GENEn~ Mitochondrial DNA Variation Among Muscidifurax spp. (Hymenoptera: Pteromalidae), Pupal Parasitoids of Filth Flies (Diptera) DAVID B. TAYLOR, RICHARD D. PETERSON II, ALLEN L. SZALANSKI,l ANDJAMES J. PETERSEN Midwest Livestock Insects Research Laboratory, USDA-ARS, Department of Entomology, University of Nebraska, Lincoln, NE 68583 Ann. EntomoI. Soc. Am. 90(6): 814-824 (1997) ABSTRACT Polymerase chain reaction-restriction fragment length polymorphism (PCR- RFLP) and sequencing analyses were used to characterize an amplicon of -625 bp in 4 of the 5 nominate species of Muscidijilrax Girault & Sanders,pupal parasitoids of muscoid Hies. A single polymorphic nucleotide site was observed among 2 samples of M. raptOr Girault & Sanders. No sequence variation was observed among 3 samples of M. raptorellus Kogan & Legner. The sequence of M. uniraptor Kogan & Legner was identical to that of M. raptOrellus. Nucleotide divergence among the Muscidifurax spp. ranged from 0.14 to 0.18 substitutions per nucleotide. Muscidijilrax zaraptor Kogan & Legner exhibited multiple haplotypes, 2 of which were char- acterized by sequencing and 4 others by PCR-RFLP. The sequencedhaplotypes differed by 0.08 nucleotide substitutions per site. Restriction site analysis indicated that nucleotide divergence ranged from 0.03 to 0.10 among all 6 haplotypes. Analysis of progeny from individual females indicated that the observed variation in M. zaraptOr was caused by multiple haplotypes within individuals rather than differentiation among individuals. These results bring to question the specific status of M. uniraptOr and indicate that the genus is native to the Western Hemisphere, and not introduced with their primary host, Musca domestica L., as previously proposed. Heteroplasmy and translocation of a portion of the mitochondrial genome to the nuclear genome are discussed as possible causes for the variation observed in M. zaraptOr. KEY WORDS Muscidijilrax, polymerase chain reaction-restriction fragment length polymor- phism, mitochondrial DNA, phylogeny WASPSIN THEgenus Muscidifurax are pupal parasi- species known only from the island of Puerto Rico toids of muscoid flies, especially the house fly, Musca (Kogan and Legner 1970). domestica L., and the stable fly, StO11WXYScalcitrans The geographic origins and phylogenetic rela- (L.), and are among the most promising biological tionships of Muscidifurax are unclear. Kogan and control agents for these flies in the confined live- Legner (1970) proposed 2 alternatives for the ori- stock environment (Miller and Rutz 1990,Petersen gins of the genus. First, they originated in the Ethi- et al. 1990). Muscidifurax includes 5 species. Mus- opian region and were introduced to the New World cidifurax raptor Girault & Sanders,the most wide- along with house flies, or 2nd, they are native to the spread,is found throughout the temperate and semi- New World and have secondarily adapted to house tropical regions of the world (Kogan and Legner flies. Kogan and Legner conclude that the "remark- 1970). The remaining 4 species are limited to the able preference of Muscidifurax spp. for house flies New World. Muscidifurax zaraptor Kogan & Legner as compared to native Nearctic flies" indicates an is found sympatrically with M. raptor in western Old World origin for the genus. Legner (1983) in- North America (Kogan and Legner 1970, Lysyk dicates that the dependence of Muscidifurax upon 1995), M. raptoroides Kogan & Legner and M. rap- the "barnyard" environment outside of Africa is torellus Kogan & Legner are found allopatrically in further evidence that the genuswas not native to the Central and South America, respectively (Kogan New World. However, to accept an Old World or- and Legner 1970). Two forms of M. raptorellus have igin of the genus and account for the 4 species been reported, one solitary and the other gregarious endemic to the New World, one must accept a very (Kogan and Legner 1970, Legner 1988). Muscidi- rapid rate of speciation following their introduction. furax uniraptor Kogan & Legner is a parthenogenic Despite the interest in these species for biological control of filth flies and the questions concerning their geographic origin, little work hasbeen done on 1 Department of Plant Pathology, University of Nebraska, Lin- the population genetics and genetic structure of coln, NE 68583. Muscidifurax. Propp (1986) used allozymes to ex- November 1997 TAYLORET AL.: mtDNA VARIATIONIN Muscidifurax SPP. 815 Table1. Strain. or MU-'c~ro% .pp. u.ed in thi. .tudy Date Source" Species Origin collected M. raptor Nebraska 1995 MLIRL New York Unknown Cornell University M. raptorellus Chile Unknown Legner via MLIRL Nebraska 1991 MLIRL 5Obp Peru Unknown Legner via MLIRL ladder M uniraptor Puerto Rico Unknown Rochester University M zaraptor Nebraska 1991-1993 MLIRL Nebraska 1995 MLIRL .MLIRL, Midwest Livestock Insect Research Laboratory, USDA-ARS, Lincoln, NE; Legner via MLIRL, originally collected by E. F. Legner and transferred to MLIRL in 1989. arnine several populations of M. raptor and M. zarap- tor. He found relatively low levels of differentiation arnong geographic isolates of the 2 species, but a high level of differentiation was observed between species. Antolin et al. (1996) used RAPD-PCR to examine 3 Muscidifurax spp. They were able to dif- ferentiate the species and associate the gregarious North American Muscidifurax sp. with M. raptorel- ius, but indicated they were unable to explore phy- logenetic relationships because of the nature of the RAPD-PCR data. The purpose of this study was to develop molec- Fig. 1. Intact amplicon from 4 Muscidifurax spp., A. ular diagnostic characters and exarnine genetic dif- mellifera, and C. mace/Lanaon 2.5% MetaPhor agaroscgel. ferentiation among Muscidifurax spp. A region of Primers were 5' ATACcrCGACGTfATfCAGA 3' and 5' the mitochondrial genome, =625 bp, including parts TCAATATCATfGATGACCAAT 3'. of the cytochrome oxidase I (COI) and II (COIl) genes as well as the entire tRNA leucine (tRNA1CU) gene was examined by polymerase chain reaction- extracted with the chloroform-phenol technique as restriction fragment length polymorphism (PCR- outlined in Taylor et al. (1996). RFLP) and sequencing. Representative samples of A region of mtDNA, =625 bp long, was amplified 4 of the 5 species in the genus (sarnples of M. using the primers 5' ATACCfCGACGTTATTC- raptoroides were not available) were included in AGA 3' (= 52792; Bogdanowicz et al. 1993) and 5' this study. TCAATATCATTGATGACCAAT 3' (K. Pruess, personal communication). The 5' ends of thcsc primers were located at bp 2773 and 3400 of thc Materials and Methods Drosophila yakuba mtDNA map (Clary and Wol- Samples. Muscidifitrax samples were obtained stenholme 1985), respectively. from established colonies (Table 1). The Chile and For amplification, 1 ILl of sample DNA was added Nebraska samples of M. raptorellus represented gre- to a reaction mixture containing 2.5 ILl of rcaction garious populations and the Peru sample was soli- buffer (Perkin-Elmer, Norwalk, cf), 2 ILl of dNTP tary. The Apis mellifera lingustica L. control sample mix (10 mM each-dATP, dTTP, dCTP and dGTP) , was collected from an apiary in Lincoln, NE, and the 1 ILl of each primer (20 mM), 1.0 unit ofTaq poly- Cochliomyia macellaria (F.) samples were from a merase (Perkin-Elmer) and deionized water to a laboratory colony originating in Fargo, ND. Mus- volume of 25 ILl. Amplifications were done in a cidifitrax colonies were maintained on freeze-killed Perkin Elmer Cetus Model 9600 thermo cycler pro- M. domesticapupae (Petersen and Matthews 1984). grammed for 35 cycles of 92°C for 1 min, 55°C for 1 Isofemale lines were initiated by isolating indi- min, and 72°C for 1 or 2 min. Amplification products vidual females (presumably mated) from the main were stored at 4°C. colony and placing them in small plastic cups (2 cm Restriction Endonuclease Digests. Twenty-seven diameter, 2 cm high) with =150 freeze-killed fly restriction enzymes-Alu I, Apo I, Ase I, Ava I, Ban pupae. Progeny emerged in 3 wk. Specimens were II, Bfa I, Bsr I, Dde I, Dpn II, Dra I, EcoR I, EcoR V, stored at -80°C. Hae III, Hinc II, Hind III, HinfI, Hpa I, Mse I, Msp I, Voucher specimens from each of the Muscidifitrax Pvu II, Rsa I, Sac I, Sau96 I, ScrF I, Ssp I, Taq I, and samples have been placed in the collection of the Xba I-were screened. Digests for polyacrylamide University of Nebraska State Museum, Lincoln. gel electrophoresis (PAGE) were done using 1.5 ILl DNA Extraction and Amplification. Pools of 5-25 ofPCR product, 0.2 ILl enzyme (New England Bio- waspswere used for each DNA extraction. DNA was labs, Beverly, MA), IX buffer (New England Bio- 816 ANNALS OF mE ENTOMOLOGICAL SOCIETY OF AMERICA Vol. 90, no. 6 Fig. 2. Apo I and Ssp I digests of Muscidifurax spp. on 2.5% MetaPhor agarosegels. labs) and water to bring the volume to 5 ILl in 200-ILI onto a 10% acrylamide gel (IX Tris:borate:EDTA tubes. For agarose gel electrophoresis, all quantities (TBE), 0.5% PhotoHo-200 [Kodak, Rochester, NY], were increased 2.5 times. Sampleswere incubated at 0.15%TEMED, and 0.05% ammonium persulfate). A 3rC for 3-16 h and stored at 4°C until further molecular size standard, pGEM (Promega, Madison, analysis. WI), was included on each gel. Hoefer (Hoefer Electrophoresis. For PAGE, 1.5 ILl of loading Scientific Instruments, San Francisco, CA) SE 600 buffer (10% Ficoll400 [Sigma,St. Louis, MO], 0.25% electrophoresis units with gels (16 by 20 cm by 0.75 Bromophenol Blue [Sigma], 50 mM EDTA, 10 mM mm) and 28 well combs were used for electrophore- Tris-HCI pH 7.5) was added to the 5-ILI digest prod- sis.Gels were run in 1 X TBE buffer at a constant 300 uct. The entire digest product (6.5 ILl) was loaded V (15 mAl gel) for 1.5 h at 20°C. Gels were stained Table 2.