The MBF protein homologs sequences from diverse species were retrieved from GenBank database and Phytozome database¹ (Goodstein et al., 2012) with BlastP using PaMBF1c amino acid sequences as a query. Multiple sequence alignment was performed using ClustalW (Thompson et al., 1994; Kumar et al., 2016). Conserved domains in each sequence were identified using NCBI conserved domain finder². The phylogenetic tree was constructed with MEGA7 software (Thompson et al., 1994; Kumar et al., 2016) from the data sets by using the maximum likelihood method based on the JTT matrix-based model. The initial tree for the heuristic search was obtained by applying the neighbor-joining method to a matrix of pair-wise distances, estimated using a JTT model. Supports for internal branches were tested by bootstrap analyses of 1000 replications.

Polytrichastrum alpinum (Hedw.) G.L.Sm. samples were collected from the King Sejong Antarctic station (62°14′29′′S; 58°44′18′′W), at the Barton Peninsula of King George Island. In vitro culture was carried out on BCD solid media (Ashton and Cove, 1977), which are successfully being used in a model moss Physcomitrella patens culture (Du et al., 2016), in a growth room operating at 23°C with 16-h light/8-h dark light cycle with a light intensity of 150 μmol m^-2S^-1.

The plates were placed in a growth chamber that operates at 22 ± 1°C and 70% relative humidity with continuous light (80–100 μmol m^-2S^-1). The seeds planted on soil (Sungro propagation mixture, Canada) were maintained in a controlled growth chamber that operates with 16-h light/8-h dark cycle. All seeds were stratified for at least 2 days at 4°C before being transferred to the growth chamber. When necessary, seedlings raised on MS were transferred to soil pots in a growth chamber at 22 ± 1°C and 50–70% relative humidity programmed with 16-h light/8-h dark light cycle (the light intensity of 80–100 μmol m^-2S^-1).

Polytrichastrum alpinum first strand cDNA was synthesized from total RNA with MMLV reverse transcriptase (Enzynomics, South Korea) and oligo(dT) primer. Full-length PaMBF1c coding sequences were amplified using gene specific primers (Supplementary Table S1). The resulting PCR products were cloned into the pENTR/D/TOPO entry vector (Invitrogen, United States) and sequenced using M-13 primers. After sequencing confirmation, the entry plasmids were LR-recombined with the gateway compatible binary destination vector, pMDC32 (Curtis and Grossniklaus, 2003), which resulted in the overexpression construct, pMDC32-35s:PaMBF1c. The construct was then transferred into Agrobacterium tumefaciens strain GV3101 via electroporation. Agrobacterium-mediated transformation of Arabidopsis thaliana (Col-0) with the PaMBF1c gene was performed through the floral dipping method (Clough and Bent, 1998). For selection of PaMBF1c-OE lines, T1 seeds were harvested from floral-dipped plants and selected on MS/agar plates containing hygromycin (25 μg/mL). Using hygromycin selection and transgene detection, homozygotes for the PaMBF1c transgene of two independent events were selected at T2 generation after segregation analysis at T3 generation (PaMBF1c-OE1 and PaMBF1c-OE2). Hygromycin resistance from each line at T2 generation was segregated to a 3:1 ratio of resistant-to-sensitive, indicating a single locus of insertion.

Polytrichastrum alpinum gametophores were transferred onto fresh agar plates of BCD medium containing mannitol (150, 300 mM) or NaCl (75, 150 mM), respectively, and incubated at 15°C for 6 h. Heat treatment was carried out by transferring colonies grown at 15°C to chambers of 37 or 42°C for 2 h.

For Arabidopsis thaliana germination experiments, at least 100–120 seeds of each genotype were planted in a media with or without stress agents and germination trends were recorded from the day after planting until all the seeds in control plates were germinated. The number of germinated seeds was expressed as the percentage of total number of seeds plated. For root growth experiments, the seeds were planted in vertical MS plates with 0.6% gelrite and allowed to grow for 4 days. Seedlings with a 1–1.5 cm long root were transferred onto a second MS vertical plate supplemented with different concentrations of salts or stress agents. For survival index under NaCl and LiCl, the shoot phenotypes of seedlings were observed every day after transferring seedlings from the control medium. Hypocotyl elongation under heat stress was assessed as previously described (Hong and Vierling, 2000). The malondialdehyde (MDA), chlorophyll, and anthocyanin content in control and stress treated plants were also determined as previously described (Heath and Packer, 1968; Lichtenthaler, 1987; Neff and Chory, 1998).

Fully expanded fourth or fifth rosette leaf with petiole from 3-week-old seedlings was placed in a 15 mL test tube containing 100 μL of deionized water and placed at 0°C in a refrigerated circulating bath (Gaon Science Instrument, South Korea). The remaining steps were carried out according to methods described previously (Lee et al., 2002). Conductivity was measured using a conductivity meter (Control Company, United States) before and after autoclaving, and electrolyte leakage was expressed as a percentage of conductivity before autoclaving over conductivity after autoclaving.

Total RNAs were isolated from plant materials using an RNA purification kit (NanoHelix, South Korea) and treated with RNase free DNase I (Qiagen, Germany). For semi-quantitative one step RT-PCR, total RNAs were added to the Hi-pure one step RT-PCR master mix (Genepole, South Korea) and the PCR reaction was performed according to manufacturer instructions. Protein phosphatase 2A (At1g13320) was used as an internal control. Quantitative real-time PCR was performed using KAPA SYBR FAST qPCR kit (Kapa Biosystems, United States) according to manufacturer instructions and run on the ABI 7500 system (Applied Biosystems, United States). The relative expression of PaMBF1c in P. alpinum was calculated by normalizing expression values with those of the housekeeping gene PaTubulin. For the Arabidopsis qRT experiment, AtClathrin gene (At4g24550) was used as an internal control. The ΔΔCt method was adapted to calculate relative gene expression (Livak and Schmittgen, 2001). The primer pairs used for amplification are shown in Supplementary Table S1.

RNA-sequencing was carried out using total RNA from 2-week-old WT, two independent PaMBF1c-OE lines (two events) and two independent AtMBF1c-OE lines (two events) under normal and salt stressed conditions (300 mM NaCl, 6 h). We used this salt stress condition because 300 mM NaCl treatment for 4–6 h in Arabidopsis brought about maximum expression of a very well characterized stress inducible gene, RD29A (Ishitani et al., 1997). Total RNAs were extracted from at least 25–30 plants of each genotype and treatment using Pure Helix total RNA purification kit (NanoHelix, South Korea) and RNase free DNase I (Qiagen, Germany). Three different biological replicates were prepared. The integrity and concentration of RNA was determined using Bioanalyzer (RIN > 6) and Qubit^® RNA BR assay kit (Life Technologies, United States). To construct the sequencing library, 1.5 μg of total RNA of each sample was used as input for the TruSeq RNA sample prep kit v2 (Illumina, United States). The libraries were validated and quantified by Bioanalyzer and qPCR quantification method, and then multiplexed with equal ratio and loaded on the flowcell of the Illumina MiSeq Reagent Kit v3 (150 cycles). Afterward, sequencing was performed on a MiSeq Sequencer system (Illumina, United States) and total 3 Gb (40M paired end reads) of sequencing data was generated (Q₃₀ > 98%). The RNA-Seq data have been deposited to the Sequence Read Archive (SRA³) under accession number SRP110226.

Basically all analyses were performed using the CLC Genomics Workbench v7 module (Qiagen, Germany). After quality and adapter trimming, raw reads were mapped to the Arabidopsis gene model annotation file of the Gene Ontology Consortium (released at August 8th, 2014). The expression values were measured in FPKM (Fragments per Kilobase of exon model per Million mapped reads) normalized values in gene level (Mortazavi et al., 2008). For statistical analysis, t-test and Baggerley’s test were performed using original and normalized read counts, and several relevant values for analysis (p-value, FDR corrected p-value, test-statistic, etc.) were calculated using the “multi-group comparison” option of the program. Through statistical analysis, differentially expressed genes were determined from a cutoff value (p-value < 0.05, corrected p-value of FDR < 0.05 and absolute value of fold change > 1.5) from pairwise comparison of normalized FPKM values between samples. Gene ontology (GO) enrichment analysis was performed using the PANTHER overrepresentation test (PANTHER version 10⁴) (Mi et al., 2013, 2016). Arabidopsis thaliana was selected as a reference organism with default settings and Bonferroni correction for multiple testing was used (p < 0.05).

All statistical comparisons between variants were determined by ANOVA (analysis of variance) and least significant differences (LSD) between variants were calculated using Statistix 8.1 computation software. Statistically significant mean values were denoted as ^∗ (p-value ≤ 0.05).

The Polytrichastrum alpinum DNA sequence encoding multi-protein bridging factor 1c (MBF1c) protein was retrieved from our unpublished transcriptome data, based on its sequence homology with AtMBF1c. The gene was designated PaMBF1c and its sequence was submitted to GenBank (Accession number, KM978992). The total length of the 432 nucleotide coding sequence comprise of 143 deduced amino acids with an estimated molecular mass of 15.7 kDa and an isoelectric point of 10.08 predicted by ExPasy bioinformatics tools for protein structure analysis⁵. It contains two distinctive conserved domains – an MBF1 domain at the N-terminal region and a helix-turn-helix (HTH) domain at the C-terminal region (Figure 1A). PaMBF1c shares 86% identity with its homolog from Physcomitrella patens (XP_001771731), 64% with Medicago truncatula (AES76734), 66% with Arabidopsis thaliana (NP_189093), 59% with Triticum aestivum (ACU43593), and 59% identity with Oryza sativa (NP_001057974). Sequence alignment analysis of PaMBF1c also suggests the presence of glutamic acid (E) at the 115th amino acid position in the HTH domain (Figure 1A). This particular amino acid is reportedly conserved across the plant species and is essential for the binding to TATA-box binding protein (Liu et al., 2007).

The PaMBF1c protein sequence was queried against proteome datasets of various plants and chlorophytes in Phytozome⁶ to better understand the phylogenetic relations of MBF1 family proteins in plants. The phylogenetic analysis demonstrated a clear divergence of MBF1c proteins from other MBF1 family proteins (MBF1a/b) (Figure 1B). Interestingly, algae have only one MBF1 gene while most land plants contain at least two types MBF1 genes (Figure 1B), suggesting a gene duplication event early in the evolution of land plant. Within the MBF1c clade, MBF1c orthologs of all non-vascular plant species were clustered together away from those of vascular plant (Figure 1B) implying MBF1c differentiation among land plants.

A number of MBF1 group genes in various species were reported to be differentially induced by various abiotic stresses (Rizhsky et al., 2002; Tsuda and Yamazaki, 2004; Kim et al., 2007). We examined PaMBF1c transcript abundance under salt, osmotic, and heat stress conditions in P. alpinum. To this end, quantitative real-time PCR (qRT-PCR) analysis was carried out using RNA isolated from P. alpinum gametophores treated with NaCl (75 mM or 150 mM for 6 h), mannitol (150 or 300 mM for 6 h), or high temperature (37 or 42°C for 2 h). The results revealed that PaMBF1c transcript levels were increased in response to various abiotic stress treatments (Figure 2); thus, we concluded that PaMBF1c is a heat, salt, and osmotic stress-responsive gene in P. alpinum.

To further investigate the functional roles of PaMBF1c in plants, we generated transgenic lines overexpressing the PaMBF1c gene under control of the 35S cauliflower mosaic virus promoter (35S:PaMBF1c). Using hygromycin resistance selection and the presence of the 35S:PaMBF1c transgene, we selected stable homozygous transgenic lines for 35S:PaMBF1c in the T4 generation (hereafter referred to as PaMBF1c-OE lines, where OE stands for overexpressing). Semi-quantitative RT-PCR analysis for PaMBF1c transcripts indeed confirmed that PaMBF1c-OE lines overexpressed PaMBF1c, as PaMBF1c transcripts were highly accumulated in transgenic plants (Figure 3A). With confirmed lines, we examined the growth and development of PaMBF1c-OE plants. Two-week old PaMBF1c-OE plants displayed better growth and development than WT plants under normal growth conditions. Fresh weight of 2-week-old transgenic plants was about 16–17 mg/seedling while that of WT was about 11 mg/seedling (Figures 3B,C). In addition, PaMBF1c-OE lines bolted 1–2 days earlier than did WT (Figures 3D,E).

To study the responses of PaMBF1c-OE plants to abiotic stress conditions, two PaMBF1c-OE lines from two independent transgenic events were subjected to different stress treatments. Germination of each line was tested on MS medium containing different kinds of stress agents: salt stress (200 mM NaCl), ionic stress (15 mM LiCl) or osmotic stress (200 mM mannitol). On the control medium, all lines germinated with percentages of 95–100% after 2–3 days of planting (Supplementary Figure S1). On mediums supplemented with 200 mM NaCl, the germination ratio of WT was decreased to 35.8%, whereas PaMBF1c-OE lines displayed 72.9–86.4% germination ratios (Figure 4A). Similarly, PaMBF1c-OE lines displayed 94–94.5% germination ratios on 15 mM LiCl MS medium, whereas WT germination ratio was reduced to 62.2% (Figure 4A). In the medium containing 200 mM mannitol, PaMBF1c-OE lines displayed slightly better germination than in WT (Figure 4A). Taken together, the results showed that PaMBF1c-OE lines germinate better than WT under salt, ionic, and osmotic stress.

Root growth is affected by various stress conditions and is often considered an index for stress sensitivity (Lee et al., 2006). Thus, we examined levels of stress tolerance by analyzing root growth of PaMBF1c-OE lines. Similarly sized seedlings (3- to 4-day-old) grown under normal condition were transferred to MS medium for salt stress (150 mM NaCl), ionic stress (15 mM LiCl), or osmotic stress (200 mM mannitol); and root elongation was measured 6 days after transfer. All lines showed similar root lengths in the control MS media (Supplementary Figure S2); however, relative root elongation in PaMBF1c-OE lines was significantly greater in the presence of 150 mM NaCl (48–48.7% vs. 34.2%), 200 mM mannitol (51.5–54.8% vs. 38.1%), or 15 mM LiCl (52.3–53.3% vs. 37%) when compared to WT (Figure 4B). Additionally, seedling survival was examined under stress conditions. Three- to four-day-old seedlings grown under normal conditions were transferred to MS medium supplemented with 200 mM NaCl for salt stress or 20 mM LiCl for ionic stress. For heat stress, 8-day-old seedlings grown on MS plates were heat treated at 45°C (Figure 4C and Supplementary Figures S3A,B). The PaMBF1c-OE lines showed significantly better survival than did WT under the presence of 200 mM NaCl (41–69% vs. 16.6%), 20 mM LiCl (73–86% vs. 40%), or heat (53–62% vs. 15%) (Figure 4C and Supplementary Figures S3C–E). We also investigated responses to cold stress in PaMBF1c-OE lines by measuring ion leakage after freezing stress treatment. At -6°C, both WT and PaMBF1c-OE showed similar ion leakage levels (Supplementary Figure S4); however, PaMBF1c-OE lines showed lower ion leakage at -9 and -12°C than WT, suggesting enhanced freezing tolerance in PaMBF1c-OE lines (Supplementary Figure S4).

We examined in detail the tolerance phenotypes of PaMBF1c-OE under salt stress by analyzing chlorophyll content, lipid peroxidation level, and anthocyanin accumulation. Chlorophyll (Chl) degradation is among the manifestations caused by salt and osmotic stress (Claeys et al., 2014; Alavilli et al., 2016). To further examine the effect of salt stress, we measured Chl content in WT and PaMBF1c-OE lines with or without salt stress. No notable differences were observed between WT and PaMBF1c-OE grown under control condition (Figures 5A,B); however, stress treatment with 75 mM or 150 mM NaCl showed that PaMBF1c-OE lines maintained higher Chl content than that of WT, as also evidenced by visual phenotypes (i.e., leaf bleaching) (Figures 5A,B). Abiotic stresses usually lead to lipid peroxidation of cell membranes which can cause irreversible damage to its functionality. MDA is considered an indicator of lipid peroxidation level (Heath and Packer, 1968). After salt stress (120 mM NaCl), the MDA content of PaMBF1c-OE lines was significantly lower in than that of WT, which suggests lower lipid peroxidation in PaMBF1c-OE (Figure 5C). Anthocyanin accumulation is known to be associated with improved drought and salt stress tolerance (Oh et al., 2011; Nakabayashi et al., 2014). Accordingly, 10-day-old PaMBF1c-OE lines grown under salt stress medium (150 mM NaCl) exhibited higher anthocyanin accumulation than WT (Figure 5D and Supplementary Figure S5). Taken all together, these results clearly demonstrated the higher salt-stress tolerance of PaMBF1c-OE lines than WT.

AtMBF1c overexpression in Arabidopsis resulted in improved heat and osmotic stress tolerance (Suzuki et al., 2005). AtMBF1c-OE lines appeared to exhibit tolerance at only low levels of salt stress (50 mM NaCl) when compared to WT (Suzuki et al., 2005). Thus, we compared the stress tolerance phenotypes of our PaMBF1c-OE lines and the AtMBF1c-OE lines under our experimental conditions. We first assessed heat tolerance with two independent lines for each overexpressor. The hypocotyl elongation assay makes use of the characteristic heat inhibition of hypocotyl elongation (Hong and Vierling, 2000). PaMBF1c-OE lines showed higher hypocotyl elongation than did WT at both 41 and 43°C (Figure 6A), suggesting hyposensitivity to heat stress in PaMBF1c-OE. Similarly, AtMBF1c-OE lines displayed higher hypocotyl elongation when compared to WT in agreement with previous reports (Suzuki et al., 2005). Hypocotyl lengths of both OE lines were comparable after heat treatment suggesting that both PaMBF1c and AtMBF1c have similar levels of increased heat tolerance (Figure 6A).

We further extended the comparison to other types of stress. We treated PaMBF1c-OE and AtMBF1c-OE lines with salt (100–200 mM NaCl), ionic (15 mM LiCl) and osmotic stress (200 mM mannitol) at germination and post-germination stages. Under salt stress, PaMBF1c-OE lines consistently displayed higher germination ratios than WT, whereas AtMBF1c-OE lines germinated at almost the same ratio as WT (Figure 6B). In addition, the germination of PaMBF1c-OE lines was better than those of WT and AtMBF1c-OE lines under ionic stress (15 mM LiCl) (Figure 6B). However, germination of each line at osmotic stress (200 and 400 mM mannitol) were largely comparable, with one PaMBF1c-OE line demonstrating a slightly higher germination ratio than the other genotypes (Figure 6B and Supplementary Figure S6). These results suggest that PaMBF1c-OE performs better in terms of germination than WT or even AtMBF1c-OE, particularly under salt and ionic stress.

Root growth comparisons also revealed that the root of PaMBF1c-OE lines grew longer under salt and ionic stresses than that of both WT and AtMBF1c-OE (Figure 6C). AtMBF1c-OE lines exhibited root growth similar to that of WT under salt and ionic stresses (Figure 6C). Under osmotic stress (200 mM mannitol), root elongation was similar between PaMBF1-OE and AtMBF1c-OE lines but was significantly enhanced in comparison with WT (Figure 6C). Improved osmotic stress tolerance in the AtMBF1c-OE line was consistent with previous reports (Suzuki et al., 2005). These results revealed that except for germination under osmotic stress, PaMBF1c-OE lines showed enhanced tolerance at both germination and post-germination stage under all tested conditions including stress conditions in which AtMBF1c did not show increased tolerance. Conclusively, these results implicate PaMBF1c as an MBF1c allele which affords better coping against multiple stresses than AtMBF1c.

In order to understand the gene expression patterns of PaMBF1c-OE lines in salt stress response and identify genes important for PaMBF1c-specific stress tolerance, we carried out RNA-Seq analysis. After sequencing with total RNA from salt-treated samples, we first defined salt-regulated genes in WT as the genes with statistically altered expression in WT under salt stress. A total of 5,360 genes were salt-regulated with 1,845 genes up-regulated and 3,515 down-regulated (Figure 7A and Supplementary Table S2). Among these salt-regulated genes, we then identified the up- or down-regulated genes in PaMBF1c-OE and AtMBF1c-OE lines under normal condition. We found that 10 genes in the PaMBF1c-OE line were already up-regulated, which included three genes that were also already up-regulated in the AtMBF1c-OE line (Table 1 and Figure 7B). We validated the expression of several genes in Table 1 by real-time PCR and found that gene expression of all tested genes was consistent with RNA-Seq results (Supplementary Figure S7).

None of the salt down-regulated genes were found already down-regulated in either PaMBF1c or AtMBF1c overexpressing lines (data not shown). The already up-regulated genes, particularly PaMBF1c-OE specific genes, might elicit quicker response to salt stress in PaMBF1c-OE lines, resulting in enhanced salt tolerance in overexpressors. Among the already up-regulated genes in PaMBF1c-OE, the FeSOD1 gene (At4g25100) was identified in a transcriptome comparison study by Taji et al. (2004) in which 77 genes were identified that showed higher expression under normal conditions in a halophyte Thellungiella halophila than in Arabidopsis. This suggests that early establishment of FeSOD1-mediated detoxification of reactive oxygen species (ROS) in PaMBF1c-OE might be an important step in the enhanced tolerance to salt stress which usually causes secondary oxidative stress.

We also compared profiles of salt-regulated genes from WT, PaMBF1c-OE and AtMBF1c-OE lines. The two overexpressors showed similar numbers of genes with altered expression under salt stress as did wild type (1,808 and 1,826up-regulated genes in PaMBF1c-OE and AtMBF1c-OE, respectively; 3,394 and 3,472 down-regulated genes in PaMBF1c-OE and AtMBF1c-OE, respectively) (Figure 7A and Supplementary Table S3). Expression of the majority of these genes was commonly regulated by salt stress in all genotypes; i.e., WT, PaMBF1c-OE, and AtMBF1c-OE lines each possessed the same set of 1,430 up-regulated and 2,475 down-regulated genes out of a total of 1845, 1,808 and 1,826 up-regulated and 3515, 3,394 and 3,472 down-regulated genes in WT, PaMBF1c-OE and AtMBF1c-OE, respectively (Figure 7A and Supplementary Table S3). At least in part, the genes with altered expression unique to PaMBF1c-OE might contribute to the enhanced salt tolerance of the PaMBF1c-OE lines.

Gene ontology enrichment analysis within the category ‘biological processes’ was conducted with salt-regulated genes using the whole Arabidopsis genome set as reference. The overall distribution pattern of salt-regulated genes in GO terms was well conserved among WT, PaMBF1c-OE, and AtMBF1c-OE lines under salt stress conditions (Figure 8A). The terms RNA metabolic process, response to stress, response to stimulus, DNA metabolic process, and carbohydrate metabolic process were significantly over- or under-represented in all three plants compared to the Arabidopsis genome (p-value of the Bonferroni correction for multiple testing < 0.05) (Figure 8A). The term protein metabolic process was significantly over-represented only in WT plants but not in PaMBF1c-OE or AtMBF1c-OE, suggesting that protein metabolism is a major biological process affected in WT plants under salt stress.

To increase the resolution of GO enrichment analysis, we divided salt-regulated genes into salt up-regulated and salt down-regulated groups and conducted separate GO enrichment analyses for each with the whole Arabidopsis genome as reference. For salt up-regulated genes, GO terms in biological processes that were significantly over-represented in PaMBF1c-OE plants were ‘lipid metabolic process,’ ‘ion transport,’ and ‘cellular amino acid biosynthetic process’ (Figure 8B). The corresponding salt up-regulated genes included in these terms were listed in Supplementary Table S4. Given the possibility that improved salt tolerance in PaMBF1c-OE might be mainly due to enhanced ionic stress tolerance, we manually surveyed genes for the term ‘ion transport’ that were significantly up-regulated in PaMBF1c-OE only. Notably, genes involved in ATP production and transport (Mitochondrial phosphate transporter [At3g48850] and ADP/ATP carrier 3 [At4g28390]) as well as ATP-dependent Ca²⁺ pumping (ECA1 [At1g07810]) were among the PaMBF1c-OE specific salt up-regulated genes (Supplementary Table S4) (Liang et al., 1997; Wu et al., 2002). Thus, enhanced salt tolerance in PaMBF1c-OE might be caused in part by promotions of ATP synthesis and Ca²⁺ signaling.

For salt down-regulated genes, the GO terms ‘cellular component biogenesis,’ ‘response to stress,’ ‘protein metabolic process,’ and ‘cholesterol metabolic process’ were significantly over-represented in WT plants under salt-stress condition but not so in PaMBF1c-OE or AtMBF1c-OE plants with the exception of ‘cellular component biogenesis’ which was over-represented in both WT and AtMBF1c-OE plants (Figure 8C). Salt down-regulated genes in these GO terms were listed in Supplementary Table S5. We reasoned that among the genes of these GO terms, those that are either salt up-regulated or not down-regulated only in PaMBF1c-OE lines would be more highly associated with enhanced salt stress tolerance in PaMBF1c-OE. And thus, we manually surveyed those groups of genes. Among the interesting findings that we noticed was that many ribosome and translation-related genes were not salt down-regulated only in PaMBF1c-OE (Supplementary Table S5), which suggests that PaMBF1c might function in promoting protein synthesis processes particularly to acquire enhanced salt stress tolerance.