Genetic sex separation in Anopheles arabiensis and the production of sterile hybrids

BRIEF COMMUNICATIONS Bulletin of the World Health Organization, 56 (3): 453-454 (1978) Genetic sex separation in Anopheles arabiensis and the production of sterile hybrids C. F. CURTIS 1 Abstract The gene for dieldrin resistance has been artificially male-linked so that females can be selectively killed with dieldrin. By intercrossing different sibling species ofthe Anopheles gambiae complex, batches consisting ofsterile males only can be reared. This seems to have potential for use in genetic control operations. Mosquitos of the Anopheles gambiae complex are the main vectors of malaria and filariasis in Africa and are therefore responsible for the transmission of more disease of man than any other insect. Within this complex, A. arabiensis (=A. gambiae species B) a is particularly difficult to control by conventional house spraying because of its tendency to rest out of doors (1). Therefore, consideration of genetic control is especially worthwhile for this species. A prerequisite for a genetic control operation is a method of eliminating females from batches of males being prepared for release so that those released do not add to the biting population. A genetic sexing system, designated R70, was produced in A. gambiae sensu stricto (= A. gambiae species A) (2). It is based on translocation of the semidominant autosomal gene for dieldrin resistance onto the Y chromosome so that when R70 males are crossed and backcrossed to homozygous dieldrin-susceptible females, the male progeny are heterozygous and the females are homozygous susceptible. Such females can be killed at the first-instar larval stage by a 24-h treatment with dieldrin at a concentration of 0.02 mg/litre leaving the males unharmed. I now report the production of a similar sexing system in A. arabiensis. Materials, methods, and results A stock of A. arabiensis was available from the Thies area of Senegal, where it is the predominant 1 Research Fellow, Ross Institute, London School of Hygiene and Tropical Medicine, London WC1E 7HT, England. a The Commission on Zoological Nomenclature has accepted the earlier described name of A. arabiensis Patton for A. gambiae species B. species of the A. gambiae complex (3). It is dieldrin resistant and was chosen as one of the parents of the proposed system. For the other parent, dieldrin- susceptible A. arabiensis were required. Such a stock was not available from Senegal and, in fact, there have been difficulties in this laboratory in maintaining stocks of dieldrin-susceptible A. arabiensis. A few susceptible individuals were available from Mauritius and, although they could not be maintained as a stock, three males were successfully artificially mated to females of a well established predominantly dieldrin-resistant laboratory stock of Nigerian origin (KCD). The progeny were mated among themselves for two generations and mated females were then isolated for egg laying. Samples of ten first-instar larvae from each female were tested with dieldrin at a concentration of 0.02 mg/litre and in two cases all ten died, indicating homozygosity for suscepti- bility. The sibs of these were reared, rechecked for susceptibility at the adult stage, and pooled to yield a susceptible homozygote stock, which was designated SB. This stock was reared in the lab- oratory, though its egg production was rather low. Adult males of the Thi6s stock were irradiated with gamma rays at 40 Gya and mated to SB females. The male progeny were backcrossed to SB females, which were isolated for egg laying. Among the egg batches laid by 18 females, five gave 38 %-65% hatch, which suggested the presence of induced translocations. When the progenies from these were tested with dieldrin, four were found to include a proportion of resistant females and they were discarded. However, when the fifth family (desig- nated T 15) was tested on 0.4% dieldrin papers for 1 h (a treatment that kills susceptible homozygotes but not heterozygotes), all 25 males survived and all 13 females died. The males were backcrossed to SB females and the progeny were reared without dieldrin treatment and mated among themselves to establish T 15 as a self-perpetuating stock. It has an egg hatchability of only about 45 %, which is assumed to be due to the presence of a male-linked a 1 Gray = 100 rad. 3708 453- 454 BRI1F COMMUNCATIONS translocation. Crosses of T 15 males to homozygous dieldrin-susceptible females of A. gambiae s.s. (16cSS stock) were made. Exposure of the adult progeny to 0.4% dieldrin papers yielded 41 live males and 35 dead females. Treatment at the first- instar stage with dieldrin at 0.02 mg/litrea yielded 160 surviving males and no females. Thus any crossing over between the dieldrin locus and the translocation is sufficiently infrequent for practical purposes. Males that survived these treatments and that were species hybrids were mated to normal females; no larvae hatched from 1333 eggs laid. Dissection of the males confirmed that this sterility was due to early interruption of spermatogenesis typical of the hybrids from an A. gambiae s.s. female x A. arabiensis male cross (4). For comparison, R70 males were crossed to SB females. The testes of the males from this species cross, with A. arabiensis as the female parent, had spermatogenesis interrupted at various stages, with some spermatids and a few spermatozoa, which conforms with the results obtained by Davidson et al. (4). When the males were mated to normal females there was 1.3% hatch from 1778 eggs laid. Discussion It is thus possible to rear batches of mosquitos consisting only of sterile males. Before release experiments with such males are considered, it would be necessary to introduce into the parental stocks the genome of wild populations from the area where the releases were planned. Even if this were done, it is questionable whether F1 species hybrids would be able to compete for mates in the field, in view of a Subsequent work has shown that with A. arabiensis dieldrin at 0.01 mg/litre is preferable for the selective killing of susceptible homozygotes without any killing of hetero- zygotes. the fact that A. arabiensis and A. gambiae s.s. have behavioural isolating mechanisms that almost completely prevent cross mating in the field (5), though not in the laboratory. Fraccaro et al. (6) have shown that in the A. maculipennis complex, a species difference in mating behaviour is under the control of the Y chromosome. If the same were true in the A. gambiae complex, male hybrids with T 15 male parents would have mating behaviour of the A. arabiensis type. To increase the chances of obtaining sterile males with A. arabiensis mating behaviour, crosses have been carried out that allow the production of hybrids with autosomes very largely of A. arabiensis origin. By the use of the sex- linked marker white eye (7), it has been arranged that the X chromosomes of the male hybrids are of A. gambiae s.s. origin, which is sufficient to ensure that they are sterile. ACKNOWLDGEMENTS I am grateful to Mamadou Sarr and C. M. Courtois for the stocks of Senegal and Mauritius origin, to Barbara Sawyer and Jill ChaLkley for technical assistance, and to the Middlesex Hospital for irradiation facilities. REFRENCES 1. WHITE, G. B. Transactions of the Royal Society o. Tropical Medicine and Hygiene, 68: 278-298 (1974) 2. CURTIS, C. F. Er AL. Mosquito news, 36: 492-498 (1976). 3. Coz, J. Cahiers O.R.S.T.O.M., serie entomologie medicale et parasitologie, 11: 3-31 (1973). 4. DAVIDSON, G. ET AL. In: Wright, J. W. & Pal, R., ed. The genetics of insect vectors of disease. Amsterdam, Elsevier, 1967, chap. 6. 5. WHITE, G. B. Nature, 231: 184-185 (1971). 6. FRACCARO, M. ET AL. Nature, 265: 326-327 (1977). 7. MASON, G. F. Genetical research, 10: 205-217 (1967).