Biological Control Potential of Native Entomopathogenic Nematodes (Steinernematidae and Heterorhabditidae) against Mamestra brassicae L. (Lepidoptera: Noctuidae)

: The largest group of cabbage plant pests are the species in the owlet moth family (Lepidoptera: Noctuidae), the most dangerous species of which is the cabbage moth ( Mamestra brassicae L.). In cases of heavy infestation by this insect, the surface of plants may be reduced to 30%, with a main yield loss of 10–15%. The aim of the present study was to assess the susceptibility of M. brassicae larvae to nine native nematode isolates of the species Steinernema feltiae (Filipjev) and Heterorhabditis megidis Poinar, Jackson and Klein under laboratory conditions. The most pathogenic strains were S. feltiae K11, S. feltiae K13, S. feltiae ZAG11, and S. feltiae ZWO21, which resulted in 100% mortality at a temperature of 22 ◦ C and a dosage of 100 infective juveniles (IJs) / larva. The least e ﬀ ective was H. megidis Wispowo, which did not exceed 35% mortality under any experimental condition. For most strains, there were signiﬁcant di ﬀ erences ( p ≤ 0.05) in the mortality for dosages between 25 IJs and 50 IJs, and between 25 IJs and 100 IJs, at a temperature of 22 ◦ C. Statistical analysis of the e ﬀ ect of temperature on mortality showed that only strain H. megidis Wipsowo exhibited signiﬁcant di ﬀ erences ( p ≤ 0.05) when applied at dosages of 50 IJs and 100 IJs.


Introduction
Plants of the mustard family (Brassicaceae) are predated on by many insect species, the largest group of which is the owlet moth (Noctuidae) family. One of the most dangerous species among this family is the cabbage moth Mamestra brassicae L. (Lepidoptera: Noctuidae), which can inflict major economic losses. Mamestra brassicae larvae are polyphagous, and apart from feeding on plants in the mustard family (cabbage, broccoli, cauliflower, Brussels sprout etc.), they also consume more than 70 plant species from 22 families (e.g., lettuce, beets, onion, potatoes, peas, tomatoes, and chrysanthemum) [1]. The first generation of M. brassicae appears in May and June. At this time, females lay eggs on the undersides of leaves, and the resulting larvae feed starting from mid-July. They initially damage external leaves by gnawing holes between veins, and later gnaw into the heads of cabbage, cauliflower, and broccoli. The larvae pass through six stages of development over the course of four to six weeks [2]. The second generation of M. brassicae is then observed in July and August. Their larvae feed from September to October [3].
On average, pests damaged 3% of the cabbage crops in Poland in 2017 and 4.5% in 2016, with the long-term (1993-2017) mean plant losses amounted to 7.2% [2]. In cases of heavy infestation, the surface area of plants may be reduced to 30%, while the main yield losses may reach 10-15%. The economic threshold for M. brassicae infestation on cabbage is four to five larvae per 50 plants. Although environmentally friendly pest control options are available, such as the biological insecticide Dipel WG (with Bacillus thuringiensis var. Kurstaki (54%) as the active substance) that targets early larval stages, pests are largely controlled using chemical treatments [4]. However, as the latter insecticides pose a risk to the environment and human health, there is an urgent need to develop alternative strategies for controlling owlet moths. One emerging option utilizes entomopathogenic nematodes (EPNs) [5][6][7][8]. EPNs are soil-inhabiting parasites that belong to the phyllum Nematoda and the families Steinernematidae and Heterorhabditidae, and have been proven to be very effective in controlling soil and above-ground insect populations [9][10][11][12][13]. EPNs are mutually associated with bacteria of the family Enterobacteriaceae; while EPNs of the family Steinernematidae usually carry bacteria of the genus Xenorhabdus, EPNs of the family Heterorhabditidae typically carry bacteria of genus Photorhabdus [14]. The third juvenile stage of EPNs is referred to as the "infective juvenile" (IJ). Once IJs of either genera have infected an insect host, they release their bacterial symbionts into the host's body and develop into fourth-stage juveniles and adults. The insect hosts die primarily because of septicaemia [15]. When their food source is depleted, the next generation of IJs exits the host insect cadaver, moves into the surrounding environment, and searches for a new host.
In this study, we examined eight strains (K10, K11, K13, ZAG11, ZAG15, ZWO4, ZWO21, and ZWO23) of Steinernema feltiae (Filipjev) and one strain (Wipsowo) of Heterorhabditis megidis Poinar, Jackson and Klein for their lethal activity against the larvae of M. brassicae. We then determined how the infectivity of these strains was impacted by temperature and concentration.

Test Insects
Larvae of M. brassicae in growth stages from L 4 to L 5 were used in laboratory tests. Observations of larval growth were carried out in a phytotron cabin with artificial light with a photoperiod L:D = 16:8, temperature of about 22 • C ± 2 • C, and relative air humidity 65 ± 5% [3,16,17]. A light trap was used to catch adult individuals from June to the end of August. The trap was systematically checked every 2-3 days.
Moths in pairs were released into glass isolators (600 cm 3 ) placed in the phytotron and covered with bolting cloth, to prevent the moths from escaping. The bottoms of the isolators were covered with discs of filter paper. The isolators were equipped with vials filled with moth feed and washed clumps of grass (perennial ryegrass-Lolium perenne L.) or inflorescences (common chickweed-Stellaria media L.) with intact root systems as a substratum for egg laying. The moth feed consisted of a 10% honey solution with drops of fruit essence. The food was replaced every second day. Eggs were checked for every day, and plant regions with eggs were removed with scissors, transferred to separate plastic containers lined with damp discs of filter paper, and then covered with cabbage or lettuce leaves. The larval development was observed daily at the same time each day. Larvae from hatching (stage L 1 ) to stage L 3 were kept in plastic boxes (160 × 120 mm) with ventilated caps, each containing 10 larvae. To provide appropriate sanitary and hygienic conditions, the filter paper was replaced every day. At the beginning of stage L 4 , the containers were half-filled with wet sand. Fresh food was provided to the larvae every day.

Nematodes
Nine strains of EPNs, isolated in 2010and 2011, were used for experiments. These strains included eight strains of S. feltiae (K10, K11, K13, ZAG11, ZAG15, ZWO4, ZWO21, and ZWO23) ( Table 1), collected from soils in southern and central Poland, and one strain of H. megidis (Wipsowo), from northern Poland [18]. EPNs were maintained on the larvae of Galleria mellonella L. (Lepidoptera: Pyralidae) in the laboratory, following the technique from Kaya and Stock [19]. The IJs were collected using White traps [20] and kept in water at 4 • C for later use.
Mamestra brassicae larvae, together with a piece of cabbage leaf, were placed on wet filter paper in a Petri dishes (Ø 9cm). Thirty-five larvae (with 5 individuals per dish) were used for each experimental condition. Nematodes were applied to the dishes at dosages of 25 IJs, 50 IJs, or 100 IJs per larva, and the dishes were placed in Sanyo incubation chambers at temperatures of 17 • C and 22 • C, in the dark. Each EPN strain was used to infect 210 insects: 35 larvae × 3 dosages × 2 temperatures.
The number of IJs was determined by counting the number of IJs in five droplets (5 µL) for a previously prepared nematode suspension in water. The selected concentrations of nematode suspension were obtained by diluting the suspension with tap water or concentrating the suspension through centrifugation [21].
Control larvae, together with a piece of cabbage leaf, were placed onto Petri dishes lined with filter paper moistened with distilled water. Five larvae were used for each dish, with 7 dishes 35 insects for each temperature, giving 35 insects per temperature and a total of 70 individuals.
The mortality rate (percent of parasitized larvae in each analyzed sample) was determined 72 h after infection by dissecting the dead larvae under an Olympus microscope at a 3-5× magnification.

Statistical Analyses
The data were statistically processed with the use of Statistica (version 10; StatSoft, Kraków, Poland). A Pearson chi-square (χ 2 ) test was used to determine the relationship between two variables i.e., between temperature and the mortality induced by particular nematode isolates in two categories (17 • C and 22 • C, versus mortality expressed as "yes" or "no"). A chi-square goodness of fit test was used to determine the effect of the initial IJ dosage (25 IJs, 50 IJs, or 100 IJs) on the mortality.

Results and Discussion
Mamestra brassicae larvae in the control condition exhibited no mortality during the course of the experiment. In contrast, seventy-two hours after application, EPN strains S. feltiae K11, S. feltiae K13, S. feltiae ZAG11, and S. feltiae ZWO21, resulted in 100% larval mortality at 22 • C and a dosage of 100 IJs. These strains also resulted in high mortalities of between 88.6% and 100% at the lower dosage of 50 IJs. Equally high mortalities of 94.3%, 94.3%, 97.1%, and 97.1% were observed when the samples were incubated at 17 • C at a dosage of 100 IJs for strains S. feltiae K11, S. feltiae ZAG11, S. feltiae K13, and S. feltiae ZWO21, respectively. The same strains also produced high mortalities of 85.7% to 91.4% at a dosage of 50 IJs.
Only two strains, S. feltiae K13 and S. feltiae ZWO21, exhibited a mortality that exceeded 50% at 22 • C at the lowest dosage of 25 IJs. For the remaining strains, at either 17 • C or 22 • C, the mortality varied between 0-48.6%. The least effective strains were H. megidis Wispowo, whose mortality did not exceed 35% in any of the experimental conditions, and S. feltiae ZWO4, which did not exceed 52% mortality ( Table 2). Statistical analysis of the effect of temperature on mortality showed that only H. megidis Wipsowo, applied at dosages of 50 and 100 IJs, was statistically significant (p ≤ 0.05) ( Table 3). The effect of nematode dosage on mortality was significantly (p ≤ 0.05) different for dosages of 25 IJs versus 50 IJs, and 25 IJs versus 100 IJs, at 22 • C for most of the strains. At 17 • C, significant differences were found in the mortality between dosages of 25 IJs and 100 IJs (for all studied strains), and between 25 IJs and 50 IJs (only for S. feltiae K10, S. feltiae K11, and S. feltiae ZWO21). Except for one strain, S. feltiae ZWO23, no significant differences in the mortality were noted between dosages of 50 and 100 IJs (Table 4).   Many studies have sought to optimize the application of EPNs for use as a potential insecticide. These studies have focused on selecting the appropriate initial dosage of nematodes, the temperature, and most importantly, the strain that induces the highest mortality in the target insect. Indeed, our previous studies showed that the susceptibility of different insects to native isolates of EPNs is highly dependent on the type of isolate used [22,23].
In this study on the susceptibility of cabbage moth larvae to EPNs, strains S. feltiae K11, S. feltiae K13, S. feltiae ZAG11, and S. feltiae ZWO21 were found to be the most pathogenic. Similarly, in our earlier studies on the susceptibility of Agrotis exclamationis L. (Lepidoptera: Noctuidae) larvae to EPNs, the most pathogenic nematode strains were S. feltiae K13 and S. feltiae ZWO21. Isolate S. feltiae ZAG11 was more efficient in controlling the larvae of M. brassicae, with a mortality following infection of 85.7-100%, than the larvae of A. exclamationis, with a mortality of 81% [22]. In a previous study, S. feltiae K11 was highly effective (89.7-100% mortality) against Pieris spp. (Lepidoptera: Pieridae) larvae [23]. Strains S. feltiae K11 and S. feltiae K13 were highly pathogenic (96.7-100%) to the larvae of Dendrolimus pini L. (Lepidoptera: Lasiocampidae) [24]. Of note, the pathogenic properties of S. feltiae ZWO4 differ depending on the infected host insect species. For M. brassicae, this strain did not induce greater than 52% mortality, and was the weakest among the strains tested in this study. In contrast, this strain induced 100% mortality in D. pini larvae [24]. A remarkable discrepancy in mortality was also demonstrated for S. feltiae ZAG15, which caused 100% mortality in the case of Pieris spp. [23], but less than 72% mortality when used against cabbage moth larvae in this study. Our studies clearly indicate that, while some strains are always effective, the pathogenicity of others is more dependent on the host.
Although M. brassicae is one of the most harmful pests for field crops, there have been no studies to date examining the potential use of native strains of EPNs, or of commercial preparations based on EPNs, to control this insect. As there is scarce data on the susceptibility of larvae of the genus Mamestra to EPNs, our results were also referred to those pertaining to other foliage feeders.
The aboveground applications of EPNs may be limited due to the harmful effects of ultraviolet radiation or desiccation on the nematodes [12]. However, optimized EPN formulations that protect against harmful environmental conditions may improve the efficacy of aboveground applications [25]. For example, the addition of anti-desiccants or other adjuvants has been reported to provide improved aboveground control of various foliar pests, including the diamond back moth Plutella xylostella (L.) (Lepidoptera: Plutellidae) [26,27], the sweet potato whitefly Bemisia tabaci (Gennadius) (Hemiptera: Aleyrodidae) [28], Spodoptera littoralis (Boisduval), and Helicoverpa armigera (Hübner) (Lepidoptera: Noctuidae) [29].
For isolates of S. feltiae, temperature did not significantly affect the mortality of the cabbage moth; in contrast, there was a temperature effect noted for H. megidis Wipsowo. This may be due to the fact that S. feltiae is adapted to cooler climates but also present in warmer areas [30]. Nevertheless, some studies [11,31,32] found that S. feltiae isolates exhibited the strongest effect against the cereal leaf beetle Oulema melanopus (L.) (Coleoptera: Chrysomelidae) and the Colorado potato beetle, Leptinotarsa decemlineata (Say) (Coleoptera: Chrysomelidae) at temperatures between 20 and 25 • C. Similarly, Bélair et al. [33] showed that temperature might significantly affect the susceptibility of Pieris rapae L. larvae. Indeed, P. rapae showed a markedly higher mortality rate at 25 • C after infection with S. feltiae UK (95.8%) or Steinernema riobrave Cabanillas, Poinar and Raulston (89.2%), than after infection with Steinernema carpocapsae Weiser (65.8%). In contrast, the mortality rates after infection with S. feltiae were 70.8% at 20 • C and only 19.2% at 15 • C. The efficiency of nematode species of the genus Heterorhabditis exhibit greater variability in response to temperature, initial IJ dosage, and host species [34].
We observed no differences in mortality between dosages of 50 IJs and 100 IJs, while marked differences were found between dosages 25 IJs and 50 IJs. Laznik [11] showed that higher nematode dosages caused higher mortality, but Steinernematidae suggested that satisfactory mortality rates were achieved even at lower dosages, which is important from an economics point of view. This conclusion was confirmed in another study [31], which did not reveal the influence of dosage on the efficacy of EPNs against the Colorado potato beetle. A similar trend was described by Andaló et al. [35], who found greater differences in the mortality between dosages of 100 IJs and 250 IJs than between 250 IJs and 500 IJs when studying the susceptibility of armyworm Spodoptera frugiperda (Smith) (Lepidoptera: Noctuidae) larvae to various EPN isolates. Furthermore, as shown by Lewis et al. [36], an excessively high density of EPNs in the host's body can increase competition, resulting in the first IJs that penetrate the host sending chemical cues to discourage further IJ invasion.
Our study indicated markedly higher mortality rates induced by isolates of Steinernema than by those of Heterorhabditis. Similarly, Gulcu et al. [37] in their study of the susceptibility of Spodoptera cilium Guenée (Lepidoptera: Noctuidae) to native EPN isolates, showed higher pathogenicity of Steinernema (77%) than of Heterorhabditis (29%) isolates. On the other hand, Salvadori et al. [38] found that the mortality of S. frugiperda, caused by four strains of Steinernematidae, ranged from 28 to 70%, while the respective efficacy of four strains of Heterorhabditidae varied between 55 and 77%. Yuksel et al. [39], in their study on the susceptibility of Peridroma saucia (Hubner) (Lepidoptera: Noctuidae) larvae to four native isolates of Heterorhabditis and Steinernema, found that the former were more effective in controlling the pest. The higher efficacy of Heterorhabditis noted by the authors cited above may be explained by the higher temperature of the experimental conditions (25-30 • C), which is favorable for the development of Heterorhabditis. Gözel and Güneş [40] confirmed this temperature effect and showed that the mortality of corn stalk borer Sesamia cretica Led. (Lepidoptera: Noctuidae) larvae treated with H. bacteriophora Poinar at 15 • C was 14%, but reached 94% at 30 • C. No differences were noted in the mortality of S. frugiperda larvae treated with either Heterorhabditis or Steinernema, although differences were observed between isolates within the same nematode species.
Current experiments were carried out at lower temperatures (17 • C and 22 • C), because most measures to control larvae of the genus Mamestra in Poland should be performed in spring and autumn. For biological pest control, it is important to select the most appropriate species or specific EPN isolate, and therefore the parasitic properties of these organisms must be first examined under laboratory conditions. These studies provide valuable information with on which species/strains of EPNs are most effective against a given insect, as the selection of the most pathogenic strains will translate into the successful biological control of harmful crop pests. This study sets the basis for future studies using EPNs for cabbage moth control.