Trichoderma atroviride Transcriptional Regulator Xyr1 Supports the Induction of Systemic Resistance in Plants

Abstract

As a result of a transcriptome-wide analysis of the ascomycete Trichoderma atroviride, mycoparasitism-related genes were identified; of these, 13 genes were further investigated for differential expression. In silico analysis of the upstream regulatory regions of these genes pointed to xylanase regulator 1 (Xyr1) as a putatively involved regulatory protein. Transcript analysis of the xyr1 gene of T. atroviride in confrontation with other fungi allowed us to determine that xyr1 levels increased during mycoparasitism. To gain knowledge about the precise role of Xyr1 in the mycoparasitic process, the corresponding gene was deleted from the T. atroviride genome. This resulted in strong reductions in the transcript levels of axe1 and swo1, which encode accessory cell wall-degrading enzymes considered relevant for mycoparasitism. We also analyzed the role of Xyr1 in the Trichoderma-Arabidopsis interaction, finding that the plant response elicited by T. atroviride is delayed if Xyr1 is missing in the fungus.

INTRODUCTION

Protection of plants from disease is of major relevance to the growing demand for plant resources for food, feed, fibers, and bioethanol (1). The response of a plant to bacteria or fungi is mediated by two main pathways (reviewed in reference 2). Systemic acquired resistance (SAR) is evoked by local infection with a pathogenic microbe and leads to the accumulation of salicylic acid (SA) and pathogenesis-related (PR) proteins, thereby increasing the long-term resistance of the whole plant to subsequent infections (3). Induced systemic resistance (ISR) is activated by rhizosphere-competent, nonpathogenic organisms, causes transient synthesis of the phytohormones jasmonic acid (JA) and ethylene (ET), and activates a defense response in distal tissues (4). Recently, it has been reported that both types of systemic responses to pathogens can be activated together, potentially enhancing plant protection (5).

Trichoderma spp. have a beneficial impact on plant protection for a number of reasons. Colonization of the roots of maize and cucumber plants with Trichoderma spp. triggers defense mechanisms in the plants, such as the production of phytoalexin, PR proteins, JA, peroxidases, or chitinases (6–11). Furthermore, during mycoparasitism, Trichoderma spp. secrete cell wall-degrading enzymes (CWDEs) such as chitinases, as well as xylanases, cellulases, and other glucanases (12–14). A correlation between the presence of these enzymes and the elicitation of pathogen response mechanisms in plants has been reported (see, e.g., reference 15). However, the expression of some CWDEs (for instance, Prb1 and ECH42) in Trichoderma spp. is not restricted to direct physical contact but seems to be preceded by an early recognition process (16, 17). In addition, some of the CWDEs from Trichoderma spp. are heterologously expressed in plants, leading to improved resistance to certain plant pathogens; for instance, expression of the endochitinase CHIT42 from Trichoderma atroviride P1 in tobacco and potato plants has resulted in high tolerance against Alternaria spp., Botrytis cinerea, or Rhizoctonia solani (18).

The identification of Trichoderma genes or gene products, other than those encoding CWDEs, involved in interactions with plants and/or plant pathogens has become the subject of more-recent studies (19, 20). In 2011, a comprehensive transcriptome-wide analysis of T. atroviride (teleomorph Hypocrea atroviridis) during its mycoparasitic interaction with R. solani revealed 175 host-responsive Trichoderma genes (21). A subsequent transcript analysis employing quantitative PCR (qPCR) confirmed that the 13 genes chosen for detailed investigation were differentially expressed during the confrontation of Trichoderma with other fungi. Among these genes, two were found to be highly induced before fungal contact: swo1 and axe1 (21). Notably, both of these genes can be considered to code for accessory CWDEs: swo1 for an expansin-like protein and axe1 for an acetyl xylan esterase. Since xylanase regulator 1 (Xyr1) has been identified as the main transactivator of a wide range of CWDEs in Trichoderma reesei (22), the possible involvement of this regulatory protein was discussed (21).

During that study, we analyzed the upstream regulatory regions (URR) of T. atroviride genes, which are differentially expressed under mycoparasitic conditions. This drew our attention to the expression of these genes in the context of the presence or absence of Xyr1. Additionally, the findings on the mycoparasitic behavior of a T. atroviride xyr1 deletion strain prompted us to examine the putative role of Xyr1 in fungal response or recognition mechanisms by plants during the interaction of T. atroviride with Arabidopsis thaliana.

MATERIALS AND METHODS

Organisms and growth conditions.

Confrontation assays were performed with the T. atroviride wild-type strain IMI206040, P1, or the xyr1 deletion strain (constructed during this study) and the plant-pathogenic fungus R. solani AG-4, B. cinerea, or Phytophthora capsici Aym2 on minimal medium (23) plates supplemented with 0.2% (wt/vol) glucose as described previously (21). For the time course of the confrontation assays, incubation was performed in ambient light; picture taking started shortly before or after the first physical contact of the T. atroviride strain with the plant pathogen and continued until full overgrowth.

For the determination of growth rates, fungi were inoculated onto potato dextrose agar (PDA) plates and minimal medium plates supplemented with 1% (wt/vol) d-xylose in light and at the ambient temperature (22°C), and colony diameters were measured at the time points given in Fig. 2.

FIG 2.

Effect of xyr1 deletion in T. atroviride on fungal growth. The growth of a xyr1 deletion strain (light shaded bars) and that of its parental strain (dark shaded bars) on minimal medium plates supplemented with 1% (wt/vol) d-xylose (A) or on PDA plates (B) were determined. Both strains were grown at room temperature for the times indicated, in hours. The results, expressed as colony diameters (mm), are means for three biological replicates. Error bars indicate standard deviations.

A. thaliana ecotype Columbia-0 (Col-0) seeds were surface sterilized with 70% (vol/vol) ethanol for 5 min and with 20% (vol/vol) household bleach for 10 min. After five washes in sterile distilled water, seeds were kept overnight at 4°C, germinated, and grown on plates containing 0.2× Murashige & Skoog medium (M0232.0001; Duchefa Biochemie, Haarlem, The Netherlands) supplemented with 2% (wt/vol) sucrose. The plates were placed vertically, and plants were grown at 22°C on a schedule of 16 h of ambient light and 8 h of darkness.

Deletion of the xyr1 gene from the T. atroviride genome.

For the construction of the deletion cassette, primers for double-joint PCR (24) were designed according to the T. reesei xyr1 deletion strategy applied by Stricker et al. (22) using the nucleotide sequence of the T. atroviride xyr1 gene (JGI protein identification [ID] 78601 [http://genome.jgi-psf.org/Triat2/Triat2.home.html]). The 5′ and 3′ regions for recombination were amplified using 5′Ta_xyr1F and 5′Ta_xyr1R (PCR1) or 3′Ta_xyr1F and 3′Ta_xyr1R (PCR2) (see Fig. S1A in the supplemental material). The marker gene encoding hygromycin B phosphotransferase was amplified from plasmid pRLMex₃₀ (25) with primers hphF and hphR (PCR3) (see Fig. S1A). The three fragments were fused with Pfu DNA polymerase (Promega, Madison, WI, USA) without primers by using 12 cycles of 30 s at 95°C, 10 min at 58°C, and 5 min at 72°C. The 4,001-bp deletion cassette was amplified by PCR using the nested_Ta_xyr1F and nested_Ta_xyr1R primers (PCR4) (see Fig. S1A) and was subsequently cloned into the pJET vector (Thermo Scientific, Waltham, MA, USA) according to the manufacturer's instructions to create pJET_DKxyr1. The linear deletion cassette was released from the plasmid by BglII digestion. Fungal transformation of T. atroviride P1 (ATCC 74058) was carried out according to the work of Peterbauer et al. (26), and mitotically stable transformants were obtained after at least three rounds of single-spore isolation.

For retransformation of the xyr1 gene into the deletion strain and investigation of the restoration of the phenotype of the parental strain, the strategy described in reference 22 was followed, with cotransformation of pXR51.1 (27), bearing the 4.3-kb xyr1 gene, and p2SR3, bearing the Aspergillus nidulans amdS gene as a marker. Briefly, transformed protoplasts were regenerated on a medium containing d-xylose (sole carbon source). The fastest-growing colonies were isolated, and their growth on d-xylose was compared as described above. Randomly chosen strains exhibited growth rates similar to that of the parental strain, strongly indicating complementation of the phenotype.

Control of correct genomic integration of the deletion cassette.

After the isolation of genomic DNA from the T. atroviride parental strain and the xyr1 deletion strain, PCRs using either primer pair 5′Ta_xyr1F and hphR (C-PCR1) or primer pair hphF and 3′Ta_xyr1R (C-PCR2) were performed (see Fig. S1A in the supplemental material). Plasmid pJET_DKxyr1 was used as the template for a negative control. Amplification products were verified by size using agarose gel electrophoresis (see Fig. S1B in the supplemental material) and were sequenced using a custom DNA-sequencing service (Microsynth, Balgach, Switzerland).

Southern blot analysis was performed using 10 μg of genomic DNA for XbaI digestion. After agarose gel electrophoresis, DNA was transferred to a membrane and was visualized with DyLight 650-labeled streptavidin, and the membrane was scanned using a Typhoon FLA 9500 scanner, according to the protocol reported in reference 28. The biotin-labeled probe was prepared as described previously (28) using the BglII-released deletion cassette as the template.

Induction of fungal resistance in A. thaliana.

Plates were inoculated with fungal spores (density, 1 × 10⁶) at a distance of 6 cm from 6-day-old A. thaliana seedlings (10 seedlings per plate). After 96 h and 120 h of incubation at 22°C on a schedule of 16 h of light and 8 h of darkness, leaves were collected and were immediately shock frozen in liquid nitrogen.

RNA extraction and cDNA synthesis.

Total fungal RNA was isolated using the guanidinium thiocyanate method according to the work of Chomczynski and Sacchi (29) and was purified with the RNeasy kit (Qiagen, Hilden, Germany). For the preparation of A. thaliana RNA, the leaves were homogenized in liquid nitrogen, and total RNA was first extracted with TRIzol (Invitrogen, Life Technologies Ltd., Paisley, United Kingdom) and then cleaned using the RNeasy kit (Qiagen). cDNA was synthesized with the RevertAid H Minus First Strand cDNA synthesis kit (Thermo Scientific) according to the manual by using 0.45 μg RNA as the template in the prior DNase I treatment. cDNA samples were diluted 1:100 before quantitative PCR (qPCR).

Selection of primers for qPCR.

Primer pairs for qPCR were designed using Primer3 software (version 0.4.0; http://frodo.wi.mit.edu/primer3/) (30). BLAST searches against the fungal genome databases of the confrontation partners R. solani (http://www.rsolani.org/), P. capsici (http://genome.jgi-psf.org/PhycaF7/PhycaF7.home.html), and B. cinerea (http://www.broadinstitute.org/annotation/genome/botrytis_cinerea/Home.html), and also against those of T. reesei (http://genome.jgi-psf.org/Trire2/Trire2.home.html), Trichoderma virens (http://genome.jgi-psf.org/Trive1/Trive1.home.html), and Neurospora crassa (http://www.broadinstitute.org/annotation/genome/neurospora/MultiHome.html) (cutoff E value, 10e−1), were conducted to ensure that only primer pairs without predicted amplification in organisms other than T. atroviride were used for qPCR. The sequences of all primers are given in Table 1.

TABLE 1.

Primers used throughout this study

Primer name	Sequence (5′ to 3′)	Use
At_PR1F	TTGTTCTTCCCTCGAAAGCTC	qPCR of A. thaliana genes
At_PR1R	AGTGACCACAAACTCCATTGC
At_PR2F	AGCTGGACAAATCGGAGTATG
At_PR2R	ACGTTGATGTACCGGAATCTG
At_PR5F	GGCTGTGTCTCTGACCTCAAC
At_PR5R	TAGCTCCGGTACAAGTGAAGG
At_PR3F	CAGACTTCCCATGAAACTACAGG
At_PR3R	GCAGTCATCCAGAACCAAATC
At_PR4F	GAGTGCTTATTGCTCCAC
At_PR4R	TTGCTACATCCAAATCCAAGC
At_PDF1.2F	TGCTTCCATCATCACCCTTATC
At_PDF1.2R	CATGTCCCACTTGGCTTCTC
At_UBQ5F	GCCGACTACAACATCCAGAAG
At_UBQ5R	ATGACTCGCCATGAAAGTCC
At_TUB9F	TTCCCGAGAGAAAGAAAG
At_TUB9R	ACAGCACGAGGAACGTATTTG
At_PP2AA3	CCTTTCCATGCTCTGTCAAAC
At_PP2AA3R	CATGCTGATACTCTGGCTGTG
Ta_xyr1F	CAACAACCACCCCTATCCTATC	qPCR of T. atroviride genes
Ta_xyr1R	TCGGATGTACTCGCAGCTAAG
Ta_actinF	AAGGACCTCTACGGCAACATTG
Ta_actinR	GACAATGGAGGGACCGCTC
Ta_sar1F	TTCAACACCTTCGATCTGGG
Ta_sar1R	GTGTCTCAGCTCATCCTCAG
5′Ta_xyr1F	TGTGCTGCTGTTATTGCTGT	Double-joint PCR
5′Ta_xyr1R	CTAATCGCCTTGCAGCACATCCCCTGGTTCAATCCGATATGCTT
3′Ta_xyr1F	TAATTCACATGGATGGTCTTTGCGTGAGCACGACTTGAACAGAA
3′Ta_xyr1R	CAATGGACGCTCTCAAAGAT
hphF	CGGGCCTCTTCGCTATTAC
hphR	ACTGATCGCAAAGACCATCC
nested_Ta_xyr1F	CCTTGTATACCGCTGCCTCT
nested_Ta_xyr1R	CCGTGCTAAGAGTCACCACA

qPCR and data analysis.

All qPCRs were performed in triplicate. Reactions were conducted in a Mastercycler ep realplex system (version 2.2; Eppendorf, Hamburg, Germany) at a final reaction volume of 15 μl containing 7.5 μl 2× iQ SYBR green mix (Bio-Rad Laboratories, Hercules, CA, USA), 100 nM forward and reverse primers, and 2 μl cDNA as the template. Amplification was accomplished using a 3-min initial denaturation step, followed by 50 cycles at 95°C for 15 s, 60°C for 20 s, and 68°C for 20 s. Amplification products were verified by subsequent melting curve analyses. PCR efficiencies, calculated by analyzing the linear regression of efficiency (LRE) (31, 32), were in the range of 100% ± 8% for each reaction. Comparative transcript level ratios were obtained according to the method of Steiger et al. (33) using the relative expression software tool (REST), version 2008 (34). The T. atroviride act1 and sar1 genes were used for data normalization on the basis of reports by Brunner et al. of their unchanged transcription during plate confrontation assays (35). UBQ5 (At3g62250), TUB9 (At4g20890), and PP2A (At1g13320) were used as reference genes for data normalization in A. thaliana transcript analyses.

RESULTS

T. atroviride increases xyr1 transcript formation before contact with a pathogen.

The upstream regulatory regions (URR) of the genes, which were previously found to be differentially expressed in T. atroviride during mycoparasitism (21), were analyzed in silico. Interestingly, in the URR of five genes (JGI transcript IDs 297887, 80187, 211243, 302587, and 84753 [http://genome.jgi-psf.org/Triat2/Triat2.home.html]), binding sites for Xyr1 [5′-GGC(T/A)₃-3′ (36)] were found. So far, nothing is known about the involvement of Xyr1 in the regulation of mycoparasitic mechanisms. To follow the transcript levels of xyr1, RNA was isolated from T. atroviride before and during contact with B. cinerea, P. capsici, R. solani, or N. crassa, or with itself as the reference condition. RT-qPCR was performed in order to determine transcript ratios. With B. cinerea, P. capsici, and R. solani, xyr1 transcript levels in T. atroviride before contact were clearly higher than those than under the control condition (confrontation of T. atroviride with itself) (Fig. 1A). Only confrontation with the nonpathogenic N. crassa strain resulted in the same level of xyr1 transcription in T. atroviride as that for a confrontation with itself (Fig. 1A). In T. atroviride mycelium harvested from the interaction zone, slightly lower xyr1 transcript levels than those for the confrontation with T. atroviride were detected if the confrontation organism was B. cinerea, P. capsici, or R. solani, while the amount detected in confrontation with N. crassa was the same as that under the control condition (Fig. 1B). However, since xyr1 expression was increased in T. atroviride before direct pathogen contact (Fig. 1A), Xyr1 is likely to be involved in mycoparasitic interactions and therefore became a target of further investigation.

FIG 1.

Comparative transcript ratios of xyr1 in T. atroviride during mycoparasitism. T. atroviride was used in plate confrontation assays with B. cinerea (TB), P. capsici (TP), R. solani (TR), or N. crassa (TN), or with itself (TT) as the reference condition. T. atroviride mycelia were harvested before contact (A) or from the interaction zone during contact (B) and were used for RNA extraction and RT-qPCR. Transcript ratios were calculated with REST 2008 by using the act and sar1 genes for normalization. The dashed lines (at a ratio of 1) indicate equal transcript amounts for the confrontations compared. Values above or below 1 indicate more or less transcript, respectively, for the confrontation of interest than for the confrontation of T. atroviride with itself. Error bars indicate 95% confidence intervals.

Deletion of xyr1 from the T. atroviride genome.

To investigate if Xyr1 plays a regulatory role in mycoparasitism in T. atroviride, the encoding gene was deleted from the genome. After several rounds of transformation, one deletion strain could be obtained. Such low transformation efficiency with xyr1 deletion has been observed previously, e.g., in T. reesei, where two Δxyr1 strains were obtained (22). This, together with the fact that T. atroviride has generally lower transformation efficiencies than T. reesei, might explain the low transformation yield. The correct genotype of the Δxyr1 strain obtained was confirmed by PCR (see Fig. S1 in the supplemental material) and Southern blot analysis (see Fig. S2 in the supplemental material), and RT-qPCR (compare Table 2) was used to demonstrate that no xyr1 transcript was formed in the deletion strain. First, the phenotype of the deletion strain was analyzed in terms of its growth behavior. On minimal medium supplemented with d-xylose, we observed a considerably lower level of growth for the xyr1 deletion strain than for its parental strain (Fig. 2A), a finding in good accordance with previous reports of the strongly reduced ability of Δxyr1 strains to metabolize d-xylose (22). On full medium, the growth behaviors of the deletion strain and the parental strain did not differ (Fig. 2B).

TABLE 2.

Comparative transcript ratios in T. atroviride during mycoparasitism before and during contact

Gene analyzed	Ratio of the transcript levela for the indicated plate confrontation conditionb to that for the reference condition
	Before contact				During contact
	TB	TR	xB	xR	TB	TR	xB	xR
xyr1	+	+	NDS	NDS	+	−	NDS	NDS
prb1	+	++	++++	++++	NDR	NDR	NDR	NDR
swo1	++	+++	+	+	−	−	−−−	−−−
axe1	+++	+++	NDS	NDS	−	−	NDS	NDS

^a+, <100-fold upregulated; ++, >100-fold upregulated; +++, >1,000-fold upregulated; ++++, >10,000-fold upregulated; −, <100-fold downregulated; −−−, >1,000-fold downregulated; NDS, no transcript detected in the sample; NDR, no transcript detected under the reference condition.

^bTB, confrontation of T. atroviride with B. cinerea; TR, confrontation of T. atroviride with R. solani; xB, confrontation of the T. atrovirideΔxyr1 strain with B. cinerea; xR, confrontation of the T. atroviride Δxyr1 strain with R. solani. The reference condition is the confrontation of T. atroviride with itself.

Xyr1 influences gene transcription under mycoparasitic conditions.

To gain insight into the relationship between the expression of xyr1 and that of some of the mycoparasitically relevant genes mentioned above, namely, prb1, swo1, and axe1, the T. atroviride parental and xyr1 deletion strains were used in plate confrontation assays with B. cinerea or R. solani. The confrontation of T. atroviride with itself again served as the reference condition. We found that before contact with the pathogen, transcript levels for all four genes investigated were higher than those under the reference condition (Table 2), a finding in accordance with earlier results of this study (Fig. 1A) and previous studies (16, 21). Interestingly, xyr1 deletion led to a loss of axe1 transcript formation, a reduced increase in the swo1 transcript level, and highly increased prb1 transcript formation relative to those under the reference condition (Table 2). We did not investigate prb1 expression in mycelia harvested from the interaction zone during contact, because it was reported previously that prb1 is contact-independently induced during the mycoparasitic response (16, 21). However, we detected fewer transcripts of swo1 and axe1 in T. atroviride during pathogen contact than under the reference condition. Again, in the Δxyr1 strain, no axe1 transcript was detected, and swo1 transcript formation was more strongly downregulated than in the parental strain (Table 2). The exact values for the comparative transcript ratios can be found in Table S1 in the supplemental material. To summarize, in T. atroviride, axe1 and swo1 are upregulated before pathogen contact, and are downregulated during contact, relative to their expression under the control condition. If Xyr1 is missing, axe1 transcript formation is fully abolished and swo1 accumulation is lower (i.e., less upregulated before contact and more strongly downregulated during contact) than in the parental strain. These results, taken together, point to the involvement of Xyr1 in the regulation of mycoparasitic mechanisms in T. atroviride.

Influence of Xyr1 on the mycoparasitic behavior of T. atroviride.

As a consequence of the findings described above, we investigated the mycoparasitic behavior of the xyr1 deletion strain compared to that of the parental strain. For this purpose, we confronted each strain with B. cinerea, P. capsici, or R. solani. As can be seen from Fig. 3, we observed elevated vitality (i.e., faster growth) and slightly stronger sporulation for the Δxyr1 strain than for its parental strain. In order to follow up the time course of the confrontation assays, corresponding pictures (starting at day 3 after plate inoculation) are provided in Fig. S3 in the supplemental material. Surprisingly, the lack of Xyr1 resulted in overall enhanced competition with the plant pathogen. This finding is the opposite of what we expected, because it seems to contradict the reduced transcript levels of axe1 and swo1, which were detected in the xyr1 deletion strain during confrontation (Table 2). However, the strong upregulation of prb1 expression may at least partly explain this observation, since it has been demonstrated previously that overexpression of this gene leads to improved mycoparasitism by Trichoderma (37).

FIG 3.

Effect of xyr1 deletion in T. atroviride on its mycoparasitic behavior. The mycoparasitic behavior of a xyr1 deletion strain (top) and its parental strain (bottom) were investigated using the plant pathogen B. cinerea, P. capsici, or R. solani. Confrontation assays were performed using minimal medium plates supplemented with 0.2% (wt/vol) glucose at room temperature for 8 days in darkness. The T. atroviride strain was placed on the left side of the plate and the plant pathogen on the right side.

Xyr1 assists induced and systemic resistance to fungi in plants.

Next, we aimed to investigate if Xyr1 might be involved in evoking a fungal resistance mechanism in plants. To test this, we inoculated A. thaliana seedlings with spores from either the T. atroviride parental strain or the Δxyr1 strain (Fig. 4A). As a reference condition we used A. thaliana seedlings incubated without fungal spores. After the isolation of RNA from leaves harvested after 96 and 120 h, we performed transcript analysis of A. thaliana genes that are considered markers of SA-dependent systemic acquired resistance (PR-1a, PR-2, and PR-5 [38, 39]) or markers of the JA/ET-dependent defense of plants (PR-3, PR-4, and PDF1.2 [40]). In A. thaliana seedlings inoculated with the T. atroviride parental strain, the SAR-related genes were induced during the earlier period, and later their transcript levels dropped close to those observed for the reference condition (without fungal spores) (Fig. 4B). In seedlings inoculated with the Δxyr1 strain, the transcript levels were below that of the reference sample after 96 h, and only after 120 h did they rise to a higher level (Fig. 4B). For ISR-related genes in A. thaliana seedlings, we observed basically the same trends as for SAR-related genes, i.e., decreasing induction over time if the T. atroviride parental strain was used and increasing induction over time if the Δxyr1 strain was used (Fig. 4C). The only difference observed from SAR-related marker genes was that transcript levels in A. thaliana seedlings inoculated with spores from the Δxyr1 strain were already higher than those under the control condition at the earlier time point. However, with the xyr1 deletion strain, higher transcript levels were detected at the later time point than at the earlier time point for all genes investigated (compare Fig. 4B and C). These opposite trends in expression patterns point to a delay of the A. thaliana defense mechanisms evoked by T. atroviride if Xyr1 is missing.

FIG 4.

Transcript analysis of marker genes for fungal defense mechanisms in A. thaliana. (A) Spores from the T. atroviride parental strain or the xyr1 deletion strain were inoculated 6 cm from 6-day-old A. thaliana seedlings and were incubated at 22°C with 16 h of light and 8 h of darkness. (B and C) After 96 and 120 h, RNA was extracted from the leaves, and RT-qPCR was performed. Transcript levels of the SAR-related A. thaliana genes PR-1a, PR-2, and PR-5 (B) and the ISR-related genes PR-3, PR-4, and PDF1.2 (C) were investigated. Transcript ratios were calculated with REST 2008 by using the UBQ5, TUB9, and PP2A genes for normalization. Black squares, incubation with the T. atroviride parental strain; blue squares, incubation with the xyr1 deletion strain. The dashed line (at a ratio of 1) indicates equal transcript amounts for the conditions compared. Values above or below 1 indicate more or less transcript, respectively, for the condition of interest than for A. thaliana seedlings without fungal spores. Error bars indicate 95% confidence intervals.

DISCUSSION

In this study, we demonstrate that Xyr1 plays an essential role in the “tripartite relationship” of the mycoparasite T. atroviride, a plant pathogen, and a plant. A graphical sketch of this interplay and of the impact of the absence or presence of Xyr1 is displayed in Fig. 5. First, we observed that the expression of Xyr1 in T. atroviride is increased specifically before contact when the fungus is confronted with a plant pathogen. Second, the prb1 gene coding for a proteinase, which was reported to be induced independently of contact (16), was more strongly induced in a Xyr1-deficient T. atroviride strain than in the parental strain upon confrontation with a plant pathogen. This upregulated expression likely contributes to the improved mycoparasitic behavior observed for the deletion strain and is in good accordance with the previously suggested strategy to overexpress prb1 for better biocontrol activity (37). An in silico analysis of the URR of prb1 revealed that within a 785-bp region upstream from the ATG, five Xyr1-binding sites [5′-GGC(T/A)₃-3′] (27, 36, 41) are present. This, in principle, would provide the prerequisite for direct regulation by Xyr1 but would imply a repressing function, which has never been described so far. It should be noted that none of these Xyr1 sites in the prb1 URR are present as inverted repeats, a pattern reported to be essential for in vivo functionality in T. reesei (27, 42). In contrast, both of the other two genes investigated, swo1 and axe1, have exactly such a motif in their URR. And in both cases, xyr1 deletion led to a decrease in transcript formation, which would suggest that Xyr1 acts directly as an activator in these two cases. In the case of prb1, alternatively, an indirect influence of Xyr1 can be considered—for example, the activation of suppression. Third, the induction of genes encoding enzymes that assist in cell wall degradation, such as axe1 and swo1, is influenced by a functional Xyr1. The presence of such enzymes and of xylanases was previously associated with the evocation of fungal defense mechanisms in plants (15). This might be an explanation for the last observation: the delayed induction of SAR- and ISR-relevant marker genes in A. thaliana seedlings in the presence of a Xyr1-deficient T. atroviride strain. Thus, the expression of functional Xyr1 is needed for fast activation of a plant's defense systems. The expression of xyr1 in T. atroviride is induced during early stages of confrontation with a plant pathogen, and for other Trichoderma spp. it has been shown that increased Xyr1 levels correlate directly with the accumulation of cellulolytic enzymes. These findings, taken together, suggest that Xyr1 is involved in the mediation of the plant-fungus recognition processes.

FIG 5.

Schematic presentation of the working hypothesis concerning the contribution of T. atroviride Xyr1 to systemic resistance in plants. CWDEs, cell wall-degrading enzymes; ISR, induced systemic resistance; SAR, systemic acquired resistance; +, positive influence on expression levels.

In addition, the findings of the present study raise questions about the ancestral function of Xyr1, as well as the question of whether its role differs in different species and if so, in what manner. Among Trichoderma spp., Xyr1 was first described in T. reesei as an essential transactivator of cellulases, xylanases, and enzymes involved in d-xylose metabolism (22). T. reesei has a saprophytic lifestyle, in contrast to the lifestyle of the biocontrol-active species T. atroviride, and is quite distantly related to mycoparasitic Trichoderma species (43). Sequence analysis of the genomes of three Trichoderma species has pointed to mycoparasitism as the ancestral Trichoderma lifestyle (44). The authors of that study showed that T. atroviride resembles the ancient state and that T. reesei evolved later and faster. This went along with a reduction in the size of the genome of T. reesei and a loss of genes, among them mycoparasitism-specific genes (44). Accordingly, one can speculate that the ancient function of Xyr1 and its regulon had to undergo a transition during evolution and that they are different for the mycoparasitic and saprophytic Trichoderma species. Reports on the function of Xyr1 homologs in other filamentous fungi support this hypothesis. In Aspergillus niger, a nonpathogenic saprophyte, the ortholog XlnR (45) governs a regulon quite similar to that of Xyr1, i.e., cellulase- and xylanase-encoding genes and genes coding for d-xylose-metabolizing enzymes (reviewed in reference 46). On the other hand, in the plant pathogen Magnaporthe oryzae, deletion of the homolog Xlr1 had only marginal effects on the expression of xylanases (47). In the head blight fungus Fusarium graminearum, though, Xyr1 did not effect the induction of cellulases (48), whereas for both pathogenic fungi, d-xylose metabolism was impaired in the corresponding Xlr1 or Xyr1 deletion strain (47, 48). Taken together, these findings suggest that Xyr1 function did evolve in different ways according to the special lifestyle of the fungus.