Identification of non-ribosomal peptide synthetase in Ganoderma boninense Pat. that was expressed during the interaction with oil palm

Basal stem rot (BSR) of oil palm is a disastrous disease caused by a white-rot fungus Ganoderma boninense Pat. Non-ribosomal peptides (NRPs) synthesized by non-ribosomal peptide synthetases (NRPSs) are a group of secondary metabolites that act as fungal virulent factors during pathogenesis in the host. In this study, we aimed to isolate NRPS gene of G. boninense strain UPMGB001 and investigate the role of this gene during G. boninense-oil palm interaction. The isolated NRPS DNA fragment of 8322 bp was used to predict the putative peptide sequence of different domains and showed similarity with G. sinense (85%) at conserved motifs of three main NRPS domains. Phylogenetic analysis of NRPS peptide sequences demonstrated that NRPS of G. boninense belongs to the type VI siderophore family. The roots of 6-month-old oil palm seedlings were artificially inoculated for studying NRPS gene expression and disease severity in the greenhouse. The correlation between high disease severity (50%) and high expression (67-fold) of G. boninense NRPS gene at 4 months after inoculation and above indicated that this gene played a significant role in the advancement of BSR disease. Overall, these findings increase our knowledge on the gene structure of NRPS in G. boninense and its involvement in BSR pathogenesis as an effector gene.

www.nature.com/scientificreports/ non-proteinogenic amino acids in their D-and L-configurations by multi-domain, multi-modular enzymes 13 . l-δ-(α-Aminoadipoyl)-l-cysteinyl-d-valine synthetase (ACVS), enniatin synthetase, cyclosporine synthetase, HC-toxin synthetase (HTS1), and bassianolide synthetase are some examples of NRPSs of plant pathogenic fungi 12,14 . The genome of many pathogenic fungi was found to contain NRPS genes which are now available on publicly available databases such as https:// mycoc osm. jgi. doe. gov/ pages/ sm-clust ers-summa ry. jsf 15 . Elicitors act as bioactive signals that are recognized by pattern recognition receptors (PRR) of plants. In the first level of inducible defense strategy against pathogens, PRRs, which initiate defense responses known as (pathogen-associated molecular patterns) PAMP-triggered immunity (PTI), caused accumulation of reactive oxygen species (ROS) in the plasma membrane 16 . For example, PebC1 (as an elicitor) caused a defense reaction in tomato against Botrytis cinerea 17 . The pathogen invasion is successful by secretion of various race-specific elicitors (effectors) directly into the plant cells, which inhibits PTI activation in the plant. In the second plant protection strategy, effector-triggered immunity (ETI) leads to expressed defense genes for the prevention of progression of the disease avirulence (avr) genes, from the aggressive pathogen, encode effectors corresponding to the resistance (R) genes in its specific host. If the fungal strain has a virulent form of avr effector, R protein is no longer able to detect the pathogen, so ETI process would be stopped 16 .
Virulent fungi destroy their host plants by various strategies including the secretion of SMs such as toxins. Some necrotrophic fungi have specific effector proteins which are active against their specific hosts, known as host-specific toxins (HSTs). HST secretion is pivotal for enhancing the pathogenicity and virulence in the host 16 . The target of toxin effectors in necrotrophic fungi is one of the main signaling or regulatory pathway that triggers gene-mediated resistance or the decrease in the regulation of defense enzymes; thereby, reducing the host sensitivity boosts against fungal attack 18 . Effectors are produced by virulent pathogens to enhance infection in the host plants 19 . To illustrate, HTS1 is an effector produced by Cochliobolus carbonum Race 1 during the pathogenicity in the susceptible genotype of hmhm maize 20 . This fungal effector disturbed the regulatory system of the plant by blocking calmodulin signalling or histone acetylation 18 . In contrast, Hm1 and Hm2 genes in maize are responsible for encoding carbonyl reductases that function to suppress the toxin; however, they are not successful 20 . A study on avirulence effector genes in Magnaporthe oryzae (the causal agent of rice blast) interacting with rice revealed that ACV1 gene (hybrid polyketide synthase (PKS)-NRPS genes) acts against the R gene (Pi33) in rice. Although the exact determination of the function of these metabolites was complicated, ACV1 expression was simultaneously linked to the appressorium-mediated penetration in the cuticle of rice 21,22 . Another example of a toxin effector produced by NRPS gene is AM-toxin that encoded by AMT1 gene in the apple pathotype of Alternaria alternata during the interaction with the host 23 . AM-toxin has two attack sites including disordering in the chloroplast's membrane by a reduction of chlorophyll substance and inhibition of photosynthetic CO 2 and damaging effect in the plasma layer 24 .
A canonical module of NRPS enzyme is composed of three different standard domains: adenylation (A), thiolation (T) or peptidyl carrier protein (PCP), and condensation/peptide-bond formation (C) which is recognized as one NRPS module 25 . Adenylate forming domain (AFD_class_I) superfamily belongs to the group of acyl-and aryl-CoA ligases, adenylation domain of NRPS 12 . The adenylate-forming enzymes catalyze an ATP-dependent in two steps. In the first step, a carboxylate substrate will be activated as an adenylate and in the next step, the carboxylate will be transferred to the pantetheine group which is a coenzyme A or an acyl-carrier protein 12 . A set of conserved sequence motifs (A1-A10) has structural and functional roles for diagnosis and binding of the substrates, which is specific in A domain 25,26 . Therefore, the A domain has a fundamental impact on the type of NRPs synthesized by NRPS and its function during pathogenicity 27 . Adenylation domain is responsible for recognition, activation and incorporation of an amino acid onto the T domain or PCP with high conserved serine which carries a 4′-phosphopantetheine [Ppant] cofactor and a thioester bond is made. The Ppant arms have to penetrate from opposite sides of the C domain to reach the conserved active-site motif HHxxxDG. The second histidine in this motif has been suggested to work as the general base to promote a nucleophilic attack of the a-amino group 28 . C domain catalyzes a condensation reaction to create peptide bonds in non-ribosomal peptide biosynthesis. This domain is located on the carboxyl side of a pp-binding domain. Therefore, C domain plays an essential role in the formation and elongation of the peptide chain in NRPSs 29 .
Three classifications are proposed for the known NRPS biosynthesis: linear (type A), iterative (type B) and non-linear (type C). Linear type as in ACV synthetase, peptaibol synthetases, and cyclosporin synthetases was identified. In an iterative type such as enniatin synthesis, the modules and domains are repeated during the building of a compound. The construction of non-linear NRPSs is more complex and varies from those of linear type. These unusual structures are unpredictable and found in many bacterial NRPSs while most fungal NRPSs are described as linear and iterative types 12 .
Several tools such as, ClustScan 30 , CLUSEAN 31 , SBSPKS 32 , SMURF 33 , PRISM 34 , SeMPI v2 35 and antiSMASH (Antibiotics and secondary metabolite analysis shell) 36 have been developed to identify and analyze the enzymatic domains in multi-modular PKSs and/or NRPSs which are the essential enzymes for the production of clinically crucial secondary metabolites. These advanced tools predict the feature of substrates which have vital roles in biosynthetic steps and the chemical structure of the final bioactive compounds based on the genome sequence 36 .
The objectives of this study were (1) to isolate, identify and characterize NRPS gene of G. boninense, (2) to determine the relationships among various fungal NRPSs, and (3) to investigate the role of NRPS in the pathogenicity of G. boninense during its interaction with oil palm.

Results
Identification of NRPS domains. With the aid of antiSMASH and SeMPI v2, contigs containing putative NRPS biosynthesis gene clusters were identified in both G. boninense strains NJ3 and G3 whole genome. One putative NRPS biosynthesis gene cluster was detected in both strains. NRPS gene in strain NJ3 was predicted in Since there was a continuity in the NRPS sequences by the antiSMASH results and the predicted NRPS regions by SeMPI v2 were located in these ranges, the NRPS regions were selected based on the antiSMASH result for the next steps and the nucleotide locations of different domains [A-C-T] were characterized by using NCBI's Conserved Domain Database (CDD) in each strain. The length of NRPS gene in the related cluster in both strains was about 14,000 bp and NRPS domains (A-T-C) region started from 6000 bp and continued till 14,000 bp.
Confirmation of NRPS domains. Seventeen PCR amplicons were sent for nucleotide sequencing. In this analysis, the length of all amplicons was corresponding to the predicted size. Overall, the length of 8322 bp nucleotides was detected using PCR as NRPS region (GBNRPS) in strain UPMGB001 and 2419 aa putative protein were predicted using AUGUSTUS software. The detected DNA region of NRPS in the UPMGB001 strain was aligned with both G. boninense strains NJ3 and G3 for more confidence (more details were included in Supplementary Tables S7 and S8). AUGUSTUS prediction, GBNRPS contains 14 intron regions. The nucleotide sequence of GBNRPS is available in NCBI with accession number MT675190. NCBI's CDD-search found superfamilies associated with the NRPS region including AFD_class_I, AMP-binding, Condensation and pp-binding ( Fig. 1). Moreover, A_NRPS_SidN3_like represented in Fig. 1, has been identified as the specific structure in the A domain of NRPS (the third adenylation domain of NRPS SidN) that led to making a specific amino acid that belonged to siderophore-synthesizing NRPS 37 .

Structural analyses of NRPS domains.
Multiple sequence alignment (MSA) among NRPS protein sequences of some basidiomycetes and strain UPMGB001 represented conserved motifs of three main NRPS domains (as shown in Fig. 2). A set of conserved sequence motifs (A1-A10) was detected for A domains according to codes reported by Eisfeld (2009) in Table 1. The side-chain of signature non-ribosomal code of SidNA3 (DXXXXXXXXK) was recognized in the A domain of basidiomycetes 25,38 ; DTSVGISKRK sequence code was detected in UPMGB001 strain. UPMGB001 presented a similar conserved region in A1 (conserved motif: LTNAEF), A2 (LRAGLLLVPID), A3 (LAYILYTSGTTGTPKGC), A4 (FDVHIAE), A5 (NFYGPSE), A6 (GELVVEGPLVGRGYIG), A7 (YRTGDLVR), A8 (GRIDTQIKLRGVRIESEGISS), A9 (LASYMRP) and A10 (NGKADA) with some variances in conserved motif codes (as demonstrated in Fig. 2). The core motif of T is the adenylation A domain of siderophore-synthesizing NRPS 37 , EntF is a NRPS synthetase component F that is responsible for secondary metabolites biosynthesis, catabolism and transport 85 . PP_binding (phosphopantetheine attachment site) is a 4′-phosphopantetheine prosthetic group is linked by a serine. The role of this prosthetic group as a 'swinging arm' for the connection of activated fatty acid and amino-acid groups to Condensation domain 29 . Condensation catalyzes a condensation process to form peptide bonds in NRP biosynthesis 86 . AFD_Class_I superfamily, Condensation superfamily, pp-binding superfamily.     13.97% beta-sheet, 4.13% beta-turn, and 40.26% random coil. All the predicted models overlapped with their templates showing alpha helices and beta sheets and gave a conformational pattern similar to that of the known NRPSs. Also, ligand ATP which is important for the catalytic activity of  Phylogenetic analysis. In this study, we performed the phylogenetic analysis of A domain that has the highest conserved domain among the other main domains (T and C domains) 11 , and also all domains of NRPS with the intent to deduce concordance between them. As presented in Fig. 4, bootstrap values at the inner nodes did not show any significant variations. Comparison between the phylogenetic dendrogram reported by Brandenburger et al. 40 and the results of this study using the same sequences showed G. boninense is placed in the same clade in both studies (Fig. 4B).
Assessment of disease severity. All the oil palm seedlings were found to be successfully infected by G.
boninense with the gradual changes in the appearance of leaves toward necrotic and growth of white mycelial at the base of stems 2 months after artificial inoculation. The un-inoculated plants (Control) did not show any sign of disease symptoms. Observation of symptoms in the infected oil palms by G. boninense was examined at every sampling interval [0, 1, 2, 4, 6 months after inoculation (MAI)] according to class disease based on a disease index value of 0-4 as represented in Fig. 5. The infected seedlings were observed with disease severity index (DSI) (%) 1 . The inoculated seedlings showed infection symptoms from 2 MAI (25%) and increased to 50% at 4 MAI. Finally, the inoculated seedlings demonstrated the highest disease severity at 6 MAI with 58.5% as illustrated in Fig. 6a. NRPS gene expression in G. boninense. The gene expression patterns of NRPS were profiled in G.
boninense from challenged root tissues of oil palm seedlings at 1, 2, 3 and 4 MAI. These intervals were selected as this time frame was crucial for infection to occur and therefore, was able to determine the involvement of NRPS    (Fig. 6b).

Discussion
NRPSs are responsible for producing secondary metabolites and their roles had been identified as virulent factors in different species of ascomycete and basidiomycete fungi 12 . The validated genomic data of G. boninense available in NCBI helped to identify the cluster of NRPS using antiSMASH and SeMPI v2 programs and led to the detection of one genomic sequence that is potentially responsible for the production of NRP. This finding is similar to the prediction of only one putative NRPS cluster in G. sinense using antiSMASH which had a similar structure to that obtained in G. lucidum 42 . The amplification of G. boninense NRPS gene (8322 bp nucleotides) revealed the detection of three major superfamilies including AFD_class_I, phosphopantetheine attachment site (pp-binding) and C domain (Fig. 1). However, there were some variations in the DNA sequences among the three strains, NJ3, G3 and UPMGB001 (Supplementary file Table S7) but all the conserved motif domains (A, T and C) have remained stable among them. These conserved motifs in these three strains have the highest similarity compared to other basidiomycetes (Fig. 2, Table 1). The comparison of NRPS conserved motifs in basidiomycete fungi displays much more variations than those of bacterial origin, even within the same species 12 . However, mutation among species is expected, all ten A motifs in G. boninense were detected with minor variances in conserved motif codes. The low sequence similarity of the A domain motif in G. boninense compared to that of other basidiomycete fungi could suggest its novelty as indicated in an earlier study on the partial cDNA of 853 nucleotides encoding NRPS from G. boninense where core motifs A2, A3 and A5 were detected and showed some variations from other species 43 . As illustrated in Fig. 2 and Table 1, the most diverse variations among A motifs based on the nonribosomal code of conserved motifs for basidiomycetes belong to A1 ((I/L/S/M)(T/S)(Y/F/N)(G/E/V/S/L/A)(E/D)(L/W/F)), and the code LTNAEF have been identified similarly in Ganoderma species in this study. Nevertheless, there was a small variation in A2, A3 and A10 between G. sinense and G. boninense. Interestingly, a minor diversity in A3 showed among UPMGB001 and G3 (LAYMLYTSGTTG) with NJ3 (LSYMLYTSGTTG) that demonstrated some mutations in the same species among different strains. Most variations in A6 motif belong to Fomitiporia mediterranea (GELLVGGRLVGRGY LN). The highest similarity among A motifs belongs to A5, A7 and A9 with NFYGPSE, YRTGDLVR and LASYMRP codes, respectively. While Stereum hirsutum has a variance in only one amino acid (NFYGPTE) in A5, Rhizopogon vesiculosus has two distinct amino acids in A7 motif code (YHTGDLAR), and Coniophora puteana has a single variation in the third amino acid in A9 motif code (LANYMRP). On the contrary, A8 code (GRIDTQ(I/V) K(L/V) RGVRIE(S/A/V) (E/D) (V/I) S(S/A)) revealed high similarity among basidiomycetes in this study. Under this evaluation, the order of similarity of core A code in G. boninense with other basidiomycete fungi from highest to lowest includes A9 and A5 > A7 > A8 > A6 > A4 > A3 > A10 > A2 > A1.
Based on findings in this study, the core motif of T domain was detected in G. boninense (Fig. 1). This core motif includes extremely conserved serine residue that the hydroxyl group operates as an attachment site for Ppant 26,29 . The MSA among UPMGB001 and other basidiomycetes showed a conserved domain GIDS, excluding R. vesiculosus and F. mediterranea with a difference in the second amino acid in this conserved region code (GVDS). As illustrated in Fig. 2, C domain with a conserved active-site motif was another domain found in this   44,45 . Besides, siderophores are secondary metabolites produced by NRPSs that are involved in iron uptake. The remarkable role of siderophores was proven in host invasion. They have, potentially, dual functions: they act not only as pathogenesis effectors but defense elicitors in host infection processes. These two functions are also recognized in some protein effectors released by fungal and bacterial pathogens 19 . The role of siderophore in the pathogenicity process by G. boninense in oil palm is unknown. Based on previous studies, siderophores synthesized by NRPS genes are crucial effectors to suppress the immune mechanism of the host plant 19 . A large number of investigations have revealed that lack of extracellular siderophores has a prominent impact on the reduction of the pathogenicity against the host 46 . The release of siderophores as toxic effectors in pathogenic fungi led to inhibiting ROS in the host attacked 47 . Detoxification of ROS is an essential function in the infection process via pathogen, such as the infection of citrus by A. alternata 16 . Studies on F. gramineasum, A. brassiciola, C. miyabeanus had demonstrated siderophores played a crucial role in the pathogenicity of wheat, Arabidopsis, and rice, respectively 48 . Consequently, a better perception of the mechanism through which siderophore impacts the oil palm immunity, and identification of interactive proteins from oil palm may contribute to the development of new oil palm genotypes resistant against G. boninense.
Siderophore producing NRPSs dependent on the domain structure are classified into types I-VI 49 . Diverse enzymes act as catalysts leading to siderophore synthesis with monomeric substrates. These enzymes are monooxygenase-mediated in two steps which consist of hydroxylation and acylation of the N-5 nitrogen atom of l-ornithine via an acyltransferase. In contrast to ascomycetes, information about siderophore production by basidiomycetes are widely unrecognized including genetic and enzymatic necessities for siderophore production. Only two reports on NRPS-related genes in basidiomycetes were found; ferrichrome and ferrichrome A biosynthesis in Ustilago maydis 50 , and ferrichrome A biosynthesis genes in Omphalotus olearius 51 .
According to prior studies, the specific putative NRPS gene (CsNPS2) in the genome of white-rot model species Ceriporiopsis subvermispora (or Gelatoporia subvermispora) is a member of an extensively distributed but formerly unknown class (type VI) of fungal siderophore synthetases 40 . A study on the type of siderophoreproducing NRPSs in C. subvermispora as a white-rot basidiomycete has shown that it can be a model for a myriad of type VI basidiomycete NRPSs 40 . Furthermore, the similarity of the this study with that of Brandenburger et al. 40 confirmed NRPS gene in G. boninense belongs to the type VI siderophore family. Phylogenic analysis displayed this type of siderophore belongs to the iterative NRPS template 40 . Specificity-conferring residues of SidN3 (SidNA3) of fungi were used in the phylogenic analysis. SidN gene is responsible for encoding the typical siderophore type, known as epichloenin. Based on the studies to date, SidNA3 has been detected only in the eukaryotic NRPS domain 52,53 . Although there is not enough information about the structure of siderophore in G. boninense, SidNA3 was detected in NRPS sequence domain in UPMGB001 strain (Figs. 1, 2). There are some reports about siderophore production in Ganoderma sp. For instance, siderophore synthesis has been studied in G. boninense by Azurol S (CAS) agar plate assay 43 , siderophores produced by Ganoderma sp. was screened using both culture supernatants and solid medium by the universal CAS agar plate assay and the result showed that siderophore synthesized by Ganoderma sp. is under the hydroxymate type 54 (there are three main classes of siderophores include hydroxamate, catecholate and carboxylates 55 ).
Ganoderma boninense is the causal agent of BSR disease, however, the precise infection pathway of G. boninense is still unraveled 5 . Thus, attempts to identify the pathogenicity genes of G. boninense involved in the infection process on oil palm is crucial. NRPS gene was expressed as early as 1 MAI 56 , as demonstrated in our study. Lee et al. 56 studied the role of NRPS at the infection stage of C. heterostrophus during the plant-pathogen interaction on maize roots. Twelve NRPSs in the genome of this fungus have been detected, and their roles as virulent factors were investigated. Each of the NRPSs was eliminated one after another, and only NPS6 has shown to be a virulent factor to maize. The removal of partial or complete NPS6 caused a reduction in virulence and an increase in H 2 O 2 sensitivity 56 . Additionally, more investigations highlighted the role of extracellular siderophore synthesis in the expression of cell wall synthesis enzymes, whilst the loss of sid1 remarkably reduced the expression of genes encoding the enzymes that are responsible for cell wall synthesis. Appressoria and conidia in C. heterostrophus have a significant role in the penetration of host cell that are synthesized by sid1 and NPS6, therefore, deletion of them strictly disturbed the pathogenicity ability in the fungus 46 . Comprehensive research on 22 NRPS genes of T. virens showed seven of them upregulated when the mycelia of the fungus were in contact with the roots of maize, and they caused internal colonization in the root during the infection on the host 57 . Investigation on differential expression of 437 genes of Valsa mail, as a destructive pathogen of apple in China, revealed the expression of NRPS transcript occurred more than 100 times during the fungal infection and that NRPS was involved with the production of toxic SM during the pathogenicity in apple 58 .
The relative fold-changes of expression in different months of root tissue colonization were calculated by comparing the expressions of NRPS in the C domain (identified in this study) and the previous study on the A domain (identified by Wong et al. 43 ); the results showed the same trend in both studies. Consistent with our investigation in Fig. 6, NRPS gene expression showed an outstanding increase of 67 change at 4 MAI when DSI of infected seedlings was relatively high (50%). Infection of the oil palm roots by G. boninense causing basal stem tissue rot appeared as typical symptoms during 5 months of observation 59 as the highest DSI was observed at 6 MAI in this study (58.5%) and Wong et al. (62%). Wong and colleagues studied the expression level of NRPS A domain in G. boninense strain PER71 using reverse transcriptase-PCR. As a result, NRPS was detected at 1-4 MAI when no apparent disease symptoms observed in infected oil palm seedlings, while disease symptoms appeared at 4 MAI with 45% DSI and reached a peak at 6 MAI with 62% DSI 43 .
The findings of this study enhance our understanding of NRPS gene structure of G. boninense and its involvement in the disease progression of BSR as an effector gene. The correlation between high disease severity and high www.nature.com/scientificreports/ expression of NRPS gene indicated that this gene in G. boninense played a significant role in the advancement of BSR disease. Moreover, these results could lead to the development of an early detection system using NRPS region as a Ganoderma-specific biomarker for efficient management of BSR disease in oil palm.

Materials and methods
Fungal strain and culture conditions. Ganoderma  Characterization of the domains present in these genes was done using NCBI's CDD 65 . A putative NRPS gene which included at least three main conserved domains (A, T and C) was confirmed for the next steps 11 . The results obtained for G3 and NJ3 strains were also used for phylogenetic analysis.
Amplification of NRPS domains. The NRPS region identified by antiSMASH and SeMPI v2 was verified by using a PCR-based approach. DNA was extracted from UPMGB001 strain that was grown on a medium containing PDA using Cetyltrimethylammonium bromide (CTAB) DNA Extraction Protocol [66][67][68]  www.nature.com/scientificreports/ sia) for both disease assessment and NRPS expression studies. Seedlings were acclimatized for 1 month in the greenhouse before artificial inoculation. Fungal Inoculum Preparation. Rubber (Hevea brasiliensis) woodblocks (RWBs) of 6 cm × 6 cm × 6 cm were purchased from Huat Hing Sdn. Bhd. (Semenyih, Malaysia), and were prepared for artificial inoculation using the direct sitting technique 77 . RWBs were used as substrate carriers to supply nutrients for G. boninense under host-free conditions. Seven days old G. boninense cultured on MEA plate was cut into small pieces and equally dispersed on the autoclaved RWBs containing 50 mL of MEA added as supplementary nutrition for G. boninense. The inoculated RWBs were kept at room temperature (28 °C) for 4 weeks in darkness until full mycelial colonization. Fungal inoculum preparation was carried out according to Govender et al. 78 .
Disease severity index (DSI). Disease assessment with destructive sampling was arranged in inoculated and control plants at 0, 1, 2, 4, 6 MAI with completely randomized design (CRD). Fifteen seedlings were used for each group of inoculated and uninoculated plants. Disease progression of the inoculated seedlings was assessed at 0, 1, 2, 4, 6 MAI according to external disease symptoms and rated based on 0-4 disease severity value (the details were summarized in Supplementary Table S10). DSI was calculated using the formula shown below based on the number of plants showing that disease class per treatment 1 .

Sample preparation for gene expression observation. Seedlings were divided into two parts; twelve
Control plants [non-inoculated, RWBs without G. boninense] and twelve inoculated (RWBs with G. boninense) plants including three independent biological replicates per treatment. Inoculated and control plants were removed from their polybags at 1, 2, 3, 4 MAI and their root tissues were used for gene expression analysis. The harvested root samples were washed with distilled water, snap-freezed with liquid nitrogen and stored at − 80 °C for RNA extraction. The expression of NRPS of G. boninense and BSR disease severity in oil palm seedlings were used to observe the probable correlation between the two parameters.
Total RNA extraction and first-strand cDNA synthesis. Extraction of total RNA from the frozen root tissue samples (n = 3) was done using RNeasy Mini Kit (Qiagen, Germany) following the manufacturer's procedure. The total RNA concentration and quality were measured using NanoDrop ND-1000 UV-Vis Spectrophotometer (Thermo Fisher Scientific, USA) and analyzed by 1% gel electrophoresis. First-strand cDNA was synthesized from DNase I-treated RNA samples using the Omniscript Reverse Transcription Kit (Qiagen, German) according to instructions from the manufacturer. The cDNA was kept in RNAse free water (DEPC water) and stored at − 20 °C. Primer design. The conserved sequence of C domain (catcacgcgctgtacgacggg) was located between forward and reverse primers. Gene-specific primer of NRPS in this study was designed on the Condensation coding region of transcript sequence to evaluate NRPS expression. The primers were designed using Primer3 plus program http:// www. bioin forma tics. nl/ according to the length between 17 and 22 bp, GC content of 35 and 65% and melting temperature between 59 and 63 °C. The amplicon size was designed to range between 120 and 250 bp. Specific primer sequences, as well as eEF2 and α Tubulin as reference genes for G. boninense 79 were listed in the Supplementary File, Table S11.

Quantitative real-time PCR (qPCR). Each qPCR reaction was performed following QuantiNova SYBR
Green PCR kit manual (Qiagen, German). Three individual technical replicates were performed for each biological replicate at respective sampling intervals and a non-template control was included to monitor crosscontamination. qPCR was performed using Eppendorf Mastercycler Real-time PCR System (Eppendorf, UK). The baseline, the cycle of threshold (C t ) and PCR efficiency were determined by LingReg-PCR software 80 . Statistical analysis for comparison between expression of the target gene (NRPS) versus the reference genes (eEf2, α Tubulin ) was computed by Pfaffl method [by the fold change equation (2 −∆∆Ct )] 81 and Relative Expression Software Tool (REST2009) 82,83 , this software changes C t values into normalized relative expression values by the fold change equation (2 −∆∆Ct ) 84 . The mean ± SEM was evaluated by one-way ANOVA analysis in Graphpad Prism version 8.0.0 software for windows (P < 0.05 showed a significant difference between the infected group and non-infected group at each interval).
Ethical approval. All local, national or international guidelines and legislation were adhered to in the production of this study.