The sterile insect technique (SIT) is an environmentally friendly, species-specific biological control tactic for the management of selected insect pests, that requires the mass production and sequential release of large numbers of sterile insects into a target population 1 . The male insects are exposed to ionizing radiation for sexual sterilization, and compete with wild males to mate with females in the field, thereby inducing sterility in the native female population, which will result in a decline in the target population in subsequent generations.

The successful implementation of SIT against the malaria vector Anopheles arabiensis relies primarily on the sterile males’ competitiveness and mating success in the field. The competitiveness of colony reared, sterilized, and released male mosquitoes is linked to numerous biological parameters, such as longevity, flight performance, spatial occupation of the habitat, available sperm complement, and mating behaviour 2 3 4 5 . Each of these parameters could be influenced by the various steps of the production process of the sterile males, three of which in particular (in addition to the rearing process and laboratory colonization 6 ) may induce a significant decline in overall quality of the An. arabiensis males. First, to eliminate potentially disease-transmitting female mosquitoes from field release material, a genetic sexing strain (GSS) for An. arabiensis (ANO IPCL1) based on a dieldrin-resistant mutation was previously developed 7 . The presence of the dieldrin resistance gene (Rdl) and/or the complexity of the chromosomal translocation induced in this strain could lead to inherent biological differences between the ANO IPCL1 males and the wild-type males. Second, during the sexing procedure at the egg stage, embryos are exposed to dieldrin and the resistant males survive while the susceptible females die. The ability to eliminate females at this early developmental stage has advantages, such as reduced production costs, labour and space requirements 7 , but may impact the biology or performance of the treated adults. Finally, the male pupae are subjected to a dose of 75 Gy of gamma or X-ray irradiation resulting in >98% sterility 8 before they emerge and are ready for release.

The GSS ANO IPCL1 has been evaluated in terms of reliability of female elimination 7 , rearing parameters 9 , strain management 7 , and radiation sensitivity 7 8 . GSSs based on dieldrin resistance have been produced in the past for An. arabiensis 10 and Anopheles gambiae 11 , but these strains no longer exist. Life history traits such as developmental parameters of the immature stages, adult size (wing length), adult fecundity and longevity were similar between the ANO IPCL1 and the wild-type An. arabiensis Dongola strain with the exception of the high intrinsic sterility observed in the ANO IPCL1 9 . Moreover, the ANO IPCL1 strain had a similar sperm production pattern as the Dongola strain, suggesting that this specific gene-translocation did not modify this reproductive parameter 12 . However, it has been observed in other strains, that there may be an intrinsic loss of vigour related to the dieldrin resistance gene 13 . Females of homozygous-resistant An. gambiae and Anopheles stephensi strains were less responsive to oviposition stimuli, produced fewer eggs per unit of blood, were less mobile when seeking hosts or oviposition sites, and responded slower to simulated predators as compared to females of heterozygous-resistant and homozygous-susceptible strains 13 14 . The resistant males were generally less successful in competing for females, which might be related to their lower mating success because their reaction to female mating cues (as to predator movements) was generally slower. These results should be considered carefully as there was no attempt to distinguish strain from resistance gene effects. However, in the light of other findings, the general fitness and quality of ANO IPCL1 must be scrutinized with a series of experiments to ensure that there are not prohibitive reductions in competitiveness.

Dieldrin (C₁₂H₈Cl₆O), a potent insecticide, is a very persistent organic pollutant known to be absorbed and stored mostly in the adipose tissue of insects and mammals 15 16 . Fourth instar larvae of the GSS ANO IPCL1 can be treated in 0.1 ppm dieldrin solutions to eliminate females, which can also be achieved at the embryonic stage by treating <12-hour old eggs in 3–4 ppm dieldrin solutions. Dieldrin adheres strongly to many surfaces, and is absorbed through the chorion of the mosquito eggs, where its residues are retained by the mosquito until adulthood 17 . Furthermore, significant differences were seen in adult male longevity after treating the eggs with the insecticide compared to untreated males (Yamada, unpublished data). Dieldrin treatment however had a positive impact on sperm production in adult males that were irradiated as pupae with the production of sperm continuing in the first week of adult life, while males that had only been irradiated as pupae without the dieldrin treatment did not produce new sperm cells during adult life 12 . It was therefore hypothesized that dieldrin treatment might have a radioprotectant effect on the An. arabiensis germinal cells.

Earlier studies documented the effect of irradiation on some Anopheles male life history traits, i.e., irradiating pupae of An. arabiensis with gamma rays (with doses ranging from 25–100 Gy) had little impact on pupal survival, adult longevity and fecundity of females mated with irradiated males 18 . Sterilizing male mosquitoes at the adult and pupal stages using X-rays (at doses ranging from 35–105 Gy) had some negative effects on adult longevity, possibly due to the stress imposed on them while in the irradiation canister, as well as somatic damage caused by irradiation 8 . Moreover, X-ray irradiation at 70 Gy decreased the number of sperm in testis in two-day old males and inhibited subsequent sperm production during adult life 12 .

The current study aimed to evaluate the mating competitiveness of male ANO IPCL1 compared to wild-type males to assess any potential prohibitive reductions in competitiveness due to the chromosomal aberrations or the Rdl gene in the GSS. Second, the effects of irradiation on the mating competitiveness of ANO IPCL1 males compared to fertile wild-type males were evaluated, and finally, the effects of the dieldrin treatments in addition to the irradiation of ANO IPCL1 eggs were assessed on adult male competitiveness when competing against untreated wild-type males. The present study was carried out in support of a feasibility study to assess the potential use of the SIT as part of an area-wide integrated pest management (AW-IPM) strategy against An. arabiensis in the Northern State of the Republic of Sudan.

The general methodology of the various experiments remained identical but according to the treatment different treated:untreated male ratios were used.

For all experiments, the recovery rate of females from the field cages ranged between 85 and 95%.

In the context of AW-IPM programmes including an SIT component, it is well known that colonization, mass rearing and irradiation processes could affect the mating competitiveness of the released males 26 27 28 29 30 31 . Male An. arabiensis males destined for release in programmes that include a SIT component, need to endure several treatments: (1) for the purpose of sex separation, a complex translocation of the Rdl gene was induced for the development of the genetic sexing strain (GSS) ANO IPCL1 7 ; (2) dieldrin treatments at immature stages are necessary for female elimination from the production line; and, (3) irradiation at pupal (or adult) stage is required for the sterilization of the males prior to release. These treatments could have a negative impact on the mating competitiveness of the males produced.

The results indicate that prohibitive reductions on competitiveness due to biological effects of the presence of the Rdl gene, and the complex translocation associated with this GSS can be precluded as the competitiveness of the ANO IPCL males was not compromised when competing with wild-type males for wild-type females in a field-cage setting. Under the experimental conditions of the current study, the ANO IPCL1 males were equally competitive compared to the wild-type strain. Similar findings were documented in the past: the chromosomal aberrations inherent in the MACHO strain of Anopheles albimanus had no negative effects on the competitiveness of the males 32 , and males of Culex tarsalis carrying a chromosomal translocation were also equally competitive as wild-type males when assessed in large laboratory cages (~4 m³) 33 .

According to the competitiveness index elicited by a comparison of the observed hatch rate with the expected hatch rate in this current study, irradiated ANO IPCL1 males (IRR GSS) exhibited a C value of around 0.5 (where a value of 1.0 suggests equal competitiveness), meaning that the males were approximately half as competitive as wild-type males when evaluated in this given scenario. Assuming that the C value obtained in this field-cage study was comparable to hypothetical values that can be expected in the field, the number of irradiated ANO IPCL1 males released in a control programme against this species would need to be doubled to arrive at an equal level of competitiveness. It needs to be noted that in an operational programme, the operational sterile to wild male ratio required in the field would probably need to be higher as the competitiveness of the sterile males is also influenced by their dispersal capacity, their spatial occupation of the habitat and their survival 34 .

Due to the variation of experimental methodologies regarding cage sizes, release ratios and adult densities, the comparison of C values from different studies as presented in the literature is difficult and non-transferable to other situations. For the same species, with the same sex ratio but in different cages, a C value of 0.76 and 0.34 was obtained for An. arabiensis Dongola males irradiated at 70 Gy released in small (30 cm) and large (2.16 m³) semi-field cages, respectively 23 . In field tests assessing male mating competitiveness for the GSS of An. albimanus (MACHO) in El Salvador against newly colonized wild-type males, a C value of 0.785 was reported 32 . A C value between 0.84 and 0.92 was obtained for an artificially infected Aedes polynesiensis strain generated by introgressing Wolbachia from Aedes riversi when evaluated in semi-field cages (2.9 m³) and conditions against F1 offspring from wild-caught A. polynesiensis females 35 . An Aedes albopictus (Stegomyia albopicta) line (ARwP), harbouring a new Wolbachia infection achieved a C value of 1.0 when they competed with naturally infected males (SR line) in laboratory cages (30 × 30 × 30 cm) and in a greenhouse (120 m³) 36 . Unfortunately, many of these reports do not indicate the variance of the C values and it is therefore difficult to deduct if C values between treatments truly differ. The coefficient of variation of the C values reported in this paper is very high making it difficult to make sound conclusions on the actual competitiveness and the differences between groups. The competitiveness index therefore was used as an indicator of mating success, which is influenced by various intrinsic or extrinsic factors and to compare treatment and control groups.

The irradiated ANO IPCL1 males additionally treated with dieldrin at the egg stage were more competitive than those that had only been irradiated. Relevant side effects of dieldrin treatments have also been observed in a study on sperm production in An. arabiensis that showed that dieldrin-treated/irradiated males had a higher level of sperm production at day 6 (post emergence) than males that were only irradiated 12 . These data suggest a radio protectant effect of dieldrin, or the dieldrin treatment procedure itself, on the An. arabiensis germinal cells. This current study seems to confirm that the somatic damage incurred by irradiation in ANO IPCL 1 males could be slightly reduced by the presence of dieldrin residues, which are known to be retained by the insect to adulthood 17 . Alternatively, the physiological stress of the dieldrin treatment potentially could have enhanced or stimulated the radiation protection mechanisms in the mosquitoes 37 . There are only a few chemicals known to have a radiation-protecting effect in mosquitoes: one of these is dimethyl sulphoxide (DMSO) that supplied to adult Anopheles atroparvus decreased the number of dominant lethal mutations induced by the X-rays, but the treatment reduced the longevity of the adult mosquitoes 38 . As observed in most of the competitiveness experiments in large cages, there was a high variability between replicates 39 , however such variability was not observed in the controls which excluded cage volume as the source of the variation. Another factor that might have contributed to the variations is the heterogeneous irradiation of the pupae. A variation of -10 and +7% of the central dose value in the stackable plates used for pupal irradiation has been measured in the X-ray irradiator 22 , meaning that a dose range of 67.5-80.25 Gy was received by the pupae. The pupae also varied slightly in age, ranging from 24- to 36-hours old. This could have contributed to the proportional variation in the final level of sterility in the males. Even if the utmost is done to avoid any variation in the absorbed irradiation dose, none of the exposed pupae would receive an equal dose, with a different impact on their biological quality as a result. The above is corroborated by another competitiveness study that obtained a high degree of variability in hatch rates of eggs of individual females mated with males irradiated as pupae, suggesting small differences in the irradiation dose received 23 . In any case, increasing the number of replicates would be advantageous though this requires a large amount of mosquitoes of homogeneous age at one time which is difficult to achieve and involves a great deal of work.

According to the irradiated: untreated male release ratios studied, a minimum ratio of 10:1 was required to reduce the cage population’s fertility by 81% and to inseminate 75% of the females present in the cage. Similar levels of induced sterility (81 ± 4%) were obtained in Aedes albopictus studies with a ratio of 5:1 irradiated to untreated males using similar semi-field cages 40 . For many other insect species, when deployed in the field as part of integrated pest management (IPM) programmes, it has been necessary to release sterile males in numbers adequate to obtain ratios in the field in favour of sterile males as these proved less competitive than wild males; In Burkina Faso, a sterile to wild male tsetse fly (Glossina palpalis gambiensis) ratio of 7:1 to 10:1 was most effective in achieving eradication of the targeted wild population 41 ; ratios of 40:1 were the target for the suppression of a codling moth (Cydia pomonella) population in British Colombia, Canada 42 ; and sterile male Mediterranean fruit flies (Ceratitis capitata) were released obtaining a 100:1 sterile: wild male ratio in the field, which resulted in a significant decline in fertility of a wild population in Guatemala 43 . Moreover, the frequency of releases will influence the efficiency of the method and have to be frequent enough to maintain a critical overflooding ratio 44 . A release of An. albimanus males in El Salvador proved effective even though sterile males were determined to be two to four times less competitive than their wild counterparts 45 .

It is interesting to note that in these experiments, the level of induced sterility in the female population by the un-irradiated GSS males was nearly as high as the induced sterility achieved by the GSS males irradiated at 75 Gy, while competitive integrity was maintained in the non-irradiated males. This leads to the assumption that due to the high natural sterility and observed equal competitiveness of ANO IPCL1 in the field cage setting, one could increase overall competitiveness of the males by reducing the irradiation dose and thereby ultimately achieving a higher level of sterility in the target population 46 . An additional advantage of this approach is that the sterility will have residual effects as any male progeny resulting from released males mated with wild females will carry forth the same level of natural sterility, producing a somewhat residual, yet self-limiting effect. Nevertheless, these assumptions must be bolstered by a series of field-based assessments before explicit recommendations can be made. The assessment of male Culex tritaeniorhynchus carrying a complex chromosomal aberration showed that they were highly competitive when released with laboratory-colonized females in a laboratory setting, but were uncompetitive when released during a field study 47 suggesting that long-term insectary maintenance can select an assortive mating behaviour. This underlines the importance of using wild females to evaluate the competitiveness of altered genotypes for a more accurate indication of the projected field performance of released males.

Moreover, even if the C values obtained here in field cages have to be treated with caution, the results presented here are a good initial indicator for the relative mating competitiveness of ANO IPCL1 males that have undergone different degrees of treatments compared to laboratory-reared wild-type males (An. arabiensis Dongola). However they require verification in a true natural setting in the field and further studies on the evaluation of the mating success of treated ANO IPCL1 males with wild females are required before any conclusions can be drawn regarding the suitability of this GSS for SIT operations, or any sound recommendations on required released sterile to wild male ratios can be made.

The authors declare that they have no competing interests.

HY, GM and DD carried out the experiments. DD conceptualized the experimental design and supervised the entire project. DD, HY and GM contributed the rearing, irradiating and dieldrin treatments of the strains. HY drafted the initial manuscript. DD and MV edited and contributed a great portion of the latter versions of the manuscript. JG supervised and supported the project. All authors read and approved the final manuscript.