Introduction

The tomato crop (Lycopersicon esculentum) is an important vegetable crop in Egypt; it belongs to Solanaceous. This plant is usually infested in Egypt with many destructive pests, especially the most destructive one which is the green peach aphid, Myzus persicae Sulzer (Hemiptera: Aphididae) Blackman and Eastop (2006). These pests transmit several viruses to many economic crops (Namba and Sylvester 1981; Berry 1998). This species causes great damages and diseases to the leaves and fruits (Filotos et al. 2004). M. persicae is a small green aphid causing a decrease in growth, shriveling of the leaves, and the death of various tissues (Namba and Sylvester 1981; Berry 1998). The green peach aphid is found worldwide, although it is less tolerant of colder climates and over winters through its eggs, which are laid in trees of the genus Prunus (Howe and Jander 2008). Destruxin is the toxin of the fungi (Metarhizium anisopliae). This toxin causes death to many serious pests, which exhibit a variety of insecticidal actions (Roberts 1981).

The toxin called destruxin is a cyclic hexadepsipeptide produced by the fungus Metarhizium anisopliae. Destruxin (DEX) causes paralysis and a speedy death to the harmful pests and also it causes suppression of the insect immune system (Odier et al. 1992). Also, destruxin causes an inhibitory activity on the leukemic cell production, decreasing the number of cells in G2/M phase (Odier et al. 1992). Destruxins (DEX) cyclic hexadepsipeptide mycotoxin which have insecticidal and phytotoxic activity. Sabbour and Shaurub (2018) found that the destruxin treatments and nano destruxin decrease the infestations with S. littoralis under laboratory conditions. Under field conditions, the nano destruxin significantly decreased the infestations number of S. littoralis in the cotton field. Thungrabeab and Tongma (2007) reviewed the research of several authors dealing with the differential susceptibilities of many natural enemies to various fungal species. They concluded that some genera or species of fungi could be specific and might inflict only certain types of hosts. They also mentioned that the fungus Beauveria bassiana was not pathogenic to Coccinella undecimpunctata and Chrysoperla carnea.

The present work aims to evaluate the efficacy of destruxin against M. persicae and their main efficient predator (C. undecimpunctata) in tomato fields in Egypt.

Materials and Methods

Insect cultures

Pests

M. persicae were reared on small potted tomato plants inside cylindrical glass cages (15-cm diameter × 40-cm height), covered with muslin, under controlled conditions (26 ± 2 °C and 65 ± 5% RH).

Predator

The stock culture of the C. undecimpunctata started with adults that` were collected from aphid-infested tomato cultivars in Nobaryia, Egypt. Each of the 5 adults was kept in 2-L glass jars. The jars were supplied with fresh lettuce leaves infested with aphids for feeding. The jars were covered with muslin cloth held in position by rubber bands. Food was renewed every other day. The jars were checked daily for eggs. The eggs were collected and transferred to Petri dishes (20-cm diameter) till hatching. Neonate larvae were transferred individually to plastic cups with ample amount of Ephestia kuehniella Zeller (Lepidoptera: Pyralidae) eggs till reaching the proposed experimental stage ( second larval instar). Unused larvae were left in 2-L glass jars (5 each) with supply with small duranta branches carrying different stages of aphids for feeding till maturation.

Destruxin

Destruxin A-760 and Destruxin A-724 obtained from Shanghai Fuang Agrochemical Co. Ltd (99.9% purity) and prepared according to Guan et al. (2008). Six aqueous concentrations of both destruxins were prepared: 2.000, 1.000, 0.750, 0.500, 0.250, and 0.125 ppm. Fresh tomato leaves were dipped in each concentration for 10 s and left to dry at room temperature. Treated leaves were offered to third nymphal instars (20 nymphs/concentration). A parallel control of non-treated tomato leaves dipped in distilled water was run. Each treatment was replicated five times. The percentage of mortality was recorded after 7 days (Sabbour and Shaurub 2018) of treatment and corrected against that of the control according to the method described by Abbott’s formula (Abbott 1925). Corrected mortality was subjected to probit analysis (Finney 1971) to determine the LC50 value. All experiments were run under the aforementioned laboratory conditions.

Treatment of pests

Fresh tomato leaves were sprayed with the desired Destruxin A-760 and Destruxin A-724 concentration (3 shots as spurts/leaf ) ( Matter et al. 1993), left to dry and placed in 1-L plastic containers (one each). Then, 20 nymphs of M. persicae were placed on each leaf. Five containers (replicates) were used/concentration/microbial pathogen (destruxin)/aphid. Each container was covered with muslin and incubated at 25 °C, thereafter, untreated leaves were introduced in the plastic containers to allow the gentle transfer of survivors to them and the previously treated leaves were discarded. Untreated leaves were placed in plastic containers sprayed with water only and used as the control treatment. The experiment was replicated four times. The percentages of mortality were calculated after 7 days and corrected and treated as above.

Treatments of predator

One-day-old adults, second instar larvae of C. undecimpunctata, and second instar nymphs of M. persicae were used for evaluation of the pathogenicity and efficacy of Destruxin A-760 and Destruxin A-724. This was accomplished by different techniques.

A—Spray technique to evaluate the contact effect

B—Feeding technique to evaluate oral toxicity, either obligatory (no choice) (exposure to treated preys only) or free choice exposure to both infected and uninfected preys, and to investigate whether the predator has the ability to distinguish between the infected, uninfected preys, and not

Spray technique

Groups (20 predators of 1-day-old adults of C. undecimpunctata or 1-day-old second nymphal stage/group) were placed in a Petri dish (19-cm diameter) and sprayed with the fungus at 2.00 ppm concentration level using a small atomizer, by three shots as spurts (Matter et al. 1993). The shots were directed to the insects at a 15-cm distance, then the insects were individually transferred gently, using tweezers, to plastic cups (5-cm diameter and 12 cm) with small water moist filter paper and aphid-infested tomato leaf. The cups were covered with muslin and incubated at 25 ° C. The filter paper and branches carrying aphids were renewed every other day. Five groups (20 individuals/group) from each stage were used for each destruxin.

The cups of each group were checked daily for insects showing signs of fungus infection. The death toll was recorded for two weeks post-treatment and mortality percentages were calculated in each case.

Obligatory and free-choice feeding techniques

Groups of 20 individuals of either adults of C. undecimpunctata or second instar nymphs of M. persicae per group were exposed either obligatory to contaminated diet (pathogen-treated aphids) or selectively to pathogen-treated and untreated aphids for 24 h. In case of free-choice feeding, 5 groups were used/pathogen/predator stage. The predator was kept starved for 4 and 6 h for M. persicae, respectively, then, each group was introduced in the middle of a 5-L glass jar with 2 branches of tafla carrying ample amounts of the pest (one branch was previously sprayed with the destruxins while the other branch was sprayed with water only). The two branches were placed on both sides of the glass jar facing each other to allow the predator individuals free-choice to feed on either treated or untreated aphids. Five glass jars (replicates) were used for each pathogen.

Regarding obligatory feeding, the same number of predators in each of 5 glass jars were used as mentioned above but offered only treated aphids. In both trials, the exposure period was 24 h. Then predators of every treatment and the control as well were transferred individually to plastic cups, offered untreated aphids, and checked daily for 14 days

Field experiments

Pests

Experiments were carried out to study the efficacy of the tested two destruxins against the target insect pests in two different areas that differ in climatic and soil factors: in Ibn Malek (El-Nobaryia region) with dry weather and sandy soil and in El-SharKia (Zagazig) with wet weather and clay soil. Tomato (Var. Bio-Bride) was planted at the first of April in an area of about 1200 m2, divided into 12 plots of 100 m2each. Four plots were assigned for each pathogen, while 4 plots were treated with water and used as controls Destruxin A-760 and Destruxin A-724 were applied at 5 ppm concentration and 5 L/plot. Treatments were performed in a randomized plot design at the sunset with a 5-L sprayer. Three applications were made at 1-week intervals at the commencement of the experiment, then 20 samples of plants were randomly collected every week from each plot and transferred to the laboratory for examination. The average number of each of the tested pests/sample/plot/treatment was calculated 20, 50, 90, and 120 days post first application. The infestation of aphids was then determined in each case.

After harvest, the yield of each treatment was weighed as kg/F. Yield loss was calculated according to the following equation:

$$ \mathrm{Yield}\ \mathrm{loss}=\frac{\mathrm{potential}\ \mathrm{yield}-\mathrm{actual}\ \mathrm{yield}}{\mathrm{Potential}\ \mathrm{yield}}\times 100 $$

Potential yield was that of which gave the best results among the tested pathogens Destruxin A-760 and was taken as a base for comparison with the other treatments.

Predator

Seedlings of tomato plants were sown in rows (ca 50 cm apart) in ca 0.5 F located in the El Sharkia Governorate. One-month-old plants were found highly infested with M. persicae. The cultivated area was divided longitudinally into 3 areas (ca Kirat/each), separated from each other by uncultivated bare and had 4 m/width. One area was used for each entomopathogen toxin tested and the check as well. Each pathogen was sprayed at the rate of 5 ppm/F, using a high pressure hand held gun. The concentration of the Destruxin A-760 and Destruxin A-724 was about 5 ppm. (This concentration was previously achieved more than 80% mortality in both pests in laboratory experiments.) Three applications were made first at 1-week intervals. Then, count of C. undecimpunctata on nymphs and adults of M. presicae were carefully counted on site in all tomato plots visually, hand picking, and finally, sweeping net (25-cm diameter). The counts were made just before the first application and 1, 2, and 3 weeks post last application. The predators were placed again after each count on their previous location in the corresponding plant site. Fifty tomato shrubs (10 from each of 5 raws) per each treated area and the control as well were arbitrarily chosen/each time interval. The average number of predators/50 plants/time interval was calculated in each case. The increase or decrease in the population of the predator/50 plants as compared with the check was calculated according to Hendrson and Tilton’s (1955) equation as follows:

$$ \%\mathrm{increase}\ \mathrm{or}\ \mathrm{decrease}\ \mathrm{in}\ \mathrm{population}\ \mathrm{density}\kern0.5em =\kern0.5em \frac{\mathrm{Ca}}{\mathrm{Cb}}\kern0.5em \times \frac{\mathrm{Tb}}{\mathrm{Ta}}-1\times 100 $$

where Ca = population density in the treated area before treatment, Cb = population density in the treated area after treatment, Ta = population density in treated before treatment, and Tb = population density in the treated area after treatment.

Results

Table 1 shows that the LC50 of M. persicae was 58 and 66 ppm when treated with Destruxin A-760 and Destruxin A-724, respectively.

Table 1 Effect of the destruxin tested against M. persicae under laboratory conditions
Full size table

Under field conditions in both localities, M. persicae infection was significantly decreased. After 20 days of applications, the infections with the target pest obtained, 14 ± 4.7 and 11 ± 7.9 individuals in Destruxin A-724. Treated plots in Ibn Malek Nobaryia (Behira) and EL Sharkia (Zagazig), respectively. The corresponding data obtained 0 and 6 ± 1.2 individuals after Destruxin A-760. Treatment plots as compared to 56 ± 6.6 and 65 ± 8.9 individuals in the corresponding treatment areas.

In plots treated with Destruxin A-760, the infestations with M. persicae significantly decreased to 23 ± 4.5 and 36 ± 7 individuals, respectively, as compared to 152 ± 7.8 and 163 ± 4.7 individuals in the control, after 120 days of applications in both two governorates. The same obtained in case of Destruxin A-724 application after 120 days, the number of target insects were significantly decreased to 36 ± 7.7 and 46 ± 6.6 in Ibn Malek Nobaryia and EL Sharkia as compared to 152 ± 7.8 and 163 ± 4.7 individuals following the same order (Table 2).

Table 2 The infestation of tomato plants with M. persicae after treatment with Destruxin A-724 and Destruxin A-760 under field conditions
Full size table

Effects of Destruxin A-760 and Destruxin A-724 on the predator on C. undecimpunctata were shown in Table 3, all of the experiment results cleared that the predator C. undecimpunctata was not affected to both destruxins treatments. This predator C. undecimpunctata proved a higher resistance against destruxin treatments Table 3. Our results show that the predator preys on nymphs which are more susceptible to that on the adult stages of C. undecimpunctata. Under laboratory conditions, the nymph’s mortality of M. persicae obtained 79.10 ± 7.61 and 21.2 ± 4.45 after treated with Destruxin A-760 and Destruxin A-724, respectively (Table 3). In case of adult stages, they significantly decreased by 0.6 times treated with the following in the same order.

Table 3 Determination of the effect of (Destruxin A-760 and Destruxin A-724) on developmental stages of C. undecimpunctata after feeding M. persicae nymphs and adults
Full size table

In general, the indirect treatments by feeding by either of the treated of the prey M. persicae only (required) or by free-choice feeding on either treated or untreated prey (selectivity) revealed that the required ingestion of Destruxin A-760-infected prey of M. persicae caused a mortality percentage of 2.56 and 2.31 times that obtained from those of the given free-choice ingestion (selection treatment) for adult and nymph of C. undecimpunctata predators, respectively. The corresponding ratios for Destruxin A-724 were about 1.35 and 1.86, respectively. This indicated that the predator, particularly the adult predator, has a greater ability to recognize between the Destruxin A-760-treated prey and non-treated ones than Destruxin A-724fungus, which indicates that the adult predator can avoid Destruxin A-760 which infected the prey much more than that of the adult predator C. undecimpunctata avoiding the Destruxin A-724-infected prey. It is worth mentioning that no death from both destruxin’s infection was encountered in the check within the experimental period.

Under field conditions the experiments results showed that the population density of the predator C. undecimpunctata in the Destruxin A-760-treated area showed 36.22 and 19.81% reductions, 1 and 2 weeks after the last application, respectively. While the corresponding densities in the control area showed a 21.33 and 2.7% increase. However, the population, 3 weeks after the last application surpassed that of the check, showing a 6.36% increase (Table 4). In the I. fumosorosea-treated area, severe reductions in the population densities (63.33 and 43.99%) were estimated in the first and second weeks after the last application, respectively. There was less reduction (21.42%) estimated 3 weeks after the last application. However, the relatively higher reductions in predator densities in the treated areas, in the first weeks after the application. percentages increased or decreased in the C. undecimpunctata population as compared with the check according to Hendrson and Tilton (1955).

Table 4 The effect of destruxins on C. undecimpunctata (all stages)/50 tomato shrubs after successive post toxin application in tomatoes
Full size table

Field application of both the two bio-insecticides showed that in the control plots, the estimated yield weights were 2169 ± 36.82 and 2000 ± 80.54 kg/F in Ebn-Malek Nobaryia (Behira) and EL Sharkia governorate (Zagazig) during the season of 2018, respectively. While in Destruxin A-760- and Destruxin A-724-treated plots, the estimated weights of the tomato yields were 3988 ± 34.31 and 3158 ± 42.57 kg/F, respectively, in the Ebn-Malek El-Nobaryia (Behira) region. In the El Sharkia (Zagazig), the untreated plots recorded 1990 ± 80.54 kg/F but the weight showed a significant increase post of the Destruxin A-760 and Destruxin A-724 treatments. The percentages of yield loss in the untreated plots were 30 and 33% in the Ebn-Malek El-Nobaryia (Behira) and the El Sharkia (Zagazig), respectively (Table 5).

Table 5 Weight of harvested tomatoes and percentage of yield loss during season 2018 post the two-destruxin treatments of M. persicae in two governorates
Full size table

Discussions

The present study showed that C. undecimpunctata exhibits relatively high and reasonable resistance to the tested entomopathogenic fungi Destruxin A-724 and Destruxin A-760 infections, respectively, even when exposed to a lethal concentration for the prey insects.

Thungrabeab and Tongma (2007) concluded that some genera of fungi could be specific and might inflict only on certain types of hosts. They reported the work of James and Lighthart (1994) who declared that the fungus N. rileyi exhibits host preferential infection in lepidopterous larvae. Also, they found that M. anisopliae, B. bassiana, and P. fumosorosea fungi have potential to infect Hippodomia converges (coccinellidae) whereas N. rileyi did not. Goettel et al. (1990) found that some commercial formulation of the entomopathogenic fungi can control aphids and thrips with low impact on non-target insects. Todorova et al. (1994) found that different strains of (B. bassiana) fungus showed different effects on the two Coleopterous predatory insects due to the host response of the insects. Sabbour and Sahab (2007) control Agrotis ipsilon and Heliothis armigera by the entomopathogenic fungi. Sabbour and Abdel-Rahman (2007) found that the two microbial control agents reduce the number of sugar beet pests under laboratory and field conditions. Sabbour (2007a, 2007b) found that the entomopathogenic fungi Nomuraea rileyi and Isaria fumosorosea proved highly pathogenic to aphids and the natural enemies Coccinella spp. were not affected by the fungi treatments. Poprawiski et al. (1998) found that Serangium parcestosrum (Coccinellidae) had lower survival potential when sprayed with Zagazig (B. bassiana) fungus than that with P. fumosorosea fungus. Shanthakumar et al. (2010) considered that in spite of the great virulence of N. rileyi against Spodoptera litura, the pathogen proved reasonable safety to Trichogramma chilonis. It did not cause a reduction in their parasitization percentages.

The present results also indicated that the predator, C. undecimpunctata, particularly adult predators can distinguish between fungus infected from non-infected preys and they almost avoid treated ones, especially if given free choice feeding. This, however, was more pronounced in the case of N. rileyi than P. fumosorosea. Such observed phenomenon in our investigations was viewed by many authors. It was mentioned that predators, when given free choice to feed upon fungus-treated or untreated aphids, predation on infected preys was less than uninfected ones (Baverstock et al. 2007). Also, Roy et al. (2010) and Goettel et al. (1990) proved that C. septempunctata adults avoid contacting with leaf and soil surface inoculated with B. bassiana fungus and mycosed cadavers. The predator was positioned away from mycosed cadaver than uninfected ones.

Nevertheless, some researches indicated several adverse effects of some entomopathogenic fungi against some natural enemies. It was considered that C. septempunctata was somewhat susceptible to B. bassiana (Haseeb and Murad 1997; Delete et al. 1995). Farag (2008) considers that some entomopathogenic formulations of B. bassiana have deleterious effects on C. undecimpunctata if applied at high-concentration levels. However, different views about the safety of entomopathogenic fungi, declared by many authors, might be due to the relative efficacy of the fungus or its isolates on pests the exhibit different susceptibilities, bionomics, and characters as well as types of assessment and application rates.

Conclusion

Destruxin A-760 and Destruxin A-724 are a promising control against M. persicae as they reduce the pest population under laboratory conditions. On the other hand, in the population under field conditions, the pathogen significantly reduced the target pests. Pathogen Destruxin A-760 and Destruxin A-724 tested non-significant effects on the predator’s populations.