C. elegans strain N95 was maintained on a nematode growth medium with Escherichia coli strain OP50 lawn. A chunk of agar containing hundreds of worms was cut and transferred onto an overnight grown OP50 lawn in a fresh Petri plate. Plates were incubated at 25°C for 3 days for nematode multiplication.

The authenticated population of M. incognita was maintained and multiplied on tomato roots (Solanum lycopersicum L. cv. Pusa Ruby) in the greenhouse. Approximately 35-day-old plants were harvested, roots washed free of soil, and used for collecting fresh egg masses, which were hatched via modified Baermann’s technique (Whitehead and Hemming, 1965) to obtain infective second-stage juveniles (J2s) required for all the experiments.

In the present study, we have taken the two Indian isolates of Arthrobotrys and Tolypocladium for species identification. Pure cultures of both the fungal isolates were grown separately in potato dextrose broth (PDB) at 25°C for 1 week. Cultures of Arthrobotrys spp. and Tolypocladium spp. were observed under an Olympus BX50 compound microscope, and morphological measurements of conidia, conidiophores, and phialides (length and width) were recorded. For molecular characterization, genomic DNA was extracted from a 1-week-old mycelial mat using the CTAB method as previously described by Wang et al. (2015). Polymerase chain reaction (PCR) analyses were carried out with genomic DNA extracted from the fungal isolates to amplify the two markers; internal transcribed spacers (ITSs) and β-tubulin using universal primers, ITS-1 (5′-TCCGTAGGTGAACCTGCGG-3′) and ITS-4 (5′-TCCTCCGCTTATTGATATGC-3′) as well as βt2A (5′-GGTAACCAAATCGGTGCTGCTTTC-3′) and βt2B (5′-ACCCTCAGTGTAGTGACCCTTGGC-3′), respectively. PCR amplification was carried out as per the procedure by Glass and Donaldson (1995). The amplified products were analyzed by electrophoresis on a 1.2% (w:v) agarose gel (Sigma Aldrich, United States) and visualized using a gel documentation system (Alpha Image Analyzer, United States). PCR products were further purified using a PCR clean-up kit (Macherey-Nagel, Germany) and sequenced (Applied Biosystems, United States). The sequences of ITS and β-tubulin generated for the tested fungi were compared with previously reported sequences in the GenBank database. All the sequences were imported for alignment by setting default parameters in the ClustalW algorithm into a MEGA X software application package. To search for homologs of both genes, sequences from each isolate were subjected to the Nucleotide Basic Local Algorithm Search Tool of the National Center for Biotechnology Information.¹ The sequences were further analyzed and submitted to GenBank. Additionally, phylogenetic trees using the sequences of ITS and β-tubulin of the tested isolates were constructed in MEGA X using the maximum composite likelihood approach considering 1,000-bootstrap replications under Kimura two-parameter distance models (Kimura, 1980). For this, Neurospora crassa and Cordyceps ophioglossoides were used as out-groups (Sung et al., 2007; Swe et al., 2008; Wright et al., 2009). In addition, the identified fungal isolates were submitted to the Indian Type Culture Collection (ITCC), which is an affiliate member of the World Federation for Culture Collections and is registered with the World Data Centre for Microorganisms (registration number 430).

Pure cultures of both the fungi were grown in six different media, viz., potato dextrose agar (PDA), cornmeal agar (CMA), Czapek malt agar (CzMA), rose bengal agar (RBA), peptone-yeast-glucose, and synthetischer nährstoffarmer agar (SNA) (HiMedia, India) in 90-mm diameter (dm) Petri plates at 25°C to examine the differences in structure, color, growth rate, and sporulation among colonies. For this, a small piece of agar measuring around 5 × 5 mm was cut from a well-established colony and placed upside down at the center of the fresh Petri plate containing different media. Three replicates (n = 3) for each medium were used, and observations were made for all the replicates separately. The fungal growth was determined at 3-, 6-, 9-, and 12-day post-inoculation (dpi) by measuring the colony diameter. The growth rate was estimated as growth rate per day (mm/day) = [the average diameter of hyphae measured in that day - 5 (diameter of original inoculum)]/days of culture, as described earlier by Wang F.H. et al. (2017).

Furthermore, the medium that supported the maximum growth rate was used to evaluate the effect of different temperatures, viz., 15, 20, 25, 30, and 35°C, by following the procedures described earlier. Additionally, a combination of the specific medium and temperature supporting maximum fungal growth was studied to evaluate the effect of different pH levels (ranging from 4 to 10). Three replicates (n = 3) were used for each treatment. The growth rate was estimated at 3, 6, 9, and 12 days after inoculation by a formula adopted from Wang F.H. et al. (2017). The mean value of the diameter of fungal hyphae was measured each day in each group and used for calculating the growth rate by deducting the original diameter of the fungal disc used for sub-culturing.

The two selected fungi were grown on a PDB medium for 10 days at 25°C with 180 rpm in an incubator shaker. Approximately 100 surface-sterilized nematodes (both M. incognita and C. elegans) were added “separately” into 5-ml Eppendorf tubes containing 3-ml fungal filtrate (FF) of each isolate. The tubes were incubated on the shaker at 120 rpm at 25°C for 3 days. The worms were washed thrice with sterile water (SW) followed by re-incubation in SW for 24 h at 25°C for revival in the nematode movement. Then, the nematodes were examined under a stereo binocular microscope to record dead and live nematodes, and nematode mortality percentage was calculated as% mortality = (number of dead nematodes/total number of nematodes) × 100. There were three replicates (n = 3) for each treatment, and observations were made for all the replicates independently. Worms in the PDB medium served as the control for comparison.

Fungal parasitism against M. incognita J2s and C. elegans L3s was evaluated in water agar plates. Briefly, fungi were grown on PDA at 25°C for 6 days, and then, 5-dm discs were transferred into 2% water-agar plates containing 1% ampicillin (100 μg ml^–1). The plates were incubated at 25°C for 3 days, and then 100 surface-sterilized worms were added to the plates separately (Contina et al., 2017). The number of captured larvae was scored after 3 days using a light microscope (40×). Three replicates (n = 3) were used for each treatment and compared with control (nematodes only in water agar medium). Parasitization was calculated as previously described by Siddiqui and Shaukat (2004) as% parasitization = (number of parasitized nematodes/total number of nematodes) × 100.

Captured nematodes of both M. incognita J2s and C. elegans L3s were also examined under a scanning electron microscope. For this, worms were fixed in 2% glutaraldehyde in 0.1-M phosphate buffer (pH 7.2) overnight, followed by 2% osmium tetroxide fixation for 6 h and dehydration using a series of ethanol as described by Den Belder et al. (1993).

One-month-old healthy tomato seedlings (S. lycopersicum L. cv. Pusa Ruby) were transplanted into 4-inch dm pots filled with 400-g autoclaved soil, sand, and farmyard manure (50:25:25), which were kept in the greenhouse. At the same time, fungal suspensions were prepared by culturing the colonies on medium, which showed a high sporulation level, and incubated at 25°C for 10 days. Subsequently, 5 ml of sterile distilled water was added to the plate and the spores were scraped using a spatula. The mixture was placed in a small beaker and stirred for 10 min, filtered using cheesecloth, and quantified using a hemocytometer on a light microscope. Finally, the fungal suspensions were adjusted to get 1 × 10⁶ spores/ml and used for inoculation around the root zone during transplantation. After 1 week, 800 J2s of M. incognita were inoculated at the rate of 2 J2s per gram of soil, and the pots were maintained in the greenhouse. Plants that received sterile distilled water served as control. Five replicates (n = 5) were used for each treatment, and observations were made for all the replicates individually. Plants were carefully harvested 45 dpi, and roots were washed free of soil. Plant growth parameters, viz., plant length, fresh weight, and dry weight were recorded. Additionally, nematode disease burden was determined as the number of galls, egg masses, and eggs per egg mass and used for deriving the nematode multiplication factor (MF) as [(total number of egg masses per plant × average number of eggs per egg mass) ÷ initial nematode inoculum level] as described previously by Hada et al. (2020, 2021a).

Species identification has been primarily made using morphological characters along with the measurement of taxonomically useful features. The Arthrobotrys isolate exhibited straw white-colored colonies with raised concentric rings along with thin and hairy rings on PDA media. The length and width of conidia and conidiophores of A. thaumasia were approximately 24.58–60 and 10.15–22.88 μm, as well as 211–446 and 2.3–5.4 μm (n = 50), respectively. Furthermore, the shape of the conidium appeared as an inverted pear with 1–3 septa. All these characteristics matched with the reported description of the fungus A. thaumasia (Supplementary Table 1 and Figure 1A), and we designated our isolate as A. thaumasia At_RK.

Similarly, in the case of Tolypocladium isolate, the colony appeared hairy with whitish cream as well as reverse yellow to pale on PDA media. The conidium was 2–4.3 and 1.3–1.7 μm in length and width, respectively. Likewise, the length and width of conidiophores were approximately 31–44 and 1.1–2.8 μm, respectively. The phialide length and width were 4.5–8.5 and 2–3.2 μm, respectively. Finally, conidia hyaline were smooth-walled, short cylindrical, straight and/or slightly curved, and both ends obtusely rounded. Comparison of all these characteristics matched with the reported description of T. cylindrosporum (Supplementary Table 1 and Figure 1B), and our isolate was designated as T. cylindrosporum Tc_RK.

To further confirm the species identity, the fungi discussed earlier were characterized using two molecular markers, ITS and β-tubulin. Sizes of ITS and β-tubulin that could be amplified in both the isolated fungi were 580 and 370 bp, respectively. Purified PCR products were sequenced and submitted to the National Center for Biotechnology Information database (Supplementary Table 2). Homology search of ITS and β-tubulin sequences using BLAST program showed that A. thaumasia At_RK sequence was 99% (accession: KT215216.1) and 97% (accession: EU977531.1) identical to earlier reported sequences of A. thaumasia, respectively. Similarly, the ITS sequence of T. cylindrosporum Tc_RK was 99.06% (accession: NR_167967.1), identical to a previously reported sequence of T. cylindrosporum.

Furthermore, we constructed an evolutionary tree based on the sequences of the ITS, and β-tubulin that demonstrated A. thaumasia At_RK was closest to A. thaumasia strain CBS 376 97 (accession: KT215216.1), whereas β-tubulin sequence was close to A. thaumasia isolate 111 (accession: EU977531). Similarly, the ITS sequence of T. cylindrosporum Tc_RK was closest to T. cylindrosporum isolate TCDAs18R1A9 (accession: MT911434.1), whereas that of β-tubulin was close to T. tropicale strain IQ214 (accession: KF747166.1) and T. tropicale strain MX338 (accession: KF747190.1). The results of phylogenetic analysis based on the sequence of ITS and β-tubulin indicated that the isolates in the present study could be different geographical strains of A. thaumasia and T. cylindrosporum (Supplementary Figure 1).

Additionally, the studied fungal isolates were submitted to ITCC with strain/accession numbers: ITCC8969 for T. cylindrosporum Tc_RK and ITCC8970 for A. thaumasia At_RK (Supplementary Table 2).

Selected fungal isolates cultured on different nutrient media showed clear variations in the growth rates and trap formation at different time intervals. The highest growth rate of A. thaumasia was observed significantly in SNA media (2.00 mm/day), followed by CMA (1.375 mm/d) and CzMA (1.272 mm/day) compared with other tested media after 3 days (Supplementary Figure 2A). Likewise, at 6 dpi, the growth rates were observed to be significantly higher in SNA (1.432 mm/day), CMA (1.405 mm/day), PDA (1.405 mm/day), and CzMA (1.395 mm/day) media, respectively. Finally, the hyphae covered the entire Petri plate, and growth rates of fungus were equal in all the media after 9 dpi (0.94 mm/day) and 12 dpi (0.71 mm/day). Concurrently, sporulation was observed, and the number of spores was counted using a hemocytometer. The maximum sporulation was noticed on RBA (30.3 × 10⁴ spore/ml) followed by CzMA medium (9.1 × 10⁴ spore/ml) (Supplementary Table 3). The response to different media shows that A. thaumasia mycelia are very sensitive to salts and amino acids because they promote trap formation at specific nutrient combinations and inhibit it partially or completely at other combinations. There are significant differences among the different media compositions (p < 0.01).

In the case of T. cylindrosporum isolate, the growth was slow in all the tested media (Supplementary Figure 2B), but PDA media comparatively showed a considerable growth rate (0.64 mm/day) at 3 and 6 dpi. Subsequently, the maximum growth rate was observed in the PDA media (0.525 mm/day) and lowest in CMA media (0.328 mm/d) after 9 and 12 dpi. Consequently, the fungal sporulation was also observed in all the tested media, and the maximum sporulation was obtained on SNA media (9006.3 × 10⁴ spore/ml) followed by PDA (7203.3 × 10⁴ spore/ml) and CMA media (3427.2 × 10⁴ spore/ml) (Supplementary Table 3 and Supplementary Figure 3A).

Furthermore, A. thaumasia grown on SNA media that provided a maximum growth rate was used to evaluate the effect of different temperatures. The data indicate that the growth rate of A. thaumasia was significantly higher at 30, 25, 20, and 15°C (0.293, 0.634, 0.61, and 0.523 mm/day) after 3, 6, 9, and 12 dpi, respectively. In the case of T. cylindrosporum that grew on PDA media, it exhibited the quickest growth rate at 20°C after 3 and 6 dpi (0.525 and 0.465 mm/day, respectively), and the highest growth was noticed after 9 and 12 dpi (0.563 and 0.475 mm/day, respectively) as compared with other treatments (p < 0.01). The fungus could not grow at 35°C after 3 and 6 dpi but could grow in the range from 15 to 30°C (Supplementary Figure 3B).

Additionally, a combination of the specific media and temperature supporting maximum fungal growth was studied to evaluate the effect of different pH levels. For this, an isolate of A. thaumasia was cultured on SNA media at 30°C. The result showed that the radial growth rate at pH 6–9 (8.575 and 8.475 mm/day) was significantly faster than other tested pH after 3 days. The highest growth under different pH recorded was at pH 9 > pH 6 > pH 7 = pH 8 > pH 10 > pH 5 > pH 4. Similarly, a culture of T. cylindrosporum grown on PDA medium at 20°C caused the highest growth rate at pH 6 (1.55–5.15 mm/day) followed by pH 9 and 10, and minimal growth was recorded at pH 4 (0.525–1.975 mm/day) (Supplementary Figure 3C).

The FF of the selected isolates were found to be effective against both M. incognita J2s and C. elegans L3s compared with control in the PDB medium. T. cylindrosporum Tc_RK and A. thaumasia At_RK caused 87.3 ± 6.02 and 57.7 ± 3.5% mortality of M. incognita J2s, respectively. Likewise, FF of T. cylindrosporum Tc_RK and A. thaumasia At_RK caused 64 ± 3.6 and 53.7 ± 2.3% mortality in C. elegans L3s (Supplementary Table 4). Both M. incognita and C. elegans worms exhibited normal behavior in the PDB control without any mortality.

Fungal parasitism against M. incognita J2s and C. elegans L3s was evaluated on water agar plates under in vitro conditions. The tested fungal isolates were found to be effective against both M. incognita and C. elegans after 3 days compared with control in the water agar plates (Supplementary Table 4). Results showed that A. thaumasia At_RK formed different types of traps to hunt the nematodes, viz., 2D adhesive network, 3D adhesive network, and non-constricting rings (Figure 2). It caused approximately 82 ± 3.6% parasitism of M. incognita J2s. During parasitization, the fungal mycelium penetrated the nematode cuticle, developed inside its body, consumed the body contents, then ruptured the cuticle, and subsequently grew out of the body (Figure 3). The fungus mostly penetrated the nematode body along the lateral lines in both the nematodes (Figures 3, 4). Similarly, the percentage of C. elegans parasitized by A. thaumasia At_RK was approximately 73 ± 4.5%. It captured C. elegans, penetrated the cuticle and grew inside the body, and then ruptured the cuticle (Figure 4). It was noticed that it colonized approximately five to six M. incognita J2s but only one to two C. elegans L3s.

Direct parasitism of T. cylindrosporum Tc_RK against both the nematodes was not as clear as in A. thaumasia. The capture and colonization of M. incognita by T. cylindrosporum were 65.2 ± 3.1% after 3 dpi and 57.7 ± 3.6% in C. elegans (Supplementary Table 4). The attachment of fungal spores on M. incognita cuticle was observed as a first step of the parasitism process (Figure 5). The fungal parasitization and spore attachment on C. elegans were significantly less compared with M. incognita, but consumption of nematode contents was detected quite well, as shown in Figure 6.

Fungal spore suspension having 1 × 10⁶ spores/ml was inoculated in the vicinity of tomato roots during transplantation. The selected fungal isolates significantly increased plant growth parameters in plant length, weight, and dry weight when compared with the control plants treated with only nematodes and SW. The maximum plant length (68.2 cm) and weight (14.92 g) were observed in the T. cylindrosporum Tc_RK-treated samples. Also, plants treated with A. thaumasia At_RK showed an average length of 66.8 cm, whereas the average weight was 13.4 g (Table 1). Interestingly, A. thaumasia At_RK-treated plants showed significantly higher dry weight (3.22 g) compared with only nematode-treated plants without the addition of fungal suspension.

Furthermore, the effect of the selected fungi was determined on nematode disease burden in tomato plants inoculated with M. incognita. Results showed that the application of fungal suspension caused a significant reduction in nematode infection compared with control. It was found that average galling was in the range of 29 ± 16.2 in A. thaumasia- and 43.4 ± 12.8 in T. cylindrosporum-treated plants compared with control, which documented approximately 142 ± 5.4 that led to a decrease in galling intensity by 80 and 69%, respectively. Corroborating this, the number of egg masses was approximately 18.6 ± 6.6 and 7.2 ± 3.8 in A. thaumasia- and T. cylindrosporum-treated plants, respectively, compared with 50.6 ± 4.5 in control plants. Likewise, the number of eggs per egg mass was 325.2 ± 31.1 and 221.2 ± 24.4 in A. thaumasia- and T. cylindrosporum-treated plants, respectively, compared with 583.4 ± 40.7 in control plants. As a consequence, approximately 63 and 86% reduction in egg masses and 44 and 62% reduction in eggs per egg mass could be obtained in A. thaumasia- and T. cylindrosporum-treated plants, respectively. Ultimately, approximately 80 and 95% decline in nematode MF was observed in A. thaumasia- and T. cylindrosporum-treated plants, respectively (Table 1).

Hexane soluble fraction of A. thaumasia and T. cylindrosporum was subjected to GC–MS analysis to identify VOCs. GC–MS analysis of hexane extract of A. thaumasia showed several peaks corresponding to 16 volatile compounds, representing 79.9% of the extract. The compounds are listed based on their retention time in Table 2. Interestingly, volatile sulfides such as n-propyl-butyl-disulfide (3.04%), bis (1-methylpropyl)-disulfide (2.80%), and dibutyl-disulfide (1.99%) were detected in the total ion chromatogram. 2-Methyl-1-pentanethiol (32.93%) was the major constituent followed by trimethyl-heptadien-4-one (23.34%), hexene-2,5-diol (5.08%), and chloro-3-butyltetrahydropyran (5.08%). Besides, dimethyl-pentanal (1.10%) and nerolic acid (1.10%) were also detected.

Likewise, GC–MS analysis of hexane extract of T. cylindrosporum also displayed various major and minor peaks corresponding to 11 VOCs. Compounds are listed as per their elution from the HP-5MS column (Table 2). Most of the peaks of the total ion chromatogram of the sample were identified as hydrocarbons and alcohols. Among these, decane (17.70%) was the most abundant. Other major compounds identified were ethyl-3-methyl-benzene (8.10%), octadecenal (7.52%), bis-(dimethyl-ethyl)-phenol (7.04%), dodecane (5.73%), hexadecanol (4.86%), and undecane (4.34%).

Furthermore, UPLC-QToF-ESI-MS analysis of a methanolic extract of A. thaumasia revealed various peaks in the total ion chromatogram; however, six metabolites were identified tentatively based on their molecular ion peaks, considering error mass (δ) value within 10 ppm (Table 3). These six metabolites were cyclo(L-Pro-L-Val) or piperazinone (1), cyclo(L-Pro-L-Leu), a pyrazine-1,4-dione derivative (2), paganin A (3), talathermophilin E (4), dactylarin (5), and trichodepsipeptide A (6). Among these, cyclo(L-Pro-L-Leu) is a homodetic cyclic peptide consisting of leucyl and prolyl amino acid residues. Furthermore, these metabolites were also confirmed from their possible fragmentation pattern in high-resolution MS. Cyclo(L-Pro-L-Val) or piperazinone was identified from its molecular ion peak at m/z 196.1201. Besides, fragmentation of the molecular ion peak resulted in daughter ion peaks at m/z 154 and m/z 126, originated due to sequential loss of isopropyl (43 amu) and carbonyl (28 amu) moieties. Another metabolite, cyclo(L-Pro-L-Leu), showed a characteristic molecular ion peak at m/z 210.1373, which on fragmentation produced possible daughter ions at m/z 195, 166, and 138, originated due to subsequent breaking of methyl (15 amu), ethyl (29 amu), and carbonyl groups (Supplementary Figure 4A). Similarly, the other four metabolites, paganin A, talathermophilin E, dactylarin, and trichodepsipeptide A, were characterized from their respective molecular ion, sodiated or adduct ion peaks at m/z 152.0940, 312.1711, 327.0921, and 711.4034.

UPLC-QToF-ESI-MS analysis of a methanolic extract of T. cylindrosporum tentatively revealed the identification of a total of nine metabolites. These metabolites, tolypocladenols A1/A2 (i), tolypyridone A (ii), tolypyridone B (iii), penicillenol B1/B2 (iv), pyridoxatin (v), terpendole E (vi), destruxin A (vii), 4-chloro-2-phenylphenol (viii), and acetamido-6-[(O-methyl-glucopyranosyl)oxy]anthraquinone (ix), were characterized based on their exact monoisotopic mass value (error mass < 10 ppm) and mass fragmentation pattern (Table 3). The first metabolite, tolypocladenols A1 or A2, was identified from its adduct ion peak at m/z 356.1852. Fragmentation of the adduct ion produced daughter ion peaks at m/z 327 and 191 originated due to the removal of two methyl (−28 amu) and side hydrocarbon chains (136 amu) (Supplementary Figure 4B). Similarly, the remaining eight metabolites were also characterized from their corresponding fragmentation pattern of the molecular ion peaks.