SEARCH

SEARCH BY CITATION

Keywords:

  • Aedes aegypti;
  • inter-population mating;
  • insemination;
  • copulation;
  • Australia;
  • laboratory colonization

ABSTRACT:

  1. Top of page
  2. ABSTRACT:
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgments
  8. REFERENCES CITED

Variability between Aedes aegypti populations in north Queensland, Australia, has the potential to impact the successful implementation of new population replacement mosquito releases for dengue control. Four Ae. aegypti colonies originating from different locations (Cairns, Mareeba, Innisfail, and Charters Towers), along with one F1 field-derived population from Cairns, were inter-crossed to determine any incompatibilities in copulation, insemination, and production of viable offspring. Greater copulation and insemination rates were observed when males recently introduced from the wild (‘Cairns-Wild’ population) were mated with long-term laboratory females. Egg viability rates for all crosses ranged from 90.2–98.2%, with no significant differences observed between crosses. Greater egg production was seen in some populations, and when corrected for wing-length, egg production was greatest in a Mareeba x Innisfail cross (19.55 eggs/mm wing length) and lowest for the Charters Towers intra-population cross (14.35 eggs/mm). Additionally, behavioral differences were observed between laboratory and wild mosquitoes from the Cairns location, suggesting possible laboratory conditioning. Finally, despite controlled larval rearing conditions, size differences between populations existed with Charters Towers mosquitoes consistently smaller than the other populations. The spread of genes or bacterial symbionts between these populations is unlikely to be hindered by pre-existing reproductive barriers.


INTRODUCTION

  1. Top of page
  2. ABSTRACT:
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgments
  8. REFERENCES CITED

The mosquito dengue vector Aedes aegypti (L.) has been present in Australia at least since the latter part of the 19th century (Lee et al. 1987), and after being widely distributed throughout eastern and northern Australia, has experienced a range retraction into tropical and sub-tropical Queensland in the northeast of the continent (Russell et al. 2009). This mosquito is abundant in the tropical cities of Townsville and Cairns but is also found in smaller settlements throughout northern and central Queensland (Mottram et al. 2009, Russell et al. 2009). In Australia, it is principally a domestic mosquito, breeding mostly in and around human settlements, utilizing artificial and natural water-filled containers for larval habitat (Lee et al. 1987). A vector of dengue throughout the cosmotropical world, Ae. aegypti is responsible for almost annual outbreaks of this disease in Queensland. In the 2008–2009 wet season, there was an epidemic in north Queensland of over 1,000 locally-acquired cases from all four dengue virus types.

In the absence of effective vaccines, mosquito control remains central to dengue control. With that, pesticide-free technologies are being developed and trialed. In Queensland, Ae. aegypti infected with symbiont Wolbachia pipientis bacteria that confer decreased dengue transmission ability and reduced lifespan have been released in trials in and around the city of Cairns to determine their ability to become established (Hoffmann et al. 2011). This establishment requires some interbreeding with existing wild field populations.

Genetic heterogeneity amongst Ae. aegypti populations in Queensland has been demonstrated (Endersby et al. 2011) with populations in the hotter and drier Charters Towers being distinct from those in the wetter and cooler locations of Cairns and Innisfail. While there is some geographic isolation between Ae. aegypti populations in different locations (see maps in Endersby et al. 2011, Bader and Williams 2012), it is unknown how well such populations interbreed when they do come into contact. This genetic heterogeneity has been considered a potential control mechanism to prevent the unwanted spread of Wolbachia-infected Ae. aegypti (Endersby et al. 2011). However, it is unknown whether the barriers to complete genetic mixing are simply due to partial geographic isolation, or are related to lower rates of mating success should males and females from different locations come together. Mating in Ae. aegypti is achieved through copulation, which occurs in flight, when males chase females. Copulation can be completed when stationary on some surface if the female lands. Males may fly near blood meal hosts (typically humans) and chase females when they approach the host (Hartberg 1971).

Further, it is also unclear whether or not location-specific lineages of Wolbachia-infected mosquitoes need to be created to facilitate effective release into other north Queensland locations. Even partially reduced mating success between populations could have a negative impact on the spread of favorable forms.

In addition, a greater understanding of the success with which Ae. aegypti from different sources copulate and produce offspring is central to the interpretation of laboratory mating studies. Despite the acknowledgement that laboratory colonization has an influence on mosquito behavior and physiology (Lorenz et al. 1984), few descriptions of how mating success changes are documented.

Here we aimed to determine the success rate of inter-population crosses of Ae. aegypti from throughout north Queensland, Australia. Inter-population mating success was determined by measuring copulation and insemination ability, and viability of eggs in the laboratory environment. We also aimed to determine the effects of laboratory colonization on mating success by crosses between laboratory and recently field-derived specimens.

MATERIALS AND METHODS

  1. Top of page
  2. ABSTRACT:
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgments
  8. REFERENCES CITED

Mosquito populations

The populations used in experiments were originally collected from four sites in northern Queensland: Cairns, Charters Towers, Innisfail, and Mareeba and then established in colony at the University of Melbourne in May, 2008 (Richardson et al. 2011). Colonies were then established at the University of South Australia in December, 2009 (Bader and Williams 2012). At the time of experimentation (2010) we estimate that the colonies had experienced approximately 40 laboratory generations.

Each of the colony origin locations is representative of differing climatic conditions found across this region. The locations differ markedly in temperature, rainfall, and humidity; the two extremes include Innisfail with highest rainfall and humidity and lowest average temperatures, while Charters Towers experiences the lowest rainfall and highest temperatures of the four locations (Bader and Williams 2012). In addition, the Charters Towers population was also known to be genetically distinct from the others (Endersby et al. 2009, Endersby et al. 2011).

Preparation of laboratory and field specimens

Aedes aegypti are known to thrive under laboratory conditions, however, as for other insects, their behavior and physiology may be altered as a result of colonization (Lorenz et al. 1984). As such, fresh field specimens were obtained from one location (Cairns) in addition to the laboratory populations in order to compare the mating behaviors of first generation laboratory-reared mosquitoes to those of their long-term laboratory counterparts.

Colonies were maintained at 28–30° C, 70–90% RH, in 12:12 (L:D) photoperiod. Adults were supplied with 15% sucrose in water and fed with human blood at least once a week. Females were provided with damp filter paper on which to lay their eggs. This filter paper was removed and changed weekly. Eggs were then stored in plastic ziplock bags and allowed seven days for embryonation to occur before being hatched.

Eggs from the Cairns field site were hatched and reared in a cage under identical conditions to those of the laboratory mosquitoes in order to obtain first-generation eggs of a field-derived population (hereafter referred to as ‘Cairns-Wild’) so we could be sure of species identification and age. It was these eggs that were used in experimentation.

Eggs from each population were hatched by flooding with rainwater for 24–36 h. First instar larvae were pipetted into trays containing 500 ml of rainwater with the starting density at 50 larvae/tray; dimensions 17.5×12×5 cm. These trays were placed inside temperature controlled cabinets set at 27° C with a photoperiod of LD 12:12 h. Larvae were fed crushed dry cat food (Kitekat Krunch, Mars Petcare, Australia) every second day at a rate of 0.8 mg per larvae (40 mg per tray). This rearing density and feeding rate was previously found to result in Ae. aegypti adult sizes similar to those reported in wild specimens, and growth rates uninhibited by overcrowding (Bader and Williams 2012).

Trays were checked daily for pupae, which were removed and stored individually in cups covered with mesh until emergence to ensure male and female adults were virgins. Once emerged, adults were held in sex-specific cups covered with mesh and cotton wool soaked in sugar water resting on top. A maximum density of ten adults per cup was used.

The above method was used for preparing the adults for both laboratory and first generation field-derived populations for use in both Experiments 1 and 2. A number of inter-population crosses were performed to determine inter-population mating success, although time and resource constraints meant that not all possible crosses were performed (Table 1).

Table 1.  Details of inter-population crosses of Aedes aegypti from Queensland, Australia. ‘Cairns-Wild’ was a first generation field-derived colony. ‘X’ indicates crosses that were performed. Thumbnail image of

Experiment 1: copulation and insemination

Five males and five females were selected for each copulation experiment (five to seven days post-emergence). Adults were transferred from the storage cup into the mating observation chamber; a transparent 10×10×10 cm airtight container (25–27° C, 70–80% RH). The adult mosquitoes had no access to fluids during the experiment. Due to the diurnal activity pattern of this species, all experiments were performed after 15:00 to coincide with a peak in flight activity (Jones 1981).

Observations of mating behaviors were carried out for a period of 1 h beginning as soon as the females were inside the container. Whenever a male and female encountered one another and there was clear physical contact between the distal portions of the abdomen, this event was counted as a successful copulation. It was not always possible to accurately distinguish between ‘copulation’ or ‘coupling’ (Gwadz et al. 1971). However, as all experiments were witnessed by the same observer, any error in judging coupling vs copulation would have been systematic.

Copulations observed in the 1-h period were tallied for both long (10–20 s) and very long (>20 s) coupling events. Coupling events of <10 s duration were not considered likely to lead to insemination and are not reported here. Observations of flight activity, rest periods, and male swarming were also recorded. Upon completion, the container was stored at −18° C to freeze the mosquitoes. Females were then dissected and their spermathecae removed and checked for the presence of sperm. Both the male and female adult mosquitoes also had one wing removed and measured using microscopy and a digital camera with morphology analysis software (NIS-Elements Br 3.1, Nikon Corporation, Tokyo, Japan). Wing length was used as a proxy for body size (Nasci 1990). Each cross experiment was repeated five times.

Experiment 2 – Fecundity and egg viability

Mating and egg laying

Five males and five females from each population (five to seven days post-emergence) were used to assess the fecundity and egg viability of various crosses, each of which was replicated five times. These adult mosquitoes were transferred together into a sealed cage (30×29×30 cm, 27° C, 70–80% RH) and allowed 24 h to mate. The mosquitoes had access to sugar water during this period.

After 24 h, female mosquitoes were transferred into individual 100 ml specimen jars with cotton wool saturated with rainwater in the base covered with a strip of wet filter paper to provide an oviposition site. The females were then blood-fed to repletion on the finger of AJR. Female reproductive success, namely the number and viability of the eggs produced, can be influenced by the size and quality of the blood meal (Spielman et al. 1969). As such, to ensure the maximum egg output from all females, each female was given enough time to feed until their abdomen was considered fully distended (Spielman et al. 1969). Males were removed and placed at −18° C for subsequent wing measurement (as in Experiment 1). Again, each cross was repeated five times.

Measuring fecundity and egg viability

Individual vials were checked daily, recording the first day eggs were observed. After ten days, all females were removed and frozen, whether they laid eggs or not. Wing lengths were measured as in Experiment 1. The filter paper holding the eggs was removed, the eggs counted and then stored in individual plastic zip lock bags for a period of at least seven days to allow for embryonation.

Eggs were then flooded with rainwater containing larval food for 24 h to stimulate hatching. The filter paper was then removed and unhatched eggs counted to determine percentage hatched. It is common for Aedes eggs to hatch in installments (Gillett 1955), meaning that some eggs may hatch on the second or third inundation. However, time constraints prevented actioning this and so unhatched eggs were instead bleached to determine viability (Bader and Williams 2011). Total percentage viability (no. hatched + no. unhatched viable / total *100) was then calculated.

Data analysis

Data from experiments one and two were analyzed to determine whether any differences in copulation and insemination frequency, and egg production and viability rates, were present between the twelve population crosses. Initially, we used a Generalized Linear Model (GLM) (STATA 11.0, College Station TX, U.S.A.) to determine the relative influence of wing length and strain of origin on total copulations observed per h. This was done for each experiment of five females with five males (n = 60 experiments), with average wing lengths used. When data were normally distributed, one-way analysis of variance (ANOVA) followed by Tukey's post-hoc analysis was performed in Minitab (Ver. 16, State College, PA, U.S.A.). When data were not normally distributed, non-parametric analyses (Kruskal-Wallis) were carried out instead. T-tests were also performed using this software.

In order to check for differences in insemination rates and egg production between inter- and intra-population crosses, contingency tables were constructed and analyzed to obtain a Chi Square statistic.

Observational results relating to mating behaviors from both experiments were also analyzed to determine whether similar patterns of behavior existed across the twelve population crosses. These included overall flight activity, timing of flight activity, and responsiveness to the opposite gender.

RESULTS

  1. Top of page
  2. ABSTRACT:
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgments
  8. REFERENCES CITED

Experiment 1 – Copulation and insemination

The GLM revealed that increasing female wing length (P= 0.021) and male strain origin (P= 0.001) were both significant predictors of the number of copulations. Male wing length (P= 0.760), female strain (P= 0.110), and encounter with a different strain (P= 0.629) were not significant factors in copulation rate.

All references to crosses in this section are written as [Population]sex. ANOVA revealed significant differences in total copulations per h (F= 2.07, df = 11, P= 0.041) but only for two crosses: Cairnsf x Cairns-Wildm (22.0) having significantly more mean copulations per h than Cairnsf x Mareebam (6.0) (Figure 1). The two crosses with the highest average total copulations (Cairnsf x Cairns-Wildm and Cairns-Wildf x Cairns-Wildm) also had the largest variation in their data (Figure 1). In short, Cairns-Wild males were involved in significantly more copulations than Mareeba males mating with the Cairns laboratory strain. Furthermore, all crosses exhibited long (10–20 s) and very long copulations (> 20 s), in approximately equal measure (Table 2). However, crosses involving Cairns-Wild males were notable for the large number of long copulations (Table 2).

image

Figure 1. Average total copulations (± SEM) for intra- and inter-population crosses of Aedes aegypti ranked from highest to lowest. Crosses with different letters were significantly different by ANOVA and Tukey's post-hoc multiple comparisons. C = Cairns, M = Mareeba, I = Innisfail, CT = Charters Towers, CW = Cairns-Wild.

Download figure to PowerPoint

Table 2.  Average copulations (± SEM) per h for each intra- and inter-population cross for different durations measured. C = Cairns, M = Mareeba, I = Innisfail, CT = Charters Towers, CW = Cairns-Wild.
Population cross (female/male)Long (10–20 s)Very long (> 20 s)
C × CW10.4 ± 5.45.4 ± 6.2
CW × CW10.2 ± 9.74.6 ± 3.0
I × CT6.4 ± 2.66.2 ± 3.2
CT × I6.6 ± 1.74.6 ± 2.9
CT × CT4.8 ± 1.85.8 ± 3.7
M × I7.0 ± 3.33.4 ± 1.3
M × M4.0 ± 2.36.4 ± 5.1
CW × C6.8 ± 1.93.4 ± 2.9
M × C4.2 ± 1.96.0 ± 2.4
I × M6.0 ± 1.62.4 ± 1.9
C × C2.4 ± 1.15.0 ± 1.4
C × M1.8 ± 0.43.2 ± 0.8

All population combinations were trialled in both directions (i.e., Population1f x Population2m and Population2f x Population1m), as the possibility of a more favorable direction for inter-population crossing for insemination success was considered. Female insemination success rates from the five replicates for each population cross were combined (n= 25) and then split into total number of females successfully inseminated and total not inseminated. Of the inter-population crosses examined, no cross direction was found to be significantly different in terms of insemination success compared to its reverse (χ2 value range 0.222 −3.03, P value range 0.187 – 0.637).

When the inter-population crosses were compared against their intra-population cross counterparts, only one combination resulted in significant differences being observed. The Cairnsf x Cairns-Wildm inter-population cross (24 inseminated, one not inseminated) was found to have a significantly greater chance of achieving insemination compared to its intra-population counterpart, Cairns-Wildf x Cairns-Wildm (18:7) (χ2= 5.357, P= 0.021).

Male behavior – differences between laboratory and wild specimens

Males from laboratory populations tended to wait for females to take flight before pursuing them, rather than fly continuously. They were very quick to respond to a female taking flight and would follow her to initiate copulation. If the copulation attempt was unsuccessful, most often they would remain in flight for a very short time before landing again. Conversely, Cairns-Wild males displayed relatively more flight during observation periods, with one to two males commonly in flight at most times. There were short periods where all five males would be resting at once, however these were short lived and only happened occasionally. When individual Cairns-Wild males did land to rest, it tended to be brief (1–2 min) before taking flight again. Male flight was also stimulated by a female flying. Once the female had landed, the Cairns-Wild males were often seen to continue flying for many minutes before resting again. Multiple Cairns-Wild males would often try to mate with the same female. This most often resulted in an unsuccessful coupling and both would fly away.

Female behavior – differences between laboratory and wild specimens

In all populations, female flight most often appeared spontaneous, although they were sometimes stimulated into flight by males flying near to them if they had not yet been mated. Post-copulation this behavior changed with females remaining motionless even when males would bump into them. Female flight always stimulated males into flight, even if only for a short period.

Some females demonstrated strong refractory behaviors following successful copulation, while others appeared receptive to numerous copulations. Unmated Cairns-Wild females exhibited refractory behaviors prior to mating on numerous occasions. Refractory behaviors prior to mating were uncommon in laboratory females.

Experiment 2 – Fecundity and egg viability

Percentage viability

All combinations yielded progeny with at least 90% average viability (Figure 2); however, the variation within each group differed between different crosses. No statistically significant differences were found between the percentage viability of each of the intra- and inter-population crosses using Kruskal Wallis tests (H= 10.60, df = 11, P= 0.478). Also, male origin did not significantly influence the percentage of viable offspring produced by the females (H= 0.81, df = 11, P= 0.368).

image

Figure 2. Percentage offspring viability (eggs that hatched or unhatched eggs containing viable embryos) (± SEM) for intra- and inter-population crosses of Aedes aegypti, Queensland Australia. C = Cairns, M = Mareeba, I = Innisfail, CT = Charters Towers, CW = Cairns-Wild.

Download figure to PowerPoint

Egg production

One-way ANOVA was performed on the total number of eggs laid per female to compare each of the intra- and inter-population crosses followed by Tukey's post-hoc analysis. Values from all replicates of each cross were combined with n ranging from 13–25 due to instances where some females did not lay any eggs.

Significant differences (P= 0.004) were found between the average total of eggs laid by females from Innisfailf x Mareebam (52.79 per female) compared to Cairnsf x Cairnsm (38.92) and Charters Towersf x Charters Towersm (36.57). Innisfail females laid significantly more eggs on average than females from the two later combinations. Finally, crosses involving Cairns-Wild females had the most variation in the total eggs laid compared to all laboratory females.

However, we expected that female size may affect the number of eggs produced. To control for different female body size and thereby isolate the influence of population crossing, the data were transformed into eggs laid/mm wing length. One way ANOVA was then performed followed by Tukey's post-hoc analysis, returning significant results (F= 2.11, df = 11, P= 0.021). It was found that Charters Towersf x Charters Towersm had significantly fewer eggs per mm wing length (14.35) than Innisfailf x Mareebam (19.55) (Figure 3). There were no further significant differences in eggs per mm wing length between the remaining crosses, although the three laboratory colony intra-population crosses were among the bottom four when ranked in descending order.

image

Figure 3. Average number of eggs laid per mm wing length (± SEM) for each Aedes aegypti intra- and inter-population cross. C = Cairns, M = Mareeba, I = Innisfail, CT = Charters Towers, CW = Cairns-Wild.

Download figure to PowerPoint

Generally, laboratory inter-population crosses appear to have more eggs per mm wing length than intra-population crosses (Figure 3) (excluding Cairns-Wildf x Cairns-Wildm). When data from all inter- and intra-population crosses (excluding Cairns-Wildf x Cairns-Wildm) were pooled into two groups and subjected to a two-sample independent T-test, inter-population crosses were indeed found to have significantly more eggs laid per mm wing length than the intra-population crosses (T=−2.98, df = 206, P= 0.002). Even when the Cairns-Wild intra-population cross data were included, the difference was still significant (T=−1.77, df = 225, P= 0.039).

Egg-laying success (number of females laying eggs)

When inter-population cross direction was examined, only the Cairns and Cairns-Wild crosses yielded significant results (P= 0.031), with Cairnsf x Cairns-Wildm leading to greater egg-laying success than the cross in the opposite direction (Table 3). All other crosses appeared to have an equal chance of laying eggs in both directions.

Table 3.  Chi Square analysis inter-population cross direction and chance of laying eggs. C = Cairns, M = Mareeba, I = Innisfail, CT = Charters Towers, CW = Cairns-Wild.
Population combination (female × male)Females (n = 16–25*) eggs laid : notχ2 P value
  1. *Total number of females per cross varied due to instances where females did not survive the experimental period. These females never laid eggs (but were inseminated) and were removed from the analyses. # Significant if P < 0.05.

C × CW / CW × C23:0 / 13:34.6710.031#
I × CT / CT × I18:3 / 18:40.1200.243
I × M / M × I19:4 / 17:81.3630.243
M × C / C × M20:3 / 20:40.1210.727

When egg laying from inter-population crosses was compared against their intra-population counterparts, a number of significant directional relationships were observed (Table 4), albeit with no clear advantage apparent for inter- and intra-population crosses.

Table 4.  Chi Square analysis intra- and inter-population cross direction and chance of laying eggs. C = Cairns, M = Mareeba, I = Innisfail, CT = Charters Towers, CW = Cairns-Wild.
Population combination (female × male)Females (n = 16–25*) eggs laid : notχ2 P value
  1. *Total number of females per cross varied due to instances where females did not survive the experimental period. These females never laid eggs (but were inseminated) and were removed from the analyses. #Significant if P < 0.05.

C × C / C × CW25:0 / 23:001.000
C × C / CW × C25:0 / 13:35.0570.025#
CT × CT / CT × I14:9 / 18:33.4160.065
CT × CT / I × CT14:9 / 18:42.4020.121
M × M / M × I22:0 / 17:88.4840.004#
M × M / I × M22:0 / 19:44.1990.040#
C × C / C × M25:0 / 20:33.4780.062
C × C / M × C25:0 / 20:44.5370.032#
M × M / C × M22:0 / 20:33.0750.080
M × M / M × C22:0 / 20:44.0160.045#
CW × CW / CW × C19:2 / 13:30.6610.416
CW × CW / C × CW19:2 / 23:02.2950.130
Size differences between populations

Analysis of wing lengths of each population (determined from measurements of all individuals used in experimentation) was performed using a one-way ANOVA followed by Tukey's post-hoc analyses. Only Charters Towers females were significantly smaller compared to the others (F= 4.73, df = 4, P= 0.001) (Table 5). The total difference in female mean wing length between the largest and smallest was 0.079 mm.

Table 5.  Grouping of female and male wing length measures following ANOVA and Tukey's post-hoc method. C = Cairns, M = Mareeba, I = Innisfail, CT = Charters Towers, CW = Cairns-Wild. Means that do not share a letter are significantly different.
Sex of PopulationPopulationnMean wing length (mm)Grouping
FemaleCW942.6468A  
 M1422.6405A  
 I922.6308A  
 C1472.6276A  
 CT932.5676 B 
       
MaleCW992.0656A  
 M1442.0447AB 
 C1492.0370AB 
 I1002.0213 BC
 CT991.9915  C

Male mosquitoes were also found to have significant differences in size between populations (F= 6.23, df = 4, P < 0.001), with Cairns-Wild males having significantly larger wing length compared to both Innisfail and Charters Towers males. In addition, Charters Towers males had significantly smaller mean wing lengths compared to other groups. The total difference in male mean wing length between the largest and smallest was 0.074 mm.

DISCUSSION

  1. Top of page
  2. ABSTRACT:
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgments
  8. REFERENCES CITED

When a number of laboratory colonies of Ae. aegypti of slightly different geographic origin within Australia were crossed, there appeared to be no physiological or behavior barriers to successful reproduction. Copulation, insemination, egg production, and egg viability rates were all high for both intra- and inter-population crosses. While high reproductive success rates were expected, we also anticipated that some crosses would yield notably reduced success that could assist in the generation of genetic heterogeneities in the wild. We did not observe any reduced mating success worthy of mention. One interpretation of these findings is that any genetic heterogeneities observed between Ae. aegypti populations in the wild (Endersby et al. 2011), are the result of geographic isolation and locally-acting adaptive selective processes that have little or no effect on reproductive physiology.

There are implications of this apparent ease of reproduction between strains of different geographic origin. Firstly, with the advent of new mosquito control technologies involving the release of transgenic and/or symbiont-carrying mosquitoes (e.g., the release of Wolbachia-infected Ae. aegypti in Queensland in 2011, Hoffmann et al. 2011), from the data we present here there seems to be no evidence to support the development of geo-specific strains for release into particular areas. Admittedly, while we have not tested the Wolbachia-infected strain used in those releases, it is likely that a single release strain should be able to mate with a variety of populations in Queensland at a high rate of success that should not impede gene flow.

However, the corollary of such ease of reproduction is that any containment of transgenics or symbiont infection after field release will not likely be assisted by reduced reproductive success of crosses. Containment will be reliant upon maintaining geographic isolation.

A significant limitation of our study is that apart from one recently derived wild strain (so-called Cairns-Wild), we studied only well established laboratory colonies. The value of studying laboratory colonies of mosquitoes, and just how representative they are of their wild counterparts, has been discussed repeatedly (Craig et al. 1961, Benedict et al. 2009). Colonies of Ae. aegypti maintain genetic heterozygosity for over 50 generations and allelic differences between colonies of different geographic origin have been described (Craig et al. 1961). Yet the process of colonization can have demonstrable impacts on physiology (Lorenz et al. 1984), so the preference at all times would be to study field material or at least recently-derived material from the wild. Hence, our findings of no significant decline in reproductive success between population crosses should be tempered by the fact that we used laboratory colonies.

Further, our measures of insemination success were obtained in a highly contrived and controlled laboratory environment. It is possible that the close proximity between males and females in our experiments led to higher levels of insemination than might be expected in the field and could have masked behavioral differences between strains that may only become apparent in flight behaviors occurring at larger scales around hosts (Hartberg 1971). The process of laboratory colonization may have selected inadvertedly for particular mating behaviors.

Indeed, we were able to demonstrate some differences between the laboratory and recently derived Cairns-Wild strain. Cairns-Wild males had greater copulation and insemination rates with laboratory females than their long-term laboratory counterparts. Crosses with Cairns-Wild males demonstrated greater egg-laying than when Cairns-Wild females mated with laboratory males. Cairns-Wild males were also significantly larger than other groups reared under the same conditions. These findings indicate that male size and behavior may change in colonization and that this may impact on reproductive vigor in the laboratory. These findings underscore the importance of ensuring that laboratory-derived transgenic males released for population suppression are able to effectively mate with wild females. Male mating competitiveness was assessed prior to the release of a transgenic Ae. aegypti in a large cage trial, in which investigators aimed to safeguard male fitness in their transgenic strain through the introduction of material from ten different sites (Wise de Valdez et al. 2011). However, we report here that time in colonization may be a more important determinant of male mating vigor than genetic diversity and is worthy of consideration by those planning future transgenic releases of Ae. aegypti.

There did appear to be enhanced vigor in inter-population crosses compared with intra-population ones. Inter-population crosses resulted in greater egg production per mm wing-length, lending support to the concept of maintaining vigor in laboratory colonies of Ae. aegypti through regular introduction of fresh field-collected material. The importance of ensuring that male mosquitoes laboratory-reared for release into the wild would be competitive in mating interactions has been recently highlighted, with calls to explicitly study the role laboratory heterozygosity plays in maintaining male reproductive fitness (Benedict et al. 2009). Although we have not studied male mating competitiveness here, we do provide novel evidence of differences between the mating success of long-term laboratory Ae. aegypti males compared with wild males.

Acknowledgments

  1. Top of page
  2. ABSTRACT:
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgments
  8. REFERENCES CITED

We thank Michael Kearney (University of Melbourne) for comments on an earlier manuscript and Kellie Richardson for supplying mosquito eggs from the University of Melbourne. Christie Bader and Gina Rau both provided technical assistance. Scott Ritchie, Petrina Johnson, and Luke Rapley (James Cook University) are all thanked for their collegiality and central role in initial specimen collection from locations in Queensland. No external funding was received to conduct this work, which was carried out as part of AJR's undergraduate Honours research program.

REFERENCES CITED

  1. Top of page
  2. ABSTRACT:
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgments
  8. REFERENCES CITED
  • Bader, C.A. and C.R. Williams. 2011. Eggs of the Australian saltmarsh mosquito, Aedes camptorhynchus, survive for long periods and hatch in instalments: implications for biosecurity in New Zealand. Med. Vet. Entomol. 25: 7076.
  • Bader, C.A. and C.R. Williams. 2012. Mating, ovariole number and sperm production of the dengue vector mosquito Aedes aegypti (L.) in Australia: broad thermal optima provide the capacity for survival in a changing climate. Physiol. Entomol. 37: 136144.
  • Benedict, M.Q., B.G.J. Knols, H.C. Bossin, P.I. Howell, E. Mialhe, C. Caceres, and A.S. Robinson. 2009. Colonisation and mass rearing: learning from others. Malaria J. 8 (Suppl 2): S4.
  • Craig, G.B., Jr., R.C. Vandehey, and W.A. Hickey. 1961. Genetic variability in populations of Aedes aegypti. Bull. Wld. Hlth. Org. 24: 527539.
  • Endersby, N.M., A.A. Hoffmann, V.L. White, S. Lowenstein, S. Ritchie, P.H. Johnson, L.P. Rapley, P.A. Ryan, V.S. Nam, N.T. Yen, P. Kittiyapong, and A.R. Weeks. 2009. Genetic structure of Aedes aegypti in Australia and Vietnam revealed by microsatellite and exon primed intron crossing markers suggests feasibility of local control options. J. Med. Entomol. 46: 10741083.
  • Endersby, N.M., A.A. Hoffmann, V.L. White, S.A. Ritchie, P.H. Johnson, and A.R. Weeks. 2011. Changes in the genetic structure of Aedes aegypti (Diptera: Culicidae) populations in Queensland, Australia, across two seasons: Implications for potential mosquito releases. J. Med. Entomol. 48: 9991007.
  • Gillett, J.D. 1955. Variation in the hatching-response of Aedes eggs (Diptera: Culicidae). Bull. Entomol. Res. 46: 241254.
  • Gwadz, R.W., G.B. Craig Jr., and W.A. Hickey. 1971. Female sexual behavior as the mechanism rendering Aedes aegypti refractory to insemination. Biol. Bull. 140: 201214.
  • Hartberg, W.K. 1971. Observations on the mating behaviour of Aedes aegypti in nature. Bull. Wld. Hlth. Org. 45: 847850.
  • Hoffmann, A.A., B.L. Montgomery, J. Popovici, I. Iturbe-Ormaetxe, P.H. Johnson, F. Muzzi, M. Greenfield, M. Durkan, Y.S. Leong, Y. Dong, H. Cook, J. Axford, A.G. Callahan, N. Kenny, C. Omodei, E.A. McGraw, P.A. Ryan, S.A. Ritchie, M. Turelli, and S.L. O'Neill. 2011. Successful establishment of Wolbachia in Aedes populations to suppress dengue transmission. Nature 476: 454457.
  • Jones, M.D.R. 1981. The programming of circadian flight-activity in relation to mating and the gonotrophic cycle in the mosquito, Aedes aegypti. Physiol. Entomol. 6: 307313.
  • Lee, D.J., M.M. Hicks, M. Griffiths, M.L. Debenham, J.H. Bryan, R.C. Russell, M. Geary, and E.N. Marks. 1987. The Culicidae of the Australasian region. Vol. 4, Entomology Monograph No. 2. Australian Government Publishing Services, Canberra .
  • Lorenz, L., B.J. Beaty, T.H. Aitken, G.P. Wallis, and W.J. Tabachnick. 1984. The effect of colonization upon Aedes aegypti susceptibility to oral infection with yellow fever virus. Am. J. Trop. Med. Hyg. 33: 690694.
  • Mottram, P., D. Gould, and P. Florian. 2009. Aedes aegypti distribution in southern Queensland 1990–2009, are they really marching toward Brisbane? Mosq. Bites Asia-Pac. Reg. 4: 4550.
  • Nasci, R. 1990. Relationship of wing length to adult dry weight in several mosquito species (Diptera: Culicidae). J. Med. Entomol. 27: 716719.
  • Richardson, K., A.A. Hoffmann, P. Johnson, S. Ritchie, and M.R. Kearney. 2011. Thermal sensitivity of Aedes aegypti from Australia: empirical data and prediction of effects on distribution. J. Med. Entomol. 48: 914923.
  • Russell, R.C., B.J. Currie, M.D. Lindsay, J.S. Mackenzie, S.A. Ritchie, and P.I. Whelan. 2009. Dengue and climate change in Australia: predictions for the future should incorporate knowledge from the past. Med. J. Aust. 190: 265268.
  • Spielman, A., M.G. Leahy Sr., and V. Skaff. 1969. Failure of effective insemination of young female Aedes aegypti mosquitoes. J. Insect Physiol. 15: 14711479.
  • Wise de Valdez, M.R., Nimmo, D., Betz, J., Gong, H., James, A.A., Alphey, L., and W.C. Black IV. 2011. Genetic elimination of dengue vector mosquitoes. Proc. Natl. Acad. Sci. USA. 108: 47724775.