Scalloped wings Is the Lucilia cuprim Notch Homologue and a Candidate for the Modi @ of Fitness and Asymmetry of Diazinon Resistance Andrew

The Scalloped wings ( S c l ) gene of the Australian sheep blowfly, Lucilia cuprina, is shown to be the homologue of the Drosophila rnelanogaster Notch gene by comparison at the DNA sequence and genetic levels. A L. cupn'na genomic fragment, which shows strong identity with the Notch (N) gene at the molecular level, hybridizes to the location of the Scl gene on polytene chromosomes. The two genes are functionally homologous; the dominant and recessive Notchlike phenotypes produced by mutations in the Scl gene allow these alleles to be classed as Nlike or Abruptexlike. The Scl gene is under investigation as a candidate for the fitness and asymmetry Modzjim ( M ) of diazinon resistance. We show that M affects the penetrance of wing and bristle phenotypes associated with two Scl alleles in a manner consistent with the M being an allele of Scl. In addition, we report a phenotypic interaction between the diazinonresistance mutation, Ropl, and the same alleles of Scl. We propose that the product of Rep-1, an esterase, may be involved in cell adhesion in developmental processes involving the Scl gene product. T HE problematic occurrence of resistance to insecticides in target and nontarget species provides fortuitous case studies for evolutionary biologists ( ROUSH and MCKENZIE 1987; MALLET 1989; ROUSH and TABASHNIK 1990). In these situations the selective agent, the insecticide, is known and the significance of changes in susceptible or resistance allele frequency is explicable in terms of the resulting phenotype. For these reasons it is possible to ask detailed questions about the evolution of insecticide resistance. A well documented case is the evolution of resistance to the organophosphorous (OP) insecticide diazinon in the Australian sheep blowfly, Lucilia cupma . Allelic substitution at a single locus ( Rop-1) , which encodes the carboxylesterase E3, is responsible for resistance (HUGHES and -0s 1985; RUSSELL et al. 1990; PARKER et al. 1991 ) . In the absence of insecticide, the relative fitness of the resistant flies is reduced in comparison with susceptible flies ( MCKENZIE et al. 1982) . In addition, the resistant flies have increased levels of asymmetry, a measure of differences between the right and left sides of a bilaterally symmetrical organism (CLARKE and MCKENZIE 1987; MCKENZIE and CLARKE 1988; MCKENZIE 1993). These fitness and asymmetry effects are proposed to be the result of developmental instability caused by the introduction of the new resistance allele into a genome Corresponding authm: Philip Batterham, Department of Genetics, University of Melbourne, Parkville, Victoria 3052, Australia. E-mail: [email protected] sity of Minnesota, St. Paul, MN 55108. 2601, Australia. ' Present addrers: Department of Genetics and Cell Biology, Univer' Present address: CSIRO Division of Entomology, Canberra, A.C.T. Genetics 143: 1321-1337 (July, 1996) coadapted for the susceptible allele (CLARKE and MCKENZIE 1987; MCKENZIE and CLARKE 1988). Continuing selection with diazinon brought the resistance allele frequency close to fixation and provided the necessary conditions for coadaptation of the resistant genome ( ROUSH and MCKENZIE 1987). Mutation of a second gene, ModiJier ( M ) , resulted in a dominant increase of the fitness of resistant flies such that, in the absence of diazinon, susceptible and resistant flies had equal fitness ( MCKENZIE et al. 1982; MCKENZIE and GAME 1987). Coincident with the change in fitness is a decrease in the level of asymmetry of resistant flies, returning asymmetry to the level of susceptible flies (CLARKE and MCKENZIE 1987; MCKENZIE and The modifier of fitness and asymmetry maps genetically to the region of the Scalloped wings locus, a gene proposed to be the homologue of the Drosophila melanogaster Notch gene (FOSTER et al. 1981 ) . This connection between Notch and Scl was originally made on the basis of similar adult wing phenotypes and recessive lethality (MADDERN et al. 1986) and is further supported by conservation of linkage groups such that both genes map close to the respective white genes of D. melanogaster and L. cuprina (FOSTER et al. 1981; WELLER and FOSTER 1993). The wellcharacterized Notch gene of D. melanogaster has a role in the determination of cell fate throughout development in a variety of tissues (reviewed by ARTAVANIS-TSAKONAS et al. 1991; FORTINI and ARTAVANIS-TSAKONAS 1993; MUSKAVITCH 1994) . The Notch gene product (Notch) is a transmembrane protein thought to CLARKE 1988). 1322 A. G. Davies rt al. mediate a cell-cell signal through protein-protein interactions at its extracellular and intracellular domains ( FEHON et al. 1990; REBAY et al. 1991; LIEBER et al. 1992; DIEDERTCH et al. 1994). An example of Notch function occurs during embryogenesis where Notch is required in the neuroectoderm to allow undetermined cells to adopt either a neural or epidermal fate. Loss of function of the Notch gene in the neuroectoderm results in the formation of an excess of neural precursor cells at the expense of epidermal precursors ( POULSON 1950; LEHMANN et al. 1983). This is known as the "neurogenic" phenotype, a phenotype that is common to mutations in any of the neurogenic genes, a group of genes that appear to act in many or all of the developmental pathways involving Notch ( LEHMANN et al. 1983) . The genetics of the Notch locus is complex, there are multiple allele classes that display pleiotropic combinations of dominant and/or recessive adult phenotypes. There are two classes with dominant adult phenotypes, the Notch ( N ) alleles, which are hypomorphic mutations, and the hypermorphic A h p t e x (Ax) alleles ( LINDSLEY and ZIMM 1992) . In addition to the recessive embryonic lethality associated with the neurogenic phenotype described above, the Nalleles display dominant adult wing notching, wing vein thickening and bristle abnormalities ( LINDSL.EY and ZIMM 1992). The Ax alleles show dominant wing vein gaps and bristle loss. These Ax alleles can themselves be divided into three classes, those that are recessive lethal ( 1-Ax) and two classes that either enhance, ( E ( N ) ) or suppress (Su ( N ) ) the wing notching phenotype of a N allele when in heteroallelic combination (FOSTER 1975; POKTIN 1975; LINDSLEY and ZIMM 1992). Here we examine the relationship between Scl and Notch at molecular, phenotypic and genetic levels and show that the two genes are homologous. A portion of the Scl gene is cloned by hybridization at low stringency using a fragment of the Notch gene as a probe. Comparison of the deduced gene products of the two genes reveals very high conservation. The genetic and phenotypic similarities between Notch and Scl are investigated by analysis of the recessive lethal and dominant adult phenotypes of loss-of-function Scl mutations and by isolation and characterization of two Abruptex-like mutations of Scl. These studies reveal functional homology between Notch and Scl. We also investigate the possibility that the modifier of fitness and asymmetry of diazinon resistance is an allele of the Scl gene. Two incompletely penetrant alleles of Scl are used as a sensitive test for interactions between M and Scl and between Rop-1 and Scl. These experiments support the allelism of M with Scl. Based on the functional homology between Notch and Scl we propose a mechanism for the observed developmental effects of the R@-1 mutation and the suppression of those effects by M. MATERIALS AND METHODS Strains used: All stocks were maintained at 27 ? 1" on standard laboratory medium unless otherwise specified. six mutations of Scl used in this study, Scl', T ( 3;6) Scl', Sei', Sc14, D f ( 3 ) w-Sc12 and D f ( 3) w-Scl', were induced by gamma irradiation and kindly supplied by G. FOSTER and G. WEILER (CSIRO Entomology, Canberra). The Sc15 mutation was induced with EMS previously in this laboratory. In the course of this study the Scl""' and Scl'"2 alleles were EMS-induced and the Sc16 and Scl' mutations were induced by gamma irradiation. The In( 3LR) 7 + 10, Scl' TU strain carries both the Scl' allele and the rusty body ( T U ) mutation on a doubly inverted chromosome 111 where the inversions overlap ( FOSTEK rt al. 1991). Two mutations of the white eyes gene, w and w'" ( white mustard eyes) ( MADDERN et al. 1986), were used in mutagenesis, mapping or as chromosome markers in recessive lethality or complementation crosses. Strains homozygous for the ru mutation or carrying a multiply-marked chromosome III [crooked bnstles ( c k ) , w"' and ru ( MAUDEFW et al. 1986) ] over the D f ( 3 ) wSc13 chromosome were used in mutagenesis experiments. Other strains used for testing combinations of ModiJier and Scl alleles were doubly homozygous for ModiJler ( M ) or wildtype ( + ) and diazinon resistance ( Rop-I) or susceptible ( + ) alleles ( M / M ; +/+, +/+; + /+ , M / M ; Rop-I /Kop-I and + / +; Rop-l/ Rop-1) in comparable genetic backgrounds. The origin of these strains is described in MCKENZIE and GAME (1987) and MCKENZIE and CIARKE (1988). Molecular biology: A standard wild-type L. cuprina genomic library (gift from TONY HOUTLIS, Biochemistry Department, the Australian National University) in hgtll was screened for Notch-hybridizing clones. Plaque lifts were hybridized with the "P-a-dATP nick translated Notch probe at 42" overnight in a solution of 30% formamide, 6X SSC, 5X Denhardts, 0.5% SDS, 200 pg/ml salmon sperm DNA. Filters were washed in 2X SSC, 0.1% SDS at 65" for 1 hr and films exposed at -70". Restriction fragments were subcloned into pBlueScript (pBSK+; Stratagene) for DNA sequencing. The doublestranded templates were amplified on a DNA Thermal Cycler (Perkin Elmer Cetus) with T3 or T7 dye primers (Applied Biosystems) , according to the instructions of the manufacturers. Sequencing gels were run and the data were read by an AB1 373A automated sequencer (Applied Biosystems) . Sequence analysis was performed using the SeqEd v1.0.3 program (Applied Biosystems) . Cytology: Trichogen cell polytene chromosomes were prepared from pupae whose parents were heterozygous ( Scl/ + ) for each Scl allele. In the case of Scl '" ' , Scl ix2 , Scl" and Scl' the parents were Scl + / + 7u so that the white tyrd non-Scalloped progeny could be excluded as some pigmentation of the eye is apparent in the pupae at the time of dissection. Chromosome preparation methods follow those of FOSTER et al. ( 1976) and BEDO ( 1982). Chromosome handing patterns were examined with phase contrast microscopy and compared with the maps of FOSTER et al. ( l98Ob). In situ hybridization to polytene chromosomes: The hybridization procedure followed the nonradioactive protocol of ENGFLS rt nl. (1986) with modifications to the chromosome denaturation step suggested by BEDO and HOMTEILS ( 1987). The same chromosome spreads were photographed before and after in situ hybridization; this allowed the position of any hybridization to be analyzed accurately as the morphology of a chromosome was more defined before the hybridization process. W i g preparation: Wings were cut from 1-2 day old flies, dehydrated briefly in 70 and 100% ethanol then mounted in Euparal (GB1 Labs), covered with a glass coverslip and flattened with light pressure. Scanning electron microscopy: Flies to be examined by Notch Homologue in L. cuprina 1323 scanning electron microscopy were desiccated at room temperature with minimal handling. After mounting-on stubs with a silver glue, the flies were coated with 140A of gold and examined at a magnification of X302-890 with a Philips SEM505 scanning electron microscope. Embryo fixation: Embryos ( 7 ? 0.5 hr old at 27") were fixed and devitellinized according to the paraformaldehyde fixation method of TAUTZ and PFEIFLE (1989). Subsequent anti-horse radish peroxidase (HRP) staining was according to steps 12-24 of protocol 96 of ASHBURNER (1989). The stained embryos were mounted in Pro-Cure 812 (Probing and Structure) according to protocol 96 of ASHBURNER (1989). The nature of the homozygous Scl lethality was investigated by analyzing embryonic progeny from parents heterozygous for each Scl allele such that one in four embryos were expected to be homozygous for the mutation at Scl. Cuticle preparation: Cuticles were prepared from embryos ( 11.75 ? 0.25-hr old at 27") by treatment with glycero1:glacial acetic acid (1:4) and mounting in Hoyer's Mountant according to the method of WIESCHAUS and NUSSLEIN-VOLHARD (1986). Mutagenesis: The method for EMS mutagenesis followed that of SMWH et al. ( 1992). Treated males were mated with c k w'" TU/ + Df( ?) w-Sc13 + virgin females. Mutagenesis using gamma irradiation was performed on 5day-old males of the ru/ ru genotype using an Eldorado 6 "Co source. A dose of 2800 rad was administered at 184 rad/min. Mutagenized males were crossed with w / w virgin females. For both types of mutagenesis, putative Scl mutants were identified following visual screening of the F, progeny. Testing Scl alleles for recessive l thality: Each of the preexisting Scl alleles were tested for recessive lethality using the cross diagrammed in Figure 1. The mutation SclAX2 was tested for lethality in a similar manner but utilized the more closely linked w"' mutation in place of the ru marker. Under this crossing regime, if a Scl allele is recessive viable, then the Scl/ Scl homozygotes can be identified as ru homozygotes ( Scl TU/ Scl T U ) . Thus an absence of rusty Scalloped flies would be an indication of recessive lethality. Complementation tests with Scl': The crosses shown in Figure 1 utilized a multiply-inverted chromosome I11 balancer, In ( ?LR) 7+ IO, which carries the Scl' allele and a mutation in ru (FOSTER et al. 1991). The inversions include the ru locus but not the Scl region. However, this balancer chromosome has been shown to significantly reduce the frequency of viable crossovers in the interval between Scl and ru (FOSTER et al. 1991). For each Scl allele tested, heterozygous male flies were mated with females bearing the double-inversion chromosome. Recombination occurs at an extremely low frequency in male L. c u p h a (FOSTER et al. 1980a), therefore recombination should be virtually eliminated between Scl and ru ensuring that only Scl homozygotes could be homozygous for ru. Temperature shift: Temperature-sensitive periods were determined by raising developing flies at 20, 25 and 27". For each temperature, white pre-pupae ( 13-15 days old at 20", 7-9 days old at 25" and 6-8 days old at 27") were either kept at that initial temperature or transferred to the alternative temperatures. Scl/ Modijkrinteractions: The effect of the modifier on the phenotype and asymmetry associated with Sc15 and SclAx' was determined by performing crosses in pairs where each Scl allele was crossed to Modajier (M/ M) and non-Modijier ( + / + ) strains in either a susceptible ( + / + ) or resistant ( R e 1 / RopI) background. Comparisons can then be made between the proportion of progeny expressing the Scl phenotype to those which do not, in the presence or absence of M. For instance, Sc15/ +;+ / + flies were crossed to M/ M;+ / + and + / +;+ / + strains at the same time so that the progeny developed under similar environmental conditions (the genetic notation used assumes that the modifier is an allele of S c l ) . The genotypes of the parents and their progeny were not revealed to the scorer to prevent bias. The effect of M on Sc14 phenotype and asymmetry was also examined but only in a susceptible background using similar pairs of crosses. Comparison between the paired crosses allows an examination of the effect of the RopI allele, if any, on the expression of the Scl phenotype. Differences in the proportion of flies expressing the Scl phenotype to those which do not can be attributed to the presence or absence of the resistance allele in flies with an otherwise identical genotype. Asymmetry: The methods followed those of CLARKE and MCKENZIE (1987) and MCKENZIE and CI.ARKE ( 1988). Asymmetry of a fly was estimated from the absolute difference in bristle count between the left and right sides of the frontal head stripe, the outer wing margin and the &+5 wing vein. The asymmetry of at least 20 flies was scored to generate a mean asymmetry value for each genotypic combination. In crosses involving Scl', asymmetry was measured for 20 Sc15/ + flies each for the two phenotypic classes, expression and nonexpression of wing notching.