively reduce Aedes aegypti population density? Trop Med Int Health 16: 965–973.
- 9. Tun-Lin W, Lenhart A, Nam VS, Rebollar-Téllez E, Morrison AC, et al. (2009) Reducing costs and operational constraints of dengue vector control by targeting productive breeding places: a multi-country non-inferiority cluster randomized trial. Trop Med Int Health 14: 1143–1153.
- 10. Kuno G (1995) Review of the factors modulating dengue transmission. Epidemiol Rev 17: 321–335.
- 11. Massad E, Coutinho FAB (2012) Vectorial capacity, basic reproduction number, force of infection and all that: formal notation to complete and adjust their classical concepts and equations. Mem Inst Oswaldo Cruz 107: 564–567.
- 12. Luz PM, Lima-Camara TN, Bruno RV, Castro MG, Sorgine MHF, et al. (2011) Potential impact of a presumed increase in the biting activity of dengue-virus-infected Aedes aegypti (Diptera: Culicidae) females on virus transmission dynamics. Mem Inst Oswaldo Cruz 106: 755–758.
- 13. Platt KB, Linthicum KJ, Myint KSA, Innis BL, Lerdthusnee K, et al. (1997) Impact of dengue virus infection on feeding behavior of Aedes aegypti. Am J Trop Med Hyg 57: 119–125.
- 14. Luz PM, Codeço CT, Massad E, Struchiner CJ (2003) Uncertainties regarding dengue modeling in Rio de Janeiro, Brazil. Mem Inst Oswaldo Cruz 87: 871–878.
- 15. Anderson RM, May RM (1991) Infectious diseases of humans: dynamics and control. Oxford University Press, UK.
- 16. Maciel-de-Freitas R, Koella JC, Lourenço-de-Oliveira R (2011) Lower survival rate, longevity and fecundity of Aedes aegypti (Diptera: Culicidae) females orally challenged with dengue virus serotype 2. Trans R Soc Trop Med Hyg 105: 452–458.
- 17. Lima-Camara TN, Bruno RV, Luz Pm, Castro MG, Lourenço-de-Oliveira R, et al. (2011) Dengue infection increases the locomotor activity of Aedes aegypti. PLoS ONE 6(3): e17690.
- 18. Takken W, Knols BGJ (1999) Odor-mediated behavior of afro-tropical malária mosquitões. Annu Rev Entomol 44: 131–157.
- 19. Walker ED, Edman JD (1985) The influence of host defensive behavior on mosquito (Diptera: Culicidae) biting persistence. J Med Entomol 22: 370–372.
- 20. Putnam JL, Scott TW (1995) Blood-feeding behavior of dengue-2 virus-infected Aedes aegypti. Am J Trop Med Hyg 52: 225–227.
- 21. Lambrechts L, Scott TW (2009) Mode of transmission and the evolution of arbovirus virulence in mosquito vectors. Proc R Soc B 276: 1369–1378.
- 22. Vazeille-Falcoz M, Mousson L, Rodhain F, Chungue E, Failloux AB (1999) Variation in oral susceptibility to dengue type 2 virus of populations of Aedes aegypti from the islands of Tahiti and Moorea, French Polynesia. Am J Trop Med Hyg 60: 292–299.
- 23. Halstead SB, Marchette NJ (2003) Biologic properties of dengue viroses following serial passage in primary dog kidney cells: studies at the University of Hawaii. Am J Trop Med Hyg 69: 5–11.
- 24. Torrentes-Carvalho A, Azeredo EL, Reis SRI, Miranda AS, Gandini M, et al. (2009) Dengue-2 infection and the induction of apoptosis in human primary monocytes. Mem Inst Oswaldo Cruz 104: 1091–1099.
- 25. Colisson R, Barciu L, Gras C, Raynaud F, Hadj-Slimane R, et al. (2009) Free HTLV-1 induces TLR7-dependent innate immune response and TRAIL relocalization in killer plasmacytoid dendritic cells. Blood 115: 2177–2185.
- 26. Miagostovich MP, Nogueira RMR, Cavalcanti SM, Marzochi KB, Schatzmayr HG (1993) Dengue epidemic in the state of Rio de Janeiro, Brazil: virological and epidemiological aspects. Rev Inst Med Trop São Paulo 35: 149–154.
- 27. Rutledge LC, ard RA, Gould DJ (1964) Studies on the feeding response of mosquitões to nutritive solutions in a new membrane feeder. Mosq News 24: 407–419.
- 28. Harbach RE, Knight KL (2002) Taxonomist`s glossary of mosquito anatomy. Marlton, NJ: Plexus Co.
- 29. Anderson RA, Brust RA (1996) Blood feeding of Aedes aegypti and Culex nigripalpus (Diptera: Culicidae) in relation to defensive behavior of Japanese quail (Coturnis japnocia) in the laboratory. J Vector Ecol 21: 94–104.
- 30. Salazar MI, Richardson JH, Sánchez-Vargas I, Olson KE, Beaty BJ (2007) Dengue virus type 2: replication and tropisms in orally infected Aedes aegypti mosquitoes. BMC Microbiology 7: 9.
- 31. Schmid-Hempel P (2005) Evolutionary ecology of insect immune defenses. Annu Rev Entomol 50 529–551.
- 32. Black 4th WC, Bennet KE, Gorrochótegui-Escalante N, Barillas-Mury CV, Fernández-Salas I, et al (2002) Flavivirus susceptibility in Aedes aegypti. Arch Med Res 33: 379–388.
- 33. Sanchez-Vargas I, Travanty EA, Keene KM, Franz AW, Beaty BJ, et al. (2002) RNA interference, arthropod-borne viroses and mosquitões. Virus Res 102: 65–74.
- 34. Sánchez-Vargas I, Scott JC, Poole-Smith BK, Franz AWE, Barbosa-Solomieu V, et al. (2009) Dengue virus type 2 infections of Aedes aegypti are modulated by the mosquito’s RNA interference pathway. PLoS Pathog 5: e1000299.
- 35. Schwartz A, Koella JC (2004) The cost of immunity in the yellow fever mosquito, Aedes aegypti, depends on immune activation. J Evol Biol 17 834: 840.
- 36. Lambrechts L, Morlais I, Awono-Ambene PH, Cohuet A, Simard F, et al. (2007) Effect of infection by Plasmodium falciparum on the melanization immune response of Anopheles gambiae. Am J Trop Med Hyg 76: 475–480.
- 37. Ciota AT, Styer LM, Meola MA, Kramer LD (2011) The costs of infection and resistance as determinants of West Nile vírus susceptibility in Culex mosquitoes. BMC Ecol 11: 23.
- 38. Joshi V, Mourya DT, Sharma RC (2002) Persistence of dengue-3 virus through transovarial transmission passage in successive generations of Aedes aegypti. Am J Trop Med Hyg 67: 158–161.
- 39. Maciel-de-Freitas R, Codeço CT, Lourenço-de-Oliveira R (2007) Daily survival rates and dispersal of Aedes aegypti females in Rio de Janeiro, Brazil. Am J Trop Med Hyg 76: 659–665.
- 40. Hurd H (2003) Manipulation of medical importante insect vectors by their parasite. Annu Rev Entomol 48: 141–161.
- 41. El Sawaf BM, El Sattar SA, Shehata MG, Lane RP, Morsy TA (1994) Reduced longevity and fecundity in Leishmania infected sandflies. Am J Trop Med Hyg 51: 767–770.
- 42. Christensen BM (1981) Effect of Dirofilaria immitis on the fecundity of Aedes trivittatus. Mosq News 41: 78–81.
- 43. Renshaw M, Hurd H (1994) The effect of Onchocerca infection on the reproductive physiology of the British blackfly, Simulium ornatum. Parasitol 109: 337–345.
- 44. Gubler DJ, Kuno G (1997) Dengue and dengue hemorrhagic fever. Wallinford UK: CAB International.
- 45. Vazeille-Falcoz M, Rosen L, Mousson L, Failloux AB (2003) Low oral receptivity for dengue type 2 viruses of Aedes albopictus from Southeast Asia compared with that of Aedes aegypti. Am J Trop Med Hyg 68: 203–208.
- 46. Chevillon C, Failloux AB (2003) Questions on viral population biology to complete dengue puzzle. Trends Microbiol 11: 415–421.
- 47. Lambrechts L, Chevillon C, Albright RG, Thaisomboonsuk B, Richardson JH, et al.. (2009) Genetic specificity and potential for local adaptation between dengue viroses and mosquito vectors. BMC Evol Biol 9, 160.
- 48. de Roode JC, Altizer S (2009) Host-parasite genetic interactions and virulence-transmission relationships in natural populations of Monarch butterflies. Evolution 64: 502–514.
- 49. Lambrecths L (2010) Dissecting the genetic architecture of host-pathogen specificity. PLoS Pathog 6: e1001019.
- 50. Lambrecths L, Failloux AB (2013) Vector biology prospects in dengue research. Mem Inst Oswaldo Cruz 107: 1080–1082.