The Endophytic Fungus Chaetomium cupreum Regulates Expression of Genes Involved in the Tolerance to Metals and Plant Growth Promotion in Eucalyptus globulus Roots

The endophytic strain Chaetomium cupreum isolated from metal-contaminated soil was inoculated in Eucalyptus globulus roots to identify genes involved in metal stress response and plant growth promotion. We analyzed the transcriptome of E. globulus roots inoculated with C. cupreum. De novo sequencing, assembly, and analysis were performed to identify molecular mechanisms involved in metal stress tolerance and plant growth promotion. A total of 393,371,743 paired-end reads were assembled into 135,155 putative transcripts. It was found that 663 genes significantly changed their expression in the presence of treatment, of which 369 were up-regulated and 294 were down-regulated. We found differentially expressed genes (DEGs) encoding metal transporters, transcription factors, stress and defense response proteins, as well as DEGs involved in auxin biosynthesis and metabolism. Our results showed that the inoculation of C. cupreum enhanced tolerance to metals and growth promotion on E. globulus. This study provides new information to understand molecular mechanisms involved in plant–microbe interactions under metals stress.


Introduction
In Chile, copper (Cu) mining has generated an over-accumulation of metals in areas surrounding Cu smelters, and Puchuncaví Valley (Valparaíso region, Chile) is a recognized place by the high environmental pollution originated mainly for Cu smelting. The toxicity of metals can generate losses of vegetal diversity and functionality of species, leading to a change of the soil characteristics and difficulty to establish vegetation [1]. However, it is known that prolonged exposure to metals can generate a selection of resistant/tolerant plant populations [2]. Eucalyptus spp. has been reported as a metal-tolerant species and used for phytoremediation process due to it fast growth; high biomass production; wide adaptability; and accumulation of high amounts of metals such as Cu, Zn, Pb, and Cd [3].
Soil microorganisms play an important role in the restoration of environments affected by contamination of metals, promoting plant growth through different mechanisms such as indole acetic

Plant Material and Experiment Design
Clonally propagated E. globulus seeds were acquired from a commercial nursery (Semillas Imperial, Los Ángeles, Chile). Seeds were germinated in vermiculite in a plant growth chamber at room temperature. After 4 weeks, uniform plants were selected and transplanted to 300 cc plastic pots with a mixture of Puchuncaví Valley soil: vermiculite (1:1 v/v) as a substrate. The soil is classified as an Entisol (Chilicauquén series) and has a pH w of 5.54 and Cu, Zn, Pb, As and Cd content (in mg Kg −1 ) of 385, 183, 135, 52 and 1.1 respectively. Pots were randomly divided into two groups, without microbial inoculation as control while the other group was inoculated with fungal strain C. cupreum according to Almonacid, et al. [20], where one slant of active mycelia was diluted in 40 mL of sterile distilled water then homogenized and vigorously agitated (10 mL of this suspension were inoculated, equivalent to 70 mg of dry mycelium). Plants were grown in a greenhouse with supplementary light provided by incandescent cool white lamps (400 umol m −2 s −1 , 400-700 nm) (Sylvania ® , Wilmigton, MA) with a Microorganisms 2019, 7, 490 3 of 15 16/8 h day/night cycle at 24/16 • C and 50% relative humidity. After 90 days post-inoculation (dpi), all plants were harvested and five biological samples were dried in an air-forced oven at 70 • C for 48 h and then were weighted to determine the biomass production. Entire roots systems were collected from five biological replicates of the control and inoculated plants. All these samples were rinsed thoroughly with distilled water, immediately frozen in liquid nitrogen and stored at −80 • C until RNA extraction.

Fungus Detection in Roots by Scanning Electron Microscopy (SEM)
For fungus visualization, six root segments collected at the end of the experiment (90 dpi) from non-inoculated and inoculated plants were cut into 1-cm pieces. Segments were obtained from the upper part (near to stem) and lower part (lateral roots). Samples were observed by SEM (Hitachi SU 3500, Japan).

RNA Extraction, Sequencing and Illumina Reads Processing
Total RNA was extracted from 70 mg of root tissue collected at the end of the experiment (90 dpi) using Spectrum ™ Plant Total RNA kit (Sigma-Aldrich, Germany) following the manufacturer's instructions. The yield and quality of the RNA isolation samples was measured using a Qubit ® 2.0 Fluorometer (Life Technolology, Carlsbad, CA), and Fragment Analyzer ™ Automated CE System (Analytical Advanced Technologies, Ames, IA). To obtain good coverage of the E. globulus transcriptome, equal quantities of individual RNAs from root tissues of five biological replicate plants were used for library construction. Complementary DNA (cDNA) libraries were constructed using the TruSeq RNA Sample Preparation kit v2 (Illumina ® , San Diego, CA) following the Illumina manufacturer's instructions and subsequently sent for sequencing to Macrogen Inc (Seoul, Korea). A total of 10 samples were sequenced in a single lane of an Illumina HiSeq 4000 platform (Illumina) in paired-end mode for 101 cycles. The resulting FASTQ files containing Illumina raw sequences were analyzed and trimmed using NGSQC Toolkit v2.3 [21], removing adaptors and low-quality reads, based on their Q-score composition, removing all reads with a content of Q>30 lower than 70% of bases.

De novo Transcriptome Assembly
Due to the endophytic nature of C. cupreum, high-quality reads were aligned with Eucalyptus grandis genome using BLASTN. Matching reads were used to construct a de novo assembly using Trinity software v2.8.3 [22]. Transcriptome was assembled on an Amazon Web Service Linux instance m4.16xlarge and downstream analysis were carried out at Centro de Modelación y Computación Científica (CMCC, Universidad de La Frontera, Chile). Refinement of transcriptome was carried out mapping reads to assembled transcripts, and relative abundance in FPKM value (Fragments per kilobase per transcript per million mapped reads) was calculated with RSEM v1.2.26 [23]. Poorly supported transcripts were removed, keeping all transcripts with a relative abundance of at least 1 FPKM for downstream analysis. Highly similar and redundant transcripts were clustered using CD-HIT-EST with a threshold of 95% [24]. In order to validate the integrity of this de novo assembly, a comparison between E. globulus assembled transcriptome was carry out by BUSCO (Benchmarking Universal Single-Copy Orthologs) against OrthodBv9 database (embryophyte), to identify highly conserved orthologous genes [25].

Functional Annotation of the Transcriptome
The resulting transcripts were aligned into the SwissProt database using BLAST+ with an e-value filter of 1-e −10 as threshold, and evaluated in hidden Markov profiles to identify any family membership and conserved domains in PFAM-A database [26]. Functional annotation and Gene Ontology (GO) terms classification (Cellular component, Biological process and Molecular function and) were performed with PANTHER system [27], using as input gene lists obtained from blast top hit using reference proteomes collection from EMBL as database.

Differential Expression Analysis
The relative abundance was calculated using RSEM through align_and_estimate_abundance.pl script and the resulting abundance for each sample were merged in a matrix and analyzed with run_DE_analysis.pl script, which involves the Bioconductor package DESeq2 in R statistical environment [28]; both scripts were contained in the Trinity package. To judge the significance of gene expression, a False Discovery Rate value (FDR) lower than 0.05 and a minimum fold change (FC) of 2 were set as thresholds. Main DEGs related to metal response and plant growth promotion were analyzed through a heat map using log10 ratio values of expression levels.

Plant Biomass and Fungus Detection in Roots
After 90 dpi, inoculated plants showed significantly higher dry biomass as compared to the control, and increased 37% in shoots and 45% in roots in relation to non-inoculated plants ( Figure 1). According to SEM images, no structures were observed in the roots of non-inoculated plants of E. globulus after harvest, while in all plants inoculated, fungal hyphae were observed ( Figure 2).

Differential Expression Analysis.
The relative abundance was calculated using RSEM through align_and_estimate_abundance.pl script and the resulting abundance for each sample were merged in a matrix and analyzed with run_DE_analysis.pl script, which involves the Bioconductor package DESeq2 in R statistical environment [28]; both scripts were contained in the Trinity package. To judge the significance of gene expression, a False Discovery Rate value (FDR) lower than 0.05 and a minimum fold change (FC) of 2 were set as thresholds. Main DEGs related to metal response and plant growth promotion were analyzed through a heat map using log10 ratio values of expression levels.

Plant Biomass and Fungus Detection in Roots.
After 90 dpi, inoculated plants showed significantly higher dry biomass as compared to the control, and increased 37% in shoots and 45% in roots in relation to non-inoculated plants ( Figure 1). According to SEM images, no structures were observed in the roots of non-inoculated plants of E. globulus after harvest, while in all plants inoculated, fungal hyphae were observed ( Figure 2).  3.2. Sequence Data, de novo Transcriptome Assembly and Annotation.
Ten libraries, which include five control samples and five samples inoculated with C. cupreum (90 dpi), with a total of 460.158.421 paired-end reads, were obtained. After sequence trimming for adapter and filtering low quality reads, it resulted in 393.371.743 high quality reads (Table 1). De novo transcriptome assembly after trimming, where transcripts with an estimated abundance lower

Sequence Data, de novo Transcriptome Assembly and Annotation
Ten libraries, which include five control samples and five samples inoculated with C. cupreum (90 dpi), with a total of 460.158.421 paired-end reads, were obtained. After sequence trimming for adapter and filtering low quality reads, it resulted in 393.371.743 high quality reads (Table 1). De novo transcriptome assembly after trimming, where transcripts with an estimated abundance lower than 1 FPKM and highly similar or redundant transcripts with a sequence similarity higher than 95% were removed, resulted in 135.155 transcripts. Transcriptome statistics such as N50 and average length values can be observed in Table 1. Transcripts of the final assembly were aligned to the Swissprot database; the homology search presented results for 37.667 sequences, corresponding to 27.9% of the total. De novo transcriptome was compared against BUSCO database, which contains information about highly conserved orthologous genes. Of the 1440 BUSCO genes, 1133 complete (78.7%), 108 fragmented (7.5%), and 199 missing genes (13.8%) were found in our assembly (Table 2).

Differential Expression Analysis
Differentially expressed genes (DEGs) were estimated using a fold change ≥ 2 and a FDR < 0.05 as cut-off between control and treatment conditions. In total, we found 709 DEGs (Figure 3), of which 403 were up-regulated and 306 were down-regulated in response to C. cupreum inoculation. However, those genes that did not match in annotation with Blast hit were removed and discarded. Therefore, we obtained 663 DEGs of which 369 were up-regulated and 294 down-regulated under C. cupreum inoculation. From DEGs, we investigated their functions carry out gene ontology analysis and classified them into three major GO terms ( Figure 4). Between the annotated transcripts, cell and organelle had the two greatest number of transcripts in cellular component terms. For the biological process term, metabolic process, cellular process, and biological regulation had the most transcripts. Within molecular function, most transcripts showed catalytic activity, binding and transporter activity. According to the PHANTER classification system, we observed that the functional class transport and transcription factor were down-regulated in treatment compared to control, where a greater number of transcripts was presented ( Figure 5).
organelle had the two greatest number of transcripts in cellular component terms. For the biological process term, metabolic process, cellular process, and biological regulation had the most transcripts. Within molecular function, most transcripts showed catalytic activity, binding and transporter activity. According to the PHANTER classification system, we observed that the functional class transport and transcription factor were down-regulated in treatment compared to control, where a greater number of transcripts was presented ( Figure 5). Among identified DEGs, we explore those involved in response to metal stress, metal transport and plant growth promotion (Table 3). In the inoculated condition, we found down-regulated genes that play an important role against metal stress, among them: metallothionein-like protein 1, peroxidase 10 precursor and heavy metal-associated isoprenylated plant protein 28 included in GO terms: metal ion binding (GO:0046872), response to oxidative stress (GO:0006979) and metal ion transport (GO:0030001), respectively. Other down-regulated genes were related to metals transport, Natural resistance-associated macrophage protein 1 (Nramp1), Nramp3, Nramp5, Nramp 6, Metal tolerance protein 4 (MTP4), and Putative Multidrug Resistance Protein (MRP) among others. Also, genes implicated in nutrient transport were down-regulated, including high affinity nitrate transporter 2.5 (Nitrate transport, GO:0015706), Inorganic phosphate transporter 1-1 (Phosphate ion transmembrane transporter activity, GO:0015114), Potassium transporter 5 (Potassium ion transport, GO:0006813), Sodium transporter HKT1 (Response to osmotic stress, GO: 0006970; Sodium transporter, GO:0006814), Magnesium transporter MRS 2-3 (Magnesium ion transport, GO:0015693), and Ammonium transporter Several up-regulated genes involved in plant growth promotion via auxin production were identified, including: Auxin-induced protein 22A, auxin-induced protein 22D, auxin-induced protein AUX22, auxin-responsive protein IAA3, auxin-responsive protein IAA4, auxin efflux carrier component 2, auxin-induced in root cultures protein 12 precursor, auxin-responsive protein SAUR50, auxin-responsive protein SAUR78, and índole-3-acetic acid-amino synthetase GH3.17, which were mainly included in GO terms: Auxin-activated signaling pathway (GO:0009734), response to auxin (GO:0009733), positive regulation of cell growth (GO:0030307), auxin homeostasis (GO:0010252) and root development (GO:0048364).
Expression patterns between control and treatment conditions of the most important DEGs related to metal response and plant growth promotion are shown in Figure 6.

Discussion
Within abiotic stress types, metal toxicity is one of the factors that causes serious deleterious effects in plants. Nevertheless, the alleviation of heavy metal toxicity by endophytic fungi could be an efficient strategy to enhance heavy metal tolerance in plants [9]. In this study, colonization of C. cupreum was detected at the end of the experiment in samples of lateral roots and near to the stem of plants, which suggests that inoculation of C. cupreum is persistent over time. Endophytic fungi establish a chemical communication with the host through sugars, fatty acids, amino acids, On the other hand, several genes associated with the transport of metals were also downregulated, among them: Nramp1, Nramp3, Nramp5, Nramp 6, MTP4 and MRP. Nramp genes play an important role in the uptake and translocation of a wide range of metal ions to the plant that include Cd, Zn, Fe, Cu and Mn, and it has been described that an up-regulation of these genes improves the accumulation of metals in Arabidopsis thaliana [34]. MTP4 has been described as a divalent cation efflux transporter that acts in the cytoplasm, being essential for the maintenance of metal homeostasis [35] and MRP realizes a similar role transporting metal ions to the vacuole as a detoxification mechanism of plants [36].
This suggests that the inoculation of C. cupreum prevents the metals present in the soil from being translocated to the roots of the plants of E. globulus, a finding which agrees with previous works where C. cupreum inoculation contributes positively to decreasing indicators of stress such as lipid peroxidation level and proline content because metal ions were adsorbed in the cell wall of the fungus [7]. However, this response mechanism depends directly on the plant-microorganism interaction, and on the type of host plant, because other studies showed that other endophytic species such as Mucor sp. despite having a protective effect on Arabidopsis arenosa plants by accumulating a lower amount of metals compared to the non-inoculated control, promotes translocation of metals from roots to shoots, which is reflected in an up-regulation of genes associated with metal transport and distribution such as HMA3, PCR2, ZIF1, and MTP1 [37]. Other transporters down-regulated were high affinity nitrate transporter 2.5, inorganic phosphate transporter 1-1, potassium transporter 5, sodium transporter HKT1, magnesium transporter MRS2-3, ammonium transporter 1 member 1 and high affinity sulfate transporter 1, which are directly related to plant nutrition. According to this, an increase in plant growth is not being promoted by an improvement in nutrient uptake. Other studies under C. cupreum inoculation reported similar results attributing the improvement in plant growth to a probable action of auxins [38]. Conversely, recent reports have shown an improvement in the absorption of nutrients under colonization of endophyte fungi. For example, the endophyte fungus Serendipita indica improved absorption and assimilation of phosphorus and nitrogen in Cunninghamia lanceolata plants under phosphorus starvation [39]. Similarly, S. indica increased the expression of genes encoding nitrate reductase in Arabidopsis thaliana [40]. While the endophyte fungus Mucor sp. increased the expression of genes related to phosphorous homeostasis in A. arenosa plants that Among identified DEGs, we explore those involved in response to metal stress, metal transport and plant growth promotion (Table 3). In the inoculated condition, we found down-regulated genes that play an important role against metal stress, among them: metallothionein-like protein 1, peroxidase 10 precursor and heavy metal-associated isoprenylated plant protein 28 included in GO terms: metal ion binding (GO:0046872), response to oxidative stress (GO:0006979) and metal ion transport (GO:0030001), respectively. Other down-regulated genes were related to metals transport, Natural resistance-associated macrophage protein 1 (Nramp1), Nramp3, Nramp5, Nramp 6, Metal tolerance protein 4 (MTP4), and Putative Multidrug Resistance Protein (MRP) among others. Also, genes implicated in nutrient transport were down-regulated, including high affinity nitrate transporter 2.5 (Nitrate transport, GO:0015706), Inorganic phosphate transporter 1-1 (Phosphate ion transmembrane transporter activity, GO:0015114), Potassium transporter 5 (Potassium ion transport, GO:0006813), Sodium transporter HKT1 (Response to osmotic stress, GO: 0006970; Sodium transporter, GO:0006814), Magnesium transporter MRS 2-3 (Magnesium ion transport, GO:0015693), and Ammonium transporter 1 member 1 (Ammonium transmembrane transport, GO:0072488). However, Sugar transporter ERD6-like 6, aquaporin TIP 2-1 and probable aquaporin PIP2-2 were up-regulated.
In addition, genes related to other abiotic and biotic stress were identified; transcription factors MYB 102, MYB 74 and probable WRKY transcription factor 72 were down-regulated. While MYB 86, Snakin-2 and Disease resistance protein RPP4 were up-regulated.
providing energy through hexose accumulation and facilitating water transport respectively, besides participating in the adaptation to several types of stress [41,42]. The storage of osmoprotectants such as sugars and amino acids allows to maintain the cellular turgor pressure necessary for cell expansion under stressful conditions [43]. Previous reports suggest that endophytic fungi can promote sugar accumulation in adverse conditions, which improves plant fitness, limiting water losses by decreasing the transpiration rate and through an osmotic adjustment [44]. This could explain the upregulation of genes related to sugar and water transport as well as Transcription factor (TF) MYB86 that regulates the stomatic aperture. Figure 6. Heat map showing differential expression of the most important up/downregulated genes between control (without microbial inoculation) and inoculated treatment (with fungal inoculation) conditions in Eucalyptus globulus roots transcriptome.
In addition, DEGs related to other types of stress were identified, including TF MYB102 and MYB74 (down-regulated), which were involved in the response to osmotic and saline stress respectively, and MYB86 (up-regulated) in the regulation of stomatal movement. It is well known that TFs can modulate the expression of genes, allowing them to respond and adapt to different environmental stimuli. MYB proteins participate in several important physiological processes, including control of the cell cycle, regulation of metabolism, synthesis of hormones, and response to several types of biotic and abiotic stresses [45]. Other genes were directly involved in plant defense response, among them, Probable WRKY transcription factor 72 (down-regulated), whose expression is induced in the presence of pathogens and also under saline and osmotic stress [46]. The role of WRKY transcription factor 72 has been described to modulate the resistance against Xanthomonas oryzae pathovar oryzae in rice plants [47]. Snakin-2 (up-regulated), a peptide with antimicrobial activity, which acts drilling membranes of the microbial cells of pathogens [48]. Wherewith, it is inferred that the inoculation of C. cupreum not only gives benefits to the plant in terms of protection against metal stress, but also plays an important role in regulating the expression of genes related to other types of biotic and abiotic stress.
In our study, we found genes related to biosynthesis and metabolism of auxins, which are responsible for the division, elongation and differentiation of the plant cell, and therefore, are directly

Discussion
Within abiotic stress types, metal toxicity is one of the factors that causes serious deleterious effects in plants. Nevertheless, the alleviation of heavy metal toxicity by endophytic fungi could be an efficient strategy to enhance heavy metal tolerance in plants [9]. In this study, colonization of C. cupreum was detected at the end of the experiment in samples of lateral roots and near to the stem of plants, which suggests that inoculation of C. cupreum is persistent over time. Endophytic fungi establish a chemical communication with the host through sugars, fatty acids, amino acids, polysaccharides, flavonoids, among others [29], and enters the plant through degradation of the cell wall or by fissure on roots [30]. Once inside the plant, the endophytic fungi can produce different chemical compounds with a beneficial effect on the performance of plants under heavy metal stress.
Exposure to metals causes the formation of reactive oxygen species (ROS) in plants, which leads to an imbalance in redox homeostasis [16]. Plants can counteract these negative effects by intracellular mechanisms such as the action of metal chelating peptides (metallothioneins and phytochelatins) and by activating antioxidant mechanisms. In our study, it was observed that inoculation of C. cupreum caused a down-regulation of genes involved in the detoxification of metals including, metallothionein-like protein 1 that acts by sequestering metals through thiol groups of their cysteine residues, whose distribution influences the capacity of union and sequestration of metals to maintain homeostasis [31]. The metallochaperone heavy metal-associated isoprenylated plant protein 28 (HIPP28), that acts against the excess of metals, binding them through their cysteine residues and transporting them to intracellular compartments, helping with the detoxification of metals and maintaining homeostasis [32], and peroxidase 10 precursor acts decomposing hydrogen peroxide generated in response to oxidative stress in addition to participating in the oxidation of reducing toxins, lignin biosynthesis, suberization and auxin metabolism, were also down-regulated [33].
On the other hand, several genes associated with the transport of metals were also down-regulated, among them: Nramp1, Nramp3, Nramp5, Nramp 6, MTP4 and MRP. Nramp genes play an important role in the uptake and translocation of a wide range of metal ions to the plant that include Cd, Zn, Fe, Cu and Mn, and it has been described that an up-regulation of these genes improves the accumulation of metals in Arabidopsis thaliana [34]. MTP4 has been described as a divalent cation efflux transporter that acts in the cytoplasm, being essential for the maintenance of metal homeostasis [35] and MRP realizes a similar role transporting metal ions to the vacuole as a detoxification mechanism of plants [36].
This suggests that the inoculation of C. cupreum prevents the metals present in the soil from being translocated to the roots of the plants of E. globulus, a finding which agrees with previous works where C. cupreum inoculation contributes positively to decreasing indicators of stress such as lipid peroxidation level and proline content because metal ions were adsorbed in the cell wall of the fungus [7]. However, this response mechanism depends directly on the plant-microorganism interaction, and on the type of host plant, because other studies showed that other endophytic species such as Mucor sp. despite having a protective effect on Arabidopsis arenosa plants by accumulating a lower amount of metals compared to the non-inoculated control, promotes translocation of metals from roots to shoots, which is reflected in an up-regulation of genes associated with metal transport and distribution such as HMA3, PCR2, ZIF1, and MTP1 [37]. Other transporters down-regulated were high affinity nitrate transporter 2.5, inorganic phosphate transporter 1-1, potassium transporter 5, sodium transporter HKT1, magnesium transporter MRS2-3, ammonium transporter 1 member 1 and high affinity sulfate transporter 1, which are directly related to plant nutrition. According to this, an increase in plant growth is not being promoted by an improvement in nutrient uptake. Other studies under C. cupreum inoculation reported similar results attributing the improvement in plant growth to a probable action of auxins [38]. Conversely, recent reports have shown an improvement in the absorption of nutrients under colonization of endophyte fungi. For example, the endophyte fungus Serendipita indica improved absorption and assimilation of phosphorus and nitrogen in Cunninghamia lanceolata plants under phosphorus starvation [39]. Similarly, S. indica increased the expression of genes encoding nitrate reductase in Arabidopsis thaliana [40]. While the endophyte fungus Mucor sp. increased the expression of genes related to phosphorous homeostasis in A. arenosa plants that developed in mining tailings [37]. Metalliferous environments are characterized by scarce vegetation, acid soils severely eroded and show high concentrations of metals, in addition to having a scarce supply of water and nutrients, which directly affects the development of plants [1]. In our study, sugar transporter ERD6-like 6, aquaporin TIP 2-1 and probable aquaporin PIP2-2 were up-regulated; these genes have important functions at the physiological level of the plant including nutrition and growth, providing energy through hexose accumulation and facilitating water transport respectively, besides participating in the adaptation to several types of stress [41,42]. The storage of osmoprotectants such as sugars and amino acids allows to maintain the cellular turgor pressure necessary for cell expansion under stressful conditions [43]. Previous reports suggest that endophytic fungi can promote sugar accumulation in adverse conditions, which improves plant fitness, limiting water losses by decreasing the transpiration rate and through an osmotic adjustment [44]. This could explain the up-regulation of genes related to sugar and water transport as well as Transcription factor (TF) MYB86 that regulates the stomatic aperture.
In addition, DEGs related to other types of stress were identified, including TF MYB102 and MYB74 (down-regulated), which were involved in the response to osmotic and saline stress respectively, and MYB86 (up-regulated) in the regulation of stomatal movement. It is well known that TFs can modulate the expression of genes, allowing them to respond and adapt to different environmental stimuli. MYB proteins participate in several important physiological processes, including control of the cell cycle, regulation of metabolism, synthesis of hormones, and response to several types of biotic and abiotic stresses [45]. Other genes were directly involved in plant defense response, among them, Probable WRKY transcription factor 72 (down-regulated), whose expression is induced in the presence of pathogens and also under saline and osmotic stress [46]. The role of WRKY transcription factor 72 has been described to modulate the resistance against Xanthomonas oryzae pathovar oryzae in rice plants [47]. Snakin-2 (up-regulated), a peptide with antimicrobial activity, which acts drilling membranes of the microbial cells of pathogens [48]. Wherewith, it is inferred that the inoculation of C. cupreum not only gives benefits to the plant in terms of protection against metal stress, but also plays an important role in regulating the expression of genes related to other types of biotic and abiotic stress.
In our study, we found genes related to biosynthesis and metabolism of auxins, which are responsible for the division, elongation and differentiation of the plant cell, and therefore, are directly involved with the growth of plants. We observed that genes classified as early response to auxins were up-regulated, whose function is to regulate cellular responses to different levels of auxins present in the plant [49]. Thereby, once the presence of auxins inside the cell is detected, several processes are triggered that modulate the expression of auxin response genes where two main families of proteins are involved, Auxin/Índole-3-acetic acid (Aux/IAA) and Auxin response factor (ARF). When there are low auxin concentrations, the repressor proteins Aux/IAA (Auxin-induced protein AUX22, Auxin-induced protein 22D, Auxin-induce protein 22A, Auxin-responsive protein IAA3 and Auxin-responsive protein IAA4) form complexes with ARF proteins which regulate the expression of auxin-responsive genes, preventing their actions as transcription factors; while, when the auxin concentration increases, it binds to other receptors (TIR1 / AFB) together with other proteins (ASK1, CUL1 and RBX) and forms a complex of ubiquitination that binds to the AUX / IAA repressor proteins and degrades them in the 26S proteasome, releasing the ARF proteins to activate or repress the transcription again [50]. Other overexpressed genes were Indole-3-acetic acid amino synthetase GH3 which were involved in the synthesis of IAA conjugates, providing a mechanism for the plant to counteract the excess of auxin, auxin efflux carrier component 2, which transports this phytohormone between different cells and tissues of the plant. Small auxin upregulated RNAs (SAURs) are the largest family of genes for early response to auxins and are closely related to cell expansion and plant growth along with regular abiotic stress tolerance responses such as saline and drought [51]. Auxin-induced roots cultures protein 12 (AIR12) was also overexpressed and has been described as an induced auxin involved in the development of lateral roots [52]. In previous studies, we observed that under the inoculation of fungal strain, plant growth is stimulated [7] what was reflected in lateral roots proliferation and biomass production respect to control, demonstrating the plant growth promoting effect of fungus C. cupreum.

Conclusions
Transcriptome changes in E. globulus inoculated with C. cupreum resulted in the detection of several genes involved in stress response to heavy metals and plant growth promotion. The protective effect showed by the inoculation of C. cupreum against metal stress is mainly due to the repression of metal transporters in the plant, which added to the ability of C. cupreum to fix metal ions on its cell surface (biosorption), preventing the generation of a ROS imbalance that subsequently triggers the activation of metal chelation mechanisms and activates the antioxidant system. Furthermore, it was found that the promoter effect of plant growth is given by a complex regulation of auxin biosynthesis and metabolism, not by an improvement in nutrient uptake.