The Queensland fruit fly, Bactrocera tryoni (Froggatt) (Diptera: Tephritidae), is the major fruit fly pest for all of eastern Australia with literature on the species dating back more than 115 years [1]. It is a major economic pest as a consequence of its ability to survive in a wide range of climatic conditions, its polyphagous nature and its destructive damage to most cultivated fruits and fruiting vegetables. The native range of B. tryoni is considered to be tropical and subtropical coastal Queensland (Qld) and northern New South Wales (NSW), however, its distribution now extends along much of the eastern seaboard and areas of inland NSW and Victoria [1]. Within NSW, B. tryoni is best suited to the climate of the coastal and northern inland areas. It can, however, thrive in less suitable areas such as the south and south-west of the state during years of favourable rainfall, with distribution shrinking back to irrigated areas during dryer years [2]. The majority of B. tryoni adults are believed to disperse up to 1 km, although larvae are readily transported in vehicles within infested fruit that pose a threat to many quarantined production areas within suitable climatic zones [3] such as those within NSW. Bactrocera tryoni also has the potential to spread internationally because of its tolerance of a wide range of climatic conditions and large host range, as well as its tendency to be dispersed by humans at the larval stage inside infested fruit [3].

The expansion of B. tryoni within Australia began as rainforests were cleared and large areas, including inland irrigation zones, were planted with susceptible fruit crops. The largely unrestricted interstate trade of fruits during the late 1890s also contributed to this pest’s spread. Infested fruits quickly become rotten and inedible causing considerable losses in production, often resulting in complete destruction of fruits rather than only cosmetic damage as is caused by many other insect pests [4]. Currently, B. tryoni is managed using mainly surveillance (trapping), bait spraying, the sterile insect technique (SIT) and chemical control as well as public education [5,6]. However, with diminishing pesticide options for the control of B. tryoni, industries are increasingly looking at other alternatives. The aim of this review is to draw-together available information on the biology, methods for study, and culturing of hymenopteran parasitoids of the Queensland fruit fly, B. tryoni. In order to assess prospects for improving biological control of this serious pest in Australia, the international literature on biological control of other fruit fly species is synthesised.

Natural enemies when applied properly are promising, environmentally-friendly and effective tools for sustainable control of arthropod pests to the extent that biological control of insect pests is one of the most cost effective and environmentally sound methods of pest management [7]. In the past 120 years, more than 200 species of exotic arthropods have been introduced on more than 5,000 occasions into 196 countries for the control of insect pests [8]. Such inoculative or classical biological control offers the advantage that well-chosen agents, compatible with the conditions into which they are released, have the ability to maintain self-perpetuating populations from generation to generation [9], providing good continuity of control. There is always a risk however, that the agent will attack non-target species. Parasitoids are, therefore, often considered a better option than predators, as the former rely on the host for development and are often more host specific [10] so reducing the risk of the agent attacking non-target species. Parasitoids are categorised based on factors such as egg development pattern in the adult female (synovigenic versus proovigenic), whether larvae develop within the host (endoparasitoid) or externally (ectoparasitoid), whether the host continues development (koinobiont parasitoid) and feeding after parasitism or is arrested (idiobiont parasitoid), and whether the adult parasitoid feeds upon the host [11]. Many parasitoid wasps (including Braconidae) have the advantage of being self-dispersing giving wide coverage in areas where other techniques such as spraying cannot be readily applied [12]. In addition, parasitoids are in no way dangerous to human health making them an attractive option for fruit fly control in urban areas such as those found in the Risk Reduction Zone (RRZ) in NSW and Victoria [13].

Another response to the potential risks of exotic biological control agents is the use of conservation biological control. Conservation biological control aims to maximise the impact of existing natural enemies and has proven effective in many crop/pest systems [14]. There has been little research attention devoted to conservation biological control against fruit flies so the major focus of the present review is in other forms of biological control. A general limitation of many forms of biological control, however, is that the natural enemies will not typically provide adequate pest suppression alone (including braconid parasitoids), thus integration with other pest management tools such as the sterile insect technique (SIT) [9,15] or bait sprays [16] is required. The use of an integrated pest management (IPM) approach is especially important in eradicating local outbreaks of B. tryoni in Australia [6,17].

Of the eight opiine braconids that occur in Australia and are known to attack B. tryoni [18], there are four that are of particular interest for use in augmentative release, not just in Australia but worldwide due largely to their ease of rearing, range of target hosts, climatic/environmental tolerance and levels of parasitism achieved. These four opiine braconids attack a range of tephritid pest species in other locations (Table 1), including several species that are considered a biosecurity risk to Australia. Fopius arisanus (Sonan) is an egg-pupal parasitoid [19]; Diachasmimorpha kraussii (Fullaway) and Diachasmimorpha tryoni (Cameron) target late second to early third instar larvae, while D. longicaudata (Ashmead) target third instar larvae [20,21,22] (Figure 1). Of these, only D. kraussii and D. tryoni are native and have been detected from far north Qld. [18] to southern inland NSW [23]. Fopius arisanus, although originally from Malaysia, was introduced from Hawaii and ranges from far north Qld, as far south as Sydney [18]; while D. longicaudata also introduced from Hawaii, has been recorded in far north Qld and Lord Howe island [18].

The first effort to introduce braconids into Australia was in 1902 for the control of the Mediterranean fruit fly (Ceratitis capitata Wiedemann) in Western Australia after unsuccessful searches for native natural enemies of this pest [33]. Between 1932 and 1938, the NSW Department of Primary Industries (DPI) attempted biological control of B. tryoni with introductions of several thousand Tetrastichus giffardianus Silv., and small numbers of Opius humilis Silv. and O. fullawayi Silv. [34]. Later, large numbers (over 205,000 in NSW alone) of Melitobia (Syntomosphyrum) indicum Silv. [34] were released, however, all of these biological control attempts failed with none of the released species establishing [34]. There is no published literature documenting augmentative release of mass-reared parasitoids, whether native or exotic, against B. tryoni in Australia.

Utilising parasitic wasps for the control of fruit flies dates back to the early 1900s [18]. Initial attempts were centred on classical biological control whereby parasitic species associated with the target pest in its place of origin were inoculated into new geographical locations. For example, D. longicaudata, originally from south-east Asia [35], was successfully introduced into Hawaii [36] and subsequently into Australia from Hawaii in 1956–1957 [37]. This species is now widely established in the east of Australia, including Lord Howe Island [18], as well as in Florida (USA), the south of Central America and throughout South America [38,39]. Fopius arisanus, also native to south-east Asia [35], was established in Hawaii and later Australia (1956–1957) [28]. Its introduction into other parts of the world including Israel, Mexico, and South America, however, was less successful, except in Costa Rica where it has achieved patchy establishment [33,39,40].

Other examples of parasitoid species used in classical biological control include the native Australian species D. kraussii and D. tryoni [18]. Diachasmimorpha kraussii has been successfully introduced to Israel and Hawaii [33,41] and while approximately 15 D. tryoni were initially released in Maui, Hawaii, around 4.1 million were later augmentatively released [42] for control of C. capitata [24,43,44]. Shortly after its introduction, D. tryoni became the most abundant parasitoid in Hawaii [45], comprising up to 33% of total parasitoids in Kula, Maui [44]. Mass releases of D. tryoni in Mazapa de Madero Canyon in Mexico between 1987 and 1989 have also been successful, substantially reducing infestations in mangoes and oranges and greatly decreasing populations of Anastrepha ludens (Loew) and Anastrepha oblique (Macquart) [39].

In recent decades, increasing interest in the active integration of parasitoids into integrated pest management programs has replaced classical biological control. This is likely to reflect not only an increased awareness of the risk of non-target impacts but also significant advancements in rearing techniques and artificial diets for rearing hosts for augmentative biological control programs [46,47]. These advances have allowed parasitoid wasps to be used as biological control agents against various fruit fly pests in augmentative release programs [48,49,50]. Augmentative biological control involves the supplemental release of parasitoids and relies on the mass-production of large numbers of parasitoids in a laboratory. Relatively few natural enemies may be released at a critical time of the season (inoculative release) sometimes with the expectation that reproductive populations will establish or literally millions may be released (inundative release). In addition, the crop/host plant system may be modified to support or augment the parasitoids, often known as habitat manipulation or cultural management. Parasitoids have been used successfully in augmentative release programs in Hawaii, Mexico, Guatemala and Israel [33,39,51]. The major species utilised in augmentative research and control programs worldwide are D. kraussii, D. longicaudata, D. tryoni and F. arisanus with the expectation that the released parasitoids will suppress the targeted pest. A particularly successful example of augmentative biological control of fruit flies occurred in Hawaii where parasitoids dramatically reduced fruit fly (mainly Bactrocera dorsalis (Hendel) and C. capitata) densities within one year [52]. Trials of C. capitata control in Hawaii using the native Australian parasitoid, D. tryoni involved releasing 4.2 million parasitoids, averaging 265,000 per week, resulting in significantly lower C. capitata per fruit compared to an untreated control area [50].

Introducing biocontrol agent species to new habitats, especially in such large numbers, may raise host specificity risks [9]. For example, imported parasitoids have been reported to form new host associations [53], sometimes with other introduced species [54]. Since the introduction of D. tryoni to Hawaii for the control of C. capitata, this parasitoid has formed host associations with two gall‑forming tephritids, the lantana gall fly, Eutreta xanthochaeta Aldrich, and the Pamakani gall fly, Procecidocharea utilis Stone, introduced to Hawaii for the control of weeds [54]. These new host associations of D. tryoni are thought to possibly be due to the introduction of F. arisanus, a competitor of D. tryoni. Diachasmimorpha kraussii was also introduced to Hawaii for the control of Bactrocera latifrons (Hendel), the solanum fruit fly. Tests of possible host associations with native and introduced tephritid fruit flies in Hawaii found that gravid D. kraussii females would oviposit into all host larvae presented [29]. These studies were performed under confined laboratory conditions in choice and no‑choice scenarios, and while it is not known whether such host associations occur in the field, the aforementioned findings illustrate the potential implications (e.g., unfavourable new host associations) of introducing a species into a new habitat and the desirability of using native parasitoids over exotics.

A limiting factor of mass-rearing most species of parasitoids (and all that attack fruit flies) is that it requires the use of live hosts, which in turn, increases production costs [55]. The quality (weight and size) of the host must be appropriately monitored as parasitoid cultures from low quality hosts tend to have a male biased sex ratio [56]. Hymenopteran parasitoids are haplodiploid with infertile eggs producing males and fertilised eggs producing females [57]. Rearing facilities aim to optimise production of female parasitoids as females are the sex responsible for exerting biological control of the target species.

The size of host larvae is also a crucial factor for parasitoid emergence. High quality hosts that allow production of comparatively large fruit fly larvae increases parasitoid emergence [56]. The age and condition of the larvae used during the rearing process is an important factor influencing both percentage emergence and sex ratio of mass reared parasitoids [58,59]. Measurements used to monitor the quality of mass reared parasitoids include: host weight, adult emergence, survival, fecundity, flight and searching behaviour [60]. Such monitoring is important because the performance of mass reared insects produced for augmentative biological control can decrease over generations due to their adaptation to laboratory conditions [61]. These changes may decrease their performance in the field and reduce the success of a parasitoid in a particular biological control program [61]. Therefore, various aspects of parasitoid biology need to be considered in order to optimise the rearing of parasitoids in large numbers.

Despite the success of parasitoids as agents in international augmentative tephritid biological control programs, parasitoids have not yet been used for augmentative release in Australia against B. tryoni or C. capitata, the two key economic fruit fly pests in Australia. Mass-rearing facilities for B. tryoni and C. capitata are located in NSW and Western Australia, respectively. These facilities currently produce flies for the sterile insect technique (SIT), and could be a source or expand production to supply host material for parasitoid rearing. As 70% of the cost of parasitoid rearing is associated with the production of host material [94] the existence of host rearing infrastructure and expertise in Australia is a benefit to the economics of parasitoid rearing.

Information on the biology of parasitoids is of great value in developing and improving mass‑rearing techniques and augmentative release programs [95]. Accordingly, to assess the potential utilisation of available parasitoid species for biological control of fruit flies in Australia, the following sections review key aspects of parasitoid biology, together with available information on the methods that are used for the study of fruit flies and their parasitoids, focusing particularly on the four major opiine braconids, D. longicaudata, D. kraussi, D. tryoni, and F. arisanus that are currently being considered for augmentative biological control.

Bactrocera tryoni poses an enormous threat to the sustainability of Australian horticulture. In particular, the Fruit Fly Exclusion Zone (FFEZ) which provides B. tryoni-free areas permitting Australian producers to export to areas which are climatically favourable to B. tryoni and therefore, are susceptible to outbreaks. If B. tryoni cannot be managed effectively in both the FFEZ and RRZ, Australia’s economy may suffer from limited exports, as well as widespread crop damage. A solution to the over reliance on chemical insecticides is to develop a more integrated system to control populations in the RRZ and other endemic areas and stop their incursion into the FFEZ and other major endemic horticultural production areas. Parasitoids offer an attractive means of achieving this.

Of the four parasitoid species currently available in Australia for possible use in an augmentative biological control program against B. tryoni, D. longicaudata and F. arisanus have the longest adult longevity and highest fecundity. However, as they cannot survive below 15 °C, they are probably unsuitable for Australia’s major horticultural production regions within the FFEZ and surrounding RRZ in inland NSW, Australia. These species also parasitise a wide range of fruit fly species and are therefore more likely to constitute a risk to non-target tephritid species, including natives and any introduced for biological control of weeds. The native parasitoids D. kraussii and D. tryoni were the only B. tryoni parasitoids detected in a survey undertaken in inland NSW [23] and are thus likely to be better suited climatically to this region. Diachasmimorpha tryoni appears to be better adapted to cooler climates than other parasitoids [50] (Table 4) and could therefore be released early in the season before large B. tryoni populations become established, or to deal with localised, early season outbreaks. Mature larvae of D. tryoni enter a winter diapause within host puparia [44] and emerge at the beginning of the season at the same time that B. tryoni emerge. Surveys have shown that D. tryoni is more abundant at higher elevations (greater than 600 metres) where temperatures are generally lower than at sea level [46]. The distribution of D. tryoni is similar to that of B. tryoni and occurs in all commercial fruit crops (except pineapple and strawberries) and many vegetable crops [121].

Importantly for augmentative releases, D. tryoni is known to be amenable to mass-rearing [50]. Additionally, although mass releases of D. tryoni have not been tested against B. tryoni, they have led to the successful control of other fruit flies such as C. capitata. Being suited to mass release in inland NSW is attractive because a widespread augmentative biological control program could decrease B. tryoni pest pressure on the FFEZ and also be used against isolated outbreaks within the FFEZ.

While the available information supports the viability of augmentative releases of D. tryoni against B. tryoni in inland NSW, there are large knowledge gaps concerning mass-rearing techniques, host specificity, and success rates. Although it is native to Australia, D tryoni has never been mass-reared for release in Australia or been field tested under Australian conditions. This information is critical to the success of biological control. Further, techniques that could optimise the mass-rearing process, such as an optimal food source for pre-release feeding and the influence of B. tryoni host size to maximise longevity and fecundity, have not been extensivelt explored for D. tryoni although literature is available for closely related species that indicates appropriate methods to fill these knowledge gaps. There have been non-target effects from the release of D. tryoni on gall forming fruit flies introduced to Hawaii as biological control agents of the weed L. camara [122]. Yet because D. tryoni is native to eastern Australia, where no gall forming insects have been recorded as hosts [110], they are unlikely to have equivalent non-target effects in Australia. Lantana camara is a pest weed in Australia, however, the biological control agents used to control it do not include gall-forming tephritids [123].

Like most parasitoids, D. tryoni is susceptible to insecticides even at exposure rates well under the recommended field application rates used for B. tryoni control [47]. Insecticides known to be harmful to D. tryoni, include carbaryl, permethrin and malathion [46]. The application of insecticides to control B. tryoni or other insect pests could significantly reduce the success of biological control against B. tryoni [47] and this needs to be considered in integrated pest management strategies. Targeted bait spraying is potentially less harmful to natural enemies because widespread application of toxins is avoided, however, baits containing sugars are potentially attractive to adult parasitoids. The combination of GF-120 bait sprays and biological control using D. tryoni has been described as a compatible control option [124]. The field effect of baits such as GF-120 has never been tested with D. tryoni, although mortality is known from direct exposure in laboratory conditions [124]. Thus, further research is required.

A more harmonious combination of control methods for B. tryoni is the release of parasitoids together with SIT [125]. These two techniques complement each other because they act on two different stages of B. tryoni (larvae and mating adult). A synergistic suppressive action can lead to local pest eradication [89]. This occurs as parasitoids tend to have a greater impact on relatively dense host populations because hosts are easy to locate and the parasitoid is able to reproduce efficiently. In contrast, SIT is expensive to use against large, dense pest populations but becomes more cost effective at lower pest densities [126]. Thus, parasitoids are generally released before sterile insects as parasitoids will suppress pest populations and reduce the number of sterile insects needed to achieve acceptable over-flooding ratios [46]. Success with SIT and biological control has been analysed in Mexico with D. longicaudata [56]. The use of both techniques helped to create fly-free zones, which allowed access to new markets valued at US $15 million [56]. There is evidence of this synergistic relationship in studies with F. arisanus and D. kraussii [40] and in the control of other insect pests such as codling moth [10]. Research into IPM strategies involving SIT and D. tryoni (or indeed other parasitoid species), however, have not been conducted in Australia and should be further researched to establish the ability of combining SIT and other biological control methods to eradicate and/or suppress B. tryoni in the FFEZ and RRZ.