Introduction

Metallothioneins (MTs) are low molecular weight, cysteine (Cys)-rich, metal-binding proteins. The first MT discovered was a cadmium (Cd)-binding peptide found in horse kidney (Margoshes and Vallee 1957). Subsequently, intensive studies on this small polypeptide revealed its wide distribution across taxonomic groups, from prokaryotes to eukaryotes (Babula et al. 2012; Cicatelli et al. 2010; Divya et al. 2018; Hassinen et al. 2011; Le Croizier et al. 2018; Sigel et al. 2009). MT sequences are highly heterogeneous in protein length and the distribution of Cys residues and do not show general homology, which has resulted in a classification system of 15 families encompassing taxa from bacteria to humans (Binz and Kägi 1999). In mammals, four tandemly clustered genes (MT1 to MT4) are known, and these four major genes originated through a single duplication event prior to the radiation of mammals (Moleirinho et al. 2011). A recent study suggested a likely association between MT functional divergence and duplication events in vertebrates (Serén et al. 2014). Iturbeespinoza et al. (2016) discovered the longest fungal MT in Tremella mesenterica, a saprophytic ascomycete. The TmMT gene has 10 exons, and it yields a 779-bp mature transcript encoding a 257-residue protein, built of repeated fragments.

Compared with MT genes in other taxa, fungal MT genes have higher heterogeneity (Binz and Kägi 1999) CUP1, the first identified fungal MT gene, found in the yeast Saccharomyces cerevisiae, is mainly involved in copper (Cu(II)) detoxification and induced by Cu(II) and silver (Ag(I)) (Butt et al. 1984; Okuyama et al. 1999). Another MT gene, CRS5, was also found in S. cerevisiae but is not homologous to CUP1. CRS5 was shown to exist mainly in the form of a Cu(II)/zinc (Zn(II)) complex in S. cerevisiae and to play a crucial role in maintaining cytoplasmic Zn homeostasis (Culotta et al. 1994; Pagani et al. 2007). In addition, four MT-like genes were identified in the yeast Yarrowia lipolytica by whole genome analysis that were predicted to have evolved through recent gene duplication events (Dujon et al. 2004). Similar MT-like proteins were also identified in ascomycete fungi, including Colletotrichum gloeosporioides, Pyrenopeziza brassicae, Podospora anserine, Heliscus lugdunensis, and Tuber melanosporum (Averbeck et al. 2001; Hwang and Kolattukudy 1995; Jaeckel et al. 2005; Loebus et al. 2013; Singh and Ashby 1998). In basidiomycete fungi, most small Cu-binding Cys-rich peptides, presumably MTs, were isolated from the ectomycorrhizal (ECM) fungus Laccaria laccata and Paxillus involutus (Howe et al. 1997). Subsequently, a gene encoding the 34-amino acid (aa) metallothionein PiMT1 was identified as a potential determinant in the responses of P. involutus to Cu(II) and Cd(II) stresses (Bellion et al. 2007; Jacob et al. 2004). Two MTs in Hebeloma cylindrosporum, which share only 40% identity, exhibit different expression responses to Cu(II) and Cd(II) overload stresses (Ramesh et al. 2009). Three MT isoforms isolated from the Ag-hyperaccumulator Amanita strobiliformis were shown to play vital roles in the sequestration of intracellular Ag(I) in the fruit body and mycelia (Osobová et al. 2011). Nguyen et al. (2017) identified two more MT coding genes: SlMTa and SlMTb in the ECM fungus Suillus luteus. Reddy et al. (2016) reported the expression levels of PaMT1 in Pisolithus albus, which encodes a 35 amino acid long polypeptide, increased higher with Cu(II) than Cd(II).

MTs have also been characterized in the arbuscular mycorrhizal (AM) fungi Gigaspora rosea, Gigaspora margarita, and Glomus intraradices (González-Guerrero et al. 2007; Lanfranco et al. 2002; Stommel et al. 2001). In addition to their pivotal role in the detoxification of heavy metals, increasing lines of evidence indicate a role of MTs or MT-like proteins in the development of fungi and plant-fungus interactions (Hwang and Kolattukudy 1995; Lanfranco et al. 2002; Johansson et al. 2004). Formation of the appressorium is essential for the penetration of C. gloeosporioides into its host, the avocado fruit (Hwang and Kolattukudy 1995). Hwang and Kolattukudy (1995) also found that the expression levels of CgMT1 and CgMT2 were significantly upregulated in the early phase of appressorium formation. The study of ECM fungi has similar results. Johansson et al. (2004) found that when P. involutus formed a symbiotic relationship with ECM fungus Betula pendula, the expression levels of P. involutus MT genes were significantly upregulated.

With the increasing number of fully sequenced fungal genomes (Floudas et al. 2012; Grigoriev et al. 2011; Kohler et al. 2015; Martin et al. 2008), MTs from various taxonomic groups can be easily retrieved. However, knowledge of the full range of fungal MT functions are unclear. As the first ECM fungus with a completely sequenced genome and the ability to form symbioses with a variety of host plants, Laccaria bicolor has become a model species for studies of symbiotic relationships between ECM fungi and woody plants and their processes of intrinsic signal recognition, signal transduction, and other molecular mechanisms (Martin et al. 2008). In this study, we identified eight MT genes in L. bicolor, present an overview of ECM fungal MTs, and described their genetic structure, phylogeny, evolution, and expression patterns in L. bicolor. In addition, we present results from complementary experiments testing the eight MTs identified in the L. bicolor genome in metal-deficient yeast mutants to determine their effects on Cu(II), Cd(II), and hydrogen peroxide (H2O2) tolerance.

Materials and methods

Organisms and growth conditions

The L. bicolor (Maire) P.D. Orton S238N strain used in this study was donated by Prof. A. G. Pardo (University of Quilmes, Argentina). The mycelia were kept in modified Melin-Norkrans (MMN; Marx 1969) medium and cultured in a dark incubator at 25 °C.

For the L. bicolor resistance to different heavy metals, a solid MMN medium was prepared, autoclaved at 121 °C for 20 min, and then heavy metal solutions were filtered by 0.22 μm filter and were, respectively, added to form different concentration gradients (Cu(II): 0, 10, 20, 50, 100, 150, 200 μM; Cd(II): 0, 1, 2, 5, 10, 20, 40 μM; Zn(II): 0, 0.5, 1.0, 1.5, 3.0, 4.5, 9.0 mM; H2O2: 0, 0.1, 0.2, 0.5, 1.0, 2.0, 5.0 mM; Ag(I): 0, 5, 10, 20, 40, 80, 150 μM). The activated fungal plugs of L. bicolor with an initial diameter of 0.9 cm were separately cultivated in the solid MMN media with different treatments in the dark at 25 °C for 28 days, and then the pictures were taken.

For transcriptional induction experiments, fungal mycelia were added to liquid MMN medium and incubated for 21 days. Three plugs of 9-mm diameter were transferred to 50 mL of fresh medium for 3 days (adaption period). External metal ions (Cu(II): 50, 100, 150 μM; Cd(II): 5, 10, 20 μM; Zn(II): 0.5, 1.5, 3.0 mM; Ag(I): 10, 20, 40 μM) and H2O2 (0.5, 1.0, 2.0 mM) were added to the medium, and the mycelia were harvested separately after 6, 12, 24, and 48 h incubation and then washed twice with distilled water.

The yeast strains DTY4 (cup1::URA3; donated by Prof. Dennis J. Thiele from Duke University, USA) and WYU (yap1::URA3; donated by Prof. Shusuke Kuge from Tohoku Pharmaceutical University, Japan), which were derived from the wild-type strain DTY3 (MATα leu2-3, 112 his3∆1 trp1-1 ura3-50 gal1 CUP1S) (Longo et al. 1996) and W303B (MATα his3 can1-100 ade2 leu2 trp1 ura3) (Kuge and Jones 1994), were used for heterologous expression analyses, respectively. Transformed yeasts were grown at 30 °C on URA+-selective SD agar medium containing (w/v) 0.7% yeast nitrogen base (Difco Laboratories, Inc., Franklin Lakes, NJ, USA), 0.005% adenine hemisulfate, 2% glucose, and 0.003% (each) essential amino acids (Sigma-Aldrich, St. Louis, MO, USA).

DNA and RNA isolation and cDNA synthesis

Fungal mycelia were frozen in liquid nitrogen, and then 100 mg of mycelia were ground for DNA/RNA extraction. Genomic DNA and total RNA were isolated using the E.Z.N.A. Plant DNA Mini Kit and E.Z.N.A. Plant RNA Mini Kit (Omega Bio-tek, Inc., Norcross, GA, USA), respectively. For RNA purification, DNA removal was achieved by incubating samples with RNase-free DNase (Omega Bio-tek) per the manufacturer’s recommendations. Total RNA concentration was determined by measuring absorbance at 260 nm, and RNA integrity was checked by formaldehyde agarose gel electrophoresis.

RNA was converted to cDNA using a RevertAid First-strand cDNA Synthesis kit (Thermo Fisher Scientific, Waltham, MA, USA). The first cDNA strand was reverse-transcribed in a 12-μL reaction from 1 μg of total RNA using anchored oligo(dT) 18 primers.

Cloning of MTs in L. bicolor

Basic Local Alignment Search Tool (BLAST) searches with reported P. involutus and H. cylindrosporum MT sequences against L. bicolor expressed sequence tags (ESTs) and genomic sequences were performed using the Joint Genome Institute (JGI) L. bicolor genome portal v2.0 (https://genome.jgi.doe.gov/Lacbi2/Lacbi2.home.html). Eight unique genes (LbMT1, LbMT2a, LbMT2b, LbMT3a, LbMT3b, LbMT3c, LbMT4, and LbMT5) representing MT homologs were detected in both the genome and the EST library. These potential MTs were subsequently cloned from both genomic DNA (as a control) and cDNA from L. bicolor. PCR amplification was performed with primer sets introduced a BamHI and an EcoRI site at the 5’ and 3’ ends, respectively (Table S2). The PCR program was as follows: an initial denaturation at 94 °C for 5 min, followed by 35 cycles at 94 °C for 30 s, 60 °C for 30 s, and 72 °C for 30 s, and a final extension at 72 °C for 10 min. PCR products were purified using a MiniBEST Agarose Gel DNA Extraction Kit (Takara Bio, Inc., Tokyo, Japan) and then directly cloned into the pMD19-T vector (Takara) and transformed into competent Escherichia coli cells. DNA sequences were determined using M13 RV and M4 primers (Thermo Fisher Scientific). The full-length cDNA sequences of eight L. bicolor MTs were submitted to the DDBJ/EMBL/NCBI DNA databank under the accession numbers KJ095785, KJ095786, KJ095787, KJ095788, KJ095789, KJ095790, KJ095791, and KJ095792.

Organ-specific expression count and promoter analysis

The upstream sequences (1,500 bp upstream of the start codon) of eight MT genes were cloned from L. bicolor genomic DNA and sequenced. Potential transcription factor binding sites (TFBSs) were predicted using Genomatix software (Genomatix GmbH, Munich, Germany) and the BIOBASE TRANSFAC database (Matys et al. 2003).

The relevant sequences of the eight cloned MTs genes were extracted from the L. bicolor EST database published on NCBI and their abundances at different growth and development stages were counted to represent these eight relative expression different developmental stages of L. bicolor. The selected plant was Douglas fir associated.

Sequence alignment and phylogenetic analysis

Multiple protein sequence alignments were performed using MUSCLE alignment software (Edgar 2004), and gaps and ambiguously aligned sites were manually removed. Phylogenetic reconstructions were performed using a maximum likelihood method using MEGA X software (Tamura et al. 2011) and PhyML 3.1 (Guindon et al. 2010) software with a bootstrap of 1000 pseudoreplicates. The appropriate substitution model was determined in MEGA X by applying the model estimator.

Real-time quantitative PCR of L. bicolor MT genes

The cDNA was normalized by dilution and amplified by qPCR using an iQ SYBR GREEN Supermix (Bio-Rad Laboratories, Hercules, CA, USA) following the manufacturer’s instructions and run on a MiniOpticon Real-Time PCR System (Bio-Rad). Laccaria tubulin-1B alpha chain gene (protein ID: 192524) was used as reference gene for normalization (Kemppainen et al. 2009). Primers are presented in Table S3. The qPCR program was as follows: 95 °C for 10 min, then 40 cycles at 95 °C for 10 s, followed by 65 °C for 30 s. Samples were run in three biological replicates with two technical replicates each. Fold changes in gene expression between exogenous stressed mycelia and free-living mycelia were calculated with the ∆∆Ct method (Pfaffl 2001). All statistical analyses were performed using the IBM SPSS 21.0 software package (IBM Corp., Armonk, NY, USA). The average value of expression multiples is ≥ 2.0 can be considered as differentially expressed. A Student’s two-tailed independent t test was used to determine the significance of the results (P < 0.05).

Yeast complementation assay

The full coding regions of the eight MT genes of L. bicolor were subcloned from pMD19-T clones into the yeast expression vector P424 (Mumberg et al. 1995). Transcription was controlled by the constitutive glyceraldehyde-3-phosphate dehydrogenase (GPD) promoter and the cytochrome C oxidase gene (CYC1) terminator of S. cerevisiae. Empty vector P424 and the P424-LbMT constructs were introduced into cup1∆ and yap1∆ cells together with the parental strains using a lithium acetate procedure (Gietz and Schiestl 2007). Transformants were utilized for serial dilution spot assays. Briefly, strains were first cultured in liquid SD-Trp medium for 24 h at 30 °C. Cell density was adjusted to an OD600 value of 1.0, and 5.0-μL serial dilutions were spotted onto SD-Trp agar plates in the presence or absence of heavy metals (Cu(II): 100 μM; Cd(II): 60 μM) and H2O2 (2.0 mM). Growth was examined by visual inspection and digital image recording after incubation at 30 °C for 3 days.

Results

Effect of heavy metals and oxidative on the growth of L. bicolor

As shown in Figure S1, Cu(II), Cd(II), Zn(II), Ag(I), and H2O2 adversely affected the growth of L. bicolor. Zn(II) had the least effect, followed by H2O2, Cu(II), and Ag(I), because L. bicolor could still grow under 9.0 mM Zn(II) stress. Cd(II) had the most obvious inhibitory effect on strain growth due to L. bicolor could only tolerate Cd(II) up to 40 μM, which was almost 1/225 of Zn(II). L. bicolor stopped growing under the Ag(I) stress of 150 μM, which was the lowest value other than Cd(II).

MT gene family in L. bicolor

ESTs of the eight LbMT genes were obtained from the NCBI EST database as counts per million transcripts for each of the given tissue and displayed as log2 transcripts per million.

To identify the potential MT-like genes in L. bicolor, MTs from P. involutus and H. cylindrosporum were selected as query sequences for BLAST searches. The whole genome (JGI L. bicolor v 2.0) revealed the presence of 8 potential MT genes and 15 protein products in L. bicolor, which were further subdivided into 5 types (Table 1). Analysis of DNA arrangements of MT exons and introns demonstrated that all eight MT genes contain three or four exons, the first of which is extremely short (only nine nucleotides; Figure 1A). Sequence alignment revealed high similarity between LbMT2a and LbMT2b, LbMT3a, LbMT3b, LbMT3c, LbMT4, and LbMT5 also had high sequence similarity. LbMT1 had particularly high sequence similarity with PiMT1, isolated from P. involutus (Figure 1B).

Table 1 Members of metallothionein family (protein IDs based on Joint Genome Institute annotations) in L. bicolor.
Full size table
Fig. 1
figure 1

Structures of L. bicolor metallothionein (MT) genes. A Exon-intron gene organization of the eight MT genes discovered in the L. bicolor genome. Exons and introns are indicated with purple boxes and lines, respectively. B Alignment of eight LbMT protein sequences. To indicate sequence similarities, MT2a and MT2b are aligned together, and MT3a, MT3b, MT3c, MT4, and MT5 are aligned in a second group. MT1 is presented alone and shows high homology with a functional MT identified in Paxillus involutus

Full size image

The eight metallothionein-like genes of L. bicolor were located on three different pseudochromosomes corresponding to linkage groups (LGs) within the genome (Labbé et al. 2008). LbMT2a and LbMT2b are clustered on LG 7, separated by 51 kb. LbMT3a, LbMT3b, LbMT3c, LbMT4, and LbMT5 are located in a cluster on LG 10 (Figure 2). LbMT1, which is a single-copy gene, is on scaffold 15. The close sequence homology among a number of proteins encoded by MT genes (Figure S2) suggests that the MTs in L. bicolor may have been generated by gene duplication. Transposable elements (TEs), reported to be widely distributed throughout the L. bicolor genome, may be major contributors to the genesis of new genes through gene duplication via mechanisms such as ectopic recombination, transposition molecular domestication, and gene retrotransposition. We analyzed the presence of TEs within the vicinity of each MT gene using an online database (https://mycor.nancy.inra.fr/IMGC/LaccariaGenome/download.php?select=anno). The highest density of TEs was found on LG 10 clustered around LbMT3a, LbMT3b, LbMT3c, LbMT4, and LbMT5, and there was also a dense cluster of TEs adjacent to LbMT2a and LbMT2b. No TEs were found within 60 kb up- or downstream of LbMT1.

Fig. 2
figure 2

Schematic illustration of the MT gene family in the L. bicolor genome. The transcriptional direction of each gene is indicated by an arrow. Genes are colored as follows: LbMT2a and LbMT2b, dark blue; LbMT3a, LbMT3b, and LbMT3c, green; LbMT4, red; and LbMT5, light blue. LbMT1 is a single-copy gene on scaffold 15 and is not shown in this figure. The sizes of genes and intergenic regions are given in kilobases (kb)

Full size image

Expression profiles of the LbMT genes

Because gene duplication often results in diversified spatiotemporal expression patterns in duplicate family members (Moleirinho et al. 2011; Byrne and Wolfe 2007; Greer et al. 2000; Gu et al. 2004), we collected EST sequences derived from various developmental stages of L. bicolor. According to the EST pattern abundance, all eight MT genes revealed diverse expression preferences among different development stages (Figure 3A). The majority of MTs (LbMT1, LbMT2a, LbMT3a, LbMT3b, LbMT3c, and LbMT4) were highly expressed in free-living mycelia. However, during the formation of ECM symbioses with poplar roots, LbMT2a and LbMT2b were expressed more highly than the other MT genes in the symbiotic mycelia, indicating an important role for these two paralogous genes in host root colonization. In Douglas fir-associated fruiting bodies of L. bicolor, LbMT3c and LbMT4 were observed to be the most highly expressed MT genes and presumably participated in metal ion homeostasis during fruiting body formation. The basal mRNA expression of LbMT5 remained relatively constant at low levels in all three stages.

Fig. 3
figure 3

Expression profiles and promoter analysis of L. bicolor MT genes. A Expression of the eight MT genes at different development stages. B In silico analysis of the promoter regions (~ 1500 bp) of the eight LbMTs genes. Letters above the rectangles indicate various transcription factor binding sites (TFBSs; Table S1), and blue rectangles represent metal response elements. Nutrient-related TFBSs in symbiosis-associated MTs (LbMT2a, LbMT2b, LbMT3c, and LbMT4) are colored as follows: carbon-responsive element, purple; nitrogen-response element, orange; and phosphorus-response element, yellow

Full size image

The cloned LbMT1 exists as a single copy in the L. bicolor S238N genome and has high homology with the previously reported MT gene PiMT1 in P. involutus (Figure 1B). Therefore, in this study, LbMT1 was selected as the research object, and its transcription level changes under different heavy metals and H2O2 stress were detected by real-time PCR (Figure 4, Figure S3). The effects of different heavy metals and H2O2 on the expression of LbMT1 were shown in Figure 4. Under the treatments of 100 μM and 150 μM Cu(II), the expression of LbMT1 significantly increased at 12 hours and then decreased in the following time. For Cd(II), the expression of LbMT1 increased significantly under the stress of 10 μM Cd(II) at 6 h and 12 h and declined at 24 h. For H2O2, the expression level of LbMT1 increased when treated by 2.0 mM H2O2 for 12 h and 1.0 mM H2O2 for 24 h, respectively, and there was no significant change in other treatments. The expression of LbMT1 did not change significantly under different Zn(II) and Ag(I) treatments during the experiment (Figure S3).

Fig. 4
figure 4

Impacts of heavy metal ions (Cu(II) and Cd(II)) and oxidative(H2O2) stress on LbMT1 expression. All expression values are described as fold differences from expression in free-living mycelia ± standard deviations. * indicates significant up- or down- regulation of MT gene expression at different processing times (6, 12, 24, or 48 h) under the same treatment concentration (P < 0.05)

Full size image

Promoters analysis

To investigate the regulatory mechanisms underlying the observed expression pattern differences, we performed in silico analyses of the respective upstream promoter regions of the LbMT genes. Differences in the types and locations of potential TFBSs in promoter regions were observed even among closely related MTs (Figure 3B; Table S1). Classic metal response elements were identified in the upstream regions of all LbMT genes except MT2b and MT4. Additionally, the LbMT2a and LbMT2b promoters contained several TFBSs potentially related to carbon and nitrogen utilization. Similar observations were found in the promoter regions of LbMT3c and LbMT4 (Figure 3B). A phosphate starvation response element was also found upstream of LbMT3c.

Heterologous expression of LbMTs in yeast

To further document the impact of MT secondary structures on their capacity to bind metal ions and detoxify oxidative stress, we expressed L. bicolor MTs in mutant S. cerevisiae strains. The S. cerevisiae cup1 locus encodes a Cu-thionein that plays a dominant role in Cu(II) sequestration. The wild-type strain DTY3 harbors a single copy of the MT gene, whereas the cup1∆ mutant strain exhibits a greater sensitivity to Cu(II) due to complete disruption of the MT gene. Figure 5 shows that the cup1∆ yeast transformed with empty vector exhibited no growth at 100 μmol L–1 Cu(II). In contrast, the Cu-sensitive phenotype of the cup1∆ mutant was fully complemented by each of the eight L. bicolor MTs. Yap1 is a transcription factor related to the mammalian AP-1 complex which positively controls various genes involved in metal tolerance and oxidative stress tolerance in yeast. Complete disruption of this transcription factor renders the yeast mutant highly sensitive to Cd(II) and H2O2 (Kuge and Jones 1994). As with the results seen in the cup1∆ mutant, expression of each of the LbMT genes in yap1∆ yeast resulted in a full complementation of Cd-sensitive cells, enabling them to tolerate higher concentrations of Cd(II) compared with the wild-type W303B. In contrast, the presence of MTs in yap1∆ had no influence on cell growth with or without excess H2O2(Figure. 5).

Fig. 5
figure 5

Functional complementation in yeast mutants on selective medium. Cellular tolerance toward Cu(II), Cd(II), and H2O2 following overexpression of eight LbMTs genes in cup1 and yap1 mutant yeast. Mutants and the corresponding wild-type (WT) strains were transformed with empty vector p424 as the negative and positive controls. Serially diluted cells were spotted on SD-Trp medium, and yeast growth was assessed after 3 days of incubation at 30 °C

Full size image

Discussion

The current availability of several genomic sequences enables the study of the evolutionary steps underlying the expansion of gene families by detailed characterization of lineage-specific expansions. Due to the high heterogeneity of MT sequences in fungi, they are generally omitted in genome sequencing projects. Using ESTs in BLAST searches querying reported basidiomycete MTs, we identified eight MT-like genes in both the genomic and EST datasets of the first fully genome-sequenced ECM fungus, L. bicolor (Martin et al. 2008). Among the eight putative LbMT genes, the conserved arrangement of exon-intron positions and high protein identities suggest that these MTs originated from gene duplications. Consistent with this finding, abundant TEs were identified around MT2a and MT2b on LG7, as well as MT3a, MT3b, MT3c, MT4, and MT5 on LG10. The L. bicolor S238N genome encodes a high percentage of TEs, which could be a major contributor to its greater quantity of multi-gene families compared with other fungal genomes (Labbe et al. 2012), as TEs are associated with gene duplication and exon shuffling (Bennetzen 2005; Morgante et al. 2005). These data are in agreement with previous observations on MT duplication in mammals (Moleirinho et al., 2011). After a duplication event, the genes that can follow distinct evolutionary paths: if a gene is actively maintained, redundant duplicates can escape purifying selection and begin accumulating loss-of-function mutations, resulting in pseudogenization; or, less frequently, particular replacements may direct this gene into novel functions (Nikolaos et al. 2021; Audrey et al. 2017). Subfunctionalization can also occur if parents and duplicates retain their functions, but become distinct and complementary in their spatiotemporal expression patterns (Zhang 2003).

Considering the large MT gene family in L. bicolor in the context of the discovery that gene duplication often results in diversified spatiotemporal expression patterns in duplicate family members, EST abundance can be utilized as a valuable metric for the basal expression patterns of the corresponding genes. Specifically, the expression levels of some MTs in L. bicolor are higher than others during the formation of ECM symbiosis, indicating an important role in plant-fungus associations. Supporting the results obtained with the EST datasets, a series of nutrient-associated TFBSs that primarily may participate in carbon, nitrogen, and phosphorus starvation responses were identified in the promoter regions of these genes. Symbiosis-regulated MTs were also discovered in both mycorrhizal fungi and certain plants (Lanfranco et al. 2002; Johansson et al. 2004; Flores-Monterroso et al. 2013; Voiblet et al. 2001). Previous studies have shown that MTs play critical roles in fungal development and plant-fungus interactions. For instance, two MT I-type genes in the plant pathogen C. gloeosporioides were uniquely expressed during appressorium formation by contact with host surface wax (Hwang and Kolattukudy 1995). Similarly, in the ascomycete fungus Magnaporthe grisea, known for causing rice blast disease, an unusual MT-like protein (22 aa long with only six Cys residues) was identified be play a novel role in the biochemical differentiation of the appressorium cell wall (Tucker et al. 2004). Furthermore, MTs were also differentially expressed during the symbiotic mycorrhizal period. Transcriptome analysis of ECM roots formed between P. involutus and B. pendula shown that a series of genes, including a fungal MT, were upregulated compared with the saprotrophic growth condition (Morel et al. 2005). In contrast, a 65-aa MT from the AM fungus G. margarita appeared to be downregulated in the symbiotic mycelia (Lanfranco et al. 2002).

Metallothionein is a low molecular weight, cysteine-rich metal binding protein (Audrey et al. 2017). MTs in different species generally have the ability to chelate heavy metals, maintain the balance of intracellular metal ion concentration, and scavenge reactive oxygen species (Sutherland and Stillman 2011). By chelating heavy metals, it reduces the accumulation and distribution of biological heavy metals to help organisms deal with heavy metal stress in the environment. For example, OsMT1e in Oryza sativa can reduce the accumulation of Cd in root and shoot (Rono et al. 2021). In ECM fungi, MTs also have a similar effect. Sacky et al. (2014) cloned three MTs genes in Hebeloma mesophaeum. Among them, HmMT1 is mainly responsible for the complexation of Zn(II) and Cd(II) in the cell and is finally transported in vacuoles and intracellular vesicles. HmMT2 and HcMT3 chelate with Ag(I) in the cytoplasm, reducing the toxic effect of Ag(I) on cells. Higher eukaryotic MTs are generally induced by a wide array of metals and stress conditions (Cobbett and Goldsbrough 2002). In contrast, each of the fungal MTs studied thus far is induced by a limited number of heavy metals, with most being induced by Cu(II) (Averbeck et al. 2001; Ramesh et al. 2009; Cobine et al. 2004; Kumar et al. 2005). This was also the case for the MTs characterized in this work, and the single-copy gene LbMT1 was chosen for further expression investigations. In contrast with PiMT1 found in P. involutus (Bellion et al. 2007), the expression of LbMT1 remained constant at high expression levels in the free-living mycelium (FLM) of L. bicolor and was induced by relatively high Cu(II) and Cd(II) concentrations, giving rise to the hypothesis that LbMT1 may play a housekeeping role in Cu(II) and Cd(II) homeostasis (Figure 4). In the study of Reddy et al. (2016), the PaMT1 gene identified from Pisolithus albus can be induced by Cu(II) or Cd(II), and the expression of PaMT1 gene under Cd(II) stress is lower than that under Cu(II) stress. This is similar to our results. In our study, LbMT1 can be induced by high concentration of Cu(II) and medium to high concentration of Cd(II), and the expression level of LbMT1 under Cd(II) stress is lower than that under Cu(II) stress. The similar results in Suillus himalayensis were also observed by Kalsotra et al. (Kalsotra et al. 2018). The expression levels of ShMT1 and ShMT1 genes of S. himalayensis under Cu(II) stress are higher than under Cd(II) stress. In the study of Courtois et al. (2020), Ag(I) was used to treat Eisenia fetida for 5 weeks, and no changes in the expression of Cd-related metallothionein and the oxidative stress-related genes were observed. We also found no effect of Ag(I) stress on the expression of LbMT1. Zn(II) is less toxic to L. bicolor (Figure S1), so it may not be more toxic to fungus. The expression of LbMT1 is also not induced by Zn(II). Liu et al. (2005) shown that the inducing activity of MTs on Cd(II) in yeast cells is mediated by oxidative stress, which is similar with our result, and it may be due to the oxidative stress caused by H2O2 after inducing the expression of LbMT1. The function of metallothionein (MT) is currently characterized as chelating metals in cells and as a free radical scavenger (Chatterjee et al. 2020). Some reagents that can induce the formation of free radicals and can also increase the expression of MTs (Bauman et al. 1991). Pakdee et al. (2019) showed that oxidative stress can increase the expression of PpMT1.2a. At the same time, some MTs also have the effect of removing H2O2 (Mierek-Adamska et al., 2019). In our study, the expression level of LbMT1 was increased when treated by 2.0 mM H2O2 for 12 h and 1.0 mM H2O2 for 24 h. This indicated that LbMT1 may be involved in the detoxification of L. bicolor under high-concentration short-term or low-concentration long-term H2O2 stress. Many MTs can be induced by Cu(II), but some MTs can be induced by Cu(II) and Cd(II), and the expression level under Cu(II) stress is higher than Cd(II) stress (Reddy et al. 2016; Kalsotra et al. 2018). These results were similar to our study. Both Cu(II) and Cd(II) can increase the expression of LbMT1, but the expression of LbMT1 under Cu(II) stress was higher than under Cd(II) stress. One possible reason is that Cd(II) has other chelating substances in the cell, such as GSH (Courbot et al. 2004). However, LbMTs still have the ability to improve the Cd(II) tolerance of L. bicolor. Yeast experiments have shown that all yeasts can grow under 60 μM Cd(II) stress after being transformed into LbMTs, but WT and yap1∆ cannot. However, replenishing the eight LbMTs genes does not improve the H2O2 tolerance of yeast. This may be because the main role of MT is to chelate heavy metals in the environment (Hauser-Davis et al. 2021), so it has no obvious replenishing effect on H2O2 stress. The sulfhydryl group of MT has nucleophilic properties and is easy to act on some electrophilic substances, especially easy to combine with free radicals, which can achieve the effect of eliminating free radicals (Chatterjee et al. 2020). Many MTs can eliminate oxidative damage (Jin et al. 2017; Zhou et al. 2014). Many studies have shown that MTs can be induced by H2O2 (Mierek-Adamska et al. 2019). Therefore, when the fungus is subjected to oxidative stress, the expression of MT will be increased to cope with the oxidative stress.

Under 50–100 μM Cu(II) or 1.0–2.0 mM H2O2 stress, the growth of L. bicolor was inhibited to a similar degree (Figure S1); therefore, there was no much difference in the expression level of LbMT1 under these four stresses, because MT’s scavenging effect on oxidative stress is weaker than that of metal chelation (Figure 5). Under Cu(II) stress, the expression level of LbMT1 was slightly higher than H2O2 stress. Under the stress of 10 μM Cd(II), the growth of L. bicolor was inhibited to a great extent (Figure S1), and excessively high concentrations of Cd(II) will inhibit the expression of most genes in ECM fungi (Ruytinx et al. 2011). Therefore, although the expression of LbMT1 was increased under 10 μM Cd(II) stress, it was still no much different from the former, even slightly lower than Cu(II) stress.

Previous studies have found that the expression of MTs genes in ECM fungi is significantly up-regulated or down-regulated in the process of ECM fungi interacting with host plants (Johansson et al.,2004). This is similar to our results. LbMT2a, LbMT2b, LbMT3c, and LbMT4 play a certain role in the symbiosis process of L. bicolor and plants. Their specific functions are still unclear. It is speculated that the nutrient exchange between them occurs during symbiosis with L. bicolor and plants. Based on the levels of induction of MTs by different metals, Capdevila and Atrian (2011) proposed the concept of MT evolution, which suggests that functional adaptations of MT genes reflect specific cellular demands for the expression level and metal specificity of a particular MT peptide.

It has been suggested that MT sequences preceded by a stretch of 10–13 aa lacking Cys residues may be a specific feature of the ECM basidiomycetes (Osobová et al. 2011). The functional importance of this feature, if any, is unknown. As for the LbMT2s, the C-terminal moiety seems to derive from other short LbMTs, as they all share conserved Cys residues, and the LbMT2 gene has conserved the 39-bp intron found in all of the short MTs in this study. Shared characteristics of LbMT1, LbMT3a, LbMT3b, LbMT3c, LbMT4, and LbMT5 proteins include their small size and the presence of 7−9 Cys residues (representing about 22% of the total aa content) with two aromatic residues. The short MTs identified in different basidiomycete species likely share a common ancestor (Reddy et al. 2014). To test the roles of L. bicolor MTs after gene duplication, we performed a yeast complementation assay. Heterologous complementation assays in yeast demonstrated that the eight LbMT genes encode functional peptides capable of conferring increased tolerance against Cu(II) and Cd(II), thereby confirming that LbMT peptides may defend fungal cells against metal ions. The data are in agreement with the results of many related studies. For example, SlMTa and SlMTb can endow S. luteus Cd(II) and Cu(II) tolerance (Nguyen et al. 2017). Previous experiments in Drosophila melanogaster demonstrated that the number of functional MT gene duplications correlates directly with Cu(II) and Cd(II) resistance (Moleirinho et al. 2011). Clustered MT genes were coregulated and conferred tolerance to multiple abiotic stressors in rice (Kumar et al. 2012). In contrast, in our study, H2O2-elicited oxidative stress handling was not affected by the expression of LbMTs in mutant yeast, suggesting that LbMTs expression responds more directly to metals than to oxidative stress.

Conclusions

In conclusion, eight MTs genes were identified and cloned from the ECM fungus L. bicolor S238N. Genomic analysis found that, except for LbMT1, there were a large number of transposon elements in the upstream and downstream 60 kb range of the other 7 LbMTs genes with high similarity. It is speculated that these MTs are formed by gene tandem duplication. Through complementary expression in the corresponding yeast mutants, it was found that it has a strong chelating ability to heavy metals Cu(II) and Cd(II), but has no replenishing ability to H2O2 stress. At the transcriptional level, the expression levels of LbMT2a, LbMT2b, LbMT3c, and LbMT4 in the symbiosis process of L. bicolor and plant roots were significantly higher than other LbMTs. Further bioinformatics analysis of the 1500 bp upstream region of these 8 genes revealed that the promoter regions of the 4 LbMTs genes have multiple transcription factor binding sites related to nutrient element response. In addition, the expression of single copy gene LbMT1 under the stress of Cu(II), Cd(II), Zn(II), Ag(I), and H2O2 were studied. LbMT1 transcript is relatively abundant in vegetative mycelium cells, and its expression is significantly increased under high concentration of Cu(II) or H2O2 stress and low concentration of Cd(II) stress.