1 THE ECOLOGY OF STABLE FLIES (DIPTERA: MUSCIDAE ) ASSOCIATED WITH EQUINE FACILITIES IN FLORIDA By JIMMY BRUCE PITZER A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2010
2 2010 Jimmy Bruce Pitzer
3 To my family and friends for their overwhelming support
4 ACKNOWLEDGMENTS I greatly acknowledge the Deployed Warfighters Program of th e Armed Forces and the Matching Deans Teaching Assistantship provided by the University of Florida Institute of Food and Agricultural Sciences, for the funding necessary for me to pursue a Ph.D. in entomology and nematology. Great appreciation is also ex tended to my major advisor, Dr. Phillip Kauf man for his wealth of knowledge, unending support and patience during my tenure as his graduate student. His financial support has allowed me to participate in several student competitions at both national and regional entomological conferences, as well as allowed the purchase of all supplies needed for the completion of my dissertation projects. I consider him both colleague and friend. Appreciation is also extended to the other members of my committee, Dr. Jerry Hogsette, Dr. Chris Geden, Dr. James Maruniak, and Dr. Saundra TenBroeck. They have all provided continued interest, support, guidance, and insight, and furthermore have been instrumental in my success. Special thanks are extended to Dr. Ale Maruniak for her undying patience and support while instructing me in the area of molecular biology. She was never too busy to lend an ear or help with my project in any way. I would like to thank Dr. Saundra TenBroeck for her assistance in acquiring permission t o use the field sites for my project. W ithout them my research would be sparse at best. Dr. Chris Geden is also extended thanks for his time taken to assist me in the collection of stable flies from which my colony was started, as well as in the design of my parasitoid projects. The wealth of knowledge in the area of veterinary entomology owned by Dr. Jerry Hogsette is insurmountable. He was never too busy or too far away to answer any question I might have had on the subject, and for this I am
5 truly g rateful. I am grateful to Dr. James Maruniak, who was instrumental in my understanding of molecular biology and always had time to help me in the preparation for an upcoming student competition. Appreciation is also extended to Dr. Michael Scharf for his assistance with the toxicological aspects of my research. I do not think any person completing a degree in this department has not truly been blessed by the efforts of Debbie Hall. She is tremendous at her job and keeps all of us in line and on track. I am truly grateful for all of the assistance she has provided to me during my stay. Drs. Jerry Butler and Don Hall must also be acknowledged for their continued support and interest in my projects as well as with my hobby as an amateur lepidopterist. Th ey always had time to chat and I will miss our conversations greatly. I would like to extend appreciation to the landowners who allowed me to conduct my research on their propert ies These people included Roy Lerman, Rod Allen, Chris Cooper, and Charlotte Weber, with special thanks to Vicki Peck, Chuck Vadnais, and Joss Cooper who assisted as farm contacts and were always interested in my progress. I was graciously allowed to work on tightly -run farms, some having very expensive champion horses; the trust and opportunity they offered the University of Florida and me did not go unnoticed. I would like to thank the crew at the Veterinary Entomology Laboratory for all of their efforts towards the completion of my dissertation projects. Mike Bentley and Dr. P eter Obenauer have become two of my best friends; their assistance with the difficult transition I had upon arriving in Florida, as well as in the countless hours of assistance they provided in the field and the laboratory will never be forgotten. Thanks are also
6 extended to Matt Geden, Chantel Carter, and Becky Hamel for assistance with counting flies and setting up any experiments I may have been attempting. Thanks are extended to Lois Wood for the many trips to town to purchase any supplies I needed fo r my project. Dr. Raj Mann was instrumental in my completion of toxicological assays and I am truly indebted to him for his assistance with them as well as any statistical interpretations I needed. I greatly appreciate the friendship and camaraderie of those in the Urban Entomology Laboratory, in particular my jogging mates Joe DiClaro and Dr. Philip Koehler. I will miss our conversations and runs around Lake Alice tremendously. Tremendous thanks are extended to all of my family and friends, in particul ar my parents, Jimmy and Carol, and my grandparents, Jim, Seu, and Lou for their continued love, support, and interest. It is my hope that I have made them all proud. In particular is the love and support of Sonia. Without her, none of this was possible The value of her assistance in my projects as well as those of the laboratory has been immeasurable to both the lab and me She has made every effort to keep my stress level low during some particularly trying times, and for this, I am truly grateful to her. Last but not least, is my buddy, Brodie. That 6 pound ball of hair will probably never know the totality of the positive effect he has had on both our lives.
7 TABLE OF CONTENTS page ACKNOWLEDGMENTS ...................................................................................................... 4 LIST OF TABLES ................................................................................................................ 9 LIST OF FIGURES ............................................................................................................ 10 ABSTRACT ........................................................................................................................ 12 CHAPTER 1 LITERATURE REVIEW OF THE STABLE FLY STOMOXYS CALCITRANS (L.) .. 14 Life History .................................................................................................................. 14 Economic Importance ................................................................................................. 16 Control ......................................................................................................................... 18 Insecticides and Traps ......................................................................................... 18 Bio logical Control ................................................................................................. 20 Insecticide Resistance ................................................................................................ 24 Distribution and Dispersal of Muscoid Fly Insecticide Resistance ..................... 25 Mechanisms of Resistance .................................................................................. 27 Research Objectives ................................................................................................... 31 2 SEASONAL ABUNDANCE OF STAB LE FLIES AND THEIR PUPAL PARASITOIDS (HYMENOPTERA: PTEROMALIDAE) ASSOCIATED WITH EQUINE FACILITIES .................................................................................................. 32 Introduction ................................................................................................................. 32 Materials and Methods ............................................................................................... 35 Results ........................................................................................................................ 39 Discussion ................................................................................................................... 42 3 THE ABILITY OF SELECTED PUPAL PARASITOIDS (HYMENOPTERA: PTEROMALIDAE) TO LOCATE STABLE FLY HOSTS IN EQUINE HUSBANDRY -GENERATED SUBSTRATES ........................................................... 57 Introduction ................................................................................................................. 57 Materials and Methods ............................................................................................... 59 Results ........................................................................................................................ 63 Discussion ................................................................................................................... 64 4 HOST BLOOD MEAL IDENTIFICATION OF STABLE FLIES COLLECTED FROM FLORIDA EQUINE FACILITIES USING A MULTIPLEX POLYMERASE CHAIN REACTION ..................................................................................................... 76
8 Introduction ................................................................................................................. 76 Materials and Methods ............................................................................................... 79 Results ........................................................................................................................ 84 Discussion ................................................................................................................... 86 5 PERMETHRIN RESISTANCE STATUS OF THE STABLE FLY IN FLORIDA: A CRITICAL UPDATE USING LABORATORY SELECTIONS AND FIELD EVALUATIONS ......................................................................................................... 103 Introduction ............................................................................................................... 103 Materials and Methods ............................................................................................. 105 Results ...................................................................................................................... 109 Discussion ................................................................................................................. 111 6 IMPLICATIONS AND FUTURE DIRECTIONS FOR RESEARCH REGARDING STABLE FLIES AND EQUINE FACILITIES ............................................................ 121 APPENDI X A UNIVERSITY OF FLORID A HEALTH CENTER INST ITUTIONAL REVIEW BOARD #342 2008 ................................................................................................... 126 B UNIVERSITY OF FLORID A INSTITUTIONAL ANIM AL CARE AND USE COMMITTEE #200801760 ....................................................................................... 132 LIST OF REFERENCES ................................................................................................. 133 BIOGRAPHICAL SKETCH .............................................................................................. 145
9 LIST OF TABLES Table page 2 -1 Total pupae collected from four Florida equ ine facilities between December 2007 and 2009. ...................................................................................................... 50 2 -2 Mean percent parasitism rates and mean percent Spalangia spp. composition recovered from stable fly and house fly pupae. ............................... 51 3 -1 Analysis of variance (ANOVA) F values for interspecies percent whole pupae, percent parasitism, and percent parasitoid-inducedmortality (PIM) of stable fly pupae. ..................................................................................................... 70 3 -2 Analysis of variance (ANOVA) F values for and intraspecies percent whole pupae, percent parasitism, and percent parasitoid-inducedmortality (PIM) of stable fly pupae ...................................................................................................... 71 4 -1 Primer sequences targeting the cytochrome b region of the mitochondrial genome of mammals. ............................................................................................. 93 5 -1 Permethrin susceptibility for several stable fly strains evaluated using insect icide -treated glass jars. .............................................................................. 117 5 -2 Permethrin susceptibility for several stable fly strains evaluated using topically applied insecticide. ................................................................................ 118 5 -3 Concentrations used and stable fly mortality results at each permethrin selection. ............................................................................................................... 119
10 LIST OF FIGURES Figure page 2 -1 Corrugated alsynite sticky trap mounted at a height of 90 cm near horse pasture, Ocala, Florida. .......................................................................................... 52 2 -2 Plastic souffl cup containing pupae individually placed into #0 gelatin capsules and held at 26 C 12:12 LD, and 70% RH for parasitoid emergence. ............................................................................................................. 53 2 -3 Mean weekly temperatures and accumulated monthly precipitation for Ocala, Florida, occurring between November 2007 and December 2009 ....................... 54 2 -4 Weekly stable fly trap collections from four equine facilities during a.) November 20072008 and b.) November 2008-2009.. ......................................... 55 2 -5 Relationship of adult stable fly trap captures for a given date, and stable fly pupae collected 2 wk prior.. ................................................................................... 56 3 -1 Release chamber (26 cm diameter, 9 cm height) filled with 3.7 L (7 cm) soiled horse bedding. ............................................................................................. 71 3 -2 Plastic souffl cups (120 ml) containing 10 ml maple wood chips used to assess parasitism when stable fly pupae were freely accessible. ....................... 72 3 -3 Comparison of three parasitoid species host attack parameters expressed as mean percent whole pupae, percent successful parasitism, and percent parasitoid -induced mortality (PIM) exposed to stable fly pupae. .......................... 73 3 -4 Evaluation of a) Muscidifurax raptorellus b) Spalangia cameroni and c) Spalangia endius searching ability on host attack rates expressed as percent whole pupae, percent successful parasitis m, and percent parasitoid-induced mortality (PIM).. ...................................................................................................... 74 4 -1 Pressure induction of a fecal droplet from a blood-fed stable fly. ........................ 94 4 -2 Feeding chambers used for stable fly blood host blood meal identification time course analysis .............................................................................................. 95 4 -3 Stable fly feeding chambers with attached cap from a 1.5 ml microcentrifuge tube fil led with 200-l host blood. .......................................................................... 96 4 -4 Pastures enclosing horses and cattle within 3 km of each equine facility. .......... 97 4 -5 Agarose gel results of time course blood meal analysis at 0 hr and 8 hr post blood -feeding on dog, horse, cattle, human, and mixed blood types .................. 99
11 4 -6 Agarose gel results of time course blood meal analysi s performed at 16 hr post blood -feeding.. ............................................................................................... 99 4 -7 Agarose gel results of time course blood meal analysis performed at 24 hr post blood -feeding.. ............................................................................................. 100 4 -8 Mean percent blood type composition determined by stable fly blood meal identification using a multiplex polymerase chain reaction. ................................ 101 4 -9 Mean percent blood type com position of farms determined by stable fly blood meal identification using a multiplex polymerase chain reaction.. ...................... 102 5 -1 Mean percent survival of several stable fly strains exposed to diagno stic permethrin concentrations applied to 60 ml glass jars. ....................................... 120
12 Abstract Of Dissertation Presented To The Graduate School Of The University Of Florida In Partial Fulfillment Of The Requirements For The Degr ee Of Doctor Of Philosophy THE ECOLOGY OF STABLE FLIES (DIPTERA: MUSCIDAE ) ASSOCIATED WITH EQUINE FACILITIES IN FLORIDA By Jimmy Bruce Pitzer May 2010 Chair: Phillip Kaufman Major: Entomology and Nematology Beginning in November 2007 and continuing until December 2009, a series of field and laboratory -based studies were conducted to increase our understanding of stable fly ecology associated wit h Florida equine facilities. Weekly stable fly surveillance was conducted at four equine facilities near Ocala, Florida using alsynite sticky traps for adults and by searching larval developmental sites for pupae. Although stable flies were collected thr oughout the year, most were captured between January and April. Correlation analysis suggested that adult stable fly presence to predict future stable fly breeding cannot be used for parasitoid releases on equine farms In contrast to studies of other livestock installations, such as cattle and poultry, 99.9% of all parasitoids collected were Spalangia spp. consisting of Spalangia cameroni Perkins (56.5%) Spalangia nigroaenea Curtis (34.0%) Sp alangia endius Walker (5.8%) and Spalangia nigra Latreille (3.7%) In addition, l aboratory evaluations assessing the ability of S. cameroni S. endius and Muscidifurax raptorellus Kogan and Legner to locate and attack stable fly hosts in a field-collect ed fly breeding substrate, suggest that Spalangia spp. are more suited to searching in habitats created by equine husbandry.
13 To assess potential movement between farms, a multiplex polymerase chain reaction (PCR) was de veloped to identify the blood meal host source of stable flies specifically those taken from cattle, horse, human, and dog hosts. O ver 60% of all blood meals identified were fro m stable flies that had previously fed on cattle, with less than 30% having fed on horses In addition, pastures enclosing cattle were up to 1.5 km from stable fly collection areas within each equine facility. Stable fly susceptibility to the commonly used pyrethroid, permethrin, was determined to assess possible resistance development in this pest. Stable fly survival after exposure to permethrin was up to 15-fold higher in f ield-collected flies than in those of a susceptible colony. Diagnostic concentration evaluations demonstrated 40 and 10% stable fly survival to permethrin residues of 3 X and 10X the LC99 of a susceptible strain. In addition, a laboratory -generated, permethrin-resistant stable fly strain demons trat ed 15 -fold greater resistance expression over that observed for the parental field -collected stable flies, after only five permethrin selections.
14 CHAPTER 1 LITERATURE REVIEW OF THE STABLE FLY, STOMOXYS CAL CITRANS (L.) Life History The stable fly, Stomoxys calcitrans (L.), is a haematophagous muscoid fly that feeds on hu mans and a wide variety of other animals. Although both sexes feed on blood, female stable flies require blood proteins for egg development Anderson and Tempelis (1970) found that female stable flies required three to seven blood meals for oviposition to occur. Stable flies feed on an array of hosts and information regarding the reproductive potential of different host blood types h as been investigated. DuToit (1975) found that egg production was highest in female stable flies fed cattle blood when compared to those that fed on blood of the pig. A study by Sutherland (1978a) confirmed the increased nutritional value of bovine blood to the stable fly, concluding that the blood of herbivores including donkeys, sheep, and horses resulted in higher egg production and survivability than that of omnivores such as pigs and dogs. Stable flie s that fed on chicken blood were unable to produce eggs. Anderson and Tempelis (1970) reported that on average, the time required to fully digest a blood meal was 2436 hours when stable flies were fed on citrated human blood and maintained at temperatures of 20-21 C. Digestion in stable flies fed cattle blood was longer, taking 46 and 70 hr when maintained at maximum and minimum temperatures of 21 and 15 C, respectively. Anderson and Tempelis (1970) also noted a decreased digestion time of 10 hr when cattle blood-fed flies were held at 25 C as well as a 2 d decrease in pre oviposition period in stable flies held at 30 C than those held at 22 C. These findings are consistent with a review by Lehane (2005), who proposed a rough average time of 48 hr for complete digestion, but noted that many
15 factors can alter this time frame, including temperature, blood meal size and host type, age, and mating and gonotrophic status. Depending on the host from which a blood meal was taken, female stable flies will enter a 5to 10day pre oviposition period (Sutherland 1978a). After this period, stable flies will seek an appropriate medium on which to oviposit. Females utilize a variety of organic larval media including spilled livestock feeds, manure, compost piles, and other mixtures of decaying organic matter (Meyer and Peterse n 1983, Broce and Haas 1999). Lysyk (1998) reported that the number of eggs produced by a single female varied depending on the temperature at which females were maintained. The highest egg production averaged 26.22 at 30 C, and was lowest at temperatures of 15 and 35 C. Lifetime fecundity was estimated to be <30 eggs for females maintained at 15 or 35 C, with a potential of >700 eggs for females maintained at 25 C. Eggs generally hatch 12 to 24 hours after they are de posited (Foil and H ogsette 1994). Survival of the resultant larvae is most dependent up on the breeding media selected by the female stable fly and the environmental conditions during development such as temperature, humidity, and pH For example, Gilles et al. (2005) demon strated that larval development varied between 13 and 71 days at 30 and 15 C, respectively. Sutherland (1978b) found the dung of various livestock, with the exception of the chicken, as well as several types of vegetable matter, to be suitable for larval development. A study by Boire et al. (1988) also demonstrated larval development in a variety of media including the dung of horses, cattle, swine, and chickens. T hey found that adding bermudagrass hay to manure i ncreased the numbers and weight of survi ving pupae for all manure types except that of cattle. The addition of hay to larval
16 substrates may play a role in increasing microbial fermentation and bacterial populations. Data collected by Romero et al. (2006) suggest that a symbiosis occurs whereby larval development is dependent upon bacterial populations. Furthermore, results indicated that female stable flies are capable of detecting bacterial cues that indicate the suitability of the larval medium. After the development period, larvae purge th e gut and begin wandering throughout the media to locate an adequate pupariation site (McPheron and Broce 1996). The pupal stage generally lasts 57 days, with commencement of eclosion depend ent on temperature. Economic Importance The stable fly is a cosm opolitan pest of humans and an assortment of other animals. The economic losses caused by this pest to livestock producers, especially those of cattle, are well documented. Bruce and Decker (1958) found that monthly reductions in milk and butterfat content averaged 0.7% per fly per cow in response to stable fly pressure. It was also shown that depressed rates of production persisted in the weeks and months after the fly season had ended. Miller et al. (1973) found that despite the annoyance behavior of cattle, tremendous fly loads of 650 -1600 per animal significantly decreased milk production in only one of six trials. The behavioral responses of dairy cattle to stable flies were also monitored in a more recent study conducted by Mullens et al. (2006). Although much time was initially spent fighting stabl e flies with head throwing, stomping, skin twitching, and tail switching, this behavior decreased over time. The results of this study suggest that c attle acclimated to the pain caused by biting flies. It was also concluded that peak stable fly loads of 3.03.5 per leg were not adequate to elicit a loss in production. Although these results are conflicting,
17 they attest to the potential detriment to cattle production systems due to annoyance behaviors caused by stable fl y feeding activity. Stable flies have been shown to affect weight gain and feed efficiency of both confinement and pastured beef cattle. Campbell et al. (1977) observed a 13.2 and 20% decrease in weight gain when growing calves were exp osed to 50 and 100 stable flies, respectively. A subsequent study conducted by Campbell et al. (1987) demonstrated the economic threshold of stable flies to be less than two per front leg when weight gain, feed efficiency, and costs of fly control were considered. Catangui et al. (1993) observed an average reduction of 10.6% in weight gains when Brahman-crossbred and English -cross exotic heifers were exposed to 32 flies per foreleg. Furthermore, Campbell et al. (2001) determined the costs incurred by producers of grazing cattle due to stable fly attacks. In this study, untreated control steers averaged 2.79 more stable flies per leg than those treated with permethrin. Flies on control animals resulted in a 19% weight gain reduction compared to those that were treated with insecticide The overall cost to producers due to attacks by flies was $33.26 per animal, or 2.33 cents per fly. Although stable flies readily attack horses, little information exists regarding the ecology of this pest associated with equine facilities. This may be due in part to the difficulty in ascertaining losses to horse producers. However, expenses to control stable fly populations on these farms are incurred through the purchase of mechanical and chemical control devices, as wel l as commercially available pteromalid pupal parasitoids. Stable flies have also been associated with disease transmission and other disorders in horses. Gortel (1998) states that stable flies are often implicated in
18 arthropod hypersensitivity reactions in horses. These reactions can cause pruritis, which worsens with age, resulting in selling or euthanizing the animal (Fadok and Greiner 1990). Gortel also reports that stable flies often transmit habronemiasis, a condition more commonly known as summer sores. Control Insecticides and T raps Stable fly control, as is the case with other muscoid flies, relies heavily on sanitation and pesticide use. Unlike the horn fly, Haematobia irritans (L.), stable flies only approach hosts two or three times a day t o feed. Furthermore, stable flies prefer to feed on the forelegs of cattle, a site not usually frequented by horn flies (Dougherty et al. 1995, Guglielmone et al. 2004). These behaviors increase the difficulty of controlling stable flies on livestock wit h insecticides. However, Hogsette et al. (1987) state that pesticides applied to cattle do provide some control of stable fly populations, although it is slower and less dramatic than that observed for horn flies. Several chemicals with different applicat ion methods have been evaluated for controlling the stable fly. Mount et al. (1966) found Baygon (Bayer 39007, methyl carbamate), naled, and fenthion to be highly effective when used against stable flies as a thermal fog. They also noted that the concent rations needed for 80% and greater mortality were as little as 1, 2, and 4%, for Baygon, naled, and fenthion, respectively. Hogsette and Ruff (1986) observed that flucythrinate-impregnated ear tags and permethrin ear tapes were effective against stable fl ies for 10 weeks. A more recent study conducted by Guglielmone et al. (2004) involving the abundance of stable flies on cattle treated with oganophosphate-impregnated ear tags, found no significant decrease in stable fly numbers. They concluded that pyre throids, such as permethrin, act as
19 repelle nts to stable flies, whereas organophosphates do not. This may account for the effectiveness of permethrin ear tapes observed by Hogsette and Ruff (1986). Efforts to control stable flies have also been attempted with several traps The evolution of the cross -configuration alsynite trap developed by Williams (1973), to the cylinder type trap developed by Broce (1988) as reported by Hogsette and Ruff (1990), is in itself evidence to the uti lity of this material to attract stable flies Several variations of the alsynite trap have been used to try to enhance collection and control of stable flies. Hogsette and Ruff (1996) found that permethrinimpregnated yarn wrapped in a vertical fashion across the length of the trap did not alter trap attractiveness, and permethrin efficacy lasted for 6 -8 weeks. Other sticky trap configurations, such as the Bite Free and EZ Trap developed by Farnam (Phoenix, AZ), have proven effective in trapping stable flies (Taylor and Berkeb ile 2006). Adhesive -coated plasticized corrugated boards in a range of colors have also been shown to effectively collect stable flies (Cilek 2003). Beresford and Sutcliffe (2006) compared the efficiency of colored corrugated plastic coated with adhesive to that of alsynite. They found white and blue plastic boards captured more flies in general, but attributed the increase to numbers of males and nulliparous females collected. Newer control devices, such as the treated targets developed by Foil and Younger (2006), may prove to be more efficient in controlling stable fly populations than traditional traps. While alsynite is effective in monitoring stable fly populations, the numbers they collect have not been shown to sufficiently control this pest. Treated targets collected 6 -fold the number of stable flies caught with alsynite sticky traps. Foil
20 and Younger (2006) suggested that the attractiveness of blue and black contrasting panels treated with insecticide may be a viable addition to stable fly cont rol programs. Biological C ontrol Increases in filth fly insecticide resistance have only compounded the pressures put on confinement livestock producers by neighboring urban communities to decrease both chemical control and fly populations. This problem h as led many producers to seek alternate control methods, such as the use of hymenopterous pupal parasitoids. Many studies have been conducted to determine the para sitoid species that prefer diff erent filth fly species, as well as their ability to locate and control fly populations. However m ost of this work has been carried out within confinement beef and dairy cattle facilities. A commonly used method to determine the parasitoid species composition in a given regio n has been to collect naturally occurrin g pupae. In addition, sentinel pupae can be placed at these sites to concurrently determine parasitoid preference for a fly species, and parasitism rates after release. Greene et al. (1989) collected house fly, Musca domestica L., and st able fly pupae fr om northwest Florida dairies to determine parasitism rates during the winter months. They found Spalangia cameroni Perkins to be the dominant species, accounting for 76 and 58% of the total parasitized stable flies and house flies, respectively. To a les ser extent, Muscidifurax spp. accounted for 11 and 36% of parasitized stable fly and house fly pupae, respectively. Differences in rates of parasitism were observed depending on habitat and substrate, with greater numbers occurring in silage than in hay or manure. A study conducted by Jones and Weinzierl (1997) found that 93 and 86% of parasitized stable fly and house fly pupae collected, respectively, resulted in a Spalangia spp.
21 Field-collected stable fly and house fly pupae from dairies in California d uring the summer also yielded significantly greater numbers of Spalangia spp. (Meyer et al. 1990). It was noted that Muscidifurax spp. parasitized sentinel house fly pupae more than three times the rate of Spalangia spp. This may have occurred because Mu scidifurax spp. have been shown to forage closer to the substrate surface than Spalangia spp. (Legner and Brydon 1966, Skovgrd and Jespersen 1999). A subsequent study conducted by Meyer et al. (1991) again revealed a greater proportion of Spalangia spp., in particular Spalangia endius Walker, attacking filth fly pupae than any other pteromalid species collected. Results also indicated that S palangia nigroaenea Curtis parasitized more house fly pupae than those of the stable fly. Furthermore, data sugges ted a preference for stable flies by S. cameroni although the difference in parasitism rate was attributed to the decomposing straw or hay substrate in which stable fly pupae were most often found. T he authors surmised that ovipositional experience might have driven this species to preferentially forage in these areas for hosts. Many livestock producers have used mass releases of commercially available hymenopterous parasitoids in attempts to control filth fly populations. However, studies conducted to evaluate the efficacy of this method have provided conflicting results Petersen et al. (1992) used freezekilled sentinel house fly pupae to determine mortality caused by fieldreared M uscidifurax zaraptor Kogan and Legner on cattle feedlots. Mean parasi tism was 37.3 and 25.9% in sentinel pupae placed at two treatment feedlots, but only 3.9% was in the two untreated feedlots. Parasitism by this species was relatively high after biweekly releases of up to 29,000 wasps. Petersen et al.
22 (1992) also d ocumen ted the capacity of M. zaraptor to locate hosts as far as 60-70 m from rearing sites. Andress and Campbell (1994) attained dissimilar results from mass releases of M uscidifurax raptor Girault and Saunders and S. nigroaenea purchased from a commercial inse ctary. Parasitoids were released at rates fivefold the insectary recommendation, with no significant effect on adult stable fly populations. Furthermore, it was determined that the cost of this treatment was greater than the losses sustained from stable fly feeding activity. Scientists evaluating the ability of M uscidifurax raptorellus Kogan and Legner to parasitize filth fly pupae found this species to utilize sentinel pupae at substantially higher rates than those pupae occurring naturally (Petersen a nd Cawthra 1995, Petersen and Currey 1996). However, the rates of parasitism in naturally occurring pupae were up to 15.5 and 37.2% for stable flies and house flies, respectively, were the highest observed in a field-based parasitoid study to date. In addition to the ease and cost effective nature of rearing this species, Floate et al. (2000) concluded that M. raptorellus could be released at 200 m intervals for uniform coverage of a confinement facility. This is three-fold the distance observed for M. z araptor by Petersen et al. (1992). Weinzierl and Jones (1998) found that weekly releases of S. nigroaenea and M. zaraptor nearly doubled parasitism rates, and increased total stable fly and house fly mortality by approximately 10%. However, total parasit ism never exceeded 13% for either fly species in feedlots where wasps were released. While data are conflicting in studies using mass parasitoid releases in open confinement animal operations, these practices have had little affect on fly populations in is olated environments as well. Kaufman et al. (2001c), using individual and paired
23 releases of M. raptorellus and Nasonia vitripennis Walker in poultry houses, found that parasitism of the house fly by the former did not significantly increase despite weekl y releases. Nasonia vitripennis became established when released alone, but decreasing rates of parasitism were observed in the presence of other parasitoids. In a subsequent release study, Kaufman et al. (2001a) confirmed this behavior when N. vitripenn is accounted for <1% of parasitized hosts in the presence of M. raptorellus and M. raptor Many factors can be attributed to the varied results obtained in studies using parasitoids to control filth fly populations. These can include season, temperature, type of substrate and moisture therein, host species composition, and the depth at which hosts are found. Geden (1999) found that differing moisture levels in poultry manure elicited strong responses in several parasitoid species. Muscidifurax raptor fo r example, preferred manure with 75% moisture independent of host density, with the highest response to samples in substrates at 45% moisture. Three Spalangia spp. were shown to prefer manure with a moisture content of 4565%, although they would attack pupae in wetter substrate when h ost numbers were low. These results suggest that releases of several parasitoid species with different moisture preferences may provide increased filth fly control over that of a single species in poultry facilities. Meyer et al. (1991) found pure manure to be the most abundant stable fly breeding sites on dairies although they observed no difference in abundance of seven parasitoid species at these locations. This suggests filth fly parasitoids may be more habitat specific than host -specific. Because f eed storage, waste removal, and unit design differ from one livestock facility to the next, different breeding habitats are created. Stable flies utilizing loose, wet alfalfa hay at one cattle feedlot, may be found readily in
24 compacted manure and grain at the next. Scenarios such as this may account in part for some of the varied success of parasitoid releases. Insecticide Resistance Muscoid flies, like many insect pests whose control relies heavily on the use of insecticides, have developed widespread re sistance to a variety of chemicals. Insecticide resistance in the house fly and horn fly has been well documented. Both species have developed resistance to most insecticides used for their control, and the mechanisms behind their ability to do so are well established. Over the past five decades, many insecticides including chlorinated hydrocarbons such as DDT, organophosphates such as tetrachlorv inph os and naled, and more recently pyrethroids such as permethrin and fenvalerate, have been used to control muscoid flies in Nebraska (Mar on et al. 2003). Shifting usage of pesticides such as these has occurred largely due to development of resistance in filth fly populations However, the dis continued use of some insecticides including organophosphates an d carbamates can be attributed to federal regulations, such as the Food Quality Protection Act of 1996 (Kaufman et al. 2001b). Insecticide resistance, coupled with the difficulty and costs of developing new chemicals with unique modes of action that are s afe, and effective, increases the need for information concerning the resistance status of these fly pests. Resistant house fly populations have been collected from cattle feedlot and dairy facilities, as well as from poultry units (Scott et al. 2000, Kaufman et al. 2001b, Mar on et al. 2003). Populations of insecticide resistant horn flies have also been collected from various confinement cattle facilities and pastures (Schmidt et al. 1985, Crosby et al. 1991, Cilek et al. 1991, Kaufman et al. 1999, Barr os et al. 2001). In areas where the gap between residential
25 areas and confinement animal facilities continues to close the mechanisms behind muscoid fly pesticide resistance must be determined as the potential for litigation looms. Selection pressure, due to years of pesticide use, has driven the genetic forces of resistance. Because pyrethroids, such as permethrin, are used to control a wide variety of arthropod pests, much work has been conducted to understand the behavioral and physiological mechanism s behind resistance to this group of insecticides Studies conducted by Lockwood et al. (1985) and Zyzak et al. (1996) documented that avoidance behaviors to pyrethroids displayed by horn flies, might be a cause of resistance to these compounds Two addi tional resistance mechanisms are widely accepted: increased rates of metabolic detoxification, and mutations in pyrethroid target sites (Huang et al. 2004). Distribution and Dispersal of M uscoid Fly Insecticide R esistance Although resistant populations of the house fly are known to occur globally, their geographical distribution must also be determined because r esistant populations may be problematic far from areas that currently employ chemical pesticides Learmount et al. (2002) stated that by 1978, populations of resistant house flies in the United Kingdom had been found for all major classes of pesticides used for their control. This included chlorinated hydrocarbons, organophosphates, organochlorines, pyrethrins and synthetic pyrethroids. Learmount et al. (2002) also found that resistant house fly populations were a widespread and increasing problem. A similar trend has been observed in the United States where permethrin has been the pesticide of choice to control muscoid flies for over 25 years. B ecause information concerning the geographical limits of resistant house fly populations is important, studies have been conducted to obtain this information. One
26 way to obtain such data is to evaluate pesticide resistance in flies collected from various confinement animal facilities in different geographical locations In a study conducted by Scott et al. (2000), house flies were collected from eight poultry facilities across New York state to evaluate their susceptibility to nine pesticides. Analysis r evealed that resistant populations were highly correlated with those poultry units that used specific insecticides. It was also determined that dispersal of resistant house flies to other poultry facilities was limited. Scott et al. (2000) concluded that management strategies against resistant fly populations of poultry units were feasible due to the relative isolation of insects within such closed facilities. In a similar study on New York state dairies Kaufman et al. (2001b) evaluated resistant populat ions of house flies. House flies collected from several dairies were exposed to seven insecticides to evaluate resistance. In contrast to the study of poultry facilities in which units were relatively isolated, levels of resistance were similar for all d airies independent of pesticide use history. The results suggest that flies from these opendesign facilities disperse, and thus resistant population strains were found for all insecticides used collectively. It was also observed that resistance levels for permethrin had increased from those detected in a 1987 survey of house flies collected from New York dairies (Kaufman et al. 2001b) Managers of c attle feedlot facilities also rely on the use of insecticides to control pest flies. Reduced weight gain and feed efficiency due to the landing or biting activity of muscoid flies such as stable flies can can result in severe economic impacts for producers (Campbell et al. 2001). Routine use of pesticides, such as permethrin, to control these pests has contributed to the resistant fly populations observed in
27 confinement animal operations. In a study conducted by Mar on et al. (2003), house flies collected from two cattle feedlots in Nebraska were evaluated for resistance to permethrin, stirofos, and methox ychlor. Although f lies from both facilities demonstrated resistance to permethrin, resistance levels to stirofos and methoxychlor were greater still. Resistance to methoxychlor, an organochlorine, and stirofos, an organophosphate, may have established in these fly populations when they were exposed to these, or other compounds with similar modes of action. It is surprising that stable flies collected from the same feedlots in Nebraska by Mar on et al. (2003) were highly susceptible to permethrin, methoxyc hlor, and stirofos (Mar on et al. 1997). Topical application of pesticides to cattle may limit chemical exposure to stable flies, which are normally found feeding on the forelegs. Data collected from cattle feedlots in Kansas by Cilek and Greene (1994), demonstrate that stable flies can become resistant to permethrin and stirofos, as well as dichlorvos. Resistant populations were found within facilities where pesticide use was infrequent or absent, suggesting dispersal of local populations from other ins tallations. Cilek and Greene (1994) also found dichlorvos resistance at facilities where it had not been used. However, Naled, a related compound, had been applied, thus conferring resistance to dichlorvos, through a mechanism known as cross -resistance. Mechanisms of R esistance A number of resistance mechanisms have been identified in muscoid flies. Most of th is work has been conducted on the genetic influences of pyrethroid resistance in house flies and horn flies. This is probably results from wide us e of permethrin for controlling insect populations and the decreasing efficacy documented over the years. House flies ha ve developed resistance to most insecticides used against them
28 Therefore, determination of the mechanisms by which this occurs is imp ortant if future management systems are to be successful. Knockdown pesticides such as DDT and pyrethroids result in rapid death of insects by acting on the nerve membrane voltage -sensitive sodium channels. However, long -term use of these classes of ins ecticides has led to knockdown resistance ( kdr ) within pest populations. A kdr mechanism was first demonstrated in horn flies using genetic crosses (Roush et al. 1986). Knockdown resistance target -site insensitivity was confirmed in horn flies when met abolic detoxification synergists failed to increase resistance in a laboratory colony (McDonald and Schmidt 1987). House flies expressing this trait were shown by Knipple et al. (1994) to harbor a kdr gene that is linked to the voltage-sensitive sodium channel gene. They concluded that a point mutation on the latter might be the basis of kdr in the house fly Further study by Smith et al. (1997) confirmed that a substitution of phenylalanine for leucine at the 1014 position of the amino acid sequence, L1 014F resulted in increased decay rates of pyrethroid induced sodium currents. Smith et al. (1997) concluded that this mutation decreased the pyrethroid effects on sodium channels sufficiently to account for kdr. Lee et al. (1999) found that a second mut ation of the kdr gene at the M918T location conferred even greater resistance to pyrethroids than that of the L1014F. Flies with this trait have been termed super -kdr The substitution of phenylalanine for leucine, as well as threonine for methionine, has also been shown in kdr and super -kdr horn flies, respectively (Guerrero et al. 1997). The frequencies of these point mutations vary with the level of resistance in house flies and horn flies (Jamroz et al. 1998, Huang et al. 2004).
29 Evidence exists tha t cytochrome P450s, in particular CYP6D1 play a role in pesticide resistance of muscoid flies as well (Seifert and Scott 2002). This cytochrome acts to detoxify pyrethroids in the house fly through a monooxygenase-mediated response (Liu and Scott 1997, 1998). This metabolic resistance mechanism was also demonstrated in horn flies by Sheppard and Joyce (1992), when pyrethroids were synergized with piperonyl butoxide. In flies that express this mechanism of resistance, an upregulation in transcription of CYP6D1 is observed. Liu and Scott (1998) found that flies of the Learn Pyrethroid Resistant (LPR) strain had 10-fold greater transcription levels of cytochrome CYP6D1 than did susceptible strains. The discovery of this cytochrome may also contribute to u nderstanding the geographical extent of resistant house fly populations. In a study by Seifert and Scott (2002), CYP6D1 was sequenced from two different house fly populations. Populations from Georgia and the original LPR strain discovered from a New Yor k state dairy were found to have identical CYP6D1, suggesting the possibility of dispersal. However, a study by Rinkevich et al. (2007) found that resistance allele frequency varied greatly between house flies collected at New York state dairies and those from Florida. The authors surmised therefore, that local factors, such as selection pressure are the driving forces behind resistance expression in house fly populations. Resistance to other pesticides such as thiodicarb and fipronil has been observed in house fly populations. Karunaratne and Plapp (1993) found that resistance to thiodicarb pesticides might be due to alterations in the activity of cholinesterases. Although this may be the mechanism behind detoxification of the chemical, cross resista nce may be the cause. In a study by Wen and Scott (1999), it was determined
30 that the LPR strain of house fly was resistant to the relatively new chemical, fipronil. The cause of cross -resistance in this case was contrary to previous results where cyclodi ene-resistant insects were shown to be susceptible to fipronil. The LPR strain used in these experiments had not before been associated with the cyclodiene chemical. Scott (1998) evaluated the effect of a newer chemical, spinosad, on both susceptible and resistant house fly populations The results of this study demonstrated that the insecticide was highly toxic to both populations, although the activity of the compound was relatively slow. It was also determined that the chemical could be synergized wit h piperonyl butoxide and S,S,S, -tributylphosphorotrithioate to increase its toxicity, while diethyl maleate synergism resulted in no effect. When synergized, it was concluded that house flies may have expressed a slight monooxygenase mediated detoxificati on of this newly developed chemical. More recently, Kristensen and Jespersen (2004) observed that field -collected house fly populations expressed small resistance factors to spinosad. However, they concluded that cross -resistance to other chemical classe s should not initially play a major role. Therefore, spinosad may offer an alternative to other conventional insecticides such as permethrin. Although spinosad could be used with permethrin in an alternating pesticide management program, Barros et al. (1999) found that such a regimen was ineffective in controlling pyrethroidresistant horn flies. Organophosphate and pyrethroid ear tags were used in a rotational management strategy, but did not improve or delay the effects of pyrethroid resistance. This i s contrary to the laboratory f indings of Byford et al. (1999), where r otations of diazinon, permethrin, and ivermectin, as well as mixtures of
31 these chemicals delayed the onset of resistance for up to 12 generations. Furthermore, a mosaic strategy employi ng pyrethroid and organophosphate ear tags did not result in decreased efficacy during a threeyear period. Research Objectives Research studies concerning various aspects of stable fly biology and ecology associated with cattle production are numerous. H owever, a large void exists in the literature exists for research concerning stable flies and other biting fly pests affecting equine production systems. Therefore, field and laboratory studies investigating several aspects of stable fly ecology associate d with equine facilities in Florida were undertaken to bridge this gap, as well as to provide a basis for further research in this area. My specific research objectives include: 1) The concurrent study of on-site adult and pupal stable fly populations to predict the presence of pupal p opulations and thus the appropriate timing of pteromalid parasitoid release s. D etermine the species composition and seasonal distribution of parasitoids attacking filth flies at Florida equine facilities. 2) Evaluation of the efficiency of stable fly host location and attack by selected pteromalid pupal parasitoids to determine potential intraand inter species differences in attack rates of hosts within a substrate and hosts made freely accessible. 3) Develop a multiplex polymerase chain reaction blood meal analysis tool to detect the hosts of stable flies collected from Florida equine facilities and determine whether stable fly host blood meal identification can be used to indicate their short -term localized movement. 4) Conduct a critically needed update on the status of permethrin susceptibilities of field -collected stable flies using diagnostic concentrations, and determine the rate at which stable flies express resistance to this insecticide by selecting for this trait in a laboratory colony
32 CHAPTER 2 SEASONAL ABUNDANCE O F STABLE FLIES AND T HEIR PUPAL PARASITOI DS (HYMENOPTERA: PTEROM ALIDAE) AT EQUINE FACILITIES Introduction Studies concerning the impact of stable flies, Stomoxys calcitrans (L.), on livestock are nu merous (Bruce and Decker 1958, Miller et al. 1973, Campbell et al. 1977, Mullens et al. 2006). However, control of this pest has been difficult using traditional tactics such as traps and insecticides. Whether this is due to factors such as the repellenc y effects of some pesticides (Hogsette and Ruff 1986), the relatively short time period spent on the host compared to other biting flies such as the horn fly, Haematobia irritans (L.) (Crosby et al. 1991), or their ability to disperse to other locations (B ailey et al. 1973, Hogsette and Ruff 1985, Chapter 4) remains to be seen. In addition, the modest control achieved using insecticides may become increasingly limited due to their expression of insecticide resistance (Chapter 5), a problem most noted in th e house fly, Musca domestica L. Pteromalid pupal parasitoids are often utilized as an alternative filth fly control measure in an integrated pest management program However, studies at cattle feedlot and dairy facilities report mixed success using parasi toids (Andress and Campbell 1994, Petersen and Cawthra 1995, Petersen and Currey 1996). R easons for this may include differences in season, temperature, moisture, host density and depth, and manure substrate types utilized by filth flies. Meyer et al. (1991) found no significant difference in the abundance of seven parasitoid species utilizing stable fly pupae in dairy cattle manure. They concluded that habitat, rather than host, played the primary role in rates of parasitism. In a study conducted by Sk ovgrd and Jespersen (1999), more Spalangia cameroni Perkins had parasitized stable flies buried deep in feed or manure,
33 whereas Muscidifurax raptor Girault and Saunders parasitized the majority of house flies located closer to the substrate surface. It w as surmised that this difference was due to accessibility of pupae and substrate rather than parasitoid preference for a particular fly species. Smith and Rutz (1991) also noted the habitat preferences of M. raptor most often found in outdoor feed and st raw, and S. cameroni most often found in loose indoor substrates. Preference for a particular habitat has been demonstrated in other Muscidifurax and Spalangia spp. using different moisture levels i n poultry manure (Geden 1999). Other factors, such as spe cies competition or dispersal patterns may also affect the outcome of a parasitoid release program. Kaufman et al. (2001a, 2001c) found that Nasonia vitripennis Walker parasitized more house fly sentinel pupae when it was utilized alone. However, when released with Muscidifurax raptorellus Kogan and Legner, host utilization by N. vitripennis decreased. While Tobin and Pitts (1999) found that M. raptorellus dispersed distances of 2 m or less in poultry facilities, Floate et al. (2000) record ed dispersal distances of up to 200 m from release sites at cattle feedlots. The seeming disparity in dispersal between these studies however, may have been due to the facility types in which they were conducted. The open nature of cattle feedlots may increase movemen t of insects such as parasitoids, as they are more exposed to weather and environmental factors such as wind. The confined nature of poultry facilities would inhibit such assisted movement. A more simple explanation for the disparities in success using parasitoids to control filth flies may be due to the difficulty in predicting the appropriate time for their release. A fly problem based on the presence of adult filth flies may not take into
34 consideration the presence of pupae. Often, assuming migration is not the primary cause, a sudden increase in adult flies indicates the eclosion of on-site pupae, thereby rendering immediate release s of parasitoids an ineffective control measure. Furthermore, depending on environmental conditions (Lysyk 1993), if a p arasitoid release is c onducted at that time, it may be weeks before suitable host pupae are present. By this time, natural mortality and dispersal of released parasitoids may result in insignificant control of any subsequent pupal populations (Skovgrd 2002). This is especially true for the stable fly, as immature development can take as long as 71 d under cool temperature s (Gilles et al. 2005). In general, under optimum conditions, the stinging activity of some parasitoids ends when pupae reach 4 d in a ge (Kaufman et al. 2001a). This can further complicate control if parasitoids are purchased from commercial insectaries, as shipment time may delay their release to a point when onsite hosts are no longer suitable. These factors necessitate the need for preliminary investigations of parasitoid activity within a particular facility prior to using pteromalid releases as a filth fly control measure (Greene et al. 1989). Differences in habitat type and suitable breeding substrate may exist between livestock installations, affecting the abundance, species composition, and parasitism rates of resident parasitoids. Although many equine producers utiliz e them for filth fly control, I w as unable to find any study on the effectiveness or species composition of pup al parasitoids at equine facilities. Therefore a study was initiated in December 2007 near Ocala, Florida, to help fill this void. The primary objectives of this study were to 1) use concurrent study of adult and pupal stable fly populations to predict p upal presence, and thus appropriate timing of
35 pteromalid parasitoid release and 2) determine the species composition and seasonal distribution of parasitoids attacking filth flies at equine facilities. Materials and Methods Equine Facilities In November 2007, a two year study of stable fly and house fly pupal parasitoid abundance of Florida equine facilities was initiated. The four facilities used in the study were located approximately 12 km north and/or west of Ocala, Florida, and approximately 8 km fr om each other. Each farm varied in acreage and number of horses: Farm 1; 1,800 acres, 350 horses, Farm 2; 200 acres, 110 horses, Farm 3; 4,500 acres, 250 horses, Farm 4; 320 acres, 120 horses. Farms 1, 3, and 4 used large wood chips (2-3 cm) and straw as bedding in stalls, while Farm 2 used small particle (0.1 0.3 cm) wood shavings. All farms used alfalfa hay as feed for horses, but only Farm 4 utilized round hay bales in pastures during the winter. Though daily removal of waste and debris from stalls w as practiced at each farm, the methods used for its disposal differed. Farms 1 and 3 disposed of accumulated horse wastes using large composting areas, with a central dumping location from which compost windrows were made. Farms 2 and 4 utilized manure s preaders to distribute horse waste products throughout pastures. All farms were chosen because they were large tracts of land with at least 100 horses, and were an appreciable distance (>0.5 km) from nearby cattle facilities, which facilitated the complet ion of my objectives in Chapter 4. Adult Stable F ly S urveillance Adult stable fly populations were monitored weekly at each farm using corrugated alsynite cylinder traps placed in different locations throughout each farm. Three traps were placed similar ly on all farms, with one trap near pastured horses, a second near barns with stabled horses, and a third near composting or manure spreading areas. Each trap was set up similarly to those
36 described by Hogsette and Ruff (1990). Briefly, a corrugated alsy nite cylinder (Olson Products Inc., Medina, OH) approximately 30 cm in height and 20 cm in diameter was mounted in a slit cut into a 5 x 5 x 122 cm wooden stake. Each wooden stake was driven 30 cm into the ground so that trap height was approximately 90 c m (Williams and Rogers 1976) (Fig 21). A translucent adhesive-coated propylene sleeve (Olson Products Inc., Medina, OH) was attached to the outer surface of each trap with metal clips and left in place for 7 d. Sleeves were replaced weekly and stable fl y numbers were recorded for each facility by date. Stable fly counts did not include sex determination as the characters used for this identification were unreliable due to damage caused by the adhesive glue and time that captured flies spent on the traps Pupal C ollection. Weekly f ilth fly pupal collections began in December 2007 and continued until December 2009 Initially, an attempt was made to use the sentinel pupae technique of Rutz and Axtell ( 1979) to assess pteromalid pupal parasitoid activity. However, this effort was abandoned as local red imported fire ant, Solenopsis invicta Buren, populations destroyed most sentinels. Thereafter, each week an attempt was made to collect at least 50 pupae from expected breeding areas at 3 -5 sites within eac h farm, collecting more pupae when possible. Regardless of availability, searching ceased after 30 minutes and the resultant pupae were removed from collected debris by floatation. All pupae were returned to the University of Florida Veterinary Entomolog y Laboratory and sorted to remove any pupae that were previously eclosed, had been damaged, or had parasitoid emergence holes using the intact pupae method of Petersen and Meyer (1985). This method was also used to calculate the percent parasitism. Immed iately after their collection, a ny intact pupae that were dark in color
37 were placed individually into #0 gelatin capsules. At this time, a ll other intact pupae were placed into 120 ml plastic souffl cups with covers and held for 3to 5 -d at 26 C, 12:12 LD, and 70% RH for adult fly eclosion. After the 3 to 5day holding period any remaining uneclosed pupae were placed individually into #0 gelatin capsules and held for 40 d at 26 C, 12:12 LD, and 70% RH for parasitoid emergence (Fig. 22). Pupae that did not produce a fly or a parasitoid were dissected to determine if adult fly mortality was due to an aborted parasitoid. The keys of Rueda and Axtell (1985a), and the pictorial guide by Gibson (2009) were used to identify all pteromalid pupal parasitoi ds. Statistical A nalysis. Stable fly trap collection data were subjected to analysis of variance (ANOVA) using the PROC GLM procedure of SAS 9. 2 (SAS Institute 2004) to determine differences in adult stable fly activity by month and by farm. Stable fly numbers by trap/month and by trap/farm were transformed using ln( n + 1), using month and farm as fixed effects in their respective analysis. These results are presented as untransformed data. Stable fly trap collections were also subjected to the PROC CO RR procedure to determine the correlation in stable fly seasonal distribution between farms. The percent contribution of each trap collection to the total number of stable flies collected on each farm was subjected to an arcsine square root transformation prior to this analysis. Multiple mean comparisons were conducted with the Ryan -Einot -Gabriel activity, weekly precipitation and temperature data was collected from the National Oceanic and Atmospheric Administration (NOAA, 2009) (Fig. 2 3 untransformed data) site, located in Ocala, Florida.
38 An attempt to correlate stable fly pupal collections with those of adult stable flies, as well as rainfall and temperature was made using the PROC REG procedure (stepwise multiple regres sion). Only weeks in which at least one stable fly pupa could be found were included in the analysis. Total adult and pupal stable fly collections from all four farms during a given week were transformed using ln( n + 1), and are presented using transform ed data. The percent parasitism was calculated as the number of emerged (identifiable) and aborted (unidentifiable) parasitoids divided by the number of intact pupae collected during a given week, at a given farm. Th e s e data w ere subjected to ANOVA to det ermine differences in parasitism rates between farms, as well as differences in parasitoid species composition during a given week, at a given farm. The data were analyzed separately for each fly species. There were only two weeks in which 50 house fly p upae were collected from Farm 4, resulting in the subsequent removal of this farm from the analysis for that fly species. Two additional ANOVA s w ere conducted to determine if differences between inter and intra-species parasitism occurred during a given m onth. Data from both stable fly and house fly pupal collections were pooled for this analysis. Only months in which 50 stable fly or house fly pupae had been collected on at least five occasions were included in the analysis, resulting in the subsequent removal of June, July, August, November, and December. All parasitoid data were transformed using an arcsine square root of the percent parasitism for a given week, at a given farm and are presented in tables as back -transformed means. Both farm and mon th were included
39 as fixed effects in their respective analysis. Multiple mean comparisons were conducted with the Ryan -Einot -Gabriel Results Adult Stable F ly S urveillance A total of 104,718 stable flies were collected from four equine facilities between November 2007 and December 2009. Although stable flies were captured in every month, significantly more (F11, 1263 = 75.98; P = <0.0001) were collected during March (74 flies/trap/week) and April (76 flies/tra p/week) than any other month. With the exception of Farm 2 in 2008 stable fly collections gradually increased beginning in January, with peak collections occurring at the end of April in both years (Fig. 2 4a, b). In 2008, peak stable fly collections oc curred in January at Farm 2, although trap captures from this farm had been increasing when the study began in November 2007. Results from the mean separation test indicated that s ignificantly more stable flies were collected from Farm 2 than any other f arm, followed by Farms 4, 3, and 1, respectively. During this study, stable fly population dynamics were highly correlated ( P <0.0001) between farms, with r values ranging from 0.66 -0.95 between years, and from 0.820.92 for both years combined. Stable F l y P arasitism During the twoyear study 12,675 stable fly pupae were collected from sites located within the four equine facilities. From these, 1,928 adult and aborted pteromalid pupal parasitoids were produced (Table 21). With the exception of one pupa producing a M. raptor all P teromalid ae collected from stable flies were Spalangia spp. No significant differences were detected between farms in overall stable fly percent parasitism or in the percent composition of Spalangia nigroaenea Curtis or Spala ngia endius Walker. Significant differences between farms in percent composition were detected for S. cameroni (F3, 56 = 3.80; P = 0.0303) and Spalangia
40 nigra Latreille (F3, 56 = 6.18; P = 0.0011). A significantly greater percentage of the parasitoids collected from Farm 4 were S. cameroni than at Farm 1. However, a significantly lower S. nigra parasitism rate was observed at Farm 4, as compared to Farm s 1 and 3. Although no significant differences were detected between farms in the percent composition of S. endius this species was recovered more often from stable fly pupae collected at Farm 1 than at Farms 2 and 3. House F ly P arasitism A total of 12,841 house fly pupae were collected from four equine facilities between December 2007 and December 2009 producing a total of 1,331 emerged and aborted pteromalid pupal parasitoids (Table 21). With the exception of one house fly pupa yielding a M. raptor and one pupa producing a Phygadeuon spp. (data not shown), all resultant parasitoids were Spalangia sp p. No significant differences in percent house fly parasitism were detected in overall percent parasitism between farms or in the percent composition of S. nigra and S. endius Significant differences between farms were detected for percent composition o f S. cameroni (F2, 25 = 3.80; P = 0.0362) and S. nigroaenea (F2, 25 = 3.98; P = 0.0315). On Farms 1 and 2, significantly more S. nigroaenea were recovered from house fly pupae than on Farm 3. Conversely, s ignificantly more S. cameroni were collected from Farm 3 than Farms 1 and 2 (Table 2-1). Monthly Parasitoid A bundance Pteromalid pupal parasitoids were recovered during every month of the study with the exception of November, when no stable fly or house fly pupae could be recovered from any farm (Table 2 -2). Our intra-species analysis indicated that there were n o significant differences between months in overall percent parasitism by all Spalangia spp. or in percent composition of S. nigra.
41 Significant differences between months were detected in percent species composition of S. cameroni (F6, 79 = 6.21; P = 0.0001), S. nigroaenea (F6, 79 = 6.94; P = 0.0001), and S. endius (F6, 79 = 8.06; P = 0.0001). A significantly greater proportion of S. cameroni were collected between January and May than during S eptember and October. Significantly more S. nigroaenea were collected during September and October than in February or March. Significantly more S. endius were collected in September than any other month during the study. Significant differences were det ected in our inter species abundance analysis for all months in which sufficient pupal recovery was achieved : January (F3, 32 = 69.40; P < 0.0001), February (F3, 56 = 106.86; P < 0.0001), March (F3, 80 = 69.05; P < 0.0001), April (F3, 92 = 32.78; P < 0.0001 ), May (F3, 20 = 5.89; P = 0.0047), September (F3, 20 = 11.22; P = 0.0002), and October (F3, 16 = 7.17; P = 0.0029). Between January and April, significantly more S. cameroni were recovered from filth fly pupae than any other species, followed by S. nigro aenea, S. nigra and S. endius During May, significantly more S. cameroni and S. nigroaenea were collected than S. endius Significantly more S. nigroaenea were collected from filth fly pupae than any other species during the months of September and Oct ober. Correlating Pupal and A dult Stable F lies. The results of the multiple regression analysis indicated that precipitation, temperature, and adult stable fly trap collections were not significant factors in determining our ability to locate pupae from f ilth fly breeding sites. However, a significant (F1, 37 = 108.18; P = 0.0001) relationship was identifi ed between adult stable fly trap collections and pupal collections from 2 weeks prior, r2 = 0.74 (Fig. 2-5).
42 Discussion Adult s table flies were collect ed from Florida equine facilities every month during the twoyear study. S ignificant differences were detected in the number of stable flies collected from each farm throughout the study. On average, traps at Farm 2 yielded the most stable flies per week followed by Farms 4, 3, and 1. Although Farm 2 used a manure spreading technique to dispose of stall waste each day, the small particle size of the wood shavings and the continuous re use of the same waste disposal sites created an optimum breeding subs trate for stable flies. Farm 4 also used a manure spreader, but here debris was cast out over long distances and did not accumulate as it did at Farm 2. However, Farm 4 was the only site to use round hay bales to feed horses, a widely recognized developm ental site for stable fly larvae on cattle farms (Broce et al. 2005, Talley et al. 2009). Not surprisingly stable fly trap collections were greatest at Farms 2 and 4. Stable fly production at Farm 3 was apparent, but did not seem to justify the numbers of stable flies collected on traps. My results from Chapter 4 suggest that many of the flies collected at Farm 3 could be occurring as emigrants from nearby cattle farms. Whether off -site breeding is the cause of the stable fly population on this farm remains unclear Historically stable fl y breeding and host -feeding have been most often associated with cattle (Hogsette et al. 1987). Stable fly activity at Farm 1 was significantly lower than any other farm. This is probably due to the Farms intense practice of daily cultural controls and composting activity. It also may be noted that this is the only farm in which chemical insecticides were not used, adding to the utility of cultural techniques as a stable fly control method. Stable fly population dy namics between farms was highly correlated suggesting that adult stable fly activity is driven by similar factors at each farm, such as weather
43 (Lysyk 1993). This is particularly evident in stable fly collections during 2009, when a non-typical peak in st able fly activity occurred in June at all farms (Fig 2 4b). Average temperatures between both years during the months preceding June were similar. However, in May 2009, accumulated precipitation was 31 cm, compared to 0.5 cm in May 2008 (Fig. 23). Ther efore, it is likely that precipitation played a larger role in the increased stable fly activity later in 2009 (Hogsette et al. 1987) This is likely to have provided a late-season developmental opportunity for stable flies otherwise maintain ed at minimal levels under average May precipitation conditions (3.2 cm) (NOAA 2009) Although several studies of stable fly activity in Florida have been conducted, most have been in an attempt to identify their breeding sites (Fye et al. 1980) and sudden appearance i n areas such as the G ulf coast beaches (King and Lenert 1936, Hogsette and Ruff 1985). In Florida, seasonal stable fly activity is usually greatest between the months of January and April, although stable flies occur throughout the year ( Gentry 2002). Mu ltiple peaks in stable fly abundance during these months can occur, depending on precipitation and available breeding habitats. This is similar to results of the present study, where stable fly collections began to increase steadily in January, with peak collections occurring in April. By early May, stable fly collections declined dramatically and were minimal for the remainder of the year except for a single lateseason peak in June of 2009 (Fig 2-4a, b). The seasonal stable fly distribution observed in our study does corroborate research co nducted by others for the state, but differs dramatically from studies in other areas of the U.S Mullens and Meyer (1987) reported that peak stable fly activity at California dairies occurred in May and June, wherea s Burg et al. (1990) found that
44 stable fly populations in Kentucky began to rise in May, with peak collections occurring during the summer months of June, July, and August. This is similar to results obtained by Broce et al. (2005) in Kansas, although bi modal peaks in stable fly activity were observed in June and again during the months of September and October. Increased stable fly activity noted by Lysyk (1993) working in Alberta, Canada was later still, occurring during the months of August and September. The later dates for more northern areas are not surprising, given the seasonal temperature patterns of those areas. To my knowledge, yearlong stable fly seasonal surveillance data have not been published for Florida, nor has it been used to indicate the presence of stable fly pupae. An attempt to correlate both stable fly life stages was made to give livestock producers a practical method to more closely predict appropriate times for release of pupal parasitoids. Regression analysis demonstrated th at while the presence of adult and pupal stable flies were highly correlated (Fig. 25), the relationship was only valid when adult trap collections were associated with pupae from 2 weeks prior. Such knowledge does little to assist in the preemptive management of this pest. There are several potential reasons for this occurrence. In the present study, stable fly populations gradually increased throughout the season; it is likely their resultant breeding also produced gradually increasing pupal numbers wi thin breeding areas. The significant relationship between adult stable flies and pupae from two weeks prior does correspond with the average time period for stable fly development from egg to pupa (Gilles et al. 2005). Additionally, t he average temperatures during the months when stable flies are most active in Florida are similar to those used for colony rearing
45 in our laboratory. Furthermore, this analysis was dependent on my ability to locate pupae scattered throughout a substrate, and their numbers l ikely varied due to the methods used for their collection. However, stable fly pupae were found in 39 of over 100 sampling attempts which is consistent with the number of weeks (16 -18/year) that stable flies are primarily active in Florida. Although h ouse flies were not the target of this study, they were opportunistically collected as pupae. Further research using the concurrent trapping of both fly species and different pupal collection methods may provide a relationship that could be used to predict the presence of subsequent pupa. Over the course of the twoyear study, nearly 100% of all pteromalid pupal parasitoids recovered from filth fly pupae were Spalangia spp., with more than 90% of the parasitoids being either S. cameroni or S. nigroaenea T his is similar to results from across the U.S., where these species made up a significantly larger proportion of recovered parasitoids than others (Greene et al. 1989, Meyer et al. 1990, Jones and Weinzierl 1997). Several studies also attest to the propensity of members of this genus to search deep within substrates for both stable fly and house fly hosts, compared to other parasitoids such as Muscidifurax spp. (Rueda and Axtell 1985b, Greene et al 1989, Skovgrd and Jespersen 1999). In addition, most st udies have been conducted at cattle feedlots or dairies (Seymour and Campbell 1993, Meyer et al. 1991), and poultry facilities (Rutz and Axtell 1981, Kaufman et al. 2001c) where fly breeding habitats can differ greatly depending on the cultural management practices of each farm. These studies often report the common occurrence of Muscidifurax spp. recovered from pupae within different livestock facilities. In the present study, pupae were never located in substrates at depths less than 3 cm, and in most c ases were
46 collected at greater depths. In addition, most of the breeding areas sampled in the current study contained porous, loose debris ; sites favorably searched by Spalangia spp. (Smith and Rutz 1991). Stable fly pupae from all farms with the excepti on of Farm 4, were most often collected from within horse dung found in discarded horse bedding. Most pupae collected from Farm 4 were located deep with in decomposing alfalfa hay n ear round bale feeding sites. These p upae were located at distances from round bales similar to observations of Talley et al. (2009). House fly pupae were most often located deep within the sandy soil beneath discarded horse bedding. These factors may explain the overwhelming occurrence of Spalangia spp. and the absence of ot her genera, such as Muscidifurax spp. Overall parasitism rates varied between farms from approximately 718% and 518% for stable flies and houses flies, respect ively but were not significantly different. Estimates of parasitism determined by published s tudies vary widely, but several report parasitism rates similar to those of the present study. Petersen and Cawthra (1995) and Petersen and Currey (1996) observed parasitism rates of M raptorellus as high as 15.5 and 37.2% for stable flies and house flie s, respectively. Weinzierl and Jones (1998) determined that weekly releases using S. nigroaenea and M uscidifurax zaraptor Kogan and Legner resulted in parasitism rates of 11.6 and 13% for stable flies and house flies, respectively In Florida, Greene et al. (1989) found that parasitism was as high as 61 and 71% for stable flies and house flies respect ively depending on the substrate from which pupae were collected. However, parasitism in the Greene et al. study was calculated using only pupae that resulted in a fly or a parasitoid, which may have inflated parasitism estimates. The use of the intact pupa method in the present
47 study may have underestimated parasitism rates, as some of the pupae held for parasitoid emergence may have been suitable for par asitoid attack when they were removed from the equine environment However, my technique does provide a moderately accurate assessment according to Petersen and Meyer (1985). All four Spalangia spp. collected during this study were recovered from both st able fly and house fly pupae (Table 21). Significant differences in parasitoid s pecies recovered from house fly pupae were detected only for Farm 3. Here, t he percent species composition of S. cameroni was greater at Farm 3 than the other farms, while t hat for S. nigroaenea was l ow er at Farm 3 than the other farms for the entire sampling period Analysis of the seasonal distribution for these species makes it unlikely that this is due to species competition, and instead suggests the differences may be due to my inability to locate pupae at Farm 3 during times when S. nigroaenea were most abundant. The significant differences determined in percent composition of Spalangia spp. were similar for parasitoids emerging from both stable fly and house fly pupae. In general, S. cameroni made up a larger proportion of the parasitoids recovered from stable flies and house flies from Farms 3 and 4, although there was no significant difference between these farms and Farm 2. Spalangia nigroaenea made up a larger pr oportion of the parasitoids recovered from Farms 1 and 2, al though no significant differences were detected in percent composition. These differences may be explained through examination of the filth fly immature developmental habitat on the four farms. Although they differed in equine waste disposal methods, the developmental habitats created by the moistened alfalfa hay and straw at Farm 3 and round hay bale debris at Farm 4, were actually quite similar. However, i n the absence of rain events, the lac k of
48 straw and hay accumulation areas at Farms 1 and 2 left horse manure debris as the only suitable area available for larval development. It may be that when encountered at equine facilities, these species demonstrate a preference for a particular habit at the presence of which is dependent on a combination of the cultural management practices of each farm and weather patterns of a given year These results are consistent with many studies demonstrating that parasitoid occurrence is not dependent on the host species, but rather the habitat in which that host is located (Rueda and Axtell 1985b, Meyer et al. 1991, Smith and Rutz 1991, Geden 1999). Analysis of the months in which Spalangia spp. are active at Florida equine facilities demonstrated that S. ca meroni were most often recovered in pupae collected between January and May, while S. nigroaenea were most often recovered in September and October. Although collect ed less often, most S. endius w ere also recovered in September. Based on this information one may speculate that the difference in seasonal distribution may be a method to avoid competition by utilizing a different temporal niche. There were only two occurrences of M. raptor throughout the study. This is probably due to the nature of the habitats created by equine husbandry that are suitable for filth fly breeding, and the unwillingness of this gen us to search within those habitats. One Phygadeuon spp. was recovered from a house fly pupae, and may be the first reported occurrence of this gen us from a Florida livestock facility. Although not included in any statistical analysis, Staphylinidae (82%) were recovered from pupae collected during October, December, and January. No adult beetle was recovered from any
49 pupae making identification bey ond family difficult. Because Aleochara spp. have been frequently identified from filth fly pupae, it is likely that our samples were of this genus. Our results demonstrate that the composition of pteromalid pupal parasitoids occurring at Florida equine f acilities is unique among similar studies. It is likely that horse producers utilizing commercially available parasitoids or parasitoid mixtures containing Muscidifurax spp. will attain little control, if any with those species. Therefore, further resear ch utilizing releases of only Spalangia spp. are needed to determine if increased fly management is possible. Our data concerning the seasonal distribution of filth fly parasitoids should assist in temporal releases of particular Spalangia spp. as well.
50 Table 2 1. Total pupae collected from four Florida equine facilities between December 2007 and 2009, with their respective mean percent parasitism rates and mean percent Spalangia spp. composition1. Farm No. Parasitoids (No. UID)2 %Parasitism (95% CI)3 % Spalangia spp. (95% CI)4 S. cameroni S. nigroaenea S. nigra S. endius Stable Fly Host 5 1 209(9) 12.0(8.8 15.7)a 55.7(42.4 68.6)b 17.7(10.6 26.2)a 6.3(2.3 12.0)a 1.7(0.2 4.5)a 2 480(41) 17.7(13.6 22.3)a 77.1(68.1 85.0)ab 15.8(9.1 23.9 )a 2.2(0.9 4.1)ab <0.1(0 0.2)a 3 705(17) 17.1(12.4 22.4)a 68.1(58.4 77.2)ab 13.9(7.9 21.3)a 8.7(5.5 12.6)a <0.1(0 0.1)a 4 448(19) 7.3(5.7 9.0)a 89.4(85.6 92.5)a 8.4(5.7 11.6)a <0.10 (0 0.2)b 0.2(0.1 0.4)a Total 1,842(86) House Fly Host 6 1 473(23) 5.1(2.8 8.3)a 33.4(19.2 49.5)b 60.3(44.6 75.0)a 0.0 (0)a 0.4(0 1.7)a 2 599(34) 17.7(9.028.6)a 21.8(9.837.8)b 69.9(54.183.6)a <0.1(0 -.03)a 1.8(0.34.4)a 3 87(5) 5.6(3.68.0)a 92.9(82.4-98.9)a 4.4(0.6-11.3)b 0.9(0 -3.4)a 0.3(0 -1.0)a 4 107(3) 2.6 n/a 92.8 n/a 7.2 n/a 0.0 n/a 0.0 n/a Total 1,266(65) 1 Means in each column within host -type followed by the same letter are not significantly different (Ryan-Einot -Gabr iel (n/a) 2 UID represents aborted parasitoids recovered during pupal dissections where identification was not possible. 3 Percent parasitism was calculated as the number of emerged and aborted parasitoids recovered divided by the total intact pupae collected at any given farm, during any given week. 4 Percent Spalangia spp. was calculated as the total number of a given species collec ted from a farm during a given week, divided by the total parasitoids recovered for that week. 5 A total of 12,675 stable fly pupae collected: Farm 1 = 1,677, Farm 2 = 2,149, Farm 3 = 3,788, Farm 4 = 5,061. 6 A total of 12,841 house fly pupae collected: Fa rm 1 = 7,164, Farm 2 = 3,646, Farm 3 = 1,226, Farm 4 = 805.
51 Table 2 2. Mean percent parasitism rates and mean percent Spalangia spp. composition recovered from stable fly and house fly pupae collected from four equine facilities near Ocala, Florida1. % Spalangia spp. (95% CI) 4 Month 2 % Parasitism (95% CI) 3 S. cameroni S. nigroaenea S. nigra S. endius Jan 6.5 (3.3 10.6)A 60.1 (53.6 66.4) Aa 38.3 (34.4 45.5)BCb <0.1 (0 0.1) Ac 0.5 (0.1 1.1)Bc Feb 17.8 (12.8 23.5)A 84.4 (79.6 88.7) Aa 9.2 (6.1 12. 8)Cb 3.2 (1.8 5.1) Ab <0.1 (0 0.07)Bc Mar 8.0 (6.2 10.1) A 85.6 (79.0 91.0) Aa 4.8 (2.3 8.0)Cb 2.7 (1.2 4.9) Ab 0.1 (0 0.5)Bb Apr 13.7 (10.2 17.7) A 66.8 (57.4 75.6) Aa 25.6 (17.2 35.0)BCb 1.3 (0.4 2.5) Ac <0.1 (0.02 0.2)Bc May 8.9 (4.7 14.2) A 62.1 (39.5 82.2) Aa 31.7 (12.8 54.6)BCab 1.2 (0.04 3.9) Ab 0.2 (0 0.6)Bb Sep 6.5 (3.3 10.6) A 8.7 (2.6 18.1)Bb 71.3 (56.0 84.4)ABa <0.1 (0 0.06) Ab 14.3 (6.4 24.6)Ab Oct 1.8 (1.2 2.6) A 9.6 (0 34.6)Bb 90.4 (65.5 100)Aa 0.0 (0) Ab 0.0 (0 ) Bb 1 Means in each column followed by the same capital letter are not significantly different while m eans in each row followed by the same lower -case letter are not significantly different (Ryan-Einot -Gabriel 0.05) ). 2 In sufficient pupae were collected in June, July, and Aug, and December to be included in the statistical analysis. No pupae were recovered from any farm in November. 3 Within farm and collection week, percent parasitism was calculated as the number of emerg ed and aborted parasitoids recovered, divided by the total intact pupae collected. 4 Within farm and collection week, percent Spalangia spp. was calculated as the total number of a given species collected, divided by the total parasitoids recovered.
52 Fi gure 21. Corrugated alsynite sticky trap mounted at a height of 90 cm near horse pasture, Ocala, Florida.
53 Figure 22. Plastic souffl cup containing stable fly or house fly pupae individually placed into #0 gelatin capsule s and held at 26 C, 12:12 LD, and 70% RH for parasitoid emergence.
54 Figure 23. Mean weekly temperatures and accumulated monthly precipitation for Ocala, Florida, occurring between November 2007 and December 2009. Data obtained from the National Oceani c and Atmospheric Administration (NOAA, 2009) site in Ocala, Florida
55 Figure 24. Weekly stable fly trap collections from four equine facilities during a.) November 20072008 and b.) November 2008-2009. The x axis represents month of collection and the y axis represents the total stable fly captures from three traps at each farm, each week. a b
56 Figure 25. Relationship of adult stable fly trap captures for a given date, and stable fly pupae collected 2 wk prior. Stepwise multiple regression document ed that adult stable fly trap captures were strongly correlated with stable fly pupal numbers 2 wk earlier ( r2 = 0.74, F1, 37 = 108.18, P = <0.0001). y = 0.8396x 0.6003 SEint = 0.55034 SE slope = 0.0807
57 CHAPTER 3 THE ABILITY OF SELEC TED PUPAL PARASITOID S (HYMENOPTERA: PTEROMALIDAE) TO LOC ATE STABLE FLY HOSTS IN EQUINE HUSBAND RY GENERATED SUBSTRATES Introduction Filth flies, in particular the stable fly, Stomoxys calcitrans (L.), continue to be a significant pest of confined and pastured livestock. Furthermore, the insecticides available for control of pest s such as the stable fly result in modest control, and are becoming increasingly limited due to federal regulation such as the Food Quality Protection Act of 1996 (Kaufman et al. 2001b) and resistance expression in some populations (Cilek and Greene 1994, Mar on et al. 1997, Chapter 5). In addition, pressure on livestock producers to control dispersing fly populations is mounting, as human population growth continues to decrease the gap between residential areas and nearby livestock operations. This can c ause a greater quandary, as the flies nuisance behavior and potential for disease transmission to urban areas may result in lit igation (Tobin and Pitts 1999). Many studies have been conducted to determine the effects of parasitoids, both released and naturally occurring, as an alternative method for filth fly control. However, the results of such studies have provided conflicting data. Geden et al. (1992) observed sentinel pupae parasitism rates as high as 65% on dairies in New York state with a parasito id release program, compared to 30% on control farms. Similarly, Petersen et al. (1992) noted up to 37% mortality of sentinel pupae on cattle feedlots where parasitoid releases occurred, while mortality at control feedlots was 4%. These results are dissi milar to results of a parasitoid study conducted on California dairies, where sentinel pupal parasitism increased from approximately 10% to only 20% when wasps
58 were released (Meyer et al. 1990). Concurrent evaluation of field-collected pupae indicated par asitism rates of only 4.4 and 12.5% for stable flies and house flies, respectively. In Nebraska, a study using parasitoid release-rates five -fold greater than the insectary -recommended amount were ineffective in reducing adult stable fly populations (Andr ess and Campbell 1994). Several factors may account for the contrasting results of parasitoid release studies conducted in different geographical areas or livestock facilities. Habitat has been shown to play a large role in the host location success of ma ny parasitoids. Filth fly breeding sites such as tightly packed feed and manure at one facility may cause wandering maggots to pupate closer to the surface or in easily accessible cracks, favoring attack by Muscidifurax spp., whereas conditions at another facility may require that parasitoids search at greater depths to locate pupae, thereby favoring attack by Spalangia spp. (Rueda and Axtell 1985b, Meyer et al. 1991, Smith and Rutz 1991). Furthermore, filth fly breeding substrates differing in abiotic fa ctors such as moisture (Geden 1999) or light (Smith and Rutz 1990) may influence the success of a parasitoid release program. Research studies concerning stable flies in Florida are numerous (King and Lenert 1936, Fye et al. 1980, Hogsette and Ruff 1985). However, most of this research has been directed towards nonanimal breeding sites and dispersal patterns, with limited research on livestock facilities (Greene et al. 1989). A large gap in research concerning equine facilities and filth flies also exist s, particularly in the area of resident pupal parasitoid populations. Additionally, in a previous study, I demonstrated that the unique
59 filth fly breeding habitats created by equine husbandry practices influence the pteromalid pupal parasitoid species com position on these farms (Chapter 2). Herein, an experiment was conducted to determine the ability of Spalangia cameroni Perkins Spalangia endius Walker and Muscidifurax raptorellus Kogan and Legner to locate and attack hosts in a standard substrate colle cted from a Florida equine facility. The primary goals of this study were to: 1) determine if the lack of field collected Muscidifurax spp. is linked to searching behavior and host -location within substrates found at Florida equine facilities; and 2) determine the intra and inter species differences in attack rates of hosts both within a substrate and those made freely accessible. Materials and Methods Stable F lies. A stable fly colony was established from wild individuals collected in February of 2007, a t the University of Florida Dairy Research Unit, in Hague, FL (UFD strain). The UFD colony flies we re maintained at 26 2 C, 12:12 LD, and 70 5% RH. Citrated bo vine blood was provided daily via saturated cotton in a 120 ml p lastic souffl cup. Gator ade wa s provided ad libitum in a 500 ml cup fitted with dental wick as a sugar and electrolyte source. Eggs we re collected one to two times weekly and ad ded to a larval medium similar to that described by McPheron and Broce (1996). A modification of thi s diet was made using maple wood chips in place of vermiculite Briefly, the diet wa s comprised of 2.8 L water, 4.0 L wheat bran, 1.2 L Teklad maple sani -chips (Harlan Laboratories, Inc. Tampa, FL) and 0.4 L fishmeal (Nelson and Sons Inc., Murray, UT) After a development period of approximately 14 d, pupae we re extracted from the rearing medium by floatation, dried, and placed in a clean 45 x 45 x 45 cm aluminum screened cage.
60 Pupal Parasitoids and T est S ubstrate. The pteromalid pupal parasitoids chos en for this experiment were S. cameroni S. endius and M. raptorellus Both S. cameroni and S. endius were obtained for this project on the day of their intended use in the experiment from colonies maintained by Dr. Christopher Geden at the USDA ARS C en ter for M edical Agriculture, and Veterinary Entomology in Gainesville, F lorida The gregarious strain of M. raptorellus used in this study was obtained from a colony maintained at the University of Florida Veterinary Entomology Laboratory in Gainesville, F lorida In our field studies, more stable flies and parasitoids were recovered from traps and pupal collections, respectively, at Farm 2 an equine facility northwest of Ocala, Florida, than from the other equine farms evaluated (Chapter 2) Therefore, F arm 2 served as the source for the soiled horse bedding substrate chosen for our laboratory assays. Soiled horse bedding at Farm 2 was composed of small particle (0.1-0.3 cm) wood shavings, and always contained varying amounts of discarded alfalfa hay, horse manure, and horse urine. To obtain uniform samples for our assays, large, 60 L plastic bins of soiled horse bedding were thoroughly mixed by shovel and divided among 3.75 L plastic zipper bags. These bags were frozen for 1week prior to use to kill a ny arthropods present at the time of collection. Parasitoid R elease C hambers. Cylindrical plastic bins (chambers) having a 26 cm diameter, a 9 cm height, and a total volume of 4.8 L, were used as arenas for the soiled horse bedding habitats for the experi ment. The chambers had tight -fitting lids modified with an 80mesh screened area. Fifty, post -feeding, UFD stable fly larvae were released on the center surface of 16 release chambers previously filled with
61 approximately 3.7 L ( habitat depth, 7 cm) soiled horse bedding, (Fig 3-1) as well as on 16, 120ml plastic souffl cups (cups) containing dry maple wood chips (Fig 3-2). Each set of 16 chambers or cups w as randomly assigned to one of four treatment groups which included: S. cameroni -release, S. endius -release, M. raptorellus -release and a no parasitoid control. This arrangement resulted in treatments containing both 4 chambers (with bedding) and 4 cups (no bedding) per experimental setup. The cups served three purposes: 1) determin ation of the time of pupariation, and thus the appropriate time for parasitoid release, 2) a way to compare parasitoid efficiency between searching for dispersed pupae in release chambers and freely accessible pupae in cups, and 3) a way to ensure that the pupae were suitab le for parasitization When at least 90% of the larvae contained in plastic cups had pupated, five female parasitoids of the appropriate species were introduced into each chamber or cup of their respective treatment (Geden 2002). Parasitoids were allowed 72 hr to search for and attack hosts (Kaufman et al. 2001a). At this time, the pupae from each chamber were recovered from the soiled horse bedding using a #6 brass sieve, and placed into new 120-ml plastic cups. Pupal recovery rates were approximately 92 100% in the chambers and 100% in the cups. Parasitoids were removed from treatment cups and a ll cups were held at 26 C, 12:12 LD, and 70% RH for 35 days for adult fly eclosion. Following the 5 day holding period, remaining uneclosed pupae were placed individually into #0 gelatin capsules and held at the same conditions for 40 d to allow parasitoid emergence ( Mann et al. 1990, Lysyk 2001). Any pupae not producing an adult stable fly or parasitoid were dissected to determine the presence or absence of partially developed parasitoids. This experiment was replicated three times, for a total of 12
62 release chambers and 12 cups for each treatment group. For each replication, five samples of the soiled horse bedding were weighed and dried to assess the mois ture content of the substrate. Statistical A nalysis. Three factors were subjected to statistical analysis to determine if differences in searching behavior and parasitism existed between parasitoid species. These factors included: 1) the percent whole pupae, or pupae that did not produce an adult stable fly divided by the total pupae recovered from a particular container, 2) the percent parasitism, or pupae that produced an adult or partially developed parasitoid divided by the total pupae recovered from a particular container, and 3) the percent parasitoid induced mortality (PIM) (Petersen et al. 1991) or difference between percent whole pupae and percent parasitism. Stable fly control mortality was assessed in release chambers and plastic cups designat ed as noparasitoid treatments. Therefore, Abbotts correction (Abbott 1925) was applied to the percent whole pupae to adjust for natural mortality factors. Each of the three response variables listed above w ere subjected to analysis of variance (ANOVA) u sing the PROC GLM procedure of SAS 9.1 (SAS Institute 2004) to determine differences between species. For each ANOVA, species and replication were included as fixed effects in the model, with container type as a variable. An additional ANOVA was conduct ed for each parasitoid to assess any within species differences in the efficiency of the aforementioned factors due to container type. All data were transformed using an arcsine square root of the percent whole pupae, percent parasitism, and percent PIM. Multiple mean comparisons were conducted with
63 the RyanEinot -Gabriel presented in figures are back -transformed means. Results Percent W hole P upae Significant differences in percent whole pupae recov ered were detected between parasitoid species in both the chambers (F4, 31 = 80.90 ; P <0.0001) and cups (F4, 31 = 6.09; P <0.00 59 ) (Table 3 1) There was a significant difference (F2, 31 = 5.28; P = 0.0106) in percent whole pupae between replications in t he cups, but not between replications in the chambers. In both the chambers and cups, significantly fewer whole pupae were recovered for M. raptorellus (3 and 76%, respectively) than either Spalangia spp., with no difference between S. cameroni ( 73 and 93 %, respectively) and S. endius (69 and 96%, respectively) (Fig 3 3a, b). The intraspecies analysis revealed that significantly fewer whole pupae were recovered from chambers than from cups for all species: S. cameroni (73 and 93%, respectively) F1, 22 = 13.15; P = 0.0015, S. endius (69 and 96%, respectively) F1, 22 = 19.80; P = 0.0002, and M. raptorellus (3 and 76%, respectively) F1, 22 = 76.25; P <0.0001 ( Table 3 2, Fig. 3 -4). Percent P arasitism Significant differences in percent parasitism were det ected only between parasitoid species in chambers (F4, 31 = 57.16 ; P <0.0001), with significantly more pupae parasitized by S. cameroni (56%) and S. endius (54%) than by M. raptorellus ( 1%) (Table 3 1, Fig. 3 -3a,b). N o significant difference was detected in parasitism in the within species analysis of S. cameroni or S. endius However, significantly more (F1, 22 = 138.15; P < 0.0001) pupae were parasitized in cups (49%) than release chambers by M. raptorellus (Table 3 2, Fig. 3 -4).
64 Percent PIM Signific ant differences in percent PIM were detected between parasitoid species in both the chambers ( F4, 31 = 11.16 ; P = 0.00 02 ) and cups (F4, 31 = 10.18 ; P = 0.000 4 ). There was also a significant difference (F2, 31 = 5.35; P = 0.0101) (Table 3 1) in percent PIM between replications for the cups, but not in the chambers. In both the release chambers and cups, percent PIM was significantly less in treatments containing M. raptorellus (2 and 20%, respectively) than either S. cameroni (14 and 39%, respectively) or S. endius (10 and 31%, respectively) with no significant difference between the two Spalangia spp. (Fig. 3 -3a,b). The intraspecies analysis revealed that percent PIM was significantly less in chambers than in cups for all species: S. cameroni (14 and 39% respectively) F1, 22 = 33.81; P < 0.0001, S. endius (10 and 31%, respectively) F1, 22 = 25.00; P < 0.0001, and M. raptorellus (2 and 20%, respectively) F1, 22 = 29.36; P <0.0001 ( Table 3 -2, Fig. 34). Discussion Our laboratory findings support those o f our field studies (Chapter 2) conducted in Ocala, Florida, where nearly 100% of all pteromalids recovered were Spalangia spp. This is particularly evident in the results observed for percent parasitism between container types (Fig. 3-3a,b). When parasi toids were forced to search for hosts buried in soiled horse bedding, pupae in release chambers containing either Spalangia spp. reproduced at a significantly higher level than M. raptorellus This wa s likely due to the depths at which most stable fly maggots ultimately pupated, and the ability of Spalangia spp. to search in the substrate, as parasitism rates between species were not significantly different in cups where pupae were freely accessible. The cylindrical release chamber was chosen to inhibit larval aggregation, as flies in our facility often pupate in the rearing container corners. This behavior would potentially alter the results
65 of the assay. Preliminary tests (data not shown) were conducted to ensure larvae distributed more or less evenly, and that aggregation was minimal. Although the depth at which every pupa recovered was not recorded from the chamber portion of the experiment p reliminary stud ies of stable fly maggot dispersal indicated that most pupation in these containers occurred at depths of 37 cm. Taken together, t his further corroborates my findings in Chapter 2, where field -collected stable fly pupae were collected at depths of 3 cm or greater and nearly 100% of all parasitoids recovered were Spalangia spp. Several studies demonstrating the effects of host dispersal and abiotic factors on pupal parasitoid activity may in part explain the findings of our laboratory experiments. A field study conducted by Rueda and Axtell (1985b) demonstrated that most M. raptor were recovered from pupae collected at depths of 3 cm or less, whereas most Spalangia spp. were collected between depths of 5 and 10 cm. King (1997), conducting laboratory evaluations of both M. raptor and S. cameroni determined that host burial greatly reduced parasit ism by the former species, whereas that of the latter was relatively unaffected. This is similar to results obtained by Floate and Spooner (2002), when pupal parasitism by three Muscidifurax spp. greatly decreased if hosts were located at depths of 1 cm o r greater. In a study conducted by Geden (2002), both S. cameroni and S. endius searched uniformly through a commonly used fly rearing medium, and regularly located hosts at 6 cm depths in this porous, relatively loose substrate. Pupae were also attacked by M. raptor at 6 cm depths, but only half as often as Spalangia spp. The substrate used in our assays more closely approximates the fly rearing medium than the dense sandy soil or manure also evaluated by Geden (2002).
66 In the present study, t he soiled horse bedding used in release chambers also included horse manure, areas from which stable fly pupae were regularly recover ed. This behavior was also noted in my field studies assessing pteromalid pupal parasitoids attacking naturally occurring stable fly pupae. Although one or two stable fly pupae could be found in these areas aggregation in horse manure was inhibited in release chambers due to its smaller particle size after mixing. The dense nature of the horse manure may have dissuaded attack by M. raptorellus at any depth in our study, while favoring attack by either S. cameroni or S. endius Smith and Rutz (1990) and Geden (1999) also have demonstrated the impact of both light and moisture on Spalangia spp. and Muscidifurax spp. In both studies, M uscidifurax spp. preferred drier substrates, whereas Spalangia spp. preferred those with higher moisture content. However, both species have been shown to prefer dimly lit conditions (Smith and Rutz 1990). Because light conditions in the present study were similar for all treatments, negating its effect, it is likely that the 60% average moisture content in the current study is another factor that resulted in the increased searching activity of Spalangia spp. over that of M. raptorellus In the present st udy, percent whole pupae and percent PIM were significantly greater for S. cameroni and S. endius than for M. raptorellus for both the chambers and the cups, although differences were greatest between species in the cups. The percent whole pupae were thos e pupae that did not initially produce an adult fly, or the overall killing effect of parasitoids. The percent PIM, or pupal mortality due to some factor other than the emergence of a parasitoid, such as host feeding, was significantly higher in release c hambers containing either Spalangia spp. than those with M. raptorellus
67 Causes for increased PIM other than host feeding may include differences in ovipositional restraint. Wylie (1971, 1972a) demonstrated that Muscidifurax zaraptor Kogan and Legner dis criminated against previously stung hosts more often than S. cameroni Petersen et al. (1991) also described this cause of PIM in detail with several Muscidifurax spp. and Spalangia nigroaenea Curtis Under various conditions, Petersen et al. (1991) foun d that Muscidifurax spp. display ed great ovipositional restraint when encountering previously stung hosts. However, S nigroaenea demonstrated little ovipositional restraint under similar conditions. Therefore, differences in ovipositional restraint may be the cause behind differences in the levels of PIM between the two genera used in the present study. Differences in PIM between the two genera were also greater in cups, than in chambers. The difference in frequency of host encounters was greater in sma ller cups where pupae were more freely accessible than in chambers which required greater searching effort This may have further accentuat ed differences in ovipositional restraint between the two genera. Furthermore, the gregarious nature of M. raptore llus may be similar to that shown by Nasonia vitripennis (Walker) (Wylie 1972b), where eggs of other species are not generally attacked. Instead, the speed and efficiency of N. vitripennis host utilization often results in starvation of competing species. Under field conditions, this may also be true for M. raptorellus T he stinging of pupae containing previously developing parasitoid larvae would not result in increased PIM, but fewer offspring, due to insufficient resources This may be a potentially disadvantageous situation for livestock producers releasing M. raptorellus during times when parasitoid to host ratios are high. U nder our conditions, if M. raptorellus did sting pupae with
68 conspecific eggs, it also would not have increased the percent PI M, but rather increased the numbers of subsequent adult parasitoids It is possible that the temporal window in the present study (3 d) would not be as adequate for decreased adult production as under field conditions where hosts and immature parasitoids of all ages are available. Further evidence of increased host encounter frequency in cups compared to chambers is provided by our intra -species analysis of the aforementioned factors (Fig. 3 -4). No significant differences in parasitism were detected betw een container types for either S. cameroni or S. endius suggesting that these species located similar numbers of hosts regardless of the searching effort required. However, this analysis revealed that M. raptorellus was less efficient in locating hosts w hen required to search in release chambers. The percent whole pupae recovered and percent PIM was significantly less in chambers than in cups for all species. This further confirms the likelihood that host encounter frequency was greater in cups than in the chambers This would accentuat e differences in host -feeding and ovipositional restraint behavior between the two genera. In a few cases, significant differences were detected between replications for percent whole pupae recovered and percent PIM. How ever, these differences only occurred between replications of cups. This discrepancy can be explained by our use of a newly established stable fly colony. Although it was our intent, it was rather difficult to collect only post -feeding third-instar stabl e flies from our rearing media. Pupariation in this colony is not completely synchronized and takes place continually between days 10 and 14 post oviposition. In general the feeding status of larval stable flies was determined by their aggregation behavi or. Because some larvae were still feeding, higher control mortality resulted from placing larvae in substrate free cups than in the
69 chambers. Those larvae still requiring nutrition to pupate were able to continue feeding in substrate -containing chambers resulting in lesser and more uniform control mortality. Several studies evaluating the behavior or ability of pteromalid parasitoids to locate hosts under laboratory conditions have been used as supporting evidence for the findings in the present study. However, most of these studies utilized small (approximately 100 ml) release chambers, or artificially -made substrate conditions to evaluate parasitoid activity. Furthermore, all of these studies utilize previously pupariated pupae, placed in the release chamber at predetermined locations. This study is to my knowledge, the first evaluation of pteromalid parasitoid host location using sizeable chambers that more closely approximate natural field conditions In addition, this is the first study allowing third-instar filth flies to wander and pupate as they would under field conditions using a substrate where known filth fly breeding occurred. My studies clearly demonstrate the ability of S. cameroni and S. endius to search more efficiently for hosts than M. raptorellus using a substrate where fly breeding occurred at a livestock farm. This experiment also corroborates our findings that host depth, due to the fly breeding substrates generated by equine husbandry practices (Chapter 2), is likely the cause for the high proportions of Spalangia spp. found in those field studies. Further experiments are needed to determine if releases of Spalangia spp. at equine facilities in Florida can increase filth fly control.
70 Table 3 1. Analysis of variance (ANOVA) F values for interspecies percent whole pupae, percent parasitism, and percent parasitoid-inducedmortality (PIM) of stable fly pupae in chambers containing soiled horse bedding, and cups with no bedding. Chambers Cups Species 1 Replication Species 1 Repli cation % Whole Pupae 2 80.90** 0.30 NS 6.09** 5.28* % Parasitism 3 57.16** 0.88 NS 2.26 NS 2.06 NS % PIM 4 11.16** 0.28 NS 10.18** 5.35* 1 Species evaluated included Muscidifurax raptorellus, Spalangia cameroni, and Spalangia endius. 2 The percent whole pupae were those pupae from which an adult stable fly did not emerge. 3 The percent parasitism were those pupae from which an adult or aborted parasitoid was recovered. 4 The percent PIM was the difference between percent whole pupae an d percent parasitism. **, P
71 Table 3 2. Analysis of variance (ANOVA) F values for and intraspecies percent whole pupae, percent parasitism, and percent parasitoid-inducedmortality (PIM) of stable fly pupae in chambers con taining soiled horse bedding, and cups with no bedding. M. raptorellus S. cameroni S. endius % Whole Pupae 1 76.25** 13.15** 19.80** % Parasitism 2 138.15** 0.78 NS 0.62 NS % PIM 3 29.36** 33.81** 25.00** 1 The percent whole pup ae were those pupae from which an adult stable fly did not emerge. 2 The percent parasitism were those pupae from which an adult or aborted parasitoid was recovered. 3 The percent PIM was the difference between percent whole pupae and percent parasitism. *, P P > 0.05. M = Muscidifurax ; S. = Spalangia. Figure 31. C hamber s (26 cm diameter, 9 cm height) filled with 3.7 L (7 cm) soiled horse bedding.
72 Figure 32. S ouffl cups (120 ml) containing 10 ml maple wood chips used to assess parasitism when stable fly pupae were freely accessible.
73 Figure 33. Comparison of three parasitoid species host attack parameters expressed as mean percent whole pupae, percent successful parasitism, and percent parasitoid -induced mortality (PIM) exp osed to stable fly pupae a) freely accessible in cups and b) dispersed in chambers containing 3.7 L (7 cm) of soiled horse bedding substrate. Within evaluation parameter, columns with the same letter are not significantly different by the Ryan -Einot -Gabri el Welsh following an analysis of variance M. = Mucidifurax S. = Spalangia. a b
74 Figure 34. Evaluation of a) Muscidifurax raptorellus b) Spalangia cameroni and c) Spalangia endius searching ability on host attack rates expressed as percent whole pupae, percent successful parasitism, and percent parasitoid-induced mortality (PIM). Stable fly host pupae were either freely accessible in cups or required searching in chambers containing 3.7 L (7 cm) of soiled horse bedding substrate. Within evaluation criteria, columns with the same letter are not significantly different by the Ryan-Einot -Gabriel Welsh multiple range a b
75 Figure 34. Continued. c
76 CHAPTER 4 HOST BLOOD MEAL IDEN TIFICATION OF STABLE FLIES COLLECTED FROM FLORIDA EQUINE FACIL ITIES USING A MULTIP LEX POLYMERASE CHAIN REACTION I ntroduction The detrimental effects of stable flies, Stomoxys calcitrans (L.), to confinement animal producers are well documented. Many studies have been conducted to ascertain the costs of stable fly attacks on dairy and beef cattle (Bruce and Decker 19 58, Miller et al. 1973, Campbell et al. 1977, Mullens et al. 2006). The effect of stable flies on other animals, such as pigs (Moon et al. 1987) and chickens (Anderson and Tempelis 1970), has also been determined. However, data concerning the ecology of this pest associated with equine facilities is limited. Sutherland (1978a) found that survival and fecundity of the stable fly was increased when fed bovine blood, although it was determined that they could survive and oviposit when fed equine blood. Fur thermore, Sutherland (1978b) found that stable fly immature mortality was less in horse manure than that of cattle manure. Although stable flies can survive under conditions provided by equine facilities, their presence at these facilities may be at least partially due to dispersal from other animal installations. Bailey et al. (1973) found that stable flies traveled up to 3.2 km in search of a blood meal. In a study conducted by Gersabeck and Merritt (1985), stable flies readily traveled upwind 0.8 km to confined horses from four release sites. Similarly to observations of Bailey et al. (1973), these flies remained in the general area because potential hosts and suitable larval substrates were available. Flies released from sites where hosts could not b e easily found were collected at distances up to 3.2 km. Additionally, Hogsette and Ruff (1985) documented that beach populations of stable
77 fl ies in Florida migrated as far as 225 km, possibly due to weather fronts. However, the understanding of stable f ly ecology and their seasonal distribution continues to be the subject of considerable debate. According to the American Horse Council (AHC 2009), horses are found in every state of the U.S., with 45 states having more than 20,000 horses. Florida ranks as having the third largest horse industry in the U.S., with 500,000 horses, contributing over $3.0 billion in goods and services, and maintaining over 38,000 full time employees. In addition, approximately 6 0% of these animals are used in showing and recre ational events. Stable flies are a common pest to horses and vector various pathogen s of veterinary importance, including those that cause pruritis and habronemiasis, the latter of which can lead to summer sores and secondary infections (Gortel 1998). Both of these conditions result in decreas ed overall animal aesthetics and thus their show value. The tremendous effort to control fly pests of horses is apparent in the plethora of chemical and mechanical products available to horse owners. Therefore, i t is surprising that research concerning stable flies and horse production is sparse, at best. Floridas rich horse industry provides a particularly suitable situation in which to monitor the localized movement of stable flies through host blood meal ident ification. Analysis of blood meals taken by haematophagous arthropods has played an important role in ascertain ing host preferences and capacity to vector blood -borne pathogens to both animals and humans (Abbasi et al. 2009, Kent 2009). The precipitin m ethod has been used to successfully identify hosts of mosquitoes (Bertsch and Norment 1983) as well as stable flies (Anderson and Tempelis 1970). Enzyme-linked immunosorbent
78 assays also have been used with success in mosquito blood meal identification (Zi nser et al. 2004). However, problems including the need for high quality antisera for the selected species and the need to refine antibodies to prevent cross -reactivity, as well as loss of blood quality over time compound the technical difficulties of th e se assays themselves (Ngo and Kramer 2003). More recently, the overwhelming increase in available DNA sequence data of various vertebrates has opened the door for newer molecular based blood meal analysis approaches such as polymerase chain reaction (PCR ) (Kent 2009). Although the use of PCR for host blood meal identification in stable flies has not been reported, the blood meals of other haematophagous arthropods have been successfully identified using a variety of laboratory techniques. Polymerase chain reaction utilizing mitochondrial DNA where many copies of the genome are present, can alleviate the problems of other bloodbased techniques that rely on more rarely encountered nuclear DNA of mammalian white blood cells, providing a more direct approa ch to host blood meal identification. Furthermore, primers targeting cytochrome b can be species specific, and have been used in the successful host blood meal identification of mosquitoes (Boakye et al. 1999, Kent and Norris 2005). Although stable flies are known pests of horses, the reasons for their occurrence in these areas either due to available breeding areas or dispersal from neighboring livestock facilities, are largely unknown. In Ocala, Florida, the large number of horse producing units offers the potential to demonstrate stable fly movement between cattle and horse installations using PCR blood meal identification. Beginning in November 2007, a study was undertaken to 1) determine the hosts of stable flies collected from
79 Florida equine facili ties and 2) determine whether stable fly host blood meal identification can be used to describe their short -term localized movement from off -farm sites. Materials and Methods Stable F ly C ollection. Between November 2007 and December 2009, weekly attempts were made to collect 10 adult blood-fed stable flies from each of four equine facilities located near Ocala, Florida as described in Chapter 2. Live a dult flies were collected from fence lines and barn walls of horse enclosures using a sweep net. Blood-f ed individuals were identified by applying light pressure to the sides of each fly, inducing production of a fecal droplet (Fig. 4 -1). Flies were considered suitable for analysis if they produced fecal droplets that were dark in color, suggesting digestio n of a recently acquired blood meal. These flies were placed individually into clean, labeled 1.5ml microcentrifuge tubes and held on ice to slow further digestion prior to processing Retained stable flies were returned to the University of Florida (UF ) Veterinary Entomology Laboratory and stored at -80 C until blood meal analysis could be performed. Time C ourse B lood M eal A nalysis Approval was granted from the appropriate UF committee for the collection of blood from each mammalian species used in s table fly feeding assays prior to beginning this study. Approval was granted from the UF Institutional Review Board (#342-2008) for collection and use of human blood in feeding assays (Appendix A). Cattle blood used in this project was collected with the approval of the UF Animal Research Committee (#018 -ANS08). Horse blood used in this project was obtained from the UF Horse Teaching Unit as part of a routine Coggins test
80 performed at the facility. The UF Institutional Animal Care and Use Committee approval was necessary for collection of dog blood (#200801760) (Appendix B) Previously non-blood-fed female stable flies (3-5 day old) were engorged on blood of individual selected host s which included cattle, horse, dog, and human, as well as a blood mixt ure of all four hosts to assess the time -dependent detection limits of the developed multiplex PCR. Stable flies from a colony established in February 2007 from individuals collected at the UF Dairy Research Unit in Hague, Florida, were reared according to the methods described in Chapter 3 Adult stable flies were mechanically aspirated from colony cages and placed into 120ml plastic feeding chambers in groups of 10 (Fig. 4-2). A 200 -l sample of blood from each host was added individually by micropip ette to the cap of a 1.5-ml microcentrifuge tube and attached to a screened area of each feeding chamber with a rubber band (Fig. 4-3). For the mixedhost feeding chambers, 50 l of blood from each host was mixed and added to a microcentrifuge cap and att ached to its respective feeding chamber. Stable flies were allowed to feed for 20 min, and held for 0, 8, 16, 24, and 48 hr post blood -feeding at 26 C, 12:12 LD, and 70% RH. At the appropriate time, flies from each designated feeding chamber were indivi dually placed into 1.5 ml microcentrifuge tubes and held at -80 C until blood meal analysis could be performed. Only stable flies that had fully engorged during the initial 20min feeding period were analyzed. E xperiments were conducted three times to e nsure the accuracy of blood meal determinations. DNA Extraction Stable fly host DNA extractions were performed with the QIAGEN DNeasy Blood and Tissue Kit (QIAGEN, Valencia, CA). The abdomen of a previously frozen blood -fed stable fly was removed and placed in a 1.5-ml
81 buffered saline, pH 7.4, and homogenized with a micro tube pestle. At this point, DNA extractions were carried out by following the insect DNA extraction protocol provided with the kit with the only modifi Primer D esign and PCR Primers (Sigma Genosys, St. Louis, MO) targeting the cytochrome b region of the mitochondrial genome of selected hosts were designed manually using a multiple alignment of sequences (CLUSTALW 2009) selected from GenBank. These included: cattle ( Bos taurus ; Accession #DQ186290), horse ( Equus caballus ; Accession #NC001640), dog ( Canis lupus familiaris ; Accession #NC002008), human ( Homo sapiens ; Acces sion #NC012920), and stable fly ( Stomoxys calcitrans; Accession #DQ533708). The horse, dog, and human hosts were selected as potential targets for PCR as they are commonly associated with equine facilities in Florida. The equine facilities in this study were chosen because they were large tracts of land (Chapter 2), known to be in the vicinity of cattle facilities. Therefore, cattle were selected as a fourth host due to their occurrence around but not on the selected equine facilities. Blood -feeding on this host was used as an indicator of potential stable fly movement between farms. Four host -specific forward primers were designed to have at least three nucleotide differences in the 3 region in the cytochrome b gene between each other, as well as the stable fly sequence. A universal reverse primer was designed in a consensus region to all four hosts, and all primers were checked for melting temperature compatibility. Additionally, these primers were selected to have expected product sizes that diffe red by approximately 100 bases (Table 41), making the source identification more certain when visualized on an agarose gel.
82 Each multiplex PCR contained final concentrations of: 20 mM Tris -HCl (pH 8.4), 50 mM KCl, 2.5 mM MgCl2 Carlsbad, CA), 0.2 pM HorseF and CattleF, 0.1 pM DogF and HumanF, 0.6 pM autoclaved water. Conditions were optimized for target DNA amplification using a touchdown PCR procedure with the following conditions: initial denaturation for 5 minutes at 95 C, followed by 12 cycles of 94 C for 30 seconds, 57 C for 30 seconds, and 72 C for 50 seconds. Two additional sets of 12 cycles each followed, using decreasing temperatures of 56 and 55 C as annealing temperatures, respectively. A final elongation was performed at the end of the 36-cycle program for 5 minutes. Amplification of stable fly host DNA by PCR was performed in a Bio Rad DNA Engine Peltier thermal cycler (Bio -Rad, Hercules, CA), followed by gel electrophoresis using an ethidium bromide stained agarose gel (1.5%) and a 100-base pair molecular weight standard (Invitrogen, Carlsbad, CA) to confirm amplification prod uct size. To confirm that amplification products were from the selected hosts, four target bands comprising each host type from both control experiments and field-collected stable flies, were removed from agarose gels and prepared for sequencing. DNA was extracted from agarose bands with the QIAGEN QIAquick Gel Extraction Kit following the instructions included with the kit Gel extractions were carried out on samples from both the time course control experiments, as well as samples from each host colle cted during the field portion of the project. The DNA obtained from gel extractions was sequenced using the Big Dye Terminator v3.1 Cycle Sequencing Kit (Applied Biosystems, Foster City, CA) using the same primer sets as in the initial multiplex PCR.
83 Th e amplification products obtained through this procedure were sent to the UF Interdisciplinary Center for Biotechnology Research (ICBR) to be analyzed using the 3130 Genetic Analyzer (Applied Biosystems, Foster City, CA). The sequences determined by the I CBR through this analysis were returned as electropherograms and edited using Sequencher 4.8 software (Gene Codes Corp., Ann Arbor, Michigan). The edited sequences of stable fly hosts from both lab and field specimens were confirmed using the Basic Local Alignment Search Tool (BLAST) available on the National Center for Biotechnology Information (NCBI 2009 ) web site. Statistical A nalysis. Data from blood meal identification of field-collected specimens w ere subjected to analysis of variance (ANOVA) using the PROC GLM procedure of SAS 9. 2 (SAS Institute 2004 ) to determine differences in host -feeding frequencies by stable flies at Florida equine facilities, as well as differences in host feeding frequencies between farms. Any farm in which at least five bl ood -fed stable flies had been collected during a weekly attempt was included in the analysis. A twosample t -test was conducted to determine if differences between host detections existed between sample collections of four or fewer flies and five or more flies. Due to the difficulty in collecting blood -fed stable flies from the facility, there was only one week in which Farm 1 met the five -fly criterion. Therefore, Farm 1 was rem oved from the between -farm host -feeding frequency analysis. Data were trans formed using an arcsine square root of the percent host type composition. All data are presented in figures as back transformed means. Both host and farm were included as fixed effects in their respective analysis. Multiple mean comparisons were conducte d with the Ryan Einot -Gabriel
84 To assist in interpretation of any differences in stable fly host -feeding frequency between farms, an effort was made to identify areas within 3 km of each farm in which horses and/ or cattle could be found. Any pasture or fenced enclosure holding either of these animal types was noted on a map created using Google Earth v. 5.0 (Fig. 4 4a, b, c, d). Results Time C ourse B lood M eal A nalysis The multiplex PCR designed for stable fly blood meal analysis was used successfully in identification of selected hosts in the laboratory time course evaluation. Animal specific amplification products were detected with 100% efficiency up to 16 hr post -blood -feeding, with no cross amplification o f nontarget hosts, or amplification of any host in nonblood -fed stable flies (Fig. 4-5, 46). Host identification was also possible at 24 hr, but in only approximately 50% of the flies assayed (Fig. 4-7). Furthermore, these amplification products were difficult to visualize No amplification products resulted from stable flies assayed at 48 hr post blood -feeding (data not shown). Using the described methods, the time course experiment demonstrates that between 24 and 48 hr post -feeding, insufficient h ost DNA for successful amplification remains in the stable fly gut. However, this technique is 100% reliable if a stable fly is tested within 16 hr post -feeding. Host DNA sequences obtained through extraction and sequencing of bands from agarose gels were entered into the BLAST program found on the NCBI website to confirm host specificity of the selected primers. All hosts were confirmed using this method, sharing 9899% identities with the intended targets in both laboratory and fieldtested stable flies
85 Field-C ollected Stable F ly H osts A total of 595 field-collected and, presumptive by visual inspection, blood-fed stable flies were subjected to blood meal analysis during this study. Of those, host amplification products were successfully detected in 350 flies representing a detection efficiency of 58.8%. However, only 291 of the 350 positive flies collected during the collection period were used in the analysis due to our statistical requirements. A two-sample t -test between collections of four or less flies and five or more flies showed no significant differences in cattle (65.8 and 85.8%, respectively) or horse (25.8 and 12.7%, respectively) host utilization. Multiple host -feeding was detected in 26 of the 350 stable flies analyzed (7.4%), with t hree of those flies having sufficient DNA from three host types. When overall host -feeding frequency of stable flies collected from equine facilities was examined, significantly more had fed previously on cattle (65%) than any other host (F3,68 = 43.90; P = <0.0001) (Fig. 4-8). Significantly more stable flies had fed previously on horses (28.8%) than dogs (0.5%) or humans (5.3%), and no significant difference was detected between dog and human feedings. Significant differences in host -feeding frequency be tween farms were also detected for cattle (F2,32 = 18.18; P = <0.0001) and horses (F2,32 = 19.27; P = <0.0001) (Fig. 4 -9). Significantly more stable flies that had previously fed on horses were collected from Farm 4 (46.3%) than from any other farm. Sign ificantly more stable flies collected from Farm 3 (98.0%) than flies collected at any other farm had previously fed on cattle. No significant differences were detected in dog or human host feeding by stable flies between farms, and these detections were i nfrequent. Although data from Farm 1 were not included in the analysis, there were 13 cattle and three human blood-
86 fed stable flies collected from this facility with no host identifications from the horse or dog. The results of the effort to identify cat tle and horse populated pastures enc ircling the four horse farms where stable flies were collected are shown in Fig. 4-4a, b, c, d. There were at least two pastures holding cattle within 2 km of all farms. In general, Farm 3 which had 98% cattle blood meals, was associated with the most cattle pastures, having a total of 20-cattle populated pastures within 3 km of fly collection sites Farm 4, with the highest horse blood meals (46.3%) was associated with the fewest nearby cattle-populated pastures, ha ving a total of five within 3 km. Horse populated pastures were abundant around all four farms where stable flies were collected. Discussion To my knowledge, stable fly host blood meal identification using multiplex PCR has not been documented. My post -f eeding, time -series assays demonstrate that successful amplification using host -specific primers targeting cytochrome b is 100% reliable up to 16 hr post -feeding under laboratory conditions, and suggests this efficiency continues to some point before 24 hr Similarly, between 24 and 48 hr post feeding DNA degradation in the stable fly gut decreases the detection efficiency from approximately 50% to 0. This is similar to mosquito host blood meal detection as performed by Kent and Norris (2005). T he autho rs surmised that differences in detection limits between their assay and others could be due to differences in extraction protocol, arthropod digestive systems, or the hosts upon which the insect had fed. Ngo and Kramer (2003) were able to detect mosquito blood meals taken from quail for up to 3 d post -blood-feeding. However, their assay utilized both primers specific to
87 cytochrome b and a restriction enzyme digestion to distinguish different avian hosts. Although not used in the study by Ngo and Kramer (2003), the efficiency of primers targeting nuclear DNA in birds might be greater because avian blood is nucleated, and thus provides more DNA amplification than similar primers for mammalian blood. The results reported here for host blood meal identifica tion are consistent with those of other studies regarding the fate of a blood meal in the stable fly gut. Anderson and Tempelis (1970) reported that on average, the time required to fully digest a blood meal was 2436 hours when fed on citrated human blood. However, stable flies used in that study were maintained at temperatures of 2021 C. Digestion of cattle blood was longer still, taking 46 -70 hr when held at cycling temperatures between 21 and 15 C (max/min). Hafez and Gamal Eddin (1959), as repor ted by Anderson and Tempelis (1970), demonstrated similar results for human blood at 21 C, but noted the decreased digestion time of 10 hr when cattle blood -fed flies were held at 25 C. Hafez and Gamal -Eddin (1959) also demonstrated a 2 d decrease in pr e oviposition period in stable flies held at 30 C than those held at 22 C. These findings are consistent with a review by Lehane (2005), who proposed a rough average time of 48 hr for complete digestion, but noted that many factors can alter this time f rame, including temperature, blood meal size and host type, age, and mating and gonotrophic status. The holding conditions in my time -series analysis may explain the reason the assay became relatively unreliable at 24 hr. However, the holding conditions in the current study more closely represent those present when stable flies are active in Florida, making any conclusions drawn from these field-collected specimens appropriate.
88 In the present study, analysis of field-collected stable flies resulted in approximately 59% detection efficiency. Many studies have documented the response of cattle to the painful bites of stable flies. Dairy cattle exhibit various defensive behaviors including stomping, head throwing, skin twitching, and tail switching (Miller et al. 1973, Mullens et al. 2006). Th e s e behavior s, including bunching, are not unique to cattle and are often observed by horse producers. Defense behaviors may, in part, explain the disparity between the laboratory and field efficiency of our assay. I n the laboratory assays, stable flies were allowed to feed to repletion, and in theory imbibed the maximum amount of DNA possible in a single feeding event. Defensive behaviors by livestock in response to painful bites often dislodge stable flies and end a feeding event abruptly causing them to move to another host or a nearby resting site. This activity is expressed in my data by the numerous multiple host detections among the field -collected stable flies. However, interrupted feeding may also result in lower volume blood meals, and therefore, decrease the time taken for digestion and DNA degradation. In addition, pressure induction of a fecal droplet to assess blood-feeding status did not guarantee the fly was actively digesting a recent blood meal, and instead, may have been within a few hours of complete digestion. The fecal droplet method was developed to thwart the problem of actively selecting flies having no visible signs of a blood meal, as well as a low labor alternative to dissections. Alth ough flies having distended, red abdomens were rarely encountered, they were also avoided, as they were likely to have fed on nearby horses. These stable flies would not provide information regarding short term localized m ovement between equine and nearby cattle farms.
89 Among the primary goals of this study was an attempt to determine the hosts utilized by stable flies collected from Florida equine facilities. Significantly more stable flies were collected that had previously fed on cattle than any other host, and on average made up 65% of the detections for any given week at any given farm. This is surprising, as all of the stable flies analyzed were collected in the direct vicinity of and at a central location within each horse facility. T hese findings do not necessarily suggest that cattle are the preferred host of stable flies. Rather, t he large proportion of cattle feedings may be due to their tolerance of stable fly biting activity over time. Most studies involving stable flies and livestock have been conducted in an attempt to quantify losses in cattle production due to the relentless biting activity of pests. While studies of growing beef cattle have demonstrated economic losses due to 1020% in weight -gain reductions (Campbell et al. 1977, 2001, Catangui et al. 1993), results of studies concerning dairy cattle have been conflicting. Bruce and Decker (1958) reported monthly reductions in milk and butterfat content of 0.7% per stable fly, per cow. Alternatively, Miller et al. (1973) and Mullens et al. (2006) demonstrated that defensive behaviors by dairy cattle did not affect production and further more, that these behaviors decreased over time. A search of the literature provided little in the area of biting fly pests and horse behavior. However, Keiper and Berger (1992) documented defensive behavior including tail switching and alternative refuge seeking by feral horses in response to horse flies and other biting fly pests. They also noted that this behavior continued whenever biting activity oc curred. Although speculation, this would suggest that horses
90 do not acclimate to stable fly biting activity as cattle do, and may have resulted in fewer collections of flies that had previously fed on horses. A s tudy of muscoid flies found at equine facil ities by Burg et al. (1990), indicated that stable flies were regularly captured using CO2baited traps, but were not observed on pastured horses. DuToit (1975) and Sutherland (1978a) determined that while stable fly longevity and reproduction was maximiz ed when reared on cattle blood, only moderate performance was achieved when reared on horse blood. In Chapter 2 of this dissertation we document the availability and use of husbandry generated equine waste products as breeding sites for stable flies. Additionally, a study by Sutherland (1978b) demonstrated that immature stable fly survival was greatest using horse manure substrates over other types, including cattle. This was confirmed in tests conducted by Boire et al. (1988), demonstrating equal or be tter immature stable fly performance on horse manure when compared to that of cattle. Further study by Jeanbourquin and Guerin (2007) found that in all tests, stable flies chose horse dung when given a choice between cattle or horse dung types. These res ults, as well as those of the present study, suggest that horse feeding by stable flies is infrequent. Whether this phenomenon is due to defensive behavior or host preference remains to be determined and requires further study. Th e s e data also suggest that stable flies are emigrating from nearby cattle installations. Although they may be breeding on off -farm sites as well, stable fly production did occur o n each farm. It may be that stable flies are highly attracted to onsite breeding areas due to their preference for horse manure substrates, occurring as intermittent biting pests that emigrate back to cattle farms for their primary blood source.
91 Because stable fly activity was not monitored at nearby cattle farms, this theory remains to be proven. How ever, the results of this study do prove stable fly movement in some capacity, between farm types. The second goal of this study was to determine whether differences in host feeding existed between farms. Although differences in cattle -feeding frequency b etween farms were detected, collections of stable flies previously fed on cattle blood predominated other blood types on all farms, with the exception of Farm 4. These results are similar to findings by Anderson and Tempelis (1970), documenting 98% of the stable flies captured on poultry farms to have previously fed on cattle. The other 2% were blood-fed on two horses and a dog, with no blood meals taken from chickens. The authors documented the nearest cattle to any poultry ranch in the investigation varied in distance from 0.4-0.8 km. Maps were created noting cattle and horse pastures to a distance of 3 km to assist in interpretation of findings in the present study (Fig. 4-4a, b, c, d). The nearest cattle pastures were located approximately 1.5, 0.8, 0.8, and 1.5 km from Farms 1, 2, 3, and 4, respectively. However, Farm 3 was associated with the most cattle pastures, followed by Farms 2, 1, and 4, respectively. That Farm 4 had the fewest cattle and the highest horse -feeding results suggests that catt le distance and abundance plays a significant role in fly movement from cattle to horse farms. Although stable fly breeding sites were located on each farm in the study, Farm 4 was also the only facility that maintained horses on round hay bales, a prefer red breeding substrate for stable flies (Broce et al. 2005, Talley et al. 2009). Furthermore, stable flies from Farm 4 were most often collected during times when on-site fly production occurred. Stable fly breeding
92 was encountered at other farms, but le ss often. However, t his does not mean that breeding attempts by flies were not as numerous. Flies were abundant and present in numbers not generally justified by the breeding activity observed, suggesting that those farms may not be producing all residen t stable flies. The observation of regularly practiced cultural controls at other farms may have also limited or destroyed stable fly development areas. Stable fly dispersal has been documented in several studies (Bailey et al. 1973, Gersabeck and Merrit 1985, Hogsette and Ruff 1985), and the distances potentially traveled in the present study are not out of line with these reports However, this is the first documentation of a method to monitor stable fly dispersal within a 24 hr time frame. The multipl ex PCR developed in this study was 100% reliable when stable flies were tested within 16 hr post -blood-feeding. Therefore, we propose a relationship between the reliability of our PCR method and the distance traveled by stable flies from nearby cattle pas tures. The results from Farm 4 and Farm 1 would suggest that within 16 hr, stable flies traveled at least 1.5 km in search of hosts or breeding sites. The results of this study have far -reaching implications. One potentially important impact of this proj ect will be in the area of integrated pest management. It is apparent that stable flies readily travel between horse and cattle farms and are capable of such dispersal within a 24 hr period. Therefore, it is likely that equine facilities are not solely r esponsible for the production of all on-site stable flies. In addition, other filth flies, such as the house fly, Musca domestica (L.), have demonstrated their ability to become resistant to many of the chemical controls used against them (Scott et al. 20 00, Mar on et al. 2003). Kaufman et al. (2001b) determined that house fly insecticide resistance
93 was widespread among dairies in New York state, and that resistance levels were similar, indicating between-farm dispersal. House fly dispersal between lives tock units in Florida may also occur, compounding fly control efforts among livestock producers. Some livestock producers utilize commercially available pteromalid pupal parasitoids to manage fly populations, with many using chemical control measures. Th ese management practices on equine facilities may be ineffective if on-site fly breeding is limited and between -farm dispersal is occurring. This study provides a foundation for further research. Blood meal analysis of stable flies collected from cattle f arms may provide additional evidence of dispersal between livestock facilities, and may direct research to areas where substantial fly breeding occurs. Modification of our procedure to include the use of microsatellites may pinpoint origins of different s table fl ies due to off -site developmental areas or dispersal, or identify farms and pastures having particularly attractive hosts. Table 4 1. Primer sequences targeting the cytochrome b region of the mitochondrial genome of mammals used to identify stabl e fly hosts with a multiplex polymerase chain reaction. Primer 5 3 Sequence Melting Temp. C Product Size Cattle ttatcatcatagcaattgcc 57.6 400 Horse ccctacatcggtactaccc 58.3 499 Dog agcctatattacggatcctatg 57.7 658 Human ctcggcttacttctcttcc 58.2 273 UnivRev agtgggygraatattatgc 58.9
94 Figure 41. Pressure induction of a fecal droplet from a blood -fed stable fly.
95 Figure 42. Feeding chambers used for stable fly host blood meal identification time course analysis. Each chamber is 120 ml and contains 10 female stable flies (3 -5 day old). Chambers were inverted during blood -feeding.
96 Figure 43. Stable fly feeding chambers with attached cap from a 1.5 ml microcentrifuge tube filled with 200-l host blood.
9 7 Figure 44. Pastures enclosing horses ( dotted) and cattle ( solid ) within 3 km of each equine facility. Rings following the bull s eye represent 1-km distances from each farm: a) Farm 1, b) Farm 2, c) Farm 3, and d) Far m 4. a b
98 Figure 44. Continued. c d
99 Figure 45. Agarose gel results of time course blood meal analysis. Outer lanes (left and rightmost) are 100base pair molecular weight standards. Lanes 2 -6 represent stable fly host blood meal identification at 0 hr post -blood-feeding on dog, horse, cattle, human, and mixed -blood types, respectively. Lanes 7 11 represent blood meal identification at 8 hr post blood -feeding on the same host blood types. Figure 46. Agarose gel results of time course blood meal analysis performed at 16 hr post blood -feeding. Lane 1 (leftmost) is a 100 base pair molecular weight standard. Lanes 2 -6 represent blood meal identification of stable flies fed on dog, horse, c attle, human, and mixed blood types, respectively. Lane 7 demonstrates the absence of non-target amplification from a nonblood-fed stable fly.
100 Figure 47. Agarose gel results of time course blood meal analysis performed at 24 hr post blood -feeding. Lane 1 (leftmost) is a 100 base pair molecular weight standard. Lanes 2 -5 represent blood meal identification of stable flies fed on dog, horse, cattle, human, and mixed blood types, respectively. Black arrows indicate faint bands present in lanes 2 and 3.
101 Figure 48. Mean percent blood type composition determined by stable fly blood meal identification using a multiplex polymerase chain reaction. Means represent percent host blood meal composition of stable flies collected across farms and dates Means with the same letter are not significantly different (RyanEinot -Gabriel 3,68 = 43.90; P = <0.0001 n = 291.
102 Figure 49. Mean percent blood type composition determined by stable fly blood meal identification using a multiplex polymerase chain reaction. Means represent percent host blood meal composition of stable flies collected from all collection events Due to the statistical constraints, Farm 1 contained insufficient data and was removed from the analys is. Within host means with the same letter are not significantly different between farms (Ryan -Einot Gabriel 2,32 = 18.18; P = <0.0001, Horse; F2,32 = 19.27; P = <0.0001.
103 CHAPTER 5 PERMETHRIN RESISTANC E STATUS OF THE STAB LE FLY IN FLORIDA: A CRITICAL UPDATE USIN G LABORATORY SELEC TIONS AND FIELD EVAL UATIONS Introduction Populations of insecticide resistant muscoid flies, especially house flies, Musca domestica L., and horn flies, Haematobia irritans (L.), are known to occur throughout the world. Scott el al. (2000) found resistant populations of house flies in poultry units to be highly correlated with insecticide use history. While the isolated nature of poultry houses may inhibit dispersal of resistant flies to other facilities, Kaufman et al. (2001b) surmised that open cattle d airies may readily promote dispersal of potentially resistant flies Although resistant populations of horn flies are found across the United States, dispersal of this pest may be limited due to its ectoparasitic feeding behavior (Schmidt et al. 1985, Crosby et al. 1991, Cilek et al. 1991, Kaufman et al. 1999, Barros et al. 2001). Reports of horn fly dispersal and resultant spread of resistance genes are sparse (Sheppard and Joyce 1992). Of the predominant muscoid livestock pests including the horn fly, house fly, stable fly, Stomoxys calcitrans (L.), and face fly, Musca autumnalis De Geer, only the horn fly and house fly have been pressured enough to demonstrate their ability to become resistant to a wide array of insecticides. Furthermore, the mechanism s behind insecticide resistance in both species appear to be similar and include expression of knockdown resistance ( kdr ) and metabolic detoxification. Knockdown resistance is conferred through point -source mutations on the sodium channel gene resulting i n target -site insensitivity This mechanism has been demonstrated in house flies and horn flies using metabolic detoxification synergists such as piperonyl butoxide and diethyl maleate (McDonald and Schmidt 1987, Scott 1998). These point mutations have
104 a lso been shown to increase resistance expression relative to their increasing frequency (Guerrero et al. 1997, Smith et al. 1997, Jamroz et al. 1998, Lee et al. 1999). If the mechanisms behind insecticide resistance are similar in house flies and horn flie s, the closely related stable fl y may also possess such abilities against insecticidal control. However, only one incidence of insecticide resistant, field-collected stable flies has been reported (Cilek and Greene 1994). Mar on et al. (1997) found small resistance factors in field populations of stable flies but noted they were only as large as those observed between laboratory colony generations. The small amount of time spent on the host, coupled with a preference for the forelegs of animals may cons iderably limit stable fly exposure to applied insecticides. Insecticide resistance in the stable fly has not been detected to a large degree in the U.S. However, a large void in the literature exists for pesticide evaluations of fieldcollected stable fli es. Therefore, it is possible that insecticide resistance in this pest is occurring, and has merely been overlooked due to the general difficulty in controlling this pest with currently used techniques Furthermore, it has been over a decade since an att empt has been made to evaluate the susceptibility of stable flies to insecticides, in particular permethrin. A study was initiated to address the need for information concerning the permethrin susceptibility of stable flies. The primary objectives for th e study were to 1) determine the ability of the stable fly to become resistant to the commonly used pyrethroid, permethrin, using laboratory selections, and 2) evaluate field -collected stable fly susceptibility to permethrin susceptibilities using diagnost ic concentrations.
105 Materials and Methods Stable F lies. Two stable fly strains were used to generate preliminary dose response values used for these experiments. A stable fly colony maintained at the USDA -ARS-CMAVE in Gainesville, FL, served as the baseli ne susceptible strain (USDA-S) from which all resistance ratios were calculated. This strain has not been exposed to pesticides for approximately 30 years. A second stable fly colony, maintained at the University of Florida (UF) Veterinary Entomology Lab oratory, was established from wild flies collected from the UF Dairy Research Unit in Hague, FL (UFD -07 ) in February of 2007. This colony, maintained under conditions described in Chapter 3, served as the parental strain from which permethrin selections w ere conducted. Wild stable flies collected as pupae and evaluated using diagnostic permethrin concentrations provided flies at a similar life stage to prevent discrepancies in survival due to age. These sites included equine Farms 1 and 4 located near Ocala, F lorida (Chapter 2) and the UF Horse Teaching Unit (HTU) in Gainesville, F lorida Insecticide -C oated G lass J ars. Glass jars (60 ml) having a 4 -cm diameter, 4.4cm height, and 67.86-cm2 inside surface area were treated with 1 m l of serially dilut ed technical grade permethrin (98%, cis:trans 47.6:50.4, Chem Service, West Chester, PA) in acetone approximately 1 hour prior to fly exposure. All dilutions were prepared from a stock solution the day prior to testing. The jars were rolled on an unheated electric hot dog cooker for 30 min to allow for evaporation of acetone and uniform pesticide coverage. After the allotted drying time, 20, 3 to 5 -day old female stable flies were anesthetized with CO2 and transferred to each treated jar for a 4 hr expo sure period. Following exposure flies in each treated jar were again anesthetized, transferred to a clean 60 ml glass jar, and fed through a screened lid with a 2 cm length of dental wick
106 soaked with Gatorade. Mortality was assessed at 4, 24, and 48 hr after the initial exposure scoring ataxic flies as dead. Coated glass jar assays contained 68 concentrations, each with 4 jars. Jars coated with acetone only served as the untreated control. The entire experiment was replicated three times for a total of 240 femal e flies tested per dilution. Topically A pplied I nsecticide A set of 1:1 serial dilutions of 68 concentrations was prepared from a stock solution the day prior to testing. For each experiment, five samples of 15 female stable flies were weighed to obtain an average fly weight for the assay. Female stable flies were aspirated from colony cages, anesthetized with CO2, and sorted into groups of 15. All 15 flies in each group were transferred to a P etri dish placed on ice and administered 1 l of their respective permethrin concentration (98%, cis:trans 47.6:50.4, Chem Service, West Chester, PA) to the thorax using a Hamilton PB-600 repeating dispenser (micro applicator) (Hamilton, Reno, NV). These flies were transferred to clean 60 ml glass jars and assessed for handling mortality scoring ataxic flies as dead There were four groups for each dilution contained four groups, and each experiment was replicated three times for a total of 180 female flies per dilution. Because the same microapplicator syringe was used for the entire assay, topical applications began with the acetoneonly control groups and proceeded from low to high concentrations In addition, the micro applicator was cleaned in triplicate with acetone between each 15-fly gro up. Mortality was assessed at 4 and 24 hr, scoring ataxic flies as dead. Diagnostic Insecticide Exposure In addition to assays used to generate standard dose mortality data, field-collected and laboratory stable fly strains were
107 evaluated with diagnosti c permethrin concentrations applied to coated glass jars using the technique of Scott et al. ( 2000). These assays were performed similarly to those described for the insecticide -coated glass jar technique However, this assay contained only four diagnost ic dilutions (98%, cis:trans 47.6:50.4, Chem Service, West Chester, PA) derived from the USDA -S stable fly colony These dilutions included four jars each of 1 -, 3 -, 10 -, and 30X concentrations, where X was the previously established LC99 determined from the USDA -S colony This assay was replicated three times when possible, for a total of 240 female stable flies tested for each dilution. Because of the difficulty in collecting enough pupae to evaluate field strains, we required that each replication hav e three jars per dilution, for a total of 120 stable flies of either sex, using more when possible. When field fly populations were evaluated for permethrin susceptibility, concurrent evaluation of the USDA -S stab le fly strain was conducted to ensure dilution accuracy. Abbotts correction (Abbott 1925) was applied to all data from this assay to adjust for control mortality. Selection for Permethrin R esistance The UFD 07 stable fly strain previously colonized in February 2007 was used in an attempt to generate a permethrin-resistant stable fly strain (UFD -PR) This colony had completed 30 generations between the time it was collected and when permethrin selections began. Stable fly selections were conducted using the LC70 value previously determined for the UFD -07 strain from dose mortality data generated with treated glass jars. Generations produced from surviving selected individuals were inadequate to complete both subsequent selections and three insecticide treated glass jar replications to re evaluate permethrin susceptibility with the Fx+1 generation Therefore only every
108 other stable fly generation was subjected to permethrin selections. Additionally, due to low adult fly numbers resulting from the high selection pressure, it was often not possi ble to conduct three experimental replications to precisely determine permethrin susceptibility prior to the next selection This problem was alleviated using a standard set of six dilutions derived during preliminary topically applied insecticide assays. This allowed for the testing of fewer individuals to generate complete dose -response data. However, only two replications using this method were performed prior to the next selection (120 fema le flies per dilution tested). Before each selection, a new L C70 value was estimated through conversion of the values determined using the topical applicator results, to those estimated for insecticide -treated glass jars. Using the relationship determined between the LD and LC values of previous evaluations with th e USDA S and UFD 07 colonies, topical applicator LD values (g permethrin/mg insect) were multiplied by a mean factor of 33 to obtain equivalent values for treated glass (g permethrin/cm2). Stable fly selections were carried out using permethrin (98%, cis :trans 47.6:50.4, Chem Service, West Chester, PA) applied to the internal surface of 1.06 L glass canning jars Jar dimensions were 7.8 cm diameter, 12.5 cm height, and 339 cm2 inside surface area. Ten jars served as permethrin-treated controls to evaluate gender selection efficiency, with five each designated to assess male and female mortality separately. Each of these jars contained 250 individuals and was monitored for mortality separately from all other selection jars. In all cases, s table flies we re exposed to permethrin treated jars for a 4 hr period, at which time they were anesthetized and transferred to a rearing cage. F lies in the gender -designated jars were transferred to a
109 30 x 30 x 30 cm rearing cage and held for 48 hr after the initial ex posure prior to mortality assessment. After the 48 hr mortality assessment surviving flies were transferred to the primary rearing cage with those from the remaining jars. The majority of the selected flies were held in jars containing 250300 mixed-sex stable flies aspirated from rearing cages with a graduated vacuum tube. Mortality of stable flies in these jars was not assessed. Approximately 10,000 individuals were exposed during each selection assay. All surviving flies were held together followin g mortality assessment of gender -sorted treatment jars. Statistical Analysis. All dose -response data was subjected to standard probit analysis using the PROC PROBIT procedure of SAS 9. 2 (SAS Institute 2004 ) to generate LC/LD values used for resistance ratio calculations, permethrin selections, and diagnostic concentrations. Values for LC/LD determined by this analysis were considered significantly different if no overlap occurred between the 95% confidence interval s (CI) Comparisons were made at both t he LC/LD50 and LC/LD90 values. Resistance ratios were calculated as the LC/LD value of a particular strain, divided by that of the susceptible USDA strain. Results Colonized Stable F ly Strains. The LC/LD50 and LC/ LD90 values determined for all fly strain s evaluated is presented in Table 5 -1 and Table 5-2 The LC70 concentrations used for each permethrin selection and the subsequent respective percent mortality are shown in Table 5 3 Using treatedglass jars, t he LC50 and LC90 values for the USDA -S colon y were 0.0013 and 0.0022 g/cm2, respectively. Due to the nonoverlap of the 95% CI, s ignificant differences were identifi ed between both LC values for the UFD -07 parental
110 colony, and the resultant UFD -PR following the fifth selection. The resistance rat ios for the UFD 07 colony f ollowing the fifth permethrin selection had increased by 7 and 12fold at LC50 and LC90, respectively, over that of the USDA -S colony. However, because low level resistance previously existed in the parental UFD 07 colony, resi stance levels in the UFD PR had only increased by approximately 3-fold over that of the parental strain. Similar to experiments using insecticide-coated g lass, resistance ratios for the field strains as well as those of the UFD -PR generations evaluated pri or to permethrin selection, demonstrated significantly higher resistance levels than the USDA -S strain. Prior to selecting for permethrin resistance, experiments using topically applied insecticide demonstrated that the UFD -07 strain possessed approximately 4 and 6-fold resistance levels over the USDA -S strain at the LD50 and LD90, respectively (Table 5 2 ). As expected, the f ive selections conducted at incr easing concentrations (Table 5-3 ), resulted in fairly consistent increases in the resistance ratio as compared to the USDA S strain With the exception of the final selection, our methods to convert microapplied insecticide LD values to those used for coated glass were accurate. Preliminary tests of the expected values for the fi fth selection resulted in 100% fly survival at 4 hr, prompting an increase in the concentration from the determined 0.016 g/cm2, to 0.044 g/cm2. Overall, five laboratory selections resulted in a 15-fold increase in resistance levels as compared to the USDA -S strain at both LD50 and LD90, respectively were measured using the topical applicator The UFD -PR strain had also increased resistance levels to 5 and 3-fold that of the parental UFD 07 strain at LD50 and LD90, respectively.
111 Field-C ollected Stable F lies Initially, a n attempt was made to colonize four stable fly strains collected from each farm used in our field studies (Chapter 2) However, rearing success was achieved only in those flies co llected from Farm 1 (Chapter 2) where s ufficient stable fly numbers were obt ained to conduct three replications using the topically applied insecticide technique To obtain the needed numbers of stable flies for assays, flies were reared in the laboratory using the procedure described in Chapter 3. Resistance ratios for flies fr om Farm 1 after five generations were approximately 9and 14 -fold that of the USDA -S strain at LD50 and LD90, respectively Stable flies from the UFD 07 UFD -PR Farms 1 and 4, and the HTU strains were also evaluated using diagnostic concentrations (Fig. 5 1). Survival at 1X LC99 USDA -S ranged between 59 and 93%. Although decreased at 3X LC99 USDA -S, survival ranged from 10% for the UFD 07 colony to 57% in stable flies collected from Farm 4. Survival at 10X LC99 USDA -S, was greatest for flies collected at Farm 4 and the HTU. In all evaluations only one fly, collected from the HTU, survived at 30X LC99 USDA S. Discussion My laboratory and field evaluations provided a critically needed update on the status of permethrin susceptibility in stable flies. T o my knowledge, only two studies, conducted more than a decade ago by Cilek and Greene (1994) and Mar on et al. (1997) document the possibility that stable flies may express insecticide resistance as do their muscoid counterparts, the house fly and horn fl y (Scott et al. 2000, Kaufman et al. 2001b, Mar on et al. 2003). The LC50 values determined in the present study were 3.5 and 13-fold greater than those used by Cilek and Greene (1994), for our UFD 07 and UFD PR colonies, respect ively Our results also differ from those of Mar on et al.
112 (1997). Furthermore, evidence from our selection experiments demonstrate that under heavy selection pressure, as few as five permethrin selections can increase resistance levels up to 5-fold that of the parental strain, w ith a comparative total increase of 15-fold that of a susceptible strain. Several known mechanisms, such as detoxification by P450 monooxygenases or hydrolases (Liu and Yue 2000), and target site mutations, such as those observed in the voltage-sensitive s odium channel gene (Smith et al. 1997, Lee et al. 1999) are poss ibly the cause of stable fly pyrethroid insecticide resistance Unfortunately, most research in this area has been conducted on the house fly and horn fly. Further study is needed to determi ne which, if any resistance mechanisms occur in the stable fly. However, during our assays using insecticide-coated glass jars, it was often difficult to achieve results that fit the model under the assumptions of probit analysis. I believe this is large ly due to the behavioral avoidance displayed by stable flies, particularly in the post -selection evaluation of the UFD PR strain where stable flies were observed on the untreated lids of the glass jars This behavior is not uncommon and has been reported as a possible explanation for resistance observed in the horn fly (Lockwood et al. 1985, Zyzak et al. 1996). This cannot completely account for the cause of resistance in the stable fly determined in the present study, as even greater resistance ratios w ere determined using topically applied insecticides a technique that mitigates bias introduced by fly behavior The differences in resistance ratio s between resistant and susceptible house fly strains of other studies (Scott and Georghiou 1985, Liu and Yu e 2000) have been greater and more dramatic than those demonstrated for the stable fly in the present
113 study. However, resistance ratio increases over the parental house fly strains used in those selection studies are similar. One reason for this occurrence may be due to differences in the level of exposure between the two pests. Stable fly control has been historically difficult, with causes stemming from the repell e nt effects of pesticides (Hogsette and Ruff 1986), to their potential dispersal habits (H ogsette and Ruff 1985, Chapter 4). Undoubtedly, selection pressure on house flies and particularly horn flies has been greater than that of stable flies Besides the small amount of time stable flies spend in commonly pesticide-treated areas such as bu ilding resting sites and on animals few pesticide applications are made to animals specifically for stable fly control Resistance development in the horn fly has been particularly problematic (Cilek et al. 1991, Byford et al. 1999, Foil et al. 2005), but is due essentially to the cons tant selection pressure th at horn flies receive as a result of their continuous onanimal behavior My studies of dispersal based on blood meal analysis of stable flies (Chapter 4) also suggest that stable flies may attack hosts and breed in different locations, increasing the difficulty of their control and provide a means for dispersal of resistance genes in the population The potential control problems attributed to stable fly dispersal have been observed in populations of the house fly. Studies of insecticide resistance in house flies of poultry facilities in New York state, demonstrated that resistance was highly correlated with the pesticide use history of each facility, possibly due to limited house fly dispersal fr om the relatively isolated environment of poultry houses (Scott et al. 2000). However, s tudy of more easily accessible dairy facilities may readily permit dispersal of resistant house flies, impacting the effectiveness of their chemical control at nearby
114 dairies (Kaufman et al 2001b). Evidence for this phenomenon has also been observ ed in the present study. Stable flies were collected from Farm 1 and evaluated both topically and with diagnostic surface concentrations. In topical assays, this colony dem onstrated 9and 14-fold resistance ratios over t he USDA -S strain at LD50 and LD90, respectively. In addition, this colony demonstrated 20% survival at the 3X LC99 USDAS. This is surprising, as this farm is the only one of those used in our field studie s where facility managers reported that insect icides were not used. This suggests that stable flies arriv ing from other areas are likely the cause for the resistance observed in flies collect ed at Farm 1. Additional data corroborating this hypothesis were the result s of our host blood meal analysis of stable flies (Chapter 4) which revealed that all flies collected from Fa rm 1 that had not fed on humans had fed on cattle an animal not present within 1.5 km of this facility Although likely, this does no t necessarily mean that stable flies on Farm 1 were predominantly from other sources within a few kilometers. Stable fly dispersal and movement continues to per plex veterinary entomologists, as data from various studies are conflicting. Stable fly disper sal of up to 225 km in Florida has been demonstrated by Hogsette and Ruff (1985) possibly due to incoming weather fronts. Because occurrence and dispersal are not static factors in stable fly biology interbreeding populations of this pest may be more wi despread than previously acknowledged. This is supported by evidence in the present study, where the UFD 07, Farm 1, and HTU stable fly strains all shared simila r diagnostic dose susceptibilities across a region spanning 91 km Further research investigating widespread stable fly susceptibilities is needed to support these findings.
115 A s tudy by Kunz (1991) demonstrated that permethrin resistance in colonized horn flies decreased 7-fold in as few as four generations when selection pressure ended. This may also account in part for the disparity between the increase in resistance of stable flies in the present study and that observed by others in the house fly. In the current study, t he number of offspring produced from the surviving adults of a previous s election w as insufficient to reevaluate permethrin susceptibility levels and perform a selection on their progeny Therefore, an intermediate generation was used to generate numbers adequate to perform the entire procedure. My highpressure selection ap proach inadvertently may have lessened the overall selection pressure, slowing the rate of increase in resistance expression. Although further study is required to determine if this effect occurs in the stable fly, the resistance ratios for Farm 1 may hav e been greater had the strain been evaluated sooner. Stable flies of the HTU strain were evaluated at the time of their capture in the present study, and demonstrated greater survivability than those colonized from Farm 1 when used in diagnostic dose expe riments (Fig 51). Like the horn fly, resistance in the stable fly may decrease in the absence of selection pressure. The results of this study demonstrate that insecticide resistance occurs in field populations of the stable fly. Although the resistance ratios determined in our selection and field evaluations are relatively low the rate at which stable flies attain ed resistance wa s similar to that reported for house fl ies Unlike the horn fly, it is possible that the dispersal habits of the stable fly have permitted the immigration of not only resistant, but also susceptible individuals within the population, decreasing the progression of resistance in this pest. Although speculation, the results of our topically applied
116 insecticide assays suggest that in the p eriod between colonization in 2007, a more than 2 -fold increase in resistance had occurred between the UFD 07 stable flies, and those collected from Farm 1 in 2009 Because these stable flies were collected from two different geographical regions further research will be required to fully support this hypothesis. The permethrin resistance levels determined in the present study are certainly lower than those reported for house fly populations collected from other parts of the U.S. Therefore, time ly research of insecticide resistance in stable flies is imperative, as this problem is still at a relatively manageable level. Further research, including continued re evaluations of permethrin susceptibilities of Florida stable flies, are needed to full y understand the present rates of increase in resistance expression of this pest. Furthermore, investigations of the mechanisms involved in stable fly resistance may elucidate information that can be used to develop newer management strategies, including the development of insecticides with alternate modes of action that will act to slow or inhibit already developing resistance expression.
117 Table 5 1. Permethrin susceptibility for several stable fly strains evaluated using insecticide treated glass jars. Fly strain n LC 50 (g/cm 2 ) a (95% CI) LC 90 (g/cm 2 ) a (95% CI) RR 50 b d RR 90 c d Slope (SE M ) USDA S 1,380 0.0013 (0.0012 0.0014) 0.0022 (0.0021 0.0024) 1.00 1.00 5.43 (0.32) UFD 07 e 1,920 0.0024 (0.0022 0.0027) 0.0101 (0.0090 0.0116) 1.85* 4.59* 2.08 (0. 09) UFD PR f 1,440 0.0092 (0.0077 0.0108) 0.0264 (0.0216 0.0345) 7.08* 12.00* 2.81 (0.28) a Values represent micrograms permethrin per cm2 applied to the inside surface area of glass jars b Resistance ratios at LC50 were calculated as the LC50 of any f ly strain divided by that of the USDA Susceptible strain. c Resistance ratios at LC90 were calculated as the LC90 of any fly strain divided by that of the USDA Susceptible strain. d Resistance ratios were considered significantly different if the 95% confi dence interval (CI ) was non overlapping. e Parental stable fly strain colonized from individuals collected at the University of Florida Dairy Research Unit in Hague, Florida. f Resultant permethrin resistant offspring after five permethrin selections targeting each generations estimated LC70 value n = total number of female stable flies evaluated for permethrin susceptibility, = significant difference between resistance of a stable fly strain compared to the USDA -S strain
118 Table 5 2. Per methrin susceptibility for several stable fly strains evaluated using topically applied insecticide Fly strain n LD 50 (g/g) a (95% CI) LD 90 (g/g) a (95% CI) RR 50 b,d RR 90 c,d Slope (SE M ) USDA S 1,200 0.0311 (0.0292 0.0333) 0.0603 (0.0548 0.0674) 1.00 1.00 4.47 (0.27) UFD 07 e 1,080 0.1178 (0.1066 0.1296) 0.3884 (0.3394 0.4555) 3.79* 6.44* 2.47 (0.14) UFD PR 1 f 720 0.0714 (0.0618 0.0812) 0.2457 (0.2063 0.3066) 2.30* 4.07* 2.39 (0.19) UFD PR 2 839 0.1596 (0.1398 0.1807) 0.5487 (0.4638 0.6749) 5.13* 9.10* 2.39 (0.17) UFD PR 3 840 0.2839 (0.2295 0.3458) 1.1037 (0.8256 1.6896) 9.13* 18.30* 2.17 (0.25) UFD PR 4 720 0.3130 (0.2813 0.3451) 0.6827 (0.6009 0.8037) 10.06* 11.32* 3.78 (0.32) UFD PR g 1,080 0.5844 (0.5433 0.6285) 1.2788 (1. 1510 1.4503) 18.79* 21.21* 3.77 (0.22) Farm 1 1,080 0.2767 (0.2523 0.3029) 0.8492 (0.7453 0.9904) 8.90* 14.08* 2.63 (0.14) a Values represent micrograms permethrin per gram insect applied by topical applicator b Resistance ratios at LC/LD50 were calcul ated as the LD50 of any fly strain divided by that of the USDA Susceptible strain. c Resistance ratios at LC/LD90 were calculated as the LD90 of any fly strain divided by that of the USDA Susceptible strain. d Resistance ratios were considered significantl y different if the 95% confidence interval ( CI ) was non overlapping. e Parental stable fly strain colonized from individuals collected at the University of Florida Dairy Research Unit in Hague, Florida. f Subsequent generations from surviving stable fly a dults of the selected parental UFD -07 strain. g Resultant permethrin resistant offspring after five permethrin selections targeting each generations LC70 value n = total number of female stable flies evaluated for permethrin susceptibility, = significa nt difference between resistance of a stable fly strain compared to the USDA -S strain
119 Table 5 3 Concentrations used and stable fly mortality results at each permethrin selection. Selected Generationa Selecting Conc. (g/cm2)b % Male Mortalityc % Female Mortalityc % Total Mortalityd UFD 07 0.0044 76.3 76.7 76.5 UFD PR 1 0.0080 80.7 78.6 79.7 UFD PR 2 0.0088 87.8 86.3 87.0 UFD PR 3 0.0160 92.0 90.5 91.2 UFD PR 4 0.0442 94.7 88.2 91.4 a Stable fly colony obtained from adult flies collected at the Univ ersity of Florida Dairy Research Unit in Hague, FL. Selections were performed on the parental strain, and every other generation thereafter using the estimated LC70 value of each generation Therefore, the selection denoting F1 was actually the F2 accordi ng to a rearing schedule. b Selecti o n concentrations were applied to 1.06L glass canning jars, in which 250300 mixed -sex adult stable flies were expos ed for a 4 hr period. c Percent male and female mortality was determined at 48 hr, from five jars of eac h sex containing 250 stable flies. d Percent total mortality was the total number of dead stable flies at 48 hr divided by a total of 2,500 separately monitored individuals used to determine sex -dependent mortality.
120 Figure 51. Mean percent survival of several stable fly strains exposed to diagnostic permethrin concentrations applied to 60 ml glass jars at 1-, 3 -, 10 -, and 30x the LC99 of a susceptible stable fly strain (USDA S).
121 CHAPTER 6 IMPLICATIONS AND FUT URE DIRECTIONS FOR RESEARCH REGARDING ST ABLE FLIES AND EQUINE FAC ILITIES The future success of stable fly, Stomoxys calcitrans (L.), management within livestock facilities likely will depend on our ability to develop and use integrated pest management (IPM) program s The problems associated with house fly control due to expression of insecticide resistance (Scott et al. 2000, Kaufman et al. 2001b) and potential dispersal of resistance genetics (Learmount et al. 2002, Maron et al. 2003) are particularly troubling when stable fly biology is consid ered. Stable flies have been shown to travel up to 3.2 km in search of a blood meal (Bailey et al. 1973, Gersabeck and Merritt 1985) and as far as 225 km depending on weather conditions (Hogsette and Ruff 1985). My study of stable fly host blood meal analysis clearly demonstrate s dispersal as far as 1.5 km in as little as 16 hr as well as the magnitude of the movement with 60% of the flies having come from cattle farms Therefore, stable fly dispersal may greatly affect the success of current control methodologies such as the release of pteromalid pupal parasitoids or adult directed insecticide use if these pests are breeding or immigrating from off -farm sites. Further research using polymerase chain reaction with microsatellite technology may elucidate the genetic relationship of stable fly populations in different geographic locations thereby accounting for dispersal and potential similarities in insecticide resistance expression Another avenue of research may involve using microsatellites of potent ial hosts at one farm, to directly pinpoint the origins of stable flies collected at another. Stable fly insecticide resistance has been demonstrated twice and only in the Midwestern U.S. (Cilek and Greene 1994, Maron et al. 1997). However, those studie s, conducted over a decade ago, document only moderate resistance ratios to permethrin
122 when compared to laboratory strains. Furthermore, Maron et al. (1997) demonstrated it was likely the observed resistance levels in field -collected stable flies were no different than the varying levels detected between laboratory colony generations. My study on the permethrin susceptibility status of Florida stable flies provides a critically needed update in this area, and demonstrated up to 7and 9 -fold greater LC/LD values than those of Cilek and Greene (1994) and Maron et al. (1997), respectively. Stable flies from an equine facility with no reported insecticide use demonstrated approximately 20% survival with diagnostic concentrations at 3X LC99 that of a suscept ible strain Additionally, survival profiles of three field -collected stable fly strains using diagnostic concentrations were similar, despite a 91km range between their collection sites. This further supports the hypothesis of long range stable fly dis persal, and su pports the p ossibility of interbreeding populations that are separated geographically. Therefore, further research to determine the widespread insecticide susceptibility status of stable flies is needed, if future efforts utilizing chemical c ontrol methods are to succeed. My studies also have shown that stable flies can readily develop resistance to a widely used pyrethroid insecticide, permethrin, in as few as five laboratory selections. Furthermore, a 15-fold increase in resistance express ion resulted even when selecting every other stable fly generation This likely decreased the potential gains in resistance had selections been performed on successive generations Although it is likely that stable fly resistance is due to mechanisms previously demonstrated in the house fly, such as target site insensitivity ( kdr ) (Smith et al. 1997, Lee et al. 1999) and metabolic
123 detoxification (Liu and Yue 2000) identification of these mechanisms is imperative to the effectiveness of chemical control d evelopment and use for this pest My studies concerning the ability of stable flies to develop resistance to permethrin are similar to th ose observed in the house fly and horn fly, and provide a grim reminder of the importance in developing effective IPM p rogram s However, research involving the use of pteromalid pupal parasitoids as an alternative filth fly control measure provides conflicting data (Meyer et al. 1990, Geden et al. 1992, Petersen et al. 1992). Many factors can affect the success of parasi toid releases including immature filth fly breeding substrate light and moisture conditions (Smith and Rutz 1990, Geden 1999) host depth and density (Rueda and Axtell 1985b, King 1997), and substrate type (Geden 2002) My study concerning pupal parasitoi ds under both field and laboratory conditions provide supporting evidence to the importance of assessing conditions of potential fly breeding areas prior to selecting a parasitoid species for release. In addition, some parasitoids may occupy different tem poral niches, whereby species specific releases may be dependent on season as well. During my field study, nearly all parasitoids collected were Spalangia spp., and consisted of Spalangia cameroni Perkins Spalangia nigroaenea Curtis, Spalangia nigra Latr eille and Spalangia endius Walker. The hypothesis that this occurrence was due to the unique filth fly larval breeding habitats created by equine husbandry practices was corroborated under laboratory conditions. In all laboratory assays, Spalangia spp. located and attacked naturally pupating stable fly hosts up to 50 -fold more often than Muscidifurax raptorellus Kogan and Legner. Therefore, under the conditions provided by equine facilities in Florida, it is likely that Muscidifurax spp. will be ineffec tive in increasing filth-fly control.
124 The areas of research detailed in this dissertation are inter -related by the underlying hypothesis that stable fly dispersal and interbreeding population dynamics are more widespread than previously thought. Therefore, insecticidal, cultural, and biological control tactics may be rendered ineffective if off -site breeding is the source of on-site stable fly activity. Instead, other elements of an IPM program such as educating nearby managers of cattle producing facilit ies in the areas of stable fly larval biology and proper livestock husbandry generated waste management should become the forefront for control of this pest. Because c ontrol of this pest may also be limited to adult stable flies due to their dispersal be havior an increase in research concerning the development of newer and more effective trapping devices such as the treated and electrified targets designed by Foil and Younger (2006) is warranted Most stable fly activity in Florida occurs in the early spring months between January and April. Therefore, Florida equine facilities with known stable fly breeding activity may benefit from pteromalid pupal parasitoid releases during these months in addition to strong cultural management of livestock generated waste M y field and laboratory studies of resident parasitoids suggest that Spalangia spp. only releases may be more effective than those of other genera, such as Muscidifurax spp. Furthermore these releases should be conducted with S. cameroni as thi s was the most abundant species during the early spring months at Florida equine facilities. The utility of continued education to livestock producers concerning the cultural control of husbandry generated wastes was demonstrated in my field studies. In those studies, the only farm with no reported insecticide use also accounted for the lowest adult stable fly trap
125 captures. This was probably due to the intense daily sanitation and composting activities practiced by the farm. My studies of the ecology of stable flies associated with equine facilities in Florida should provide a basis for further research in an area relatively uninvestigated by the entomology field. Additionally, evidence provided by this dissertation suggests that the problems most often associated with horn flies and house flies such as insecticide resistance are increasing in the stable fly Similar study of other livestock facilities, such as dairy and feedlot cattle systems may assist in increasing our understanding and therefore, our chances of effectively controlling this economically important pest.
126 APPENDIX A UNIVERSITY OF FLORID A HEALTH CENTER INST ITUTIONAL REVIEW BOA RD #3422008
132 APPENDIX B UNIVERSITY OF FLORID A INSTITUTIONAL ANIMAL CARE AND USE COMMITT EE #200801760
133 LIST OF REFERENCES Abbasi, I., R. Cunio, and A. Warburg. 2009. Identification of blood meals imbibed by phlebotomine sand flies using cytochrome b PCR and reverse line blotting. Vector Borne Zoonot ic Dis. 9: 79-86. Abbott, W. S. 1925. A method of computing the effectiveness of an insecticide. J. Econ. Entomol. 18: 265-267. [AHC] American Horse Council. 2009. National economic impact of the U.S. horse industry. (http://www.horsecouncil.org). Anderson, J. R. and C. H. Tempelis. 1970. Precip itin test identification of blood meals of Stomoxys calcitrans (L.) caught on California poultry ranches, and observations of digestion rates of bovine and citrated human blood. J. Med. Entomol. 7: 223229. Andress, E. R., and J. B. Campbell. 1994. Inundative releases of pteromalid parasitoids (Hymenoptera: Pteromalidae) for the control of stable flies, Stomoxys calcitrans (L.) (Diptera: Muscidae) at confined cattle installations in west central Nebraska. J. Econ. Entomol. 87: 714 -722. Bailey, D. L., T. L. Whitfield, and B. J. Smittle. 1973. Flight and dispersal of the stable fly. J. Econ. Entomol. 66: 410411. Barros, A. T. M., M. W. Alison, Jr., and L. D. Foil. 1999. Evaluation of a yearly insecticidal ear tag rotation for control of pyrethroid -resis tant horn flies (Diptera: Muscidae). Vet. Parasitol. 82: 317325. Barros, A. T. M., J. Ottea, D. Sanson, and L. D. Foil. 2001. Horn fly (Diptera: Muscidae) resistance to organophosphate insecticides. Vet. Parasitol. 96: 243-256. Beresford, D. V. and J. F. Sutcliffe. 2006. Studies on the effectiveness of coroplast sticky traps for sampling stable flies (Diptera: Muscidae), including a comparison of alsynite. J. Econ. Entomol. 99: 10251035. Bertsch, M. L., and B. R. Norment. 1983. The host -feeding pat terns of Culex quinquefasciatus in Mississippi. Mosq. News. 43: 203206. Boakye, D. A., J. Tang, P. Truc, A. Merriweather, and T. R. Unnasch. 1999. Identification of bloodmeals in haematophagous Diptera by cytochrome B heteroduplex analysis. Med. Vet. Ent omol. 13: 282287. Boire, S., D. E. Bay, and J. K. Olson. 1988. An evaluation of various types of manure and vegetative material as larval breeding media for the stable fly. Southwest. Entomol. 13: 247249.
134 Broce, A. B. 1988. An improved alsynite trap for stable flies Stomoxys calcitrans (Diptera: Muscidae). J. Med. Entomol. 25: 406-409. Broce, A. B. and M. S. Haas. 1999. Relation of cattle manure age to colonization by stable fly and house fly (Diptera: Muscidae). J. Kans Entomol. Soc. 72: 6072. Broce, A. B., J. Hogsette, and S. Paisley. 2005. Winter feeding sites of hay in round bales as major developmental sites of Stomoxys calcitrans (Diptera: Muscidae) in pastures in spring and summer. J. Econ. Entomol. 98: 23072312. Bruce, W. N. and G. C. Decker. 1958. The relationship of stable fly abundance to milk production in dairy cattle. J. Econ. Entomol. 51: 269274. Burg, J. G., F. W. Knapp, and D. G. Powell. 1990. Seasonal abundance and spatial distribution patterns of three adult muscoid (Di ptera: Muscidae) species on equine premises. Environ. Entomol. 19: 901904. Byford, R. L., M. E. Craig, S. M. DeRouen, M. D. Kimball, D. G. Morrison, W. E. Wyatt, and L. D. Foil. 1999. Influence of permethrin, diazinon and ivermectin treatments on insect icide resistance in the horn fly (Diptera: Muscidae). Int. J. Parasitol. 29: 125135. Campbell, J. B., J. E. Wright, R. Crookshank, and D. C. Clanton. 1977. Effects of stable flies on weight gains and feed efficiency of calves on growing or finishing rat ions. J. Econ. Entomol. 70: 592 -594. Campbell, J. B., I. L. Berry, D. J. Boxler, R. L. Davis, D. C. Clanton, and G. H. Deutscher. 1987. Effects of stable flies (Diptera: Muscidae) on weight gain and feed efficiency of feedlot cattle. J. Econ. Entomol. 80: 117-119. Campbell, J. B., S. R. Skoda, D. R. Berkebile, D. J. Boxler, G. D. Thomas, D. C. Adams, and R. Davis. 2001. Effects of stable flies (Diptera: Muscidae) on weight gains of grazing yearling cattle. J. Econ. Entomol. 94: 780783. Ca tangui, M. A., J. B. Campbell, G. D. Thomas, and D. J. Boxler. 1993. Average daily gains of brahman-crossbred and E nglish X exotic feeder heifers exposed to low, medium, and high levels of stable flies (Diptera: Muscidae). J. Econ. Entomol. 86: 1144 -1150. Cilek, J. E. 2003. Attraction of colored plasticized corrugated boards to adult stable flies, Stomoxys calcitrans (Diptera: Muscidae). Fla. Entomol. 86: 420-423.
135 Cilek, J. E., and G. L. Greene. 1994. Stable fly (Diptera: Muscidae) insecticide resistance in Kansas cattle feedlots. J. Econ. Entomol. 87: 275-279. Cilek, J. E., C. D. Steelman, and F. W. Knapp. 1991. Horn fly (Diptera: Muscidae) insecticide resistance in Kentucky and Arkansas. J. Econ. Entomol. 84: 756762. [CLUSTALW] Kyoto University Bioinformatics C enter. 2009. Multiple sequence alignment by CLUSTALW. (http://align.genome.jp/). Crosby, B. L., R. L. Byford, and H. G. Kinzer. 1991. Insecticide resistance in the horn fly Haematobia irritans (L.), in New Mexico: Survey and c ontrol. Southwest. Entomo l. 16: 301 309. Dougherty, C. T., F. W. Knapp, P. B. Burrus, D. C. Willis, and P. L. Cornelius. 1995. Behavior of grazing cattl e exposed to small populations o f stable flies ( Stomoxys calcitrans (L. ) ). Appl. Anim. Behav. Sci. 42: 231248. DuToit, G. D. G 1975. Reproductive capacity and longevity of stable flies maintained on different kinds of blood. J. S. Afr. Vet. Assoc. 46: 345347. Fadok, V. A., and E. C. Greiner. 1990. Equine insect hypersensitivity: S kin test data and biopsy results correlated with clinical data. Equine Vet. J. 22: 236-240. Floate, K. D. and R. W. Spooner. 2002. Parasitization by pteromalid wasps (Hymenoptera) of freezekilled house fly (Diptera: Muscidae) puparia at varying depths in media. J. Econ. Entomol. 95: 908911. Floate, K. D., P. Coghlin, and G. A. P. Gibson. 2000. Dispersal of the filth fly parasitoid Muscidifurax raptorellus (Hymenoptera: Pteromalidae) following mass releases in cattle confinements. Biol. Control. 18: 172178. Foil, L. D., and J. A. Hogsette. 1994. Biology and control of tabanids, stable flies and horn flies. Rev. Sci. Tech. Off. Int. Epizoot. 13: 11251158. Foil, L. D., and C. D. Younger. 2006. Development of treated targets for controlling stable flies (Diptera: Muscidae). Vet. Parasitol. 137: 311 -315. Foil, L. D., F. Guerrero, M. W. Alison, and M. D. Kimball. 2005. Association of the kdr and superkdr sodium channel mutations with resistance to pyrethroids in Louisiana populations of the horn fly, Haematobia irritans (L.). Vet. Parasitol. 129: 149158. Fye, R. L., J. Brown, J. Ruff, and L. Buschman. 1980. A survey of Northwest Florida for potential stable fly breeding. Fla. Entomol. 63: 246251.
136 Geden, C. J. 1999. Host location by house fly (Diptera: Muscidae) parasitoids in poultry manure at different moisture levels and host densities. Environ. Entomol. 28: 755760. Geden, C. J. 2002. Effect of habitat depth on host location by five species of parasitoids (Hymenoptera: Pteromalidae, Chalcidae) of house flies (Diptera: Muscidae) in three types of substrates. Environ. Entomol. 31: 411-417. Geden, C. J., D. A. Rutz, R. W. Miller, and D. C. Steinkraus. 1992. Suppression of house flies (Diptera: Muscidae) on New York and Maryland dairies using releases of Muscidifurax raptor (Hymenopte ra: Pteromalidae) in an integrated management program. Environ. Entomol. 21: 1419-1426. Gentry, A. B. 2002. Evaluation of protection systems and determination of seasonality for mosquitoes and biting flies at the University of Florida Horse Teaching Unit M. S. thesis, University of Florida, Gainesville. Gersabeck, E. F. and R. W. Merritt. 1985. Dispersal of adult Stomoxys calcitrans (L.) (Diptera: Muscidae) from known immature developmental areas. J. Econ. Entomol. 78: 617-621. Gibson, G. A. P. 2009. Revision of new world Spalangiinae (Hymenoptera: Pteromalidae). Zootaxa. 2259: 1159. Gilles, J., J. F. David, and G. Duvallet. 2005. Temperature effects on development and survival of two stable flies, Stomoxys calcitrans and Stomoxys niger niger (Dip tera: Muscidae), in La Runion Island. J. Med. Entomol. 42: 260265. Gortel, K. 1998. Equine parasitic hypersensitivity. Equine Pract. 20: 14 16. Greene, G. L., J. A. Hogsette, and R. S. Patterson. 1989. Parasites that attack stable fly and house fly ( Diptera: Muscidae) puparia during the winter on dairies in Northwestern Florida. J. Econ. Entomol. 82: 412415. Guerrero, F. D., R. C. Jamroz, D. Kammlah, and S. E. Kunz. 1997. Toxicological and molecular characterization of pyrethroid-resistant horn fli es, Haematobia irritans : Identification of kdr and super -kdr point mutations. Insect Biochem. Mol. Biol. 27: 745755. Guglielmone, A. A., M. M. Volpogni, O. R. Quaino, O. S. Anziani, and A. J. Mangold. 2004. Abundance of stable flies on heifers treated f or control of horn flies with organophosphate impregnated ear tags. Vet. Med. Entomol. 18: 10 -13.
137 Hafez, M., and F. M. Gamal-Eddin. 1959. On the feeding habits of Stomoxys calcitrans (L.) and sitiens Rond., with special reference to their biting cycle in nature. Bull. Soc. Entomol. Egypte. 43: 291-301. Hogsette, J. A. and J. P. Ruff. 1985. Stable fly (Diptera: Muscidae) migration in Northwest Florida. Environ. Entomol. 14: 170175. Hogsette, J. A. and J. P. Ruff. 1986. Evaluation of flucythrinate a nd fenvalerate -impregnated ear tags and permethrin ear tapes for fly (Diptera: Muscidae) control on beef and dairy cattle in Northwest Florida. J. Econ. Entomol. 79: 152-157. Hogsette, J. A. and J. P. Ruff. 1990. Comparative attraction of four different fiberglass traps to various age and sex classes of stable fly (Diptera: Muscidae) adults. J. Econ. Entomol. 83: 883886. Hogsette, J. A. and J. P. Ruff. 1996. Permethrin-impregnated yarn: L ongevity of efficacy and potential use on cylindrical fibergla ss stable fly (Diptera: Muscidae) traps. J. Econ. Entomol. 89: 1521-1525. Hogsette, J. A., J. P. Ruff, and C. J. Jones. 1987. Stable fly biology and control in Northwest Florida. J. Agric. Entomol. 4: 111. Huang, J., M. Kristensen, C. L. Qiao, and J. B Jespersen. 2004. Frequency of kdr gene in house fly field populations: correlation of pyrethroid resistance and kdr frequency. J. Econ. Entomol. 97: 1036-1041. Jamroz, R. C., F. D. Guerrero, D. M. Kammlah, and S. E. Kunz. 1998. Role of the kdr and su per -kdr sodium channel mutations in pyrethroid resistance: correlation of allelic frequency to resistance level in wild and laboratory populations of horn flies ( Haematobia irritans ). Insect Biochem. Mol. Biol. 28: 1031-1037. Jeanbourquin, P. and P. M. Guerin. 2007. Chemostimuli implicated in selection of oviposition substrates by the stable fly Stomoxys calcitrans. Med. Vet. Entomol. 21: 209216. Jones, C. J. and R. A. Weinzierl. 1997. Geographical and temporal variation in pteromalid (Hymenoptera: P teromalidae) parasitism of stable fly and house fly (Diptera: Muscidae) pupae collected from Illinois cattle feedlots. Environ. Entomol. 26: 421-432. Karunaratne, K. M., and F. W. Plapp, Jr. 1993. Biochemistry and genetics of thiodicarb resistance in th e house fly (Diptera: Muscidae). J. Econ. Entomol. 86: 258264.
138 Kaufman, P. E., J. E. Lloyd, R. Kumar, and T. J. Lysyk. 1999. Horn fly susceptibility to diazinon, fenthion, and permethrin at selected elevations in Wyoming. J. Agric. Urban Entomol. 16: 141 157. Kaufman, P. E., S. J. Long, and D. A. Rutz. 2001a. Impact of exposure length and pupal source on Muscidifurax raptorellus and Nasonia vitripennis (Hymenoptera: Pteromalidae) parasitism in a New York poultry facility. J. Econ. Entomol. 94: 998 -1003 Kaufman, P. E., J. G. Scott, and D. A. Rutz. 2001b. Monitoring insecticide resistance in house flies (Diptera: Muscidae) from New York dairies. Pest Manag. Sci. 57: 514 -521. Kaufman, P. E., S. J. Long, D. A. Rutz, and J. K. Waldron. 2001c. Parasitism rates of Muscidifurax raptorellus and Nasonia vitripennis (Hymenoptera: Pteromalidae) after individual and paired releases in New York poultry facilities. J. Econ. Entomol. 94: 593-598. Keiper, R. R., and J. Berger. 19 9 2. Refuge -seeking and pest avoidance by feral horses in desert and island environments. Appl. Anim. Ethol. 9: 111120. Kent, R. J. 2009. Molecular methods for arthropod bloodmeal identification and applications to ecological and vector -borne disease studies. Mol. Ecol. Resour. 9: 4 -18. K ent, R. J. and D. E. Norris. 2005. Identification of mammalian blood meals in mosquitoes by a multiplexed polymerase chain reaction targeting cytochrome B. Am. J. Trop. Med. Hyg. 73: 336342. King, B. H. 1997. Effects of age and burial of house fly (D iptera: Muscidae) pupae on parasitism by Spalangia cameroni and Muscidifurax raptor (Hymenoptera: Pteromalidae). Environ. Entomol. 26: 410415. King, W. V., and L. G. Lenert. 1936. Outbreaks of Stomoxys calcitrans ( L. ) (dog flies) along Floridas north west coast. Fla. Entomol. 19: 3339. Knipple, D. C., K. E. Doyle, P. A. Marsella Herrick, and D. M. Soderlund. 1994. Tight genetic linkage between the kdr insecticide resistance trait and a voltage -senstive sodium channel gene in the house fly. P roc Nat l. Acad. Sci. U S. A. 91: 24832487. Kristensen, M., and J. B. Jespersen. 2004. Susceptibility of spinosad in Musca domestica (Diptera: Muscidae) field populations. J. Econ. Entomol. 97: 1042-1048. Kunz, S. E. 1991. Dynamics of permethrin resistance i n a colony of horn flies (Diptera: Muscidae). J. Med. Entomol. 28: 6366.
139 Learmount, J., P. Chapman, and A. Macnicoll. 2002. Impact of an insecticide resistance strategy for house fly (Diptera: Muscidae) control in intensive animal units in the United Ki ngdom. J. Econ. Entomol. 95: 1245-1250. Lee, S. H., T. J. Smith, D. C. Knipple, and D. M. Soderlund. 1999. Mutations in the house fly Vssc1 sodium channel gene associated with super -kdr resistance abolish the pyrethroid sensitivity of Vssc1/tipE sodium c hannels expressed in X enopus oocytes. Insect Biochem. Mol. Biol. 29: 185194. Legner, E. F., and H. W. Brydon. 1966. Suppression of dung-inhabiting fly populations by pupal parasites. Ann. Entomol. Soc. Am. 59: 638651. Lehane, M. 2005. Managing the bl ood meal. pp 84-115. The biology of bloodsucking in insects. Cambridge University Press, Cambridge, New York. Liu, N., and J. G. Scott. 1997. Inheritance of CYP6D1mediated pyrethroid resistance in house fly (Diptera: Muscidae). J. Econ. Entomol. 90: 14 781481. Liu, N., and J. G. Scott. 1998. Increased transcription of CYP6D1 causes cytochrome P450mediated insecticide resistance in the house fly. Insect Biochem. Mol. Biol. 28: 531-535. Liu, N., and X. Yue. 2000. Insecticide resistance and cross -resi stance in the house fly (Diptera: Muscidae). J. Econ. Entomol. 93: 12691275. Lockwood, J. A., R. L. Byford, R. N. Story, T. C. Sparks, and S. S. Quisenberry. 1985. Behavioral resistance to the pyrethroids in the horn fly, Haematobia irritans (Diptera: M uscidae). Environ. Entomol. 14: 873880. Lysyk, T. J. 1993. Seasonal abundance of stable flies and house flies (Diptera: Muscidae) in dairies in Alberta, Canada. J. Med. Entomol. 30: 888-895. Lysyk, T. J. 1998. Relationships between temperature and life history parameters of Stomoxys calcitrans (Diptera: Muscidae). J. Med. Entomol. 35: 107119. Lysyk, T. J. 2001. Relationships between temperature and life history parameters of Muscidifurax raptorellus (Hymenoptera: Pteromalidae). Environ. Entomol. 30: 982 -992. Mann, J. A., R. E. Stinner, and R. C. Axtell. 1990. Parasitism of house fly (Musca domestica ) pupae by four species of Pteromalidae (Hymeno ptera): E ffects of host parasitoid densities and host distribution. Med. Vet. Entomol. 4: 235243.
140 Mar on, P. C. R. G., G. D. Thomas, B. D. Siegfried, and J. B. Campbell. 1997. Susceptibility of stable flies (Diptera: Muscidae) from Southeastern Nebraska beef cattle feedlots to selected insecticides and comparison of 3 bioassay techniques. J. Econ. Entomol. 90: 293298. Mar on, P. C. R. G., G. D. Thomas, B. D. Siegfried, J. B. Campbell, and S. R. Skoda. 2003. Resistance status of house flies (Diptera: Muscidae) from Southeastern Nebraska beef cattle feedlots to selected insecticides. J. Econ. Entomol. 96: 1016 1020. McDonald, P. T., and C. D. Schmidt. 1987. Genetics of permethrin resistance in the horn fly (Diptera: Muscidae). J. Econ. Entomol. 80: 433437. McPheron, L. J. and A. B. Broce. 1996. Environmental components of pupariation -site selection b y the stable fly (Diptera: Muscidae). Environ. Entomol. 25: 665671. Meyer, J. A. and J. J. Petersen. 1983. Characterization and seasonal distribution of breeding sites of stable flies and house flies (Diptera: Muscidae) on Eastern Nebraska feedlots an d dairies. J. Econ. Entomol. 76: 103108. Meyer, J. A., B. A. Mullens, T. L. Cyr, and C. Stokes. 1990. Commercial and naturally occurring fly parasitoids (Hymenoptera: Pteromalidae) as biological control agents of stable flies and house flies (Diptera: M uscidae) on California dairies. J. Econ. Entomol. 83: 799-806. Meyer, J. A., T. A. Shultz, C. Collar, and B. A. Mullens. 1991. Relative abundance of stable fly and house fly (Diptera: Muscidae) pupal parasites (Hymenoptera: Pteromalidae; Coleoptera: Staphylinidae) on confinement dairies in California. Environ. Entomol. 20: 915-921. Miller, R. W., L. G. Pickens, N. O. Morgan, R. W. Thimijan, and R. L. Wilson. 1973. Effect of stable flies on feed intake and milk production of dairy cows. J. Econ. Entomol. 66: 711713. Moon, R. D., L. D. Jacobson, and S. G. Cornelius. 1987. Stable flies (Diptera: Muscidae) and productivity of confined nursery pigs. J. Econ. Entomol. 80: 10251027. Mount, G. A., C. S. Lofgren, and J. B. Gahan. 1966. Malathion, naled, fen thion, and B ayer 39007 thermal fogs for control of the stable fly (dog fly), Stomoxys calcitrans (Diptera: Muscidae). Fla. Entomol. 49: 169173. Mullens, B. A., and J. A. Meyer. 1987. Seasonal abundance of stable flies (Diptera: Muscidae) on California d airies. J. Econ. Entomol. 80: 1039 1043.
141 Mullens, B. A., K. S. Lii, Y. Mao, J. A. Meyer, N. G. Peterson, and C. E. Szijj. 2006. Behavioural responses of dairy cattle to the stable fly, Stomoxys calcitrans in an open field environment. Med. Vet. Entomol. 20: 122 137. [NCBI] National Center for Biotechnology Information. 2009. B LAST a basic local alignment search tool. (http://blast.ncbi.nlm.nih.gov). Ngo, K. A., and L. D. Kramer. 2003. Identification of mosquito bloodmeals using polymerase chain reac tion (PCR) with order -specific primers. J. Med. Entomol. 40: 215222. [NOAA] National Oceanic and Atmospheric Administration. 2009. Historical weather data archives. (http://cdo.ncdc.noaa.gov). Petersen, J. J. and J. A. Meyer. 1985. Evaluation of methods presently used for measuring parasitism of stable flies and house flies (Diptera: Muscidae) by pteromalid wasps (Hymenoptera: Pteromalidae). J. Kans. Entomol. Soc. 58: 8490. Petersen, J. J. and J. K. Cawthra. 1995. Release of a gregarious Muscidifu rax species (Hymenoptera: Pteromalidae) for the control of filth flies associated with confined beef cattle. Biol. Control. 5: 279284. Petersen, J. J. and D. M. Currey. 1996. Timing of releases of gregarious Muscidifurax raptorellus (Hymenoptera: Pteromalidae) to control flies associated with confined beef cattle J. Agric. Entomol. 13: 55-63. Petersen, J. J., M. A. Catangui, and D. W. Watson. 1991. Parasitoid -induced mortality of house fly pupae by pteromalid wasps in the laboratory. Biol. Control. 1 : 275-280. Petersen, J. J., D. W. Watson, and B. M. Pawson. 1992. Evaluation of field propagation of Muscidifurax zaraptor (Hymenoptera: Pteromalidae) for control of flies associated with confined beef cattle. J. Econ. Entomol. 85: 451-455. Rinkevich, F. D., R. L. Hamm, C. J. Geden, and J. G. Scott. 2007. Dynamics of insecticide resistance alleles in house fly populations from New York and Florida. Insect Biochem. Mol. Biol. 37: 550 558. Romero, A., A. Broce, and L. Zurek. 2006. Role of bacteria in th e oviposition behaviour and larval development of stable flies. Med. Vet. Entomol. 20: 115-121. Roush, R. T., R. L. Combs, T. C. Randolph, J. MacDonald, and J. A. Hawkins. 1986. Inheritance and effective dominance of pyrethroid resistance in the horn fly (Diptera: Muscidae). J. Econ. Entomol. 79: 11781182.
142 Rueda, L. M., and R. C. Axtell. 1985a. Guide to common species of pupal parasites (Hymenoptera: Pteromalidae) of the house fly and other muscoid flies associated with poultry and livestock manure. N. C. Agric. Res. Serv. Tech. Bull. 278. 88pp Rueda, L. M., and R. C. Axtell. 1985b. Effect of depth of house fly pupae in poultry manure on parasitism by six species of pteromalidae (Hymenoptera). J. Entomol. Sci. 20: 444-449. Rutz, D. A., and R. C. Axtel l. 1979. Sustained releases of Muscidifurax raptor (Hymenoptera: Pteromalidae) for house fly (Diptera: Muscidae) control in two types of caged-layer poultry houses. Environ. Entomol. 8: 11051110. Rutz, D. A., and R. C. Axtell. 1981 House fly ( Musca domestica ) control in broiler -breeder poultry houses by pupal parasites (Hymenoptera: Pteromalid ae): I ndigenous parasite species and releases of Muscidifurax raptor Environ. Entomol. 10: 343345. SAS Institute. 2004. Version 9.1. SAS Institute, Cary, NC. Schmidt, C. D., S. E. Kunz, H. D. Petersen, and J. L. Robertson. 1985. Resistance of horn flies (Diptera: Muscidae) to permethrin and fenvalerate. J. Econ. Entomol. 78: 402406. Scott, J. G. 1998. Toxicity of spinosad to susceptible and resistant strai ns of house flies, Musca domestica. Pestic. Sci. 54: 131 133. Scott, J. G., and G. P. Georghiou. 1985. Rapid development of high-level permethrin resistance in a field-collected strain of the house fly (Diptera: Muscidae) under laboratory selection. J. E con. Entomol. 78: 316-319. Scott, J. G., T. G. Alefantis, P. E. Kaufman, and D. A. Rutz. 2000 Insecticide resistance in house flies from caged-layer poultry facilities. Pest Manag. Sci. 56: 147153. Seifert, J. and J. G. Scott. 2002. The CYP6D1 allele is associated with pyrethroid resistance in the house fly, Musca domestica. Pestic. Biochem. Phys iol 72: 40 -44. Seymour, R. C. and J. B. Campbell. 1993. Predators and parasitoids of house flies and stable flies (Diptera: Muscidae) in cattle confinements in West Central Nebraska. Environ. Entomol. 22: 212-219. Sheppard, D. C., and J. A. Joyce. 1992. High levels of pyrethroid resistance in horn flies (Diptera: Muscidae) selected with cyhalothrin. J. Econ. Entomol. 85: 1587-1593.
143 Skovg rd, H. 2002. Dispersal of the filth fly parasitoid Spalangia cameroni (Hymenoptera: Pteromalidae) in a swine facility using fluorescent dust marking and sentinel pupal bags. Environ. Entomol. 31: 425431. Skovgrd, H., and J. B. Jespersen. 1999. Activ ity and relative abundance of hymenopterous parasitoids that attack puparia of Musca domestica and Stomoxys calcitrans (Diptera: Muscidae) on confined pig and cattle farms in Denmark. B ull Entomol. Res. 89: 263-269. Smith, L., and D. A. Rutz. 1990. The influence of light and moisture gradients on the attack rate of parasitoids foraging for hosts in a laboratory arena (Hymenoptera: Pteromalidae). J. Insect Behav. 4: 195-208. Smith, L., and D. A. Rutz. 1991. Seasonal and relative abundance of hymenopterous parasitoids attacking house fly pupae at dairy farms in Central New York. Environ. Entomol. 20: 661668. Smith, T. J., S. H. Lee, P. J. Ingles, D. C. Knipple, and D. M. Soderlund. 1997. The L1014F point mutation in the house fly Vssc1 sodium channel c onfers knockdown resistance to pyrethroids. Insect Biochem. Mol. Biol. 27: 807812. Sutherland, B. 1978a. Nutritional values of different blood diets expressed as reproductive potentials in adult Stomoxys calcitrans ( L. ) (Diptera: Muscidae). Onderstepoort J. Vet. Res. 45: 209212. Sutherland, B. 1978b. The suitability of various types of dung and vegetable matter as larval breeding media for Stomoxys calcitrans (L. ) (Diptera: Muscidae). Onderstepoort J. Vet. Res. 45: 241243. Talley, J., A. Broce, and L. Zurek. 2009. Characterization of stable fly (Diptera: Muscidae) larval developmental habitat at round hay bale feeding sites. J. Med. Entomol. 46: 1310-1319. Taylor, D. B. and D. Berkebile. 2006. Comparative efficiency of six stable fly (Diptera: M uscidae) traps. J. Econ. Entomol. 99: 1415-1419. Tobin, P. C. and C. W. Pitts. 1999. Dispersal of Muscidifurax raptorellus Kogan and Legner (Hymenoptera: Pteromalidae) in a high-rise poultry facility. Biol. Control. 16: 68-72. Weinzierl, R. A., and C. J Jones. 1998. Releases of Spalangia nigroaenea and Muscidifurax zaraptor (Hymenoptera: Pteromalidae) increases rates of parasitism and total mortality of stable fly and house fly (Diptera: Muscidae) pupae in Illinois cattle feedlots. J. Econ. Entomol. 9 1: 11141121.
144 Wen, Z. and J. G. Scott. 1999. Genetic and biochemical mechanisms limiting fipronil toxicity in the LPR strain of house fly, Musca domestica Pestic. Sci. 55: 988992. Williams, D. F. 1973. Sticky traps for sampling populations of Stomox ys calcitrans J. Econ. Entomol. 66: 1279 -1280. Williams, D. F and A. J. Rogers. 1976. Vertical and lateral distribution of stable flies in Northwestern Florida. J. Med. Entomol. 13: 9598. Wylie, H. G. 1971. Observations on intraspecific larval com petition in three hymenopterous parasites of fly puparia. Can. Entomol. 103: 137142. Wylie, H. G. 1972a. Oviposition restraint of Spalangia cameroni (Hymenoptera: Pteromalidae) on parasitized house fly pupae. Can. Entomol. 104: 209214. Wylie, H. G. 19 72b. Larval competition among three hymenopterous parasite species on multiparasitized house fly (Diptera) puape. Can. Entomol. 104: 11811190. Zinser, M., F. Ramberg, and E. Willott. 2004. Culex quinquefasciatus (Diptera: Culicidae) as a potential West Nile virus vector in Tucson, Arizona: B lood meal analysis indicates feeding on both humans and birds. J. Insect Sci. 4: 1-3. Zyzak, M. D., R. L. Byford, M. E. Craig, and J. A. Lockwood. 1996. Behavioral responses of the horn fly (Diptera: Muscidae) to selected insecticides in contact and noncontact environments. Environ. Entomol. 25: 120129.
145 BIOGRAPHICAL SKETCH Jimmy Pitzer was born in Tracy, California, where he resided for 22 years. After graduating high school in 1997, Jimmy began his undergraduate education at San Joaquin Delta College in Stockton, California. Although he was accepted for admission to the Animal Biology program at the University of California at Davis, Jimmy took the opportunity to travel to New Mexico with his parents in Septem ber of 2000. He decided to enroll at New Mexico State University in Las Cruces, New Mexico in July of 2001. During his undergraduate study at NMSU, Jimmy enrolled in Parasitology, where he met Dr. Ronnie Byford. His overwhelming interest and enthusiasm f or the field of veterinary entomology, prompted Dr. Byford to hire Jimmy as a laboratory aide at the NMSU Veterinary Entomology Laboratory. Jimmy graduated with a B.S. in Animal Science in December of 2003, and continued study as a m asters student under the instruction of Dr. Byford. Jimmys m asters research focused on determining the potential vectors of West Nile virus in Doa Ana County New Mexico. His work in this area was completed in December of 2006, for which he received an M.S. degree in animal s cience. Jimmy directed his focus on the field of veterinary entomology and pursuit of further education in this area of research. In January of 2007, Jimmy was accepted as a Ph.D. student at the University of Floridas Entomology and Nematology Departm ent, under the instruction of Dr. Phillip Kaufman. Jimmys research involves the ecology of stable flies associated with Florida equine facilities, including the study of their preferred hosts, insecticide susceptibility, and pupal parasitoids. He expect s to graduate in May of 2010.
146 Jimmy has presented the findings of his research at various scientific conferences including the annual meetings of the Entomological Society of America, the Southeastern Branch of the Entomological Society of America, and the Livestock Insect Workers Conference. He has also participated in the Student Science Training Program offered by the University of Florida for two years as mentor of high school students wishing to gain experience in scientific research at the collegiat e level. His ultimate career goal is to earn a position as a research scientist and professor at a major university.