For the irradiation of insects, gamma rays are usually used due to their high energy and penetration. The most common sources of gamma rays are the radioisotopes ⁶⁰Co and ¹³⁷Cs as both have a long half-life and emit high-energy gamma rays. ⁶⁰Co is more easily manufactured and is therefore more often used. In conventional self-shielded irradiators (e.g. the Gammacell 220^®, MDS Nordion, Ottawa, Canada, Figure 1), the sample chamber is surrounded by several rods or "pencils" of the isotope. The dose rate of the cell is determined by the activity of the source and the absorbed dose delivered to the insects is controlled by adjusting the exposure time 21. The sample chamber volume of this machine is 3.7 L. The dose rate distribution in the chamber is not uniform and accordingly, insects receive different dose rates when placed at different positions in the chamber with the dose rate being most uniform towards the centre of the chamber. Besides gamma rays, X rays and accelerated electron beams can also be used to irradiate insects. X rays of appropriate energy have similar penetration as gamma rays, and they have been used in a number of studies on Anopheles irradiation 22232425, but the use of electron beams has not been reported.

Dosimetry is used to quantify the dose received by the irradiated insects. The selection of a suitable dosimetry system depends on several considerations including: dose range of interest, ease of handling, expertise available, cost, and uncertainty that is inherent in the system. For SIT programmes, a radiochromic film dosimetry system has been proposed 26.

During the successive stages of spermatogenesis, germ cells multiply and differentiate. In short, spermatogenesis includes the following developmental stages: primordial germ cells, spermatogonia (primary and secondary), spermatocytes (primary and secondary), spermatids and spermatozoa (mature sperm cells). Spermatogenesis is cystic and within one testis, cysts of different stages of development can be found. Germ cells within a single spermatocyst are more or less at the same stage of development. As spermatocysts mature, they break down and release spermatozoa into the sperm reservoir, situated in the anterior section of the testis.

Spermatogenesis occurs mainly during the larval and pupal stages but mosquito species differ in the timing of the process 27. In Anopheles culicifacies 28, a small number of mature spermatozoa are already present in the sperm reservoir in the late pupal stage and many mature spermatocysts occupy the testis. During the last hour of the pupal stage and the first hours after emergence, spermatocyst maturation continues and in newly emerged males, spermatozoa make up 45% of the testis volume in Anopheles stephensi 29 and 41% in An. culicifacies 28, and spermatocysts in various stages of development (mature or not) are present 282930. These spermatocysts continue to mature and release sperm into the sperm reservoir during adult life, and the percentage of the testis occupied with sperm increases with age 282930.

Radiation-induced dominant lethal mutations arise as a result of chromosomal damage in the treated cells 17. An excellent overview on the induction of dominant lethal mutations by irradiation or chemosterilization is provided by LaChance 31. A dominant lethal mutation occurring in a germ cell does not affect the maturation of the cell into a gamete or the participation of the gamete to form the zygote but causes the death of the developing embryo 17. In general, the earlier stages of spermatogenesis (spermatocytes and spermatogonia) are more radiosensitive than later stages (spermatids and spermatozoa) in terms of irreversible damage, and radiation can result in the death of the developing cell 213233. Irradiation of the later stages results in dominant lethal mutations in spermatozoa that lead to embryonic mortality after fertilization 323435. Irradiation also damages somatic cells, with those undergoing mitosis being the most sensitive 33. Reduced longevity is one of the most commonly observed results of somatic damage 33. Other effects of irradiation can be more subtle. A study in the male house fly Musca domestica showed that irradiation induced considerable changes in the fine structure of the fibrillar flight muscle and caused damage to the flight muscle mitochondria; the damage persisted longer in flies irradiated with higher doses 36.

To reduce somatic damage, insects should be irradiated at, or near to, the completion of their development, i.e. in mosquitoes the late pupal and adult stages. Eggs 3738 and larvae have been irradiated 37, but this caused unacceptable mortality in the treated insects. In general, somatic damage is less pronounced in adults compared to pupae 39, although much depends on the dose and pupal age. However, handling and irradiation of pupae is considered easier due to their relative robustness compared to the adults. In Anopheles, under laboratory conditions, the pupal stage lasts between 25-52 h 40, depending on species and rearing conditions. Following irradiation of Anopheles pharoensis pupae at different ages 41, it was shown that emergence and longevity of adult mosquitoes irradiated as pupae older than 15 h did not differ from un-irradiated insects, even when the dose was high (120 Gy). The irradiation of pupae aged 1-5 h drastically decreased the emergence rate. Similar results were obtained in Anopheles gambiae s.s. 39. In Anopheles quadrimaculatus, the irradiation of young pupae (1-4 h) with 90 Gy resulted in normal emergence 42 but longevity, measured as the percent survival after three days, was greatly reduced. The irradiation of adults following emergence at night is generally performed from the next day onwards (i.e. adults > 12 h old).

In experimental settings, pupae can be irradiated in small wells or Petri dishes, lined with wet cotton wool covered with filter paper 4143 allowing for the irradiation of relatively large numbers of pupae (~500). Adults, however, are much more fragile and require careful handling. Prior to irradiation, adults can be inactivated by chilling 44 which allows them to be confined in a small space within the irradiator so that dose variation can be reduced and mechanical damage to the insects minimized. In small-scale studies, adults can usually tolerate the chilling and stacking for the irradiation but in operational campaigns, very large numbers of insects will have to be irradiated and new protocols will be required. One system has been developed 3944 that allows relatively large numbers of adults to be irradiated simultaneously (~7,000-14,000), and for recent field releases of Aedes albopictus, pupae were irradiated in a device that could hold ~20,000 individuals 45.

To determine the optimal dose for released insects, a wide range of doses is used to generate dose response data. Initially, it is important to confine the insects in a small volume in the centre of the irradiation chamber to ensure a dose uniformity ratio of < 1.1 (where dose uniformity = highest dose/lowest dose). In operational programmes, this precision in dose distribution cannot be obtained as very large numbers of insects will need to be irradiated and the programme managers will need to define the range of acceptable doses. When determining the optimal dose, effects on sterility, longevity, and importantly, competitiveness need to be taken into account 18.

The level of sterility induced in irradiated males is measured by mating the males with un-irradiated virgin females. Eggs are then collected from females individually or en masse and checked for hatching. Unhatched eggs are presumed to have died due to a dominant lethal mutation (after correcting for the control sterility naturally present in the colony 46). Control sterility in laboratory colonies was reported to be 16% in Anopheles arabiensis 24, 10% in An. stephensi and 15% in An. gambiae s.s. 47.

When the residual fertility is plotted on a logarithmic scale against dose, an insight into the number of dominant lethal mutations in a cell is provided 1631. A linear response indicates a "one-hit" relationship whereas departures from linearity indicate a "multi-hit" relationship (i.e. two or more independent events in the same cell produce a single dominant lethal event 16).

Reduced longevity is often a result of radiation-induced somatic damage 33, and this must be measured, ideally, under conditions that induce stress to emphasize any differences. Specifically, male survival during the first days of adult life is important as this is the period when mating is expected to occur after release.

The ability of irradiated males to locate, compete for, and successfully couple with and inseminate the wild females is as important as their level of induced sterility 48. Mating competition experiments are performed to study how well males are able to compete against un-irradiated males for females. Initially, competition experiments are carried out in the laboratory, but field or large outdoor cage tests must also be conducted to reveal those effects that are not evident under laboratory conditions 49. Ideally, irradiated males are competed against wild males for wild females in a semi-field setting as wild males are likely to perform poorly under laboratory conditions.

To perform competition experiments, un-irradiated males and virgin females are introduced into a cage in a 1:1 ratio and irradiated males are introduced at equal and higher ratios. Males will compete for the females and hatching data are collected from eggs laid en masse or from egg batches collected from individual females that are separated after mating. When eggs are collected en masse, a method has been developed 50 for determining a point estimate of competitiveness for sterilized insects. This value, usually called the Fried index, can be determined provided egg hatch data are known for control (Ha) (i.e. un-irradiated females mated with un-irradiated males) and sterile (Hs) matings (i.e. un-irradiated females mated with irradiated males). The competitiveness index (C) is then estimated as C = ((Ha-Ee)/(Ee-Hs))*(N/S); where Ee is observed hatch, N = number of un-irradiated males, and S = number of irradiated males 51. Moreover, procedures have been developed 52 to calculate an estimate of the variance of the C value where a number of replicates have been run, which permits detection of significant differences between values. The Fried index is independent of the ratio of un-irradiated to irradiated males but the variance depends strongly on the ratio and has the lowest value when half the observed matings are by irradiated males.

The levels of sterility induced in several Anopheles species are shown in Table 1. In majority of the studies, pupae were irradiated and at a range of ages between 0-32 h. Although it is desirable to irradiate pupae as late as possible, most studies used pupae around 24 h old as this is the most convenient age for irradiation under normal laboratory rearing conditions and light regimes. Adults were irradiated from less than 24 to 96 h old. The doses administered ranged between 5-120 Gy with some studies using a wide dose-range whilst others tested fewer, depending on the goal of the study. Unfortunately, most studies do not specify whether dosimetry was used to verify the absorbed dose.

The relationship between induced sterility and log dose in insects is sigmoid in form and follows the pattern of a logistic response curve. At lower doses, an approximately linear relationship between dose and induced sterility is observed while at higher doses, the curve flattens such that increasing amounts of radiation are required for proportionally smaller increases in sterility 181954. In general for Anopheles, doses of around 100-120 Gy induce more than 98% sterility (Table 1). At a dose of 80 Gy, more than 90% sterility is observed and at a dose of 50 Gy, sterility exceeds 70%. This is confirmed in chromosome rearrangements studies; when adult males were irradiated with 40-45 Gy, the level of sterility ranged between 75-87% 2455. However, a considerable amount of variation is observed when comparing sterility levels between species (Table 1; 56). A dose of 50 Gy applied to pupae induced 76% sterility in An. arabiensis, 84% sterility in Anopheles albimanus, and 80% sterility in An. stephensi, while in An. pharoensis the level of sterility induced was 96% (Table 1). However, at the higher doses of 100-120 Gy, more than 98% sterility was induced for all species (Table 1).

In a few studies, both pupal and adult stages were irradiated, and the level of sterility induced in each was determined (Table 1). At the high dose of 120 Gy, equal levels of sterility were found for pupal and adult irradiation in An. gambiae s.s., although one experiment suggested pupae to be more radioresistant. In An. arabiensis, pupae were slightly more radioresistant at the doses of 60-80 Gy compared to adults 56. Anopheles stephensi pupae and adults were irradiated in two separate studies; at 50 Gy, a higher level of induced sterility was reported for the adult irradiation. At 80 Gy, this difference was no longer observed.

Longevity was measured in a small number of studies and scored as daily mortality under normal rearing conditions. In most studies, differences in longevity of irradiated males compared to un-irradiated males were small with only trends being reported. Pupal irradiation in An. pharoensis resulted in a non-significant reduction in longevity after irradiation at doses of 100-130 Gy 41; while in another study, a non-significant increase in longevity after irradiation with 5-70 Gy was reported 43. In An. arabiensis, longevity of males irradiated as pupae with 25-100 Gy was increased, similar, or reduced compared to un-irradiated males but differences were small and longevity of males irradiated as adults was similar to un-irradiated males 56. However, in An. stephensi, the longevity of males irradiated as pupae at 80 Gy was significantly reduced 57. When pupae (22-27 h) and adults (< 24 h) of An. gambiae s.s. were irradiated with a high dose of 120 Gy, an increased mortality for the irradiated pupae 24 h after irradiation was reported compared to zero mortality in the irradiated adults 39.

In general, the mating ability of males does not seem to be adversely affected by irradiation. The number of eggs produced by the females mated to males irradiated over a wide dose-range as pupae 223741 or adults 58 is similar to un-irradiated insects; this suggests that sperm transfer from irradiated males to females was normal. Insemination rates were determined 2256 and in An. albimanus, irradiated males inseminated females at an equal rate as un-irradiated males 22. However, in the irradiated males × irradiated females group at 80 Gy, a greatly reduced insemination rate was observed 22. In An. arabiensis males irradiated as pupae, a weak but significant negative correlation was observed between dose and insemination, while this was not observed after adult irradiation 56.

The ability of irradiated sperm to compete with normal sperm was assessed in An. pharoensis 41. Males irradiated with 120 Gy as pupae were allowed to mate for a number of nights with virgin females after which they were removed and replaced with normal males. Females laid sterile eggs indicating that either remating did not occur or the first mating took precedence. Storage of sperm in the testes did not restore fertility when males were mated at five days after pupal irradiation with 120 Gy 41, and no recovery of fertility was observed when males irradiated as pupae or adults were remated to a second batch of virgin females after 8-9 days 39. In addition, storage of sperm in the spermathecae from males irradiated as adults did not restore fertility when females oviposited twice 39. A study in the honeybee, Apis mellifera, showed that mature sperm could not repair radiation damage even after prolonged storage of up to one year in the spermatheca 59.

Competition experiments in the laboratory have been performed in some anopheline species. Irradiated males were introduced in various ratios into rearing cages with un-irradiated males and virgin females. In most cases, the un-irradiated males and females were laboratory-reared individuals with the exception of one study 60 which used wild-caught material. Mating was allowed for a number of nights, and males were introduced soon after irradiation 4247 or some days after emergence 57.

A selection of data on competitiveness experiments in anophelines is presented in Table 2. Most studies used only one high dose and no comparison was made between higher and lower doses with two exceptions 5761. Irradiation almost always had a negative impact on competitiveness of the males, especially so if the pupal stage was irradiated, resulting in higher egg hatch than would be expected if the irradiated males were equally competitive with the un-irradiated males. Egg hatch could be lowered by increasing the ratio of irradiated to un-irradiated males (Table 2) but it should be noted that where the experimental egg hatch is close to either control value (because the ratio used is too high or too low), the variance increases rapidly and the C values become meaningless 5262.

High doses have a negative effect on competitiveness. In An. stephensi, males irradiated as pupae with 80 Gy were 1.7 times more competitive than males irradiated as pupae with 120 Gy 57. In An. arabiensis, the use of the higher irradiation dose of 120 Gy resulted in a reduced competitiveness compared to 70 Gy for pupal and adult irradiation 61. In Culex quinquefasciatus, the irradiation of adults with a dose of 50 Gy resulted in high competitiveness, but as the dose increased the competitiveness decreased 63.

Where both the pupal and the adult stages were irradiated with 120 Gy, males irradiated as adults were more competitive than males irradiated as pupae (Table 2). In Cx. quinquefasciatus, a dose of 80 Gy applied to pupae resulted in lower competitiveness when measured at a ratio of 1:1 compared to that for adult irradiation with a slightly lower dose of 75 Gy 763.

Some field studies on competitiveness of radiation-sterilized mosquitoes have been performed, although few with anophelines. In An. quadrimaculatus, pupae irradiated with 120 Gy and released as adults were not able to induce sterility in the target population after prolonged releases due to behaviour differences as a result of the colonization process 64. In Culex tarsalis, males irradiated with 50 Gy (95% induced sterility) as adults were competitive with un-irradiated males (measured at a ratio of 1:1) from the laboratory or from field populations, in small cages indoors or large cages outdoors 65. Males irradiated with 70 Gy were also competitive in small cages (not tested in field cages 65). In a small release of Cx. tarsalis (i.e. single release of 13,000 males), males irradiated with 60 Gy were effective in inducing some sterility in the target population 66. The continuous release of Aedes aegypti males (i.e. 4.6 million in 43 weeks) sterilized as pupae did not, however, result in population suppression, but the dose applied was high (110-180 Gy; 67).

An important factor during radiation is the oxygen level as oxygen molecules form free radicals that induce biological damage 68. Irradiation in a low oxygen environment reduces genetic and somatic damage and consequently, higher doses are needed to induce sterility levels comparable to those induced in air, although it is often observed that competitiveness and longevity are improved despite the higher dose required. Two strategies are commonly used to reduce oxygen levels. Irradiation under hypoxia brought about by respiration of pupae kept in sealed bags is routinely performed with Mediterranean fruit fly pupae. Nitrogen has been used experimentally in tsetse 69 and routinely in Western Australia for Mediterranean fruit fly 70. Prior to irradiation, the container that holds the insects is flushed with nitrogen for some time after which irradiation follows. Beneficial effects, i.e. long-term survival, of irradiation in a nitrogen environment were demonstrated in Mediterranean fruit fly 68 and tsetse 697172.

Irradiation of mosquitoes in nitrogen has been performed in one anopheline and two culicine species. No beneficial effects of pupal radiation in An. gambiae s.s. 39, or pupal or adult radiation in Cx. quinquefasciatus 63 in nitrogen were reported. As expected, insects required much higher doses to achieve adequate sterility in nitrogen compared to air but at those higher doses, only a marginal improvement in competitiveness was observed in Culex adults (competitiveness was not assessed in An. gambiae s.s.). However in Ae. aegypti, irradiation in nitrogen was shown to be beneficial 73. Competitiveness of males irradiated at 35, 70 or 100 Gy in nitrogen was equal to that of un-irradiated males, while males irradiated in air at the same doses were less competitive. Irradiation in nitrogen did increase egg hatch to some extent, but at 100 Gy, in nitrogen as well as air, 100% sterility was achieved.

Radioprotectors are substances which when present during irradiation diminish its effects. A wide range of radioprotectors is available with various modes of action 74. A range of protectors including amino-acids, cysteamine (aminothiol), diaminoethanetetraacetic acid (EDTA), and 2-aminoethyl isothiuronium bromide (AET) were tested on Cx. quinquefasciatus 75. Pupae were soaked for a number of hours in the compounds, pre- and post-irradiation. None of the tested radioprotectors seemed to have a beneficial effect on the competitiveness or sterility of the irradiated males. However, little absorption of the radioprotector is expected in the non-ingesting pupal stage. Another protector, dimethyl sulphoxide (DMSO), ingested in the adult stage before irradiation, decreased the induction of dominant lethal mutations by X rays in Anopheles atroparvus 76. However, DMSO is toxic and even at low concentrations a reduced life span was observed.

Another potential class of radioprotectors are anti-oxidants which neutralize free radicals and thus prevent damage. One of these is nordihydroguaiaretic acid (NDGA), a reducing agent that replaces the naturally present reducing agent glutathione, whose amount decreases with the age of an organism 77. NDGA administered to the larval diet of Ae. aegypti increased the longevity of both sexes over un-treated controls by 42-64% 78. The use of antioxidants has not been studied in mosquito irradiation.

Even though it is known that irradiation reduces competitiveness, other methods to induce sterility are controversial or not in use. Chemosterilants offer high levels of sterility with more competitive insects compared to irradiation 757 but their use requires special safety precautions which are difficult to implement under field conditions. Even though public opinion led to the disappearance of chemosterilants for mosquito control in the seventies, they continue to be used against other pests. In 1991, a large field trial to eradicate introduced sea lampreys Petromyzon marinus from the Great Lakes (USA) was initiated and lasted for several years. Radiation sterilization was considered but yielded unsatisfactory results regarding male survival and competitiveness hence, male lampreys were sterilized with the chemosterilant bisazir. However, the hazards of handling were acknowledged and alternative strategies were explored to eliminate the use of mutagenic agents 79.

Another potential class of sterilants are insect growth regulators (IGRs). IGRs have been used as sterilizing agents against housefly M. domestica 80, blowfly Lucilia sericata 81 (triflumuron) and the Mediterranean fruit fly 82 (lufenuron) using impregnated targets. The compounds were effective in inducing some sterility, although dependent on the dose administered 81. Males could induce sterility in unexposed females, although with various degrees of success, and this was attributed to either the result of direct impairment of sperm development or transfer of active ingredient during mating to the females 81. Also, in the tsetse fly Glossina morsitans morsitans sterility could be introduced in mates when males were exposed to certain juvenile hormone mimics as pupae 83. The potential use of IGR's as a sterilant for mosquitoes remains unknown.