Trichoderma species from rhizosphere of Oxalis corymbosa release volatile organic compounds inhibiting the seed germination and growth of Echinochloa colona

2.1 Isolation of Trichoderma spp. from the rhizosphere soil

In the present study, Trichoderma spp. were isolated from the rhizosphere soil of O. corymbosa, which were collected from sugarcane fields in Guangxi province, China. One gram of soil from each sample was taken in a 250 mL conical flask with 99 mL of sterile distilled water and grinding beads, and then agitated for 10 min to prepare 10⁻¹ dilution. This suspension was used for serial dilutions up to 10⁻⁵ using serial dilution method. Two hundred microliter of the suspension from 10⁻³, 10⁻⁴ and 10⁻⁵ were plated separately on petri plates containing potato dextrose agar (PDA) and incubated at 25 °C in the dark. For microscopic morphology, photographs were taken with a Nikon Ni-E fluorescence microscope (Tokyo, Japan) connected to a DP controller digital camera.

2.2 DNA extraction, PCR amplification and sequencing

Fungal mycelia grown on PDA for 4–7 d was scraped off and collected into 1.5 mL tube. Genomic DNA was extracted using a modified CTAB protocol described in the reference (Guo et al., 2000). The internal transcribed spacer (ITS) and translation elongation factor 1-alpha (tef 1) were amplified with the following two primer pairs: ITS1 (5′-TCCGTAGGTGAACCTGCGG-3′) and ITS4 (5′-TCCTCCGCTTATTGATATGC-3′) for ITS (Kullnig-Gradinger et al., 2002) and TEF1-728F (5′-CATCGAGAAGTTCGAGAAGG-3′) (Carbone and Kohn, 1999) and TEF1-LLErev (5′-AACTTGCAGGCAATGTGG-3′) (Jaklitsch et al., 2005) for tef 1.PCR reactions were performed in 25 μL mixture of 12.5 μL 2 × Taq PCR StarMix, 8 μL ddH₂O, 2 μL DNA and 1.25 μL of each primer using a BIO-RAD T100TM Thermal Cycler. The thermocycling program for ITS was 94 °C for 2 min or tef 1 was 95 °C for 5 min, followed by 40 cycles of 94 °C for 30 s, 55 °C annealing for 30 s, 72 °C for 1 min, and 10 min at 72 °C. The PCR products were visualized using a 1% agarose gel. Successfully amplified PCR products were sequenced by Sangon Biotech (Shanghai) Co., Ltd.

2.3 Phylogenetic analyses

For sequences alignment, the obtained sequences were initially checked with BioEdit (Hall 1999) and ran BLAST searches in the NCBI database. The ITS and tef 1 sequences of reference Trichoderma species were retrieved from the NCBI database. Nectria berolinensis (CBS 127382) and Nectria eustromatica (CBS 121896) were selected as out-group. All sequences were aligned using Clustal W within the MEGA v.7.0.26 software (Kumar et al., 2016). Phylogenetic analysis included maximum-likelihood (ML)) and Bayesian inference (BI) methods using RAxML v.7.2.8 (Stamatakis, 2006). In the case of the ML analysis, the GTRCAT model was implemented and clade stability was determined based on 1000 bootstrap (BS) replicates of rapid ML guided replications performed in Raxml v.7.2.8. BI analysis adopted MrBayes v. 3.2.1 following the scheme of Cheng et al. (Cheng et al., 2015) and basing on the Markov Chain Monte Carlo (MCMC) method (Ronquist et al., 2012). Each locus was optimized as a partition in the combination analysis. Mrmodel test v. 2.3 (Posada and Crandall, 1998) used Akaike Information Criterion (AIC) to determine the best fit evolutionary model (GTR + I + G). Two MCMC chains were run in the random tree with a total of 1 million generations sampling once every 100 generations. When the average standard deviation of the separation frequency was<0.01 of the sampled trees’ burn-in phases from the BI analysis. The first 25% of the samples were discarded and posterior probability (PP) values were calculated. The trees were viewed and edited with Figtree v.1.3.1 (Rambaut and Drummond, 2010). The branch support values of PP/BS were higher than 0.6/60%, which were shown at the nodes.

2.4 Inhibition activity of Trichoderma spp. against E. colona using double-plate chamber method

In this experiment, six Trichoderma strains were selected for testing inhibitory activity against E. colona according to the method of Fernando (Fernando et al., 2005) with some modifications. Briefly, E. colona seeds were surface sterilized in a 75% ethanol and 2% bleach solution. E. colona was taken to the petri dish (diameter 9 cm) containing 4 mL of autoclaved 0.8% agar and covered with another petri dish containing Trichoderma spp. cultured on PDA for two days, and then two plates were sealed with Parafilm to obtain a double-plate chamber. The average distance between autoclaved 0.8% agar surface of bottom dish and PDA agar surface of top dish was 1.5 cm (Fig. 1). The mixture was incubated for 140 h at 28 °C (±2) under in the dark. The control plates had only autoclaved PDA agar in upper dish. The germination of E. colona was recorded at 24 h intervals by visual inspection (Hong et al., 2018).

Close

Fig. 1

Inhibition activity of Trichoderma spp. against E. colona using double-plate chamber method.

Export to PPT

The same method was used to planta assays, which autoclaved soil was in replaced of 0.8% agar to culture E. colona at bottom petri dish. The mixture was incubated for 72 h at 28 °C (±2) under in the dark. The control plates had only autoclaved PDA agar at top petri dish. The germination of E. colona was recorded at 24 h intervals by visual inspection. All treatments were performed in triplicate.

2.5 Extraction and analysis on E. colona endogenous hormone

The indole acetic acid (IAA), GA, Zeatin (ZT), and ABA of E. colona were determined by HPLC. Accurate 0.2 g sample was grinded in a mortar, extracted with 1 mL pre-cooled 70–80% methanol solution (pH = 3.5) overnight at 4 °C, and centrifuged at 4 °C 12000 × g for 10 min. The supernatants were collected, and the residue was extracted again using 0.5 mL 70–80% methanol solution at 4 °C for 2 h. The supernatants were combined and evaporated to about 0.5 mL with rotary vacuum gibberellins at 40 °C. The evaporated samples were repeatedly extracted using petroleum ether for 2–5 times. The combined petroleum ether phase was added triethylamine to adjust pH = 8.0, added polyvinylpyrrolidone (PVPP), shocked for 20 min, and centrifugated at 4 °C 12000 × g for 5 min. The supernatant was adjusted pH to 3.0 with hydrochloric acid, extracted with ethyl acetate 3 times and combined with the ester phase. The ethyl acetate phases were combined, dehydrated, and concentrated to dryness under reduced pressure. The dry extracts were dissolved in 1 mL ethyl acetate, vortexed, and filtered with 0.22 μm Millipore filter membrane for HPLC analysis.

The samples were detected by HPLC (SHIMADZU LC-20A) equipment with PDA detector and C18 column (250 × 4.6 mm, 5 μm) at 35 °C. Mobile phases were acetonitrile/water (1% formic acid) (V:V = 45:55) at a flow rate of 0.8 mL/min. The injections volume was 10 μL and the total running time was 30 min. The IAA, GA3, ZT and ABA of the E. colona were compared with the standards. All treatments were performed in triplicate.

2.6 Identified VOCs of Trichoderma spp. using SPME-GC–MS

To detect the VOCs, the Trichoderma spp. was incubated with in sealed glass bottles containing 10 mL PDA medium at 25 °C for 140 h. The VOCs released by Trichoderma spp. incubated at 60 °C were adsorbed for 1 h with a 2 cm solid-phase microextraction (SPME) fiber with DVB/CAR/PDMS film (Supelco, Bellefonte, PA, USA). The VOCs was desorbed from SPME and analyzed by GC–MS (Agilent 7890 N/5975, USA) equipped with HP-5MS capillary column (30 m × 0.25 mm ID; film thickness 0.25 μm). The oven program was as following: 40 °C for 5 min, at a rate of 4 °C/min to 120 °C for 3 min, then desorbed at 10 °C/min to 220 °C for 5 min, and a final at 30 °C/min to 280 °C for 3 min. In the operating conditions, helium was used as the carrier gas (1 mL/min) and the total running time was 48 min. The mass fragments were analyzed using electron ionization at 70 eV and a scan rate of 1.9 scans⁻¹. Fragments were read from 40 to 450 m/z. The VOCs were identified according to NIST17. L mass spectrometry database.

2.7 Effect of synthetic compounds on E. colona growth

The candidate VOCs from Trichoderma spp. were purchased from Aladdin Biochemical Technology Co., Ltd. (Shanghai, China) and evaluated individually for their inhibition ability to plant growth. For the volatile-exposure bioassay, we performed the divided petri dishes (I-plates) assays. To exactly evaluate activity of candidate VOCs, E. colona was pre-geminated in the 0.8% agar. The E. colona seedlings were grown for 72 h in half of an I-plate as described above. Each compound was dissolved in ethanol to a final concentration of 50, 100, 200, 500 and 1000 μM, and applied to the other half of the plate, respectively. The ethanol was control treatment in the other half of the plate. The plates were incubated for 72 h lighting (12-h light–dark cycles) at 27 ℃ (±2) to allow plant growth. Then the shoot length, root length and fresh weights of plants were measured in 3 days after incubation. All bioassays were conducted three times with 20 seedlings per treatment.

2.8 Statistical analysis

The data were subjected to analyses of variance (ANOVA) by using SPSS 24.0 (SPSS Inc., USA). For percentage data, an arcsine transformation was applied prior to ANOVA. The least significant difference (LSD) test was used to calculate the inhibition assay for shoot length, treatments of the VOCs and control at α < 0.05 (Kamaruzzaman et al., 2021). Significant differences were also based on independent-sample T test (P < 0.05, P < 0.01). All experiments were performed at least twice.

3.1 Colony morphology and molecular identification of Trichoderma spp.

Total 44 Trichoderma strains were isolated from the rhizosphere soil of O. corymbosa. Five strains were selected and one ineffective strain was control in this study. Trichoderma strains were cultured on PDA and SNA at 25 °C under alternating 12 h light and 12 h darkness. Colonies were photographed after 4 days on PDA and SNA. Conidia were photographed after 14 days of incubation on SNA (Fig. 2). After 4 days, the colony rapidly developed white, and villous hyphae radiated to the peripheral. Massive conidiation started on the third day of culture. The juvenile conidia appeared yellowish-green and gradually turned dark green within 24 h especially for TZ9-107 and TZ10-103. On the fourth day, the hyphae were nearly full of radiating petri dishes, with the exception of TZ9-107. After 14 days, the mycelia of Trichoderma strains were significantly reduced, transparent and smooth. The conidia were smooth, spherical, or ovoid-shaped, translucent to appeared yellowish-green.

Close

Fig. 2

Morphological characteristics of Trichoderma spp.. Colony on PDA (A1-F1), Colony on SNA (A2-F2), conidia on SAN (A3-A5). Scale bar are 100 µm, 20 µm, 10 µm, respectively.

Export to PPT

To identify six Trichoderma strains, the sequences of ITS and tef1 regions from 64 strains representing different clades of Trichoderma, were analyzed by the methods of BI and ML (Table 1). The ML analysis with the BI analysis showed consistent tree topology, therefore the ML phylogeny was used as basal tree shown in Fig. 3. Phylogenetic analysis sequence data supported combined ITS and tef 1 that the TZ1-113 subclades corresponding to T. koningiopsis (PP/BS: -/63) pertain to the Koningii clade, the TZ2-102 and TZ14-110 subclades corresponding to T. afroharzianum (PP/BS: 0.99/84, 0.79/-) pertain to the Harzianum clade, the TZ3-103 subclades corresponding to T. atroviride pertain to the Atroviride clade, the TZ9-107 subclades corresponding to T.virens (PP/BS: 0.96/96) pertain to the Virens clade, the TZ10-103 subclades corresponding to T. asperelloides (PP/BS: 0.99/99) pertain to the Hamatum clade. Finally, 6 strains were considered as 5 different species in Trichoderma according to morphological features and sequences analysis.

Close

Fig. 3

Phylogenetic tree of Trichoderma species based on the combined gene sequences of ITS and tef 1. Maximum‐likelihood (ML) tree constructed with ITS and tef 1 sequences of 64 strains representing the Trichoderma. The Bayesian posterior probabilities greater than 60% (PP) and RAxML bootstrap support values greater than 60% (BS) are given at the nodes (PP/ BS). Nectria berolinensis (CBS 127382) and Nectria eustromatica (CBS 121896) were used as the outgroup. The present isolates are marked in bold. Note, Hypocrea atroviridis is a heterotypic synonym of Trichoderma atroviride.

Export to PPT

3.2 Inhibition of E. colona by Trichoderma spp

In the double-plate chamber experiment, germination of E. colona was significantly suppressed by the Trichoderma spp. compared with the control. The inhibition percentages were nearly 90% (Fig. 4A, Table 2). This result indicated that VOCs were the major regulators inhibiting germination of E. colona.

Close

Fig. 4

Inhibition of E. colona growth by VOCs emitted Trichoderma spp.

Export to PPT

Close

Fig. 4

Inhibition of E. colona growth by VOCs emitted Trichoderma spp.

Export to PPT

To demonstrate the effect of VOCs, we created the in planta bioassay system. In the system, shoot growth of E. colona was evidently suppressed by the Trichoderma spp. compared with the control, and the inhibition percentages were 100% (Fig. 4B). These results showed that the VOCs of Trichoderma had inhibitory effect on the growth of E. colona.

3.3 Identification of volatile organic compounds produced by Trichoderma spp

Since particular VOCs might play key roles in inhibiting plant growth, we characterized the VOCs emitted by 6 Trichoderma strains were collected using SMPE and analyzed by GC–MS. The background compounds were excluded from the control (only PDA). Venn diagrams were constructed using control and the VOCs from five inhibitory Trichoderma strains (Chen et al., 2021) (Fig. 5A). Statistics showed that the compound dimethyl phthalate (1) was excluded from control. Venn diagram showed that six common compounds in five different inhibitory Trichoderma strains. The common compounds were 2,4,6-trichloroanisole, dimethyl phthalate and ethanol, hexadecanoic acid, ethyl ester, linoleic acid ethyl ester and phenylethyl alcohol (Fig. 5B). Compared to VOCs in non-inhibitory Trichoderma strain, some common compounds were identified as ethanol, hexadecanoic acid, ethyl ester, linoleic acid ethyl ester and phenylethyl alcohol (Fig. 5C). Ethanol was excluded according to Venn diagrams anlysis. In addition, 2,4,6-trichloroanisole was not present in the VOCs of the non-inhibitory strain. We found that the area of linoleic acid ethyl ester, hexadecanoic acid, ethyl ester and 2,4,6-trichloroanisole was lower content in VOC of the non-inhibitory strain than five strains with inhibitory activity. Results showed that five compounds were screened as candidate bioactive compounds (Table 3). For detailed five compounds chromatographic information, see the supplementary materials (Fig. S2-6).

Close

Fig. 5

Analysis of VOCs produced by Trichoderma spp. The abscissa of the bar plot below the Venn diagram is the Trichoderma strain number, the ordinate is the number of VOCs.

Export to PPT

Table 3 The volatile organic compounds (VOCs) produced by Trichoderma spp.

N°	Compound	Strain	Peak Area (%) 1	RT (min) 2	Purity 3	Molecular Formula (MF) 4	Molecular weight (MW) 5	Chemical structure
1	Phenylethyl alcohol	TZ1-113	0.72	12.765	≥99%	C8H10O
		TZ2-102	36.9	12.825
		TZ3-103	2.93	12.749
		TZ9-107	1.42	12.765			122.073
		TZ10-103	2.83	12.782
		TZ14-110	53.29	12.754
2	2,4,6-Trichloroanisole	TZ1-113	0.43	20.113	≥98%	Cl3C6H2OCH3
		TZ2-102	0.79	20.125
		TZ3-103	0.56	20.103			209.941
		TZ9-107	5.59	20.108
		TZ14-110	0.48	20.103
3	Hexadecanoic acid, ethyl ester	TZ1-113	0.6	35.563	≥99%	C18H36O2
		TZ2-102	3.99	35.541
		TZ3-103	0.43	35.563			284.272
		TZ9-107	0.84	35.547
		TZ10-103	0.15	35.558
		TZ14-110	13.01	35.53
4	10(E),12(Z)-Conjugated linoleic	TZ2-102	0.73	36.85	≥80%	C18H32O2
		TZ3-103	0.38	36.85			280.24
		TZ10-103	0.13	36.85
		TZ14-110	2.14	36.845
5	Linoleic acid ethyl ester	TZ1-113	0.56	37.041	≥97%	C20H36O2
		TZ2-102	1.76	37.031
		TZ3-103	0.83	37.041
		TZ9-107	1.00	37.031			308.272
		TZ10-103	0.28	37.036
		TZ14-110	8.24	37.02

Note: RPA ¹ = Relative Peck Area; RT ² = Retention Time; Purity ³ = Compound of standard for purity; MF ⁴ = Molecular Formula; MW ⁵ = Molecular Weight.

3.4 Inhibition effect of VOCs on E. colona

To determine their potential biological activity, the pure standards of five VOCs were purchased to test their inhibitory activity against E. colona. These results showed that the five compounds had different degrees of inhibition on plant growth, comparing with ethanol control treatment. The inhibitory rates for shoot length was approximately 3%, 26%, 28%, 31%, 21% (50, 100, 200, 500 and 1000 μM) by conjugated linoleic acid, 9%, 17%, 24%, 52% (100, 200, 500 and 1000 μM) by 2,4,6-trichloroanisole, 45%, 39%, 41% (200, 500 and 1000 μM) by linoleic acid ethyl ester, 14%, 19% (500 and 1000 μM) by phenylethyl alcohol and 14% (1000 μM) by hexadecanoic acid, ethyl ester compared to growth under control conditions (Fig. 6A and B). Our results suggested that the VOCs emitted by Trichoderma were able to inhibit the growth of the shoot in the main root of E. colona.

Close

Fig. 6

Inhibition effect of VOCs on E. colona.

Export to PPT

Close

Fig. 6

Inhibition effect of VOCs on E. colona.

Export to PPT

3.5 The level of phytohormones in E. colona

We quantitatively determined IAA, ABA, GA and ZT accumulation in E. colona seedlings with Trichoderma treatments. As expected, the VOCs led to a significant difference in the endogenous levels of IAA, ABA and ZT in E. colona. Compared with the controls, ZT increased from 0.376 to 0.751, IAA from 3.251 to 9.101, and ABA from 2.903 to 8.721. However, there were no significant difference in the GA between treatments and control (Table 4). The above results indicated that the VOCs emitted by Trichoderma could inhibit growth of E. colona by regulating the levels of IAA, ZT and ABA in E. colona.