Spodoptera frugiperda eggs were procured from a local maize field in Chiang Mai, Thailand (18.7883° N, 98.9853° E). Eggs were placed in a ventilated plastic box (25×15×16 cm) and observed daily until hatching. Neonates (from first–third instars) were maintained on fresh and tender maize leaves in a plastic container. Small plastic boxes (11×8×10 cm) were used for the fourth to fifth instars to avoid cannibalism until pupation. The insect cultures were maintained at a temperature of 27°C and a 12:12 photoperiod (L:D).

From our previous research, we collected several potent strains of entomopathogenic fungi, particularly M. anisopliae (Vivekanandhan et al., 2020a; Vivekanandhan et al., 2020b) (see Supplementary Figures 1A, B ). Potato dextrose agar (PDA) medium was employed to maintain the fungal culture at 27°C for 14 days. After the incubation time, a sterile loop was used to gently scrape the surface of the agar plate to remove conidia (Vivekanandhan et al., 2022a). After aggressively shaking the mixture for three minutes with a 0.05% Tween 80 solution, the hyphae were separated using a sterile millipore cloth. These conidia were spun for twenty minutes to ensure that they were free of clumps. Fungal conidia were counted using a Neubauer hemocytometer chamber, and the active dosage of conidia was estimated using light microscopy (Olympus DSX510i, Japan) at 40X magnification. By diluting the original solution in sterilized distilled water, five distinct concentrations were obtained: 1.5×10³, 1.5⁵×10⁴, 1.5×10⁵, 1.5×10⁶, and 1.5×10⁷ conidia/mL. Insecticidal bioassays were executed with these different concentrations.

Under laboratory conditions, S. frugiperda larvae (second-fifth) were tested for larvicidal activity with M. anisopliae conidial concentration for up to 9 days. Fresh, immature maize leaves were dipped with various concentrations of fungal conidia (1.5×10³, 1.5×10⁴, 1.5×10⁵, 1.5×10⁶, and 1.5×10⁷ conidia/mL), and the treated containers were air dried for 10-15 minutes. Then, twenty-five (second to fifth) instar larvae of S. frugiperda were individually transferred into a (5×5×5 cm) plastic bioassay container. In the negative control group, 0.05% Tween-80 was used. Every day, the larvae were fed fresh, surface-sterilized maize leaves. Larval mortality was subsequently assessed for nine days at three-day intervals. Each concentration comprised three replications, and each replication contained twenty-five S. frugiperda larval (second-fifth) instars. The treatment bioassay container was incubated at a temperature of 27 ± 2°C.

Infected and control larvae were placed in Petri dishes lined with moist filter paper, and incubated at 27 ± 2°C. The fungal colonization of larvae was examined using a light microscope. After 5-7 days of incubation, all larval stages of S. frugiperda were evaluated for entomopathogenic fungal mycelium growth.

The hemocyte levels of leaf dip treated and control larvae were investigated. Three days post-treatment, hemolymph was collected from the control and experimental groups. The hemolymph was gently inverted two to three times in a 1.5-mL plastic tube with 1 mL of ice-cold anticoagulant buffer (0.01 mL EDTA, 0.1 mL glucose, 0.62 mL NaCl, and 0.26 mL citric acid; pH 4.6) to prevent breakdown. A modified Neubauer hemocytometer (Rohem, India) and a 50-X phase contrast microscope (Magnus MLXi, Japan) were used to count the hemocytes in each corner. The following formula was used to calculate the total hemocyte count:

X-Total number of cells counted; Dilution factor-20; Depth factor-10.

Live S. frugiperda control and conidia-treated larvae were washed with double-distilled water and wiped with tissue paper to remove excess water. Control and conidia-treated larvae were homogenized in 50 mL of ice-cold sodium phosphate buffer (20 mM, pH 7) to determine enzyme activity levels. The homogenates were previously centrifuged (8000 g for 20 min at 4°C), and the supernatants were used for further investigation. The homogenates were maintained on ice until use in subsequent experiments.

After larvicidal experiments, control and conidia-treated S. frugiperda larvae were separated. Excess water was wiped away before cleaning the larvae. Ellman et al. (1961) employed acetylcholine iodide as a substrate to evaluate the acetylcholine esterase activity of the larval homogenate. 850 mL of 100 mM sodium phosphate buffer was mixed with 50 mL of larval tissues (pH 7.5). Each reaction mixture contained 50 mL of 10 mM DTNB and 50 mL of 12.5 mM cholinergic iodide, which were kept at room temperature for 5 minutes. A suitable blank and a Multiskan EX (200–240 V) spectrophotometer were used to measure the sample’s optical density at 405 nm.

We used the method described in Van Asperen (1962) to investigate the activity of α- and β-carboxylesterase enzyme responses in larval homogenate. For the tests, 30 mL of homogenate were combined with 1 mL of sodium phosphate buffer (pH 7.0, 100 mM) containing 250 mM of α- and β-naphthyl acetate. This solution was then maintained at room temperature for 30 minutes. Each reaction mixture was then treated with 400 µL of 0.3% Fast Blue B in 3.3% SDS to block the enzymatic process. The combinations were allowed to remain at room temperature for 15 minutes to develop the distinct color. Multiskan EX-200-240V (Thermo Scientific) and the correct reagent of blank were used to determine the optical density of the sample. The activities of α- and β-carboxylesterase were then determined using a standard blot curve with naphthol as a control.

Eudrilus eugeniae were maintained at 27 ± 2°C on agricultural waste and cow dung. For experimental purposes, earthworms were removed from the soil by hand sorting. Earthworm bodies were washed with tap water to remove dirt. Earthworm cocoons were counted before being placed on separate bedding.

The artificial soil toxicity test was performed in accordance with the method detailed in Vivekanandhan et al. (2022a). The artificial soil was composed of 69% fine sand, 12% crushed sphagnum peat, and 19% kaolinite clay. Small amounts of fine sand, a chemical insecticide (Monocrotophos 250 ppm), and different amounts of fungal conidia were mixed together for each test concentration. After 0.65 kg of dirt and 15 mature earthworms had been added to each 500-mL glass jar, the jars were sealed. The jars were lightly covered with polypropylene covers to allow for ventilation. Jars were kept at a constant temperature of 27 ± 2°C and a relative humidity of 85–90%. Earthworm mortality was assessed after 9 days of treatment.

Individually, 15 adult E. eugeniae were tested for a period of 9 days with either fungal conidia, the organophosphate insecticide Monocrotophos, or distilled water (negative control). Both the control and treated earthworms were fixed in 3% formalin for 3 hours at 4°C after 9 days of treatment. The blocks were cooled for 3 hours at 4°C before being sliced into 2.5-mm-thick ribbons with a microtome (Leica, Germany). Ehrlich’s hematoxylin and eosin were used to stain E. eugeniae sections. After staining, the slides were left for drying and later viewed under a light microscope at 40X (Olympus CH20i/India).

The mortality data were subjected to analysis of variance (ANOVA) using logarithmic, arcsine, and square root transformed percentages. The mean of five replications was used to estimate the observed readings. The disparity in the data values among the various treatment groups for larvae was evaluated using Tukey’s multiple range test (p < 0.05) in conjunction with the Minitab®17 software. Sigma Plot 12 (Microcal Software) was employed to calculate the statistical changes for the estimation of mid-gut enzyme. The estimation of the lethal dosage (LC₅₀) against larvae after nine days was performed using the Probit study in conjunction with the Minitab®17 software package, with a 95% confidence interval.

There was a significant difference in larval mortality between the control and conidia-treated groups ( Figures 1A–D ). For second instar larvae, the various fungal conidia concentrations produced 37-69% mortality three days post treatment (df: 5, F _(5,12) =144.071; p≤ 0.01), 60–89% mortality six days post treatment (df: 5, F _(5,12)=264.862; p≤ 0.01), and 62-97% mortality nine days post treatment (df: 5, F _(5,12)=194.480; p≤ 0.01). When compared to the control ( Figure 1A ), significant differences were observed. For third instar larvae, the fungal conidia produced 25-55% mortality three days post treatment (df: 5, F _(5,12)=69.700; p≤ 0.01), 48-78% mortality six days post treatment (df: 5, F _(5,12)=142.322; p≤ 0.01), and 50-89% mortality nine days post treatment (df: 5, F _(5,12)=101.656; p≤ 0.01) When compared to the control ( Figure 1B ), significant differences were observed. Notable difference in larvicidal activity were observed for forth instar larvae. The fungal conidia showed 18-50% mortality three days post treatment (df: 5, F _(5,12)=210.270; p≤ 0.01), 36-66% mortality six days post treatment (df: 5, F _(5,12)=454.450; p≤ 0.01), and 45-77% mortality nine days post treatment (df: 5, F _(5,12)=188.677; p≤ 0.01). When compared to the control ( Figure 1C ), significant differences were observed. For fifth instar larvae, fungal conidia produced 9-42% mortality three days post treatment (df: 5, F _(5,12)=208.200; p≤ 0.01), 29-61% six days post treatment (df: 5, F _(5,12)=195.425; p≤ 0.01), and 44-72% mortality nine days post treatment (df: 5, F _(5,12)=153.492; p≤ 0.01). When compared to the control ( Figure 1D ), significant differences were observed. Overall, significant differences in larval mortality were observed between control and treatment for all five different concentrations of fungal conidia against second to fifth larval stages of S. frugiperda.

Spodoptera frugiperda larvae were subjected to M. anisopliae fungal conidia at various concentrations (1.5×10³, 1.5×10⁴, 1.5×10⁵, 1.5×10⁶, and 1.5×10⁷ conidia/mL). When the concentrations of fungal conidia were increased, haemocyte levels gradually decreased. After three days of treatment, fungal conidia at a higher dose of 1.5×10⁷ conidia/mL reduced haemocyte count by 25.01% and was statistically different from the control (df 5; F _(5,12) = 272.004; P≤ 0.01) ( Figure 2 ).

The biochemical effects of entomopathogenic fungal conidia on S. frugiperda larvae were investigated. As conidia concentrations increased, the amount of AchE enzyme activity decreased. A maximal level of AchE was expressed at a concentration of 1.5×10⁷ conidia/mL (df 5; F _(5,12) = 170.162; p≤ 0.01) ( Figure 3 ). The enzyme level in the control larvae was found to be normal. Enzyme expression was mainly dose-dependent, with larvae exposed to entomopathogenic fungal conidia displaying lower levels of AchE enzymes ( Figure 3 ).

Results showed that the treatment of M. anisopliae fungal conidia against larvae of S. frugiperda resulted in a lower level of α and β-carboxylesterase enzyme activity ( Figures 4A, B ), with the lowest level achieved after three days of treatment. Conidia of M. anisopliae had a dose-dependent effect on the α and β-carboxylesterase enzyme activity in larvae. When larvae were exposed to fungal conidia, their α-carboxylesterase activity was considerably reduced compared to control (5.88 to 2.6 mg protein/mL of homogenate) (df 5; F _(5,12) = 139.597; p≤ 0.01). ( Figures 4A, B ). A similar finding was observed with β-carboxylesterase, which showed a decrease in activity (from 4.9 to 1.9 mM/protein/mg/min) (df = 5; F_{(5, 12)} = 172.778; p≤ 0.01) in S. frugiperda larvae ( Figures 4A, B ).

The results of the artificial soil bioassay revealed that entomopathogenic fungi had no effect on the beneficial non-target species E. eugeniae. The results suggests that the mortality rate ranged from 2 to 8% with the treatment of fungal conidia and was not significantly different to control individuals. Fungal conidia thus had no significant effect on earthworm mortality (df = 5; F _{(5, 12)} = 5.133; p 0.01). In contrast, more than 93% of earthworms died because of the monocrotophos treatment. The chemical pesticide, Monocrotophos caused high mortality on E. eugeniae when compared to fungal conidia treatment (df = 4; F _{(1, 4)} = 420.250; p 0.01). Monocrotophos was substantially more toxic to the earthworm when compared to M. anisopliae conidial concentrations of 1.5×10⁷conidia/mL. Furthermore, M. anisopliae fungal conidia were not pathogenic to earthworms after 9 days of treatment. The monocrotophos and control groups were statistically different (df= 4; F _{(1, 4)} =420.250; p≤ 0.01). Monocrotophos was found to be the most dangerous to earthworms when compared to fungal conidia and the negative control ( Figure 5 ).

The histopathological findings revealed that M. anisopliae conidia are not toxic to earthworms as compared to Monocrotophos. In the fungal conidia group, treated earthworm gut tissues did not contain any fungal conidia, and the fungal conidia were unable to disintegrate the earthworm epidermis. The fungal conidia had no toxicity effects on earthworm gut tissues; all tissues including the epidermis, circular muscle, setae, mitochondrion, and intestinal lumen tissues were normal and similar to the negative control. In contrast, treatment with monocrotophos resulted in significant damage to the earthworm tissues including the epidermis, circular muscle, setae, mitochondrion, and intestinal lumen tissues ( Figure 5 ).

Metarhizium anisopliae conidia with a green color mycelium completely invaded infected S. frugiperda larvae (second to fifth instar). In contrast, control insects were healthy and showed no signs of fungal infection. M. anisopliae fungal mycelium has been found to be more capable of proliferating and infiltrating host tissues, resulting in dose-dependent mortality against larvae. Larvae were found to be especially vulnerable to fungal conidia at a concentration of 1.5×10⁷ conidia/mL^-1.