Microtubule Dynamics Plays a Vital Role in Plant Adaptation and Tolerance to Salt Stress

Although recent studies suggest that the plant cytoskeleton is associated with plant stress responses, such as salt, cold, and drought, the molecular mechanism underlying microtubule function in plant salt stress response remains unclear. We performed a comparative proteomic analysis between control suspension-cultured cells (A0) and salt-adapted cells (A120) established from Arabidopsis root callus to investigate plant adaptation mechanisms to long-term salt stress. We identified 50 differentially expressed proteins (45 up- and 5 down-regulated proteins) in A120 cells compared with A0 cells. Gene ontology enrichment and protein network analyses indicated that differentially expressed proteins in A120 cells were strongly associated with cell structure-associated clusters, including cytoskeleton and cell wall biogenesis. Gene expression analysis revealed that expressions of cytoskeleton-related genes, such as FBA8, TUB3, TUB4, TUB7, TUB9, and ACT7, and a cell wall biogenesis-related gene, CCoAOMT1, were induced in salt-adapted A120 cells. Moreover, the loss-of-function mutant of Arabidopsis TUB9 gene, tub9, showed a hypersensitive phenotype to salt stress. Consistent overexpression of Arabidopsis TUB9 gene in rice transgenic plants enhanced tolerance to salt stress. Our results suggest that microtubules play crucial roles in plant adaptation and tolerance to salt stress. The modulation of microtubule-related gene expression can be an effective strategy for developing salt-tolerant crops.


Introduction
Plant adaptation to environmental stress is regulated by cascades of molecular networks, including stress perception, signal transduction, metabolic adjustment, and the regulation of stress-responsive gene expressions, to reestablish cellular homeostasis, such as osmotic and ionic homeostasis, and protect proteins and cell membranes by using heat shock proteins (Hsps), chaperones, late embryogenesis abundant (LEA) proteins, osmoprotectants, and free-radical scavengers [1]. Plant cells have adapted to salt stress by changing cell wall composition [2,3]. Extensin, a significant cell wall glycoprotein, is cross-linked with phenolics by reactive oxygen species (ROS) accumulation to stiffen the cell wall when plant cells are exposed to salt stress [2]. The RhEXP4, expansin A4 of analysis of expressed proteins by proteomics is valuable for understanding the mole mechanisms underlying plant adaptation and tolerance to salt stress.
Our previous metabolite profiling study using salt-adapted Arabidopsis callus pension-cultured cells reveals that various cellular processes, including cell wall thic ing, play essential roles in plant salt adaptation [40]. In this proteomics study, we reve that major differentially expressed proteins (DEPs) identified from salt-adapted cells functionally associated with cytoskeleton and cell wall biogenesis. Structural and phological changes of plant cells mediated by cytoskeleton and cell wall biogenesis tions are vital for adaptation and tolerance to salt stress.

Morphological Features of Salt-Adapted Callus Suspension-Cultured Cells
Plants exhibit growth inhibition and impediment of tissue development in resp to salt stress because of a deficit of cell wall extensibility [41]. When we compared phologies between control cells (A0) and salt-adapted cells (A120; adapted to 120 NaCl), we observed that the A120 cells showed distinct morphological changes comp with A0 cells, including spherical or ellipsoidal and isodiametric shapes (Figure 1a) ditionally, newly divided A120 cells stuck together in small clumps. Vacuole size an cytoplasmic volume in A120 cells were significantly reduced compared with those i cells ( Figure 1a). These data suggested that plant suspension cells have changed their phology to adapt to long periods of salt stress. To understand the molecular mecha underlying cell morphology changes during salt adaptation, we identified DEPs in adapted A120 cells by proteomics analysis. Additionally, we characterized their biolo functions by molecular genetic analysis using Arabidopsis mutants and transgenic plants (Figure 1b).

Overview of Proteomic Profiles in Salt-Adapted Cells
Crude proteins were extracted from A0 and A120 cells grown in normal media A0 cells) and saline media with 120 mM NaCl (for A120 cells) for 8 days after subcu using the trichloroacetic acid/acetone/phenol extraction protocol [42] and quantified u a 2D-Quant Kit (GE Healthcare, Waukesha, WI, USA). Representative two-dimens gel electrophoresis (2-DE) images from three biological replicates of A0 and A120 cell displayed in Figure 2. With a cut-off point as a p-value of < 0.05 for the differential ex sion between A0 and A120 cells, 50 DEP spots were identified by matrix-assisted desorption/ionization time-of-flight mass spectrometry (MALDI-TOF/TOF MS) (Tab

Functional Network Analysis of Differentially Expressed Proteins
To understand the biological functions and modes of action of 50 DEPs in plant salt adaptation, we analyzed putative physical interactions of DEPs using the Cytoscape software platform (https://cytoscape.org/) (accessed on 1 April 2021) and the IntAct database (https://www.ebi.ac.uk/intact/) (accessed on 1 April 2021) (Figure 4). The Cytoscape with large databases of protein-protein, protein-DNA, and genetic interactions is a powerful software for studying the prediction of a physical interaction network in model organisms [43]. Out of 50 DEPs, the physical interactions of 34 DEPs were identified from this analysis. The largest cluster was the "cell structure-associated cluster," including 12 DEPs (red ellipse) in the functional network. The proteins in this cluster were mainly involved in the regulation of cell structures, including both cytoskeleton functions, such as actin filaments (ACT7, ADF3, and FBA8) and MTs (TUB3, TUB4, and TUB9), and secondary cell wall biogenesis (CCoAOMT1) (Figure 4). Even though TUB7 and PCAP1 proteins were highly induced in A120 cells (Table 1), they were not identified in functional network analysis. This is probably due to the lack of physical interaction information identified so far. PCAP1, also known as MT-destabilizing protein 25 (MDP25), functions as a negative regulator in hypocotyl cell elongation [44]. Additionally, other DEPs physically interacted with various functional proteins clustered in the ROS-associated cluster (CTIMC and ANNAT1; green ellipse), drought-and ABA-associated cluster (GRF3 and RBG2; purple ellipse), temperature-associated cluster (HSP70-1, HSP70-9, HOP2, and CSP2; orange ellipse), and transcriptional/translational system-associated cluster (PBD1, RPN11, PAP1, and CPN20; gray box) ( Figure 4). The connectivity of protein interaction networks suggested that significant cellular and molecular changes in plant adaptation to salt stress might be associated with the plant cytoskeleton and cell wall biogenesis, affecting cell structure changes.

Expression Patterns of Cytoskeleton-Related Genes in Salt-Adapted Cells
To confirm the results of proteomics and bioinformatics analyses suggesting the crucial roles of cell structure-related proteins in plant salt adaptation (Table 1 and Figure 4), we tested the expressions of 12 genes encoding DEPs. This belonged to the cytoskeleton and cell wall biogenesis functions between A0 and A120 cells using quantitative real-time PCR (qRT-PCR). We also included two cytoskeleton-related genes, TUB7 and PCAP1, in

Expression Patterns of Cytoskeleton-Related Genes in Salt-Adapted Cells
To confirm the results of proteomics and bioinformatics analyses suggesting the crucial roles of cell structure-related proteins in plant salt adaptation (Table 1 and Figure 4), we tested the expressions of 12 genes encoding DEPs. This belonged to the cytoskeleton and cell wall biogenesis functions between A0 and A120 cells using quantitative real-time PCR (qRT-PCR). We also included two cytoskeleton-related genes, TUB7 and PCAP1, in the gene expression analysis.
Although TUB7 was not identified in functional network analysis (Figure 4), its mRNA was more abundant in A120 cells ( Figure 5). Since four TUB genes, TUB3, TUB4, TUB7, and TUB9, were induced in the protein and mRNA levels in A120 cells, MT-related proteins play essential roles in plant adaptation to salt stress. Figure 5. Transcript levels of cell structure-related genes in control (A0) and salt-adapted (A120) cells. Total RNAs were extracted from the A0 and A120 cells. Transcript levels were determined by quantitative real-time PCR (qRT-PCR). UBQ10 was used as a quantitative control for qRT-PCR. Error bars represent the standard deviation (SD) of three independent replicates. Asterisks indicate significant differences in the A0 cells (* p-value < 0.5; **, p-value ≤ 0.01, Student's t-test).

The Effect of the Loss-of-Function β-Tubulin Genes in Salt Stress Response
Recent evidence indicates that the regulation of MTs' destabilization and reorganization is essential for plant adaptation to salt stress [17,45,46]. Furthermore, the largest protein family among the cell structure-associated cluster is related to the β-tubulin family proteins, including TUB3, TUB4, TUB7, and TUB9 (Table 1 and Figure 4). To characterize Figure 5. Transcript levels of cell structure-related genes in control (A0) and salt-adapted (A120) cells. Total RNAs were extracted from the A0 and A120 cells. Transcript levels were determined by quantitative real-time PCR (qRT-PCR). UBQ10 was used as a quantitative control for qRT-PCR. Error bars represent the standard deviation (SD) of three independent replicates. Asterisks indicate significant differences in the A0 cells (* p-value < 0.5; **, p-value ≤ 0.01, Student's t-test).

The Effect of the Loss-of-Function β-Tubulin Genes in Salt Stress Response
Recent evidence indicates that the regulation of MTs' destabilization and reorganization is essential for plant adaptation to salt stress [17,45,46]. Furthermore, the largest protein family among the cell structure-associated cluster is related to the β-tubulin family proteins, including TUB3, TUB4, TUB7, and TUB9 (Table 1 and Figure 4). To characterize the physiological functions of β-tubulin in salt stress responses, we isolated T-DNA insertion mutants of Arabidopsis β-tubulin genes (tub3, SALK_073132; tub4, SALK_204506; tub7, SALK_026797; tub9, SALK_015876). Wild-type (WT, ecotype Col-0) and four tub mutant plants were grown on MS medium for 5 days and then transferred to the soil to test mutant phenotypes under salt stress conditions. After 9 days, we supplied water containing 130 mM NaCl to the soil once a week for 4 weeks. The tub4 mutants displayed strongly tolerant phenotypes, such as enhanced plant height and late wilting of leaves, to salt stress compared to WT plants (Figure 6b). In contrast, tub9 mutants were hypersensitive to salt stress with a quickly wilting phenotype compared with WT plants (Figure 6d). Furthermore, tub3 and tub7 mutants showed similar phenotypes, including plant height and wilting of leaves, to WT plants under salt stress conditions (Figure 6a,c). These results suggested that both TUB4 and TUB9 play significant roles in plant adaptation and tolerance to salt stress, but their mode of function differs.  (Figure 6b). In contrast, tub9 mutants were hypersensitive to s stress with a quickly wilting phenotype compared with WT plants (Figure 6d). Furth more, tub3 and tub7 mutants showed similar phenotypes, including plant height and w ing of leaves, to WT plants under salt stress conditions (Figure 6a,c). These results su gested that both TUB4 and TUB9 play significant roles in plant adaptation and toleran to salt stress, but their mode of function differs.

The Effect of TUB9 Overexpression in Rice During Salt Stress
The hypersensitive phenotype of Arabidopsis tub9 mutants to salt stress suggests th the overexpression of Arabidopsis TUB9 gene can enhance crop tolerance to salt stress. confirm this, we generated transgenic rice plants overexpressing Arabidopsis TUB9 ge under the control of the CaMV 35S promoter (TUB9-OX). The Arabidopsis TUB9-OX co struct was transformed into rice ("Ilmi" cultivar) embryogenic callus, and three indepen ent TUB9-OX T1 lines were selected by hygromycin B resistance and RT-PCR analys Under normal conditions, TUB9-OX transgenic plants were shorter than WT plants (F ure 7a). Besides plant height, other morphological phenotypes of TUB9-OX transgen plants were comparable with WT plants. Ten-day-old WT and TUB9-OX transgenic pla seedlings were transferred into MS liquid media containing 120 mM NaCl. After 7 da of salt treatment, salt-treated WT and TUB9-OX transgenic plants were recovered in liqu MS medium without NaCl for 10 days. The TUB9-OX transgenic plants had greener leav and higher heights than WT plants (Figure 7b). The number of rice transgenic plants w green leaves in WT and TUB9-OX transgenic lines was counted in the recovery stage af salt treatment to calculate the survival rate. The survival rate of TUB9-OX transge plants was approximately 40%; however, most WT plants had no green leaves (Figu  7b,c). These results suggested that Arabidopsis TUB9 gene functions as a positive regula in plant adaptation to salt stress and can enhance plant tolerance to salt stress.

The Effect of TUB9 Overexpression in Rice during Salt Stress
The hypersensitive phenotype of Arabidopsis tub9 mutants to salt stress suggests that the overexpression of Arabidopsis TUB9 gene can enhance crop tolerance to salt stress. To confirm this, we generated transgenic rice plants overexpressing Arabidopsis TUB9 gene under the control of the CaMV 35S promoter (TUB9-OX). The Arabidopsis TUB9-OX construct was transformed into rice ("Ilmi" cultivar) embryogenic callus, and three independent TUB9-OX T 1 lines were selected by hygromycin B resistance and RT-PCR analysis. Under normal conditions, TUB9-OX transgenic plants were shorter than WT plants (Figure 7a). Besides plant height, other morphological phenotypes of TUB9-OX transgenic plants were comparable with WT plants. Ten-day-old WT and TUB9-OX transgenic plant seedlings were transferred into MS liquid media containing 120 mM NaCl. After 7 days of salt treatment, salt-treated WT and TUB9-OX transgenic plants were recovered in liquid MS medium without NaCl for 10 days. The TUB9-OX transgenic plants had greener leaves and higher heights than WT plants (Figure 7b). The number of rice transgenic plants with green leaves in WT and TUB9-OX transgenic lines was counted in the recovery stage after salt treatment to calculate the survival rate. The survival rate of TUB9-OX transgenic plants was approximately 40%; however, most WT plants had no green leaves (Figure 7b,c). These results suggested that Arabidopsis TUB9 gene functions as a positive regulator in plant adaptation to salt stress and can enhance plant tolerance to salt stress.

Discussion
Salt stress disrupts cell division in leaves and roots through various cellular nisms, such as calcium ion, ROS, and ABA-dependent responses [47]. The change lular morphology, such as cell proliferation and cell expansion, are essential for p aptation and tolerance to salt stress [47,48]. However, cellular and molecular mech of morphological changes during salt adaptation have not been well elucidate study demonstrated that salt-adapted A120 cells showed morphological changes, spherical or ellipsoidal and isodiametric shapes, compared with control A0 cells 1a). Results of GO and network analysis using proteomics data showed that man identified from salt-adapted cells were associated with regulating cell structures, ing cytoskeleton and cell wall biogenesis (Figures 3 and 4). Moreover, our gene exp and molecular genetic analyses revealed that β-tubulin family proteins play posi negative roles in plant adaptation and tolerance to salt stress (Figures 6 and 7). Ou suggest that β-tubulin MTs are vital components in modulating plant adaptation erance to salt stress.

Molecular Functions of Differentially Expressed Proteins in Salt-Adapted Cells
This study elucidated the molecular mechanisms underlying plant adaptation longed salt stress by comparative proteomics between control and salt-adapted ce previous proteomics studies conducted using suspension cells demonstrated tha ular mechanisms of suspension cells in salt stress response are complicated but si those studied at the whole plant level [49,50]. Using proteomics, we identified 5 including 45 up-regulated and 5 down-regulated proteins, in salt-adapted cells co with control cells (Table 1). Functional network analysis revealed that the identifie were included in various functional clusters, but many of them in cell structure-as clusters, including cytoskeleton and cell wall biogenesis functions (Figure 4).

Discussion
Salt stress disrupts cell division in leaves and roots through various cellular mechanisms, such as calcium ion, ROS, and ABA-dependent responses [47]. The changes in cellular morphology, such as cell proliferation and cell expansion, are essential for plant adaptation and tolerance to salt stress [47,48]. However, cellular and molecular mechanisms of morphological changes during salt adaptation have not been well elucidated. This study demonstrated that salt-adapted A120 cells showed morphological changes, such as spherical or ellipsoidal and isodiametric shapes, compared with control A0 cells (Figure 1a). Results of GO and network analysis using proteomics data showed that many DEPs identified from salt-adapted cells were associated with regulating cell structures, including cytoskeleton and cell wall biogenesis (Figures 3 and 4). Moreover, our gene expression and molecular genetic analyses revealed that β-tubulin family proteins play positive and negative roles in plant adaptation and tolerance to salt stress (Figures 6 and 7). Our results suggest that β-tubulin MTs are vital components in modulating plant adaptation and tolerance to salt stress.

Molecular Functions of Differentially Expressed Proteins in Salt-Adapted Cells
This study elucidated the molecular mechanisms underlying plant adaptation to prolonged salt stress by comparative proteomics between control and salt-adapted cells. The previous proteomics studies conducted using suspension cells demonstrated that molecular mechanisms of suspension cells in salt stress response are complicated but similar to those studied at the whole plant level [49,50]. Using proteomics, we identified 50 DEPs, including 45 up-regulated and 5 down-regulated proteins, in salt-adapted cells compared with control cells (Table 1). Functional network analysis revealed that the identified DEPs were included in various functional clusters, but many of them in cell structure-associated clusters, including cytoskeleton and cell wall biogenesis functions (Figure 4).

Cell Structure-Associated Cluster
The plant cell surface comprises the cell wall, plasma membrane, and cytoskeleton [41]. Plant cytoskeletons play essential functions in plant tolerance and survival to salt stress [17,18]. Many up-regulated proteins in A120 cells were MTs and actin filamentrelated proteins (Table 1 and Figure 4). The ACT7 (AT5G09810), ADF3 (AT5G59880), and FBA8 (AT3G52930) proteins were involved in the actin cytoskeleton. Actin cytoskeletons are composed of two classes, which are vegetative (ACT2, ACT7, and ACT8) and reproductive (ACT1, ACT3, ACT4, ACT11, and ACT12). ACT7 transcription is high in vegetative organs and induced by auxin [51]. The act11 mutant decreases pollen germination and increases pollen tube growth by increasing the actin turnover rate [52]. However, the lossof-function ACT2 mutant vegetative class affects root hair growth but is not complemented by overexpressing ACT7, even if they are of the same classes [53]. Additionally, ACT7 physically interacts with ACT1, ACT11, ACT12, actin-depolymerizing factor 6 (ADF6), and actin-interacting protein 1-2 (AIP1-2) (Figure 4). ADF3 (actin-depolymerizing factor 3) depolymerizes F-actin and acts as a crucial regulator in plant defense response to biotic stress [54]. Abiotic stresses also regulate the protein and gene expression of ADFs. OsADF proteins in rice leaves are highly accumulated because of drought stress [55]. OsADF3 protein is induced by salt stress in two rice cultivars (Oryza sativa L. cv. Nipponbare and Oryza sativa L. cv. Tainung 67) [56,57]. FBA8, which encodes fructose-bisphosphate aldolase 8, is involved in actin polymerization and various abiotic stress responses, such as salt, drought, ABA, and temperature stresses [50,58]. MT dynamics, polymerization and depolymerization, are necessary for cellular processes of plant tolerance and adaptation to salt stress [17]. Our results revealed that TUB3, TUB4, TUB7, and TUB9 proteins, involved in MT depolymerization and reorganization, play vital roles in plant adaptation and tolerance to salt stress (Table 1 and Figure 4). It was also reported that TUA6 (α-chain tubulin 6) and TUB2 (β-chain tubulin 2) proteins are highly expressed in Arabidopsis roots in response to salt stress [59].

Temperature-Associated Cluster
Heat shock 70 kDa protein 1 (HSP70-1), a key component in protein folding, plays a vital role in stomatal closure and seed germination and response to ABA stress. Mitochondrial HSP70-9 protein is involved in iron-sulfur protein biogenesis. Cold shock protein 2 (CRP2) protein plays different roles as a negative regulator in response to cold stress and as a positive regulator in salt stress response. Hsp70-Hsp90 organizing protein 2 (HOP2) influences plant adaptation to prolonged heat stress. These four proteins compose functional networks via physical interaction with chaperon regulators, including heat stress transcription factor A-1 (HSF1A), suppressor of G2 allele of skp1 (SGT1) homolog B (SGT1B), Bcl-2-associated athanogene 3 (BAG3) and 5 (BAG5), and cytokinin response factor 3 (CRF3) (Figure 4).

The Role of Microtubules in Plant Adaptation and Tolerance to Salt Stress
MTs are fixed in the plasma membrane and composed of a greater part of plant interphase arrays [9,15]. The cortical MT arrays are involved in plant response to various abiotic stresses, especially salt stress [9,18]. Plants increase salt tolerance by regulating depolymerization and reorganization of the cortical MTs [48]. MAP65-1 acts as a positive regulator in plant salt tolerance by promoting cortical MT reorganization [68]. Calcium ions reorganize the damage of MT arrays in the salt stress response of plant cells [17]. The lossof-function sos3, a calcium sensor in the salt stress response, mutant shows hypersensitivity to salt stress due to the irregular organization of MTs [26]. Plants with salt-susceptible phenotypes have a lower concentration of calcium ions than that of salt-tolerant plants [69]. Our proteomic analysis showed that the four β-tubulin family proteins, including TUB3, TUB4, TUB7, and TUB9, were induced in salt-adapted A120 cells compared with control A0 cells (Table 1 and Figure 4). Additionally, the mRNA levels of TUB3, TUB4, TUB7, and TUB9 genes were higher in A120 cells than in A0 cells ( Figure 5). Our results suggest that the elevation of β-tubulin mRNAs and protein levels can affect MT functions and enhance plant adaptation to salt stress. In our molecular genetic analysis, the loss-of-function tub4 mutant showed enhanced tolerance to salt stress. In contrast, the tub9 mutant was more hypersensitive than WT plants ( Figure 6). The overexpression of TUB9 in rice plants enhanced the plant's tolerance to salt stress ( Figure 7). Interestingly, tub4 and tub9 mutant plants showed opposite phenotypes in response to transiently applied salt stress, even though TUB4 and TUB9 protein levels were higher in cells that have adapted to salt stress for a long time. These results suggest that TUB4 and TUB9 proteins play different roles in plant responses to short-term and long-term salt stresses. It was also reported that short-term and long-term salt stress have different effects on the actin filament assembly and disassembly [25]. It would be worthwhile to dissect the biological functions of TUB4 and TUB9 in plant adaptation and tolerance to salt stress in further studies.
Altogether, our results suggest that β-tubulin proteins play different roles in plant adaptation and tolerance to salt stress by regulating MT depolymerization and reorganization. Therefore, changes in MT dynamics in plant cells would be essential for cellular processes to enhance the adaptation and tolerance to salt stress. Furthermore, morphological changes in salt-adapted suspension cells are at least partly due to the changes in MT dynamics.

Growth Conditions of Callus Suspension Cells
Salt-adapted callus suspension cells were generated from Arabidopsis thaliana (Col-0 ecotype) roots as described in detail in a previous study [40]. Callus suspension cells were maintained at 23 • C in the dark with gentle shaking (140 rpm).

Proteomic Profiling Using Two-Dimensional Gel Electrophoresis
Total protein was isolated from 5 g of A0 and salt-adapted cells (A120) using trichloroacetic acid/acetone/phenol extraction protocol described in detail in a previous study [42]. Total soluble proteins were quantified using the 2D-Quant Kit (Amersham Biosciences Europe GmbH, Freiburg, Germany). Two-dimensional gel electrophoresis was performed with Protean IEF cell (Bio-Rad, Hercules, CA, USA) for the first-dimensional isoelectric focusing using immobilized pH gradient strips (24 cm, pH 4-7; Bio-Rad Laboratories, Hercules, CA, USA), and with the Protean Xi-II Cell system (Bio-Rad Laboratories, Hercules, CA, USA) for the second-dimensional sodium dodecyl sulfate-polyacrylamide gel electrophoresis. After Coomassie brilliant blue staining, gel images were taken using a GS-800 Imaging Densitometer Scanner (Bio-Rad Laboratories, Hercules, CA, USA) and analyzed using PDQuest v.7.2.0 (Bio-Rad Laboratories, Hercules, CA, USA). All experiments were performed in three independent biological replicates, and the volume of each spot was detected and normalized to a relative density. Proteins showing a statistically significant difference (p < 0.05) between A0 and A120 cells were identified. For protein identification, differential protein spots visualized in the gel were excised and subjected to in-gel digestion as described previously [42]. Protein identification was performed by MALDI-TOF/TOF MS using the ABI 4800 Plus TOF-TOF Mass Spectrometer (Applied Biosystems, Framingham, MA, USA). Fifty proteins were identified, of which peptide and fragment mass tolerance was fixed at 100 ppm. The high confidence interval displayed statistically reliable search scores (more than 95% confidence) corresponding to protein's experimental isoelectric point (pI) and molecular weight.

Analysis of Quantitative Real Time PCR (qRT-PCR)
Total RNA was extracted from A0 and A120 cells using the RNeasy Plant Kit (Qiagen, Valencia, CA, USA) following the manufacturer's protocol. To remove genomic DNA contaminants, extracted RNA was treated with DNaseI (Thermo Fisher Scientific, Waltham, MA, USA). One µg of total RNA was used the first strand of cDNA synthesis using a cDNA synthesis kit (Invitrogen, Carlsbad, CA, USA), according to the manufacturer's protocol.
The qRT-PCR analysis was performed using the QuantiMix SYBR (PhileKorea, Seoul, Korea), and the relative values of indicated gene expression were automatically calculated using the CFX96 real-time PCR detection system (Bio-Rad Laboratories, Hercules, CA, USA) by applying normalization of the expression of UBQ10. The qRT-PCR was performed using the following conditions: 50 • C for 10 min, 95 • C for 10 min; followed by 50 cycles at 95 • C for 15 s, 60 • C for 15 s, and 72 • C for 15 s. The gene specific primers in qRT-PCR analysis are listed in Supplementary Table S1.

Generation of Transgenic Rice Plants
To generate the transgenic rice plants overexpressing the Arabidopsis TUB9 gene, we cloned the full-length cDNA (1335 bp) of the Arabidopsis TUB9 gene into pH2GW7 vector under the control of CaMV 35S promoter. The TUB9-OX construct was introduced into Agrobacterium tumefaciens (LBA4404) by electroporation. We used a modified version of the general rice-transformation protocol [70]. Transgenic TUB9-OX (T1) plants were selected on MS medium containing hygromycin B and then transferred to soil and allowed to self-pollinate.

Salt Stress Treatment
In Arabidopsis, 5-day-old WT seedlings, tub3, tub4, tub7, and tub9 plants grown on MS media were transferred to soil. After 9 days, we supplied water containing 130 mM NaCl to the soil once a week for 4 weeks. Photographs of each representative of 12-16 individual plants were taken to analyze plant phenotypes. In rice, 10-day-old WT seedlings and TUB9-OX plants germinated in MS media containing hygromycin B were transferred into MS liquid medium with 120 mM NaCl. After 7 days, plants were recovered in MS solution without NaCl for 10 days. Photographs were taken to represent 8-10 individual plants to analyze plant phenotypes.

Statistical Analyses
Statistical analyses, including Student's t-test, were performed using Excel 2010. qRT-PCR analysis was performed in three independent biological replicates, and the average values of 2 ∆∆ CT were used to determine expression differences. Data were indicated as means ± standard deviation (SD). Error bars indicate SD.

Conclusions
This study suggests that the morphological changes of plant cells are an essential cellular process for adaptation to prolonged salt stress. We revealed that various protein families involved in various cellular processes play a role in salt adaptation response using proteomic analysis. Furthermore, gene expression and molecular genetic analyses demonstrated that β-tubulin proteins play an important role in plant adaptation and tolerance to salt stress. Altogether, our results suggest that the dynamics of depolymerization and reorganization of tubulin MTs play critical roles in plant adaptation to salt stress.