Functional analysis of eliciting plant response protein Epl1-Tas from Trichoderma asperellum ACCC30536

Eliciting plant response protein (Epl) is a small Trichoderma secreted protein that acts as an elicitor to induce plant defense responses against pathogens. In the present study, the differential expression, promoter analysis, and phylogenetic tree analysis of Epl1-Tas (GenBank JN966996) from T. asperellum ACCC30536 were performed. The results showed Epl1-Tas could play an important role in the interaction between T. asperellum ACCC30536 and woody plant or woody plant pathogen. Furthermore, the effect of the Escherichia coli recombinant protein rEpl1-e and the Pichia pastoris recombinant protein rEpl1-p on Populus davidiana × P. alba var. pyramidalis (PdPap) was studied. In PdPap seedlings, rEpl1-e or rEpl1-p induction altered the expression levels of 11 genes in the salicylic acid (SA, three genes), jasmonic acid (JA, four genes) and auxin (four genes) signal transduction pathways, and five kinds of enzymes activities The induction level of rEpl1-p was significantly higher than that of rEpl1-e, indicating that rEpl1-p could be used for further induction experiment. Under 3 mg/mL rEpl1-p induction, the mean height of the PdPap seedlings increased by 57.65% and the mean lesion area on the PdPap seedlings leaves challenged with Alternaria alternata decreased by 91.22% compared with those of the control. Thus, elicitor Epl1-Tas could induce the woody plant resistance to pathogen.

A plant-mediated response has been confirmed as a component of bio-control using Trichoderma spp. 1-3 , for example, T. virens on cotton and maize 4 , T. asperellum on soybean 5 , and T. harzianum on beans 6 . Some small molecular weight proteins secreted by Trichoderma spp. can induce plant defense responses. One of the largest groups of proteins secreted by Trichoderma is the small secreted cysteine-rich proteins (SSCPs), which are identified as being 100 aa long and contain four or more cysteine residues 7 . SSCPs are subdivided into four groups: (i) Hydrophobins and hydrophobin-like proteins; (ii) elicitor-like proteins; (iii) proteins with similarity to MAP kinase repressed secreted protein 1 (MRSP1); and (iv) SSCPs with no attribution to any functional category 7,8 .
Eliciting plant response protein (Epl) secreted by Trichoderma is an elicitor-like SSCP belonging to the cerato-platanin family, and has four cysteines that form two disulfide bonds 9 . Recently, Epl was confirmed to act as an elicitor to induce plant defense responses 5 , which provided insights into the mechanisms underlying the processes of Trichoderma-plant recognition, defense elicitation, and induction of resistance. Small molecular protein Sm1 (also known as Epl1) from T. virens Gv29-8 is non-toxic to plants and microbes. Instead, native, purified Sm1 from T. virens Gv29-8 triggered the production of reactive oxygen species in rice and cotton seedlings, and induced the expression of defense-related genes, both locally and systemically, in cotton 10 . Overexpression of Sm1 in T. virens Gv29-8 significantly enhanced levels of disease protection of maize seedlings challenged with the pathogen Colletotrichum graminicola, and this protection was associated with notable induction of jasmonic acid and green leaf volatile-biosynthetic genes, which demonstrated that the activity of a functional elicitor was required for T. virens-mediated induced systemic resistance (ISR) in maize 4 . In addition, glycosylation of T. virens Sm1 maintained the protein in a monomeric form, which elicited ISR, and deglycosylation led to the formation of an Sm1 dimer, which did not elicit ISR 11 . Similarly, overexpression of Epl1 in T. atroviride IMI206040 promoted disease resistance against many pathogens, and induced the expression of the peroxidase and the α-dioxygenase encoding genes in tomato 12 . Moreover, the absence of Epl-1 in T. harzianum ALL42 affected not only the recognition of T. harzianum ALL42 as a symbiotic fungus by beans 6 , but also the transcription of tomato defense-related genes during the T. harzianum-tomato interaction 13 , which indicated that Epl-1 could be important for plant protection in Trichoderma. Besides, EplT4 from T. asperellum T4 was transformed into Pichia pastoris, and soybean leaves were protected by the recombinant EplT4 monomer against the pathogen Cercosporidium sofinum 5 . Epls have been studied in T. virens, T. atroviride, T. harzianum, and T. asperellum, only in the presence of herbage plant, including dicot (cotton, tomato, bean and soybean) and monocot (rice and maize) plants. However, there are no reports about the effect of Epl as an elicitor on woody plants.
In our previous study, the eliciting plant response protein gene Epl1-Tas from T. asperellum ACCC30536 was cloned. It encodes a protein comprising 138 aa with a molecular weight of 12.6 kDa. And the Escherichia coli recombinant protein rEpl1 (rEpl1-e) was obtained and could promote poplar tissue-culture seedling growth. In the present study, the transcription of Epl1-Tas in T. asperellum ACCC30536 was studied using quantitative real-time reverse transcription polymerase chain reaction (qRT-PCR) under eight inducing conditions. Similar sequences to Epl1-Tas and their promoters were analyzed in four Trichoderma genomes to further certify the genome wide biocontrol function of Epl1-Tas. Epl1-Tas was also recombinantly expressed in P. pastoris (rEpl1-p). Under rEpl1-e or rEpl1-p induction, the transcription of 11 genes related to hormone signaling and 5 physiological enzymes activities in Populus davidiana × P. alba var. pyramidalis (Pdpap) seedlings were detected, and the disease resistance against phytopathogen Alternaria alternata and the growth of PdPap transplanted seedlings were also investigated. The results could provide theoretical support and a practical reference for the development of biological small molecular protein elicitors from T. asperellum in forests.

Results
Promoter analysis and multiple sequence alignment of Epl1-Tas. The promoter of Epl1-Tas is 1,476 bp in length. Many cis-elements known to be involved in stress responses such as MYB, MYC, and WRKY motifs were found in the Epl1-Tas promoter (see Supplementary Fig. 1). The disease resistance element ' ASF1MOTIFCAMV' (S000024) appears three times and the disease resistance response element 'BIHD1OS' (S000498) appears once in promoter region of Epl1-Tas ( Supplementary Fig. 1), suggesting that Epl1-Tas might be regulated by the transcription factors that bind to MYB motifs.
Multiple sequence alignment of similar sequences from 25 fungal strains revealed that the amino acid sequences of these proteins were conserved, with identities ranging from 50 to 92%. Epl1-Tas from T. asperellum ACCC30536 shared the highest similarity (92%, E value = 7e-95) with a known elicitor Epl1 (ABF73692) from T. atroviride ATCC74058 9,14 . Furthermore, four cysteine residues that form two disulfide bonds were highly conserved among the 25 sequences (Fig. 1).
Differential expression of Epl1-Tas under eight inducing conditions. The response to eight biotic stresses of the eliciting plant response protein gene Epl1-Tas in T. asperellum ACCC30536 was investigated using qRT-PCR (Fig. 2). Epl1-Tas transcription was upregulated strongly at 72 h in response to both 1% A. alternate cell walls and 5% A. alternate fermentation liquid, by 76.64 (2 6.26 )-fold and 150.12 (2 7.23 )-fold, respectively (Fig. 2D,E). Epl1-Tas transcription was also upregulated in response to the 1% stem powder and 1% leaf powder of PdPap seedlings, by 362.04 (2 8.5 )-fold and 44.94 (2 5.49 )-fold at 12 and 8 h, respectively (Fig. 2G,H). Interestingly, the  transcription of Epl1-Tas was mainly upregulated under carbon source starvation (Fig. 2B). However, Epl1-Tas transcription was inhibited by minimal medium (MM) (0.5% glucose and 0.5% ammonium sulfate), nitrogen source starvation, and 1% root powder. The lowest transcription levels were −1.7 (2 −3.70 ), −1.43 (2 −3.43 ) and  Table 1) that were predicted to be members of the small molecular cerato-platanin protein family. There were three Epls in each Trichoderma spp. genome. All the Epls were predicted to be acidic proteins, had the same number of amino acids (138 aa), and contained one intron and two exons ( Table 1). They were divided into four groups by phylogenetic analysis (Fig. 3). Epl1s (Epl1-Tas, Sm1-Tvi, Epl1-Tat, and Epl1-Tha) from the four Trichoderma species were divided into two groups (Group1 and Group 4 in the phylogenetic tree), in which two Epl1s (Epl1-Tat and Epl1-Tha) were reported to be elicitors. The other eight Epl2s and Epl3s from the four Trichoderma species were placed in Group 2 and Group 3 in the phylogenetic tree ( Fig. 3), which were not reported as elicitors.
Promoter analysis showed that the promoter region of Epl1-Tas and its similar sequences contained many stress response elements; for example, disease resistance (S000024), disease resistance responses (S000498), and drought and low temperature responsive (S000418, S000135) (Fig. 4). Moreover, many hormone response elements, such as gibberellin (S000030), ethylene (S000037), and salicylic acid (S000142), were also identified in the promoters (Fig. 4). In particular, dehydration protein (S000409) with 24.22% and disease resistance (S000024) with 19.53% appeared most frequently in the promoters, and thus might play the most important roles in the bio-control function of Epl protein. The promoter of Epl1-Tas contains 14 elements and thus could have more functions than the promoter of Sm1-Tvi (reported), which only contains five elements (Fig. 4). Moreover, the disease resistance element ' ASF1MOTIFCAMV' (S000024) appeared three times and the disease resistance response element 'BIHD1OS' (S000498) appeared once in promoter region of Epl1-Tas, which indicated that Epl1-Tas could participate in the disease resistance response.
Among the three homologous Epls (Epl1-Tas, Epl2-Tas and Epl3-Tas) from T. asperellum ACCC30536 genome, Epl1-Tas has the most elements in its promoter (Fig. 4). Furthermore, only Epl1-Tas showed a high expression level in four inducing conditions, while Epl2-Tas had a lower expression level and the expression of Epl3-Tas barely detectable under the four inducing conditions (see Supplementary Table 2). Taken together, these results suggested that Epl1-Tas has multiple functions in T. asperellum ACCC30536.

SDS-PAGE analysis of the recombinant Epl1-Tas expressed in P. pastoris (rEpl1-p). Compared
with the control transformant GS115-pPIC9K, the transformant GS115-Epl1 showed a clear protein band with a molecular mass of approximately 12.6 kDa (Fig. 5). This result indicated that the recombinant protein rEpl1-p was successfully synthesized in P. pastoris.

Effect of recombinant rEpl1-e or rEpl1-p on 11 hormone signal related genes of PdPap seedlings.
To determine whether the recombinant rEpl1-e (Epl1-Tas expressed by E. coli obtained in our previous study) or rEpl1-p could stimulate a response in PdPap seedlings, the transcription levels of 11 genes related to plant hormone signaling were studied using qRT-PCR (Figs 6 and 7). The transcription levels of three genes in the SA signal transduction pathway (Fig. 6L) in PdPap seedlings were significantly induced by rEpl1-e ( Fig. 6A-C) and rEpl1-p ( Fig. 7A-C). The expression of NPR1 (encoding nonexpresser of PR genes 1) was downregulated from 6-12 h under rEpl1-e induction, and then upregulated to the same level as the control (Fig. 6A). The expression of NPR1 showed a transient decrease at 6 h under rEpl1-p induction, and then also increased to the same level as the control (Fig. 7A). The expression of TGA (encoding TGACGTCA cis-element-binding protein) was mainly downregulated under rEpl1-e and was lower than that of the control (Fig. 6B); however, the expression of TGA was mainly upregulated under rEpl1-e with a peak at 57.68 (2 5.85 )-fold compared with the control (Fig. 7B). Compared with the downregulated control, the expression of PR1 (encoding pathogenesis-related 1) was upregulated under rEpl1-e (
Effect of recombinant rEpl1-e or rEpl1-p on enzymes activities in PdPap seedlings. To gain a further insight into the PdPap seedlings response to rEpl1-e or rEpl1-p, the activities of five kinds of enzymes in the PdPap seedlings were analyzed under rEpl1-e or rEpl1-p induction (Fig. 8). The enzyme activities showed obvious differences under rEpl1-e or rEpl1-p induction. Compared with the control, the activities of the  growth-related indole acetic acid oxidase (IAAO) in PdPap seedlings slowly decreased under rEpl1-e (Fig. 8A) or rEpl1-p (Fig. 8B) induction. In addition, the activities of the defense-related peroxidase (POD), superoxide dismutase (SOD), polyphenol oxidase (PPO) and phenylalanine ammonia lyase (PAL) in PdPap seedlings were all higher than those of the control under rEpl1-e or rEpl1-p induction (Fig. 8C-J). The peaks under rEpl1-e induction were 77.01, 34.08, 110.42, and 66.37-fold higher than the control at 7, 7, 1 and 3 d, respectively (Fig. 8C,E,G,I). The peaks under rEpl1-p induction were 108.53, 104.06, 122.42, and 41.43-fold higher than the control at 5, 3, 1, and 7 d, respectively (Fig. 8D,F,H,J). In summary, recombinant rEpl1-e and rEpl1-p had important effects on growth-related and defense-related enzymes activities in PdPap seedlings. The induction effect of rEpl1-p was also significantly higher than that of rEpl1-e, based on significance analysis.

Effect of rEpl1-p on growth and disease resistance of PdPap transplanted seedlings.
To determine the ability of Epl1-Tas to promote the growth and induce disease resistance of PdPap, different concentrations rEpl1-p were used to induce PdPap transplanted seedlings (Fig. 9). The growth rate of the PdPap transplanted seedlings varied under the different concentrations of rEpl1-p (Fig. 9A). Compared with the control, the mean heights of the PdPap transplanted seedlings increased significantly under induction by 3 and 5 mg/mL rEpl1-p, by 57.65% and 51.18%, respectively (Fig. 9A').
Moreover, the disease-resistance ability of the PdPap seedlings against A. alternate also varied under the different concentrations rEpl1-p (Fig. 9B). Compared with the control, the mean lesion areas of leaves decreased significantly under 0.5, 1, 3, or 5 mg/mL rEpl1-p induction (Fig. 9A'). Especially, under induction by 3 and 5 mg/mL rEpl1-p, the mean lesion areas of leaves decreased by 91.22% and 93.17%, respectively (Fig. 9B'). Thus, rEpl1-p promoted PdPap transplanted seedlings growth and induced resistance to the pathogen A. alternata in PdPap seedlings, especially at 3 mg/mL.

Discussion
Previously, differential expression of Epl1 in Trichoderma was studied only in the presence of dicot and monocot herbage plants 4,9,10 . Sm1 (Epl1) from T. virens Gv29-8 was upregulated in the presence of the cotton and maize plants early in the interaction 4,10 , and Sm1 from different Trichoderma showed increased expression in presence of bean seeds 15 . In the present study, we also showed that Epl1-Tas from T. asperellum ACCC30536 was upregulated in the presence of the woody plant PdPap seedlings, stems, and leaves (Fig. 2G,H), which, combined with the results of previous studies, suggested that upregulated expression of Epl1 participates in the interaction between Trichoderma and plants ubiquitously. However, this increase in the expression of Epl1-Tas was not observed in the presence of plant roots, as it was for Sm1 from T. harzianum T37 15 (Fig. 2F), Furthermore, the transcription of Epl1 from T. atroviridis ATCC74058 was induced cell walls from the plant pathogen Rhizoctonia solani 9 . Similarly, in this study, a high transcription level of Epl1-Tas was induced by cell walls and fermentation liquid from the PdPap pathogen A. alternata. Thus, Epl1-Tas of T. asperellum ACCC30536 might not only play an important role during the interaction with herbage plants or their pathogenic fungi, but also in the interaction with woody plants and their pathogenic fungi, indicating the enriched functions of Epls.
T. jecorina, T. virens, and T. atroviride each have three paralogs of Epls (sm1) 7,14,16 . In our analysis, T. asperellum and T. harzianum also have three Epls (Table 1), which indicated that Epls as elicitor are ubiquitous in the Trichoderma genus and could be important for the bio-control of Trichoderma. In addition, the elicitor Sm1 from T. virens Gv29-8 (Sm1-Tvi) could induce local and systemic defenses in plants, including rice, cotton and maize 4,10-12,15,17 . Our analysis showed that compared with the promoter of Sm1-Tvi, the Epl1-Tas promoter has more types and numbers of regulatory motifs (6 types; 14 numbers vs. 4 types; 5 numbers), and the disease resistance element ' ASF1MOTIFCAMV' (S000024) appears three times and the disease resistance response element 'BIHD1OS' (S000498) appears once in promoter region of Epl1-Tas (Fig. 4). This indicated that Epl1-Tas might have more functions than Sm1-Tvi, and might play an important role in bio-control of T. asperellum ACCC30536, as suggested by Pozo et al. 18 . Furthermore, compared with its homologs in T. asperellum ACCC30536 (Epl2-Tas and Epl3-Tas), the Epl1-Tas promoter has more types and numbers of regulatory motifs (6 type; 14 numbers vs.  N) The plant hormone signal transduction of the KEGG software from the Kanehisa laboratory (http://www.genome.jp/kegg-bin/show_pathway?map04075). Different capital letters represent significant differences among different time points of treatment or the control group; different lowercase letter represent significant differences between the treatment and the control at the same time point; *significant difference between rEpl1-e and rEpl1-p treatments at the same time point. All significances were at p < 0.5. All experiments were performed three times. The quantitative data are presented in Supplementary  Table 3.
In the phylogenetic tree, Epl1s (Epl1-Tas, Sm1-Tvi, Epl1-Tat and Epl1-Tha) from four species of Trichoderma were divided into two groups (Group1 and Group 4), in which two Epl1s (Epl1-Tat and Epl1-Tha) were reported to be elicitors (Fig. 3). However, the other eight Epl2s and Epl3s from the four species were placed in Group 2 and Group 3 (Fig. 3), which were not reported as elicitors. The highest expression of Epl1-Tas among Group 4 Epl1s under four inducing conditions (see Supplementary Table 2) further indicated that Epl1 proteins play an important role in different Trichoderma, while Epl2 and Epl3 had no obvious effect. Thus, the promoter (Fig. 4), differential expression (see Supplementary Table 2), and phylogenetic tree (Fig. 3) analyses further suggested that Epl1-Tas functions in stimulating plant defense responses as an elicitor.
Furthermore, to study the function of Epl1-Tas from T. asperellum ACCC30536 to the woody plant PdPap seedlings, we examined the transcription of 11 genes related to hormone signal in PdPap seedlings under E. coli recombinant rEpl1-e or yeast recombinant rEpl1-p (Fig. 5) induction, respectively. Trichoderma spp. colonization could induce a transient increase in expression of defense-related genes of plant 19 . That observation was in agreement with our results that 11 hormone signaling-related genes in PdPap seedlings showed transiently upregulated expression under rEpl1-e or rEpl1-p induction (Figs 6 and 7). In the SA signal transduction pathway, NPR1 plays an important role in both innate immunity and systemic acquired resistance (SAR), and PR1 is often associated with plant host resistance. Although NPR1 is a key component in the induction of SAR-related PR1 expression, Different capital letters represent significant differences among different time points of treatment or the control group; different lowercase letter represent significant differences between the treatment and the control at the same time point; *significant difference between rEpl1-e and rEpl1-p treatments at the same time point. All significances were at p < 0.5. All experiments were performed three times. The quantitative data are presented in Supplementary Table 4. its overexpression did not lead to PR1 expression in Arabidopsis 20 . Similarly, although the transcription of NPR1 in PdPap seedlings was downregulated under rEpl1-e or rEpl1-p induction (Figs 6A and 7A), PR1 was upregulated compared with the control group, which was consistent with the results of Mou et al. 20 . The transient peak of PR1 was 2210.26 (2 11.11 )-fold higher at 2 d under rEpl1-e induction (Fig. 6C), and the transient peak of PR1 was 9607.86 (2 13.23 )-fold at 12 h under rEpl1-p induction (Fig. 7C), which showed rEpl1-e and rEpl1-p could trigger SAR in PdPap seedlings. In addition, in the JA signal transduction pathway, Octadecaniod-derivative responsive catharanthus AP2-domain protein gene ORCA3 is a jasmonic acid responsive gene and regulates jasmonate-responsive expression, which could induce ISR in the host plant 21,22 . In this study, ORCA3 showed higher expression in PdPap under rEpl1-e or rEpl1-p induction (Figs 6G and 7G). The highest peaks of ORCA3 were 349.71 (2 8.45 )-fold at 2 d under rEpl1-e induction (Fig. 6G) and 5007.93 (2 12.29 )-fold at 2 h under rEpl1-p induction (Fig. 7G), respectively, which showed that rEpl1-e and rEpl1-p, similar to jasmonic acid, could also trigger ISR in PdPap seedlings. These results showed that rEpl1-e and rEpl1-p affected SA and JA signal transduction pathways in PdPap seedlings; however, the transcription level under rEpl1-p induction was significantly higher than under rEpl1-e induction.
Alterations to defense-related enzyme activities in the leaves of PdPap seedlings were detected under rEpl1-e or rEpl1-p induction. When plants suffer damage, more O · 2 − is produced, in which case, SOD dismutates O · 2 − , and the subsequent excessive H 2 O 2 induces the upregulation of POD, which showed that POD and SOD activities would increase to avoid injury to plant cells [23][24][25] . In this study, the overall trend of the POD and SOD activities in leaves of PdPap seedlings were similar under rEpl1-e or rEpl1-p induction, and their activities were obviously higher than those of the control (Fig. 8C-F). POD and PPO act synergistically during enzymatic browning because PPO can promote POD activity by generating H 2 O 2 from the oxidation of phenolic compounds 25 . In addition, expression of PAL has been reported to be activated by the JA Signaling pathway 4,19,26 . These enzymes play an important role in plant defense, increasing the toxicity to invading pathogens and pests 25 . In this study, the activities of defense-related enzymes PPO and PAL in PdPap seedlings increased under rEpl1-e or rEpl1-p induction compared with those in the control group (Fig. 8G,I,H,J). Epl1-Tas as elicitor triggered ISR and SAR in PdPap seedlings resulting in the physiological response of PdPap seedlings. Moreover, the defense-related enzymes activities under rEpl1-p induction were also significantly higher than those under rEpl1-e induction. Perhaps the E. coli system expresses the target protein in inclusion bodies 27 , which can cause functional decreases after denaturation and renaturation, while the P. pastoris system has been engineered to yield high amounts of extracellularly secreted proteins with very few contaminating native proteins 17 , which explain why the yeast recombinant rEpl1-p had a better induction effect on PdPap seedlings. Therefore, we chose rEpl1-p further to induce PdPap transplanted seedlings challenged with pathogen A. alternata.
In induction test, the mean heights of the PdPap transplanted seedlings increased significantly under induction by 3 and 5 mg/mL rEpl1-p, by 57.65% and 51.18%, respectively (Fig. 9A,A'). This was consistent with the changes of auxin signal related genes (Fig. 7H-K) and IAAO activity (Fig. 8B) in PdPap seedlings leaves under rEpl1-p induction. Auxin acts by directly binding to the TIR1/AFB proteins, and promoting the degradation of transcriptional repressors called IAA/AUX proteins 28 . The original auxin response gene, GH3, could enhance sensitivity to IAA in root inhibition assays 29 . The relationship between the levels of IAA and IAAO activity is negatively correlated 30 . Auxin can regulate auxin-responsive genes 31 . Epl1-Tas, like auxin, could promote PdPap transplanted seedling growth. Moreover, the stronger transplanted seedlings had the stronger disease resistance. Under 3 and 5 mg/mL rEpl1-p induction, the mean leaf lesion areas decreased significantly, by 91.22% and 93.17% compared with that of the control, respectively (Fig. 9B,B'), which provided further evidence that the elicitor of defense response, Epl1-Tas, could induce protection of PdPap seedlings against attack by A. alternata.
In summary, the transcription of Epl1-Tas could be induced by woody plant PdPap seedlings stems and leaves and by the cell wall and fermentation liquid of its pathogen A. alternata. Disease resistance element ' ASF1MOTIFCAMV' (S000024) and disease resistance response element 'BIHD1OS' (S000498) were found in the promoter region of Epl1-Tas. Epl1-Tas was also successfully expressed in P. pastoris GS115. The recombinant proteins rEpl1-e and rEpl1-p both affected the transcription of 11 genes related to JA, SA, and auxin signal transduction pathways and the activities of five enzymes in seedlings of the woody plant PdPap. However, rEpl1-p had a better induction effect than rEpl1-e, making it more suitable for woody plant stimulation. Furthermore, rEpl1-p not only promoted PdPap transplanted seedling growth, but also protected PdPap transplanted seedling against attack by the pathogen A. alternata. This study increased our understanding of the role of Epl1-Tas in biocontrol and provides a practical reference for applications of Epl1-Tas in forests.

Materials and Methods
Strains, vectors and plant materials. T. asperellum ACCC30536 was obtained from the Agricultural Culture Collection of China (ACCC). P. pastoris GS115 and vector pPIC9K (Invitrogen, USA) were used for eukaryotic expression. Aseptic Populus davidiana × P. alba var. pyramidalis (PdPap) tissue-culture seedlings were used for the induction experiment. The stems, roots, and leaves of aseptic PdPap seedlings were used as inducing substrates for T. asperellum ACCC30536. Poplar pathogenic fungus A. alternata (the causative agent of poplar leaf wither) was used to prepare a carbon source in the inducing media for T. asperellum.
Promoter analysis and multiple sequence alignment of Epl1-Tas. The 1,500 bp upstream fragments of the coding regions were obtained as predicted promoter regions. Regulatory motifs in these regions were predicted using the Promoter Database of Saccharomyces cerevisiae (http://rulai.cshl.edu/SCPD/) and the STRING tool 32 .

Differential expression of Epl1-Tas in T. asperellum under eight inducing conditions. The tran-
scription of Epl1-Tas in T. asperellum ACCC30536 under eight inducing conditions was studied, namely minimal medium (MM) with 0.5% (w/v) glucose and 0.5% (w/v) ammonium sulfate 33 , carbon starvation medium, nitrogen starvation medium, and variable carbon source in MM as follows: 1% (w/v) root powder, 1% (w/v) stem powder, or 1% (w/v) leaf powder from PdPap seedlings, 1% (w/v) powdered cell walls; and 5% (v/v) fermentation supernatant of A. alternata for 10 d. Root, stem and leaf powders, fungal phytopathogen cell walls, and fungal phytopathogen fermentation supernatants were prepared according to a previous study 34 . All experiments were performed in triplicate.
Total RNA was extracted from the mycelia using the Trizol reagent (Invitrogen, Carlsbad, CA, USA) and then subjected to qRT-PCR. The qRT-PCR products were analyzed using the IQ5 real-time PCR detection system (Bio-Rad Laboratories Co., Ltd., Shanghai, China). The expression level of Epl1-Tas was calculated from the threshold cycle according to 2 −ΔΔCT method 32 . The qRT-PCR primers are shown in Supplementary Table 5.

Sequences similar to Epl1-Tas and their promoters from four Trichoderma species genomes.
The Epl1-Tas amino acid sequence was used as a query to search for similar coding sequences using tBLASTN in the genome databases of T. asperellum ACCC30536, T. virens Gv29-8, T. atroviride ATCC74058, and T. harzianum CBS 226.95 genomes (http://genome.jgi-psf.org/). Bioinformatic analysis was performed according to the methods of Fan 32 . The phylogenetic tree was constructed using the neighbor-joining method in the MEGA 5.10 program 32 . The promoter analysis method is detailed in the "Promoter analysis and multiple sequence alignment of Epl1-Tas" section.
Differential expression of three Epls in T. asperellum ACCC30536 genome under four inducing conditions. The transcription of Epl1-Tas and its two homologous genes in the T. asperellum ACCC30536 genome was analyzed in C-starvation (MM with 0.5% (w/v) ammonium sulfate), N-starvation (MM with 0.5% (w/v) glucose) and SXY (variable carbon source in MM as follows: 1% (w/v) root powder, 1% (w/v) stem powder, or 1% (w/v) leaf powder from PdPap seedlings) 35 . All experiments were performed in triplicate.

P. pastoris transformations and SDS-PAGE analysis.
The ORF of Epl1-Tas was amplified using the primers Epl1p1: 5′-ATCGG AATTCGATACGGTCTCCTACGACAC-3′ (EcoRI site underlined) and Epl1p2: 5′-CGATGCGGCCGCGAGGCCGCAGTTGCTCACGGC-3′ (NotI site underlined), digested with the indicated enzymes, ligated into the expression vector of pPIC9K, and transferred into E. coli Top10 competent cells to select positive clones. The recombinant vector pPIC9K-Epl1 was transferred into P. pastoris GS115 to induce expression according to the instructions of the Pichia expression kit (Invitrogen, Cat. No. K1710-01, USA). P. pastoris transformant GS115-Epl1 and the control transformant GS115-pPIC9K (P. pastoris GS115 transformed with empty vector pPIC9K), were induced with 0.5% (v/v) methanol per 24 h at 30 °C and 200 rpm, respectively. The supernatant of the transformant cultures was harvested at 1, 2, 4 and 6 h. The supernatants were collected by centrifugation at 10,000 × g at 4 °C for 10 min. After the addition of 1× loading buffer, the supernatant or cells were boiled for 5 min, centrifuged for 10 min at 8,000 rpm and loaded into a 12% SDS-PAGE gel 36 . Recombinant E. coli proteins were purified according to a pervious study 37 .
ScientiFic RepoRts | (2018) 8:7974 | DOI:10.1038/s41598-018-26328-1 Differential expression of hormone signal related genes of PdPap seedlings under recombinant rEpl1-e or rEpl1-p induction. The rEpl1-e obtained in our previous study 38 and the rEpl1-p obtained in present study were used for the induction experiment. PdPap tissue-culture seedlings were cultured in WPM medium containing 10% (v/v) recombinant rEpl1-e or rEpl1-p. The recombinant protein was heat inactivated, and the PdPap tissue-culture seedlings were cultured in WPM medium containing 10% (v/v) heat-denatured protein as the control 35 . The leaves were harvested at 0 h, 6 h, 12 h, 1 d, 2 d, 5 d, and 7 d for RNA extraction. The elongation factor (Ef), β-tubulin, and actin genes were used as internal references to normalize the amount of total RNA present in each reaction. The qRT-PCR primers was listed in Supplementary Table 6. All experiments were performed three times.
Extraction and detection for enzymes of PdPap under rEpl1-e or rEpl1-p induction. The treatment of the PdPap tissue-culture seedlings with rEpl1-e or rEpl1-p is detailed in the "Differential expression of hormone signal related genes of PdPap seedlings under recombinant rEpl1-e or rEpl1-p induction" section and the leaves were harvested after induction for 0, 0.5, 1, 2, 3, 5, and 7 d. Fresh tissue (0.5 g) was weighed from the above samples in each treatment with three replicates. The sample was frozen in liquid nitrogen and ground on ice with 10 mL of 100 mM pre-cold phosphate buffer (pH 6.0) containing 1% (w/v) dithiothreitol for enzyme extraction. IAAO, POD, SOD, polyphenol oxidase 30 , and PAL were extracted and their activities determined according to previous studies 25,30 . A. alternata resistance assay of PdPap seedlings leaves under rEpl1-p induction. The PdPap tissue-culture seedlings were transplanted into aseptic soil for 10 d. The rhizosphere of PdPap transplant seedlings were watered using the supernatant of a GS115-Epl1 culture containing final concentrations of 0, 0.5, 1, 3 and 5 mg/mL rEpl1-p, respectively. The growth of PdPap transplanted seedlings were observed after treatment for 30 d, and the plant height of PdPap transplant seedlings was measured. Meanwhile, the same position of leaves selected randomly per plant was inoculated with A. alternata of 1 × 10 3 spores/mL (the optimum infection spore concentration). The disease spots were observed and the lesion areas were determined using Image Pro Plus 6.0 software after 10 d 5 . Each treatment contained five repetitions, and the experiment was repeated twice.