Assessing the Various Antagonistic Mechanisms of Trichoderma Strains against the Brown Root Rot Pathogen Pyrrhoderma noxium Infecting Heritage Fig Trees

A wide range of phytopathogenic fungi exist causing various plant diseases, which can lead to devastating economic, environmental, and social impacts on a global scale. One such fungus is Pyrrhoderma noxium, causing brown root rot disease in over 200 plant species of a variety of life forms mostly in the tropical and subtropical regions of the globe. The aim of this study was to discover the antagonistic abilities of two Trichoderma strains (#5001 and #5029) found to be closely related to Trichoderma reesei against P. noxium. The mycoparasitic mechanism of these Trichoderma strains against P. noxium involved coiling around the hyphae of the pathogen and producing appressorium like structures. Furthermore, a gene expression study identified an induced expression of the biological control activity associated genes in Trichoderma strains during the interaction with the pathogen. In addition, volatile and diffusible antifungal compounds produced by the Trichoderma strains were also effective in inhibiting the growth of the pathogen. The ability to produce Indole-3-acetic acid (IAA), siderophores and the volatile compounds related to plant growth promotion were also identified as added benefits to the performance of these Trichoderma strains as biological control agents. Overall, these results show promise for the possibility of using the Trichoderma strains as potential biological control agents to protect P. noxium infected trees as well as preventing new infections.


Introduction
Fungi are the most dominant group of plant pathogens, causing some of the world's most serious plant diseases on a wide range of crops and pose a continuous threat to global food security [1][2][3][4]. One such fungus is Pyrrhoderma noxium which causes the destructive plant disease known as brown root rot in tropical and sub-tropical regions, such as Africa, Asia, Central America, the Caribbean, Australia and Oceania. In the case of Australia, sub-tropical conditions in the state of Queensland (QLD) have been reported to facilitate the establishment of this disease [5]. Its highly invasive nature has already killed many significant trees including several heritage fig trees found in the public spaces of Brisbane city, the capital of QLD [5]. Infected trees in urban parks, along the roads and power lines are imposing a great risk to surrounding people and properties. This necessitated the need to identify suitable disease control measures to manage its spread.
Thus, studying the antagonistic mechanisms of Trichoderma species against the pathogen in vitro gives insight into its effectivity as a biological control agent for environmental applications. Accordingly, two indigenous Trichoderma strains (#5001 and #5029) were tested to: (1) evaluate the antagonistic mechanisms of the strains against P. noxium strains; (2) study the expression of the biological control associated genes during their confrontation with P. noxium; (3) characterise the antifungal activity of volatile compounds emitted from these Trichoderma strains; (4) study the plant growth promoting traits of these Trichoderma strains which may be associated with plant growth promotion. Identification of effective biological control agent would protect the economically and culturally significant plants from P. noxium infection and also minimise the application of hazardous chemical control measures against this pathogen.

Genome Sequencing for the Molecular Identification of Trichoderma Strains
For the molecular identification, the Trichoderma strains (#5001 and #5029) were cultured on Potato Dextrose Agar (PDA, Oxoid, Hampshire, UK) and incubated at 28 • C for 5 days prior to the DNA extraction. DNA extraction was performed using DNeasy Plant mini Kit (Qiagen, Hilden, Germany) and following the manufacturer's instructions. DNA samples were then sent to the Australian Genome Research Facility (AGRF) in Brisbane, QLD, for whole genome sequencing on NovaSeq 6000 platform. The data generated with the Illumina bcl2fastq pipeline version 2.20.0.422 were used in the downstream analysis.

Assembly Preparation for Downstream Application
De novo genome annotation of both Trichoderma genomes were performed using GeneMark-ES [76] and AUGUSTUS [77] in Funannotate (v1.7.4) gene prediction pipeline (https://funannotate.readthedocs.io/en/latest/ accessed on 15 January 2022). Then, validated by mapping the RNAseq data of the reference strain T. reesei (PRJNA15571) to the genes to support the predictions. The newly annotated gene models were evaluated for completeness using BUSCO v5 in protein mode against the ascomycota_odb10 database. 3) as out group were down-loaded from the GenBank database (as of December 2021). Protein sequences were analysed with OrthoFinder v. 2.3.8, which grouped them into orthogroups [78,79]. Gene families that contained only one gene for each species (single copy orthologs) were selected to construct a phylogenetic tree using FastTree program integrated in OrthoFinder [80]. The phylogenetic trees of the Trichoderma isolates were constructed by maximum likelihood method [81]. Finally, the tree was visualized using iTOL [82].

Microscopic Investigation of the Mycelial Interaction between P. noxium and Trichoderma Strains
The mycelial interaction between Trichoderma strains and P. noxium was observed under an automated upright microscope system (LEICA DM5500 B) [83]. The P. noxium plug was inoculated at 0.5 cm distance from a coverslip embedded into the PDA at a 45 • angle. After two days of incubation, a plug of the strains #5001 and #5029 (separately) was placed 1 cm away from P. noxium on the plate and incubated for 2 days at 28 • C. Once both fungi were grown onto the coverslip, fungal mycelia were stained with lactophenol cotton blue and observed for the mycoparasitism-related structures produced by Trichoderma strains, e.g., coiling and appressorium-like structures.

Detection of the Antagonistic Activities of Trichoderma Strains against P. noxium Using a Dual Culture Assay
The antagonistic activity of Trichoderma strains (#5001 and #5029) was tested against five different P. noxium strains: A, B, D, E and F using a dual culture assay [84]. Trichoderma and P. noxium plugs were inoculated on a PDA plate away from each other [84]. Due to the slow growth rate, P. noxium was inoculated two days before the inoculation of Trichoderma strains. After two days, a Trichoderma plug was placed 2 cm from the edge of the P. noxium colony. As a control, the pathogen and the biological control agents (BCAs) were placed in a similar manner on a fresh PDA plate without reciprocal fungal inoculations and incubated at 28 • C for 5 days. All pairings were carried out in quadruplicate. Growth inhibition percentage for tested phytopathogenic fungi was determined [84].

Assessing the Inhibitory Effect of Volatile Compounds Produced by Trichoderma on P. noxium Growth
To study the effect of volatile compounds of Trichoderma strains on P. noxium (A, B, D, E and F), the sandwich plate method described by Dennis and Webster, 1971b [85], was used. A Trichoderma plug (3 days old) was cultured on PDA medium and incubated for 4 days at 28 • C. After that, the lids of the Trichoderma culture plates were replaced by the PDA plates (bottom part) freshly inoculated with a P. noxium plug, sealed with three layers of parafilm, and incubated at 28 • C for 3 days. Controls of P. noxium were prepared by sandwiching with an un-inoculated PDA plate. All pairings were carried out in quadruplicate. Effect of volatile compounds produced by BCAs on the pathogen was evaluated by comparing the growth area of P. noxium in the treatment plates with the control plates [85].

Changes in the Volatile Compounds Produced by Trichoderma Strains before and after the Confrontation with P. noxium
To identify the volatile compounds produced by the pure cultures of Trichoderma strains (#5001 and #5029), each strain was prepared by placing 2 mm agar plugs in the center of a headspace vial (20 mL) containing PDA medium. To study the changes in the volatile compounds produced by Trichoderma strains during the interaction with pathogen, co-cultures were prepared by placing two 2 mm agar plugs of the strains #5001 and 5029 (separately) and P. noxium (strain B was used as representative of P. noxium) at the center of the headspace vial and incubating for 7 days at 28 • C [86]. Due to the slow growth rate, P. noxium was inoculated one day before the inoculation of Trichoderma. Pure cultures of P. noxium were also prepared in headspace vials and used for identifying the volatile compounds produced by P. noxium in the co-culture setup.
Uninoculated PDA medium was used as blank [86]. All the inoculated vials and controls were incubated for 7 days at 28 • C. Volatiles in the headspace of the above cultures were analyzed using Solid Phase Microextraction (SPME). Fibers (75 µm Carboxen/PDMS, Merck) were inserted in the headspace vial, and exposed for 15 min. Volatiles were desorbed from the fiber when it was injected into the inlet port and analyzed with a gas chromatograph (GC) (Agilent 6890 Series) coupled to a mass spectrometer (MS) (Agilent 5975) and fitted with a silica capillary column (Agilent, model HP5-MS, 30 m × 250 µm ID × 0.25 µm film thickness). Gas chromatograph conditions for acquiring data were-inlet temperature: 250 • C, carrier gas: helium at 51 cm/s, split ratio 13:1, transfer-line temperature: 280 • C, initial temperature: 40 • C, initial time: 2 min, rate: 10 • C/min, final temperature: 260 • C, final time: 6 min. The MS was held at 280 • C in the ion source and the scan rate was 4.45 scans/s. Tentative identities were assigned to peaks with respect to the National Institute of Standards and Technology mass spectral library. Mass spectra of peaks from different samples with the same retention time were compared to ensure that the compounds were indeed the same. Negative control results were compared with volatiles from blank PDA control and common substances were removed during interpretation of results.

Effect of Diffusible Compounds Produced by Trichoderma Strains on P. noxium
To examine the effects of diffusible (non-volatile) compounds produced by the strains # 5001 and #5029 on P. noxium growth, biological control agents and pathogen (P. noxium strain B) were co-cultured on PDA and incubated at 28 • C for 7 days. After the incubation period, culture medium was cut into small pieces and the diffusible compounds were extracted using ethyl acetate [87]. Three extractions were performed at 24-h intervals on a shaker at 120 rpm at 20 • C. All the extracts were filtered through a Whatman No. #1 filter paper [87]. The ethyl acetate extracts were concentrated on a rotary evaporator at room temperature and further evaporated using Genevac. The dried crude extracts were measured and dissolved in appropriate volume of ethyl acetate to get the concentrations of 500, 1000, 1500, 2500, 5000 and 10,000 ppm and used for fungicidal assay using paper disc diffusion technique [88].
To determine the fungicidal activity of the crude extract against P. noxium, an agar plug from a 3-day-old mycelial culture of P. noxium strain B was inoculated at the center of a fresh PDA plate. Two sterile disks were kept both sides of the inoculated plug at 2.5 cm distance. One of the disks was saturated with 100 µL of the crude extract and 10 µL of Dimethyl sulfoxide (DMSO) and the other disk was saturated with solvents (100 µL ethyl acetate and 10 µL DMSO) and used as the control. The plates were incubated at 28 • C for 5 days. The fungicidal activity of crude extract at different concentrations was assessed in terms of the area of fungal inhibition and expressed as the percentage of the inhibition of fungal growth as mentioned above. The fungicidal assay was performed with three replicates.

The Biocontrol-Associated Genes Expression in Trichoderma Strains during the Interaction with P. noxium
In order to study changes in the expression level of biological control related genes in Trichoderma strains (#5001 and #5029), a confrontation assay was conducted using P. noxium [22]. The strain B was used as the P. noxium representative in this study. Six biological control associated genes (chitinase 33 (chit), endochitinase 42 (endo), β-1,3-endoglucanase (bgn), trypsin-like protease pra1 (pra), subtilisin-like protease prb1 (prb) and QID74 protein (qid)) previously reported in Trichoderma spp. [27] for the expression upon pathogen confrontation were studied during the interaction between Trichoderma strains (#5001 and #5029) and P. noxium strain B. PDA plates were overlaid with cellophane and were inoculated with agar plaques of the P. noxium strain B and the strains # 5001 or #5029 (separately) as explained above and incubated for 3 days at 28 • C. The membrane was used to facilitate the removal of the Trichoderma mycelia at the contact point from the plate for RNA extraction [22,24]. For the controls, Trichoderma strains (#5001 and 5029) were grown alone on PDA plates at the same growth conditions, while the peripheral hyphae were used for RNA extraction.
For RNA extraction, mycelia (90 mg) of the Trichoderma strains #5001 and #5029 from the treatment and control plates were immediately placed in Eppendorf tubes with Zirconia silica beads (few of the 23 mm beads and a small quantity of 0.1 mm diameter beads) and frozen in liquid nitrogen for 30 s and then placed into a Qiagen TissueLyser adapter for efficient grinding for 40 s at 23,000 rpm. The liquid nitrogen freezing and grinding steps were repeated twice and the adapter positions were reversed in the TissueLyser each time. The ground mycelia were used for total RNA extraction using RNeasy Plant Mini Kit (Qiagen, Hilden, Germany) as per manufacturer's instructions. NanoDrop was used for RNA quantification.
Genomic DNA elimination step followed by cDNA synthesis was carried out using the reaction components provided in the QuantiTect ® Reverse Transcription Kit (Qiagen, Hilden, Germany) and following manufacturer's instructions. The cDNA concentration was quantified using a NanoDrop. The changes in the expression level of biological control associated genes of Trichoderma during the confrontation with P. noxium was analysed using qPCR techniques. This method is widely used for the quantification of gene expression [22,24]. Biological control associated genes in strains #5001 and #5029 were identified via homology sequence search against the biological control genes of T. harzianum CBS 226.95 (V1.0) previously reported [24]. The mRNA of T. harzianum CBS 226.95 (V1.0) were obtained from Genbank (https://www.ncbi.nlm.nih.gov/genbank/ as of 15 December 2021) and the gene transcripts (chit, endo, bgn, pra, prb and qid) were identified via locating the specific primer sequences reported [24]. The identified transcripts of T. harzianum were individually BLAST against the blast library of mRNA sequences from the strains #5001 and #5029 created in Qiagen CLC Genomics Workbench version 20.0. The homology sequences (with lowest E-value and highest percentage of identity) in the transcript of Trichoderma strains #5001 and #5029 were selected for primer designing. The specific primers were designed using Primer 3 (http://primer3.ut.ee/ accessed on 15 January 2022) (Table S1) and manufactured by Merk (Australia). Quantitative real-time PCR was used for amplifying the transcripts from individual genes using QuantiTect SYBR Green PCR Kit (Qiagen, Hilden, Germany). Each reaction consisted of 10 µL QuantiTect SYBR Green PCR master mix, 1 µL each of forward and reverse primers (final concentration of 0.5 µM), 2 µL cDNA, and 6 µL Rnase-free water. qPCR conditions were: one cycle of initial denaturation step at 95 • C for 15 min to activate the HotStarTaq DNA Polymerase, followed by 35 cycles of denaturation at 95 • C for 15 s, annealing at 59 • C for 30 s and extension at 72 • C for 30 s. High resolution melting (HRM) was carried out over the ranging from 65 • C to 95 • C with temperature increasing steps of 0.5 • C every 4 s. RNase-free water was used in non-template control [24].
Genes encoding elongation factor 1-α and β-actin were used as internal references for normalization of gene expression in Trichoderma [24,89]. Relative expression levels of individual genes were calculated from the threshold cycle using the Pfaffl method for efficiency correction [90]. A standard curve was generated to calculate the gene-specific PCR amplification efficiency and correlation coefficient for each gene [24].

Effect of Volatiles Produced by Trichoderma Strains on the Growth Parameters of Arabidopsis thaliana
The effect of VOCs of Trichoderma on plant growth parameters was studied using modified double plate-within-a-plate method described by Lee et al. 2016 [54]. Arabidopsis thaliana was used as model plant [54]. Two small Petri dishes (50 mm) with PDA were placed in a larger plate (140 mm) and 50 mL of Murashige and Skoog salt (MS) medium (pH 5.7) containing 3% sucrose and 0.7% agar was poured into the larger plate ( Figure S3). A. thaliana seeds were surface sterilised for 5 min in 70% ethanol, soaked for 7 min in 10% sodium hypochlorite solution, and rinsed four times with distilled water. Finally, seeds were kept in distilled water and vernalized for 2 days at 4 • C in dark. Sterilised seeds were inoculated on MS media and incubated for 3 days. Ten germinated seeds selected from contamination free plates were reinoculated on the MS media portion established in the double plate-within-a-plate setup ( Figure S3). In the meantime, small plates in the double plate-within-a-plate units were inoculated with 5 µL of a Trichoderma spore suspension (containing 1 × 10 6 spores). For controls, plants were exposed to the PDA alone. The VOCs mediated plant growth promotion is often associated with exposure period [54]. Therefore, this study was conducted for a period of 12 days to provide enough volatile exposure to the plants. All the plates were sealed and incubated at 25 • C. The experiment was performed in a complete randomized block design with 2 treatments (Trichoderma strains #5001 and #5029) and 1 negative control with 5 replicates (corresponding to 10 plants/plate).
The effect of VOCs on the growth parameters of A. thaliana was evaluated by comparing the fresh mass, shoot length, root length and total chlorophyll content of plants exposed to Trichoderma strains with unexposed plants. Ten plates were pooled and biomass was measured (n = 5, corresponding to 50 plants). Total chlorophyll content of A. thaliana was determined spectrophotometrically [54]. Fifty milligrams of shoots collected from each treatment were ground and submerged overnight in 100 µL of 80 % acetone in the dark at 4 • C. The following day, absorbance readings were measured at 645 nm (A 645 ) and 663 nm (A 663 ) using an EnSpire™ multimode plate reader [52,54]. The total chlorophyll content was calculated from the following equation [(8.02)(A 663 ) + (20.2)(A 645 )]V/1000 × W, where V is volume and W is plant fresh mass [52,54].

Indole 3-Acetic Acid Production by Trichoderma Strains
The IAA producing ability of Trichoderma strains was studied in potato dextrose broth (PDB, Oxoid, Hampshire, UK) supplemented with L-tryptophan [91]. Filter (0.2 µm membrane filter, Whatman) sterilised tryptophan (0.2 % w/v) was incorporated after sterilising the media at 15 psi at 121 • C for 15 min. During the preparation of 0.2% Ltryptophan aqueous solution, a few drops of NaOH (1 M) was used for increasing solubility of tryptophan. After adding the L-tryptophan, culture medium pH was adjusted to 5.6 with filter sterilized HCl (1 M).
Culture flasks containing 50 mL of PDB broth with L-tryptophan were inoculated with two agar plugs removed from a 3-day old Trichoderma culture plate and the flasks were incubated at 28 • C at 120 rpm [92,93]. Cell-free culture filtrates collected on day 7 were used for testing the indole 3-acetic acid production. IAA was measured by mixing 1 mL of the cell-free culture filtrate with 2 mL of Salkowski reagent (2% 0.5 M FeCl 3 in 35% HClO 4 ) [94]. The reaction mixture was incubated at room temperature for 20 min and the absorbance was recorded at 530 nm in a UV-Vis spectrophotometer [95]. IAA produced by the Trichoderma was determined from the standard curve. The experiment was performed in triplicate, IAA produced by the Trichoderma strains are presented as mean values with a standard deviation of three replicates.

Siderophore Production by Trichoderma Strains
The siderophore production by Trichoderma strains was studied on universal siderophore assay medium using chrome azurol S (CAS). The medium was prepared according to the procedure described by Schwyn and Neilands, 1987 [96] and Louden et al. 2011 [97]. A culture plug removed from 3-day old Trichoderma plate was placed at the centre of CAS agar medium and incubated at 28 • C for 7 days. After the incubation period, plates were observed for evidence (orange halo) of siderophore production.

Statistics
The effect of total volatiles produced by Trichoderma strains on the growth of different P. noxium strains (A, B, D, E and F) were analysed via one-way ANOVA followed by Tukey's test. The differences in inhibitory effect caused by the Trichoderma strains # 5001 and #5029 were compared using the paired t-test. Similarly, the inhibitory effect displayed by the Trichoderma strains on the growth of different P. noxium strains during the confrontation was studied using one-way ANOVA followed by Tukey's test. The differences in mycoparasitism caused by the Trichoderma strain # 5001 and #5029 were compared using the paired t-test. The effect of crude extract of Trichoderma strains on the growth of P. noxium strains B at different concentrations was analysed via one-way ANOVA followed by Tukey's test. The differences in inhibitory effect caused by the crude extract of Trichoderma strains # 5001 and #5029 were compared using the paired t-test. The results from gene expression study were analysed using a Student's t-test to compare the changes in expression levels between control and treatments. The effect of VOCs of Trichoderma strains on A. thaliana was compared between the strains and also with control via one-way ANOVA followed by Tukey's test. The IAA and siderophore producing ability of the strains # 5001 and #5029 were compared using the paired t-test. All the statistical analyses were performed using SPSS statistics 26 [41].

Molecular Identification of Trichoderma Strains via Whole Genome Sequencing
The Illumina sequencing generated 222 million, 150-bp paired-end reads for Trichoderma strain #5001 and 202 million, 150-bp paired-end reads for Trichoderma strain #5029. The genome of Trichoderma strain #5001 assembled in to 116 scaffolds with a total assembly length 32.126 Mb (N50 = 493.501 KB and N90 = 196.64 KB), a GC content of 53.33%, and a largest scaffold size of 1.473 Mb (Table 1). Based on Kmer-based genome size estimation size of the Trichoderma strain #5001 genome was 31,444,261 bp. In total, 8812 protein coding genes were predicted in the Trichoderma strain #5001 genome. The analysis showed the presence of 99% BUSCO completeness. Trichoderma strain #5029 genome consists of 85 scaffolds sequence with a total assembly length 32.132 Mb (N50 = 599.273 KB and N90 = 264.963 KB), a GC content of 53.33%, and a largest scaffold size of 1.76 Mb (Table 1). Genome size of Trichoderma strain #5029 was estimated as 31,502,229 bp. In total, 8969 protein coding genes were predicted in the Trichoderma strain #5029 genome. The analysis showed the presence of 99.06% BUSCO completeness. These assembly quality metrics indicate that we have achieved a high level of contiguity and completeness for both these genomes in comparison with other published assemblies. The program OrthoFinder identified 158,696 gene clusters, from which 4653 orthologous genes were shared across all 15 Trichoderma species. From these shared gene clusters, 3059 orthologous genes present as single copies were used for the phylogenetic analysis. Trichoderma strain #5001 and #5029 form a distinct cluster with T. reesei (GCF 000167675.1) (Figure 1)  The program OrthoFinder identified 158696 gene clusters, from which 4653 orthologous genes were shared across all 15 Trichoderma species. From these shared gene clusters, 3059 orthologous genes present as single copies were used for the phylogenetic analysis. Trichoderma strain #5001 and #5029 form a distinct cluster with T. reesei (GCF 000167675.1) (Figure 1) indicating they are most closely related to T. reesei.

Microscopic Investigation of the Mycelial Interaction between P. noxium and Trichoderma Strains
During the interaction between Trichoderma strain #5001 (T) and P. noxium (P), Trichoderma coiled around the pathogen (Figure 2a) and formed appressoria like structures (Figure 2b). As observed with the strains #5001, the strain 5029 also used the similar mycoparasitic structures against P. noxium. Moreover, surface colonisation by Trichoderma on the host mycelia was also observed under the microscope (Figure 2c).

Microscopic Investigation of the Mycelial Interaction between P. noxium and Trichoderma Strains
During the interaction between Trichoderma strain #5001 (T) and P. noxium (P), Trichoderma coiled around the pathogen (Figure 2a) and formed appressoria like structures ( Figure 2b). As observed with the strains #5001, the strain 5029 also used the similar mycoparasitic structures against P. noxium. Moreover, surface colonisation by Trichoderma on the host mycelia was also observed under the microscope (Figure 2c).  The results show that the Trichoderma strains used in this study were effective in in-

Detection of the Antagonistic Activities of Trichoderma Strains against P. noxium in Dual Culture Assay
The results show that the Trichoderma strains used in this study were effective in inhibiting P. noxium growth ( Figure S1). The growth inhibition caused by Trichoderma strains was significantly different between P. noxium strains (F = 7.493, p < 0.001) (Figure 3a). The growth inhibition caused by the strains #5001 on P. noxium was significantly (p = 0.04) higher than the inhibition caused by the strain #5029. The growth inhibition caused by the strain #5001 on P. noxium strain B was significantly higher (F = 15.391, p < 0.001) than P. noxium A (62%), D (60%), E (60%) and F (60%). Whereas the inhibition caused by the strain #5029 on P. noxium strains B and A was significantly higher (F = 17.7, p < 0.001) than the strains E (67%), F (65%) and minimum in D (61%) (Figure 3a).

Inhibitory Effect of Volatile Compounds of Trichoderma Strains on P. noxium Growth
The results show that P. noxium growth was inhibited by the volatile compounds emitted by the Trichoderma strains. The growth inhibition caused by the volatile compounds of Trichoderma strains was significantly different between P. noxium strains (F = 3.104, p = 0.033) (Figure 3b). The inhibition caused by the strain #5001 was not significantly

Inhibitory Effect of Volatile Compounds of Trichoderma Strains on P. noxium Growth
The results show that P. noxium growth was inhibited by the volatile compounds emitted by the Trichoderma strains. The growth inhibition caused by the volatile compounds of Trichoderma strains was significantly different between P. noxium strains (F = 3.104, p = 0.033) (Figure 3b). The inhibition caused by the strain #5001 was not significantly different (p = 0.924) from the inhibition caused by the strain #5029. The growth inhibition caused by the VOCs of the strain #5001 on P. noxium strains B and E was significantly higher (F = 4.886, p = 0.019) than the inhibition on P. noxium strains A (58%), D (53%) and minimum in F (49%). Whereas the inhibition caused by the VOCs of the strain #5029 on P. noxium strain B was significantly higher (F = 5.99, p = 0.010) than strain F (56.9%), D (56.12%), A (55.2%) and minimum in E (52.9%) (Figure 3b).

Changes in the Volatile Compounds Produced by Trichoderma Strains before and after the Confrontation with P. noxium
The volatile compounds produced by strain #5001 before and after confrontation with P. noxium strain B are shown in Table 2 and Figure 4, whereas volatile compounds produced by strain #5029 before and after confrontation with P. noxium strain B are shown in Table 2 and Figure 5. The compounds identified in pure cultures of Trichoderma strains and in co-culture treatments were members of the compound classes; alcohol, aldehyde, ester, diterpenoid alcohol and sesquiterpene ( Table 2). A total of 21 compounds were tentatively identified using the NIST library for the pure culture of strain #5001 (Table 2), whereas 19 compounds were identified in co-culture set up of strain #5001-P. noxium. On the other hand, a total of 19 compounds were identified for the pure culture of strain # 5029, while 18 compounds were identified in co-culture set up of strain #5029-P. noxium. The volatile profiles of the strain #5001 and co-culture of the strain #5001-P. noxium were, nearly the same: 16 of 21 compounds were in common. Similarly, 15 of 19 compounds from the strain #5029 were detected in the emission profile of co-culture of strain #5029-P. noxium. On the other hand 13 out of 16 compounds seen in a pure P. noxium culture, furan, βnecoclovene, aristolene, thujopsene, 2-isopropenyl-4a,8,dimethyl-1,2,3,4a,5,6,7-octahydronaphthalene, δ-gurjunene, β-elemene, β-caryophyllene, 8-isopropenyl-1,5-dimethyl-cyclodeca-1,5-diene, cadiene, elixene and a unidentified sesquiterpene ( Figure S2 and Table S2) were not detected in the co-cultures of P. noxiumstrain #5001 or in the co-cultures of P. noxium-strain #5029. Overall, co-culture profiles showed high similarity with the profile of the Trichoderma strains rather than of P. noxium (Table 2 and Figures 4 and 5). This confirmed the inhibitory effect of Trichoderma strains on P. noxium growth in the co-culture set up.
The most dominant volatile compound produced by the pure cultures of strain # 5001 and #5029 was ethanol, at a retention time of 1.02 min with relative percent area of 22.72% and 31.18%, respectively, (Table 2, Figures 4 and 5), While, the relative percentage area of ethanol was increased to 33.98% and 36.3%, respectively in the co-culture setup of the strain #5001-P. noxium and strain #5029-P. noxium. The second most dominant compound produced by the strain #5001 and #5029 was β-cedrene, at 11.72 min with relative percent area of 19.95% and 18.31%, respectively. The emission of β-cedrene was slightly reduced to 19.28% and 16.5% in the co-culture of strain #5001-P. noxium and in the co-culture of strain #5029-P. noxium, respectively. In addition, isobutanol produced by pure cultures of strains #5001 and #5029, at retention time of 1.33 min with relative percent area of 7.90% and 6.60%, respectively, was also reduced to 4.24% in the co-culture of strain #5001-P. noxium and 4.53% in the co-culture of strain #5029-P. noxium. There were three new sesquiterpene compounds (α-copaene, β-himachalene, α-longipinene) detected in the co-culture set up of the strain #5001-P. noxium and in the cultures of the strain #5029-P. noxium were not detected when the strains were grown alone ( Table 2, Figures 4 and 5). α-cubebene and acoradiene were unique in the co-culture set up of strain #5001-P. noxium, whereas 2-methylbutyl acetate was unique in the co-culture assay of strain #5029-P. noxium.   Table 2.   Table 2.  Table 2.

Effect of Diffusible Compounds of Trichoderma Strains on P. noxium Growth
The diffusible compounds emitted by the Trichoderma strains showed growth inhibition on P. noxium strain B (Figure 3c). The growth inhibition caused by the crude extract was significantly different (F = 815.24, p < 0.001) among the tested concentration. The inhibition caused by the crude extracts of the strain #5001 was not significantly different from the inhibition by the extracts of strain #5029 (p = 0.58). The inhibition caused by the crude extract of the strain #5001 (F = 536.84, p < 0.001) and #5029 (F = 499.85, p < 0.001) were significantly high at 10,000 ppm compared to other tested concentrations. Whereas the lowest inhibition by the extracts from both the strains was observed at 500 ppm (Figure 3c).

The Biocontrol-Associated Genes Expression in Trichoderma Strains during the Interaction with P. noxium Strain B
The gene expression levels measured between the strains were significantly different (p < 0.04). Expression of chit was higher than other biological control related genes expression levels measured in both Trichoderma strains ( Figure 6). Particularly, relative expression of chit gene was increased by 11.6 times in Trichoderma strain 5029 and by 8.4 times in Trichoderma strain #5001 (Figure 6), While relative expression of endo, bgn, pra, prb and qud genes were increased by 3.3, 2.5, 3.1 and 2.9 times, respectively in Trichoderma strain 5001 and by 1.5, 4.4, 2.4 and 2 times, respectively in Trichoderma strain #5029 ( Figure 6). The chit, qid and bgn genes of Trichoderma strain #5029 were expressed at 1.38, 3.13 and 1.74 times higher, respectively, than expressions detected in Trichoderma strain #5001 ( Figure 6). Whereas endo, pra and prb genes of the strain #5001 expressed 2.2, 1.3 and 1.4 times higher, respectively, than expression measured in Trichoderma strain #5029 ( Figure 6). tion on P. noxium strain B (Figure 3c). The growth inhibition caused by the crude extrac was significantly different (F = 815.24, p < 0.001) among the tested concentration. The in hibition caused by the crude extracts of the strain #5001 was not significantly differen from the inhibition by the extracts of strain #5029 (p = 0.58). The inhibition caused by th crude extract of the strain #5001 (F = 536.84, p < 0.001) and #5029 (F = 499.85, p < 0.001) wer significantly high at 10000 ppm compared to other tested concentrations. Whereas th lowest inhibition by the extracts from both the strains was observed at 500 ppm (Figur 3c).

The Biocontrol-Associated Genes Expression in Trichoderma Strains during the Interaction with P. noxium Strain B
The gene expression levels measured between the strains were significantly differen (p < 0.04). Expression of chit was higher than other biological control related genes expres sion levels measured in both Trichoderma strains ( Figure 6). Particularly, relative expres sion of chit gene was increased by 11.6 times in Trichoderma strain 5029 and by 8.4 times in Trichoderma strain #5001 (Figure 6), While relative expression of endo, bgn, pra, prb and qu genes were increased by 3.3, 2.5, 3.1 and 2.9 times, respectively in Trichoderma strain 500 and by 1.5, 4.4, 2.4 and 2 times, respectively in Trichoderma strain #5029 ( Figure 6). The chit qid and bgn genes of Trichoderma strain #5029 were expressed at 1.38, 3.13 and 1.74 time higher, respectively, than expressions detected in Trichoderma strain #5001 ( Figure 6) Whereas endo, pra and prb genes of the strain #5001 expressed 2.2, 1.3 and 1.4 times higher respectively, than expression measured in Trichoderma strain #5029 ( Figure 6). . Figure 6. Expression patterns of biocontrol-associated genes after confrontation with P. noxium strain B in (a,c) Trichoderma strain #5001 and (b,d) Trichoderma strain #5029. Elongation factor 1−α and β−actin genes (tef and act, respectively) were used as references to normalize biological control associated genes expression in Trichoderma strain #5001 and #5029. Asterisks indicate significant differences by Student's t-test (p < 0.05).

Effect of Volatiles Produced by Trichoderma Strains on the Growth Parameters of Arabidopsis thaliana
The plants from the shared atmosphere with the Trichoderma strains showed improved growth than control plants ( Figure S3). The root length of the plants, which shared the space with the strains #5001 and #5029 were significantly (F = 1986.4, p < 0.001) higher than control plants (Figure 7). Similarly, shoot length (F = 3362.7, p < 0.001), chlorophyll content (F = 41.7, p < 0.001) and fresh mass (F = 565, p < 0.001) were also significantly improved in the plants which shared the space with the strains #5001 and #5029 (Figure 7). The root length and chlorophyll content measured different from control (F = 1986.4, p < 0.001) but were not significantly different between the treatments of the strain #5001 and #5029 (Figure 7). However, shoot length and fresh mass measured in the treatment of Trichoderma strain #5001 was significantly (F = 1986.4, p < 0.001) higher than the strain # 5029 (Figure 7). . Expression patterns of biocontrol-associated genes after confrontation with P. noxium strain B in (a,c) Trichoderma strain #5001 and (b,d) Trichoderma strain #5029. Elongation factor 1−α and β−actin genes (tef and act, respectively) were used as references to normalize biological control associated genes expression in Trichoderma strain #5001 and #5029. Asterisks indicate significant differences by Student's t-test (p < 0.05).

Effect of Volatiles Produced by Trichoderma Strains on the Growth Parameters of Arabidopsis thaliana
The plants from the shared atmosphere with the Trichoderma strains showed improved growth than control plants ( Figure S3). The root length of the plants, which shared the space with the strains #5001 and #5029 were significantly (F = 1986.4, p < 0.001) higher than control plants (Figure 7). Similarly, shoot length (F = 3362.7, p < 0.001), chlorophyll content (F = 41.7, p < 0.001) and fresh mass (F = 565, p < 0.001) were also significantly improved in the plants which shared the space with the strains #5001 and #5029 (Figure 7). The root length and chlorophyll content measured different from control (F = 1986.4, p < 0.001) but were not significantly different between the treatments of the strain #5001 and #5029 (Figure 7). However, shoot length and fresh mass measured in the treatment of Trichoderma strain #5001 was significantly (F = 1986.4, p < 0.001) higher than the strain # 5029 (Figure 7).

Indole 3-Acetic Acid Production by Trichoderma Strains
Both of the strains used in this study produced IAA in the L-tryptophan supplemented media. The IAA produced by the strain #5029 (6.13 ± 0.31 µg/mL) was significantly higher (p = 0.03) than the IAA produced by the strain #5001 (4.49 ± 0.19 µg/mL).

Siderophore Production by Trichoderma Strains
Visible orange halo formation surrounding the Trichoderma growth on CAS agar media indicate that both the strains produced siderophores on CAS agar media. Halo zone developed by the strain #5029 (18.33 ± 0.5 cm 2 ) was significantly larger (p = 0.017) than the strain #5001 (14.2 ± 0.45 cm 2 ), indicating that the siderophores production by the strain #5029 was most efficient than the strain #5001 ( Figure 8).

Indole 3-Acetic Acid Production by Trichoderma Strains
Both of the strains used in this study produced IAA in the L-tryptophan supplemented media. The IAA produced by the strain #5029 (6.13 ± 0.31 µg/mL) was significantly higher (p = 0.03) than the IAA produced by the strain #5001 (4.49 ± 0.19 µg/mL).

Siderophore Production by Trichoderma Strains
Visible orange halo formation surrounding the Trichoderma growth on CAS agar media indicate that both the strains produced siderophores on CAS agar media. Halo zone developed by the strain #5029 (18.33 ± 0.5 cm 2 ) was significantly larger (p = 0.017) than the strain #5001 (14.2 ± 0.45 cm 2 ), indicating that the siderophores production by the strain #5029 was most efficient than the strain #5001 ( Figure 8).

Indole 3-Acetic Acid Production by Trichoderma Strains
Both of the strains used in this study produced IAA in the L-tryptophan supplemented media. The IAA produced by the strain #5029 (6.13 ± 0.31 μg/mL) was significantly higher (p = 0.03) than the IAA produced by the strain #5001 (4.49 ± 0.19 μg/mL).

Siderophore Production by Trichoderma Strains
Visible orange halo formation surrounding the Trichoderma growth on CAS agar media indicate that both the strains produced siderophores on CAS agar media. Halo zone developed by the strain #5029 (18.33 ± 0.5 cm 2 ) was significantly larger (p = 0.017) than the strain #5001 (14.2 ± 0.45 cm 2 ), indicating that the siderophores production by the strain #5029 was most efficient than the strain #5001 ( Figure 8).

Discussion
Overall results indicate that Trichoderma strains used possessed different mechanisms that provided effective antagonism against Pyrrhoderma noxium. The Trichoderma strains used in this study produced coil around P. noxium, while the flattened, enlarged hyphal tips produced by the strains might involve in the direct penetration and ultimately, lead to leakage of cytoplasm and death. Findings also confirm that the strains #5001 and #5029 can grow faster and easily compete for space and nutrients with the slower growing P. noxium, which can thus lead to the elimination of the pathogen as previously reported by other researchers [98][99][100][101].
The induced expression of qid genes in the strains #5001 and #5029 indicate that Trichoderma strains were able to sense the presence of P. noxium and prepared for defense by protecting their cell wall from the lytic enzymes of P. noxium. Similar observations were also made for T. harzianum against F. solani and T. virens against F. oxysporum [23,24]. While induced expression of chit and endo genes in Trichoderma strains #5001 and #5029 indicate that chitinolytic enzymes produced by the strains possibly played a significant role in inhibiting the growth of P. noxium ( Figure 6). The induced expression of chitinase encoding gene in recombinant T. harzianum strain was previously reported to increase its antagonizing activity against R. solani [21]. Similarly, expression of chitinase gene (chi46) of Trichoderma reesei PC-3-7 in E. coli showed lytic activity against the cell wall of Fusarium oxysporum [102]. In a different study, induced expression of endo and chit genes were detected in T. harzianum, when it was cultured in the medium supplement with inactivated F. solani cell wall [26]. Therefore, the induced expression of the genes encoding the chitinase and endochitinases in the strains #5001 and #5029 may be associated with degradation of chitin in the cell wall of P. noxium.
Over expression of prb and pra in the strains #5001 and #5029 indicate that proteinases may play an important role in biological control by the Trichoderma strains used in the study ( Figure 6). Induced expression of subtilisin-like protease prb1 gene was previously detected in T. harzianum transformants during the confrontation with Rhizoctonia solani and also in the presences of cell walls of R. solani [29]. Moreover, in a greenhouse study, these transformants significantly reduced the disease caused by R. solani in cotton plants [29]. Similarly, the expression of trypsin-like protease pra1 gene was previously reported in T. atroviride during physical contact with R. solani [103]. Therefore, induced expression of the gene encoding the proteases in the strains #5001 and #5029 may degrade the protein components in the cell wall of P. noxium and cause cell lysis.
The crude extracts of Trichoderma strains #5001 and #5029 were effective in inhibiting the growth of P. noxium (Figure 3c). However, further studies are underway to identify the antifungal metabolites present in the crude extracts of the strains #5001 and #5029. In this study, the VOCs of the strains #5001 and #5029 were also efficient in inhibiting the growth of P. noxium (Figure 3b), therefore these strains may be able to be used as a biofumigation source to manage P. noxium over the application of toxic fungicides. Similarly, an inhibitory effect by the VOCs of T. atroviride, T. harzianum and T. virens on P. noxium was reported by Schwarze et al. 2012 [5] and Burcham et al., 2017 [104]. Inhibitory effects of VOCs of T. reesei on Colletotrichum capsici, Aspergillus flavus, A. niger, Cladosporium cucumerium, Fusarium solani, F. udum, Rhizoctonia solani, Rhizoctonia bataticola, Rhizopus stolonifera and Pythium species have also been observed by several researchers [105,106].
The VOC profiles of the strains #5001 and #5029 were dominated by sesquiterpenes. These results were also aligned with Guo, (2020) study [107], where 52.69% of the volatile profile of T. reesei QM6a was dominated by sesquiterpenes. Sesquiterpenes compounds are lipophilic compounds, therefore can cause osmotic control failure in cell membranes of competitors [108]. On the other hand, they can also act as a solvent to facilitate the transportation of toxic compounds through the membrane into the cell [108]. It remains to be determined if any of these sesquiterpenes inhibit P. noxium. However, sesquiterpenes from a number of Trichoderma species have been reported to have antibacterial and antifungal activities [109][110][111]. The second most dominant VOC detected in this study was β-cedrene and this compound produced by T. longibrachiatum was reported for antifungal activity against C. lagrnarium and B. cinerea [112,113]. Cedrene production by T. harzianum, T. hamatum, T. reesei and T. velutinum was also documented by researchers [107,108,114]. The β-cedrene produced by the strains # 5001 and # 5029 may have antifungal properties, and this finding is worthy of further studies. The 3-methyl-1-butanol, 2-methyl-1-propanol, ethanol and ester compound detected in this this study were also produced by various Trichoderma species and reported to have antimicrobial activity [54,55,58,[115][116][117][118].
Compared to monocultures, co-cultivation settings were found to induce specific changes in the fungal VOC emissions, indicating that there is an interaction between P. noxium and Trichoderma strains. Similar results were observed during the interaction between T. hamatum and Laccaria bicolor [114]. Altered VOC profiles were also observed during the interaction between A. alternata with F. oxysporum and L. bicolor with T. harzianum, T. hamatum and T. velutinum [114,120]. Production of new VOCs and concentration changes in co-culture settings may be required for the inhibition of a competitor. 13 out of 16 compounds seen in a pure P. noxium culture ( Figure S2 and Table S2) were not detected in the co-cultures of strain #5001-P. noxium or in the co-cultures of strain #5029-P. noxium. Overall, co-culture profiles showed high similarity with the profile of the Trichoderma strains rather than of P. noxium (Table 2 and Figures 4 and 5). This confirmed the inhibitory effect of Trichoderma strains on the growth of P. noxium growth in the co-culture set up and also the inhibitory effect on the production of P. noxium's volatile compounds. Therefore, the new compounds detected in the co-culture set were high, likely produced by Trichoderma strains.
In addition to the antifungal property, VOCs of Trichoderma species were also reported to be associated with plant growth promotion [50][51][52][53][54]. This study clearly shows that VOCs from the strains #5001 and #5029 were effective in increasing the plant biomass, chlorophyll content, shoot and root length (Figure 7). Similar results were reported for the VOCs of T. atroviride, T. virens and T. asperellum on growth parameters of A. thaliana [60] and T. viride on tomato seedlings [54]. The results obtained in this study contrast the results reported by Nieto-Jacobo et al. 2017 [60], where VOCs of T. reesei did not show any effects on plants. This contrasting result may be due to the shorter volatile exposure period, the media used for growing Trichoderma (MS media) and strain specificity.
The Trichoderma strains used in this study also produced siderophores. Therefore, if these strains were used in the environmental applications they might possibly compete and suppress the growth of P. noxium via rapidly acquired available iron before the pathogen. Furthermore, siderophores producing Trichoderma species were also reported to help the transformation of insoluble Fe 3+ to plant absorbable Fe 2+ and boosts the plant uptake [37,63,64], while, the IAA produced by the Trichoderma strains #5001 and #5029 may asset the recovery of P. noxium infected trees via modifying the root architecture, resulting in development of new roots which aid the efficient nutrient and water uptake by the plant.

Conclusions
Mycoparasitic ability of Trichoderma strains #5001 and #5029 against P. noxium was confirmed and the results showed that during the initial stage of interaction Trichoderma hyphae often coil around the P. noxium and attach to the host via appressorium like structure. After these mycelial interactions, the genes related to the production of cell wall degrading enzymes are expressed and facilitate the penetration of Trichoderma hyphae through the P. noxium cell wall. Similarly, the volatile and diffusible compounds produced by Trichoderma strains also played significant roles in inhibiting the growth of P. noxium. It is conceivable that, during the interaction between Trichoderma and P. noxium, all these mechanisms may be operative simultaneously in order to inhibit the growth of the pathogen. The ability of Trichoderma strains to produce IAA and siderophores indicate the possibility of these biological control agents to promote plant growth in environmental applications. The overall study confirmed that Trichoderma strains can be used as potential biological control agents against P. noxium in environmental applications especially for protection of heritage fig trees.
Supplementary Materials: The following supporting information can be downloaded at: https: //www.mdpi.com/article/10.3390/jof8101105/s1, Figure S1: Growth inhibitory effect of the (b) Trichoderma strain #5001, (c) strain #5029 on P. noxium strain B after 5 days of inoculation in dual culture (T: Trichoderma and P: P. noxium), (a) P. noxium strain B grown alone; Figure S2: Chromatogram of volatile compounds produced by strain B (above zero level); chromatogram of volatile compounds detected in the PDA media control (below zero level). Numbers refer to compounds listed in Table S2; Figure S3: Effect of volatile compounds on the growth of Arabidopsis thaliana in a shared atmosphere with Trichoderma strains #5001 and #5029 for 12 days. (a) Control plants exposed to PDA medium, (b) plants exposed to Trichoderma strain #5001 and (c) plants exposed to the strain #5029; Table S1: Primer pairs used for gene expression analysis by qPCR technique; Table S2: Volatile compounds produced by strain B tentatively identified through SPEM GC/MS analysis.