Pone.0062711 1.7

Fitness of Transgenic Mosquito Aedes aegypti MalesCarrying a Dominant Lethal Genetic System Blandine Massonnet-Bruneel1*, Nicole Corre-Catelin1, Renaud Lacroix1,2, Rosemary S. Lees3¤, Kim Phuc Hoang3, Derric Nimmo2, Luke Alphey2,3, Paul Reiter1* 1 Unite´ Insectes et Maladies Infectieuses, Institut Pasteur, Paris, France, 2 Oxitec Ltd, Oxford, United Kingdom, 3 Department of Zoology, University of Oxford, Oxford, OX513A is a transgenic strain of Aedes aegypti engineered to carry a dominant, non-sex-specific, late-acting lethal geneticsystem that is repressed in the presence of tetracycline. It was designed for use in a sterile-insect (SIT) pest control systemcalled RIDLH (Release of Insects carrying a Dominant Lethal gene) by which transgenic males are released in the field tomate with wild females; in the absence of tetracycline, the progeny from such matings will not survive. We investigated themating fitness of OX513A in the laboratory. Male OX513A were as effective as Rockefeller (ROCK) males at inducingrefractoriness to further mating in wild type females and there was no reduction in their ability to inseminate multiplefemales. They had a lower mating success but yielded more progeny than the wild-type comparator strain (ROCK) when onemale of each strain was caged with a ROCK female. Mating success and fertility of groups of 10 males—with different ratiosof RIDL to ROCK—competing for five ROCK females was similar, but the median longevity of RIDL males was somewhat(18%) lower. We conclude that the fitness under laboratory conditions of OX513A males carrying a tetracycline repressiblelethal gene is comparable to that of males of the wild-type comparator strain.
Citation: Massonnet-Bruneel B, Corre-Catelin N, Lacroix R, Lees RS, Hoang KP, et al. (2013) Fitness of Transgenic Mosquito Aedes aegypti Males Carrying aDominant Lethal Genetic System. PLoS ONE 8(5): e62711. doi:10.1371/journal.pone.0062711 Editor: Kristin Michel, Kansas State University, United States of America Received November 2, 2012; Accepted March 22, 2013; Published May 14, 2013 Copyright: ß 2013 Massonnet-Bruneel et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, whichpermits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Funding: RL, RSL, HKP, DN, and LA are or have been employees or students of Oxitec Ltd and/or the University of Oxford. Oxitec and the University of Oxfordhold intellectual property relating to the subject matter of this paper. This work was supported by Institut Pasteur (Paris) and funded in part by a grant to theRegents of the University of California from the Foundation for the National Institutes of Health through the Grand Challenges in Global Health initiative. Thisstudy was partially funded by EU grant FP7-261504 EDENext and is catalogued by the EDENext Steering Committee as EDENext112 (http://www.edenext.eu). Thecontents of this publication are the sole responsibility of the authors and don't necessarily reflect the views of the European Commission. All other authors havedeclared that no competing interests exist. This does not alter the authors' adherence to all the PLOS ONE policies on sharing data and materials.
Competing Interests: RL, RSL, HKP, DN, and LA are or have been employees or students of Oxitec Ltd and/or the University of Oxford. Oxitec and the Universityof Oxford hold intellectual property relating to the subject matter of this paper. All other authors have declared that no competing interests exist. This does notalter the authors' adherence to all the PLOS ONE policies on sharing data and materials.
* E-mail: [email protected] (BMB); [email protected] (PR) ¤ Current address: Insect Pest Control Laboratory, Joint FAO/IAEA Division of Nuclear Techniques in Food and Agriculture, International Atomic Energy Agency,Vienna, Austria OX513A is a transgenic strain of Aedes aegypti engineered to carry a dominant, repressible, non-sex-specific, late-acting lethal Dengue is the most important arbovirus transmitted by genetic system, together with an Act5C-DsRed2 fluorescent mosquitoes; 2.5 billion people live in areas at risk of epidemic marker [17]. It is intended for use in a sterile-insect pest control transmission [1,2]. The principal urban vector of dengue, yellow system called RIDLH (Release of Insects carrying a Dominant fever and chikungunya is Aedes aegypti. No vaccines are available for Lethal gene or genetic system) [18]. Without tetracycline, larvae dengue or chikungunya, so mosquito control is the only option for carrying one or more copies of the OX513A insertion develop reducing transmission. In recent decades, however, conventional normally but die at pupation. This late-acting lethality has methods of control have proven insufficiently effective [1–3] so theoretical advantages over the early-acting lethality characteristic there is an urgent need for new and innovative strategies.
of other sterilisation methods (e.g. radiation, chemicals, Wolbachia- Transgenic insects are receiving increasing attention for the induced cytoplasmic incompatibility), at least if there are density- control of mosquito-borne diseases [4,5]. Two broad classes of dependent effects before the late-lethal phase [17,19]. If reared in strategy have been proposed [6]: (i) population reduction, for the presence of tetracycline (e.g. 30 mg/ml), the lethal gene is example by variants of Sterile Insect Technique (SIT) [7–11] and; repressed; tetracycline therefore acts as an ‘antidote' or repressor (ii) replacement of the wild population by insects that are of the lethal system to allow the RIDL strain to be reared under refractory to pathogens [12,13]. It is now feasible to create defined conditions. The proposed strategy [9,11,20,21] is to mass- transgenic strains using transposons, fluorescent proteins and rear homozygous RIDL Ae. aegypti mosquitoes and release males to tissue- or stage-specific promoters [14,15], and several species of mate with wild females in the field. Each egg fertilised by a RIDL culicine and anopheline mosquitoes have been transformed male carries the transgene and therefore dies; the RIDL males are therefore effectively sterile [11]. If females only mate once, as is PLOS ONE www.plosone.org May 2013 Volume 8 Issue 5 e62711 Fitness of Transgenic RIDL Aedes aegypti Males generally assumed for Ae. aegypti, then the female's entirereproductive output is destroyed by mating to a RIDL male andshe is herself therefore effectively sterilised. However, femalemonogamy is not a requirement of the approach, though wheremultiple mating is common, post-copulatory effects such as spermcompetition are also relevant.
The success of any SIT-like vector control strategy depends on the performance of the organisms released. Assessing parametersof fitness in the laboratory is the first and necessary step beforeperforming any field releases. Marrelli [22] reviewed three studiesof fitness in transgenic mosquitoes. Briefly, in cage experiments,Catteruccia [23] showed that the transgenic allele frequencydecreased through time, when introduced into mixed cages oftransgenics and wild-type insects, in four strains of Anophelesstephensi expressing the enhanced green fluorescent protein (EGFP)or DsRed. Irvin [24] found that estimates of survivorship,longevity and fecundity for three strains of Ae. aegypti homozygousfor transposase genes and EGFP were lower than for the wild Figure 1. Egg production per female with different proportions strain. However, Marcelo Jacobs-Lorena's lab reported that while of ROCK and RIDL males. For each ratio of strains (ROCK/RIDL): mean a set of transgenic An. stephensi lines expressing a bee venom number of eggs with Standard Errors (6SE). Above the figure, values (n) component had significant fitness problems, another set expressing indicate the no. of females laying eggs. There is a significant differencein the mean number of eggs across the five ratio of strains, bars with a synthetic peptide did not [25]. Later work showed that such lines different letters are significantly different (Generalized linear model: could even have a net fitness advantage in certain circumstances, p = 0.00561).
albeit highly artificial ones [26,27]. The lower fitness observed for homozygous transgenic mosquitoes in some studies could either bedue to (i) insertional mutagenesis and/or negative effects of eggs than the other ratios (Figure 1; GLM, ROCK control: transgene products or (ii) inbreeding and the harmful effects of p = 0.00561, other treatments: p.0.05).
homozygous recessive genes [22]. The study of Moreira [25] wasdesigned to distinguish between these two hypotheses and their Mortality and emergence results suggested that transgenesis is not always deleterious if In the 10 =/5 R experiment, mean mortality of offspring for all inbreeding is minimised; Allen et al [28] reached a similar transgenic crosses (n = 83) was 85.5%, i.e. mean adult emergence conclusion for transgenic Cochliomyia hominivorax.
was 14.5% (Table 2): of the heterozygous RIDL adults collected, We report on the first laboratory studies of selected fitness 164/496 (33%) were females and 332/496 (67%) were males parameters for homozygous Ae. aegypti RIDL males viz: (i) mating (Table 2). Overall emergence was the highest for the ROCK competitiveness between RIDL and ROCK males for ROCK control (98.1%) and the lowest for the RIDL control (15%) and, as females, (ii) insemination rate and (iii) adult male longevity. The expected, decreased as the proportion of RIDL males increased OX513A was originally generated in a ROCK background, thus making this the most suitable wild type comparator strain forassessing the impact of transgenesis on fitness. We also assessed the Mating competitiveness lethality of the RIDL construct in heterozygous RIDL/wild type In the 2 =/1 R experiment, there was a significant deviation progeny reared without tetracycline, i.e. the progeny of wild type from expectation in the observed frequencies of transgenic and females mated with RIDL males. We discuss implications for the non-transgenic matings (Table 1; x2 = 6.75, df = 1, p = 0.009): of suppression of Ae. aegypti populations in the field.
the 48 such matings, 15 were transgenic (31%) and 33 were nontransgenic (69%).
In the 10 =/5 R experiment, there were no significant differences between expected and observed transgenic vs. non- Fertilisation and oviposition transgenic mating for the different proportions of ROCK/RIDL In the 2 =/1 R experiment, all females (n = 146) took a blood males (x2 test: 8 ROCK/2 RIDL: x2 = 3.18, df = 1, p = 0.075, meal and 137 (93.8%) laid eggs. Of the 9 that failed to lay eggs, 3 n = 34; 5 ROCK/5 RIDL: x2 = 0.9, df = 1, p = 0.343, n = 40 and had sperm in 2 of their three spermathecae. Thus, 140/146 2 ROCK/8 RIDL: x2 = 0.012, df = 1, p = 0.912, n = 27). In both (95.8%) of females had been inseminated successfully. In the 10 =/ mating experiments (n = 336), there was only one case (0.30%) in 5 R experiment, all females (n = 236) took a blood meal and 199 which both fluorescent and non-fluorescent progeny were (84.3%) laid eggs. Of the 37 females that failed to lay eggs, 1 observed, which presumably represents a female mating both a (2.7%) (a ROCK control) had no sperm, 32 (86.3%) had sperm in RIDL and a ROCK male.
two spermathecae, and 4 (11%) had sperm in all three.
In the 2 =/1 R experiment, there were more eggs laid in Insemination rate transgenic crosses than in non- transgenic crosses (Table 1; Mann- When contact was limited to 24 hours, there were no significant Witney test, U = 1863, p = 0.043). In the 10 =/5 R experiment, differences between the number of females fertilized by RIDL there was no significant difference between the number of eggs laid males; 4.9060.60 females (n = 150, range 1–7) or ROCK males; in transgenic crosses than in non transgenic crosses (Table 2; 5.260.32 females (n = 150, range 4–7), x2 = 0.043, df = 1, Mann-Witney test, U = 4402, p = 0. 41). There was a significant p = 0.84. The same applied when mosquitoes were kept together difference in the number of eggs across all the five different ratios for several days. RIDL males fertilised 4.7060.68 females of strains, the ROCK control had a significantly higher number of (n = 150, range 0–8) and ROCK males fertilised 3.660.60 females PLOS ONE www.plosone.org May 2013 Volume 8 Issue 5 e62711 Fitness of Transgenic RIDL Aedes aegypti Males Table 1. Mating competitiveness experiment 2 =/1 R.
For total replicates, the RIDL and ROCK controls and the actual competition experiment (RIDL vs. ROCK), values are given for the number of crosses ( = no. of femaleslaying viable eggs) and for the mean number of eggs laid per female with Standard Errors (6SE). When comparing data from the total replicates, ROCK females fertilisedby RIDL males laid more eggs (Mann-Witney, U = 1863, p = 0.043) whereas in the competition experiment (RIDL vs. ROCK), there were fewer observed fertilisations byRIDL males than expected (x2 test, x2 = 6.75, df = 1, p = 0.009).
doi:10.1371/journal.pone.0062711.t001 (n = 150, range 0–6), x2 = 1.473, df = 1, p = 0.23. The number of valid (Shapiro-Wilk: raw data, p = 0.001; log-transformed data, full spermathecae among inseminated females was also not p = 0.0062). In this experiment, a significant difference between significantly different between strains (For 24 h exposure to males: RIDL and ROCK male longevity is not clear.
RIDL: 2.19 (60.077) vs. ROCK: 2.39 (60.119), Mann-WhitneyU test: U = 16.5, p = 0.1841; for indefinite exposure to males: Anomalous survival of heterozygous RIDL progeny RIDL: 1.94 (60.228) vs. ROCK: 1.80 (60.260), Mann-Whitney U test: U = 25, p = 0.7279). The average number of spermathecae Of 4265 larvae hatched for this experiment, 788 (18.4%) inseminated in the ‘‘24 h'' experiment was not significantly higher survived to adulthood. Of these, 104/350 (29.7%) females and than during the ‘‘until death'' experiment (Experiment 24 h: 2.28 195/438 (44.5%) males survived for one week or more. Thus, (60.072) vs. Experiment until death: 1.87 (60.169), Mann- surviving one-week old adults represented about 7% (2.5% females Whitney U test: U = 265, p = 0.0646).
and 4.5% males) of hatched larvae compared to 99.5% of ROCKlarvae in the control. This rate was unexpectedly high, markedly Adult male longevity greater than in published rates for this strain (3–5%, Phuc et al., Longevity (LT50) was higher when males were maintained 2007 and unpublished data). Comparison of procedures revealed without females. The median longevity of RIDL males was that whereas larvae in our studies were reared on a commercial cat approximately 18% lower than that of ROCK males whether held food (Purina ONEH, Nestle´e Purina PetCare France, Rueil- with or without females (Figure 3). For males only, there was a Malmaison, France), larvae in previous published and unpublished significant difference in longevity between replicates for RIDL studies had been fed a brand of fish food widely used in mosquito males (Log-rank test, p = 0.0014) but not for ROCK males (Log- insectaries (TetraMinH, Tetra GmbH, Melle, Germany). We rank test, p = 0.38). The RIDL vs. ROCK difference, when therefore ran a side-by-side comparison of the two procedures. Of differences for replicates within types of males were allowed for, the 9847 larvae hatched and fed on cat food, 1818 (18%) survived were significant for both the raw data (GLM, p,0.0001) and log- to adulthood. Of the 10413 larvae hatched at the same time but transformed data (GLM, p,0.0001). However, the assumption of fed on fish food, 402 (3.9%) survived to adulthood—consistent normality for both GLMs was not valid (Shapiro-Wilk, p,0.0001).
In this experiment, there is evidence that RIDL males have a with the previous observations of Phuc et al. (2007).
reduced longevity if differences between cages are ignored (Log-Rank, p = 0.0004; Wilcoxon, p,0.0001; PH likelihood, p = 0.001).
For males with females, there were no significant differences Our studies demonstrated that the key aspects of the fitness of between replicates for RIDL males (Log-Rank, p = 0.29) and Ae. aegypti RIDL males carrying a tetracycline repressible lethal ROCK males (Log-Rank, p = 0.42). However, the comparison gene was comparable to that of ROCK males, an encouraging RIDL vs. ROCK that ignored differences between replicates gave step towards the application of this transgenic strain and genetic different results (Log-Rank, p = 0.060; Wilcoxon, p = 0.012; PH control strategy. This conclusion is supported by recent field data likelihood, p = 0.11). We therefore performed the same analysis as showing that OX513A males can compete for mates with wild above. When differences for replicates within types of males were males in the field, and that sustained release can suppress a target allowed for, the RIDL vs. ROCK differences were significant for field population of Aedes aegypti [29,30].
both the raw data (p = 0.026) and log-transformed data (p = 0.006).
However, the assumption of normality for both GLMs was not Table 2. Mating competitiveness experiment 10 =/5 R.
Mean emergence rate This dataset includes 3 replicates of 6 ROCK/4 RIDL and 4 ROCK/6 RIDL. For transgenic and non transgenic crosses: number of crosses ( = no. of females laying viableeggs), mean no. of eggs laid per female with Standard Error (6SE), total no. of emerging males and females and mean emergence rate (no. of adults/no. of larvae)with Standard Error (6SE). There was no significant difference in the number of eggs laid by females fertilised either by RIDL or ROCK males (Mann-Witney, U = 4402,p = 0. 41).
doi:10.1371/journal.pone.0062711.t002 PLOS ONE www.plosone.org May 2013 Volume 8 Issue 5 e62711 Fitness of Transgenic RIDL Aedes aegypti Males Figure 2. Egg hatch rate for different proportions of ROCK andRIDL males. For each ratio of strains (ROCK/RIDL): mean hatching rate Figure 3. Lifespan of RIDL and ROCK males. Proportions of Ae.
(no. of larvae/no. of eggs, white bars) and mean emergence rate (no. of aegypti males surviving when (i) 150 males (RIDL or ROCK) were caged adults/no. of larvae, black bars) with Standard Errors (6SE). Above the alone (grey lines) or (ii) when 30 males (RIDL or ROCK) were caged with figure, values (n) indicate the no. of females laying viable eggs.
120 ROCK females (black lines). Data are the average of two replicates for each experiment (males with or without females). The LT50 values( = median longevity values) were 39 for ROCK males alone, 32 for RIDLmales alone, 11 for ROCK males caged with females and 9 for RIDL Fertilisation and oviposition males caged with females. All mosquitoes were maintained off-TET.
Nearly 100% of ROCK females in the mating competition experiments were fertilised, and there was only one case of mixed in other words no indication that the RIDL males were less progeny (0.3%). This was presumably the result of a double competitive than wild type. However, in the 2 =/1 R experiment, fertilisation, though a similar outcome could be obtained if the there were fewer transgenic matings than expected. The reason for RIDL strain contained heterozygotes. We have no reason to these apparently contradictory results is not clear. Laboratory suspect this alternative explanation here, which would in any case experiments of this type inevitably differ considerably from natural not affect the interpretation. Monogamy [31,32], or at least first- conditions. In the wild, females can probably choose between male paternity [33], is considered typical for Aedes aegypti, though more than two males, so the 10 =/5 R experiment may have been Gwadz and Craig found that inadequate transfer of semen from more ‘natural' in that regard. On the other hand, the 0.54 l mating male Ae. aegypti can result in females remating [34]. Helinski et al arena constrains the mosquitoes to a higher density than in the [35] revisited the question of polyandry in large field cages finding wild; this effect may be more pronounced for the 10 =/5 R 14% of females had engaged in multiple matings. In laboratory mating tests similar to those described here, double mating haspreviously been observed by the authors at 0–6% of total Adult male longevity inseminations (data not shown). Clearly, under certain conditions, Irvin et al [24] found that one homozygous Ae. aegypti strain out Ae. aegypti females can fertilise eggs using sperm from more than of the three had reduced adult longevity but Moreira et al [25] one male. Nearly three times as many eggs were laid per female in found no significant difference in the survival of two heterozygous the 2 =/1 R experiment (mean = 120.3) than in the 10 =/5 R strains of An. stephensi. Note that when homozygous RIDL males experiment (mean = 42.3). This may have been due to a reduction are maintained off-TET, there should be an additional cost in the quantity of blood ingested due to crowding. Alternatively, because adults express the lethal gene. The protocol used – larvae the sperm of RIDL males may be higher in quality and/or reared on-TET and adults held off-TET, was to mimic the quantity because ROCK females fertilized by RIDL males laid conditions to which RIDL males would be exposed if reared and more eggs than those fertilized by ROCK males. These data released into the field in a control program. In our study, the indicate that RIDL males are as effective as ROCK males at median longevity of newly-emerged RIDL males was slightly inducing refractoriness to remating in wild type females.
reduced (18%) relative to ROCK males. This modest reduction issimilar to that seen for related molecular constructs in the Insemination capacity Mediterranean fruit fly Ceratitis capitata (0, 13 and 21% reduction The maximum number of females fertilized by RIDL and for three different transgenic lines [40]).
ROCK males was similar to that in other studies [36–38], andshowed no reduction in the ability of RIDL males to inseminate multiple females, relative to wild type.
Fitness can be defined as the relative success of an individual in passing its genes to the next generation. For mosquitoes, it can be Mating competitiveness estimated as (i) survival, measured as larval biomass productivity, Andreasen and Curtis [39] found that Anopheles stephensi and An.
development time, adult emergence, larvae/adult survival, and (ii) gambiae males irradiated as adults were as competitive as non- reproduction, including parameters such as fecundity, fertility, irradiated males, but were less competitive when irradiated as mating competitiveness. We did not find significant differences pupae. In the 10 =/5 R experiment, there was no difference between RIDL and ROCK in fecundity or mating capacity. In between expected and observed frequencies of transgenic crosses, addition there was no significant difference in mating competi- PLOS ONE www.plosone.org May 2013 Volume 8 Issue 5 e62711 Fitness of Transgenic RIDL Aedes aegypti Males tiveness when several RIDL males competed with several ROCK as noted. Larval density was maintained at ca. 250–300 larvae per males. However, longevity of RIDL males was lower than for tray. For rearing of the RIDL strain, 30 mg/litre of tetracycline ROCK males and RIDL males were less competitive in the hydrochloride (SigmaH) was added to the rearing water, and to the experiment 2 =/1 R.
10% sugar water and blood offered to adults. Insectaries were The lower survival of RIDL males could be due to inbreeding, maintained at 26uC (61uC) and 60% (610%) relative humidity and/or the expression of the lethal gene during adult stage, as with 12-hour light/dark cycle. All data were analysed with the adults were off-TET. In addition, the RIDL strain expresses the software package SPSS version 13 (SPSS Inc., Chicago, IL). In DsRed2 protein under the control of the ubiquitous Actin5C addition, the SAS System for Windows (8.02) was used for adult promoter. Ubiquitous promoters may have a stronger impact on male longevity analyses.
fitness than tissue- or stage-specific promoters [22].
Mating competitiveness ROCK larvae were reared off-TET, RIDL larvae were reared Large quantities of tetracycline and other antibiotics are used to on-TET and pupae transferred to individual 15 ml plastic tubes.
boost growth in factory-reared chickens. The label on the cat food Males and females were separated by gender and caged for 2–3 that we used states ‘‘made with selected chicken'' and the list of days to attain sexual maturity. Mating experiments were ingredients included a minimum of 16% chicken plus dehydrated conducted in cardboard cylindrical cages (diameter 8.5 cm, height poultry protein, hydrolysed liver (source not specified, possibly 9.5 cm, volume 539 ml) for seven days. As adults, all mosquitoes chicken) and ‘‘animal fats'', ingredients that are derived from were maintained off-TET.
processed poultry offal and bone-meal. By contrast, TetraMinH Mosquitoes were introduced to the cylinder cages after comprises fish, molluscs, crustaceae and vegetable materials immobilization at 4uC. Females were introduced after males.
presumably free of tetracycline contamination.
Females were offered off-TET heparinated (1000 IU/ml) rabbit Studies have shown that poultry products (even those used for blood via a ParafilmH membrane [44] after seven days. Males human consumption) may contain oxytetracycline, tetracycline were removed right after blood feeding. One day after blood and chlortetracycline [41–43] at relevant concentrations (e.g. 1– feeding, females were transferred into individual cages. Eggs were 2 mg/ml [42]) and there is little doubt that the presence of such collected on wet cotton disks (make-up removers) and dried for at compounds gave rise to the anomalous survival of insects reared least 3 days in the laboratory before hatching. Spermathecae of on cat food. In nature it is highly unlikely that larvae of Ae. aegypti females that did not oviposit were examined for sperm. Eggs were would ever be contaminated with tetracycline because it is a submerged for 48 hours and larvae reared off-TET no more than container-breeding species, not present in ground pools or other 1 month after egg laying. First or second instar larvae were sites where contamination with tetracycline is possible; in its transferred to individual wells of 96-well plates and screened for original habitat, it breeds in tree-holes and other natural DsRed2 fluorescence. Mortality of the larvae was recorded.
containers but it has adopted the urban, peri-domestic environ- Mating competitiveness: 2 =/1 R.
We compared mating ment by breeding in artificial containers — discarded tyres, success of two males paired with one ROCK female. Three sets of buckets and cans, flower pot saucers etc. — hence its importance 60 cages, each with one ROCK female, contained either (i) two as a highly effective urban vector of yellow fever, chikungunya and ROCK males, (ii) two RIDL males or (iii) one RIDL male and one ROCK male. Survival of the adults was monitored daily. Cages In summary, the mating competitiveness of a strain of Ae. aegypti containing dead mosquitoes were eliminated. Eggs were hatched with a late-acting, tetracycline-repressible gene was comparable to off-TET and mortality was recorded. Including all the data, we that of the ROCK strain. Our results encouraged us to continue compared the number of eggs in transgenic crosses (RIDL male) to our studies in more realistic settings, and using a more wild-type non-transgenic crosses (ROCK male) using the Mann-Whitney genetic background. We introgressed the OX513A insertion into a test. After removing the RIDL and ROCK controls from the Mexican-derived strain and found this derivative to have good dataset, we tested whether the observed frequencies of transgenic mating competitiveness in the field [30] and indeed used it to and non-transgenic crosses differed from the expected (equal suppress a target field population of Ae. aegypti [29]. In addition, we frequency, based on a null hypothesis of equal competitiveness of are investigating dispersal and survival of male and female the two male genotypes) using the x2 test.
mosquitoes in the field. Such data are essential to optimise controlstrategies and field releases in the future. As we have shown that Mating competitiveness: 10 =/5 R.
We assessed the mating late-acting RIDL insects have a comparable fitness as the ROCK competitiveness of RIDL vs. ROCK males when caged with five strain from which it was produced, this adds to the growing ROCK females, using a range of males of the two strains (ratio of evidence that transgenic mosquitoes can be produced without strains ROCK/RIDL: 10/0, 8/2, 5/5, 2/8, 0/10), with eight gross effects on fitness.
replicates of each. We recorded the number of eggs, the number oflarvae hatching off-TET, the mortality of the resulting pupae. All surviving pupae were transferred to tubes and sexed if theyreached adulthood. We compared the number of eggs in transgenic to non-transgenic crosses using the Mann-Whitney Throughout this report, transgenic OX513A mosquitoes test. We analysed the difference between the numbers of eggs laid homozygous for the RIDL construct are referred to as ‘‘RIDL'' across the different ratio of strains using the Kruskal-Wallis test.
unless mentioned otherwise. In addition, rearing of strains with or Lastly, we compared the expected frequencies of transgenic and without tetracycline (TET) are referred to as ‘‘rearing on/off- non transgenic crosses with the observed frequencies, for each type TET'' and maintenance of adults without TET in the sugar water of crosses, using the x2 test. The expected probabilities of a are referred to as ‘‘off-TET''.
transgenic cross ranged from 0 (10 ROCK/0 RIDL), 0.2 (8 All experiments were conducted with the RIDL and ROCK ROCK/2 RIDL), 0.5 (5 ROCK/5 RIDL), 0.8 (2 ROCK/8 strains. Larvae, ca. 250–300 per tray (20630 cm, 1.5 litres), were RIDL) to 1 (0 ROCK/10 RIDL).
fed on commercial chicken-based cat food (Purina ONEH), except PLOS ONE www.plosone.org May 2013 Volume 8 Issue 5 e62711 Fitness of Transgenic RIDL Aedes aegypti Males Insemination rate longevity of RIDL vs. ROCK males alone, and with females; (i) We tested whether RIDL and ROCK males could fertilise equal between replicates in both experiments and (ii) between RIDL vs.
numbers of ROCK females. RIDL larvae were reared on-TET ROCK for both experiments (with and without ROCK females).
and RIDL and ROCK adults were maintained off-TET. Newly When differences between replicates were found, we used a emerged adults were separated by gender and aged in cages for generalised linear model (GLM) on (i) the raw data and (ii) the log- three to four days. Two experiments were performed, with 10 transformed data. The Shapiro-Wilk test was used to test for replicates of each: (i) one male (either RIDL or ROCK) caged with normality of GLM residuals.
15 females for 24 h and (ii) one male (either RIDL or ROCK)caged with 15 females until male's death. When experiments Survival of heterozygous RIDL progeny reared off-TET ended, females were killed by freezing and assessed for the Heterozygous RIDL eggs were hatched and reared off-TET.
presence of sperm. We compared the insemination of RIDL and Larval, pupal and adult mortality was recorded. Surviving pupae ROCK males by analysing (i) the number of fertilized females and were transferred to cages (30630630 cm) for recording pupal and (ii) the number of spermathecae containing sperm using the Mann- adult mortality. 200 ROCK pupae were placed into another Whitney test.
identical cage for control. One week after pupae were put into thecages, the number of surviving of adults was recorded.
Adult male longevity off-TET We tested for variation in the longevity between RIDL and ROCK males when kept (i) with ROCK females or (ii) without.
RIDL larvae were reared on-TET and RIDL and ROCK adults We thank Catherine Lallemand for technical assistance and Institut were maintained off-TET. We set up two replicates, using Pasteur for additional funding. We thank Christl Donnelly with help with 30630630 cm cages, of each of the following: (i) 150 RIDL statistical analyses of adult male longevity. RIDL is a registered trademarkof Oxitec Ltd.
males, (ii) 150 ROCK males, (iii) 120 ROCK females with 30RIDL males, (iv) 120 ROCK females with 30 ROCK males. Deadadults were collected and counted every three days until the last Author Contributions male died. LT50 values ( = median longevity in days) were Conceived and designed the experiments: BMB DN LA PR. Performed the estimated. The Log-rank test, the Wilcoxon test and the experiments: BMB NCC RL RSL KPH DN. Analyzed the data: BMB DN Proportional Hazards (PH) likelihood test were used to compare PR. Wrote the paper: BMB LA PR.
1. WHO-TDR (2006) Scientific Working Group Report on Dengue. Geneva: 20. Alphey L (2002) Re-engineering the sterile insect technique. Insect Biochem Mol WHO. TDR/SWG/08 TDR/SWG/08. 162 p.
Biol 32: 1243–1247.
2. WHO-TDR (2009) Dengue: guidelines for diagnosis, treatment, prevention and 21. Alphey L, Andreasen M (2002) Dominant lethality and insect population control. Geneva: WHO.
control. Mol Biochem Parasitol 121: 173–178.
3. Reiter P, Gubler DJ (1997) Surveillance and control of urban dengue vectors. In: 22. Marrelli MT, Moreira CK, Kelly D, Alphey L, Jacobs-Lorena M (2006) Gubler DJ, Kuno G, editors. Dengue and Dengue Hemorrhagic Fever.
Mosquito transgenesis: what is the fitness cost? Trends Parasitol.
Wallingford, UK: CABI.
23. Catteruccia F, Godfray HC, Crisanti A (2003) Impact of genetic manipulation 4. Alphey L, Beard CB, Billingsley P, Coetzee M, Crisanti A, et al. (2002) Malaria on the fitness of Anopheles stephensi mosquitoes. Science 299: 1225–1227.
control with genetically manipulated insect vectors. Science 298: 119–121.
24. Irvin N, Hoddle MS, O'Brochta DA, Carey B, Atkinson PW (2004) Assessing 5. Christophides GK (2005) Transgenic mosquitoes and malaria transmission. Cell fitness costs for transgenic Aedes aegypti expressing the GFP marker and Microbiol 7: 325–333.
transposase genes. Proc Natl Acad Sci U S A 101: 891–896.
6. Alphey L (2009) Natural and engineered mosquito immunity. Journal of Biology 25. Moreira LA, Wang J, Collins FH, Jacobs-Lorena M (2004) Fitness of anopheline mosquitoes expressing transgenes that inhibit Plasmodium development. Genetics 7. Nolan T, Papathanos P, Windbichler N, Magnusson K, Benton J, et al. (2011) 166: 1337–1341.
Developing transgenic Anopheles mosquitoes for the sterile insect technique.
26. Lambrechts L, Koella JC, Boe¨te C (2008) Can transgenic mosquitoes afford the Genetica 139: 33–39.
fitness cost? Trends in Parasitology 24: 4–7.
8. Catteruccia F, Crisanti A, Wimmer E (2009) Transgenic technologies to induce 27. Marrelli MT, Li C, Rasgon JL, Jacobs-Lorena M (2007) Transgenic malaria- sterility. Malaria Journal 8: S7.
resistant mosquitoes have a fitness advantage when feeding on Plasmodium- 9. Alphey L, Nimmo D, O'Connell S, Alphey N (2008) Insect population infected blood. Proc Natl Acad Sci U S A 104: 5580–5583.
suppression using engineered insects. In: Aksoy S, editor. Transgenesis and the 28. Allen M, Berkebile D, Skoda S (2004) Postlarval fitness of transgenic strains of management of vector-borne disease. Austin, Texas: Landes Bioscience. pp. 93– Cochliomyia hominivorax (Diptera: Calliphoridae). J Econ Entomol 97: 1181–1185.
29. Harris AF, McKemey AR, Nimmo D, Curtis Z, Black I, et al. (2012) Successful 10. Benedict M, Robinson A (2003) The first releases of transgenic mosquitoes: an suppression of a field mosquito population by sustained release of engineered argument for the sterile insect technique. Trends Parasitol 19: 349–355.
male mosquitoes. Nature Biotechnol 30: 828–830.
11. Alphey L, Benedict M, Bellini R, Clark GG, Dame DA, et al. (2010) Sterile- 30. Harris AF, Nimmo D, McKemey AR, Kelly N, Scaife S, et al. (2011) Field insect methods for control of mosquito-borne diseases: an analysis. Vector Borne performance of engineered male mosquitoes. Nat Biotechnol 29: 1034–1037.
Zoonotic Dis 10: 295–311.
12. Sinkins SP, Gould F (2006) Gene drive systems for insect disease vectors. Nat 31. Craig G Jr (1967) Mosquitoes: Female Monogamy Induced by Male Accessory Rev Genet 7: 427–435.
Gland Substance. Science 156: 1499–1501.
13. Gould F, Schliekelman P (2004) Population genetics of autocidal control and 32. Gwadz R, Craig G Jr (1968) Sexual receptivity in female Aedes aegypti. Mosquito strain replacement. Annu Rev Entomol 49: 193–217.
News 28: 586–593.
14. Fraser MJ (2012) Insect Transgenesis: Current Applications and Future 33. George JA (1967) Effect of mating sequence on egg-hatch from female Aedes Prospects. Annual Review of Entomology 57: 267–289.
aegypti (L.) mated with irradiated and normal males. Mosquito News 27: 82–86.
15. Wimmer E (2003) Applications of insect transgenesis. Nat Rev Genet 4: 225– 34. Gwadz RW, Craig GB Jr (1970) Female polygamy due to inadequate semen transfer in Aedes aegypti. 30: 355–360.
16. Morrison NI, Alphey L (2012) Genetically modified insects for pest control: an 35. Helinski MEH, Valerio L, Facchinelli L, Scott TW, Ramsey J, et al. (2012) update. Outlooks Pest Management submitted.
Evidence of polyandry for Aedes aegypti in semifield enclosures. Am J Trop Med 17. Phuc HK, Andreasen MH, Burton RS, Vass C, Epton MJ, et al. (2007) Late- Hyg 86: 635–641.
acting dominant lethal genetic systems and mosquito control. BMC Biol 5: 11.
36. Hausermann W, Nijhout HF (1975) Permanent loss of male fecundity following 18. Thomas DD, Donnelly CA, Wood RJ, Alphey LS (2000) Insect population sperm depletion in Aedes aegypti (L.). J Med Entomol 11: 707–715.
control using a dominant, repressible, lethal genetic system. Science 287: 2474– 37. Foster WA, Lea AO (1975) Renewable fecundity of male Aedes aegypti following replenishment of seminal vesicles and accessory glands. J Insect Physiol 19. Atkinson MP, Su Z, Alphey N, Alphey LS, Coleman PG, et al. (2007) Analyzing 21: 1085–1090.
the control of mosquito-borne diseases by a dominant lethal genetic system. Proc 38. Jones JC (1973) A study on the fecundity of male Aedes aegypti. J Insect Physiol 19: Natl Acad Sci U S A 104: 9540–9545.
PLOS ONE www.plosone.org May 2013 Volume 8 Issue 5 e62711 Fitness of Transgenic RIDL Aedes aegypti Males 39. Andreasen MH, Curtis CF (2005) Optimal life stage for radiation sterilization of 42. Ko¨rner U, Ku¨hne M, Wenzel S (2001) Tetracycline residues in meat and bone Anopheles males and their fitness for release. Med Vet Entomol 19: 238–244.
meals. Part 1: methodology and examination of field samples. Food Addit 40. Gong P, Epton M, Fu G, Scaife S, Hiscox A, et al. (2005) A dominant lethal Contam 18: 293–302.
genetic system for autocidal control of the Mediterranean fruitfly. Nat 43. Moats W (1999) The effect of processing on veterinary residues in foods. Adv Biotechnol 23: 453–456.
Exp Med Biol 459.
41. Ku¨hne M, Ko¨rner U, Wenzel S (2001) Tetracycline residues in meat and bone 44. Mishra K, Kumar Raj D, Hazra RK, Dash AP (2005) A simple, artificial- meals. Part 2: the effect of heat treatments on bound tetracycline residues. Food membrane feeding method for the radio-isotope labelling of Aedes aegypti Addit Contam 18: 593–600.
polypeptides in vivo. Ann Trop Med Parasitol 99: 803–806.
PLOS ONE www.plosone.org May 2013 Volume 8 Issue 5 e62711

Source: http://www.edenext.eu/content/download/32490/283708/version/1/file/EDENext112.pdf