9 Methods for Assessment of Contaminants of Invertebrate Biological Control Agents and Associated Risks

Mark S. Goettel and G. Douglas Inglis Lethbridge Research Centre, Agriculture and Agri-Food Canada, 5403–1st Avenue South, Lethbridge, Alberta T1J 4B1, Canada (email: [email protected]; [email protected]; fax number: +1-403-382-3156)

Abstract

With the importation or transport of any commodity, there exists the hazard that unwanted organisms or substances (i.e. ‘contaminants’) will be conveyed and introduced. Invertebrate biological control agents (IBCAs) can be contaminated with numerous biotic and abiotic enti- ties such as parasitoids, hyperparasitoids, pathogenic and/or non-pathogenic microorgan- isms, other organisms, pesticide residues, unwanted packaging materials, etc. Therefore, assessment of the risk posed by the contaminant must be addressed in the commerce of IBCAs. In this chapter, we provide an overview of possible contaminants of IBCAs and of the methods used to detect them. We consider two major factors when assessing risk. These are: (i) whether the IBCA is field collected or insectary reared; and (ii) whether the IBCA is exotic, being introduced for classical biological control or is indigenous and to be used for inundative biological control. We conclude that minimal risk is posed by contaminants of commercially mass-produced IBCAs, that are established in the area of use and are to be used inundatively. For such IBCAs, we recommend that the standards established for impor- tation of most commodities, such as many foodstuffs, plants, vegetables, fruits etc. (i.e. qual- ity control assurances by the producers) be adopted. Field-collected IBCAs, on the other hand, have a much higher potential for harbouring unknown contaminants that may repre- sent a risk. We recommend that feral IBCAs to be released outside of the area from which they were collected should be kept for at least one generation under quarantine, if at all pos- sible, and that the appropriate quarantine protocols are applied. This would allow the detec- tion and elimination of biotic contaminants. We stress that the key to the regulation of IBCAs is to address the extent of the possibility that a contaminant could pose a hazard to the com- modity or to the environment of the commodities’ final destination, and, if warranted, to ensure that such harm does not take place. The extent to which measures for prevention of transfer of contaminants are implemented must be weighed in relation to the present transfer of unknown or unwanted substances by other means. If the cautionary approach is strictly implemented for all possible contaminants, then almost certainly the international move- ment of IBCAs would grind to a halt. The ramifications of this must be weighed against the presently known benefits of IBCAs in our agriculture and forestry industries.

©CAB International 2006. Environmental Impact of Invertebrates for Biological Control of Arthropods: Methods and Risk Assessment (eds F. Bigler et al.) 145 146 M.S. Goettel and G.D. Inglis

Introduction could affect the IBCA’s efficacy, the health of the user, or which could become estab- In the importation or transport of any com- lished or pollute the new environment. modity, there is always a concern that Possibilities include pathogenic or non- ‘contaminants’ will be conveyed and intro- pathogenic microorganisms, parasitoids, duced. Before we proceed, it is first neces- hyperparasitoids, misidentified inverte- sary to define what constitutes a brates, pesticide residues, unwanted pack- contaminant. According to Merriam- aging materials, etc. In this chapter, we Webster’s Medical Dictionary (2003), a con- characterize possible contaminants as taminant is defined as: ‘a substance that either microorganisms, invertebrates or abi- contaminates; to contaminate is to soil, otic agents. stain or corrupt by contact; to tarnish; to pollute; contamination is the act of conta- minating or polluting including (either Microorganisms intentionally or accidentally) unwanted substances or factors.’ The key word in this Microorganisms are ubiquitous and they definition is ‘unwanted’. Restricting the are always found in association with both definition to unwanted presents a cadre of field-collected and mass-reared inverte- problems. What is ‘unwanted’ as far as an brates, including IBCAs. Their associations invertebrate biological control agent (IBCA) are complex, and their associations with is concerned? Human-pathogenic bacteria IBCAs can be considered as either inciden- associated with IBCAs may be unwanted, tal, mutualistic, pathogenic or commensal- yet the introduction of a small number of istic. It is important to emphasize that cells of a relatively weak human-patho- these categories are not necessarily exclu- genic microorganism does not necessarily sive of each other. A number of arthropods pose a serious threat. Microorganisms are vector mammalian (e.g. West Nile Virus) ubiquitous and no invertebrates are devoid and plant pathogens. Furthermore, micro- of them unless special measures are taken. organisms associated with IBCAs may be How does one determine whether a micro- pathogens of invertebrates, plants or verte- organism falls into the ‘unwanted’ cate- brates, including humans. Groups of micro- gory? To determine this, the definition of organisms dealt with in this chapter contaminant must also address risk, and include the viruses, bacteria, fungi and this paper defines contamination as the protozoa. For convenience, we also include inclusion of any unwanted substance or the nematodes in this section. Numerous factor (i.e. a contaminant) in the commerce examples of pathogens of beneficial arthro- of IBCAs that poses an unacceptable risk. pods are provided by Vinson (1990), and of In defining unacceptable risk, we limit our mass-produced IBCAs by Bjørnson and discussion to the impacts of contaminants Schütte (2003). on the health of the IBCAs or on humans, and their potential impact on ecosystems Viruses (e.g. introduction of non-indigenous micro- organisms). We also compare risk assess- Viruses are obligate, intracellular ments applied to other invertebrates in pathogens that consist of double-stranded some OECD (Organization for Economic or single-stranded nucleic acid (DNA or Cooperation and Development) countries. RNA) encased in a protective coating called a capsid. Collectively, the nucleic acid and capsule are termed a nucleocap- Contaminants Associated with sid. Depending on the virus, some nucleo- Invertebrates capsids are enclosed within a lipid envelope. Virions are the infectious unit of Using the above definitions, a contaminant a virus. In enveloped viruses, the virions could be one of numerous factors that consist of the nucleocapsid (i.e. nucleic Methods for Assessment of Contaminants of Invertebrate BCAs 147

acid and capsid) and the envelope. In non- by the hymenopteran parasitoid, Cotesia enveloped viruses, the virions are com- marginiventris (Cresson) (Hymenoptera: prised only of the nucleocapsid. Viruses do Braconidae), from infected to healthy not possess the ability to replicate them- Spodoptera larvae (Lepidoptera: selves independently of a living host, and Noctuidae) (Hamm et al., 1985). Viral thus cannot be cultured on microbiological pathogens are also present in numerous media. They, in essence, highjack the meta- mite and insect species that are used in bolic machinery of the host cell and trick it biological control (Bjørnson and Schütte, into producing progeny viruses. A number 2003). For instance, cytovirus and nuclear- of entomopathogenic viruses produce polyhedroviruses are known from the occlusion bodies (OBs), in which the viri- aphid predator, Chrysoperla (Neuroptera: ons are embedded within a paracrystalline Chrysopidae) (Martignoni and Iwai, 1986). protein matrix. The OBs protect the virions Unidentified, non-occluded virus particles (i.e. from ultraviolet light) and increase were observed in the yolk of predatory persistence of the virions outside of the Neoseiulus cucumeris (Oudemans) host; their formation has important conse- (Mesostigmata: Phytoseiidae) and quences for their disease-producing poten- Phytoseiulus persimilis Athias-Henriot tial. They may also serve an important (Mesostigmata: Phytoseiidae) mites; how- function in the infection process. ever, the effects of these on predatory effi- Most of the viruses associated with cacy were not established (Bjørnson et al., insects belong to one of 12 viral families, 1997). More information on entomopatho- but many remain unclassified. Of particular genic viruses can be found in Granados concern to IBCAs are viruses in the families and Federici (1986), Adams and Bonami Baculoviridae, Poxviridae, Parvoviridae, (1991a), Tanada and Kaya (1993), Miller Reoviridae and Polydnaviridae. Some of (1997), Hunter-Fujita et al. (1998) and these viruses possess restricted host ranges Miller and Ball (1998). affecting insects in a specific genus, whereas others can affect a variety of hosts Bacteria belonging to different orders. For most viruses of IBCAs, the primary route of The bacteria represent a very large and infection is through the alimentary canal diverse group of prokaryotes. There are two after ingestion. However, other routes of main types of prokaryotes, the archaeabac- infection do occur (e.g. mechanical intro- teria and the eubacteria (collectively duction of virions on infested ovipositors). referred to as bacteria). Although all bacte- In some instances, viruses are restricted to ria lack a nucleus and organelles, they are specific tissues (e.g. midgut epithelium), very diverse, both in morphology and but other viruses spread systemically, thus physiology. Some are single-celled, while affecting the entire body. As a general rule, others form filaments or aggregates. They viruses that are restricted to specific tissues may be spherical, rod-shaped, spiral or incite chronic disease, whereas systemi- lobed. Their size varies in diameter from cally transmitted viruses often cause acute 0.1 to more than 15 ␮m (filaments up to disease. Virions are typically released into 200 ␮m). Most produce a well-defined cell the environment in frass or from cadavers. wall. The archaebacteria differ from the Epizootics in natural populations are com- eubacteria in many important respects, mon and, periodically, entire colonies can such as: (i) their cell walls lack the carbo- be wiped out in mass-rearing operations. hydrate, peptidoglycan; (ii) their lipid Viruses are often intimately associated bilayer membranes consist of branched with both parasitoids and predators and chain hydrocarbons linked by ether link- their hosts and prey (Vinson, 1990). ages to glycerol; and (iii) many archaebac- Parasitoids are often implicated in the teria live in extreme environments and are transmission of the virus to the host. For very difficult to culture. Phylogenetic stud- example, an ascovirus can be transmitted ies indicate that the archaeabacteria are 148 M.S. Goettel and G.D. Inglis

close relatives of the eukaryotes. The vast Muscidifurax, Nasonia and Trichogramma, majority of bacteria associated with IBCAs and in predators such as Adalia, are eubacteria, and they can be divided Phytoseiulus, Neoseiulus and Metaseiulus into two groups based on cell wall mor- (Stouthamer et al., 1999). In a survey of phology (i.e. Gram positive or negative). pest and beneficial arthropods studied by Most eubacteria are saprotrophs, but some researchers at Agriculture and Agri-Food are facultative or obligate pathogens, and Canada, infections of Wolbachia were some form mutualistic symbioses with detected in 40 of the 65 species examined. IBCAs. A number of bacteria associated Taxa within the Acari (Tetranychidae), with arthropods are pathogenic to verte- Anoplura (Haematopinidae, Linognathidae, brates, including humans. Pediculidae) Coleoptera (Chrysomelidae, Normally, IBCAs are never devoid of Curculionidae), Diptera (Muscidae, bacteria, whether they are feral or reared in Calliphoridae), Hymenoptera (Braconidae, captivity. Saprotrophs catabolize non- Encyrtidae, Pteromalidae, Tricho- living organic matter and are common grammatidae), Mallophaga and Siphon- within the alimentary canal of IBCAs, as aptera (Pulicidae) were all infected (G. well as on their external integuments. Kyei-Poku and K. Floate, Alberta, 2004, Although they may be commensalistic, in personal communication). some instances they have been shown to be Serratia marcescens Bizio (Entero- beneficial to the arthropod, providing a bacteriales: Enterobacteraceae) is a com- degree of protection from pathogenic mon contaminant in reared insects. In microorganisms. They may also be general, it is not a very virulent pathogen, pathogens themselves (i.e. facultative), able causing disease only when insect vigour is to infect arthropods under stress. This is a greatly reduced (Sikorowski and Lawrence, common scenario in rearing settings (Inglis 1997). For example, Lighthart et al. (1988) and Sikorowski, 2005a). Other bacteria are found that a high-temperature pulse (i.e. a more specialized pathogens. The best physiological stressor) before inoculation known of insect-pathogenic bacteria is the with S. marcescens greatly increased the spore-forming Bacillus thuringiensis (Bt) susceptibility of Metaseiulus occidentalis Berliner (Baciliales: Bacillaceae). Other (Nesbitt) (Mesostigmata: Phytoseiidae) to entomopathogenic bacteria are found in the the bacterium. Greany et al. (1977) found genera Bacillus, Aeromonas, Clostridium, that optimizing Opious longicaudatus Paenibacillus, Photorhabdus, Pseudomonas, Ashmead (Hymenoptera: Braconidae) host- Rickettsia, Rickettsiella, Serratia, Wolbachia rearing conditions greatly reduced para- and Xenorhabdus. Some are opportunistic sitoid mortality attributed to bacteria, pathogens, whereas others are highly including S. marcescens. evolved pathogens. Other bacterial taxa Readers are referred to Tanada and Kaya also form symbioses with arthropods, but (1993), Charles et al. (2000), Glare and their effect is beneficial to the host and the O’Callaghan (2000) and Siegel (2000) for bacterium (i.e. a mutualistic symbiosis); in more information on entomopathogenic many situations, the presence of the bac- bacteria. terium is essential to the survival of the arthropod. Fungi Wolbachia is commonly associated with a diverse array of organisms. It is an intra- Fungi represent a diverse assemblage of cellular parasite, and it may have pro- non-phylogenetically related microorgan- found negative effects on the reproductive isms (representing at least three fitness of IBCAs, although not necessarily Kingdoms). They are grouped together on host fitness (Zchori-Fein et al., 2000). since they are all eukaryotes, they usually Among beneficials, Wolbachia is ubiqui- produce hyphae and possess rigid cell tous and it has been found in parasitoids walls, and they are all heterotrophs (i.e. such as Aphytis, Encarsia, Lysiphlebus, organisms that utilize organic matter as a Methods for Assessment of Contaminants of Invertebrate BCAs 149

source of energy). Some fungi are adapted Protozoa to existence in liquid environments and The protozoa are also a diverse assemblage produce unicellular growth forms (i.e. of non-phylogenetically related eukaryotic yeasts). Reproduction may be sexual or microorganisms. They can exist in mutual- asexual; some types of fungi produce both istic symbioses with insects. For example, types of spores, others produce either sex- the hindgut in termites houses protozoa ual or asexual spores. Fungi are primarily that hydrolyse cellulose. This is an obligate decomposers of non-living organic matter, symbiosis, and neither the protozoa nor the but some have evolved highly specialized termites can survive without each other. relationships with arthropods. Some form Protozoan pathogens of arthropods are typ- mutualistic relationships with those such ically single-celled organisms possessing as leafcutting ants or Ambrosia beetles, varied characteristics and little taxonomic which cultivate fungi as a source of food, affinity among groups (Solter and Becnel, and with polyphagous chrysopid adults, 2000). Many species are obligate pathogens which utilize yeasts to provide essential and have complicated life cycles, some nutrients. Other relationships with fungi with intermediate hosts. Most infections are benign or detrimental. Most ento- are chronic and non-lethal, but typically mopathogenic fungi are members of two result in reduced fecundity. divisions, the Zygomycota and the The phylum Apicomplexa contains the Ascomycota. Some are obligate pathogens insect-pathogenic gregarines. The most infecting specific species, some are less commonly encountered are eugregarines specialized, able to infect a variety of host with species within Gregarina, whereas species, whereas others are facultative Farinocystis, Mattesia and Ophryocystis are pathogens, only able to infect insects that neogregarines commonly producing lethal are immunocompromised. Fungal epi- infections in dipteran, coleopteran and zootics are common in some insect species, hemipteran hosts. Most neogregarines have while others are rarely affected. Fungi are narrow host ranges; however, others, such unique among the insect pathogens as their as Farinocystis tribolii Weiser and Mattesia primary route of entry into the host is via grandis McLaughlin (Neogregarinorida: the external integument. Lipotrophidae), have a relatively wide host Many beneficial invertebrates are sus- range. Many neogregarines are found as ceptible to entomopathogenic fungi contaminants in insectaries. Members of (Goettel et al., 1990; Bjørnson and Schütte, the phylum Ciliophora are usually found 2003; Vestergaard et al., 2003). For in the larval and adult stages of dipterans. instance, Neozygites spp. have been found The two most common genera are infecting Neoseiulus and Macrochelus. Lambornella and Tetrahymena. The phy- Some species, such as Beauveria bassiana lum Rhizopoda includes the amoebas such Balsamo (Vuillemin) (Hypocreales: as Malameba locustae (King and Taylor) – Clavicipitaceae), are ubiquitous and have a infecting acridids, and Malpighamoeba very wide host range, infecting many mellificae Prell (Amoebida: Amoebidae) – IBCAs. This fungus is often found infecting infecting honeybees. overwintering coccinelids. Some fungal The phylum Microsporidia have tradi- taxa may also adversely affect humans, but tionally been considered to be primitive pro- this is primarily restricted to rearing set- tozoa. However, recent evidence indicates tings in which facultative pathogens colo- that they are actually highly evolved intra- nize organic materials (e.g. insect diets) cellular fungi (Keeling and Fast, 2002). and propagules are released into the rear- Nonetheless, we discuss them here as proto- ing environment (Inglis and Sikorowski, zoa, largely because they are traditionally 2005a,b). Readers are referred to Samson et handled with this phylogenetically diverse al. (1988), Tanada and Kaya (1993) and group of microorganisms. The microsporidia Butt et al. (2001) for more information on is a large group (approximately 1000 species) entomopathogenic fungi. of obligate intracellular pathogens affecting a 150 M.S. Goettel and G.D. Inglis

variety of vertebrates and invertebrates. externally on the exoskeleton or internally Approximately 600 species have been in the reproductive, respiratory, digestive reported as infecting insects. Insects in virtu- or excretory systems, or within the haemo- ally all orders possess members in which coel, where they subsist causing very little microsporidial infections have been docu- or no apparent damage to their host. Some mented. Some species of entomopathogenic of these commensal nematodes are microsporidia possess a narrow host range phoretic, utilizing insects for dispersal. (e.g. one host species), whereas other species Other nematodes, including free-living possess a wide host range, which includes nematodes, are saprotrophs, and may uti- vertebrates. lize insect cadavers merely as a nutrient Protozoa are of concern as pathogens of source. Many nematodes are animal and IBCAs; they primarily incite sublethal, plant parasites that use the insects as vec- debilitating disease, although acute disease tors. Examples include those nematodes may occur in some instances and they are responsible for onchocerciasis, eyeworm common in mass-reared IBCAs. The most and elephantiasis in humans, and for common entomopathogenic genus is canine heartworm, while plant-parasitic Nosema, which has more than 150 nematodes vectored by insects include described species reported from at least 12 those responsible for pine wilt and red ring insect orders. For instance, N. muscidifu- disease of coconut. racis Becnel and Geden (Microsporidia: The most commonly encountered ento- Nosematidae) is prevalent in mass-reared mopathogenic nematodes are included in Muscidifurax used for fly control (Geden et three major families, the Mermithidae, al., 1995). Infected parasitoids have an Steinernematidae and Heterorhabditidae. extended developmental time, are shorter All are obligate insect pathogens and gain lived and have a much reduced fecundity. entry through the cuticle, spiracles, mouth, Microsporidians have been found in anus or via mechanical injury. many genera of beneficial insects, includ- Although steinernematid and het- ing Coccinella, Cotesia, Encarsia, and erorhabditid nematodes are generally non- Phytoseiulus (Bjørnson and Shütte, 2003), specific insect pathogens, natural Metaseiulus (Olsen and Hoy, 2002) and epizootics caused by entomopathogenic Tachinaephagus (Ferreira del Almeida et nematodes are relatively rare in nature, and al., 2002). Recently, a microsporidian was nematodes are not commonly encountered found to be responsible for the decline of in rearing settings. In addition, infections the weed biological control weevils, in field populations of beneficial insects Neochetina eichorniae Warner and N. such as predators and parasitoids are rare, bruchi Hustache (Coleoptera: Erirhinidae), even after inundative nematode applica- originally introduced from South America tions (Akhurst, 1990). and mass produced in Florida for control of water hyacinth (ARS, 2004). This microsporidian was found to decrease sur- Invertebrates vival rates of the weevils by 30%, and their reproductive capacity by 60 to 70%. The Invertebrates are almost always associated original source of this contaminant within in one way or another with other insects. the rearing facility is not known. Sweep samples from field collections are an attestation to the biodiversity of insects within ecosystems. Insects can harbour Nematodes commensals such as phoretic mites; pseu- Thirty families of nematodes within six doscorpions and the like; ectoparasites orders are associated with insects. The such as parasitic mites; and endopara- most common association between nema- sitoids such as tachinid flies, or endo- todes and insects is apparently commensal- hyperparasitoids such as braconid wasps. istic. Such nematodes can be found Insect–insect associations can be mutualis- Methods for Assessment of Contaminants of Invertebrate BCAs 151

tic, such as the classic relationship taminant, especially if it is a species that is between aphids and some ants. not the same target host species, or if it sig- Hyperparasitoids often occur in IBCAs. nificantly increases the pest population in For instance, Mesochorus curvulus the area of introduction. For example, Thompson (Hymenoptera: Ichneumonidae) whitefly puparia can accompany ship- is a hyperparasitoid of Peristenus spp. ments of Encarsia puparia. In addition, (Hymenoptera: Braconidae), a biological other incidental species could co-occur control agent of European lygus bugs, Lygus with the IBCA. For instance, a number of rugulipennis Poppius and L. pratensis soil-dwelling mites could co-occur in ship- Linnaeus (Hemiptera: Miridae) (Day, 2002); ments of the predatory soil mites, various aphidiine braconids and aphelinids Hypoaspis spp. in the genera Aphelinus, Aphidius, Ephedrus, Lysiphlebus and Trioxys are par- asitized by a variety of hyperparasitoids; Abiotic contaminants many Encarsia species are facultative hyperparasitoids of other primary para- Inanimate compounds or agents that are sitoids; and convergent ladybird beetles either detrimental to the efficacy of the field-collected in California and exported as agent or to the safety of the user or environ- IBCAs may be parasitized by Dinocampus ment of introduction could conceivably spp. (Hymenoptera: Braconidae). accompany biological control agents. The IBCAs can be contaminated with species list of possible inanimate contaminants is of similar appearance or a species can be limitless. Unless intentionally introduced, shipped in error. For instance, predatory existence of contaminants that may harm mite species in the genus Amblyseius the biological agent itself are more probable (Acari: Phytoseiidae) look very much alike, than those that may harm the environment even to taxonomic specialists. European or user. Such compounds could include species of Orius are very similar to species chemically contaminated packaging mater- such as Orius insidiosus (Say) (Hemiptera: ial, inappropriate substrates that harbour the Anthocoridae), and could easily be insects, pesticide residues, etc. Inanimate shipped mistakenly to an importer who contaminants that harm the environment or might not recognize the error. Whenever user are more difficult to contemplate. any host plant material has to be shipped However, some possible contaminants in with insects, it can be difficult to find and this category could include toxic com- exclude eggs or young instars of predators pounds used to rear or decontaminate the that are small, inconspicuous and/or hid- insects from microorganisms. For example, den in leaf folds, under veins or in debris. fumigation of leafcutting bee cocoons with In shipments of aphids, such problem paraformaldehyde is carried out prior to predators have included syrphids and export, in order to decontaminate the cells cecidomyiids, in particular. of spores of Ascosphaera aggregata Skou Another problem that may occur is (Ascomycota: Ascosphaeraceae). Improper hitch-hiking facultatively polyphagous or aeration after fumigation could result in the saprophytic mites on field-collected host build-up of potentially dangerous levels material, sometimes phoretic on the target of formalin gases within the shipping insects. It can be very difficult to exclude containers, which in turn could potentially all of these from every shipment. Such be harmful to the recipient of the ship- arthropods have caused problems from ment. time to time in the quarantine cultures of olive fruit flies and artificial diets for other insects at the USDA European Biological Diagnostic and Detection Techniques Control Laboratory in Montpellier, France. Finally, the species used as a host in the The ability to detect and intercept the mass production of the IBCA can be a con- importation or transfer of contaminants 152 M.S. Goettel and G.D. Inglis

associated with IBCAs will very much and moulting insects are often erroneously depend on the diagnostic and detection suspected of being diseased. Insects sus- techniques available for the specific conta- pected of being diseased should first be minants in question. Here, we provide an examined externally, followed by dissec- overview of the techniques that are tion and macroscopic examination of the presently available for the detection of internal organs and tissues, at first with the such contaminants. naked eye, and thence with the aid of a stereomicroscope. Although this can pro- vide valuable information on the identity Microorganisms of the pathogen, it is usually insufficient to make a conclusive identification. Detailed Many commercialized techniques have observations involving microscopic exami- been developed for quick diagnosis of cer- nation of suspect tissues are usually tain microorganisms, mostly for rapid and required. Non-destructive diagnosis can routine diagnosis for presence of pathogens sometimes be made by examination of the of human concern. They can be divided haemolymph, faecal pellets or meconium. into methods involving: microscopy; Microscopic examinations are made classical microbiology; physiological char- using light or electron microscopy. The acters; protein detection and characteriza- first procedure usually entails use of a wet- tion; and nucleic acid detection and mount, wherein the whole insect, or spe- characterization. These methods can be cific tissues, are gently crushed in a drop of applied in vitro (if a microorganism is cul- water between the microscope slide and turable) and/or in vivo. Since many of the cover slip. Some entomopathogens can be microorganisms of concern incite disease, easily observed at the light microscope diagnostic methods involved with patho- level using many of the light microscopy genesis have been developed. They are: (i) techniques available. In some instances, differential diagnosis, where signs and tissues must be prepared for histological symptoms and postmortem changes are examination using standard sectioning and compared in a systemic manner between staining techniques for light and electron different diseases to distinguish one dis- microscopy. ease from another; (ii) preliminary diagno- While nematodes, fungi, bacteria and sis, which is the first cursory examination protozoa can be observed readily at the of a diseased insect; (iii) tentative diagno- light microscopy level with phase contrast sis, which is made after general macro- without the use of stains, use of differential scopic and microscopic examination and stains can greatly enhance visualization of some cursory laboratory tests; and (iv) these, as well as other, pathogens. For more definitive diagnosis, in which a final con- information on histological techniques clusion is made, and the disease-causing used to diagnose entomopathogens, the organism is identified. Facts to be collected reader is referred to Becnel (1997) and on which to base the definitive diagnosis Evans and Shapiro (1997). The reader is include: (i) history of disease; (ii) physical referred to Lacey and Brooks (1997) for a examination; and (iii) ancillary examina- key to the major groups of ento- tion. It is also important to stress that quan- mopathogens. titative assessments (i.e. how many Molecular methods for detecting the microorganisms are present) are critical in presence of microorganisms have advanced many risk-assessment schemes. tremendously in recent years and are rou- Common symptoms of the presence of tinely being used to detect microorganisms pathogens within IBCAs include sluggish- in the agri-food, veterinary and medical ness, reduced or cessation of feeding, sciences and related industries. The two colour change and reduced fecundity. primary strategies use either immunologi- Unfortunately, many of these symptoms are cal or nucleic acid-based methodologies. also associated with the moulting process, Immunological detection of insect Methods for Assessment of Contaminants of Invertebrate BCAs 153

pathogens relies on the application of specific template (e.g. a gene). This can mono- or polyclonal antibodies to antigens then be extrapolated to numbers of of the microorganism produced in mam- pathogens (e.g. virions) present. The reader mals or birds. Methods such as enzyme- is referred to reviews of Innis et al. (1990), linked immunosorbent assay (ELISA) are Persing (1996) and Caetano-Anollés and frequently used. In ELISA, the breakdown Gresshoff (1997) for more information on of a substrate bound to the antibody by an PCR-based detection of pathogens. enzyme causes a colour change, indicating In the following sections, we briefly out- the presence of the antigen. One of the line the techniques presently available for major problems with immunological meth- detection of specific groups of microorgan- ods is poor sensitivity to low amounts of isms from invertebrates. For detailed pro- antigens. cedures, the readers are referred to Poinar Nucleic acid-based methods for micro- and Thomas (1984), Lacey (1997) and organism detection can be much more sen- Inglis and Sikorowski (2005a,b), and refer- sitive. Initially, hybridization methods, ences therein. including Northern (i.e. RNA) and Southern (i.e. DNA) blots were used to Viruses qualitatively detect insect pathogens (St Leger and Joshi, 1997, and references All viruses are obligate parasites and they therein). Another, more powerful, method are typically detected and/or quantified in that relies on hybridization is fluorescence situ or following extraction of virions from in situ hybridization (FISH). The most insect tissues. The application of molecular powerful and widely adopted method uses detection methods targeting viral proteins the polymerase chain reaction (PCR). PCR or nucleic acids is now commonplace. multiplies specific regions of nucleic acid. However, microscopic examination using Following the amplification, nucleic acid light or electron microscopy can still pro- specific to the microorganism can be vide valuable information on the aetiology detected by electrophoresis with or with- of viral diseases. Light microscopy can be out hybridization. The use of nested PCR is used to visualize occlusion bodies, but is often necessary to obtain adequate levels of limited to observing tissue abnormalities sensitivity while providing specific ampli- (e.g. hypertrophy of the midgut epithelium) fication. Although PCR can be exception- for viral diseases in which occlusion bod- ally sensitive and specific, it must be ies are not produced. More specific proto- stressed that extreme care must be taken in cols for virus diagnosis and identification developing PCR methods. The inclusion of can be found in Adams and Bonami an internal amplification control (IAC) is (1991b), Tompkins (1991), Evans and becoming mandatory for diagnostic PCR; Shapiro (1997) and Inglis and Sikorowski an IAC is a non-target nucleic acid (2005a). sequence present in the same sample reac- tion tube, which is coamplified simultane- Bacteria ously with the target sequence. If an IAC is not included, it is unknown whether a neg- Bacteria associated with arthropods are pri- ative response represents a true or false marily saprotrophs (including facultative negative (i.e. the reaction could be inhib- pathogens), but some are obligate parasites. ited due to malfunction of the thermal Detection and quantification of sapro- cycler, incorrect PCR mixture, poor poly- trophic bacteria is often accomplished by merase activity and/or the presence of isolation of cells using selective or non- inhibitory materials). Recently, the applica- selective media. For qualitative assess- tion of real-time quantitative PCR (RTQ- ments of bacteria, insects are typically PCR), in which the amplification process is homogenized and the homogenate plated monitored in real time, has made it is pos- on an appropriate agar medium. Individual sible to estimate the initial quantity of a colonies are then subcultured to ensure 154 M.S. Goettel and G.D. Inglis

purity. Once in pure culture, bacteria are incubated for more than several days, as typically identified based on morphologi- eventually saprotrophic fungi will over- cal (e.g. cell shape and cell wall structure), come any insect cadaver placed under physiological (e.g. assimilation of carbohy- humid conditions. However, if an ento- drates) and/or molecular (e.g. 16S rDNA mopathogen was the cause of death, it will sequence) characters. Quantitative assess- usually surface on the cadaver before ments of cell densities using microbiologi- saprotrophic fungi do. Some entomophtho- cal methods typically involve the use of ralean fungi do not readily grow in culture, the dilution spread-plate or most probable and therefore identification must be made number methods. Detection of intracellular from material obtained directly from the obligate parasites can typically be made cadaver. only by using molecular techniques. This If diagnosis is necessary prior to death, usually involves the visualization of cells the insect can be sacrificed and a wet of a particular taxon by microscopy using mount of the haemocoel can be examined in situ hybridization, or the application of for the presence of hyphal bodies (some- conventional PCR-based detection meth- times termed ‘blastospores’), which are ods. For conventional PCR, taxon-specific essentially short fragments of mycelium. primers (i.e. short segments of DNA that The appearance and size of the hyphal anneal to complementary sequences in the bodies may provide some evidence as to target nucleic acid) using universal genes the type of fungus involved; however, posi- (e.g. 16S rRNA genes), or specific to genes tive identification to the genus level is usu- unique to the taxon of interest, are used. ally not possible. For example, to detect Wolbachia infec- Many of the fungi associated with tions, primers specific to genes encoding arthropods are saprotrophic, and thus they proteins on the surface of the cell wall are can be cultured on microbiological media. used (Zhou et al., 1998; Stouthamer et al., As with the bacteria, fungi are typically 1999; Kyei-Poku et al., 2003). The use of isolated on selective or non-selective real-time quantitative PCR is becoming media. Identifications of filamentous fungi more popular in detection of bacteria in (i.e. fungi producing hyphae) are primarily situ; this method also allows for the quan- based on sporogenesis. In contrast, the tification of fastidious bacteria and obligate identification of yeasts primarily relies on parasites. physiological characters. However, molecu- More information on entomopathogenic lar methods are becoming more widely bacteria and their detection can be found used to characterize fungi. The most com- in Tanada and Kaya (1993), Klein (1997), monly used genes are the 18S and 26S Thiery and Frachon (1997), Charles et al. rRNA genes and regions between them. (2000) and Inglis and Sikorowski (2005a). Accurate quantification of filamentous fungi using microbiological media is prob- lematic given their growth form. For exam- Fungi ple, many fungi are r-selection organisms Entomopathogenic fungi are most easily producing copious quantities of asexual diagnosed on insect cadavers. If an ento- spores. Each propagule is capable of pro- mopathogenic fungus is suspected, and no ducing a colony on an agar medium, and outward growth of mycelia is visible on the therefore microbiological quantification cadaver, placement of the cadaver in a ster- methods grossly overestimate biomass of ile humid chamber will induce outward such fungi. Some fungi associated with growth of mycelia and production of coni- arthropods (e.g. many entomophthoralean dia on the cadaver surface. The fungus can and all laboulbinalean fungi) cannot be then be isolated into pure culture by asep- cultured on agar media. Therefore, they tically transferring the conidia or hyphae to must be allowed to sporulate directly on an appropriate agar medium. Care must be the cadaver before they can be identified taken to ensure that the cadaver is not microscopically. Methods for Assessment of Contaminants of Invertebrate BCAs 155

For more information on entomopatho- microsporidia (Weiss and Vossbrinck, genic fungi and their diagnosis, readers are 1999), the most likely target for primers to referred to Poinar and Thomas (1984), detect entomopathogenic microsporidia is Samson et al. (1988), Goettel and Inglis the universal 18S rRNA gene. Primers to (1997), Humber (1997), Lacey and Brooks other genes have been used to detect (1997), Papierok and Hajek (1997), Butt et human-pathogenic taxa, and these may al. (2001) and Inglis and Sikorowski prove useful in detecting some ento- (2005a). mopathogenic microsporidia as well (Weiss and Vossbrinck, 1999). Keys to entomopathogenic protozoa and Protozoa more details on classical diagnosis of Characterization of entomopathogenic pro- infected insects can be obtained by con- tozoa is difficult and currently relies sulting Brooks (1988), Undeen and Vávra almost exclusively on microscopic charac- (1997), Becnel and Andreadis (1999), ters. However, detection of spores within Solter and Becnel (2000) and Inglis and the insect is relatively easy. It is often pos- Sikorowski (2005a). sible to make preliminary diagnoses through direct observation of the squashed Nematodes cadaver in wet mounts, where large num- bers of spores are usually visible. Spores Nematodes can be easily visualized under can readily be detected in smears using magnifications of 10 to 100 ϫ under a Giemsa, Trichome or Buffalo Black stains stereomicroscope. In some cases, the nema- with bright-field microscopy, or using cal- todes can be seen within the body of the cofluor with fluorescence microscopy intact insect. The insect can be dissected to (Vavra and Chalupsky, 1982; Didier et al., liberate the nematodes, but identification 1994; Inglis and Sikorowski, 2005a). In of most species of entomopathogenic addition, there have been efforts to develop nematodes requires the adult stage. indirect antibody detection of However, the stages that are present microsporidia (Didier et al., 1994; Green et within, or emerge from, the host are usu- al., 2000). ally not the adult stage and must be held Development of molecular characters to under appropriate conditions until they detect entomopathogenic protozoa in situ mature. The reader is referred to Gaugler would simplify diagnostics; however, this and Kaya (1990) and Kaya and Stock (1997) area of investigation is still in its infancy. for more details and for a key to ento- Primers to detect specific human-patho- mopathogenic nematodes. genic enteric microsporidia either in water or in faecal samples have been developed (Muller et al., 2001; Dowd et al., 2003). Invertebrates However, the tremendous diversity of ento- mopathogenic microsporidia and the lack Most invertebrate contaminants such as of economic incentive to develop primers ectoparasitoids, commensals and inciden- have hindered the development and appli- tals are visible to the naked eye. cation of PCR-based detection methods for Consequently, close examination of the entomopathogenic taxa. One example shipment or anaesthetized biological con- where PCR has been applied for detection trol agents with the aid of a hand lens of an entomopathogenic microsporidian is would reveal the presence of such inverte- Thelohania solenopsae Knell, Allen and brates. Early stages of parasitism are very Hazard (Microsporidia: Thelohaniidae) difficult to detect and may be facilitated by infections of red imported fire ants, molecular methods. For instance, Ratcliffe Solenopsis invicta Buren (Hymenoptera: et al. (2002) used PCR to detect the pres- Formicidae) (Valles et al., 2002; Milks et ence of early-stage parasitoids in fly pupae. al., 2004). As with other taxa of On the other hand, it is possible to detect 156 M.S. Goettel and G.D. Inglis

later stages through dissection and exami- sible contamination of IBCAs by abiotic nation, by the naked eye or under magnifi- contaminants. Such contamination would cation. However, it is difficult to observe be very difficult to predict a priori. hyperparasitoids through dissection, as their larvae are inside their primary para- sitoid, which is inside the primary host. Defining the Risk Posed by IBCA Another method would be to rear the Contaminants insects through a generation, as virtually all invertebrate parasites will have completed In risk assessment, risk is usually defined their life cycle and emerged as adults as ‘hazard ϫ probability’ (Zadoks, 1998). A within the lifespan of their host. For hazard is any imaginable adverse effect instance, European Peristenus species are that can be identified. Once a hazard has obtained as cocoons that have emerged been identified, it is then necessary to from parasitized, field-collected lygus assign a probability or likelihood of occur- nymphs and these are shipped to North rence. With biological entities such as America. The cocoons usually require over- microorganisms, hazards typically remain wintering in quarantine in North America imprecise. Furthermore, assigning a prob- before emerging the following spring. The ability that the hazard will occur to benefi- hyperparasitoid Mesochorus is screened out cial organisms is difficult. As a result, at emergence from the Peristenus. Recently, decisions regarding risks associated with Ashfaq et al. (2005) developed PCR primers biological organisms are rarely based which were used successfully to detect purely on scientific data. Nevertheless, rel- Mesochorus spp. within Peristenus within atively strict approval standards are cur- the lygus primary hosts. rently imposed on the application of Proper recognition and identification of microorganisms (e.g. plant protection prod- the IBCA is necessary to prevent accidental ucts) in many jurisdictions throughout the introduction of similar-appearing species. world, and guidelines for commerce in Voucher specimens can be sent to special- IBCAs are being considered for implemen- ists for taxonomic verification (see tation. Therefore, one way or another, risk Stouthamer, Chapter 11, this volume). assessments must be made. Observations on behaviour and life history attributes can also often signal the possibil- ity that one is dealing with a contaminating Microorganisms species. Pathogens of invertebrates or plants Abiotic contaminants Many microorganisms have been devel- oped as commercial microbial control As mentioned above, the types or nature of agents, or have been used in the classical conceivable inanimate contaminants that biological control of pest insects, with no could potentially affect the agent’s efficacy or minimal impact on biodiversity or envi- or harm the environment of introduction ronmental health (Laird et al., 1990; are virtually limitless, especially if the con- Goettel and Hajek, 2001; Goettel et al., tamination is intentional. Detection of abi- 2001; Hokkanen and Hajek, 2003). There otic agents, such as toxins and poisons, are also numerous examples of how ento- that may affect an agent’s efficacy is diffi- mopathogens can be used safely in con- cult; however, these often can be narrowed junction with IBCAs (e.g. Laird et al., 1990; down in many cases to suspected sources Hokkanen and Hajek, 2003). of contamination (e.g. fumigation at point Microorganisms pathogenic to IBCAs of arrival, etc.). It is beyond the scope of are a primary concern as far as the efficacy this chapter to cover the methods that of the IBCAs themselves is concerned. would be required to determine the pos- Disease incited by pathogens is often detri- Methods for Assessment of Contaminants of Invertebrate BCAs 157

mental to the host, resulting in reduced into four categories based on the threat longevity and death. Pathogenic micro- they represent to human and animal health organisms are divided into obligate or fac- (Health Canada, 2004). These include: (i) ultative pathogens. As a general rule, level 2 pathogens, which represent a mod- obligate invertebrate pathogens possess erate individual risk and limited commu- narrow host ranges, whereas facultative nity risk; (ii) level 3 pathogens, which pathogens infrequently incite disease in represent a high individual risk but a low vertebrate hosts. Furthermore, facultative community risk; and (iii) level 4 pathogens are ubiquitous, whereas obligate pathogens, which represent a high individ- pathogens are typically intimately associ- ual risk and a high community risk. Of the ated with their hosts. In addition to direct human pathogens, level 2 microorganisms effects on IBCAs, the possibility of trans- are most commonly found associated with mission to other invertebrates exists for invertebrates. These include representa- pathogens possessing wide host ranges. tives of bacteria, fungi, protozoa, nema- IBCAs can conceivably be contami- todes and other parasitic microorganisms. nated with plant pathogens, especially if Examples of level 2 pathogens sometimes host plant material is transported along found in association with invertebrates with the IBCA. For instance, concerns include: Aspergillus spp., Bacillus cereus have been raised regarding the possibility Frankland and Frankland (Bacillales: of Macrolophus caliginosus Wagner Bacillaceae), Clostridium spp., Crypto- (Hemiptera: Miridae) transmitting pepino coccus spp., Enterobacter spp., Lacto- mosaic virus to tomatoes (Bolckmans, bacillus spp., Micrococcus spp., 2003). Pseudomonas spp., Salmonella spp., The risk is dramatically greater if the Serratia spp., Staphylococcus aureus contaminating microorganism is exotic (i.e. Rosenbach (Bacillales: Staphylococcaceae), it does not already occur in the area of Streptococcus spp., and Yersinia spp. introduction) rather than indigenous (i.e. it Although they are capable of inciting dis- is already present in the area of introduc- ease in animals, level 2 pathogens are tion), and it is capable of establishing itself unlikely to be a serious hazard to healthy in the area of introduction. Certainly, humans, the community, livestock or the exotic pathogens as contaminants of IBCAs environment (Health Canada, 2004). In are a potential hazard, and therefore pose a healthy animals, exposure levels required higher risk and should be avoided. The dif- to incite infection are typically high. ficulty lies in assessing the probability of Furthermore, effective treatment and pre- their occurrence and of quantifying the ventive measures are available and the risk hazard they represent. The reader is of spread is limited. Therefore, the risk referred to Cook et al. (1996), Goettel and posed by level 2 pathogens associated with Hajek (2001) and Hajek et al. (2003) for fur- IBCAs is very low. For instance, some ento- ther discussions on the potential risks of mopathogenic microsporidia can infect the introduction of exotic pathogens. vertebrates (e.g. a Nosema species that infects mosquitoes also infects the cooler body parts of mice such as the tail, ears Pathogens of vertebrates and feet), but the zoonotic risk of insect- Insects reared in captivity frequently pos- pathogenic microsporidia is considered sess a bacterial microflora more typical of minimal at present. that found associated with humans. Since Most fungi associated with reared humans typically carry human-pathogenic insects originate from decomposing vegeta- bacteria within their gastro-intestinal tion. Some are human pathogens, and their tracts, respiratory organs, skin or hair, it is proliferation on organic matter (e.g. artifi- not surprising that insects reared in captiv- cial diets) and subsequent liberation of ity also carry human-pathogenic micro- large numbers of propagules can impact organisms. Human pathogens are classified negatively on the health of insectary work- 158 M.S. Goettel and G.D. Inglis

ers. Fungi such as Aspergillus, Penicillium, pathogen of grasshoppers, Entomophaga Rhizopus and a variety of yeast and yeast- praxibuli Humber, Milner and Soper like organisms can colonize insect diets (Entomophthorales: Entomophthoraceae), and may be hazardous to employees. Such in a classical biological control programme fungi may be capable of infecting humans for control of native grasshoppers in North directly, they may produce secondary America, might competitively displace or metabolites which can be toxic to humans even cause extinction of the native if they are ingested, or they can act as aller- Entomophaga grasshopper pathogens. gens. Inhalation of airborne fungal propag- However, in reality, infection levels were ules can cause allergic rhinitis or sinusitis, low and declining, suggesting that the hypersensitivity pneumonitis due to sensi- pathogen had little chance of establishing tization to fungal spores, and/or organic itself (Bidochka et al., 1996). To date, there dust syndrome caused by inhalation of is no evidence of displacement of an large quantities of toxin-containing micro- indigenous pathogen due to introduction of bial particles. Although such problems a microorganism for classical biological may be apparent in insectaries, they should control. The advent of molecular diagnostic normally not pose a problem as far as cont- techniques that enable one to track particu- aminants of IBCAs are concerned. Unless lar genotypes of a pathogen provides an contaminated diet is present with the opportunity to conduct more detailed stud- IBCA, quantities of fungal propagules car- ies on the potential of competitive dis- ried on the external exoskeletons of insects placement of native entomopathogenic or in their alimentary canal would typi- microorganisms by non-indigenous ones. cally be small. In some instances, insects that come in contact with faeces from humans or live- Invertebrates stock may be contaminated with more seri- ous pathogens (e.g. verotoxigenic Throughout history, many invertebrates Escherichia coli (Migula) Castellani and have been either intentionally or uninten- Chalmers (Enterobacteriales: Enterobacter- tionally introduced into new ecosystems, iaceae) or Campylobacter jejuni (Jones et where they have caused detrimental effects al.) Véron and Chatelain (Campylo- or become serious pests (Pimentel, 2002). bacterales: Campylobacteraceae). Such Furthermore, most predators and para- pathogens would be more prevalent in feral sitoids have the potential to seriously affect insects, but normally this would not occur the efficacy of the IBCA in question. in reared insects. Consequently, every effort must be made to avoid the presence of unknown inverte- brate contaminants in shipments of IBCAs. Competitive displacement Evidence indicates that introduction of most microorganisms into an ecosystem Abiotic contaminants (e.g. soil) has only a transient effect on the indigenous microflora (Alabouvette and As mentioned above, contamination due to Steinberg, 1998). The most common micro- abiotic elements would be very difficult to organisms associated with mass-rearing of predict a priori. Unless intentional, it is insects (see above) are ubiquitous and difficult to conceive that such contami- would not normally pose a hazard to the nants would pose a hazard beyond that of microbial flora if conveyed with the IBCA. the user or immediate vicinity of use. However, concerns have been raised regard- Details on source and treatment of the ing possible displacement of indigenous IBCAs prior to shipment would aid in the obligate insect pathogens. For instance, identification for possible presence of cont- Lockwood (1993) suggested that the intro- aminants. By and large, abiotic contami- duction of an exotic obligate fungal nants should pose a minimum risk. Methods for Assessment of Contaminants of Invertebrate BCAs 159

Guidelines for Assessing the Risk Australia (AQIS, 2004), quarantine stan- dards for importation of living organisms Although risk cannot be scientifically are generally specific to recognized defined, standards based on the precau- pathogens, and do not encompass non- tionary principle and familiarity are typi- pathogenic contaminants. cally relied upon. However, application of In considering risk acceptance of conta- such standards may not be relevant to cont- minants in IBCAs, the following points aminants associated with IBCAs. The should be considered. amount of effort used to detect potential ● IBCAs used are diverse and their pro- contaminants should be in direct propor- duction involves substantially different tion to the risks they pose to the user, to methods. the environment and to the IBCA itself. ● Microbial contaminants associated with A number of recommendations regard- insects are diverse and, in many ing risk assessment of microorganisms instances, their biology is poorly under- were agreed upon by participants of the stood. ‘Microbiological Plant Protection Products ● The risk of introduction of an obligate Workshop on the Scientific Basis for Risk pathogen of an IBCA is higher if the Assessment’, held in Stockholm, Sweden IBCA is field-collected than if it was (Anon, 1998). One of the six points agreed laboratory-reared. upon is directly relevant to contaminants ● There are currently no quality control of IBCAs. Within the production control (QC) standards for contaminants associ- heading, it was indicated that the ‘level of ated with IBCAs. acceptable contaminants should be judged ● Given the diversity of contaminants from a “risk acceptance” point of view’. encountered, logistics of testing are dif- The goal is to define ‘risk acceptance’ with ficult. respect to contaminants associated with ● Where testing of IBCA for contaminants IBCAs. is applied, the methods and comprehen- Guidelines for regulation of IBCAs must siveness of testing vary tremendously. be addressed and implemented relative to ● Contaminants associated with the IBCA the commerce of other commodities, typically occur in relatively small num- including invertebrates. For instance, cur- bers. rently there are no regulations for the ● Epizootics of disease in an insect popu- importation of many invertebrates such as lation are dependent on more than all species of aquatic snails, leeches, scor- simply dose. pions, spiders, the German cockroach, the ● There is global commerce in plants and Russian cockroach and Drosophila animals, yet for the most part, no or melanogaster Meigen (Diptera: minimal standards exist for contami- Drosophilidae) into many countries such nants associated with these entities. as Canada. As far as amphibians and rep- ● There is global movement of people tiles are concerned, the present Canadian with no standards applied as far as Food Inspection Agency (CFIA) policy human pathogens or commensal micro- reads ‘Please be advised that amphibians organisms are concerned. and reptiles (excluding turtles and tor- toises) are no longer regulated under the Health of Animals Regulations and as a result, no CFIA import permit is required, Recommendations nor a health certificate and no inspection will normally be done at the border. We consider that routine contamination Imports are permitted from any country, by incidental or commensal micro- for any use, to any destination in Canada’ organisms is to be expected and no addi- (CFIA, 2004). Even in countries with very tional precautions are needed. Although strict quarantine standards, such as such organisms may not necessarily be 160 M.S. Goettel and G.D. Inglis

‘wanted’, they are inevitable and in most exporters must be very vigilant regarding instances should pose a minimal risk. contaminants that affect efficacy of their Such organisms would normally not be product. Furthermore, many mass- very different to those that are found in produced IBCAs destined for export are numerous commodities that are exported relatively cosmopolitan species for use in large quantities around the world. In against cosmopolitan pests. Therefore we addition, abiotic contaminants should consider that minimal risk is posed by also be of minimal risk and should not contaminants of mass-produced IBCAs normally warrant special consideration. that are established in the area of use and Exceptions would be in situations where are to be used inundatively. It is expected there is fear of deliberate sabotage (e.g. that there is minimum concern regarding bioterrorism). contaminants if the IBCAs come from a The contaminants which warrant con- well-established and reputable rearing sideration are biotic agents that pose a source. However, we recommend that the threat to the IBCAs themselves, or to the principles established for importation of ecosystem of introduction. We consider most commodities such as many food- two factors that could affect the risk posed stuffs, plants, vegetables, fruits, etc. be by a contaminant in an IBCA that could be adopted for reared IBCAs (i.e. mandatory used when assessing risk. These are: (i) documentation of the QC status of IBCAs whether the IBCA is field-collected or with regard to contaminants). This would insectary-reared; and (ii) whether the eliminate the need to scrutinize every insect is exotic, being introduced primar- shipment. With respect to contaminants ily for classical biological control or is and risk, the key factors to consider are: indigenous and used primarily for (i) correct identity of the IBCAs; (ii) QC inundative biological control. Risk of data on natural enemies of the IBCAs (e.g. presence of pathogens within commer- pathogenic microorganisms and para- cially produced IBCAs should presumably sitoids); and (iii) information on the rear- be low, as good QC and pathogen manage- ing systems used (e.g. source of insects, ment should be an integral part of mass quarantine status, rearing systems, etc.), rearing (Inglis and Sikorowski, 2005a). In which would be important in gleaning contrast, use of feral insects provides an information on the potential for contami- increased risk, as it is difficult to predict nation (e.g. plant material used to rear or detect the presence of pathogens or IBCAs that may be potentially contami- parasitoids in feral populations. nated by phytopathogens). Although no recognized standards Field-collected IBCAs, in contrast to exist to date, there are attempts to adopt insectary-reared IBCAs where appropriate international QC standards for mass- QC standards have been applied, have a produced IBCAs (van Lenteren, 2003). much higher potential for harbouring Adoption of such QC standards will facil- unknown parasitoids, pathogens or other itate the detection of contaminants asso- contaminants. There is also a much ciated with mass-reared IBCAs that may greater potential that these may include have a detrimental impact on their effi- misidentified contaminants (i.e. similar- cacy, especially if sound detection appearing species). Certainly, every effort methodologies for microorganisms are must be made to prevent introduction of applied (Inglis and Sikorowski, 2005a,b). contaminants that could affect the IBCA As part of QC, natural enemies of the itself, or that could become established IBCAs should be routinely screened for in and become a pest per se. Consequently, commercial rearing insectaries. The unless already well established in the area impacts on success of the IBCA, and of introduction, field-collected IBCAs therefore of customer satisfaction, are warrant much stricter scrutiny for conta- such that the IBCA producers and minants. It is standard practice in many Methods for Assessment of Contaminants of Invertebrate BCAs 161

countries that such agents be held in an commodity final destination, and, if war- approved containment or quarantine facil- ranted, ensure that such harm does not ity (e.g. ARS, 1991) prior to release, and take place. The extent to which measures we recommend that this practice be for prevention of transfer of contaminants adopted for most field-collected IBCAs are implemented must be weighed in being introduced for the first time to an relation to the present transfer of eco-region. Ideally, such insects should be unknown or unwanted substances by kept for at least one generation under other means. For example, presently there quarantine. This would allow detection are no regulations for the importation of and elimination of any parasitoids and/or many invertebrates. Consequently, one pathogens that may have been included must compare the possibility of introduc- with the imported insects. For insects that tion of contaminants via IBCAs with can not be reared or for which there are other methods (i.e. transportation of limited numbers, representative samples people, forestry and agricultural prod- plus suspect individuals can be sacrificed ucts, etc.). for a contaminant (i.e. pathogen) check Certainly, regulations regarding importa- (see Inglis and Sikorowski, 2005a for the tion of invertebrates to be used in biologi- strategies used to detect and eliminate cal control must not be more stringent than entomopathogens from insectary-reared those for other organisms, as far as most IBCAs). The numbers used would depend contaminants are concerned. Exceptions upon their relative risk of harbouring may be those substances, more specifically pathogens or parasites. Such information microbial and other living organisms that may be obtained from monitoring the par- may be detrimental to the environment of ent feral populations. Strict QC and moni- introduction, especially those that could toring of viability will facilitate the become established. elimination of entomopathogens. We have identified two major points However, the detection of ento- that need to be considered in assessing mopathogens inciting disease in IBCAs as potential risk: (i) whether IBCAs are field- part of QC protocol is often overlooked. collected or mass-reared in an insectary; Furthermore, the detection and assess- and (ii) whether they are indigenous and ment of the risk represented by a contami- destined for use primarily in inundative nant requires considerable expertise. We biological control, or whether they are recommend that insect-rearing personnel exotic and destined for use primarily in obtain the appropriate training in the classical biological control. As a mini- methodologies used for diagnosis, and to mum, it is evident that QC procedures for assess the potential risk posed by contam- commercialized IBCAs should include inates, or alternatively, to obtain assis- tance from specialists. This is not only monitoring for entomopathogens. Field- essential in QC of IBCAs to be used in bio- collected IBCAs destined for use in classi- logical control programmes, but is a pre- cal biological control warrant a higher requisite for applying strategies within degree of scrutiny. rearing settings for management of ento- mopathogens and disease. Acknowledgements

Conclusions We wish to thank the following for their help in providing information and sugges- The key to regulation of IBCAs is to tions in completing this chapter: James address the extent of the possibility that a Becnel, Dave Gillespie, Kim Hoelmer, Jeff contaminant could pose a hazard to the Littlefield, Charles Pickett and Charles commodity, or to the environment of the Vossbrink. 162 M.S. Goettel and G.D. Inglis

References

Adams, J.R. and Bonami, J.R. (1991a) Atlas of Invertebrate Viruses. CRC Press, Boca Raton, Florida. Adams, J.R. and Bonami, J.R. (1991b) Preparation of invertebrate viruses and tissue for examination. In: Adams, J.R. and Bonami, J.R. (eds) Atlas of Invertebrate Viruses. CRC Press, Boca Raton, Florida, pp. 9–30. Akhurst, R.J. (1990) Safety to non-target invertebrates of nematodes of economically important pests. In: Laird, M., Lacey, L.A. and Davidson, E.W. (eds), Safety of Microbial Insecticides. CRC Press, Boca Raton, Florida, pp. 233–240. Alabouvette, C. and Steinberg, C. (1998) Fate of microorganisms introduced into soil, effects on autochthonous communities and activities. In: Alabouvette, C., Möllby, R., Steffen, M. and Zadoks, J.C. (eds) Proceedings Microbial Plant Protection Products – Workshop on the Scientific Basis for Risk Assessment. KEM, Stockholm, Sweden, pp. 57–64. Anon (1998) Agreed Document. In: Alabouvette, C., Möllby, R., Steffen, M. and Zadoks, J.C. (eds) Proceedings Microbial Plant Protection Products – Workshop on the Scientific Basis for Risk Assessment. KEM, Stockholm, Sweden, pp. 65–68. AQIS (2004) Australian Quarantine Inspection Service, http://www.aqis.gov.au (accessed 26 May 2005). ARS (1991) Biological Control Quarantine: Needs and Procedures. Publication ARS-99, USDA, Beltsville, Maryland. ARS (2004) Study reveals cause of disease in beneficial insects. Agricultural Research, April 2004, http://www.ars.usda.gov/lis/AR/archive/apr04/insect0404.htm (accessed 26 May 2005) Ashfaq, M., Erlandson, M. and Braun, L. (2005) Hyperparasitism by Mesochorus spp. (Hymenoptera: Ichneumonidae) in Peristenus sp. (Hymenoptera: Braconidae) and development of PCR primers for hyperparasitoid detection. Biological Control 32, 371–377. Becnel, J.J. (1997) Complementary techniques: preparations of entomopathogens and diseased speci- mens for more detailed study using microscopy. In: Lacey, L.A. (ed.) Manual of Techniques in Insect Pathology. Academic Press, San Diego, California, pp. 338–353. Becnel, J.J. and Andreadis, T.G. (1999) Microsporidia in insects. In: Wittner, M. (ed.) The Microsporidia and Microsporidiosis. American Society for Microbiology, Washington DC, pp. 447–501. Bidochka, M.J., Walsh, S.R.A., Ramos, M.E., St Leger, R.J., Silver, J.C. and Roberts, D.W. (1996) Fate of biological control introductions: monitoring an Australian fungal pathogen of grasshoppers in North America. Proceedings of the National Academy of Sciences USA 93, 918–921. Bjørnson, S. and Schütte, C. (2003) Pathogens of mass-produced natural enemies and pollinators. In: van Lenteren, J.C. (ed.) Quality Control and Production of Biological Control Agents. CABI Publishing, Wallingford, UK, pp. 133–165. Bjørnson, S., Steiner, M.Y. and Keddie, B.A. (1997) Birefringent crystals and abdominal discoloration in the predatory mite Phytoseiulus persimilis (Acari: Phytoseiidae). Journal of Invertebrate Pathology 69, 85–91. Bolckmans, K.J.F. (2003) State of affairs and future directions of product quality assurance in Europe. In: van Lenteren, J.C. (ed.) Quality Control and Production of Biological Agents. CABI Publishing, Wallingford, UK, pp. 215–224. Brooks, W.M. (1988) Entomogenous protozoa. In: Ignoffo, C.M. and Mandava, N.B. (eds) Handbook of Natural Pesticides, Vol V, Microbial Pesticides, Part A. CRC Press, Boca Raton, Florida, pp. 1–149. Butt, T.M., Jackson, C.W. and Magan, N. (2001) Fungi as Biocontrol Agents. CABI Publishing, Wallingford, UK. Caetano-Anollés, G. and Gresshoff, P.M. (1997) DNA Markers: Protocols, Applications, and Overviews. Wiley and Sons, New York. CFIA (2004) Pet Imports Canadian Food Inspection Agency, Animal Products, Animal Health and Production Division, http://www.inspection.gc.ca/english/anima/heasan/import/petse.shtml (accessed 26 May 2005). Charles, J.-F., Delécluse, A. and Nielsen-LeRoux, C. (2000) Entomopathogenic Bacteria: From Laboratory to Field Application. Kluwer Academic Publishers, Dordrecht, The Netherlands. Cook, R.J., Bruckart, W.L., Coulson, J.R., Goettel, M.S., Humber, R.A., Lumsden, R.D., Maddox, J.V., McManus, M.L., Moore, L., Meyer, S.F., Quimby, P.C., Stack, J.P. and Vaughn, J.L. (1996) Safety of microorganisms intended for pest and plant disease control: A framework for scientific evalua- tion. Biological Control 7, 333–351. Methods for Assessment of Contaminants of Invertebrate BCAs 163

Day, W.H. (2002) Biology, host preference, and abundance of Mesochorus curvulus (Hymenoptera: Ichneumonidae), a hyperparasite of Peristenus spp. (Hymenoptera: Braconidae) parasitizing plant bugs (Miridae: Hemiptera) in alfalfa-grass forage crops. Annals of the Entomological Society of America 95, 218–222. Didier, E.S., Orenstein, J.M., Aldras, A., Bertucci, D., Rogers, L.B. and Janney, F.A. (1994) Comparison of three staining methods for detecting microsporidia in fluids. Journal of Clinical Microbiology 33, 3138–3145. Dowd, S.E., John, D., Eliopolus, J., Gerba, C.P., Naranjo, J., Klein, R., Lopez, B., de Mejia, M., Mendoza, C.E. and Pepper, I.L. (2003) Confirmed detection of Cyclospora cayetanesis, Encephalitozoon intestinalis, and Cryptosporidium parvum in water used for drinking. Journal of Water Health 1, 117–123. Evans, H. and Shapiro, M. (1997) Viruses. In: Lacey, L.A. (ed.) Manual of Techniques in Insect Pathology. Academic Press, San Diego, California, pp. 18–53. Ferreira de Almeida, M.A., Gedfen, C.J., Boohene, C.K., Becnel, J.J. and Pires do Prado, A. (2002) Microsporidosis of Tachinaephagus zealandicus Ashmead (Hymenoptera: Encyrtidae) Memorias do Instituto Oswaldo Cruz, Rio de Janeiro 97, 527–530. Gaugler, R. and Kaya, H.K. (1990) Entomopathogenic Nematodes in Biological Control. CRC Press, Boca Raton, Florida. Geden, C.J., Long, S.J., Rutz, D.A. and Becnel, J.J. (1995) Nosema disease of the parasitoid Muscidifurax raptor (Hymenoptera: Pteromalidae): prevalence, patterns of transmission, man- agement, and impact. Biological Control 5, 607–614. Glare, T.R. and O’Callaghan, M. (2000) Bacillus thuringiensis: Biology, Ecology and Safety. John Wiley and Sons, Chichester, UK. Goettel, M.S. and Hajek, A.E. (2001) Evaluation of non-target effects of pathogens used for manage- ment of arthropods. In: Wajnberg, E., Scott, J.K. and Quimby, P.C. (eds) Evaluating Indirect Ecological Effects of Biological Control. CABI Publishing, Wallingford, UK, pp. 81–97. Goettel, M.S. and Inglis, G.D. (1997) Fungi: Hyphomycetes. In: Lacey, L.A. (ed.) Manual of Techniques in Insect Pathology. Academic Press, San Diego, California, pp. 213–249. Goettel, M.S., Poprawski, T.J., Vandenberg, J.D., Li, Z. and Roberts, D.W. (1990) Safety to nontarget invertebrates of fungal biocontrol agents. In: Laird, M., Lacey, L.A. and Davidson, E.W. (eds) Safety of Microbial Insecticides. CRC Press, Boca Raton, Florida, pp. 209–232. Goettel, M.S., Hajek, A.E., Siegel, J.P. and Evans, H.C. (2001) Safety of fungal biocontrol agents. In: Butt, T., Jackson, C. and Magan, N. (eds) Fungal Biocontrol Agents – Progress, Problems and Potential. CABI Publishing, Wallingford, UK, pp. 347–375. Granados, R.R. and Federici, B.A. (1986) The Biology of Baculoviruses. CRC Press, Boca Raton, Florida. Greany, P.D., Allen, G.E., Webb, J.C., Sharp, J.L. and Chambers, D.L. (1977) Stress-induced septicemia as an impediment to laboratory rearing of the fruit fly parasitoid Biosteres (Opius) longicaudatus (Hymenoptera: Braconidae) and the Caribbean fruit fly Anastrepha suspensa (Diptera: Tephritidae). Journal of Invertebrate Pathology 29, 153–161. Green, L.C., LeBlanc, P.J. and Didier, E.S. (2000) Discrimination between viable and dead Encephalitozoon cuniculi (microsporidian) spores by dual staining with Sytox Green and Calcofluor White M2R. Journal of Clinical Microbiology 28, 3811–3814. Hajek, A.E., Delalibera, I. Jr. and Butler, L. (2003) Entomopathogenic fungi as classical biological con- trol agents. In: Hokkanen, H.M.T. and Hajek, A.E. (eds) Environmental Impacts of Microbial Insecticides. Kluwer Academic Publishers, Dordrecht, The Netherlands, pp. 15–34. Hamm, J.J., Nordlund, D.A. and Marti, O.G. (1985) Effects of a nonoccluded virus of Spodoptera frugiperda (Lepidoptera: Noctuidae) on the development of a parasitoid, Cotesia marginiventris (Hymenoptera: Braconidae). Environmental Entomology 14, 258–261. Health Canada (2004) Laboratory Biosafety Guidelines. Publication No. 4252. Minister of Public Works and Government Services, Ottawa, Ontario, Canada. Hokkanen, H.M.T. and Hajek, A.E. (2003) Environmental Impacts of Microbial Insecticides. Kluwer Academic Publishers, Dordrecht, The Netherlands. Humber, R.A. (1997) Fungi: identification. In: Lacey, L.A. (ed.) Manual of Techniques in Insect Pathology. Academic Press, San Diego, California, pp. 154–185. Hunter-Fujita, F.R., Entwistle, P.F., Evans, H.F. and Crook, N.E. (1998) Insect Viruses and Pest Management. John Wiley and Sons, Chichester, UK. 164 M.S. Goettel and G.D. Inglis

Inglis, G.D. and Sikorowski, P.P. (2005a) Entomopathogens and their impact on insect rearing. In: Schneider, J.C. (ed.) Principles and Procedures for Rearing Quality Insects. Mississippi State University, Massachusetts (in press). Inglis, G.D. and Sikorowski, P.P. (2005b) Microbial contaminants and their impact on insect rearing. In: Schneider, J.C. (ed.) Principles and Procedures for Rearing Quality Insects. Mississippi State University, Massachusetts (in press). Innis, M.A., Gelfand, D.H., Sninski, J.J. and White, T.J. (1990) PCR Protocols: A Guide to Methods and Applications. Academic Press, San Diego, California. Kaya, H.K. and Stock, S.P. (1997) Techniques in insect nematology. In: Lacey, L.A. (ed.) Manual of Techniques in Insect Pathology. Academic Press, San Diego, California, pp. 281–324. Keeling, P.J. and Fast, N.M. (2002) Microsporidia: biology and evolution of highly reduced intracellu- lar parasites. Annual Review of Microbiology 56, 93–116. Klein, M.G. (1997) Bacteria of soil-inhabiting insects. In: Lacey, L.A. (ed.) Manual of Techniques in Insect Pathology. Academic Press, San Diego, California, pp. 101–116. Kyei-Poku, G., Benkel, B., Goettel, M.S. and Floate, K. (2003) Elimination of Wolbachia from Urolepis rufipes (Ashmead) (Hymenoptera: Pteromalidae) with heat and antibiotic treatments: implications for host reproduction. Biocontrol Science and Technology 13, 341–354. Lacey, L.A. (1997) Manual of Techniques in Insect Pathology. Academic Press, San Diego, California. Lacey, L.A. and Brooks, W.M. (1997) Initial handling and diagnosis of diseased insects. In: Lacey, L.A. (ed.) Manual of Techniques in Insect Pathology. Academic Press, San Diego, California, pp. 1–15. Laird, M., Lacey, L.A. and Davidson, E.W. (1990) Safety of Microbial Insecticides. CRC Press, Boca Raton, Florida. Lighthart, B., Sewall, D. and Thomas, D.R. (1988) Effect of several stress factors on the susceptibility of the predatory mite, Metaseiulus occidentalis (Acari: Phytoseiidae), to the weak bacterial pathogen Serratia marcescens. Journal of Invertebrate Pathology 52, 33–42. Lockwood, J.A. (1993) Environmental issues involved in biological control of rangeland grasshoppers (Orthoptera: Acrididae) with exotic agents. Environmental Entomology 22, 503–518. Martignoni, M.E. and Iwai, P.J. (1986) A Catalogue of Viral Disease of Insects, Mites, and Ticks. US Department of Agriculture, Forest Service, Pacific Northwest Research Station, General Technical Report PNW-195. Merriam-Webster Medical Dictionary (2003) http://www.intelihealth.com/IH/ihtIH/WSIHW000/ 9276/9276.html (accessed 26 May 2005). Milks, M.L., Sokolova, Y.Y., Isakova, I.A., Fuxa, J.R., Mitchell, F., Snowden, K.F. and Vinson, S.B. (2004) Comparative effectiveness of light-microscope techniques and PCR in detecting Thelohania solenopsae (Microsporidia) infections in red imported fire ants (Solenopsis invicta). Journal of Eukaryote Microbiology 51, 187–191. Miller, L.K. (1997) The Baculoviruses. Plenum Press, New York. Miller, L.K. and Ball, L.A. (1998) The Insect Viruses. Plenum Press, New York. Muller, A., Bialek, R., Kamper, A., Fatkenheuer, G., Salzberger, B. and Franzen, C. (2001) Detection of Microsporidia in travelers with diarrhea. Journal of Clinical Microbiology 39, 1630–1632. Olsen, L.E. and Hoy, M.A. (2002) Heat curing Metaseiulus occidentalis (Nesbitt) (Acari: Phytoseiidae) of a fitness-reducing microsporidium. Journal of Invertebrate Pathology 79, 173–178. Papierok, B. and Hajek, A.E. (1997) Fungi: Entomophthorales. In: Lacey, L.A. (ed.) Manual of Techniques in Insect Pathology. Academic Press, San Diego, California, pp. 187–212. Persing, D.H. (1996) PCR Protocols for Emerging Infectious Diseases. ASM Press, Washington DC. Pimentel, D. (2002) Biological Invasions. CRC Press, Boca Raton, Florida. Poinar, G.O. Jr. and Thomas, G.M. (1984) Laboratory Guide to Insect Pathogens and Parasites. Plenum Press, New York. Ratcliffe, S.T., Robertson, H.M., Jones, C.J., Bollero, G.A. and Weinzieri, R.A. (2002) Assessment of parasitism of house fly and stable fly (Diptera: Muscidae) pupae by pteromalid (Hymenoptera: Pteromalidae) parasitoids using polymerase chain reaction assay. Journal of Medical Entomology 39, 52–60. Samson, R.A., Evans, H.C. and Latgé, J-.P. (1988) Atlas of Entomopathogenic Fungi. Springer-Verlag, Berlin, Germany. Siegel, J.P. (2000) Bacteria. In: Lacey, L.A. (ed.) Field Manual of Techniques in Invertebrate Pathology. Kluwer Academic Publishers, Dordrecht, The Netherlands, pp. 209–230. Methods for Assessment of Contaminants of Invertebrate BCAs 165

Sikorowski, P.P. and Lawrence, A.M. (1997) Major Diseases of Heliothis virescens and Helicoverpa zea in Mississippi Field and Insectaries. Mississippi Agriculture and Forestry Experiment Station Technical Bulletin 218. Solter, L.F. and Becnel, J.J. (2000) Entomopathogenic Microsporida. In: Lacey, L.A. (ed.) Field Manual of Techniques in Invertebrate Pathology. Kluwer Academic Publishers, Dordrecht, The Netherlands, pp. 231–254. St Leger, R.J. and Joshi, L. (1997) The application of molecular techniques to insect pathology with emphasis on entomopathogenic fungi. In: Lacey, L.A. (ed.) Manual of Techniques in Insect Pathology. Academic Press, San Diego, California, pp. 367–394. Stouthamer, R., Breeuwer, J.A. and Hurst, G.D. (1999) Wolbachia pipientis: microbial manipulator of arthropod reproduction. Annual Review of Microbiology 53, 71–102. Tanada, Y. and Kaya, H.K. (1993) Insect Pathology. Academic Press, London. Thiery, I. and Frachon, E. (1997) Identification, isolation, culture and preservation of entomopatho- genic bacteria. In: Lacey, L.A. (ed.) Manual of Techniques in Insect Pathology. Academic Press, San Diego, California, pp. 55–77. Tompkins, G.J. (1991) Purification of invertebrate viruses. In: Adams, J.R. and Bonami, J.R. (eds) Atlas of Invertebrate Viruses. CRC Press, Boca Raton, Florida, pp. 31–40. Undeen, A.H. and Vávra, J. (1997) Research methods for entomopathogenic protozoa. In: Lacey, L.A. (ed.) Manual of Techniques in Insect Pathology. Academic Press, San Diego, California, pp. 117–151. Valles, S.M., Oi, D.H. and Williams, D.F. (2002) Detection of Thelohania solenopsae (Microsporidia: Thelohaniidae) in Solenopsis invicta (Hymenoptera: Formicidae) by multiplex PCR. Journal of Invertebrate Pathology 81, 196–201. van Lenteren, J.C. (2003) Quality Control and Production of Biological Control Agents. CABI Publishing, Wallingford, UK. Vavra, J. and Chalupsky, J. (1982) Fluorescence staining of microsporidian spores with the brightener ‘Calcofluor White M2R’. Journal of Protozoology 29, 530. Vestergaard, S., Cherry, A., Keller, S. and Goettel, M. (2003) Hyphomycete fungi as microbial control agents. In: Hokkanen, H.M.T. and Hajek, A.E. (eds) Environmental Impacts of Microbial Insecticides. Kluwer Academic Publishers, Dordrecht, The Netherlands, pp. 35–62. Vinson, S.B. (1990) Potential impact of microbial insecticides on beneficial arthropods in the terres- trial environment. In: Laird, M., Lacey, L.A. and Davidson, E.W. (eds) Safety of Microbial Insecticides. CRC Press, Boca Raton, Florida, pp. 43–64. Weiss, L.M. and Vossbrink, C.R. (1999) Molecular biology, molecular phylogeny, and molecular diag- nostic approaches to the microsporidia. In: Wittner, M. and Weiss, L.M. (ed.) The Microsporidia and Microsporidiosis. ASM Press, Washington DC. Zadoks, J.C. (1998) Risk analysis of beneficial micro-organisms – wild types and genetically modi- fied. In: Alabouvette, C., Möllby, R., Steffen, M. and Zadoks, J.C. (eds) Proceedings Microbial Plant Protection Products – Workshop on the Scientific Basis for Risk assessment. KEM, Stockholm, Sweden, pp. 9–38. Zchori-Fein, E., Gottlieb, Y. and Coll, M. (2000) Wolbachia density and host fitness components in Muscidifurax uniraptor (Hymenoptera: Pteromalidae). Journal of Invertebrate Pathology 75, 267–272. Zhou, W., Rousset, F. and O’Neil, S. (1998) Phylogeny and PCR-based classification of Wolbachia strains using wsp gene sequences. Proceedings of the Royal Society of London, Series B 265, 509–515.