MTA, an RNA m6A Methyltransferase, Enhances Drought Tolerance by Regulating the Development of Trichomes and Roots in Poplar.

N6-methyladenosine (m6A) is the most prevalent internal modification present in the mRNAs of all higher eukaryotes, where it is present within both coding and noncoding regions. In mammals, methylation requires the catalysis of a multicomponent m6A methyltransferase complex. Proposed biological functions for m6A modification include pre-mRNA splicing, RNA stability, cell fate regulation, and embryonic development. However, few studies have been conducted on m6A modification in trees. In particular, the regulation mechanism of RNA m6A in Populus development remains to be further elucidated. Here, we show that PtrMTA (Populus trichocarpa methyltransferase) was colocalized with PtrFIP37 in the nucleus. Importantly, the PtrMTA-overexpressing plants significantly increased the density of trichomes and exhibited a more developed root system than that of wild-type controls. Moreover, we found that PtrMTA-overexpressing plants had better tolerance to drought stress. We also found PtrMTA was a component of the m6A methyltransferase complex, which participated in the formation of m6A methylation in poplar. Taken together, these results demonstrate that PtrMTA is involved in drought resistance by affecting the development of trichomes and roots, which will provide new clues for the study of RNA m6A modification and expand our understanding of the epigenetic molecular mechanism in woody plants.


Cloning and Obtaining Transgenic Plants of PtrMTA
To elucidate the potential function of PtrMTA in ligneous plants, we constructed a phylogenetic tree of MTA using DNAMAN software. We found that PtrMTA (Potri.001G085200.1) had 99% homology with PagMTA and PeMTA ( Figure S1). We further found that MTA was highly conserved among a variety of plant species through multiple sequence alignments, such as Populus trichocarpa, 84K poplar, Populus euphratica, Oryza sativa, Prunus persica, Glycine max, Solanum lycopersicum, Fragaria vesca, and Arabidopsis thaliana. (Figure S2).
To further study the potential biological functions of PtrMTA, we cloned the MTA gene from Populus trichocarpa and constructed a 35S overexpression vector. Next, we performed genetic transformation through the leaf disc method [46] ( Figure 1A). The leaves can differentiate into adventitious buds under light culture conditions ( Figure 1B). Through screening for resistance, we initially obtained PtrMTA-overexpressing transgenic plants ( Figure 1C). Every transgenic line was verified by PCR (polymerase chain reaction) and qRT-PCR (quantitative real-time PCR) ( Figure 1D, E). In the literature, these two methods are usually used to verify transgenic plants. Through the verification of the two methods, we obtained transgenic poplar 84K (P. alba X P. glandulosa), i.e., 14  To further study the potential biological functions of PtrMTA, we cloned the MTA gene from Populus trichocarpa and constructed a 35S overexpression vector. Next, we performed genetic transformation through the leaf disc method [46] ( Figure 1A). The leaves can differentiate into adventitious buds under light culture conditions ( Figure 1B). Through screening for resistance, we initially obtained PtrMTA-overexpressing transgenic plants ( Figure 1C). Every transgenic line was verified by PCR (polymerase chain reaction) and qRT-PCR (quantitative real-time PCR) ( Figure 1D, E). In the literature, these two methods are usually used to verify transgenic plants. Through the verification of the two methods, we obtained transgenic poplar 84K (P. alba X P. glandulosa), i.e., 14 lines that overexpressed PtrMTA. . All asterisks denote significant differences: ** p < 0.01.

Subcellular Localization of PtrMTA
Previous studies indicated that Arabidopsis MTA, an m 6 A methyltransferase, was localized in the nucleus [13]. Therefore, PtrMTA, a homolog of MELLT3, might localize to the same cellular

Subcellular Localization of PtrMTA
Previous studies indicated that Arabidopsis MTA, an m 6 A methyltransferase, was localized in the nucleus [13]. Therefore, PtrMTA, a homolog of MELLT3, might localize to the same cellular structures. To determine the subcellular localization of PtrMTA, we constructed a 35S::PtrMTA-GFP fusion protein (green fluorescent protein) expression vector and transiently transformed it into tobacco leaves. Using laser scanning confocal microscopy (LSCM), we found that PtrMTA-GFP was localized in the nucleus (Figure 2A). We further transiently transformed tobacco leaves with the 35S::PtrMTA-GFP fusion protein and 35S::PtrFIP37-mCherry fusion protein. LSCM observation showed that 35S::PtrMTA-GFP and 35S:: PtrFIP37-mCherry were co-localized in the nucleus ( Figure 2B). structures. To determine the subcellular localization of PtrMTA, we constructed a 35S::PtrMTA-GFP fusion protein (green fluorescent protein) expression vector and transiently transformed it into tobacco leaves. Using laser scanning confocal microscopy (LSCM), we found that PtrMTA-GFP was localized in the nucleus (Figure 2A). We further transiently transformed tobacco leaves with the 35S::PtrMTA-GFP fusion protein and 35S::PtrFIP37-mCherry fusion protein. LSCM observation showed that 35S::PtrMTA-GFP and 35S:: PtrFIP37-mCherry were co-localized in the nucleus ( Figure  2B).

Overexpression of PtrMTA Affected Trichome Development
To determine whether PtrMTA affected the development of trichomes in poplar, we selected PtrMTA-overexpressing plants (OE-PtrMTA-14, OE-PtrMTA-10, OE-PtrMTA-6) and wild type (WT) plants for observation of leaf trichomes. Using scanning electron microscopy (SEM), we found that the trichomes of poplar were single-celled, non-glandular trichomes that did not have branches ( Figure 3A). Furthermore, we counted the number of trichomes in the 20 areas (1500 µm * 1000 µm) of leaves in each line. In the transgenic line OE-PtrMTA-14, the number of trichomes was about 13, which was slightly more than in the WT ( Figure 3B). The number of trichomes of transgenic line OE-PtrMTA-10 was about 18 ( Figure 3C). Compared with the WT, the number of trichomes in the transgenic line OE-PtrMTA-6 was significantly increased and was about two times that of the WT ( Figure 3D). More significantly, through statistical analysis, we found the number of trichomes in the PtrMTA-overexpressing plants was significantly greater than that in the WT ( Figure 3E and Table S1). Our results illustrated that overexpression of PtrMTA can increase the density of trichomes in poplar. To determine whether PtrMTA affected the development of trichomes in poplar, we selected PtrMTA-overexpressing plants (OE-PtrMTA-14, OE-PtrMTA-10, OE-PtrMTA-6) and wild type (WT) plants for observation of leaf trichomes. Using scanning electron microscopy (SEM), we found that the trichomes of poplar were single-celled, non-glandular trichomes that did not have branches ( Figure 3A). Furthermore, we counted the number of trichomes in the 20 areas (1500 μm * 1000 μm) of leaves in each line. In the transgenic line OE-PtrMTA-14, the number of trichomes was about 13, which was slightly more than in the WT ( Figure 3B). The number of trichomes of transgenic line OE-PtrMTA-10 was about 18 ( Figure 3C). Compared with the WT, the number of trichomes in the transgenic line OE-PtrMTA-6 was significantly increased and was about two times that of the WT ( Figure 3D). More significantly, through statistical analysis, we found the number of trichomes in the PtrMTA-overexpressing plants was significantly greater than that in the WT ( Figure 3E and Table  S1). Our results illustrated that overexpression of PtrMTA can increase the density of trichomes in poplar. . All asterisks denote significant differences: **p < 0.01.

PtrMTA-Overexpressing Poplar had a More Developed Root System
To investigate whether PtrMTA affected root development in poplar, we selected WT, OE-PtrMTA-14, OE-PtrMTA-10, and OE-PtrMTA-6 seedlings aged 20 days and potted the seedlings for 50 days. At the seedling stage, the PtrMTA-overexpressing plants showed a more developed root system than that of the WT ( Figure 4A). The root length of PtrMTA-overexpressing plants was significantly longer than that of the WT ( Figure 4B). In addition, the number of root tips in the PtrMTA-overexpressing plants was about five, which was also significantly higher than that of the WT ( Figure 4C). Moreover, we examined the PtrMTA-overexpressing poplars that had been grown for 50 days and found that their root systems were significantly more developed than those of the WT ( Figure 4D). Compared with the WT roots, the root length of PtrMTA-overexpressing plants (about 10 cm) was longer ( Figure 4E). The root fresh and dry weights of different lines were measured. We found that the root weight of PtrMTA-overexpressing plants was significantly higher than that of the WT ( Figure 4F). In summary, PtrMTA-overexpressing plants had a more developed root system compared with the WT.
To investigate whether PtrMTA affected root development in poplar, we selected WT, OE-PtrMTA-14, OE-PtrMTA-10, and OE-PtrMTA-6 seedlings aged 20 days and potted the seedlings for 50 days. At the seedling stage, the PtrMTA-overexpressing plants showed a more developed root system than that of the WT ( Figure 4A). The root length of PtrMTA-overexpressing plants was significantly longer than that of the WT ( Figure 4B). In addition, the number of root tips in the PtrMTA-overexpressing plants was about five, which was also significantly higher than that of the WT ( Figure 4C). Moreover, we examined the PtrMTA-overexpressing poplars that had been grown for 50 days and found that their root systems were significantly more developed than those of the WT ( Figure 4D). Compared with the WT roots, the root length of PtrMTA-overexpressing plants (about 10 cm) was longer ( Figure 4E). The root fresh and dry weights of different lines were measured. We found that the root weight of PtrMTA-overexpressing plants was significantly higher than that of the WT ( Figure 4F). In summary, PtrMTA-overexpressing plants had a more developed root system compared with the WT.

OE-PtrMTA Poplar Possessed Drought Tolerance
We further analyzed the approximately 3000 bp promoter sequence upstream of the PtrMTA ATG using the https://www.ncbi.nlm.nih.gov/pubmed/11752327 website. Interestingly, we found that the PtrMTA promoter contained MYB binding sites involved in drought induction and cis-acting elements involved in stress responses (Table S2). Furthermore, to investigate whether PtrMTA increased the drought resistance of poplar, PtrMTA-overexpressing poplar and WT poplar plants were cultured under the same conditions and were then subjected to drought treatment to reduce the soil RWC (relative water content) from 70%. On day 2, some leaves of the WT poplars were slightly withered, while the PtrMTA-overexpressing (OE-PtrMTA) poplars appeared normal ( Figure 5A). On day 5, the leaves of the WT poplars were more seriously withered than those of the PtrMTA-overexpressing poplars ( Figure 5A). that the PtrMTA promoter contained MYB binding sites involved in drought induction and cis-acting elements involved in stress responses (Table S2). Furthermore, to investigate whether PtrMTA increased the drought resistance of poplar, PtrMTA-overexpressing poplar and WT poplar plants were cultured under the same conditions and were then subjected to drought treatment to reduce the soil RWC (relative water content) from 70%. On day 2, some leaves of the WT poplars were slightly withered, while the PtrMTA-overexpressing (OE-PtrMTA) poplars appeared normal ( Figure 5A). On day 5, the leaves of the WT poplars were more seriously withered than those of the PtrMTAoverexpressing poplars ( Figure 5A).
As the malondialdehyde (MDA) content is an effective indicator of the oxidative damage of the cytomembrane, we measured the changes in the MDA content in leaves of WT and PtrMTAoverexpressing poplars under drought stress. The results showed that the MDA content of OE-PtrMTA poplars was lower than that of WT poplar under drought stress ( Figure 5B). Superoxide dismutase (SOD) is an important antioxidant that removes reactive oxygen species. Therefore, we tested the activities of SOD in OE-PtrMTA and WT poplars before and after drought treatment. The results showed that the activity of SOD was significantly higher in OE-PtrMTA poplars than in the WT under drought stress ( Figure 5C). Except that the soil RWC remained at 70%, the control plants were maintained under the same culture conditions, and no significant differences in stem diameter and plant height were found between OE-PtrMTA poplars and the WT ( Figure S3). Therefore, we can conclude that OE-PtrMTA transgenic poplars had better tolerance to drought stress.

PtrMTA Affected the Level of m 6 A
To study whether PtrMTA affected the level of m 6 A, we selected WT, OE-PtrMTA-14, OE-PtrMTA-10, and OE-PtrMTA-6 seedlings aged 30 days and extracted RNA to measure the level of As the malondialdehyde (MDA) content is an effective indicator of the oxidative damage of the cytomembrane, we measured the changes in the MDA content in leaves of WT and PtrMTA-overexpressing poplars under drought stress. The results showed that the MDA content of OE-PtrMTA poplars was lower than that of WT poplar under drought stress ( Figure 5B). Superoxide dismutase (SOD) is an important antioxidant that removes reactive oxygen species. Therefore, we tested the activities of SOD in OE-PtrMTA and WT poplars before and after drought treatment. The results showed that the activity of SOD was significantly higher in OE-PtrMTA poplars than in the WT under drought stress ( Figure 5C). Except that the soil RWC remained at 70%, the control plants were maintained under the same culture conditions, and no significant differences in stem diameter and plant height were found between OE-PtrMTA poplars and the WT ( Figure S3). Therefore, we can conclude that OE-PtrMTA transgenic poplars had better tolerance to drought stress.

PtrMTA Affected the Level of m 6 A
To study whether PtrMTA affected the level of m 6 A, we selected WT, OE-PtrMTA-14, OE-PtrMTA-10, and OE-PtrMTA-6 seedlings aged 30 days and extracted RNA to measure the level of m 6 A. Compared to the WT, the level of m 6 A was significantly higher in PtrMTA-overexpressing plants ( Figure 6A). Furthermore, we also detected the m 6 A level in the roots, stems, and leaves of the WT and overexpressing transgenic seedlings. Importantly, we found that the level of m 6 A in the roots of PtrMTA-overexpressing plants was higher than that of the WT, and the m 6 A content in the roots of OE-PtrMTA-6 was about twice as high as that of the WT ( Figure 6B). It is worth mentioning that the level of m 6 A in stems of PtrMTA-overexpressing plants was much higher than that of the WT ( Figure 6C). In addition, the m 6 A content in leaves was lower than that in the roots and stems, but the m 6 A content in PtrMTA-overexpressing plants was still significantly higher than that in the WT ( Figure 6D). These results indicate that PtrMTA is a component of m 6 Figure 6A). Furthermore, we also detected the m 6 A level in the roots, stems, and leaves of the WT and overexpressing transgenic seedlings. Importantly, we found that the level of m 6 A in the roots of PtrMTA-overexpressing plants was higher than that of the WT, and the m 6 A content in the roots of OE-PtrMTA-6 was about twice as high as that of the WT ( Figure 6B). It is worth mentioning that the level of m 6 A in stems of PtrMTA-overexpressing plants was much higher than that of the WT ( Figure 6C). In addition, the m 6 A content in leaves was lower than that in the roots and stems, but the m 6 A content in PtrMTA-overexpressing plants was still significantly higher than that in the WT ( Figure 6D). These results indicate that PtrMTA is a component of m 6 A methyltransferase and affects the level of m 6 A.

Discussion
RNA m 6 A is a conserved and abundant mRNA modification that affects many aspects of RNA metabolism [34]. In 2015, Chen et al. found that when the level of m 6 A modification was lowered. It was found to impede cell reprogramming seriously [47]. In mammals, loss of function in the core component of the m 6 A methyltransferase complex, including METTL3 and WTAP, leads to embryonic death and reduces the levels of m 6 A [48,49]. In Arabidopsis, AtMTA (a homolog of METTL3) and AtFIP37 (a homolog of WTAP) are localized in the nucleus [13]. Moreover, when the fluorescent fusion proteins MTA-YFP and FIP37-CFP were transiently transformed into Allium cepa epidermal cells, it was found that the two proteins were localized to the same nuclear position [13]. Therefore, AtMTA and AtFIP37 were co-localized in the nucleus. Importantly, AtMTA and AtFIP37 are critical for early embryonic development [13,40,41]. T-DNA knockout of MTA or FIP37 led to the stagnation of embryonic development during the global stage and also reduced levels of m 6 A in different tissues [13,40]. In Arabidopsis, the demethylases include an AlkB family of 13 homologous

Discussion
RNA m 6 A is a conserved and abundant mRNA modification that affects many aspects of RNA metabolism [34]. In 2015, Chen et al. found that when the level of m 6 A modification was lowered. It was found to impede cell reprogramming seriously [47]. In mammals, loss of function in the core component of the m 6 A methyltransferase complex, including METTL3 and WTAP, leads to embryonic death and reduces the levels of m 6 A [48,49]. In Arabidopsis, AtMTA (a homolog of METTL3) and AtFIP37 (a homolog of WTAP) are localized in the nucleus [13]. Moreover, when the fluorescent fusion proteins MTA-YFP and FIP37-CFP were transiently transformed into Allium cepa epidermal cells, it was found that the two proteins were localized to the same nuclear position [13]. Therefore, AtMTA and AtFIP37 were co-localized in the nucleus. Importantly, AtMTA and AtFIP37 are critical for early embryonic development [13,40,41]. T-DNA knockout of MTA or FIP37 led to the stagnation of embryonic development during the global stage and also reduced levels of m 6 A in different tissues [13,40]. In Arabidopsis, the demethylases include an AlkB family of 13 homologous proteins [29,50,51]. ALKBH9B participates in defense against viral infections [50]. ALKBH10B is an mRNA m 6 A demethylase that affects vegetative growth and floral transition [51]. In our study, we transiently transformed the 35S::PtrMTA-GFP fusion protein into tobacco leaves and found that PtrMTA targeted the nuclear spots. In addition, we simultaneously transferred 35S::PtrMTA-GFP and 35S::PtrFIP37-mCherry into tobacco leaves and found that the two proteins were co-localized in the nucleus. In addition, we found that the level of m 6 A in PtrMTA-overexpressing plants increased significantly compared to the WT. This indicated that PtrMTA is a component of m 6 A methyltransferase and can affect the level of m 6 A.
Trichomes and roots are often associated with plant resistance [52]. A trichome is a hairy structural appendage extending from the epidermal tissues of aerial parts; trichomes include glandular and non-glandular hairs, linear trichomes, and branched trichomes [53]. Trichomes have a variety of biological functions, such as reducing transpiration. Moreover, trichomes are involved in defense against adverse conditions, such as cold, drought, high temperature, strong radiation, as well as disease infestation [53]. Glandular hairs can secrete chemicals, such as alkaloids, aromatic oils, and resins, to defend against biotic and abiotic stresses [53]. Non-glandular hairs play an important role in resisting drought and mechanical damage [54]. The root system is the organ through which a plant directly senses water signals from the soil and absorbs soil water [55]. The root length and depth directly affect the absorption and utilization of soil water and nutrients [56]. In addition, the root tip number affects the range of the root system depth and breadth in space and also affects the ability to absorb water [57]. Root length, root tip number, and root weight are all closely related to plant drought resistance [58]. In this study, the results of SEM showed that overexpression of PtrMTA significantly increased the density of trichomes in poplar. In addition, we found that the root length, root tip number, and root weight of PtrMTA-overexpressing plants were significantly increased compared to WT plants.
These results indicate that PtrMTA is closely related to the development of trichomes and plant roots. More importantly, we found that PtrMTA-overexpressing plants had stronger drought resistance than the WT under drought treatment. We speculate that PtrMTA may be involved in the regulation of drought resistance.
The balance between the generation and elimination of reactive oxygen species in a cell would be broken under stress conditions [59]. Excess reactive oxygen species first attack the cell membrane system, thus destroying the cell membrane [60,61]. MDA is the final product of cytoplasmic membrane peroxidation. MDA can combine with proteins on the cell membrane to inactivate it, thereby destroying the structure and function of the biological membranes [60]. Plants have a membrane protection system that can remove excess reactive oxygen species from cells [60,62]. This membrane protective system is an antioxidant system composed of many enzymes, such as SOD [62]. SOD can protect the biological membrane system by removing harmful reactive oxygen species [63]. Therefore, the MDA content and SOD activity can reflect the resistance of plants to a certain extent [59]. In this study, we found that PtrMTA-overexpressing plants were more resistant to drought stress compared to WT plants. Therefore, we measured the MDA content and SOD activities of PtrMTA-overexpressing plants before and after drought treatment. After drought treatment, the MDA content was significantly lower than that of the WT, which may be related to the strong drought tolerance of PtrMTA-overexpressing plants.
In addition, the SOD activity of PtrMTA-overexpressing plants was obviously higher than that of the WT, which may be related to resistance to stress.

Plant Materials and Plant Growth Conditions
In this study, the experimental materials-Tobacco, Populus trichocarpa, and 84K poplar-were planted in a greenhouse of the Beijing Forestry University at 25 • C under a 16 h light (240 µmol m -2 s -1 ) and 8 h dark. Potted Populus trichocarpa were watered based on evaporation demand and were regularly irrigated with water.

cDNA Cloning and Plasmid Construction
Whole seedlings of poplar were collected and stored in liquid nitrogen. The total RNA was isolated using an RNA Isolation Kit (TIANGEN, Beijing, China). In the last step, potentially contaminating DNA was removed by treatment with DNase I. Four micrograms of total RNA was used for the reverse transcription reaction with a First-Strand cDNA Synthesis Kit (TIANGEN, Beijing, China) according to the protocol. The cDNA of PtrMTA and PtrFIP37 was cloned by PCR. The DNA polymerase we used was ex-Taq (Takara, Japan). The initial denaturing time was 95 • C for 5 min, followed by 35 cycles at 95 • C for 30 s, 56 • C for 30 s and 72 • C for 2.5 min, with a final extension at 72 • C for 5 min. Furthermore, the PtrMTA sequence was inserted between the KpnI (Takara, Japan) and BamHI (Takara, Japan) sites of the pCAMBIA-2301 and pCAMBIA-2301-GFP vectors. We transformed the constructed pCAMBIA-2301-MTA and pCAMBIA-2301-MTA-GFP vectors into Agrobacterium tumefaciens GV3101. The transformed Agrobacterium was then applied to Luria-Bertani (LB) solid plates with 50 mg/L kanamycin and 50 mg/L rifampicin antibiotics and incubated at 28 • C for 2-3 days. A single clone was picked and subsequently verified by PCR. The PtrFIP37 sequence was inserted between the KpnI (Takara, Japan) and XbaI (Takara, Japan) sites of the pCAMBIA-1300-mCherry vector. pCAMBIA-1300-FIP37-mCherry Agrobacterium transformation and PCR were performed using the same method as above. The primers are in Table S3.

qRT-PCR (Quantitative Real-Time Polymerase Chain Reaction) Analysis
RNA from transgenic plants and WT plants was extracted using an Adelaide RNA Kit (Adelaide, China). The RNA was reverse transcribed into cDNA according to the method provided by the One-Step gDNA Removal and cDNA Synthesis SuperMix (TransGen Biotech, China). qRT-PCR was performed using SYBR Green Mix (Takara, Japan) in an optical 96-well plate. As an internal control, 18S rRNA was used in qRT-PCR. The qRT-PCR instrument we use was Bio-Rad-CFX96 Touch, America. The 2-DDCt analysis method was used to analyze the relative expression of PtrMTA. The initial denaturing time was 95 • C for 2 min, followed by 40 cycles at 95 • C for 5 s, 60 • C for 5 s, and 95 • C for 5 s. The primers are in Table S3.

Phylogenetic and Domain Analyses of PtrMTA
The homologous amino acid sequences of MTA in different plant species were obtained from the NCBI database (https://www.ncbi.nlm.nih.gov/) and Phytozome database (https://phytozome.jgi. doe.gov/pz/portal.html). In NCBI, PtrMTA homologous sequences were obtained by the BLASTn method (Table S4). In addition, the phylogenetic tree of PtrMTA was simply created using DNAMAN. DNAMAN was used for comparison of PtrMTA amino acid sequences and confirmed conserved structures.

Genetic Transformation of Poplar and Identification of Transgenic Seedlings
A single colony of Agrobacterium was shaken in LB medium at 28 • C overnight until the culture density reached an OD 600 of 0.6-0.8. Agrobacterium cells were harvested by centrifugation at 5000 rpm for 5-8 min. The harvested cells were suspended in a 1 2 -strength MS solution (pH 5.8) containing 5% (w/v) sucrose and acetosyringone (100 µM) as a conversion solution (Song et al., 2019). The pretreated leaves were placed in the transformation solution and soaked for about 15 min, during which time shaking was required every 3 min. The infested leaves were placed on sterile filter paper, and extra Agrobacterium was aspirated. The leaves were placed in 1 2 -strength MS medium for 3-5 days under dark culture. The dark-treated leaves were placed in a medium containing kanamycin for differentiation culture. When the shoots grew to 1 cm after 4-5 weeks, they were cut off and transferred to a resistant screening medium. The obtained transgenic plants were screened for DNA identification and qRT-PCR identification. The primers are in Table S3.

Subcellular Localization
The Agrobacterium was collected by centrifugation for 5 min at 5000 g. The cultures were resuspended in a medium containing 5 mM MES, 100 mM MgCl 2 (pH 5.6), and 0.1 mM acetylsyringone [64]. The Agrobacterium cells were then injected into the abaxial surfaces of the tobacco leaves by using a needleless syringe. The plants were then cultured in a dark environment for 24 h. The transiently transformed tobacco leaves were fixed in sterile water and imaged with a Leica 20X or 40X water immersion objective, white light laser, and a hybrid detector with a resolution of 1024 × 1024 pixels. All images shown in this article are original images taken from the Leica SP8.

SEM Analysis of Poplar Leaves Trichomes
SEM (Hitachi S4800, Japan) was used to study the trichome density of 40-day-old WT and PtrMTA-overexpressing poplar leaves. We selected potted seedlings that had grown for 40 days and picked the sixteenth leaf of each plant. Then we cut out the middle area near the main vein of the leaf. Poplar leaves were trimmed to a size of 1 cm 2 . Samples were then immediately drop fixed in formalin-acetic acid-alcohol (FAA was composed of 5 mL 38% formaldehyde, 5 mL acetic acid and 90 mL 70% alcohol). The samples were dehydrated successively with 30%, 50%, 60%, 70%, 80%, and 90% ethanol for 30-40 min and were dehydrated overnight with anhydrous ethanol. A dryer was used for the dehydrated samples (Leica EM CPD300, Germany). Dry samples were sputter-coated with gold (110 s). Finally, the samples were imaged using a scanning electron microscope. We counted the number of trichomes in the 20 areas (1500 µm * 1000 µm) of leaves in WT and transgenic plants.

Drought-Tolerance Experiments
Three plants of the 3 transgenic lines and one WT poplar plant were grown in a greenhouse at 25 • C during the day and 22 • C at night. The greenhouse was illuminated for 16 h per day. All plants were grown in suitable pots with a diameter of 13.5 cm. All plants were subjected to a short-term drought treatment in which the soil RWC (relative water content) decreased from 70% [65,66]. The control plants were kept under the same conditions.

Determination of the m 6 A Level in RNA
RNA from transgenic and WT seedlings was extracted using an Adelaide RNA Kit (Adelaide China). The RNA concentrations of different samples were determined using a NanoPhotometer NP80 (Implen, Germany). The m 6 A level was determined using an Aderr kit (Epigentek, America). m 6 A (%) = (SampleOD-NcOD)/S/(PcOD-NcOD)/P*100% (Nc: Negative control; Pc: Positive control; S: RNA sample inputs, ng; P: Positive control inputs, ng).

Conclusions
This study shows that the m 6 A methyltransferase MTA from Populus trichocarpa was localized in the nucleus and co-localized with PtrFIP37 in the nucleus. Moreover, the overexpression of PtrMTA significantly affected the level of m 6 A and increased the density of trichomes. PtrMTA-overexpressing plants had a more developed root system. More importantly, the overexpression of PtrMTA can enhance the drought resistance of poplar. Therefore, our research not only enriched the study of the m 6 A function in Populus but also provided a basis for exploring the epigenetic molecular mechanism in woody plants.
Supplementary Materials: The following are available online at http://www.mdpi.com/1422-0067/21/7/2462/s1, Figure S1. Phylogenetic tree of different MTA protein family members and amino acid sequence alignment of PtrMTA; Figure S2. MTA was highly conserved among various plant species; Figure S3. Growth status of wild type (WT) and PtrMTA-overexpressing plants under normal conditions; Table S1. Statistics on the number of trichomes in the 20 areas (1500 µm * 1000 µm) of leaves in wild type (WT) and transgenic plants; Table S2. Analysis of the PtrMTA promoter functional components; Table S3 Primers used in this study; Table S4 GenBank IDs of the PtrMTA homologs proteins.