The sampling activities were conducted at FGV plantations, in Malaysia. The root samples were collected within the radius of the palm canopy and stored at 4˚C prior to the isolation process. Endophytic Trichoderma was isolated from healthy oil palm roots and cultured on Martin’s Rose Bengal (MRB) agar, one of the selective media for Trichoderma isolation [9]. All the root samples were surface sterilized [10]. Root samples were cut into 10 cm lengths and cleaned under running tap water to remove the debris. About 0.5 cm was then removed from both tips. The remaining roots were cut into 3 cm lengths. Roots were surface sterilized in 10% sodium hypochlorite for two minutes, then in 50%, 70%, 90%, and 99% ethanol for 30 seconds each, before rinsing three times with sterile distilled water (SDW). The root samples were then allowed to dry on sterilized filter paper. Once dried, 1 cm of the ends of each section was discarded. The cuticle from the middle section was removed, cut into two sections of 0.5 cm each and placed on MRB. Petri dishes were incubated at 28˚C ± 2˚C for 7 days. The distinguished individual colonies were subcultured onto Potato Dextrose Agar (PDA) (Oxoid) plates and were then incubated for 7 days. The pure cultures were finally maintained until further testing.

A total of 14 endophytic Trichoderma isolates that was successfully isolated were screened for their antagonistic potential against Ganoderma boninense (PER71), obtained from Malaysian Palm Oil Board (MPOB) through dual culture assay [11]. Approximately 5 mm plug agar of PER71 was subcultured at 1 cm from the 90 mm Petri plate’s circumference, while 5 mm plug agar of endophytic Trichoderma was inoculated opposite the PER71 agar plug (1 cm from circumference of Petri plate). Another Petri plate, which was only subcultured with PER71, was prepared as control. All the endophytic Trichoderma isolates in dual culture assay were incubated at 28˚C ± 2˚C for seven days. Percentage of inhibition radial growth (PIRG) was then measured based on the formula as below:

( % ) PIRG = [ ( R1 − R2 ) / R1 ] × 1 00

where, R1 = radial growth of PER71 (control); R2 = radial growth of PER71 in dual culture plate.

The number of days required by endophytic Trichoderma to completely overgrow PER71 was recorded. The shortest number of days indicates a significant antagonistic effect in dual culture assay [12]. Their ability to overgrow and inhibit the growth of PER71 was then scored according to Bell’s scale [13]. where:

R1 = 100% overgrowth;

R2 = 75% overgrowth;

R3 = 50% overgrowth;

R4 = locked at the point of contact.

To evaluate the potential of PER71 to proliferate, the inhibition zone between endophytic Trichoderma and PER71 in a dual culture assay was plugged and subcultured onto a new PDA. The ability of PER71 to grow on the new PDA was observed.

Scanning electron microscopy (SEM) observation was conducted at Quasi-S Technology Sdn. Bhd. The inhibition zone from plates with the highest PIRG was examined for their hyphal interaction activity through SEM. The plug agar at the inhibition zone was cut into 1 cm³ and fixed in 1% gluteraldehyde fixative for 12 - 24 hours at 4˚C. Phosphate buffer was then used to wash the sample. The sample was serially dehydrated with acetone concentrations; 30%, 40%, 50%, 60%, 70%, 80%, 90% and 100% for 20 minutes. After that, the samples were dried and coated with gold before viewing through SEM.

The double plate method was used to test the production of volatile compounds by endophytic Trichoderma candidates against G. boninense PER71. A 5 mm plug agar of all examined endophytic Trichoderma (5 days old) and PER71 (7 days old) were cultured on the center of PDA at 28˚C ± 2˚C. After 24 hours, the Petri plate covers of both Trichoderma and PER71 were removed. The plate containing endophytic Trichoderma culture was placed upside down on top of a plate containing PER71 culture. Both plates were tightly sealed together with parafilm to avoid any gap. The presence of gap between the plates may not only prohibit physical contact between the fungi but also prohibit the formation of volatile compounds by the tested Trichoderma. Another plate subcultured with PER71 alone was used as control. All culture plates were incubated at 28˚C ± 2˚C for 7 days to observe the PIRG.

Culture filtrates methods against G. boninense PER71 were used to detect the production of non-volatile chemicals by endophytic Trichoderma isolates [14]. A 5 mm plug agar of tested endophytic Trichoderma (5 days old) was inoculated into a 250 mL conical flask containing 100 mL sterilized potato dextrose broth (PDB) (Oxoid). The inoculated flasks were shaken on an orbital shaker at 28 ± 2˚C for 7 days at 150 rpm. The harvested culture filtrate was first filtered through a filter paper and finally filtered through a filtrate membrane filter sized 0.45 µm. The final filtrate of endophytic Trichoderma was collected and mixed with double strength molten sterile PDA (50˚C) at ratio 1:1. The mixture of PDA and culture filtrate were then poured into a Petri plate and allowed to solidify. A 5 mm plug agar of PER71 was cultured on the center of a Petri plate agar containing the mixture of filtrate and PDA. A plug agar of PER71 was cultured on PDA without any filtrate as control. The cultured plates were then incubated at 28˚C ± 2˚C and PIRG was measured after 7 days.

The species of two isolates of endophytic Trichoderma that showed significant potential in suppressing G. boninense PER71 in the in vitro study was identified. Mycelia and spores of the isolates were scraped from the agar plates and inoculated in 100 mL PDB before incubated at 30˚C for five days. The five-day old mycelia were then filtered and pulverized in liquid nitrogen. Genomic DNA was isolated from pulverized fungal mycelia according to in-house protocol. Terra^TM PSR DNA polymerase (Clontech) was used for the amplification of internal transcribed spacer (ITS) of fungal genomic DNA using forward primer 5’ - TCC GTA GGT GAA CCT GCG G - 3’ and reverse primer 5’ - GGT GCG TGT TTC AAG AC - 3’. 16S rDNA was amplified from bacterial genomics DNA using forward primer 5’ - GGT TAC CCT ACG ACT T - 3’ and reverse primer 5’ - AGA GTT TGA TCM TGG CTC AG - 3’. PCR product was purified, then ligated in pGEM-T Easy Vector (Promega) and cloned in E. coli. Vectors carrying the insert were sent out to Apical Scientific Sdn. Bhd. for DNA sequencing service. Nucleotide sequence was analyzed at SIRIM and then searched against Genbank database of the National Center of Biotechnology Information (NCBI) (https://www.ncbi.nlm.nih.gov/) using Basic Local Alignment Search Tool (BLAST) programme [15]. Species identification was determined according to the homology search results from NCBI database. The nucleotide sequence of the two isolates was deposited to NCBI database. A phylogenetic tree was performed using MEGA (Molecular Evolutionary Genetics Analysis) version 9.0, generated through neighbour joining with bootstrap analysis using 1,000 replicates [16]. Published sequences available in NCBI were retrieved and used for phylogenetic analysis.

All the in vitro screening studies were conducted in Completely Randomized Design (CRD) with five replications. Data were analyzed with SAS® Software. The significant data was determined using Least Significant Different (LSD) test at 5% probability level. The percentage data were transformed into Arcsine transformation [17] before subjected to ANOVA.

A total of 14 endophytic Trichoderma isolates were successfully recovered from healthy oil palm root samples collected from various locations across the FGV plantation in Malaysia (Table 1). After 3 to 7 days of being incubated at 28 ± 2˚C, the colonies of endophytic Trichoderma have emerged on the MRB agar (Figure 1). The morphology of the mycelium initially appeared as white hyphae and later developed into green conidiation on the fifth day after plating [8].

The antagonistic effect of endophytic Trichoderma isolates to suppress the mycelium growth of pathogenic G. boninense PER71 was observed through a dual culture assay. All endophytic Trichoderma isolates were found to be capable of suppressing the mycelium growth of PER71 with PIRG between 70.86% to 93.14% after 7 days of incubation. Isolate M103 has recorded the highest PIRG at 93.14% (Table 2). The experiment was extended for another 7 days to observe their capability to overgrow the PER71 culture. Overall, all endophytic Trichoderma

Means in the same column with different alphabet (s) are significantly different (p ≤ 0.05) according to LSD.

isolates are able to overgrow PER71 with Bell’s Ranking at R1. Isolates M103 was determined as the fastest growing culture to overgrow PER71 after 9 days, compared to other isolates which only overgrow the pathogen at day 14 (Figure 2). PER71 was also not able to grow on the inhibition zone of all tested Trichoderma isolates.

The mycoparasitism activity caused by isolate M103 on pathogenic PER71 was examined through Scanning Electron Microscope (SEM), specifically on the inhibition zone. The pictograph shows the mycelia of M103 had grown and coiled the mycelia of PER71, eventually penetrating the hyphae and lysed them (Figure 3).

Endophytic Trichoderma isolates that have recorded PIRG higher than 80% in dual culture assay were further tested for their ability to produce volatile compounds. A total of 9 isolates were determined producing volatile compounds, indicated by the successful suppression of G. boninense PER71 growth on plates. The PIRG recorded was in the range of 36.00% to 65.00% after seven days of incubation (Table 3). The highest suppression was 65%, caused by the volatile compound produced by M103, while the lowest suppression was recorded from the volatile compound produced by M108, with PIRG of 36.00%. This study shows that the production of volatile compounds may be varied from the different isolates of endophytic Trichoderma which eventually produce different levels of suppression on PER71.

The antibiosis effect of non-volatile compounds secreted by the endophytic Trichoderma isolates against G. boninense PER71 were determined via cultures filtrated assay. The percentage of inhibition caused by the tested isolates was

Means in the same column with different alphabet(s) are significantly different (p ≤ 0.05) according to LSD.

Means in the same column with different alphabet(s) are significantly different (p ≤ 0.05) according to LSD.

from 16.09% to 98.18% (Table 4). In this assay, M108 has produced a toxic non-volatile compound, which caused 98.18% inhibition on PER71 on plates, which was the highest PIRG recorded compared to other isolates. Contrarily, M103 which was found to be the most outstanding isolate in previous dual culture assay and production of volatile compound experiments, has produced a non-volatile compound that only caused the lowest inhibition on PER71 (16.09%).

The ITS sequence data of isolates M103 and M108 were compared with several possible homologous sequences from NCBI GenBank database (Table 5). Based on the sequence analysis, isolates M103 and M108 isolates were identified as

Trichoderma asperellum and T. harzianum at 99% similarity with NCBI Genbank accession number given in the phylogenetic studies of sequence analysis data. The generated neighbour joining based on the phylogenetic tree shows that endophytic Trichoderma are clustered together according to their species (Figure 4).