Commercial products based on Bacillus thuringiensis recommended for controlling the population of Plutella xylostella in different brassica crops.
The diamondback moth, Plutella xylostella (L.) (Lepidoptera: Plutellidae), is one of the most serious pests of cultivated Brassicaceae worldwide [1,2]. This crucifer specialist may have its origin in Europe , South Africa , or East Asia , but is now present worldwide wherever its host plants exist .
In the first instar, the larvae enter into the leaf parenchyma and feed between the upper and lower surfaces of leaves creating mines. In the second instar, the larvae leave the mines, and from the second to the third instar, they feed on the leaves, destroying the leaf tissue except for the upper epidermis, leaving transparent “windows” in the leaves. Fourth-instar larvae feed on both sides of the leaves . This insect has a short life cycle, around 18 days, and its population may increase up to 60-fold from one generation to the next . Studies indicate that the moths can remain in continuous flight for several days while covering distances up to 1000 km per day, but how the moths survive at such low temperatures and high altitude is not known . In eastern Canada, annual populations of diamondback moths originate from adult migrants from the United States .
P. xylostella was the first crop insect reported to be resistant to dichloro-diphenyl-trichloroethane (DDT), only 3 years after the start of its use , and subsequently it has shown significant resistance to almost every insecticide applied in the field, including new chemical compounds [11,12]. In addition, diamondback moth has the distinction of being the first insects to develop resistance in the field to the bacterial insecticide Bacillus thuringiensis [13,14]. The resistance of P. xylostella populations to B. thuringiensis has been observed by [15-23] in the USA (Florida, Hawaii, and New York), Central America (Mexico, Costa Rica, Guatemala, Honduras, and Nicaragua), and Asia (Japan, China, Malaysia, and the Philippines). In Brazil,  it was documented this pest’s resistance in environments where B. thuringiensis is commonly used as a bioinsecticide.
This has prompted increased efforts worldwide to develop IPM programs for P. xylostella, based principally on new management tactics that are not yet used in the field for this pest [8,25,26]. In this chapter, we give an overview of the association of P. xylostella with its host plants and natural enemies, and describe management strategies and practices for control of the diamondback moth.
2. Tactics for integrated pest management
2.1. Biological control
Biological control can be defined as the use of one type of organism to reduce the population density of another. Biological control has been used for approximately two millennia, and has been widely used in pest management since the end of the nineteenth century . The following types of biological control can be distinguished: natural, conservative, inoculative (or classical), and augmentative. Natural biological control involves the reduction of pest organisms by their natural enemies and has been occurring since the evolution of the first terrestrial ecosystems, 500 million of years ago . It takes place in all of the world’s ecosystems without any human intervention, and, in economic terms, is the greatest contribution of biological control to agriculture . Conservation biological control consists of human actions that protect and stimulate the performance of naturally occurring enemies . In inoculative biological control, natural enemies are collected in an exploration area (usually the area of origin of the pest) and then released in new areas where the pest was accidentally introduced. In augmentative biological natural control, natural enemies are mass-reared in biofactories for release in large numbers to obtain immediate pest control .
2.1.1. Entomophagous agents: parasitoids and predators
Parasitoids can be defined as insects that are only parasitic in their immature stages, kill their host in the process of development, and have free-living adults that do not move their hosts to nests or hideouts .
All stages of the diamondback moth are attacked by numerous parasitoids and predators, with parasitoids being the more widely studied. Over 90 parasitoid species attack the diamondback moth . Egg parasitoids belonging to the polyphagous genera Trichogramma and Trichogrammatoidea contribute little to natural control and require frequent mass releases. Larval parasitoids are the most predominant and effective. Many of the effective larval parasitoids belong to two major genera, Diadegma and Cotesia; a few Diadromus spp., most of which are pupal parasitoids, also exercise significant control . The majority of these species come from Europe where the diamondback moth is believed to have originated . In countries near Brazil, such as Argentina, P. xylostella larval parasitoids collected in the field include the species Diadegma insulare (Cresson) (Hymenoptera: Ichneumonidae), Oomyzus sokolowskii (Kurdjumov) (Hymenoptera: Eulophidae), and C. plutellae (Kurdjumov) (Hymenoptera: Braconidae) .
Seven species of parasitoids were observed in a P. xylostella population on cabbage crops in the Brasilia region of Brazil, with the two most common species being Diadegma liontiniae (Brethes) (Hymenoptera: Ichneumonidae) and Apanteles piceotrichosus (Blanchard) (Hymenoptera: Braconidae). Cotesia plutellae (Kurdjumov) (Hymenoptera: Braconidae) and Actia sp., previously more abundant, had become very minor parasitoids. Six species of hyperparasitoids emerged from D. liontiniae and A. piceotrichosus, showing a high diversity of natural enemies in this region of recent colonization by P. xylostella .
In organically farmed kale in Pernambuco, Brazil, seven natural enemies of P. xylostella were observed: three parasitoids, C. plutellae Kurdjumov (Hymenoptera: Braconidae), Conura pseudofulvovariegata (Becker) (Hymenoptera: Chalcididae) and Tetrastichus howardi (Olliff) (Hymenoptera: Eulophidae), and four predators, Cheiracanthium inclusum (Hentz) (Araneae: Miturgidae), Pheidole sp. Westwood (Hymenoptera: Formicidae), and nymphs and adults of Podisus nigrispinus (Dallas) (Hemiptera: Pentatomidae) .
Several studies have been conducted in Brazil to examine whether these entomophagous agents of the diamondback moth could be used as a biological control for this pest in crucifer crops.
Parasitoids of the genus Trichogramma are among the entomophagous agents that have already been studied for P. xylostella. The species T. pretiosum Riley (Hymenoptera: Trichogrammatidae), Tp8 strain, can parasitize approximately 15 P. xylostella eggs in the first or second generation when reared in this host under laboratory conditions, with 100% emergence, and 10 to 11 days for adult emergence . Eggs of two P. xylostella populations, one reared on kale leaves and the other on broccoli leaves, were exposed to the T. pretiosum Tp8 strain, and the number of parasitized eggs was 5.8–9.4 on kale and 3.2–8.4 on broccoli . Furthermore, the optimal way to mass rear this parasitoid in the laboratory is to use eggs glued to blue, green, or white colored cards .
The impact on non-target species, particularly Trichogramma, of insecticides for P. xylostella control should be analyzed because some are toxic to these parasitoids in crucifers. Endosulfan and etofenprox, classified as class-4 toxic products, are extremely toxic to the parasitoids. Triflumuron, classified as a non-toxic product, is selective for these parasitoids in the eggs of P. xylostella . The combination of chemicals or natural insecticidal products from vegetables with certain cultivars of crucifers enables more effective management of the diamondback moth, particularly in the case of the interaction between pyroligneous extract and cabbage. However, the interaction among cultivars and products can be detrimental to the effectiveness of T. pretiosum and T. exiguum, and thus requires a careful evaluation to minimize the impact on these natural enemies . Bioinsecticides based on B. thuringiensis for controlling P. xylostella can influence the parasitoid T. pretiosum in the moth’s eggs. The application of isolates of B. thuringiensis on P. xylostella larvae influenced the parasitism of T. pretiosum in eggs of subsequent pest generations .
Another parasitoid of P. xylostella larvae, which has been studied in Brazil, is O. sokolowskii. The duration of the immature stage of these parasitoids can range from 12.9 to 31.6 days at 28 and 18°C, respectively, and the number of adults emerged per pupa of P. xylostella varies between 7.3 and 12, with a sex ratio of between 0.86 and 0.91 . During a year, the number of generations of O. sokolowskii is always higher than that of P. xylostella, suggesting that O. sokolowskii could develop up to 24 generations per year while the diamondback moth could reach 20 annual generations . Furthermore, the O. sokolowskii parasitoid is able to disperse and parasitize P. xylostella throughout a kale field up to 24 meters from the release point .
Another larval parasitoid studied in Brazil for P. xylostella is A. piceotrichosus, which was collected in the Rio Grande do Sul State. Its immature stage was observed to last 14.6 to 15.5 days and its adult longevity was found to be 12.7 to 13.4 days .
Among the stink bug predators, P. nigrispinus has great potential for use in P. xylostella control. P. nigrispinus has been reported preying on P. xylostella in crucifer crops , and, furthermore, this predator consumed on average 10.9 larvae or 5.5 pupae in 24 h . Adults of Orius insidiosus (Say) (Hemiptera: Anthocoridae) has been reported consuming 5.9 diamondback moth eggs in 24 h .
2.1.2. Entomopathogens: Bacteria
The occurrence of P. xylostella populations of resistant to certain active ingredients, like synthetic and biological insecticides, has caused a considerable increase in research directed at developing tactics for Integrated Pest Control based on economic, social, and ecological parameters [21,45-47).
Recent studies on control strategies and population reduction of P. xylostella using microorganisms has been increasingly cited in the scientific community, with emphasis on the entomopathogenic bacterium B. thuringiensis Berliner (1911) [48-51,39].
This entomopathogen can be easily found in different environments [52,53], and it is characterized by a variety of strains, each forming one or more protein crystals (Cry) and cytolytic toxins  that have insecticidal activity and determined its efficiency as a control on certain agricultural pests. Another type of insecticidal protein that can be synthesized by some strains of B. thuringiensis is “Vegetative Insecticidal Proteins” (Vip), whose insecticide action spectrum operates in different insect species .
A long history of intensive research has established that their toxic effect is due primarily to their ability to form pores in the plasma membrane of the midgut epithelial cells of susceptible insects [56,57]. The presently available information still supports the notion that B. thuringiensis Cry toxins act by forming pores, but most events leading to their formation, following binding of the activated toxins to their receptors, remain relatively poorly understood .
Strains of B. thuringiensis can produce from one to five toxins that represent a large variability in toxicity and interfere in the expression levels and the spectrum control of insects, and differ in their specificity to certain species . For example, the Cry proteins are show high toxicity to insects of the orders Lepidoptera, Coleoptera, Hymenoptera, Diptera, Orthoptera, and Mallophaga, and to other organisms such as nematodes and mites [60,54,61].
Among the different protein crystals identified in insect control, 59 toxins were tested against 71 Lepidoptera species . The broadest range of toxins was tested against P. xylostella (43 toxin types), which was one of only 12 species that were tested against 15 toxins or more .
In Brazil, P. xylostella is controlled using entomopathogenic bacteria in phytosanitary applications of formulation products properly registered for a particular crop, most commonly biological products containing B. thuringiensis var. kurstaki, which expresses Cry1Aa, Cry1Ab, and Cry1Ac toxins  (Table 1).
However, the low variability in the number of toxins related to formulated biological products, combined with a high number of applications in the field, puts selection pressure on the population of P. xylostella and, consequently, expression of resistance of this pest to protein crystals has been observed since the 1990s [20,24].
According to , mixed formulations of different bacteria or isolates of B. thuringiensis that have a wide variety of Cry toxins, organized in isolation or together, have the ability to reduce selection pressure and, consequently, the development of new cases of resistance in populations of P. xylostella.
To improve the biological control of P. xylostella using this entomopathogenic bacterium, several studies have initially focused to on the characterization of new strains of B. thuringiensis, with the objective of discovering more efficient insecticides and implementing them in new formulations [65,66,51]. In a study conducted by  using stored grains and different strains of B. thuringiensis from soils of several regions of Brazil, there was high mortality (98–100%) of second-instar larvae of P. xylostella. These results have demonstrated that a high variability of Cry genes in the same strain can constitute a substantial tool for resistance management of this pest, with subsequent use in the synthesis of new biological products.
In pathogenicity tests, the strains behave in different ways, and few of them are able to cause total mortality in the insects analyzed. In research conducted by , approximately 19% of the strains tested caused total mortality to second-instar larvae of P. xylostella between 24 and 48 hours.
In this case, in addition to pathogenicity and virulence tests, researchers should analyze the sublethal effects of these strains on the remaining individuals, an important parameter in the toxicological evaluation of B. thuringiensis strains [67,68].
Many biological characteristics of P. xylostella may be influenced by the sublethal effects of these toxins, causing discernible changes in insect behavior, such as appetite loss, decreased movement with subsequent paralysis, change in the tegument color from bright green to dark yellow or dark brown, and loss of reaction to touch [69,51].
According to  and , the most pronounced biological changes observed between phytosanitary applications with strains and commercial products based on B. thuringiensis were in the viability of larvae and pupae and the weight of pupae. The biological characteristics less influenced by these strains were related to the caterpillar and pupal period and sex ratio .
The behavior of strains or commercial products based on B. thuringiensis that result in individuals surviving phytosanitary application, but that provide sublethal effects in subsequent generations, may be a significant tool for Integrated Pest Management , the objective of which is to improve management of the pest through interactions with other control methods, such as biological control with predators and parasitoids, which will reduce the population density due to sublethal effects caused by strains of B. thuringiensis. The remaining pests may be a food source and host for other insects considered beneficial to agriculture, and can help maintain and assist the populations of these arthropods in different crops.
The Integrated Management of P. xylostella based on biological control with the entomopathogenic bacterium B. thuringiensis is an important method for reducing the population density of this pest in brassica crops. However, the use of this control must be well planned, because there are populations of this pest resistant to biological products, necessitating the use of certain methods of resistance management to eliminate these harmful individuals and, perhaps, prevent future problems with the development of resistant populations that can undermine the whole program of rational control of this pest.
2.1.3. Entomopathogens: Fungi
There is no fungus-based bioinsecticide registered for crucifer crops in Brazil; however, some entomopathogenic fungi have been studied to determine their potential as a biological control agent for P. xylostella. Among the fungi that have been studied for their activity against P. xylostella, Paecilomyces tenuipes caused the highest mortality to third-instar P. xylostella larvae, with an LC50 of 1.09 × 106 spores/mL at 25°C .
The most active crude protein extract, isolated from the CNZH strain of Isaria fumosorosea, produced 83.3% mortality in third-instar larvae 6 days post treatment . Furthermore, it has been found that a synergism exists between the fungus I. fumosorosea and a plant secondary chemical, and that larval deaths were directly related to the concentration of each component in the mixtures and their cumulative effect was evident for an extended period .
In addition to the species already mentioned, the fungi Metarhizium anisopliae and Beauveria bassiana were also studied for the control of P. xylostella. Several strains from Benin were tested and found to cause 94% larval mortality . Strains obtained in Brazil have also been tested and caused mortality to P. xylostella larvae ranging from 70% to 96% .
2.1.4. Entomopathogens: Nematodes
Research on the control of Lepidoptera with entomopathogenic nematodes has focused on the diamondback moth . Field studies on cabbage in Java (Indonesia) confirmed that Steinernema carpocapsae can be used as a substitute for ineffective chemical insecticides . Diamondback moth eggs are deposited and the emerging larvae feed on the underside of the leaves. The control of young caterpillars with entomopathogenic nematodes can therefore be optimized by directing the nematode spray to the lower side of the leaves . The use of a surfactant for lowering the surface tension and of a polymer for increasing the viscosity significantly improved nematode performance against P. xylostella . The performance of these adjuvants is, however, influenced by the spray application technique .
2.2. Chemical control
The chemical control method, recommended as one of the tools or tactics of Integrated Pest Management, is still the main strategy for reducing pest populations among crucifer producers. This preference is due to the practicality, speed, and efficiency of controlling insects considered agricultural pests, particularly P. xylostella .
The chemical groups used to control this pest have great variability in terms of the active ingredient, formulation, and toxicological and environmental classes (Table 2).
Among the pesticides recommended for different brassicas, the chemical group of pyrethroids represents one of the most important for P. xylostella control. Chemical control of P. xylostella using a synthetic pyrethroid is recommended when larval density exceeds an economic threshold, which varies in relation to the growth stage of the crop and environmental conditions [79,80]. However, the inappropriate use of these chemical products has considerably increased the frequency of resistance in different diamondback moth populations to some types of active ingredients of this chemical group [81,82,24,83]. According to  and , P. xylostella populations are considered very prone to developing resistance to some active ingredients. In addition to lowering the pesticide efficiency, increasing the frequency of application may not lead to a significant reduction in crop damage.
This may be due to the biological characteristics of this species, the life cycle of which is short when compared to that of other insects, and to the cultural practice of constantly applying pesticides with the same active ingredients in more concentrated doses, without providing a chemical molecule rotation or an appropriate dosage as listed on the label of the phytosanitary product used .
In the context of Integrated Pest Management, cultural, physical, plant resistance, biological, and chemical control methods may be important strategies in the success of the P. xylostella control program . Techniques such as crop residue removal, management of the interval between crops, use of tolerant cultivars, use of sprinkler irrigation, application of plant and biological products and reduction in the number of pesticide applications by measuring the economic injury level, used harmoniously and consciously, can provide significant improvement in the quality of products and the system in which the culture is embedded [86-90,83].
After a rational application of chemical controls, the first response observed in the field is the high larval mortality of P. xylostella in direct proportion to the commercial product concentration recommended for the determined culture [91,83]. Another response to phytosanitary application is a significant alteration in the life cycle of the insect, principally the larval period, because many chemical compounds present in insecticides affect the process of ecdysis, interfering with the transition between instars, and thereby act as a growth regulator .
Among the types of insecticides recommended for the control of P. xylostella, growth regulators have been found to have low interference with the activity of predators, parasitoids, and entomopathogenic fungi, because they do not affect the embryogenesis and reproduction of this pest, which is important since the parasitoid larvae live inside the pest’s eggs before emerging as adults [85,90,38]. This is important principally because the physiological selectivity of this chemical group makes them more toxic to the pest than to the biological control agent [92,93,38,94,26].
Insecticides of plant origin are also a very important group for the population management of this pest. Among these, neem extract (Azadirarachta indica) has shown significant results in the control of P. xylostella, affecting the growth, larval mortality during ecdysis, oviposition, deformation in pupae and adults, and the physiological processes of reproduction, such as inadequate egg maturation and infertility, that interfere with larval hatching [95,90,83,38,96].
In this context, managing the population of P. xylostella using chemical control methods can be a very interesting strategy if well used, because of the large number of chemical groups with different active ingredients, which enables a chemical molecule rotation and prevents the development of resistance. These products can be used with other control techniques to reduce the number of applications of pesticide and improve the quality of the final product. Another very important consideration in choosing the chemical product is its selectivity, because many chemicals have high selectivity for the host but not for biological control agents, which contributes to the maintenance of populations considered beneficial to the integrated management of P. xylostella.
2.3. Plant resistance
The crop forms a template for various interactions between pests and their environment, and varietal resistance to pests is a key component for stabilizing an IPM system .
Plants have a bewildering array of responses to herbivory, broadly categorized as direct and indirect defenses and tolerance . Some primary wax components, including specific long-chain alkyl components, have allelochemical activity that influences the host acceptance behavior of P. xylostella larvae . Furthermore, glucosinolates, a category of secondary products, are found primarily in species of the Brassicaceae. When tissue is damaged, for example by herbivory, glucosinolates are degraded in a reaction catalyzed by thioglucosidases, called myrosinases, which are also present in these species. This causes the release of toxic compounds such as nitriles, isothiocyanates, epithionitriles and thiocyanates. The glucosinolate-myrosinase system is generally believed to be part of the plant’s defense against insects, and possibly also against pathogens .
Among various cultivars of crucifers observed, the cabbage 'chato de quintal' showed a high level of the substance glucobrassicin, and was classified as moderately resistant to P. xylostella .
Several studies have been conducted in Brazil to determine the crucifer cultivars resistant to P. xylostella for use in the management of this pest. Among the crucifers that are marketed in Brazil—cabbage cultivars, broccoli, kale, and cauliflower—cabbage cultivars were more resistant, and kale cultivars were more susceptible to diamondback moth . When compared only cultivars of kale, it was found that ‘Ribeirão Pires I-2620’ was the most susceptible to two generations of diamondback moth .
The use of silicon in the integrated management of diamondback moths may help to reduce the use of pesticides. Silicon damages the jaws of larvae, limiting ingestion and causing high mortality .
2.4. Cultural control
The current pest management tactics pursued by growers focus on the protection of crucifer seedlings, using both cultural and chemical means, in some seasons in the established crops . Because of the failure of insecticides to control the diamondback moth, interest is growing in the use of cultural controls in commercial crucifer production. Some of the classical control measures that have been tried with some success are intercropping, use of sprinkler irrigation, trap cropping, crop cover rotation, and clean cultivation .
The mortality of P. xylostella was significantly higher with the intercropping of Chinese cabbage (Brassica chinensis) with garlic (Allium sativum) and lettuce (Lactuca sativa) than in monocultures of Chinese cabbage. These results suggest that intercropping can suppress the diamondback moth populations for a long period rather than just the short term . Furthermore, studies conducted in Brazil of the intercropping of cabbages with other crop plants (cabbage and green onion, cabbage and cilantro, and cabbage, green onion, and cilantro) did not reduce the rate of parasitism of P. xylostella larvae by O. sokolowskii, which makes it promising for diamondback moth biological control; however it did not interfere with cabbage colonization by the diamondback moth .
A study investigating the impact of irrigation systems on diamondback moth infestation in cabbage noted that when irrigation water was applied by sprinkler-irrigation, diamondback moth infestations were reduced by 37.5–63.9% compared with a drip-irrigated control .
Glucosinolates are biologically active natural products characteristic of crucifers, and crucifer-specialist insect herbivores, such as P. xylostella, frequently use glucosinolates as oviposition stimuli. Benzylglucosinolate-producing tobacco plants were more attractive for oviposition by female P. xylostella than wild-type tobacco plants. As newly hatched P. xylostella larvae were unable to survive on tobacco, these results represent a proof-of-concept strategy for rendering non-host plants attractive for oviposition by specialist herbivores with the long-term goal of generating efficient dead-end trap crops for agriculturally important pests .
With regard to crop cover for crucifers, a broccoli cover-cropping system (cereal rye) resulted in fewer leaves, smaller plants, and a slightly reduced yield when compared to the other systems. Strip-cropping broccoli with potatoes did not convey any agronomic advantages. Gross margin analysis revealed that on a total system basis, a 2.2% yield improvement or a 7% price premium was required to make the cover crop system perform as well as conventional practice .
Another study looked at the effect of two diversification strategies, one a broccoli/potato (Solanum tuberosum) strip crop comprising 1.65-m (tractor width) replications of two rows of potatoes and two rows of broccoli, and the other a cereal rye (Secale cereale) cover crop, which formed a sacrificial planting that was killed and rolled flat to minimize weed competition and improve the agronomic performance of the subsequent broccoli crop. In this case, it was observed that P. xylostella eggs, and the subsequent larvae and pupae, were less abundant on broccoli with the cover crop, probably due to interference with host location and oviposition processes. The strip crop had no effect on broccoli crop yield .
2.5. Sex pheromones
The potential for using synthetic sex pheromone traps as a simple and practical method of monitoring population densities of insect pests has been investigated in many crop systems. Sex pheromones of P. xylostella have already been synthesized for use in the management of this pest in crucifers . Thus, trap catches can be used to forecast infestations during periods that coincide with high P. xylostella infestations .
Currently, pheromone-baited traps in the Prairie Pest Monitoring Network are used to detect and survey  the arrival of migrating moths. Recent research has shown that capture of male moths in pheromone-baited traps in the Prairie Pest Monitoring Network is correlated with moderate, but not low, densities of the immature stages of the diamondback moth sampled in the same fields . Then exists the potential to develop commercially available pheromone-baited traps as tools that can predict the ephemeral nature of diamondback moth population densities in the prairies and inform producers of key thresholds and timing for control efforts .
When placed on Delta sticky traps, the artificial sex pheromone Bioplutella, marketed in Brazil, efficiently captured males of the diamondback moth and could be used for monitoring this pest .
3. Final remarks
As shown above, the management of pests on crucifers in Brazil has largely been dependent on synthetic pesticides, used prophylactically or in response to P. xylostella occurrence, although cultural practices have also played some role in the control of the diamondback moth . The general lack of understanding of interactions between crucifers and their invertebrate pests and between pests and their natural enemies has resulted in a lack of alternative integrated options for growers. Growers would rely less heavily on the prophylactic and reactive application of broad-spectrum pesticides if they were provided with knowledge and training in identifying natural enemies and using economic thresholds. Furthermore, we again emphasize glucosinolates, their breakdown products, and plant volatile compounds as key components in these processes , which have been considered beneficial in the past and hold great promise for the future of integrated pest management.