Role of the HaHOG1 MAP Kinase in Response of the Conifer Root and But Rot Pathogen (Heterobasidion annosum) to Osmotic and Oxidative Stress

The basidiomycete Heterobasidion annosum (Fr.) Bref. s.l. is a filamentous white rot fungus, considered to be the most economically important pathogen of conifer trees. Despite the severity of the tree infection, very little is known about the molecular and biochemical aspects related to adaptation to abiotic stresses. In this study, the osmotic and oxidative tolerance as well as the role of the HaHOG1 Mitogen Activated Protein Kinase (MAPK) gene were investigated. The transcript levels of the yeast orthologues GPD1, HSP78, STL1, GRE2 and the ATPase pumps ENA1, PMR1, PMC1 known to have an important role in osmotolerance were also quantified under salt osmotic conditions. The HaHOG1 gene was used for a heterologous expression and functional study in the Saccharomyces cerevisiae Δhog1 strain. Moreover, the phosphorylation level of HaHog1p was studied under salt osmotic and oxidative stress. The result showed that H. annosum displayed a decreased growth when exposed to an increased concentration of osmotic and oxidative stressors. GPD1, HSP78, STL1 and GRE2 showed an induction already at 10 min after exposure to salt stress. Among the ATPase pumps studied, PMC1 was highly induced when the fungus was exposed to 0.2 M CaCl2 for 60 min. The heterologous expression of the HaHOG1 sequence in yeast confirmed that the gene is able to restore the osmotolerance and oxidative tolerance in the S. cerevisiae hog1Δ mutant strain. The HaHog1p was strongly phosphorylated in the presence of NaCl, KCl, H2O2 but not in the presence of CaCl2 and MgCl2. The GFP-HaHog1p fusion protein accumulated in the nuclei of the S. cerevisiae hog1Δ cells when exposed to high osmotic conditions but not under oxidative stress. These results provide the first insights about the response of H. annosum to osmotic and oxidative stress and elucidate the role of the HaHOG1 gene in such conditions.


Introduction
Living cells are able to respond as well as adapt to changes in the environmental conditions. A wide range of biotic and abiotic stresses often affect the fitness of a specific organism, by altering the metabolism and biology, thus compromising the development and the growth. Fungi are naturally exposed to a diverse range of environmental conditions which could differ depending on the fungal lifestyle and ecological niche. Among abiotic stresses, changes in the osmolarity and oxidative stress level can greatly affect the viability of the organism. Fungi possess several intracellular pathways that are able to respond to alterations in the osmolarity and oxidative conditions. Stress-activated protein kinases (SAPK) are mitogen activated protein kinases (MAPKs) which are phosphorylated in a variety of stress related conditions [1].
The MAPK core module is composed by three main MAPKs: a primary MAP kinase kinase kinase (MKKK), a MAP kinase kinase (MKK) and finally a MAP kinase (MAPK) [1]. MKKKs are activated either by MAPK kinase kinase kinase (MKKKK) mediated phosphorylation or by interacting with small GTPbinding proteins. When activated, MKKKs actively phosphorylate the second member MKK. Once activated, the MKK recognises and phosphorylates a conserved Thr-X-Tyr motif on the final MAPK effector. The MAPK can then specifically induce the expression of transcription factors or activate other protein kinases, phospholipases and cytoskeleton-associated proteins [1] [2]. Human and plant pathogenic fungi have evolved different ability to tolerate a wide range of osmotic and oxidative stressors. The HOG (High Osmolarity Glycerol response) is one of the most investigated stress related pathways which was first described in Saccharomyces cerevisiae [3].
Earlier studies have elucidated the two component mechanism of the HOG pathway in the budding yeast; a phosphorelay system which is composed of Sln1p-Ypd1p and Ssk1p and known to specifically activate a MAP kinase cascade composed by Ssk2/ 22p-Pbs2p and Hog1p [4]. When the osmotic pressure increases in the media, the Hog1p localizes in the nucleus and activates specific transcription factors to promote an initial response and adaptation [5]. In the budding yeast, oxidative stress caused by t-BOOH and H 2 O 2 can also stimulate the phosphorylation of Hog1p mediated by Pbs2 in a similar manner to what happens during osmolarity condition [6]. Human pathogens have developed specific mechanisms to defend themselves especially from high oxidative conditions which they encounter during the colonization process. The MAPK Hog1p of the human pathogen Candida albicans is actively phosphorylated when the fungus is exposed to oxidative (10 mM H 2 O 2 ) and osmolarity (1.4 M NaCl) stress conditions [7].The Cryptococcus neoformans serotype A and D hog1DA and hog1DD mutant strains show both high sensitivity to osmolarity condition (1 M and 1.5 M KCl). On the other hand, HOG1 ortholog gene has a different role in oxidative stress response in the two serotypes: while the hog1DA mutant is hypersensitive to 2 mM and 3 mM H 2 O 2 , the hog1DD is resistant [8]. In Aspergillus nidulans the sakA (HOG1 ortholog) MAPK is activated by phosphorylation in both osmotic (0.6 M KCl) and oxidative (0.3 mM H 2 O 2 ) stresses even if DsakA mutant strain does not show any reduced growth in hyperosmotic medium compared to the wild type [9]. The hyphal growth in the plant pathogen Fusarium proliferatum DFphog1 mutants is reduced when the fungus is exposed to different stressors like H 2 O 2 , NaCl and the non-ionic sorbitol in a concentration dependent manner [10].
In the rice pathogen Magnaporthe grisea, the Dosm1 mutant shows an impaired growth when exposed to hyperosmotic conditions (0.4 M NaCl, sorbitol or KCl) and fails to accumulate arabitol compared to the wild type, thus showing a correlation between OSM1 activity and accumulation of suitable polyols to protect the fungus from the osmotic shock [11]. In the citrus necrotrophic fungus Alternaria alternata, the AaHOG1 mutants show enhanced sensitivity to oxidants like 7.5 mM tert-butyl-hydroxyperoxide, 30 mM H 2 O 2 , and 2 mM menadione, and high ionic osmotic conditions, like 1 M KCl and NaCl salts [12]. A copy of the HOG1 homologue gene is present in the Dead Sea fungus Eurotium herbariorum. The EhHOG1 gene from E. herbariorum is able to restore the normal osmotolerant phenotype when introduced in S. cerevisiae Dhog1 mutant strain [13]. Moreover, complemented S. cerevisiae Dhog1 strain with EhHOG1 sequence shows a higher growth capacity in Li + supplemented media compared to the S. cerevisiae wild type [13]. This result suggests that EhHOG1 gene is probably part of those specific mechanisms that allow E. herbariorum to adapt to high Li+ content of the Dead Sea water [13].
Mutants with impaired growth in high osmolarity condition have been described in the model ascomycete fungus Neurospora crassa [14]. The os-2 mutant strain is sensitive to the fungicide fludioxonil and the growth is inhibited in media supplemented with 4% NaCl. The os-2 gene encodes a S. cerevisiae Hog1p homologue MAPK that confers resistance to hyperosmotic conditions and sensitivity to phenylpyrrole fungicides [14]. HOG1 homologue gene has also been studied in the endophyte fungus Epichloë festucae [15]. The E. festucae sakA cDNA when introduced into the Schizosaccharomyces pombe sty1 mutant strain is able to restore the stress sensitive defect when exposed to osmotic (KCl, sorbitol, NaCl) and oxidative (H 2 O 2 ) stressors [15]. Like N. crassa os-2 mutant strain, the E. festucae DsakA shows enhanced sensitivity to osmotic stress and resistance to the fungicide fludioxonil, but not increased sensitivity to oxidative stress [15].
Despite extensive research on signal transduction cascades and their role in fungal growth and pathogenicity, almost nothing has been done in this respect on the conifer pathogen H. annosum. H. annosum (Fr.) Bref. sensu lato species complex is the causative agent of root and butt rot disease and it is the most important pathogen of conifer trees in the northern hemisphere [16] [17]. Three intersterile species have been described in Europe, the P, S and Ftypes named according to their host species preferences (pine, spruce and fir respectively). The European P-type Heterobasidion annosum (Fr.) Bref. prefers trees of the genus Pinus as its host, but also infects many other conifer and broadleaved tree species. The European S-type Heterobasidion parviporum Niemelä & Korhonen, infects mainly Norway spruce (Picea abies (L.) Karsten) and seldom other tree species. Based on studies on the extent of decay caused by H. parviporum and H. annosum in 60 year old Norway spruce stems, H. parviporum shows higher specialization for spruce as a host [18]. The European F-type Heterobasidion abietinum Niemelä & Korhonen infects species of the genus Abies [16] [19]. The economical losses caused by Heterobasidion species in Europe are estimated to be around 800 million euro annually [17]. In Finland, H. parviporum and H. annosum cause severe damages in stands of Norway spruce (P. abies) and Scots pine (Pinus sylvestris L.).
In this study, the osmotolerance and oxidative tolerance of the tree pathogen H. annosum P-strain were characterized. We also described the role of the HaHOG1 MAP kinase, the yeast HOG1 homologues MAPK which has been demonstrated to be involved in osmotolerance and oxidative tolerance in many other fungi. This study also confirms the possibility to use a heterologous system to study the function of H. annosum genes thus overcoming the limitations imposed by the lack of an efficient DNAtransformation system.

Fungal strains and growth conditions
H. annosum P-type (isolate 03012 kindly provided by Kari Korhonen, METLA Finnish Forest Research Institute, Finland) was maintained in MEG agar plates (0.5% Malt extract, 0.5%Glucose and 2% agar) and grown at room temperature in laboratory conditions. Fungal culture experiments were all performed at room temperature in MEG agar plates or MEG liquid media. Transformed Escherichia coli for plasmid replication was grown in LB plates or LB liquid media at 37uC. The S. cerevisiae strains used in this study were the wild type BY4742 (Euroscarf acc. num. Y10000: MATa; his3D1; leu2D0; lys2D0; ura3D0) and the Dhog1 mutant YLR113w (Euroscarf acc. num. Y12724: BY4742:MATa; his3D1; leu2D0; lys2D0; ura3D0; YLR113w::kanMX4). Both strains were maintained in yeast extract-peptone-dextrose (YPD) 2% agar plates at 30uC or grown in YPD liquid media under shaking at 28uC. YLR113w strain carrying the pYES2 plasmid was selected and maintained in synthetic defined (SD)-URA 2 selective media.

Gene cloning
The H. annosum P-type HOG1 gene homolog was retrieved from JGI Heterobasidion genome browser (http://www.jgi.doe.gov/) with a BLASTp search using the S. cerevisiae Hog1p protein sequence from Saccharomycete Genome Database (SGD, http://www.yeastgenome.org/) as a query. Full-length gene sequence was PCR amplified from H. annosum P-type cDNA using specific primers (Table 1) and the following PCR program: 95uC 3 min, 30 cycles (95uC 1 min-57uC 30 sec-72uC 1 min), 72uC 10 min. HaHOG1 full-length cDNA was purified and cloned into pGEMT-easy vector (Promega, Finland) according to manufacturer's instruction. The gene was sequenced and the sequence was submitted to GenBank database (accession number JN127357). HaHOG1 cDNA was cloned into pYES2 vector (Invitrogen) by single restriction reaction with EcoRI (New England Biolabs) creating pYES2-HaHOG1 construct. Correct orientation was checked by restriction analysis using XhoI (New England Biolabs).

Effect of salts and H 2 O 2 on H. annosum growth
To study the effect of different osmotic and oxidative conditions on H. annosum, the fungal growth was quantified on MEG 2% agar plates supplemented with either calcium chloride (CaCl 2 ), potassium chloride (KCl), magnesium chloride (MgCl 2 ), or sodium chloride (NaCl) at concentration ranging from 0.05 M to 0.5 M. For NaCl salt, 1 M concentration was also tested. For the oxidative conditions, hydrogen peroxide (H 2 O 2 ) was added at final concentration ranging from 1 mM to 5 mM. The plates were sealed with parafilm and incubated at room temperature. The growth of the fungus was measured at regular intervals, and monitored for approximately 3 weeks post inoculation. Four replicates for each different concentrations were prepared and the colony radius was measured from three different directions from the center in each plate. The plates for the control contained MEG 2% agar without salt or oxidative stressors.

S. cerevisiae transformation
S. cerevisiae Dhog1 mutant strain (YLR113W) was transformed by electroporation either with the empty pYES2 vector (Dhog1+-pYES2) or with the pYES2-HaHOG1 vector (Dhog1+pYES2-HaHOG1). Briefly: YLR113W mutant strain was grown in YPD media at 28uC until OD 600 = 3. The cultures were centrifuged at 3000 rcf at 4uC for 5 min then washed twice with sterile MilliQ water and resuspended in 5 ml 1 M sorbitol and kept on ice. A further centrifugation at 2000 rcf at 4uC for 5 min was performed and finally cells were resuspended in 150 ml of 1 M sorbitol. Cell suspension (40 ml) of the S. cerevisiae were transferred in 0.2 cm path cuvette (Bio-Rad) and electroporated (BIO-RAD Gene-Pulser XCell, V = 1.5 KV, 25 mF, 200 Ohm). Electroporated cells were immediately resuspended in 1 ml of 1 M sorbitol and 200 ml were plated in SD-URA 2 selective plates for colony selection. Plates were incubated at 30uC until colonies appeared.

S. cerevisiae complementation experiment
The S. cerevisiae wild type strain was grown in 5 ml YPD media while the Dhog1+pYES2 and the Dhog1+pYES2-HaHOG1 mutant strain were grown in 5 ml SD-URA 2 liquid media at 28uC over night. For the osmotolerance experiment, YPD 2% agar plates supplemented either with 2% glucose or 2% galactose+1% raffinose were prepared for each condition to be tested: 0.5 M and 1 M of NaCl, KCl, MgCl 2 , and CaCl 2 . For the oxidative tolerance experiment, YPD 2% agar plates supplemented with 2% galactose+1% raffinose were used with either 3 mM, 4 mM or 5 mM H 2 O 2 . The wild type, Dhog1, Dhog1+pYES2 and Dhog1+-pYES2-HaHOG1 strains were quantified with a hemocytometer and a dilution series (10 5 , 10 4 , 10 3 , 10 2 cells) were spotted in a row for each strain. The plates were incubated at 30uC for 4-10 days to allow comparison between the wild type and the mutant strains.

RNA extraction and cDNA synthesis
Total RNA was extracted from H. annosum P-type with a modified CTAB protocol [20]. Briefly, the mycelium was filtered from the culture using Miracloth (ChalBiochem), wrapped in aluminum foil, and immediately frozen in liquid nitrogen. For each sample, 3 ml CTAB extraction solution (2% (w/v), 100 mM Tris-Cl pH 8, 20 mM EDTA pH 8, 1,4 M NaCl, 2% (v/v) 2mercaptoethanol) was added to the pulverized mycelium and the mixture was incubated for 30 min at 65uC. All samples were extracted twice with an equal volume of chloroform-isoamyl alcohol (24:1) and centrifuged at 10000 rcf for 10 min at 20uC. Selective RNA precipitation was performed by adding 1/4 volume of 10 M LiCl. After overnight incubation at 4uC, samples were centrifuged for 30 min and the pellet was resuspended in 500 ml SSTE (1 M NaCl, 0.5% SDS, 10 mM Tris-Cl, 1 mM EDTA, prewarmed at 65uC). A further extraction with an equal volume of chloroform-isoamyl alcohol (24:1) was performed as described above. The upper phases were recovered, and total RNA were precipitated with two volumes of absolute ethanol overnight at 220uC. The samples were then centrifuged at 10000 rcf for 15 min at 20uC. The total RNA pellet was washed with 80% ethanol, left to dry for 30 min and then resuspended in 50 ml nuclease free water. RNA samples were stored at 280uC. Total RNA was retrotranscribed as follows: 1 mg total RNA was treated with DNase (Promega), incubated at 37uC for 30 min and final DNase inactivation at 65uC for 10 min. Random primers (0.1 mg, Fermentas) were added to the mixture and the reaction was incubated at 65uC for 5 min followed by a quick cool down on ice. The retrotranscription was performed with RevertAid Reverse Transcriptase (200 U, Fermentas) according to the manufacturer's instruction. The cDNA was then diluted 20 or 40 times to be used for the quantitative real-time PCR (qPCR) or used without dilution for semi-quantitative PCR.
HaHOG1 phosphorylation experiment, total protein extraction and western blot H. annosum was grown in liquid MEG media for 4 weeks at room temperature. The salt osmotic stress was induced by adding either NaCl, KCl, MgCl 2 or CaCl 2 at 0.5 M final concentration to the liquid fungal culture. The oxidative stress was induced by adding H 2 O 2 at 5 mM final concentration. After each stress induction, Table 1. List of primers used in this study.

Gene
Forward Primer Reverse Primer

Gene expression analysis by Quantitative real-time PCR (qPCR)
The GPD1, HSP78, STL1, GRE2, ENA1, PMR1, PMC1 and GAPDH gene models in H. annosum were found by BLASTp search in the JGI Heterobasidion genome browser using the S. cerevisiae genes from the Saccharomyces Genome Database (SGD, http:// www.yeastgenome.org/) as query. Internal primers for qPCR analysis were designed using the Universal ProbeLibrary Assay Design Center (Roche, http://www.roche.com). The primers used in this study are shown in Table 1. LightCycler 480 SYBR Green I Master (Roche) was used with 5.5 ml of the diluted cDNA sample in a 15 ml total reaction volume. The following cycles were used in the LightCyclerH 480 Instrument II (384 wells plates, Roche): preincubation at 95uC for 5 min, denaturation 94uC for 10 sec (4.8uC/s), annealing at 59uC for 10 sec (2.5uC/s), extension at 72uC for 10 sec (4.8uC/s), 40 cycles of amplification and final extension at 72uC for 3 min. The Ct values were automatically calculated using the LIGHTCYCLER 480 software, the transcript levels were normalized against GAPDH expression and the fold change was calculated based on the control with the Pfaffl method [21].

Gene expression analysis by Semi-Quantitative PCR
The ENA1, PMR1 and PMC1 gene expression was visualized on ethidium bromide agarose gel by semi-quantitative PCR. The three pumps and the internal reference GAPDH were amplified using the same primers listed in Table 1. The PCR reaction mixture was set as follows: 1 ml cDNA (undiluted, see above), 2.5 ml DreamTaq TM Green Buffer (Fermentas), 0.5 ml dNTPs (10 mM each, Fermentas), 1 ml forward primer, 1 ml reverse primer, 0.65 ml DreamTaq TM Green DNA Polymerase (Fermentas) and water to 25 ml total volume. For ENA1, PMR1 and PMC1 the PCR cycle was as follows: 95uC for 3 min, denaturation 95uC for 30 sec, annealing 59uC for 30 sec, extension 72uC for 30 sec, 26 cycles of amplification and final extension at 72uC for 10 min. For the internal reference gene GAPDH the same conditions were used but with only 20 amplification cycles due to the higher initial amount of transcript. To visualize the transcript level, 10 ml of the reaction mixture were then loaded on ethidium bromide agarose gel and photographed under UV light.

GFP-HaHog1p fusion protein
The HaHOG1 gene was fused to the GFP for subsequent subcellular localization. HaHOG1 was amplified from pYES2-HaHOG1 vector using specific primers to include the NotI restriction site at the 39 end of the gene ( Table 1). The GFP was amplified using specific primers to include an overlapping region with the HaHOG1 at the 39 end and the HindIII restriction site at the 59 end of the reporter gene ( Table 1). The two genes were fused together by a ''fusion PCR'' using PhusionH High-Fidelity DNA Polymerase (Finnzymes) and the following PCR program: pre-incubation at 98uC for 30 sec, denaturation 98uC for 10 sec, annealing at 58uC for 20 sec, extension at 72uC for 1 min, 30 cycles of amplification and final extension at 72uC for 10 min.
The GFP-HaHOG1fusion fragment was purified from gel and double digested at both ends using HindIII and NotI restriction enzymes in a single reaction according to the manufacturer's instruction. The fragment was then ligated into the pYES2 vector restricted with the same two enzymes generating the pYES2-GFP-HaHOG1 construct. The construct was transformed into the S. cerevisiae YLR113W Dhog1 mutant strain using the same procedures as for the yeast complementation experiment (see above). The yeast cells were photographed using a fluorescent microscope (Leitz, Laborlux S) equipped with a digital camera (Olimpus, DP50-CU).

Growth of H. annosum under salt osmotic and oxidative stress conditions
H. annosum displayed a decreased growth when exposed to osmolarity stress condition (Figure 1). The fungus can grow in media supplemented with NaCl, KCl, MgCl 2 and CaCl 2 with a concentration ranging from 0.1 M to 0.5 M. At 1 M concentration in all the tested salts, no growth was observed on the plates. The divalent salts (magnesium and calcium chloride) had a stronger inhibitory effect on the fungal growth compared to the monovalent salts (sodium and potassium chloride).
H. annosum tolerated a concentration of hydrogen peroxide ranging from 1 mM to 5 mM ( Figure 2). The fungal growth capacity decreased as the amount of H 2 O 2 increased in the media. At 1 mM of hydrogen peroxide, the fungal growth was slightly inhibited compared to the control where the plate was covered by the fungal mycelium after 9 day post inoculation. At 3 mM H 2 O 2 the fungal mycelium grew much slower and the petri-plate was over grown with hyphae at 15 days post inoculation (i.e. 6 days later than the control). At the highest concentration of 5 mM of hydrogen peroxide a much stronger inhibition was observed but the fungus was able to recover gradually with slow growth which started 12 days after initial inoculation.

Gene expression under salt stress conditions
The transcriptional regulation of 4 genes (GPD1, HSP78, STL1 and GRE2) known to be associated with osmotic stress was investigated. The four genes were found up-regulated at 10 min after salt addition (NaCl, KCl, MgCl 2 and CaCl 2 ). The upregulation was stronger for HSP78 for which the level of the transcript was up to 3-fold induced compared to the control (Figure 3). GPD1 showed a moderated up-regulation at 10 min after osmotic stress induction in all the conditions with an average induction of 2-fold compared to the control. The transcript level for STL1 and GRE2 was up-regulated compared to the control but not as strong as the other genes studied. The transcript level of three ABC transporters named ATPase ENA1, PMR1 and PMC1 were also quantified under different salt osmotic conditions by qPCR. When H. annosum was exposed to NaCl, KCl and MgCl 2 the three pumps were expressed but no differences in the expression level was detected over time (data not shown). On the other hand, in the presence of the divalent salt CaCl 2 neither ENA1 nor PMR1 showed any induction compared to the control although they were actively expressed to a certain level (Figure 4 A). However, the ATPase pump PMC1 showed an increased expression at 60 min after the addition of CaCl 2 to the H. annosum culture (Figure 4 A). The results were confirmed by the semiquantitative PCR where an induction of PMC1 over time in CaCl 2 was also detected (Figure 4 B). In the semi-quantitative PCR gel picture, the pump transcript was almost absent after 10 min post calcium chloride addition but its transcript level increased after 30 min and 60 min (Figure 4 B). The GAPDH gene displayed a stable transcript expression thus providing a good reference gene for expression studies under salt conditions (Figure 4 B).

S. cerevisiae complementation with HaHOG1 under different salt conditions
A complementation experiment in S. cerevisiae Dhog1 mutant strain was performed with the H. annosum HaHOG1 sequence. The result showed that the H. annosum HaHOG1 gene is able to restore the capacity of the S. cerevisiae Dhog1 mutant strain to grow in high osmotic conditions ( Figure 5). All the yeast strains were able to grow equally in standard YPD media supplemented with glucose (Control, Figure 5). The S. cerevisiae wild type strain (wt, Figure 5) is able to grow in media supplemented with 0.5 M and 1 M of sodium chloride (NaCl, Figure 5). In the same conditions, the Dhog1 mutants (Dhog1 and Dhog1+pYES2, Figure 5) show a decreased growth, especially at the higher concentration of salts (1 M). However, the yeast Dhog1 mutant strain carrying the pYES2-HaHOG1 vector (Dhog1+pYES2-HaHOG1, Figure 5) displayed the same osmotolerance compared to the wild type. The capacity to restore the osmotolerance is more evident when the expression of the HaHOG1 gene in the pYES2 vector was induced in the presence of galactose in the media (+GAL, Figure 5). A partial complementation can be seen even under glucose repression condition when the yeast was exposed to 1 M of NaCl (+GLU, Figure 5). Probably a basal level of HaHOG1 transcript was still present under the repression condition and this was enough to provide partial complementation in the mutant yeast. All the S. cerevisiae strains exposed to 0.5 M and 1 M of potassium chloride (KCl, Figure 5) displayed the same phenotype and osmotolerance as observed for the sodium chloride.
On the other hand, a stronger growth inhibition was observed in the presence of the divalent salts magnesium chloride (MgCl 2 , Figure 5) and calcium chloride (CaCl 2 , Figure 5) at the concentration of 0.5 M and 1 M. The wild type strain was able to grow at 0.5 M MgCl 2 both in glucose and galactose media but it showed a remarkable reduced growth at higher concentration of salt (1 M) particularly in galactose condition. This result can be explained by the fact that galactose represents a poorer utilizable sugar compared to glucose. Furthermore, there were evidences related to the beneficial effect of the glucose regarding the osmotolerance of fungi (C. albicans) when exposed to a hyperosmotic condition [22]. The osmotolerance of Dhog1+pYES2-HaHOG1 strain at 0.5 M MgCl 2 is comparable to the wild type only when the HaHOG1 gene was induced by the galactose (+GAL, Figure 5). The strongest toxicity effect among the salts is caused by CaCl 2 . Under calcium chloride osmotic condition, the only growth observed was related to the wild type strain at 0.5 M CaCl 2 in the presence of glucose (+GLU, Figure 5).

S. cerevisiae complementation with HaHOG1 in oxidative conditions
The yeast strains were grown in YPD plate where the concentration of hydrogen peroxide ranged from 0.25 mM to 20 mM. A wide range of concentration was used to determine the sensitivity of S. cerevisiae strains to the oxidative stress. The results show that all the yeast strains (wt, Dhog1, Dhog1+pYES2 and Dhog1+pYES2-HaHOG1, Figure 6) tolerated concentration of hydrogen peroxide up to 3 mM with no visible effect on the yeast cell survival. At the concentration of 4 mM H 2 O 2 a remarkable difference in the yeast cell viability was seen between the mutant strains (Dhog1, Dhog1+pYES2, Figure 6) and the complemented strain (Dhog1+pYES2-HaHOG1, Figure 6). Interestingly, the complemented strain shows a better fitness compared to the wild type strain when exposed to 0.4 mM of hydrogen peroxide. At 5 mM H 2 O 2 the yeast cell viability was considerably decreased for all the yeast strains but still a better growth could be seen related to the complemented strain (Dhog1+pYES2-HaHOG1, Figure 6) compared to the mutant strains (Dhog1, Dhog1+pYES2, Figure 6).

HaHog1 phosphorylation level
The monoclonal antibody used in this study was able to reveal a sharp and specific band close to 46 KDa (Figure 7). The same antibody has been used in other studies and it showed crossreaction among a broad range of species. Based on the literature data and the molecular weight which is close to the expected size (42.4 KDa, http://web.expasy.org/compute_pi/), we concluded that the band corresponds to the phosphorylated form of the H. annosum HaHog1p (phospho-HaHog1p). The western blot results showed an increased level of phospho-HaHog1p at 1 min and at 3 min after NaCl and KCl salt addition respectively compared to the control (Figure 7). The amount of phospho-HaHog1p increased gradually over time to reach the highest level between 10 min to 30 min in both salt conditions. However, in the presence of the divalent salts (CaCl 2 and MgCl 2 ) the amount of phospho-HaHog1p was considerably lower compared to the monovalent salts (NaCl and KCl) treatment ( Figure 7). Indeed, a weak signal related to the phospho-HaHog1p appeared later at 30 min upon calcium salt addition and at 60 min upon magnesium salt addition.
In the case of H 2 O 2 treatment, the phospho-HaHog1p signal was detected already after 1 min. The signal increased over time to reach a maximum at 3 min followed by a gradual decrease of signal intensity. However, a higher amount of phospho-HaHog1p compared to the control was still present at 60 min after hydrogen peroxide exposure (Figure 7).

Subcellular localization of the GFP-HaHog1p in yeast
To study the subcellular localization of the HaHog1p protein in yeast, a GFP-HaHog1p fusion protein was generated. The yeast cells expressing GFP were observed in unstressed and stressed conditions under fluorescent microscope. The result revealed that the GFP fused to the N-terminal of the protein was still functional and the fusion protein localized in the cytoplasm in the unstressed yeast cells without any evident accumulation in any subcellular compartments (Control, Figure 8). The GFP in the N-terminal position did not alter the function of the HaHOG1 gene since it can still restore the osmotolerance in high salt conditions when the construct was transformed into the S. cerevisiae Dhog1 mutant strain (data not shown). When the yeast cells were exposed to 0.2 M of the different salts, the GFP-HaHog1p accumulated in the nucleus of the stressed cell within the first 30 min. after the addition of the salt (NaCl, KCl, MgCl 2 and CaCl 2 , Figure 8). The nuclear accumulation was confirmed by the co-localization with the DAPI staining (data not shown). Interestingly, no clear nuclear accumulation could be seen in the yeast cells exposed to 5 mM of H 2 O 2 after 30 min following the addition of the oxidative stressor ( Figure 8).

Discussion
In this study, we investigated the response and the putative role of the MAPK HaHOG1 in the basidiomycete H. annosum and its capacity to overcome osmolarity and oxidative stress conditions. A heterologous system with S. cerevisiae as a model host organism was used to study the function of the HaHOG1 gene. The osmotolerance of H. annosum to osmotic stressors was investigated under different salt conditions. The results revealed that H. annosum was able to tolerate salt concentration less than 0.5 M. Tolerance capacity to higher salt concentration have been reported in other fungal species. C. albicans, C. glabrata and Debaryomyces hansenii show a higher tolerance level when exposed to NaCl osmotic stress compared to H. annosum [23]. In the plant pathogen Botrytis cinerea the wild type strain was able to grow in media supplemented with 1.5 M NaCl [24]. In another study, the wild type strain of the causative agent of Southern corn leaf blight, Cochliobolus heterostrophus, was able to tolerate concentrations of KCl up to 0.75 M [25]. In our study, monovalent salts (NaCl and KCl) inhibited the growth of H. annosum to a less extent compared to the divalent salts (CaCl 2 and MgCl 2 ). This result could be partly explained by the stronger osmotic pressure generated by divalent salts compared to the same concentration of the monovalent salts. On the other hand, Mg 2+ and Ca 2+ ions are known to have important intracellular roles in eukaryotes and are involved in critical intracellular function. In particular, Ca 2+ ions act as a secondary messenger in pivotal intracellular pathways involved in Figure 3. Expression of the genes GPD1, HSP78, STL1 and GRE2 in Heterobasidion annosum exposed to high concentration of different salts. The transcript levels of four putative H. annosum genes involved in the HaHOG1 osmotic pathway were quantified in the mycelium exposed to 0.5 M of either NaCl, KCl, MgCl 2 or CaCl 2 in liquid culture. RNA was extracted from the fungal mycelium, cDNA was synthesized and qPCR was performed. Fold change variation of the genes compared to the control was calculated using Pffafl method (GAPDH as internal reference was used). Three biological replicates were used for each treatment. Bars represent standard deviation. doi:10.1371/journal.pone.0031186.g003 cellular metabolism, growth and intracellular signal transduction [26]. In Neurospora crassa for instance, the role of calcium in the hyphal tip elongation has been described. Tip-high Ca 2+ gradient and the role of actin cytoskeleton are pivotal for the site and direction of apical growth [27]. Very little studies have been carried out about the role of Mg 2+ in fungal growth and development. Recently, in the wheat pathogen F. graminearum, it was shown that magnesium ions at low concentration (2 mM Mg 2+ ) could inhibit almost completely the trichothecenes biosynthesis, an important class of mycotoxins [28]. Thus, high Ca 2+ and Mg 2+ ions in the media have the ability to affect the fungal growth not just by altering the cellular turgor but even by affecting intracellular pathways which are important for the fungal metabolism and biology. In a recent study, the levels of Na + , K + , Ca 2+ and Mg 2+ were quantified in the sapwood, heartwood, reaction zone and decay zone in Norway spruce infected by H.
annosum [29]. It was shown that the above mentioned ions accumulated in the reaction zone (i.e. the area in the living wood with an active defense against the pathogen) as well as in the decay zone. In particular, K + was detected at the highest amount in the decay area (4844 mg kg 21 ), followed by Ca 2+ (1952 mg kg 21 ), Mg 2+ (491 mg kg 21 ) and finally Na + (20 mg kg 21 ) [29]. These data indicate that H. annosum encounters a high saline environment during its colonization process which suggests that it may have developed intracellular responses to be able to adapt in those unfavourable conditions.
No studies have been carried out about the capacity of H. annosum to tolerate high level of salts in the media. Consequently we investigated the expression of several putative genes which could be involved in osmotic stress tolerance according to literature data [30]. The NAD-dependent glycerol-3-phosphate dehydrogenase GPD1 was shown to be an important induced gene Figure 4. Expression of the ATPase pumps ENA1, PMR1, and PMC1 in Heterobasidion annosum exposed to high concentration of calcium chloride. The transcript levels of three putative H. annosum ATPase pumps were quantified in the mycelium exposed to 0.2 M CaCl 2 for 10, 30, 60 min in liquid culture. (A) RNA was extracted from the fungal mycelium, cDNA was synthesized, and qPCR was performed. Fold change variation of the genes compared to the control was calculated using Pffafl method (GAPDH as internal reference was used). Three biological replicates were used for each time point. Bars represent standard deviation. (B) A representative semi-quantitative PCR was performed on the three pumps using the same cDNA as for the qPCR. 10 ml of the PCR reaction mixture were equally loaded on ethidium bromide gel and photographed under UV light. Stable expression of the internal reference GAPDH transcript is shown. doi:10.1371/journal.pone.0031186.g004 during osmotic stress response [31].The GPD1 induction in H. annosum after 10 min post salt addition was found to be close to 2fold compared to the control. Our results correlates with the GPD1 induction as it was described in S. cerevisiae 15 min after salt osmotic stress [32]. The HSP78 gene encodes a mitochondrial heat shock protein and the transcript was found up-regulated in heatshocked S. cerevisiae cells when grown in a non-fermentable carbon source [33]. It was found to be up regulated also in saline stress [30] and high hydrostatic pressure [34]. In H. annosum this is the first evidence of HSP78 induction under salt stress with a transcript level as high as 3-fold compared to the level in the non-stressed mycelia. The transcript levels of two more genes, STL1 and GRE2, were quantified. STL1 encodes for a glycerol/H + symporter in S. cerevisiae as demonstrated in a previous study [35]. In C. albicans the STL1 expression was induced when the cells were exposed to osmotic shock (1 M NaCl) but a basal level of mRNA was still detected in the presence of either glucose or glycerol in minimal media (without stress) [36]. The major difference in the STL1 expression from the non-pathogenic S. cerevisiae and the pathogenic C. albicans is that in the pathogenic fungus STL1 is constitutively expressed in the presence of fermentable carbon source too [36]. In our study, the constitutive STL1 expression in the presence of glucose could be the reason why we did not observe a strong induction of the putative H. annosum STL1 transcript since the fungus was grown in media supplemented with glucose. Another reason could be that 10 min after salt addition is still too early to Figure 5. Complementation experiment using the Heterobasidion annosum HaHOG1 gene in the S. cerevisiae Dhog1 mutant strain under salt osmotic conditions. The H. annosum HaHOG1 gene was cloned into the pYES2 vector under the control of the galactose inducible GAL1 promoter and expressed in the osmosensitive S. cerevisiae Dhog1 mutant. Wild type and osmosensitive mutant were grown on YPD liquid media while the plasmid carrying yeasts were grown on selective SD-URA 2 liquid media at 28uC. Four different salts were tested (NaCl, KCl, MgCl 2 and CaCl 2 ) with two different concentration (0.5 M and 1 M). For each salt two carbon sources were also used: either glucose (+Glu) to repress or galactose (+Gal) to induce the HaHOG1 expression. Standard YPD media was used as control (Control). The four yeast strains used were as follow: wild type (wt), osmosensitive yeast strain (Dhog1), osmosensitive yeast strain carrying the empty pYES2 plasmid (Dhog1+pYES2) and osmosensitive yeast strain carrying the pYES2 plasmid with HaHOG1 gene under the GAL1 promoter (Dhog1+pYES2-HaHOG1). Cells from each strain were quantified with a hemocytometer, 10-fold diluted suspensions were prepare (10 6 , 10 5 , 10 4 , 10 3 cells/ml) and 10 ml were spotted on the different plates. The plates were incubated at 30uC for 4-10 days to allow comparison between the wild type and the mutant strains. doi:10.1371/journal.pone.0031186.g005 see a strong effect on STL1 expression level caused by the osmotic stress. We also decided to quantify the expression level of the homologue of the S. cerevisiae methylglyoxal reductase GRE2 [37]. As for STL1, we did not observe a strong GRE2 transcript induction at 10 min post salt addition. In a previous microarray study, GRE2 was shown to be induced in S. cerevisiae stressed cells at 10 min after osmotic shock imposition [38]. The discrepancy between literature data and our results could be explained by the different approaches used (microarray instead of quantitative PCR) or by the different organism studied.
The transcript levels of three different cellular channels (ENA1, PMR1 and PMC1 yeast orthologues in H. annosum) was also quantified by qPCR after exposure of the fungal cells to the different salts. No significant differences in the transcript level were found for ENA1 and PMR1 in the presence of NaCl, KCl and MgCl 2 (data not shown). The yeast ENA1 is an ATPase pump which is responsible for Na + /K + efflux to keep the intracellular iron concentration at low level [39]. This type of pump is induced in the presence of high Na + or K + in the media in alkaline condition [40]. H. annosum was grown in acidified media (pH 5) and in low pH values it has been shown that other type of channels (electroneutral Na + /H + and K + /H + antiporters) are probably responsible for Na + /K + cytosol depletion [40]. No induction at the transcript level was observed for the PMR1 transcript either. The yeast PMR1 is a Ca 2+ ATPase pump responsible for the import of Ca 2+ into the Golgi compartment and for the proper functioning of the secretory pathway [41]. Another pump, the yeast PMC1 which is a vacuolar Ca 2+ ATPase, is also responsible to keep the intracellular Ca 2+ at a physiological state. Strong induction related to PMC1 has been shown when the PMR1 is not functional [42]. We observed a strong induction of PMC1 homolog in H. annosum when exposed to 0.2 M CaCl 2 for 60 min and this result provides the first evidence of the potential role of the PMC1 pump in calcium homeostasis in this fungus. The level of PMC1 transcript was very low in H. annosum in non-stress conditions and it was induced most probably to increase the sequestration in the vacuolar compartment of calcium thus keeping the intracellular concentration at an acceptable level (typically 0.1 mM). It should be emphasized that all the genes described in this study have not previously been functionally investigated. The results showed in this section were the first evidence about their possible role in the adaptation of H. annosum to osmotic stress. In future studies, the possibility to generate H. annosum knock-out mutants for the above Figure 6. Complementation experiment using the Heterobasidion annosum HaHOG1 gene expressed in the S. cerevisiae Dhog1 mutant strain under oxidative conditions. The H. annosum HaHOG1 gene was cloned into the pYES2 vector controlled by the galactose inducible GAL1 promoter and expressed in the osmosensitive yeast Dhog1 mutant. Wild type and osmosensitive mutant were grown on YPD liquid media while the plasmid carrying yeasts were grown on selective SD-URA 2 liquid media at 28uC. Different concentrations of hydrogen peroxide (H 2 O 2 ) were used (3 mM, 4 mM and 5 mM). Galactose was used as carbon source to induce HaHOG1 gene expression on the pYES2 vector. Standard YPD media was used as control (YPD). The four yeast strains used were as follow: wild type (wt), osmosensitive yeast strain (Dhog1), osmosensitive yeast strain carrying the empty pYES2 plasmid (Dhog1+pYES2) and osmosensitive yeast strain carrying the pYES2 plasmid with HaHOG1 gene under the GAL1 promoter control (Dhog1+pYES2-HaHOG1). Cells from each each strain were quantified with an hemocytometer, 10-fold diluted suspensions were prepare (10 6 , 10 5 , 10 4 , 10 3 cells/ml) and 10 ml were spotted on the different plates. The plates were incubated at 30uC for 4-10 days to allow comparison between the wild type and the mutant strains. doi:10.1371/journal.pone.0031186.g006 Figure 7. Phosphorylation level of the Heterobasidion annosum HaHog1p exposed to high concentration of different salts at different time points. The H. annosum mycelium was exposed to 0.5 M of NaCl, KCl, MgCl 2 and CaCl 2 in liquid culture and the total protein were extracted at 1, 3, 10, 30 and 60 min after the salt addition. 7 mg of total proteins were loaded on 10% SDS polyacrylamide gel for protein separation. The separated total proteins were transferred to a nitrocellulose membrane and the anti-phospho-p38 monoclonal antibody was used to detect the phosphorylated form of the HaHog1p. doi:10.1371/journal.pone.0031186.g007 mentioned genes will probably give more information about their precise role in the osmostress tolerance.
The sensitivity of H. annosum to oxidative stress was tested using hydrogen peroxide in the culture media. The fungus shows a decreased growth when the peroxide concentration increased with the highest inhibition at 5 mM. Similar to the osmotic stress, the sensitivity to oxidative stress varied between different fungi. B. cinerea can tolerate higher concentration of H 2 O 2 up to 10 mM [24]. The plant pathogen C. heterostrophus can tolerate higher oxidative stress level with a range up to 20 mM of H 2 O 2 . The fungal lifestyle and its ecological niche could also have a profound effect on their response and tolerance to oxidative stress. The necrotrophic fungi B. cinerea and C. heterostrophus are actively and continuously exposed to the plant immune responses during the infection process. H. annosum is equally a necrotroph capable of killing living conifer tissues of all ages. It is also able to survive and adapt in heartwood tissues that contain many other toxic compounds. It would be of particular interest to investigate the ability of H. annosum to grow in the presence of toxic phenolic compounds which are abundant in the heartwood tissues. The differences in the resistance and susceptibility of H. annosum when compared to other fungal species would also provide some insights about the different oxidative tolerance ability based on host and infection process.
The activation of the H. annosum HaHog1p in osmotic and oxidative conditions was assessed by western blot using a monoclonal antibody able to recognize the phosphorylated form of the MAPK (phospho-HaHog1p). In the presence of the monovalent salts (NaCl and KCl), we observed the most strong and rapid HaHog1p activation within the first 60 min. A similar activation and kinetic pattern was also detected in other fungal species exposed to NaCl for the same period of time (0 to 60 min) [8] [43]. In this study, we showed that KCl, a less toxic salt compared to NaCl, can also activate the HaHOG1 pathway by phosphorylation of the HaHog1p MAPK. However, the sodium salt induced phosphorylation earlier at 1 min compared to the potassium salt for which the activation was induced at 3 min after stressor addition. A possible explanation could be the additive presence of the toxicity effect exerted by the Na + ions compared to the K + ions. Interestingly, both divalent salts, CaCl 2 and MgCl 2 at 0.5 M, were able to inhibit the HaHog1 phosphorylation. A very weak signal was recorded at 30 min for CaCl 2 and at 60 min for MgCl 2 after salt addition. In a recent study, the Hog1p activation was detected in S. cerevisiae exposed to 300 mM of CaCl 2 [44] and the authors proposed a model for the activation of the HOG1 pathway by extracellular Ca 2+ ions. In our study, we used a higher concentration of CaCl 2 (0.5 M): such a high concentration of Ca 2+ ions could have altered dramatically the cell physiology thus compromising the fungal stress adaptation pathways by inhibiting or delaying the HaHog1p phosphorylation. On the other hand, the impact of MgCl 2 salt on the osmotic balance and on the physiology on the fungal cell has been less investigated. In our study, the effect of 0.5 M of MgCl 2 was similar to that of CaCl 2 causing an overall inhibition of the HaHog1p phosphorylation. The reason for such inhibition was not clear and this merits further investigation. The oxidative stress exerted by the use of hydrogen peroxide caused the HaHog1p to be clearly phosphorylated. The activation pattern in H. annosum is very similar to what was observed in C. albicans exposed to 10 mM of H 2 O 2 [7]. In both fungi the phosphorylated Hog1p showed a peak around 2-10 min after oxidative stress exposure followed by a decrease in the signal intensity to a minimum at 60 min.
As no efficient DNA-transformation is available for H. annosum, a functional study of the HaHOG1 gene was therefore carried out