Root morphogenic pathways in Eucalyptus grandis are modified by the activity of protein arginine methyltransferases
© The Author(s). 2017
Received: 15 December 2016
Accepted: 1 March 2017
Published: 9 March 2017
Methylation of proteins at arginine residues, catalysed by members of the protein arginine methyltransferase (PRMT) family, is crucial for the regulation of gene transcription and for protein function in eukaryotic organisms. Inhibition of the activity of PRMTs in annual model plants has demonstrated wide-ranging involvement of PRMTs in key plant developmental processes, however, PRMTs have not been characterised or studied in long-lived tree species.
Taking advantage of the recently available genome for Eucalyptus grandis, we demonstrate that most of the major plant PRMTs are conserved in E. grandis as compared to annual plants and that they are expressed in all major plant tissues. Proteomic and transcriptomic analysis in roots suggest that the PRMTs of E. grandis control a number of regulatory proteins and genes related to signalling during cellular/root growth and morphogenesis. We demonstrate here, using chemical inhibition of methylation and transgenic approaches, that plant type I PRMTs are necessary for normal root growth and branching in E. grandis. We further show that EgPRMT1 has a key role in root hair initiation and elongation and is involved in the methylation of β-tubulin, a key protein in cytoskeleton formation.
Together, our data demonstrate that PRMTs encoded by E. grandis methylate a number of key proteins and alter the transcription of a variety of genes involved in developmental processes. Appropriate levels of expression of type I PRMTs are necessary for the proper growth and development of E. grandis roots.
KeywordsEucalyptus grandis Root hair initiation Root architecture Cytoskeleton
Post-translational modifications of proteins, including phosphorylation, acetylation, methylation or ubiquitination, have significant effects on both the structure and the function of proteins. Methylation of proteins typically occurs at arginine or lysine residues, catalysed by protein arginine methyltransferases (PRMTs) or protein lysine methyltransferases (PKMTs), respectively. Both families of enzymes use S-adenosyl-L-methionine (SAM) as a methyl donor to add one or more methyl groups to amines within the protein residue. Such modifications alter the end use of the protein [1, 2]. There are 11 identified members of the PRMT family, including the plant specific PRMT10 , which are classified into four different types based on the site(s) at which the arginine residue is methylated . Type I PRMTs, produce monomethyl arginines (MMA) as an intermediate and asymmetric dimethyl arginines (aDMA). Type II PRMTs produce MMAs as an intermediate and symmetric dimethyl arginines (sDMA), while Type III PRMTs only make MMAs . Type IV PRMTs methylate the secondary amine on the arginine and have only been described in yeast .
PRMTs are well conserved throughout all eukaryotic cells and their downstream effects include altered transcription, RNA processing, transport and translation, signal transduction, DNA repair, chromatin structure and cellular differentiation [3, 4]. Defects in PRMT expression have also been implicated in serious mammalian diseases [5, 6]. Protein targets of PRMTs are often histones resulting in altered gene transcription , although many non-histone methylated proteins have been identified [2, 8]. These non-histone proteins are often involved in RNA binding or transcription [9, 10], but also include cytoplasmic proteins involved in various cellular developmental processes . While these proteins have been well characterised in mammalian- and yeast-based systems, our understanding of PRMT activity in regulating plant development and signalling is still developing . In plants, Type I PRMTs have been implicated in the alteration of transcription through the methylation of histones (PRMT1 homologues; [11, 12]). These histone modifications have been found to affect flowering time, among other processes (PRMT4 or PRMT10 homologues; [13–15]). PRMTs are also involved in RNA processing and ribosomal biogenesis in Arabidopsis (PRMT3 homologue; ). AtPRMT5, a Type II PRMT, has been found to affect pre-mRNA splicing [17, 18], flowering time [18, 19], salt stress tolerance , primary root length , root stem cell maintenance during DNA damage  and circadian rhythms .
The majority of PRMT research in plants has involved the annual plant Arabidopsis thaliana. While many genomic resources exist in Arabidopsis, we were interested in annotating the PRMTs in a perennial tree model and characterizing their expression and effects on root formation. With a recently sequenced genome  and the ability to manipulate it genetically, the economically important tree model E. grandis presents a useful system to study the effects of PRMTs in a longer-lived plant species. We found that the E. grandis genome encodes a set of seven PRMTs, giving it one of the smallest complements of PRMTs in a sequenced plant species, and that these PRMTs are expressed in all major plant tissues. Using both chemical inhibition and transgenic modification of PRMT activity in E. grandis, we explored the role of Type I PRMTs in root growth and development. We found that transgenic repression of EgPRMT1, 3, 4 or 10 homologues result in a similar phenotype: interruption of the normal growth and branching of plant roots. Additionally, over-expression of EgPRMT1 causes abnormal root hair extension. We demonstrate that plant roots over-expressing EgPRMT1 have increased methylation of β-tubulin, which has been proposed to affect microtubule stability in neurons  and is a likely contributor to the root hair phenotype. Transcriptomic and proteomic data show that PRMTs act as key regulators of gene networks and pathways involved in the control of root growth and morphogenesis. Given the essential role of the root system, the study of PRMTs will be an important avenue of research to understand not only root patterning but also other aspects of plant health and nutrition.
E. grandis encodes PRMT-like genes that exhibit similar expression patterns in major plant tissues
E. grandis plant tissues contain a diverse set of asymmetrically dimethylated proteins
List of methylated proteins found in E. grandis root or leaf tissues
Methylated Peptide Sequences Identified by Mass Spectrometry
Predicted MW (kDa)
14-3-3-like protein A
5-methyltetrahydropteroyltriglutamate-homocysteine methyltransferase A
Aspartic proteinase nepenthesin-like protein B
ATP synthase subunit beta, mitochondrial A,B
Endochitinase A2 A, B
Glyceraldehyde-3-phosphate dehydrogenase A,B
Heat shock 70 kDa protein A
Heat shock 80 kDa protein A
Malic Enzyme A
Nucleoside diphosphate kinase 1A
Oxygen evolving enhancer protein 1A
ATP synthase delta chain, chloroplastic B
ATP synthase gamma chain, chloroplastic B
ATP synthase subunit beta, chloroplastic A,B
Basic endochitinase A B
Basic endochitinase C B
Carbonic anhydrase, chloroplastic B
Chlorophyll a-b binding protein A
Cytochrome B6 A
Fructose-bisphosphate aldolase A
Glutamine synthetase B
Mediator of RNA Polymerase II Transcription Subunit
Malate dehydrogenase, glyoxysomalB
Oxygen-evolving enhancer protein 3–2, chloroplastic B
Phosphoglycerate kinase A,B
Photosystem I reaction center subunit IV B, chloroplastic B
Ribulose bisphosphate carboxylase large chain A
165YGRPLLGCTIKPK 195GGLDFTKDDENVNSQPFMR 340DITLGFVDLVRDDFIEKDR 436DLAREGXDIIR
Ribulose bisphosphate carboxylase small chain A,B
Ribulose bisphosphate carboxylase/oxygenase activase A
274EENPRVPIIVTGNDFSTLYAPLIR 311EDRIGVCMGIFR 351ARVYDDEVRK
Ribulose-phosphate 3-epimerase, chloroplastic B
Thylakoid lumenal 29 kDa, chloroplastic A
Chemical inhibition of PRMT activity within roots alters root morphology
DMNS treated root tips have altered transcript abundance of genes related to root development
Transgenic roots differentially expressing PRMT genes show altered growth and lateral root formation
Transgenic roots expressing 35S::EgPRMT1 show increased β-tubulin methylation
List of methylated proteins found in E. grandis root tissues with elevated expression of EgPRMT1 or EgPRMT10
Methylated Peptide Sequences Identified by Mass Spectrometry
Predicted MW (kDa)
Endoplasmin/Heat Shock protein 90 kDa
RuBisCO large subunit- binding protein alpha
66 SISISISISIRPVR 346 RLEKAATYDEIK
Probable aldo-keto reductase
Armadillo/beta catenin repeat protein
40S Ribosomal Protein
Transaldolase family protein
Probable Aldo-keto reductase
ATP synthase subunit beta
Beta tubulin 5
Elongation factor 1-alpha
Heat shock protein 80 kDa
131 NRTTPSYVAFTDTER 558 FELSGIPPAPR
Beta Tubulin 5
Elongation factor 1-alpha
Heat shock protein 70 kDa
38 NRTTPSYVAFTDTER 478 DNNLLGKFELSGIPPAPR, 540 TTPSYVGFTDTER
Methylation of β-tubulin by EgPRMT1 was demonstrated using immunoprecipitation and Western blotting. Protein extracts from control and 35S::EgPRMT1 roots were immuno-precipitated with a β-tubulin antibody and the resulting protein was visualized using Western blots with both the ASYM24 and β-tubulin antibodies (Fig. 7b). The relative amount of methylated β-tubulin is, on average, 14x higher in the 35S::EgPRMT1 root sample than in the control (expressed as a fold-change of methylated tubulin over total β-tubulin calculated from Western blot band intensities).
We have demonstrated an important role for Type I PRMT homologues in the development of root tissues in E. grandis, an economically important model tree species. Unlike the better studied A. thaliana, E. grandis encodes only one PRMT4 homologue and one PRMT1 homologue, giving it the smallest complement of PRMT homologues out of the plant species considered. All seven of the identified PRMT homologues in E. grandis are expressed in a similar pattern in the various plant tissues investigated, even though methylated proteins vary by tissue. This suggests that despite similar gene expression profiles, PRMT substrates may vary between tissues and/or that PRMT proteins are subject to a level of control beyond transcription. The activity of PRMTs can be modified through phosphorylation [4, 35, 36], regulation by PRMT-binding proteins [37, 38] or by automethylation [39, 40]. Through mass spectrometry a broad selection of proteins from both root and leaf tissue were identified as having methylated arginine residues. Of these proteins, a large number are enzymes involved in photosynthesis or cellular respiration. Many of these have been previously described as having methylated arginine residues in other systems [41, 42]. Other proteins identified (e.g. Hsp70, 14-3-3, tubulin) have been associated with roles in maintenance of root architecture [30–32, 43, 44]. Previous work in tomato roots identified Hsp70 as methylated , and α- and β- tubulin were reported to be methylated in mammalian tissues , however methylation of 14-3-3 like proteins has not been previously described. It is likely that there are many more peptides and proteins methylated in E. grandis that were not detected in this study, but our results demonstrate that PRMTs of E. grandis appear to target regulatory proteins in a tissue-specific manner. Overall, the overlap between discovered methylated proteins in our system and those previously described indicate that there is some consistency and conservation in the targets of PRMTs, regardless of organism.
Chemical inhibitors of PRMT activity caused a reduction in growth when applied to E. grandis roots. Microscopic observation of the roots showed that the lack of growth in the roots may be due to a disturbance in root tip maintenance as both the meristem and elongation zones were significantly shortened. The results of the inhibitor experiments were complemented with transgenic mis-expression of Type I PRMTs in E. grandis roots, with RNAi silenced roots having both reduced growth and fewer lateral roots. Previous studies in fungal model systems have demonstrated a similar effect of PRMT activity on cellular extension and tissue growth, with deletions of PRMT1 or PRMT3 analogues resulting in significantly reduced hyphal elongation or hyphal branching, respectively [47, 48]. The transcriptomic analysis presented here of roots treated with DMNS – inhibiting only PRMT1 activity - showed a large number of significantly regulated genes, 128 of which have proposed roles in root development, including meristem maintenance according to PFAM enrichment analysis. A large number of enzymes involved in the formation of cell walls (eg. laccases, xyloglucan endotransglucosylases or cellulose synthases) are differentially regulated in tissues inhibited in PRMT1 activity. This may play a role in the lack of growth seen in treated and transgenic roots. Additionally, many regulated proteins are protein kinases or transcriptional regulators. Therefore, we find that PRMTs not only methylate and change the function of important signalling molecules directly (as seen with the protein data) but that they also affect the transcription of other regulatory molecules. Thus PRMTs are likely to be positioned near the top of a signalling cascade as major regulators of cellular processes.
We found that the transgenic mis-expression of four different Type I PRMT genes elicited similar root phenotypes. Additionally, recent work has shown that repression of PRMT5 (Type II PRMT) in Arabidopsis also results in shorter roots [17, 22]. This is an argument both for and against redundancy within the PRMT family as all of these PRMTs seem to accomplish a similar end result and yet are unable to “make up” for the loss of one another. A recent study by Hernandez et al.  demonstrated that the morphological and transcriptomic phenotype of A. thaliana to the loss of either AtPRMT4 or AtPRMT5 also had significant overlaps. One reason for this could be the tendency of PRMTs to form homo- or hetero-dimers with each other, thus multiple PRMTs may be necessary for the same methylation event . Alternatively, the proper function of a protein or signalling pathway may require methylation at multiple arginine residues by separate PRMTs and thus the loss of any one PRMT is deleterious. Finally, it is probable that PRMTs have such a crucial role in plant processes that the loss of any one of them severely compromises the health and vitality of the plant root causing the observed short root phenotype, though this may occur via a different mechanism for each PRMT homologue.
In mammalian systems, PRMT1 is the most active arginine methyltransferase, accounting for up to 80% of asymmetric arginine methylation , and preferentially methylates arginine residues within Glycine-Arginine rich (GAR) motifs . Within young E. grandis seedlings, EgPRMT1 is not the most highly expressed PRMT gene although Western blot analysis of 35S::EgPRMT1 root proteins shows a greater increase in methylated proteins as compared to 35S::EgPRMT10 roots. This latter result could be due to a bias of the ASYM24 antibody generated against the GAR motif preferentially methylated by PRMT1 . Arguing against this bias, however, was the finding that the majority of the peptides with methylated arginine residues identified by mass spectrometry did not encode a GAR motif: only 12% of the methylated arginine residues identified had a neighbouring glycine residue. Therefore, while it is possible that PRMT1 plays a predominant role in asymmetric methylation of E. grandis proteins as has been observed in mammalian systems, it is likely that it methylates more than just GAR motifs.
Overexpression of EgPRMT1 results in a root hair phenotype not seen with the other E. grandis PRMTs studied. Normal root hairs grow only from the tip in a finely balanced cooperation between microtubules and the actin cytoskeleton . Tubulin has been found to control root morphology in Arabidopsis, where reduced levels of tubulin correlated with aberrant microtubule assembly, laterally expanded root width, reduced root growth and altered root hair density and morphology . Both actin and tubulin proteins have been identified as containing methylated arginine residues in mammalian systems [25, 46, 54]. Our own analysis identified α- and β-tubulin as methylated in 35S::EgPRMT1 tissues. Further, immunoprecipatation and Western blot analysis demonstrated a 14 times increase in the relative amounts of methylated β-tubulin within 35S::EgPRMT1 tissues, identifying β-tubulin as a substrate of EgPRMT1. As increased methylation has been proposed to cause microtubule destabilisation in mammalian tissues , it is reasonable to suggest that the over-methylation of β-tubulin in EgPRMT1 overexpressing plants could be one of the factors contributing to the abnormal root hair phenotype. Additionally, xyloglucan endotransglycosylase activity is necessary for root hair initiation in Arabidopsis  and our transcriptomic data demonstrate that several xyloglucan endotransglycosylases are upregulated in inhibitor treated tissues, implicating EgPRMT1 as a repressor of their expression. Thus, the observation of rounded root hairs in 35S::EgPRMT1 could occur directly, as a result of excessive methylation of cytoskeletal proteins, and/or indirectly through alteration of cellular signalling.
PRMTs are important, well-conserved proteins found in the genomes of all eukaryotic organisms described to date. Eucalyptus grandis encodes its own set of seven PRMT homologues that are expressed in all plant tissues and methylate a variety of proteins involved in photosynthesis, cellular respiration and signalling. Investigation into the role of Type I PRMTs in the development of E. grandis roots demonstrates a crucial role for these proteins in the growth and branching of plant roots and root hair initiation. Our results concerning the impact of EgPRMT1 gene expression on root hair morphology also demonstrate that, while PRMTs are crucial in many cellular processes, their over-production can also have negative effects. Therefore, these genes must be carefully regulated within cells. The mechanisms by which PRMTs alter root morphology, however, require further studies as well as an investigation into the roles of the individual plant proteins methylated by PRMTs and their downstream effects.
Construction of PRMT phylogeny in model plant species
PANTHER protein classification PTHR11006 and the PFAM conserved domain PF05185 were used to retrieve all PRMT containing sequences in the genomes of E. grandis, A. thaliana, O. sativa, G. max, P. trichocarpa, and S. purpurea from Phytozome v10.3 (phytozome.jgi.doe.gov: accessed 29/10/2015). A phylogenetic tree was constructed using the online tool ‘Phylogeny.fr’. All of the PRMT-like protein sequences were downloaded from the Phytozome database. PRMT homologue assignment was based on prior annotation of the A. thaliana genome (http://www.arabidopsis.org/).
PRMT expression in plant tissues
Eucalyptus grandis (W. Hill ex Maiden) seeds were obtained from the Commonwealth Scientific and Industrial Research Organisation (CSIRO, Clayton, Vic., Australia) tree seed center (Seedlot 21068). They were sterilized in 30% H2O2 for 10 min, followed by several washes in sterile water. Seeds were germinated on 1% agar media and transferred after one month to MS media. A sterile cellophane membrane was placed on the surface of the MS media to prevent the roots from growing into it. Plants were grown on the MS media for another six weeks with a 16 h photoperiod and at a temperature of 25 °C and were then harvested. Three independent biological replicates of leaves, stem, shoot apex and roots were harvested separately and immediately frozen in liquid nitrogen. RNA was extracted from the tissues using the Qiagen RNeasy Plant Mini kit as per manufacturer’s instructions. RNA was made into cDNA with the iScript cDNA synthesis kit (Bio-Rad). The SensiFAST SYBR No-ROX kit (Bioline) was used for qPCR analysis on a Corbett Rotor-gene 6000 RT-PCR cycler. Relative expression levels were calculated as the difference in expression as compared to the control genes, Eucgr.C00350.2 and Eucgr.K02046.1. Primer sequences used can be found in Additional file 4: Table S4. Expression values of PRMTs in mature Eucalyptus tissues was retrieved from https://phytozome.jgi.doe.gov/pz/portal.html#!info?alias=Org_Egrandis (accessed November 2016).
Protein extraction and mass spectrometric identification of methylated proteins in plant tissues
A mass of 250 mg of fresh E. grandis leaf or root tissues was harvested and snap frozen in liquid nitrogen. Tissues were then ground immediately after freezing in a sintered glass tissue grinder in either ice cold 1x IP Lysis/Wash Buffer (Thermo Scientific) supplemented with 1 mM plant protease inhibitor cocktail (Sigma Aldrich; Cat#P9599) for immunoprecipitation or in 50 mM Tris HCl (pH 8)/1% SDS/50 μM DTT/1% PVPP/1 mM plant protease inhibitor cocktail for total protein extraction and Western blotting. Grinding was performed on ice for no more than 3 min after which the soluble protein was quantified using the Qubit total protein analysis kit according to manufacturer’s instructions (Life Technologies). Protein extract was then diluted to 1 μg/μL using the extraction buffer and either used immediately for immunoprecipitation or was mixed with NuPAGE LDS buffer (Life Technologies) and snap frozen in liquid nitrogen for Western blotting.
Immunoprecipitation was performed using the Pierce IP immunoprecipitation kit (Thermo-Scientific; Cat#26148) where ASYM24 (Merck-Millipore; Cat#07-414) or anti-β-tubulin (Abcam; Cat#ab6046) was cross-linked to the agarose resin. Approximately 900 μg of total protein was added to the agarose slurry and incubated at room temperature for 2 h. Following incubation, rinses and elution of the bound proteins were performed according to manufacturer’s instructions.
For protein separation by electrophoresis, equal amounts of total protein were separated on a 4–20% Mini-PROTEAN® TGX™ gradient gel for 2 h at 80 V. The gel was stained overnight, either with Brilliant Blue G- Colloidal stain or SYPRO Ruby, at room temperature with shaking. Coomassie stained gels were destained in milli-Q water for 5 h. Protein bands were excised and further destained for 10 min in a 1:1 solution of 25 mM ammonium bicarbonate and acetonitrile. This was repeated until bands appeared colourless, and was followed by incubation with acetonitrile for 20 min. Bands in SYPRO Ruby stained gels were visualised by UV Transilluminator, bands were excised and then destained as per the Coomassie stained bands. All further treatments were identical for gel bands stained with either staining method.
The gel bands were incubated with 10 mM DTT in milli-Q water for 1 h at 37 °C to reduce cysteine residues, then with 25 mM IAA in milli-Q water for 1 h at 37 °C to alkylate the cysteine residues. Proteins were digested by the addition of Trypsin Gold (Promega) with a final enzyme concentration of 5 μg/mL and incubated overnight at 37 °C. While trypsin was used to cleave proteins, it has been reported to produce inconsistent  or ineffective cleavage at di-methylated arginine sites . The digestion solution was collected into new low-binding tubes and peptides were further extracted by adding a 1:1 solution of acetonitrile and 0.1% formic acid and sonicated for 10 min. This process was repeated twice. Peptides were dried under vacuum (Waters) and were then resuspended in 0.1% formic acid.
LC-MS/MS analysis was performed on a Xevo QToF mass spectrometer from Waters (Micromass,UK) fed by a nanoAquity UPLC (Waters Corp., Milford, MA, USA) at the Western Sydney University mass spectrometry facility. 3 μl of digested peptides were loaded onto a nanoAquity UPLC Symmetry C18 trapping column (1.7 μm, 180 μm × 20 mm) and then separated and eluted from the column using a binary gradient program at a flow rate of 5 μl/min and desalted at this flow rate for 3 min. The peptides were washed off the trap at 400 nL/min on to a Waters C18 BEH analytical column (75 μm x 100 mm), packed with 1.7 μm particles with 130 Å pore size. After separation, the peptides were analysed using tandem mass spectrometry, implementing an emitter tip that tapers to 10 μm at 2300 V. Mobile phase A was 0.1% formic acid in water; and mobile phase B was 0.1% formic acid in acetonitrile. The nano-UPLC gradient was as follows: 0 min, 97:3 A/B; 5 min, 97:3 A/B; 75 min, 40:60 A/B; 85 min, 10:90 A/B; 97 min, 97:3 A/B; 110 min, 97:3 A/B. The mass spectrometer was operated in positive ESI mode with capillary voltage of 3.5 KV, cone voltage of 40 V, source temperature of 80 °C. Targeted MS/MS data or DDA (data dependent acquisition) data were acquired by continuously scanning for peptides of charge state 2+ to 4+ with an intensity of more than 50 counts per second; with a maximum of three ions in any given 3 s scan. Selected peptides were fragmented and the product ion fragment masses were measured. The data were acquired by the software Masslynx (Version 4.1, Micromass, UK).
The acquired DDA data from Masslynx with “RAW” extension, were converted to PKL files by Proteinlynx Global Server (PLGS) for analysis using the Mascot Daemon database (Australian Proteomics Computational Facility, Melbourne, Australia).
The MS/MS data files were searched against MSPnr100 database with trypsin as the enzyme. The following parameters were considered in the Mascot Editor tab for identification of the peptides: maximum missed cleavage of 3, peptide charge state of 2+ and 3+, peptide mass tolerance of 0.5 Da in MS and MS/MS data base, fixed modification: carbamidomethyl (C) and variable modifications: oxidation (M), mono and di-methylation of arginine (R). An ion score of 20 was applied to all results in order to filter out low probability matches. Peptides were also matched using Peaks Studio Software (version 7.5; Bioinformatics Solutions Inc., Waterloo, Ontario, Canada) by uploading the E. grandis protein list translated from the primary transcripts  and performing a database search to identify proteins and post-translational modifications. The following parameter settings were used: fixed modifications: carbamidomethyl (C); variable modifications: oxidation (M), mono and di-methylation of arginine (R); enzyme: trypsin; number of allowed missed cleavages: 3; peptide mass tolerance; 100 ppm; MS/MS mass tolerance: 0.5 Da; and peptide charge state: 2+ and 3+. Proteins identified as uncharacterised proteins were matched to their function by performing a BLAST search (Uniprot) and homology of the identified residues was checked using the alignment tool (Uniprot).
A recent report by Hart-Smith et al.  warned of false positive detection of methylated peptides when using ethanol, methanol and/or isopropanol in the preparation of large-scale mass spectrometry. The methods used here were different to the ones described by Hart-Smith  and the results were validated in the absence of alcohol-based staining. Furthermore, bacterially expressed tubulin β (which lacks methylation, Sigma; Cat#SRP5148) analysed from stained or unstained protocols, did not yield any falsely identified methylated arginine residues using the mass spectrometry protocol described above.
Western blot analysis
For Western blot analysis, equal amounts of protein were separated on a 4–20% Mini-PROTEAN® TGX™ gel for 2 h at 80 V. Proteins were then transferred onto a Nitrocellulose membrane (Bio Rad) for 1 h at a constant voltage of 35 V using an X-Cell II Blot Module in an Invitrogen Novex™ Mini-Cell. The buffer used was NuPAGE transfer buffer (Cat# NP0006). The membranes were blocked for 1 h in 3% skim milk powder dissolved in Tris-buffered saline (TBS; 50 mM Tris-Cl/150 mM NaCl pH 8.0). After blocking, the membrane was washed with TBS 3x followed by overnight incubation at 4 °C in the primary antibody diluted in TBS as follows: 1:500 dilution of ASYM24 (for detection of asymmetrically methylated proteins) or 1:500 dilution of anti-tubulin (as a loading control protein). The membranes were then washed 3x in TBS and then incubated for 2 h at room temperature in an HRP-conjugated anti-rabbit antibody (Merck-Millipore; Cat#12-348). Following another 3x washes in TBS, the membranes were treated with Clarity™ ECL substrate for 5 min and the chemiluminescent signal was captured using a VersaDoc Imaging System (Bio Rad). All Western blots presented in the main text are representative images of multiple blots. Each sample type was extracted in a minimum of biological triplicates with the exception of transgenic tissues, which were each treated as a single transformed line.
Inhibitor trials – roots only
Young E. grandis seedlings were grown as above – germinating for one month on 1% agar and then growing for an additional six weeks on MS media covered with a sterile cellophane membrane. Their roots were cut and allowed to commence regrowing to ensure an active growth phase. At T0, the starting length and position of the roots were recorded. Roots were treated with either AdOx (5 nM-5 μM), AMI-1 (50 nM-50 μM), 2,3-dimethoxynitrostyrene (30 nM-30 μM in 1% DMSO), 1% DMSO control or water control and root growth was measured 24 h later. The first inhibitor, AdOx, is a general methylation inhibitor and indirectly blocks the reaction providing the methyl group that is added by methyltransferases to their respective substrates including arginine. Thus all methylation processes, including lysine methylation, are inhibited by AdOx. The second inhibitor, AMI-1, more specifically inhibits protein arginine methylation by specifically blocking PRMTs but not PKMTs . The more recently described DMNS only inhibits the activity of PRMT1 (and PRMT8, not present in E. grandis; ). Alignment of PRMT protein sequences from human and E. grandis genomes shows that EgPRMT1 has the appropriate cysteine residue to be inhibited by DMNS, while the other E. grandis PRMTs do not (Additional file 5: Figure S1). One millilitre of solution was administered directly to each root system daily over the test period of two weeks. Between 17 and 30 replicates were performed for each treatment type. Lateral rooting was analyzed after two weeks of treatment. Root tip samples were taken and observed with a Zeiss Stemi 2000-C stereomicroscope (Germany). All plants were alive and healthy at the time of harvest, indicating that the dosage of inhibitors was sub-lethal.
Transcriptomic analysis of DMNS treated roots
Three representative biological replicates of roots from DMNS treated roots and 1% DMSO treated control roots were snap frozen in liquid nitrogen and the RNA was extracted using the Qiagen RNeasy Plant Mini kit as per manufacturer’s instructions. Transcriptional analyses of all tissues were performed using RNA-Seq via conventional poly-A library preparation for Illumina sequencing (TruSeq RNA Library Prep v2, Illumina). Library construction and 100-bp paired-end reads sequencing was performed by the Western Sydney University Next Generation Sequencing Facility on three independent biological replicates of DMSO-control treated roots and DMNS treated roots. The samples were indexed and run on four high-output lanes (paired end, 1.4 billion perfect reads) of an Illumina Hi-Seq 2000 flow-cell. Raw reads were trimmed for quality and aligned to the primary transcripts of the E. grandis genome taken from https://phytozome.jgi.doe.gov/pz/portal.html#!info?alias=Org_Egrandis (accessed November 2015; ) using CLC Genomics Workbench 7. For mapping, the minimum length fraction was 0.9, the minimum similarity fraction 0.8, and the maximum number of hits for a read was set to 10. The unique and total mapped reads number for each transcript were determined, and then normalized to reads per kilobase of exon model per million mapped reads. CLC software was used to determine the statistical significance of gene expression. To identify functional domains that were significantly enriched, PFAM annotation of the differentially regulated genes was analyzed by the dcGO program http://supfam.org/SUPERFAMILY/cgi-bin/dcenrichment.cgi (accessed December 2015). Only FDR-corrected of PFAM slim ‘specific’ and ‘highly specific’ annotations with p-values < 0.001 were assigned as significantly over-represented in the data set (Additional file 2: Table S2 and Additional file 6: Table S3).
Production of PRMT 35S::overexpression and RNAi transgenic roots
EgPRMT1, EgPRMT10, EgPRMT3 and EgPRMT4 were cloned from cDNA synthesized using iScript (Bio Rad) from total RNA extracted from E. grandis roots using RNeasy plant extraction kit according to manufacturer’s instructions (Qiagen). The amplified fragments were gel purified and ligated into pDONR222 and sequence verified. Positive inserts were then transferred to pH2GW7 (35S::) or to pH7GWIWG2(II) (RNAi) using Gateway Gene Cloning (Life Technologies) and transformed into Agrobacterium rhizogenes isolate K599. E. grandis seedlings were grown from seed to one month old on 1% agar media. The root system of the plant was cut off and the resultant wound was dipped in freshly grown A. rhizogenes expressing the plasmid of choice, or wild type control. Dipped plants were embedded in MS media and left for one week, placed upside down, in a growth cabinet with a constant temperature of 25 °C and a 16 h photoperiod. Plants were then transferred to fresh MS media supplemented with 150 μg/mL Timentin and grown under the same conditions . Transformed roots typically emerged within one or two weeks. After one months of growth, the transgenic roots were harvested. Harvested roots were scanned and measured (using ImageJ), analysed by microscopy (Zeiss Stemi 2000-C, Germany) and frozen at −80 °C. RNA was extracted from select transformed roots, made into cDNA and analyzed by qPCR to confirm altered expression of the transgene. Growth data for each individual transformed line is given in Additional file 3: Figure S2. Antibodies that cross-react with E. grandis PRMT proteins are only commercially available for EgPRMT1. Therefore, we were only able to confirm increased abundance of the EgPRMT1 protein in the transgenic roots (Additional file 7: Figure S3). RNAi lines were not similarly probed as total protein recovered was not sufficient to detect EgPRMT1 signal.
Asymmetric dimethyl arginine
Arginine methyltransferase inhibitor 1
Copy deoxyribonucleic acid
Data dependent analysis
Glycine Arginine rich
Liquid chromatography tandem mass spectrometry
Lithium dodecyl sulphate
Polyacrylamine gel electrophoresis
Protein lysine methyltransferase
Protein arginine methyltransferase
Quantitative polymerase chain reaction
Ribonucleic acid interference
Symmetric dimethyl arginine
Sodium dodecyl sulphate
Ultra performance liquid chromatography
We would like to thank Fiona Koller and Jasmine Grinyer for technical assistance, Peter Gresshoff for the K599 strain of A. rhizogenes and the mass spectrometry facility at Western Sydney University. The authors would like to acknowledge the Western Sydney University Confocal Bio-Imaging Facility for access to its instrumentation and staff. Salix purpurea sequence data were produced by the US Department of Energy Joint Genome Institute and used with permission from LB Smart.
We would like to thank the Australian Research Council for funding to JMP (DE 150100408) and the Hawkesbury Institute for the Environment and Western Sydney University for research scholarships to AEL and SB.
Availability of data and materials
The datasets acquired and/or analysed during the current study are available from the corresponding author on reasonable request.
JMP, SCP and KLP conceived and designed the experiments; KLP, AEL, SB and JMP performed the experiments; KLP, AEL, JMP, SCP and ICA analyzed the data, KLP wrote the article with contributions from all the authors. All authors read and approved the final manuscript.
The authors declare that they have no competing interests.
Consent for publication
Ethics approval and consent to participate
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Open AccessThis article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.
- Bedford MT, Clarke SG. Protein arginine methylation in mammals: who, what, and why. Mol Cell. 2009;33:1–13.View ArticlePubMedPubMed CentralGoogle Scholar
- Lee YH, Stallcup MR. Minireview: protein arginine methylation of nonhistone proteins in transcriptional regulation. Mol Endocrinol. 2009;23(4):425–33.View ArticlePubMedPubMed CentralGoogle Scholar
- Ahmad A, Cao X. Plant PRMTs broaden the scope of arginine methylation. J Genet Genomics. 2012;39:195–208.View ArticlePubMedGoogle Scholar
- Bedford MT. Arginine methylation at a glance. J Cell Sci. 2007;120:4243–6.View ArticlePubMedGoogle Scholar
- Yang Y, Bedford MT. Protein arginine methyltransferases and cancer. Nat Rev Cancer. 2013;13:37–50.View ArticlePubMedGoogle Scholar
- Mirto L, Piller S. Protein arginine methylation: a pivotal factor and a target for novel HIV-therapeutics, other viruses and multiple diseases. HIV Ther. 2010;4(1):65–81.View ArticleGoogle Scholar
- Liu C, Lu F, Cui X, Cao X. Histone methylation in higher plants. Annu Rev Plant Biol. 2010;61:395–420.View ArticlePubMedGoogle Scholar
- Wei H, Mundade R, Lange KC, Lu T. Protein arginine methylation of non-histone proteins and its role in diseases. Cell Cycle. 2014;13:32–41.View ArticlePubMedGoogle Scholar
- Blackwell E, Ceman S. Arginine methylation of RNA-binding proteins regulates cell function and differentiation. Mol Reprod. 2011;79:163–75.View ArticleGoogle Scholar
- Bedford MT, Richard S. Arginine methylation: an emerging regulator of protein function. Mol Cell. 2005;18:263–72.View ArticlePubMedGoogle Scholar
- Yan D, Zhang Y, Niu L, Cao X. Identification and characterization of two closely related histone H4 arginine 3 methyltransferases in Arabidopsis thaliana. Biochem J. 2007;408:113–21.View ArticlePubMedPubMed CentralGoogle Scholar
- Ahmad A, Dong Y, Cao X. Characterization of the PRMT gene family in rice reveals conservation of arginine methylation. PLoS ONE. 2011;6(8):e22664.View ArticlePubMedPubMed CentralGoogle Scholar
- Niu L, Lu F, Pei Y, Liu C, Cao X. Regulation of flowering time by the protein arginine methyltransferase AtPRMT10. EMBO Rep. 2007;8:1190–5.View ArticlePubMedPubMed CentralGoogle Scholar
- Niu L, Zhang Y, Pei Y, Liu C, Cao X. Redundant requirement for a pair of PROTEIN ARGININE METHYL TRANSFERASE4 homologs for the proper regulation of Arabidopsis flowering time. Plant Physiol. 2008;148:490–503.View ArticlePubMedPubMed CentralGoogle Scholar
- Liew LC, Singh MB, Bhalla PL. An RNA-Seq transcriptome analysis of histone modifiers and RNA silencing genes in soybean during floral initiation process. PLoS One. 2013;8(10):e77502.View ArticlePubMedPubMed CentralGoogle Scholar
- Hang R, Liu C, Ahmad A, Zhang Y, Lu F, Cao X. Arabidopsis protein arginine methyltransferase 3 is required for ribosome biogenesis by affecting precursor ribosomal RNA processing. Proc Natl Acad Sci. 2014;111:16190–5.View ArticlePubMedPubMed CentralGoogle Scholar
- Deng X, Lu T, Wang L, Gu L, Sun J, Kong X, Liu C, Cao X. Recruitment of the NineTeen complex to the activated spliceosome requires AtPRMT5. Proc Natl Acad Sci U S A. 2016;113:5447–52.View ArticlePubMedPubMed CentralGoogle Scholar
- Deng X, Gu L, Liu C, Lu T, Lu F, Lu Z, Cui P, Pei Y, Wang B, Hu S. Arginine methylation mediated by the Arabidopsis homolog of PRMT5 is essential for proper pre-mRNA splicing. Proc Natl Acad Sci U S A. 2010;107:19114–9.View ArticlePubMedPubMed CentralGoogle Scholar
- Wang X, Zhang Y, Ma Q, Zhang Z, Xue Y, Bao S, Chong K. SKB1-mediated symmetric dimethylation of histone H4R3 controls flowering time in Arabidopsis. EMBO J. 2007;26:1934–41.View ArticlePubMedPubMed CentralGoogle Scholar
- Zhang Z, Zhang S, Zhang Y, Wang X, Li D, Li Q, Yue M, Xu Y. Arabidopsis floral initiator SKB1 confers high salt tolerance by regulating transcription and pre-mRNA splicing through altering histone H4R3 and small nuclear ribonucleoprotein LSM4 methylation. Plant Cell. 2011;23:396–411.View ArticlePubMedPubMed CentralGoogle Scholar
- Pei Y, Niu L, Lu F, Liu C, Zhai J, Kong X, Cao X. Mutations in the type II protein arginine methyltransferase AtPRMT5 result in pleiotropic developmental defects in Arabidopsis. Plant Physiol. 2007;144:1913–23.View ArticlePubMedPubMed CentralGoogle Scholar
- Li Q, Zhao Y, Yue M, Xue Y, Bao S. The protein arginine methylase 5 (PRMT5/SKB1) gene is required for the maintenance of root stem cells in response to DNS damage. J Genet Genomics. 2016;43:187–97.View ArticlePubMedGoogle Scholar
- Sanchez SE, Petrillo E, Beckwith EJ, Zhang X, Rugnone ML, Hernando CE, et al. A methyl transferase links the circadian clock to the regulation of alternative splicing. Nature. 2010;468(7320):112–6.View ArticlePubMedGoogle Scholar
- Myberg AA, Grattapaglia D, Tuskan GA, Hellsten U, Hayes RD, Grimwood J, Jenkins J, Lindquist E, et al. The genome of Eucalyptus grandis. Nature. 2014;510:356–62.Google Scholar
- Piller S, Jwad N, Hejazi L, Gamsjaeger R, Sucher N. Protein arginine methylation of tubulin beta decreases binding of taxol in Neuro2a cells. FASEB J. 2015;29(1):s717.16.Google Scholar
- Lamesch P, Berardini TZ, Li D, Swarbreck D, Wilks C, Sasidharan R, Muller R, Dreher K, et al. The Arabidopsis information resource (TAIR): improved gene annotation and new tools. Nucleic Acids Res. 2012;40(Database issue):D1202–10.View ArticlePubMedGoogle Scholar
- Ouyang S, Zhu W, Hamilton J, Lin H, Campbell M, Childs K, Thibaud-Nissen F, Malek RL, et al. The TIGR rice genome annotation resourse: improvements and new features. Nucleic Acids Res. 2007;35(Database issue):D883–7.View ArticlePubMedGoogle Scholar
- Schmutz J, Cannon SB, Schlueter J, Ma J, Mitros T, Nelson W, Hyten DL, Song Q, et al. Genome sequence of the palaepolyploid soybean. Nature. 2010;463(7278):178–83.View ArticlePubMedGoogle Scholar
- Tuskan GA, Difazio S, Jansson S, Bohlmann J, Grigoriev I, Hellsten U, Putnam N, Ralph S, et al. The genome of black cottonwood, Populus trichocarpa. Science. 2006;313:1596–604.View ArticlePubMedGoogle Scholar
- Sung DY, Guy CL. Physiological and molecular assessment of altered expression of Hsc70-1 in Arabidopsis. evidence for pleiotropic consequences. Plant Physiol. 2003;132:979–87.View ArticlePubMedPubMed CentralGoogle Scholar
- Cazale A-C, Clement M, Chiarenza S, Roncato M-A, Pochon N, Creff A, Marin E, Leonhardt N, Noel LD. Altered expression of cytosolic/nuclear HSC70-1 molecular chaperone affects development and abiotic stress tolerance in Arabidopsis thaliana. J Exp Bot. 2009;60:2653–64.View ArticlePubMedGoogle Scholar
- He Y, Wu J, Lv B, Li J, Bao Z, Xu W, Baluska F, Shi W, Shaw PC, Zhang J. Involvement of 14-3-3 protein GRF9 in root growth and response under polyethylene glycol-induced water stress. J Exp Bot. 2015;66:2271–81.View ArticlePubMedPubMed CentralGoogle Scholar
- Kumar V, Parkhi V, Kenerley CM, Rathore KS. Defense-related gene expression and enzyme activities in transgenic cotton plants expressing an endochitinase gene from Trichoderma virens in response to interaction with Rhizoctonia solani. Planta. 2009;230:277–91.View ArticlePubMedGoogle Scholar
- Bao Y, Kost B, Chua NH. Reduced expression of alpha-tubulin genes in Arabidopsis thaliana specifically affects root growth and morphology, root hair development and root gravitropism. Plant J. 2001;28:145–57.View ArticlePubMedGoogle Scholar
- Higashimoto K, Kuhn P, Desai D, Cheng X, Xu W. Phosphorylation-mediated inactivation of coactivator-associated arginine methyltransferase 1. Proc Natl Acad Sci. 2007;104:12318–23.View ArticlePubMedPubMed CentralGoogle Scholar
- Liu F, Zhao X, Perna F, Wang L, Koppikar P, Abdel-Wahab O, Harr MW, Levine RL, Xu H, Tefferi A. JAK2V617F-mediated phosphorylation of PRMT5 downregulates its methyltransferase activity and promotes myeloproliferation. Cancer Cell. 2011;19(2):283–94.View ArticlePubMedPubMed CentralGoogle Scholar
- Jelinic P, Stehle J-C, Shaw P. The testis-specific factor CTCFL cooperates with the protein methyltransferase PRMT7 in H19 imprinting control region methylation. PLoS Biol. 2006;4:e355.View ArticlePubMedPubMed CentralGoogle Scholar
- Singh V, Miranda TB, Jiang W, Frankel A, Roemer ME, Robb VA, Gutmann DH, Herschman HR, Clarke S, Newsham IF. DAL-1/4.1 B tumor suppressor interacts with protein arginine N-methyltransferase 3 (PRMT3) and inhibits its ability to methylate substrates in vitro and in vivo. Oncogene. 2004;23:7761–71.View ArticlePubMedGoogle Scholar
- Dillon MBC, Rust HL, Thompson PR, Mowen KA. Automethylation of protein arginine methyltransferase 8 (PRMT8) regulates activity by impeding S-adenosylmethionine sensitivity. J Biol Chem. 2013;288(39):27872–80.View ArticlePubMedPubMed CentralGoogle Scholar
- Singhroy DN, Mesplede T, Sabbah A, Quashie PK, Falgueyret J-P, Wainberg MA. Automethylation of protein arginine methyltransferase 6 (PRMT6) regulates its stability and its anti-HIV-1 activity. Retrovirology. 2013;10:73.View ArticlePubMedPubMed CentralGoogle Scholar
- Alban C, Tardif M, Mininno M, Brugiere S, Gilgen A, et al. Uncovering the protein lysine and arginine methylation network in Arabidopsis chloroplasts. PLoS One. 2014;9(4):e95512.View ArticlePubMedPubMed CentralGoogle Scholar
- Fisk JC, Li J, Wang H, Aletta JM, Qu J, Read LK. Proteomic analysis reveals diverse classes of arginine methylproteins in mitochondria of trypanosomes. Mol Cell Proteomics. 2013;12:302–11.View ArticlePubMedGoogle Scholar
- Denison FC, Paul AL, Zupanska AK, Ferl RJ. 14-3-3 proteins in plant physiology. Semin Cell Dev Biol. 2011;22(7):720–7.View ArticlePubMedGoogle Scholar
- Lozano-Duran R, Robatzek S. 14-3-3 proteins in plant-pathogen interactions. Mol Plant Microbe Interact. 2015;28(5):511–8.View ArticlePubMedGoogle Scholar
- Nover L, Scharf K-D. Synthesis, modification and structural binding of heat-shock proteins in tomato cell cultures. Eur J Biochem. 1984;139:303–13.View ArticlePubMedGoogle Scholar
- Guo A, Gu H, Zhou J, Mulhern D, Wang Y, Lee KA, Yang V, Aguiar M, Kornhauser J, Jia X. Immunoaffinity enrichment and mass spectrometry analysis of protein methylation. Mol Cell Proteomics. 2014;13(1):372–87.View ArticlePubMedGoogle Scholar
- Feldman D, Ziv C, Gorovits R, Efrat M, Yarden O. Neurospora crassa protein arginine methyl transferases are involved in growth and development and interact with the NDR kinase COT1. PLoS One. 2013;8(11):e80756.View ArticlePubMedPubMed CentralGoogle Scholar
- Wang G, Wang C, Hou R, Zhou X, Li G, Zhang S, Xu J-R. The AMT1 arginine methyltransferase gene is important for plant infection and normal hyphal growth in Fusarium graminearum. PLoS One. 2012;7(5):e38324.View ArticlePubMedPubMed CentralGoogle Scholar
- Hernando CE, Sanchez SE, Mancini E, Yanovsky MJ. Genome wide comparative analysis of the effects of PRMT5 and PRMT4/CARM1 arginine methyltransferases on the Arabidopsis thaliana transcriptome. BMC Genomics. 2015;16:192.View ArticlePubMedPubMed CentralGoogle Scholar
- Le DD, Fujimori DG. Protein and nucleic acid methylating enzymes: mechanisms and regulation. Curr Opin Chem Biol. 2012;16:507–15.View ArticlePubMedPubMed CentralGoogle Scholar
- Rawal N, Rajpurohit R, Lischwe MA, Williams KR, Paik WK, Kim S. Structural specificity of substrate for S-adenosylmethionine protein arginine N-methyltransferases. BBA Protein Struct Mol Enzym. 1995;1248(1):11–8.View ArticleGoogle Scholar
- Côté J, Boulanger M-C, Bedford MT, Richard S. Sam68 RNA binding protein is an in vivo substrate for protein arginine N-methyltransferase 1. Mol Biol Cell. 2003;14:274–87.View ArticlePubMedPubMed CentralGoogle Scholar
- Sieberer BJ, Ketelaar T, Esseling J, Emons AMC. Microtubules guide root hair tip growth. New Phytol. 2005;167:711–9.View ArticlePubMedGoogle Scholar
- Pang CNI, Gasteiger E, Wilkins MR. Identification of arginine- and lysine-methylation in the proteome of Saccharomyces cerevisiae and its functional implications. BMC Genomics. 2010;11:92.View ArticlePubMedPubMed CentralGoogle Scholar
- Vissenberg K, Fry SC, Verbelen JP. Root hair initiation is coupled to a higly localized increase of xyloglucan endotransglycosylase action in Arabidopsis roots. Plant Physiol. 2001;127:1125–35.View ArticlePubMedPubMed CentralGoogle Scholar
- Martin G, Ostareck-Lederer A, Chari A, Neuenkirchen N, Dettwiler S, Blank D, Ruegsegger U, Fischer U, Keller W. Arginine methylation in subunits of mammalian pre-mRNA cleavage factor I. RNA. 2010;16(8):1646–59.View ArticlePubMedPubMed CentralGoogle Scholar
- Hart-Smith G, Yagoub D, Tay AP, Pickford R, Wilkins MR. Large-scale mass spectrometry-based identifications of enzyme-mediated protein methylation are subject to high false discovery rates. Mole Cell Proteomics. 2016;15(3):989–1006.View ArticleGoogle Scholar
- Cheng D, Yaday N, King RW, Swanson MS, Weinstein EJ, Bedford MT. Small molecule regulators of protein arginine methyltransferases. J Biol Chem. 2004;279:23892–9.View ArticlePubMedGoogle Scholar
- Dillon MBC, Bachovchin DA, Brown SJ, Finn MG, Rosen H, Cravatt BF, Mowen KA. Novel inhibitors for PRMT1 discovered by high-throughput screening using activity-based fluorescence polarization. ACS Chem Biol. 2012;7:1198–204.View ArticlePubMedPubMed CentralGoogle Scholar
- Macrae S, van Staden J. Agrobacterium rhizogenes-mediated transformation to improve rooting ability of eucalypts. Tree Physiol. 1993;12:411–8.View ArticlePubMedGoogle Scholar
- Mizrachi E, Hefer CA, Ranik M, Joubert F, Myberg AA. De novo assembled expressed gene catalog of a fast-growing Eucalyptus tree produced by Illumina mRNA-Seq. BMC Genomics. 2010;11:681.
- Plett JM, Kohler A, Khachane A, Keniry K, Plett KL, Martin F, Anderson IC. The effect of elevated carbon dioxide on the interaction between Eucalyptus grandis and diverse isolates of Pisolithus sp. is associated with a complex shift in the root transcriptome. New Phytol. 2015;206:1423–36.