The wilt affected banana corms from different banana-cultivating provinces of Tamil Nadu, India pertaining to Lakshmipuram village of Theni province (10° 11′ 16.62″ N77° 47′ 34.4112″E/ latitude 10.187950/ longitude 77.792892), Chinnamanur village of Theni province (9° 50′ 22.776″N77° 22′ 58.0692″E/ latitude 9.839660/ longitude 77.382797), Thondamuthur village of Coimbatore province (11°00′35″N 76°49′41″E/ latitude 10.9905/ longitude 10.9905), Sirumugai village of Coimbatore province (11° 19′ 16.9″N, 77° 0′ 18.8″E/ latitude 11.3183/ longitude 77.0066), Ammapalayam village of Salem province (11° 40.7123′ N 78° 7.4319′ E/ latitude 11.678539/ longitude 78.123865), Athani village of Erode province (11.5232° N, 77.5120° E/ latitude 11.515219/ longitude 77.452367), Andhiyur village of Erode province (11° 34′ 37.4628″N 77° 35′ 15.8280″E/ latitude 11.577073/ longitude 77.587730), and Gobichettipalayam village of Erode province (11.4548° N, 77.4365° E/ latitude 11.450410/ longitude 77.430036) were collected. Pathogenic Fusarium spp. associated with banana cultivars susceptible to Panama wilt of banana were isolated as per the protocol described by Nelson et al. (1983). Genomic DNA of Fusarium was extracted from the mycelium of pure culture through the CTAB method (Griffith and Shaw, 1998). Genomic DNA was used as a template for PCR amplification of Foc isolates using ITS1 (5′-TCCGTAGGTGAACCTGCGG-3′) and ITS4 (5′-TCCTCCGCTTATTGATATGC-3′) primers (White et al., 1990). Further to confirm the Foc races, secreted in Xylem (SIX) Six13c 343 F (5′CAGCCTCCTAGCGTCGAAAA 3′) and Six13c R (5′CCGTGATGGGGTACGTTGTA 3′) were used (Czislowski et al., 2018). The program cycle comprised of initial denaturation (95°C) for 2 min, followed by 40 cycles of denaturation (95°C) for 1 min, annealing at 58°C for 1 min, extension for 1 min at 72°C, and with a final extension at 72°C. Gel electrophoresis and staining was done by loading 10 μl of PCR product on 1% agarose gel in TAE buffer at 80 V for 50 min at 25°C. A 1 kb DNA ladder was used to determine the size of amplified genomic products. PCR products were photographed using a gel documentation system. The amplified genomic product was sequenced by Eurofins Genomics Biotech Pvt. Ltd., Bangalore, India. Gene homology searches were performed using NCBI BLAST. Sequences were compared with different Foc isolates retrieved from the GenBank database. Newly obtained sequences were submitted to the GenBank database (New York, USA) and accession numbers were obtained. Phylogenetic analysis was performed with MEGA7 software (Tamura et al., 2007).

An antagonistic bacterial endophyte was isolated from 11-month-old Foc-resistant banana cv. Yengambi KM5 (AAA) maintained at the Banana Field Gene Bank, Tamil Nadu Agricultural University, Coimbatore (latitude: 11° 07′ 3.36″ N, longitude: 76° 59′ 39.91″ E), Tamil Nadu, India. The identity of antagonistic bacterial endophytes against Foc KP was confirmed using a 1.5 kb full-length 16S rRNA gene: 8 F (5′AGAGTTTGATCCTGGCTCAG-3′), 1492 R (5′GGGTTACCTTGTTACGACTT-3′). Antagonistic isolates YEBBR6, YEBN2, YEBRH5, YEBRT4, YEBFR1, and YEBFL6 that inhibited Foc KP were confirmed as Bacillus velezensis (MT372157), Bacillus albus (MT120179), Achromobacter xylosoxidans (MK258170), Beijerinckia fluminensis (MK263670), Acinetobacter refrigeratoris (MT326234), and Bacillus endophyticus (MT326238), respectively (Saravanan et al., 2021a).

The efficacy of bacterial endophyte B. velezensis YEBBR6, isolated from resistant banana cv. Yengambi KM5 (AAA) was tested in vitro against F. oxysporum f. sp. cubense Foc KP (NCBI accession no. MW 436477). The ability of the antagonist B. velezensis to suppress Foc KP was assessed through a dual culture technique. A 9 mm mycelial disc was excised using a sterile cork borer from a 7-day-old culture of Foc KP. It was placed on one side of a Petri plate containing PDA medium, 10 mm away from the periphery. The bacterial endophyte (24 h old) was streaked on the medium 10 mm away from the periphery, exactly opposite the mycelial disc of Foc KP. The plates were incubated at 28 ± 2°C for 7 days. The VOCs/NVOCs produced by B. velezensis YEBBR6 in PDA medium from the zone of inhibition were extracted by excising the agar from the zone of inhibition using a sterile scalpel. Excised agar with VOCs/NVOCs was blended with HPLC-grade acetonitrile in 1:4 ratios (5 g agar in 20 ml of HPLC grade acetonitrile). The mixture was sonicated twice for 30 s at 30% of the power of the sonicator for homogenization. After homogenization, samples were centrifuged and filtered to remove solid particles. The samples were dried in a vacuum flash evaporator (Rotrva Equitron Make). After removing the eluent, the final product was dissolved in 1 ml of HPLC-grade methanol (Cawoy et al., 2015). The difference in VOCs/NVOCs profile produced during the interaction of B. velezensis YEBBR6 with Foc KP was characterized with PDA control, pathogen-inoculated control and bacterial antagonist-inoculated control through GC/MS (GC Clarus 500 Perkin Elmer Analysis) using the NIST version 2005 MS data library.

Liquid formulation was developed by inoculating single colonies of B. velezensis YEBBR6, B. albus YEBN2, Achromobacter xylosoxidans YEBRH5, Beijerinckia fluminensis YEBRT4, Acinetobacter refrigeratoris YEBFR1, and Bacillus endophyticus YEBFL6 in Luria Bertani (LB) broth. Inoculated LB broth was incubated at 28 ± 2°C for 72 h in an orbital shaker at 150 rpm. The concentration of bacterial cells in the culture broth was assessed by measuring the absorbance at 600 nm in a bio-spectrophotometer (Eppendorf Make) at an optical density (OD) of 1.0 ABS. Later, bacterial suspension in LB broth pertaining to six different bacterial antagonists was blended with 10 ml of glycerol (1%), 10 ml of tween 20 (1%), and 1% poly vinylpyrrolidone (PVP) supplied from Sigma–Aldrich. The resultant mixture of individual bacterial antagonists was mixed by incubation in an orbital shaker at 150 rpm for 10 min. The formulation of individual bacterial antagonists was adjusted to 8 × 10⁸ cfu/ml (Vinodkumar et al., 2017).

Micropropagated banana cv. Karpooravalli (ABB) plantlets were biohardened with a liquid formulation of antagonistic bacterial endophytes (8 × 10⁸ cfu/ml). Micropropagated plantlets maintained in the protrays were biohardened on days 15 and 30 by drenching the root zone with 1% bacterial formulation. Biohardened plantlets were planted in polybags filled with EC (< 0.6 mS / cm) and pH (7)-stabilized sterile Cocopeat and incubated in the mist chamber. After 1 month of primary hardening, the micropropagated plantlets were subjected to secondary hardening. During secondary hardening, banana plantlets were transplanted in polybags (10 cm × 15 cm) containing one part of sterile red soil, one part of sterile sand, and one part of sterile farmyard manure mixture. The plantlets were hardened twice at 15-day intervals with 1% liquid formulation of antagonistic bacterial endophytes (8 × 10⁸ cfu/ml) in the root zone until saturation and maintained in a greenhouse for further studies. Micropropagated banana cv. Karpooravalli (ABB) plantlets drenched with water served as untreated control.

The photosynthetic rate (μmol CO₂ m⁻² s⁻¹), transpiration rate (mmol H₂O m⁻² s⁻¹), stomatal conductance (m mol m⁻²s), chlorophyll stability index, and relative water content (%) were recorded on the fully expanded leaves of banana plantlets using the LI-COR (LI-6400XT) Portable Photosynthesis System (PPS). The leaf chamber was the open type and measurements were taken at 10:00 h (IST). While taking measurements, photosynthetically active radiation of 1,500 μmol m⁻² s⁻¹ was maintained with an inbuilt light source. A temperature of 25 ± 5°C, relative humidity of 65 ± 5%, and reference carbon dioxide at a concentration of 380 mol CO₂ mol air⁻¹ were also maintained. Observations were recorded after the plant reached steady-state photosynthesis (15 days after transferring to the greenhouse).

The experiment consisted of six treatments, viz., B. velezensis YEBBR6, B. albus YEBN2, Achromobacter xylosoxidans YEBRH5, Beijerinckia fluminensis YEBRT4, Acinetobacter refrigeratoris YEBFR1, and Bacillus endophyticus YEBFL6, replicated three times with 10 plantlets per replication. Inoculum of Foc KP was multiplied in quarter-strength Potato Dextrose Broth (PDB) incubated for 7 days at 28 ± 2°C. Conidial load in the culture broth was assessed using a hemocytometer. Concentration of conidia was adjusted to 1 × 10⁶ spores per ml (Catambacan and Cumagun, 2021). Biohardened micropropagated banana cv. Karpooravalli (ABB; Pisang Awak) plantlets with different bacterial endophytes were challenged with conidial suspension of Foc KP. Simultaneously, untreated control was also maintained. Banana plantlets biohardened with different bacterial endophytes challenged with Foc KP were watered regularly until saturation of soil moisture. Eight weeks after inoculation of Foc KP, the degree of disease severity was assessed. It was assessed based on leaf yellowing using a modified disease rating scale (Dita et al., 2014). The intensity of rhizome discoloration was scored by examining the longitudinal section of the pseudostem to the rhizome using a modified rating scale (Carlier et al., 2003).

Percent disease severity was calculated as:

One centimeter of banana root bits were fixed in 2.5% glutaraldehyde and 2.5% formaldehyde prepared in 0.1 M phosphate buffer of pH 7.2 (Liu et al., 2014) for 24 h at 4°C. Samples were exposed to a mild vacuum so as to ensure rapid infiltration of the fixative into the specimens. The fixed specimens were washed three times with 0.1 M phosphate buffer (pH 7.2) by incubation at 23°C for 30 min each time. The specimens were infiltrated overnight with 30% glycerol in water as a cryoprotectant. Later, specimens were frozen in liquid nitrogen, then cross-sectioned with a sterile scalpel. Later, specimens were post-fixed in 1% OsO4 in 0.1 M phosphate buffer (pH 7.2) for 1 h. Subsequently, samples were washed three times in deionized water and incubated for 30 min for each wash. Next, specimens were dehydrated in 100% ethanol and air-dried to remove the ethanol residues. The dried specimens were mounted on aluminum stubs and sputter-coated with gold. SEM photographs pertaining to colonization of the rhizoplane by B. velezensis YEBBR6 and Foc KP and hyperparasitic interaction of B. velezensis YEBBR6 on Foc KP were documented with Hitachi S-3500N 143 SEM at 15 kV.

Total RNA was extracted from the roots of banana biohardened with B. velezensis YEBBR6 challenged with Foc KP using Trizol (Sigma Aldrich) at 0 h, 24 h, 48 h, and 72 h after Foc KP inoculation (Chomczynski and Sacchi, 1987). Similarly, RNA was extracted from untreated healthy control, B. velezensis YEBBR6 alone, and Foc KP-inoculated control. RNA extracted from different treatments was made up to 3,000 ng/l using Nanodrop (Eppendorf Make, Germany). According to the manufacturer's protocol, respective RNA from different treatments was digested using DNase I (Sigma Aldrich, USA). Further, the quality of RNA was determined by measuring the absorbance value at an A260/A280 ratio. RNA was converted to cDNA with the ThermoFischer Scientific-RevertAid First Strand cDNA Synthesis Kit (cat. # K1622). A ratio of 1.8 + 0.2 indicated the best quality of nucleic acid. The cDNA was diluted 10-fold and used for qRT-PCR analysis. It was performed in a BIO-RAD CFX manager system. The reaction mixture for qRT-PCR comprised 3 μl of cDNA template, 10 μl of SYBR Green master mix (KAPA SYBR@FAST for LightCycler 480, Cat-KK4610), 0.8 μl of 10 μM forward primer, and 0.8 μl of 10 μM reverse primer. The final volume was made up to 20 μl using nuclease free water. The PCR program included denaturation at 95°C for 10 min, amplification for 40 cycles at 95°C for 30 s, 60°C for 30 s, and 72°C for 30 s. It was followed by standard melting temperature analysis. The major defense gene transcripts assessed for the induction of resistance response against Foc KP infection were WRKY33, mitogen-activated protein kinase (MAPK), chitin elicitor receptor kinase (CERK 1), lipoxygenase (LOX), and phenylalanine ammonia lyase (PAL). For each defense gene expression study, three biological replicates and two technical replicates were maintained throughout the study. The fold changes in gene expression were calculated using the formula ΔΔCt = Δ Ct sample– Δ Ct reference. The relative fold changes in the transcript level were represented graphically by converting the ΔΔCT value to 2^−ΔΔCt (Livak and Schmittgen, 2001). Statistical analysis for relative fold change was performed using TIBCO Spotfire Analyst version 7.11.1.

Defense-related proteins in banana cv. Karpooravalli (ABB) plantlets that had a predominant role against Foc KP infection were further investigated for their interacting partners using the STRING database (Szklarczyk et al., 2021). Protein sequences of WRKY33, MAPK, CERK1, PAL, and LOX were used as a query against the double haploid Pahang (Musa acuminata) banana genome to understand the interacting protein partners. Information regarding interacting partners was obtained based on text mining, experiments, databases, co-expression, neighborhood, gene fusion, and co-occurrence results. Query protein and their interacting protein domain information in addition to the number of interacting partners belonging to a particular domain in the network were analyzed. A degree-sorted network was constructed using transcription factor, defense related genes, and their interacting proteins. Protein domains of all the interacting partners were analyzed to understand the functional relevance and their association with query protein. A medium confidence level of 0.40 and minimum of 10 interactors were used as parameters for the construction of the network.

Analysis of protein-protein interaction network between WRKY33, MAPK, CERK1, PAL, and LOX proteins was obtained using the STRING database. It was exported to Cytoscape 3.9 version for analysis. Using STRING and enrichment map applications in Cytoscape, STRING enrichment was performed and a map was constructed. Functional enrichment was performed by merging all the five proteins with their interacting partners.

Amplification of internal transcribed spacers of eight different isolates of Foc for ITS 1 and ITS 4 regions with specific primers yielded an expected amplicon size of 560 bp. Nucleotide sequences of amplicons pertaining to eight different isolates confirmed the identity of Foc. The sequences subjected to multiple alignments were submitted to GenBank and were provided with accession numbers, viz., Foc KP MW436477, Foc KP MW 436476, Foc NP1 MW436482, Foc NP2 MW436483, Foc KP2 MW436485, Foc RS2 MW436581, Foc RS3 MW436484, and Foc RS1 MW43658. Phylogenetic analysis of Foc isolates revealed the presence of three major clusters. Cluster 1 comprised Foc KP MW 436476, Foc KP MW436477, and Foc NP1 MW436482. Cluster 2 had five isolates including Foc NP2 MW436483, Foc KP2 MW436485, Foc RS2 MW436581, Foc RS3 MW436484, and Foc RS1 MW43658. Cluster 3 comprised Fusarium oxysporum f.sp. lycopersici (Fol) as an out group (Supplementary Figure S1). Based on the pathogenicity study, Foc KP was most virulent compared to other isolates. Sequencing of Six13c for the isolate Foc KP confirmed the presence of race 4 bearing the accession number MW323435. It had similarity with Foc isolate BRIP62892 bearing the accession number KX435021 and also matched with VCG group 0122 pertaining to the Cavendish group (AAA) of Philippines origin. Hence, Foc KP was used throughout the study.

Bacillus velezensis YEBBR6 inhibited the mycelial growth of Foc KP up to 63% compared to untreated control (Supplementary Figure S2). Other bacterial endophytes, B. albus YEBN2, A. xylosoxidans YEBRH2, Beijerinckia fluminensis YEBRT4, Acinetobacter refrigeratoris YEBFR1, and B. endophyticus YEBFL6 were not equally effective as B. velezensis YEBBR6 in the suppression of mycelial growth of Foc KP in vitro.

Fifteen bioactive metabolites were characterized through GC/MS from the zone of inhibition during the ditrophic interaction between B. velezenzis YEBBR6 and Foc KP. They included dihydro acridine, nonanol, hexadecanoic acid, oleic acid, clindamycin, 5-hydroxymethylfurfural, azulene, 3 amino-4 hydroxy phenyl sulfone, campholic acid, procyclidine, linoelaedic acid, acetylvaleryl, allobarbital, aminomorpholine, and 3-thiazolidine carboxamide (Supplementary Figure S3 and Supplementary Table S1). Six bioactive metabolites were produced by B. velezensis YEBBR6 in the absence of Foc KP. They were identified as 1,3-propanediol, clindamycin, 4H-pyran, pentanoic acid, acetaldehyde, and hexadecanoic acid (Supplementary Figure S4 and Supplementary Table S2). Bioactive metabolites produced by Foc KP were identified as 2-propen-1-ol, 3H-pyrazol-3-one, hexanoic acid, cyclohexan, 1H-azonine, trioxsalen, and butanamide (Supplementary Figure S5 and Supplementary Table S3). The Venn diagram of differentially expressed bioactive metabolites during the interaction of bacterial endophyte B. velezensis YEBBR6, either with Foc KP or without Foc KP, and Foc KP alone revealed that B. velezensis YEBBR6 produced six bioactive metabolites. But, during the interaction of B. velezensis YEBBR6 with Foc KP, 15 bioactive metabolites were produced. Comparison of bioactive metabolites between B. velezensis YEBBR6 alone and B. velezensis YEBBR6 with Foc KP revealed the production of two bioactive metabolites in common, viz., hexadecanoic acid and clindamycin (Figure 1).

Micropropagated banana plantlets treated with B. velezensis YEBBR6, B. albus YEBN2, A. xylosoxidans YEBRH2, Beijerinckia fluminensis YEBRT4, Acinetobacter refrigeratoris YEBFR1, and B. endophyticus YEBFL6 increased plant growth parameters (Figure 2). Among the bacterial endophytes, B. velezensis YEBBR6 significantly increased pseudostem height (38.63 cm), pseudostem width (5.9 cm), leaf area (1.290 m²), root length (54.75 cm), and leaf emergence rate (Phyllochron) (6.0), compared to 11.21 cm, 3.2 cm, 0.09 m², 36.1 cm, and 12.0 in control plants, respectively (Figure 3A and Supplementary Figures S6–S10). B. velezensis YEBBR6-treated banana plants also enhanced the photosynthetic rate to 24.61 mol m⁻² s⁻¹ against 13.09 mol m⁻² s⁻¹ in non-bacterized control plants. Further, transpiration rate and stomatal conductance in B. velezensis-treated banana plants was 3.95 mol.cm⁻² s⁻¹, 1.01 mmol m⁻² s, while it was 2.64 mol.cm⁻² s⁻¹, 0.06 mmol m⁻² s in non-bacterized control plants. Biohardened banana plants had higher levels of total chlorophyll (1.29 mg/g¹), chlorophyll stability index (88.6%), and relative water content (83.3%) than non-bacterized control plants (Supplementary Table S4).

Banana plantlets biohardened with B. velezensis YEBBR6, B. albus YEBN2, Achromobacter xylosoxidans YEBRH2, Beijerinckia fluminensis YEBRT4, Acinetobacter refrigeratoris YEBFR1, and B. endophyticus YEBFL6 against Foc KP indicated that banana plantlets biohardened with B. velezensis at 10 ml/plant had zero incidence of wilt. But, the banana plantlets biohardened with B. albus and Acinetobacter refrigeratoris had 10 % wilt against 30 % in banana plants biohardened with B. endophyticus. However, 100 % wilt incidence was observed in inoculated control (Figure 3B and Supplementary Table S5). As B. velezensis YEBBR6 was effective in the suppression of Fusarium wilt, it was subsequently used for further studies.

Analysis of the roots of banana plantlets biohardened with B. velezensis YEBBR6 using scanning electron microscopy confirmed the colonization of bacterial cells on the root surface. Agglomerates of bacterial cells resulted in the formation of biofilm on the root surface (Figure 4A). Besides, the roots of banana plantlets biohardened with B. velezensis YEBBR6 followed by challenge inoculation with Foc KP witnessed hyperparasitism of Foc KP mycelium and microconidia by the bacterial cells of B. velezensis (Figures 4B,C). However, in pathogen-inoculated control, proliferation of microconidia was noticed along the root zone of banana plantlets (Figure 4D).

Biohardening of banana plantlets with B. velezensis YEBBR6 challenged with or without Foc KP altered the expression of the WRKY transcription factor, MAPK, chitin elicitor receptor kinase, lipoxygense, and PAL genes responsible for plant defense. Irrespective of different treatments, the WRKY 33 transcript was downregulated in all the treatments at 0 h. The transcription rate of the WRKY 33 gene was affected immediately after inoculation with Foc KP and in biohardened plants. The level of the transcript in Foc KP-inoculated control increased after 24 h and declined after 48 and 72 h. However, the transcript of WRKY 33 was upregulated in banana plantlets biohardened with B. velezensis at 24, 48, and 72 h after treatment. Interestingly, a 1.87-fold increase of the WRKY 33 gene transcript was observed in banana plantlets biohardened with B. velezensis YEBBR6 challenged with Foc KP. But, in untreated healthy control, upregulation of WRKY 33 (0.33-fold) was noticed only at 72 h (Figure 5A).

The expression level of MAPK transcripts varied between different treatments. Banana plantlets biohardened with B. velezensis YEBBR6 challenged with Foc KP increased the expression of MAPK transcripts up to 2.32-fold after 72 h in comparison with biohardened plants that were not challenged with Foc KP. Besides, only a 0.69-fold change of the WRKY 33 transcript was noticed in pathogen-inoculated control after 48 h and it decreased further after 72 h of inoculation (Figure 5B).

Induction of the chitin elicitor receptor kinase (CERK 1) transcript associated with innate immune response was initiated after 24 h of inoculation with pathogen Foc KP. However, upregulation was more pronounced after 72 h in plantlets biohardened with B. velezensis YEBBR6 coupled with challenge inoculation of Foc KP. The expression of the CERK1 transcript was 2.2-fold higher than in biohardened plants that had not been inoculated with Foc KP. Furthermore, the expression level of CERK1 transcripts was reduced in untreated healthy control compared to pathogen-inoculated and biohardened plantlets challenged with Foc KP (Figure 5C).

Lipoxygenase (LOX) is a key enzyme involved in the induction of the immune response in plants, therefore attempts were made to understand the ability of B. velezensis YEBBR6 to modulate immune response against Foc KP. Biohardened plants challenged with Foc KP increased the transcript level of fatty acid dioxygenase LOX up to 2.6-fold after 72 h of inoculation. The expression of the LOX transcript was 1.6-fold greater in B. velezensis YEBBR6 and was downregulated after 72 h. Comparison on the expression of the LOX transcript in inoculated control reflected a 1.7-fold change after 72 h, while in healthy control, there was no increase in the LOX gene transcripts (Figure 5D).

Assessing the expression of phenylalanine ammonia lyase (PAL) revealed a significant increase of PAL transcripts in B. velezensis YEBBR6 biohardened banana plantlets challenged with Foc KP. After 48 h of challenge inoculation with Foc KP, the level of induction of the PAL transcript was 1.9 times higher than other treatments. After 48 h, the PAL transcript in Foc KP-inoculated control was observed only up to 0.6-fold, and decreased after 72 h. However, the activity of PAL was downregulated in healthy control (Figure 5E).

STRING analysis was performed to understand the functional association of defense-related genes with other proteins and their conserved domains. From our analysis, most of the interacting proteins for each query protein had similar conserved domains and were clustered together in the network. Protein sequences of defense-related genes were used as input to retrieve the interacting partners based on the double haploid Pahang (Musa acuminata) banana genome. Details on the query protein and their interacting protein domain and number of interacting partners in the network are listed in Supplementary Table S6. The degree-sorted network constructed using transcription factor and defense-related genes and their interacting proteins are illustrated in Figure 6.

Protein-protein interaction of WRKY transcription factor indicated the involvement of a large protein family with diverse functions. They were expressed in response to pathogens, elicitors, and defense-related phytohormones such as salicylic acid (SA) or jasmonic acid (JA). Protein-protein interaction of MAPK revealed that it belongs to the kinase family and interacted with G protein domains and protein tyrosine phosphatase domains similar to chitin elicitor receptor kinase 1-like protein. Besides, functional domains of five out of ten interacting proteins were uncharacterized in the given network. CERK 1 protein pertaining to the protein kinase family protected plants at multiple layers against invading pathogens by conferring PAMP-triggered immunity (PTI) or effector-triggered immunity (ETI). Regardless, CERK proteins interacted with the LysM domain and guanine nucleotide-binding domain (Supplementary Table S6). Protein interaction of LOX indicated that it interacted with proteins containing LOX, phospholipase, and PLAT/LH2 conserved domains. These proteins were associated with membrane or lipid-associated proteins. Similarly, protein-protein interaction of PAL reflected that chalcone synthase (CHS) was essential for the formation of 4,2′, 4′,6′-tetrahydroxychalcone, responsible for the biosynthesis of anthocyanin pigments, anti-microbial phytoalexins, and flavonoid inducers of Rhizobium nodulation genes. Nonetheless, stilbene synthases (STSs) occurred in a limited number of unrelated plants and synthesized the backbone of stilbene phytoalexins that have antifungal properties and contribute to pathogen defense. Despite the other genes, PAL also induced defense against pathogens by triggering other enzymes linked with different pathways.

Merging of WRKY33, MAPK, CERK1, PAL, and LOX resulted in the formation of an enlarged network comprising of 52 nodes. Three isoforms of guanine nucleotide-binding proteins interacted commonly with CERK and MAPK protein targets. Apart from this, there was no common interacting proteins between the five protein targets. Functional enrichment analysis resulted in the formation of more than five clusters. A heat cluster map was generated based on the functional relevance of each of the proteins. In the enrichment figure, each cluster was linked with the query protein to indicate the commonality and the difference observed with the five query proteins and their interaction partners. Based on the enrichment map, domain, and pathway information of five proteins, CERK1 was linked with the LysM domain and guanine nucleotide-binding protein domain. LOX was linked with linoleic acid, alpha-linolenic acid, arachidonic acid metabolism, ether lipid metabolism, glycerophospholipid metabolism, lipoxygenase, phospholipase, and PLAT/LH2 conserved domains. The MAPK gene was associated with G protein domains and protein tyrosine phosphatase domains. While, PAL was coordinated with AMP binding, chalcone synthase, polyketide synthase, and thiolase-like domains. The WRKY transcription factor was associated with the VQ5 domain and WD40 repeat containing domain (Figure 7).