Abstract

The diamondback moth (DBM), Plutella xylostella, is a serious world-wide pest of Brassica crops which has developed resistance to all categories of chemical insecticides and to toxins of the bacterium, Bacillus thuringiensis. Potential sources of novel control options for DBM include the use of microbial agents such as entomopathogenic fungi. Infective fungal spores, or conidia, are fragile and short-lived. Therefore, the identification of robust alternative inoculum sources and the development of novel formulation and application techniques can contribute to realising their potential. Under certain environmental conditions, some entomopathogenic fungi, including the common DBM pathogen Zoophthora radicans, produce specialised resting spores which could have potential as an alternative commercial inoculum as they are robust and long-lived. Research is underway to determine the mechanisms of resting spore production and germination to underpin their effective exploitation.

Keywords

Plutella xylostella, entomopathogenic fungi, microbial control, resting spores

Introduction

The diamondback moth (DBM), Plutella xylostella L. (Lepidoptera: Plutellidae), occurs throughout the world on cruciferous plants. In some areas of the world, DBM populations are effectively regulated by natural enemies and do not cause economic damage (Wilding 1986). However, in many regions its high reproductive rate coupled with heavy selection pressure for insecticide resistance make it a very severe pest. Rapid development of resistance to all categories of insecticides including toxins of the biological agent Bacillus thuringiensis (Bt) and increased awareness of the environmental consequences of excessive pesticide use has led to the investigation of novel, non-chemical methods for DBM control (Shelton et al. 1997).

Microbial agents for control of DBM

Microbial control agents with potential against DBM include bacteria (such as Bacillus thuringiensis), viruses and entomopathogenic fungi. Some of these organisms contribute to the natural regulation of DBM populations in the field and their exploitation has focussed on augmentation and conservation strategies to improve their natural efficacy. Others are not commonly found in DBM populations, but are highly pathogenic to DBM and, therefore, if they can be produced, formulated and applied, also have potential as microbial control agents.

Bacteria

Of all the microbial control agents developed, Bacillus thuringiensis has been the most commercially successful (Lacey & Goettel 1995). This is partly because it can be used in a very similar way to conventional chemical insecticides; it is fast acting, can be produced on inexpensive media, applied inundatively by conventional equipment, has a long shelf life and minimal effects on non-target organisms (Schacter 1999, Lacey & Goettel 1995). Potentially this has also contributed to the limitations in its use associated with resistance. Bacillus thuringiensis does not recycle in insect populations and kills through the action of a crystalline protein toxin (Cry proteins) and not via invasion and growth within the insect tissue. As such resistance can, and has, evolved to the toxin in the same way as resistance develops to chemical insecticides when selection pressure is high. Integration of B. thuringiensis with other biologicals/microbials can contribute to resistance management strategies in the same way as integration of conventional insecticides with other control strategies.

Viruses

In the field, DBM can be infected by two types of virus, nuclear polyhedrosis viruses (NPV) and granulosis viruses (GV). Unlike Bt, viruses can recycle in DBM populations developing epizootics (epidemics) and are relatively persistent in the environment. They can be slower to kill than Bt and some have a wide host range and/or are difficult to mass produce (Granados & Williams 1986, Shapiro 1986). Like Bt they can be applied by conventional means as inundative sprays and have been used successfully in this way to control DBM. Unformulated DBM granulosis virus (Plxy GV), applied at weekly intervals at a rate of 3.0 x 10¹³ occlusion bodies/ha, controlled DBM on kale in Kenya more effectively than available chemical insecticides (Grzywacz et al. 2001).

Fungi

There are approximately 750 species of fungi from 56 genera that infect arthropods (Hawkesworth et al. 1995). Insect pathogenic fungi are mostly found in the orders Moniliales (Deuteromycotina: Hyphomycetes syn. Deuteromycetes) and the Entomophthorales (Zygomycotina: Zygomycetes) (Flexner & Belnavis 1998). DBM populations are commonly regulated by two entomophthoralean species, Zoophthora radicans and Erynia blunckii, but are also susceptible to several species of Hyphomycetes which are not usually found in DBM populations. These include Beauveria bassiana, Paecilomyces fumosoroseus and Metarhizium anisopliae (Wilding 1986).

Like viruses, entomopathogenic fungi are ubiquitous and in appropriate hosts are capable of natural recycling. Unlike the other microbials discussed, they cause infection by direct penetration through the host cuticle without the requirement for ingestion (Lacey & Goettel 1995). This is advantageous as it limits the potential for the target to avoid consuming a lethal dose, but it also means that fungi are reliant on appropriate environmental conditions to infect and multiply. Like viruses, speed of kill can be variable and host specificity varies between species and even among isolates of a single species (Pell et al. 1993, Yeo et al. 2001). Hyphomycetes can have broad host ranges in contrast to Entomophthorales which are usually highly host specific (Pell et al. 2001).

The development of fungal entomopathogens as biological control agents has been the subject of considerable research, particularly since the 1970s. However, there are only limited examples of currently available marketed products (Shah & Goettel 1999). Exploitation of fungi, like other microbial agents, has focussed on using them in a similar way to conventional insecticides, i.e. as an inundative spray application or 'mycoinsecticide' with no requirement for secondary cycling. For example, Beauveria bassiana (Mycotrol^®) applied to seedlings grown in a nursery was effective at controlling DBM before they were transplanted into the field (Shelton et al. 1998). In open field trials in the USA, B. bassiana significantly reduced the numbers of DBM larvae when used alone (Vandenberg et al. 1998) and when integrated with Bt could control three lepidopteran pests on brassicas (Vandenberg et al. 1999). This approach reduces the number of applications of Bt and therefore contributes to resistance management.

Entomophthoralean fungi such as Z. radicans are common natural enemies of DBM and contribute to the natural regulation of DBM populations worldwide (examples in Pell et al. 2001). The high natural pathogenicity to DBM, strain specificity, relatively fast speed of action, pre-mortality impacts on DBM biology and potentially positive interactions with insect natural enemies means that Z. radicans has excellent potential as a biological control agent of DBM when exploited in an appropriate integrated strategy (Pell et al. 1993, Furlong et al. 1997, Yeo et al. 2001). They develop epizootics which can eliminate DBM populations at a local level clearly demonstrating its potential as a microbial control agents. However, epizootics are unpredictable and can be too late to prevent crop damage, so augmentation has been attempted. Mass production of mycelial material for augmentation is possible (McCabe & Soper 1985), but is currently limited by production and economic constraints.

An early attempt at inundative release was made by Kelsey (1965) in New Zealand. A spray comprised of macerated and diluted Z. radicans-infected larvae was applied to two DBM-infested Brassica fields. The time taken to give adequate control was not reduced by such sprays when the fungus was already present, but Kelsey suggested that there was merit to introducing the fungus into uninfected caterpillar populations. Kelsey (1965) also observed that only one spray was necessary per season, indicating evidence of natural recycling within the population, thereby suggesting that an inoculative approach could be appropriate.

Inoculative augmentation strategies for Z. radicans include the development of auto-dissemination where DBM behaviour is manipulated using pheromones to encourage transmission and the establishment of population regulating epizootics in DBM populations before they reach damaging levels (Pell et al. 1993, Furlong et al. 1995, Vickers et al. 2001). This approach exploits the ability of Z. radicans to transmit in DBM larval populations even at low host densities (Furlong & Pell 2001).

Underpinning ecological studies on the fungus are essential to identify appropriate strategies (inoculation, inundation and/or conservation) for the effective exploitation of the selected approaches. More imaginative control opportunities are only possible when the positive and negative attributes of the organism are considered and addressed. The potential of fungi as insect control agents is affected by the many biotic and abiotic constraints on the ability of fungi to infect their target hosts (Lacey & Goettel 1995), but all these constraints could ultimately be overcome if the right propagule, formulated in an optimum fashion is introduced into the population at the right time and in the right manner. One challenge to the development of Z. radicans as a microbial control agent is to identify and produce alternative propagules for release.

Future research on Z. radicans in New Zealand

Zoophthora radicans produces two spore types: conidia for dispersal and resting spores (azygospores) for persistence. In addition, conidia may either produce secondary conidia, or capilliconidia (Pell et al. 1993). Conidia, which are the dominant infective propagule in nature, are fragile, short lived and subject to environmental desiccation (Furlong & Pell 1997, Uziel & Shtienberg 1993). The potential of conidia as the basis of a commercial product is therefore limited by their rapid environmental desiccation. In contrast, resting spores, which are the long term survival structure of the fungus, are thick-walled and robust, long-lived, environmentally stable, and have a period of dormancy. Resting spores, therefore have potential as an alternative commercial inoculum for use in augmentation (inoculative and mycoinsecticide) and conservation approaches.

Individual isolates of Z. radicans differ in their ability to form resting spores in infected cadavers; some form resting spores in few or no cadavers, whereas others form resting spores in many, under similar conditions (Glare 1988, Pell et al. 1993, Yeo et al. 2001). Resting spore production rates increase at low temperature and high humidity, high inoculum density, differing host age or physiological condition and when hosts are infected with more than one isolate (Perry et al. 1982, Glare et al. 1989), but the conditions for resting spore production are not fully understood. Perry & Fleming (1989), found resting spores of Z. radicans (= Erynia radicans) germinated after storage for >2 months at 4°C or by natural overwintering.

Research is, therefore, ongoing to determine the mechanisms of production and germination of resting spores by Z. radicans in DBM to facilitate the use of resting spores as an alternative commercial inoculum for use in microbial control of DBM.

Acknowledgements

In New Zealand, project funding and support was received from HortResearch and AgResearch. FJLS would like to thank HortResearch for funding conference registration and attendance. JKP and HY were supported by an Alma Baker Fellowship. JKP also receives funding from the Department of Environment, Food and Rural Affairs, UK. IACR-Rothamsted receives grant aided support from the Biotechnology and Biological Sciences Research Council of the UK.

References