Ectopic Expression of a Glycine soja myo-Inositol Oxygenase Gene (GsMIOX1a) in Arabidopsis Enhances Tolerance to Alkaline Stress

Myo-inositol participates in various aspects of plant physiology, and myo-inositol oxygenase is the key enzyme of the myo-inositol oxygenation pathway. Previous studies indicated that myo-inositol oxygenase may play a role in plant responses to abiotic stresses. In this study, we focused on the functional characterization of GsMIOX1a, a remarkable alkaline stress-responsive gene of Glycine soja 07256, based on RNA-seq data. Using quantitative real-time PCR, we demonstrated that GsMIOX1a is rapidly induced by alkaline stress and expressed predominantly in flowers. We also elucidated the positive function of GsMIOX1a in the alkaline response in the wild type, atmiox1 mutant as well as GsMIOX1a-overexpressing Arabidopsis. We determined that atmiox1 mutant decreased Arabidopsis tolerance to alkaline stress, whereas GsMIOX1a overexpression increased tolerance. Moreover, the expression levels of some alkaline stress-responsive and inducible marker genes, including H+-Ppase, NADP-ME, KIN1 and RD29B, were also up-regulated in GsMIOX1a overexpression lines compared with the wild type and atmiox1 mutant. Together, these results suggest that the GsMIOX1a gene positively regulates plant tolerance to alkaline stress. This is the first report to demonstrate that ectopic expression of myo-inositol oxygenase improves alkaline tolerance in plants.


Introduction
Soil alkalinity is one of the most serious environmental factors negatively affecting plant production in northeast China [1,2]. Alkalization of soil due to NaHCO 3 and Na 2 CO 3 can be more destructive than soil salinization caused by neutral salts, such as NaCl and Na 2 SO 4 [3]. Alkaline stress not only involves ion injury, excessive levels of reactive oxygen species (ROS) and osmotic stress [4,5] but also acts as a high-pH stressor. A high-pH environment surrounding roots may inhibit ion uptake and disrupt ion homeostasis in plant cells [6]. Consequently, growth and photosynthesis are negatively affected [7]. However, to date, tolerance to alkaline stress has not been intensively studied in plants. Glycine soja 07256 (G. soja 07256) can germinate and grow in soil at pH 9.02 and survive in nutrient solutions containing 50 mM NaHCO 3 [8], indicating that this plant has developed unique molecular and physiological mechanisms to adapt to these stress conditions. Therefore, G. soja 07256 is an ideal candidate for screening stress-resistance genes and studying the molecular mechanisms of plant stress tolerance.
Myo-inositol oxygenase (MIOX; E.C. 1.13.99.1) is a monooxygenase that catalyzes the conversion of myo-inositol into D-glucuronic acid (D-GlcUA). MIOX proteins are highly conserved and present in nearly all eukaryotes. Many animals, including rats, use this product to synthesize ascorbic acid (AsA). MIOX first degrades myo-inositol to D-GlcUA. This step is followed by reduction to L-gulonic acid and ring formation to gulonolactone, which is finally oxidized to AsA [9]. Plants have also established the pathway for biosynthesizing D-GlcUA with MIOX activity as the committed step, whereas plants further use this product and activate it into UDP-glucuronic acid (UDP-GlcUA). UDP-GlcUA is an important precursor for several nucleotide sugars that are used to synthesize cell wall polysaccharides. The finding was supported by 3 H-myoinositol labeling studies [10,11] as well as the analysis of knockout mutants in Miox1 and Miox2 of Arabidopsis [12,13]. Therefore, one of the important functions of MIOX involves controlling the level of myo-inositol, which is a precursor for many inositolcontaining compounds that are involved in various physiological and biochemical processes, such as growth regulation, cell membrane biogenesis, hormonal regulation and stress signaling [14]. MIOX studies in plants were previously limited because the content of MIOX was difficult to measure in plant extracts and no molecular sequences were available. Recently, peptide sequences were obtained by purifying the enzymes from pig kidneys [15] and the yeast Cryptococcus lactativorus [16], thereby facilitating the subsequent cloning of the corresponding genes. With the exception of the D-mannose/L-galactose (Man/Gal), D-galacturonate (GalU) and Lgulose pathways, an animal-like route for AsA synthesis in plants has been identified [17]. Some studies have suggested that MIOX gene overexpression leads to a 2-to 3-fold increase in AsA in transgenic Arabidopsis lines [18]; Zhang et al. demonstrated that overexpression of a purple acid phosphatase resulted in an increase in D-GlcUA in Arabidopsis [19]. Because the product of the MIOX reaction was D-GlcUA, their data suggested that this metabolite is readily converted to AsA in Arabidopsis, resembling the pathway characterized in animals. However, recent studies reported that MIOX controls the level of myo-inositol without increasing AsA content [20,21], which was also confirmed by J. Duan et al. in a study reporting that OsMIOX (Oryza sativa L. cv. IRAT109) overexpression improves drought tolerance in rice, but the level of AsA was not altered before and after stress [22]. Therefore, whether MIOX has a positive role in AsA biosynthesis in plants remains unclear.
We analyzed RNA-seq data from G. soja 07256 treated with 50 mM NaHCO 3 (pH 8.5) and observed that GsMIOX1a was among those genes most strongly up-regulated (raw sequence data are available at http://www.onekp.com/samples/single.php?id=LXGM). In this study, we aimed to identify the functional and regulatory roles of GsMIOX1a in plant responses to alkaline stress. We analyzed alkaline stress-induced transcript accumulation of GsMIOX1a in a time-dependent manner as well as the expression profiles of GsMIOX1a in various organs of G. soja 07256. To verify the positive role of GsMIOX1a in the alkaline response, we functionally validated gene expression under conditions of alkaline stress in wild type (WT), GsMIOX1aoverexpressing (OX) Arabidopsis and an Arabidopsis mutant with a T-DNA insertion of a homologous gene, atmiox1. Furthermore, we assessed the physiological mechanism by which GsMIOX1a OX lines exhibit increased alkaline tolerance. Collectively, these results provide direct evidence that GsMIOX1a positively regulates the plant alkaline stress response and will facilitate further studies of the biological functions and molecular mechanisms of plant MIOX in response to abiotic stresses. Seeds of G. soja 07256 and G. soja 50109 were obtained from the Jilin Academy of Agricultural Sciences (Changchun, China). G. max Suinong 28 and G. max Hefeng 55 were obtained from the Chinese Crop Germplasm Information System.
To analyze gene transcript accumulation induced by alkaline stress, after treatment with 98% sulfuric acid for 10 min, the seeds of G. soja 07256 were washed five times with sterilized water and then germinated and grown in 1/4 Hoagland solution for 3 weeks at 24 to 26°C and a 16-h light:8-h dark cycle. The roots of 21-day-old seedlings were submerged in 1/4 Hoagland solution containing 50 mM NaHCO 3 (pH 8.5). Samples of roots were harvested at 0, 1, 3, 6, 12 and 24 h after treatment. Then, the samples were frozen in liquid nitrogen and stored at -80°C for RNA extraction.
For GsMIOX1a expression analysis in different soybean varieties, seeds of G. soja 07256, G. soja 50109, G. max Suinong 28 and G. max Hefeng 55 were placed on each petri dish with wet filter paper to accelerate germination for 2 days. Germinated seedlings were then transferred into 1/4 Hoagland solution. Three weeks after sowing, seedlings were transferred into 1/4 Hoagland solution with 50 mM NaHCO 3 . Equal amounts ofroots were sampled at 0 and 6 h.
Wild type Arabidopsis thaliana (Columbia ecotype) and the atmiox1 mutant (SALK_082070) were obtained from the Nottingham Arabidopsis Stock Centre (NASC). Arabidopsis seeds were germinated and grown on 1/2 MS media with 0.8% agar or in pots filled with a 1:1:1 mixture of vermiculite: peat moss: perlite in a growth chamber under controlled environment conditions (21 to 23°C, 16-h light:8-h dark cycle).

GsMIOX1a cDNA cloning
The full-length GsMIOX1a gene was obtained via its homologous gene in G. max (Gly-ma07g01660). Briefly, total RNA was extracted from 21-day-old G. soja 07256 seedlings using an RNAprep Pure Plant Kit (Tiangen, China). Then, the cDNA was generated with a Super-Script III Reverse Transcriptase Kit (Invitrogen, Carlsbad, CA, USA). Gene-specific primers, GsMIOX1a-F: 5'-ATGACTATCCTCATTGAGCAATCTGATC-3' and GsMIOX1a-R: 5'-TCACCACTTCAGCTTTGCAGGGA-3', were designed according to the corresponding gene sequence from G. max, and the full-length CDS region of GsMIOX1a was obtained by polymerase chain reaction (PCR) amplification. The PCR products were cloned into the pGEM-T cloning vector (Promega, Madison, WI, USA), and the resulting plasmids were sequenced.

Quantitative real-time PCR analysis
Total RNA extraction from different tissues and cDNA synthesis were performed as described above. Quantitative real-time PCR was performed on a Stratagene MX3000p real-time cycler using SYBR Green. One reaction (10 μL) consisted of 1 μL of cDNA (diluted 1: 5), 5 μL of SYBR Green Real-time PCR Master Mix (TOYOBO, USA), 400 nM of each primer, and 3.2 μL of ddH 2 O. The housekeeping genes AtACTIN2 (F: 5'-TTACCCGATGGGCAAGTC-3'; R: 5'-GCTCATACGGTCAGCGATAC-3') and GsGAPDH (glyceraldehyde 3-phosphate dehydrogenase; F: 5'-GACTGGTATGGCATTCCGTGT-3'; R: 5'-GCCCTCTGATTCCTCCTTGA -3') were used for internal normalization of Arabidopsis and G. soja genes, respectively [23]. For calculations, the reaction efficiency of the individual well was computed from the original data. Results were averaged over three triplicates and normalized using the calculated levels of AtAC-TIN2 or GsGAPDH transcripts. The relative intensities were calculated and normalized as previously described [24].
Characterization of the atmiox1 mutant PCR and semi-quantitative reverse transcriptase (RT)-PCR analyses were performed to characterize the AtMIOX1 gene T-DNA insertion mutant of Arabidopsis (atmiox1, SALK_082070). First, the forward (FP: 5'-GTGAAACCCTACAAGCAGAAAGC-3') and reverse (RP: 5'-CGATTTCAAGCCGTCCACA-3') primer pair was used to determine if the T-DNA insertion was homozygous, and the FP and left border (LB: 5'-GATTTGGGTGATGGTTCACGTAGTG-3') primer pair was used to verify that the T-DNA was inserted into the AtMIOX1 gene. In addition, semi-quantitative RT-PCR using the FP and RP primer pair was performed to further identify whether the AtMIOX1 gene was efficiently silenced by the T-DNA insertion in the atmiox1 mutant. Briefly, total RNA was extracted from 2-week-old WT and atmiox1 seedlings and then reverse transcribed to generate cDNA. The yield of cDNA was normalized according to the PCR signal generated from the internal standard housekeeping gene ACTIN2 amplified with 25 cycles starting with 0.2 μl of the cDNA solution. AtMIOX1 gene expression levels were analyzed by PCR amplification for 30 cycles.

Generation of GsMIOX1a OX Arabidopsis thaliana plants
To generate GsMIOX1a OX lines, the complete coding region of GsMIOX1a was cloned into the pCAMBIA330035Su vector as described [25] by using primers F: 5'-GGCTTAAUATGAC TATCCTCATTGAG CAA-3' and R: 5'-GGTTTAAUTCACCACTTCAGTTTGC-3' and introduced into the Agrobacterium tumefaciens strain GV3101. Then, the floral dip method was used to transform Arabidopsis thaliana [26]. The T 0 generation seeds were grown on selective solid medium containing 50 mg/L glufosinate ammonium (Sigma-Aldrich), and positive transformations were confirmed by PCR using primers designed based on the coding DNA sequence (CDS) region of GsMIOX1a. Homozygous OX lines from the T 3 generation that displayed 100% resistance to glufosinate ammonium were confirmed by semi-quantitative RT-PCR analysis as described above. Primer pair was GsMIOX1a-F: AGGCGTGAAGGAGTG GAGAAC-3' and GsMIOX1a-R: GGATAAGGCCAGTTAAGTGCAA-3' and the housekeeping gene ACTIN2 served as an internal control. PCR conditions were 30 cycles for GsMIOX1a and 25 cycles for ACTIN2.

Phenotype analysis of the GsMIOX1a OX and atmiox1 mutant lines
Seeds from the WT, atmiox1 and GsMIOX1a OX lines were surface sterilized with 5% (w: v) sodium hypochlorite for 7 min and subsequently washed with sterilized distilled water 6 to 8 times. The seeds were then stored in the dark at 4°C for 3 days to break seed dormancy. In plate germination assays, the seeds of the WT, atmiox1 and GsMIOX1a OX lines (independent lines #4, #16, and #20) were germinated on 1/2 MS agar medium supplemented with 11 mM NaHCO 3 . The germination profiles were observed with respect to radicle emergence for 7 consecutive days. On the 3rd day, photographs were taken to demonstrate the growth performance of each line. In total, 120 seeds were used for each experiment, and all experiments were repeated at least three times. For soil alkaline stress treatment, three-week-old WT, atmiox1 and GsMIOX1a OX seedlings were irrigated with 100 mM NaHCO 3 (pH 8.5) every 3 days for a total of 12 days for alkaline stress.
To measure the free proline concentration, peroxidase (POD) activity and AsA content, leaf samples were obtained from 5-week-old plants under control conditions as well as 100 mM NaHCO 3 treatment. For free proline measurement, leaves (approximately 0.25 g) were extracted with 5 mL of 3% sulfosalicylic acid, and measurements were obtained using the ninhydrin assay [27]. The proline concentration was estimated at 520 nm using a spectrophotometer (UV-2550, Shimadzu, Japan). For POD, the oxidation of guaiacol was determined by measuring the increased absorbance values at 470 nm for 60 s (linear phase) [28]. The 1.8 mL reaction mixture contained 1.45 mL of 0.05 mmolÁL -1 phosphate buffer (pH 7.0), 0.05 mL of 2% guaiacol, 0.05 mL of enzyme extract and 0.25 mL of 2% H 2 O 2 . A unit of enzyme activity was defined as the change in the absorbance value per minute per mg fresh weight. The AsA content was measured as described by Arakawa et al [29]. Approximately 0.30 g of leaves was frozen in liquid nitrogen, homogenized in 6 mL of 5% trichloroacetic acid (TCA) solution, and centrifuged at 15,000×g for 15 min. The supernatant was used to measure the AsA content using 0.5% 4, 7-diphenyl-1, 10-phenanthroline.

Results
Isolation and bioinformatic analysis of the GsMIOX1a gene GsMIOX1a was isolated using homology-based cloning. The 921-bp coding sequence encodes a 306-amino acid protein that corresponds to a polypeptide with a predicted molecular weight (MW) of 35.6 kDa. The deduced protein sequence of GsMIOX1a was used as a query for a BLASTP search against the G. max protein database Phytozome v9.1 (http://www.phytozome. net/). From this search, we identified 5 putative MIOX protein sequences in soybean, and these genes were named according to the existing numbering system used in the phylogenetic tree of Arabidopsis. The characteristics of the G. max and Arabidopsis MIOX family genes, including the gene ID, full CDS length, genomic length, protein length, molecular weight and isoelectric point (pI), are presented in Table 1. Furthermore, we aligned the GsMIOX1a amino acid sequence with the MIOX amino acid sequences from other species (S1 Fig). The analysis revealed a high degree of sequence identity, including 100% identity with GsMIOX1 (KHN39268). Because the GsMIOX1 coding sequence was unavailable, we hypothesized that these sequences might represent the same gene. Then, a phylogenetic tree was constructed based on the above sequences using MEGA 5.0. The phylogenetic tree indicated that the amino acid sequences of monocot MIOX1 and dicot MIOX1 were in separate clusters. GsMIOX1a appeared in the dicot cluster and exhibited a closer relationship to the MIOX1 genes from other plants than to MIOX2, MIOX4 or MIOX5 from other species (Fig 1).

GsMIOX1a expression is alkaline stress inducible and tissue specific
To explore the biological function of GsMIOX1a in response to alkaline stress, we initially investigated GsMIOX1a transcript accumulation under alkaline stress by quantitative real-time PCR analysis. As shown in Fig 2A, under alkaline stress, GsMIOX1a expression levels increased and reached a maximum level at 6 h that was approximately 42-fold higher than that at 0 h. This result was consistent with our previous RNA-seq data ( Figure A in S1 File), which also strongly suggested a role of GsMIOX1a in the alkaline stress response. Because we screened GsMIOX1a as a putative gene from G. soja 07256 involved in the response to alkaline stress, we further investigated the GsMIOX1a response to alkaline stress in other soybean varieties, including G. soja 07256, G. soja 50109, G. max Suinong 28 and G. max Hefeng 55. G. max Suinong 28 and G. max Hefeng 55 are Chinese soybean cultivars that exhibit much lower adaptability to stress compared to wild soybean. G. soja 50109 was also a kind of wild soybean but didn't reported to be alkaline resistant. We determined the transcript expression levels of GsMIOX1a using roots of four soybean variety seedlings under 50 mM NaHCO 3 treatment by quantitative real-time PCR (Fig 2B). Under alkaline stress, the relative transcript abundance of GsMIOX1a increased by 32.8-fold in G. soja 07256 and 5.7-fold in G. soja 50109. By contrast, the expression of GsMIOX1a was relatively unchanged in G. max Suinong 28 and G. max Heheng 55, which exhibit reduced tolerance to alkaline stress. This result suggests that GsMIOX1a is an important responsive gene in alkaline-resistant soybean varieties.
As a step toward elucidating GsMIOX1a gene function in G. soja 07256, we surveyed GsMIOX1a expression profiles across different organs using quantitative real-time PCR. G. soja 07256 seedlings were planted and maintained under normal conditions, and total RNA was isolated from different organs. As shown in Fig 2C, the GsMIOX1a transcript was detected in roots, stems, old leaves, episperms, epicotyls, hypocotyls and flowers; however, minimal GsMIOX1a transcript was observed in young leaves. Among these examined tissues, increased GsMIOX1a transcript levels accumulated in flowers, as determined by quantitative real-time PCR. The quantitative real-time PCR results were confirmed by three independent experiments. To clarify the regulatory role of GsMIOX1a gene in Arabidopsis, we used constitutive OX lines in which expression was controlled by the strong constitutive CaMV35S promoter (Fig 3A). T 0 generation GsMIOX1a transformed plants were initially screened for glufosinate ammonium resistance, and T 1 generation OX lines, which displayed a 3:1 (resistant: sensitive) segregation pattern in their progeny, were selected for further analysis. Three homozygous T 3 OX Arabidopsis lines (#4, #16, #20) as determined by PCR that were 100% resistant to glufosinate ammonium were obtained. Semi-quantitative RT-PCR analysis confirmed increased GsMIOX1a transcript levels in the OX lines compared with WT plants (Fig 3B). We also examined an AtMIOX1 gene knockout mutant in Arabidopsis, atmiox1, carrying a T-DNA insertion in the 3'UTR located 146 bp downstream of the last exon (Fig 3C). We verified the genotype of mutant plants by PCR tests using the FP and RP primers for the homozygous analysis and the LB and RP primers for the T-DNA insertion analysis. As shown in Fig 3D (top lane), when assessing the T-DNA insertion using the LB and RP primers, no clear DNA band was observed for WT, but a clear DNA band was observed for the T-DNA insertion in the mutant. By contrast, no DNA band appeared for atmiox1 when we performed the PCR test with the FP and RP primers for the homozygous analysis. These results indicated that the T-DNA insertion of the atmiox1 mutant was homozygous. To verify that the AtMIOX1 gene was efficiently silenced in the atmiox1 line, we assessed the expression of AtMIOX1 using semi-quantitative RT-PCR analysis. In contrast to WT Arabidopsis, no AtMIOX1 transcripts were detected in the atmiox1 line (Fig 3E), indicating that AtMIOX1 was effectively silenced in the atmiox1 line.

GsMIOX1a positively regulates plant tolerance to alkaline stress
To investigate the possible role of GsMIOX1a in plant stress adaptation and determine whether GsMIOX1a improves the tolerance of the plant to alkaline stress, the phenotypes of the WT, atmiox1 and OX Arabidopsis were evaluated. WT, atmiox1 and two independent GsMIOX1a OX lines (#4, #16) with similar GsMIOX1a transcript accumulation levels were chosen. Initially, seeds were planted in 1/2 MS solid medium, and no significant difference in the germination rate was observed among WT, atmiox1 or GsMIOX1a OX lines. However, radicle emergence from the seeds of atmiox1 was inhibited more severely, whereas the GsMIOX1a OX lines performed better than WT lines in 1/2 MS solid medium under alkaline stress (Fig 4A). When treated with 11 mM NaHCO 3 , the germination rates of GsMIOX1a OX seeds were maintained at relatively high levels (#4, 70% and #16, 65.5%) on the 3rd day compared with WT (44.1%), but the germination rate of the atmiox1 seeds was reduced to 27.4% (Fig 4B). These results suggested that the alkaline tolerance of atmiox1 was decreased at the seed germination stage, whereas tolerance increased in the GsMIOX1a OX lines. In addition, these results further confirmed the positive role of GsMIOX1a in the alkaline stress response.
We also investigated the growth pattern of soil-grown mature plants under alkaline stress. The seeds were planted in pots and allowed to grow normally for three weeks with routine watering. Then, the 3-week-old plants were irrigated with 100 mM NaHCO 3 solution every three days for 12 days, and the growth pattern was monitored. The WT plants developed chlorotic symptoms and became yellow and purple, whereas the atmiox1 plants exhibited more severe symptoms and began to die. By contrast, the GsMIOX1a OX lines remained green and survived better than control plants (Fig 5A). Several stress-related physiological and biochemical parameters were also assessed in WT, atmiox1 and OX plants. Alkaline conditions inhibit plant growth due to a high pH and increased ROS and osmotic stress [30]. POD is one of the most important scavenging enzymes in the ROS-scavenging systems of plants [31]. Free proline, an osmoprotectant, aids in reducing the detrimental effects of alkaline conditions on cellular components [32]. Therefore, free proline concentrations along with POD activity under alkaline stress treatment were measured to understand the physiological mechanisms responsible for the increased alkaline tolerance of the GsMIOX1a OX lines and decreased alkaline tolerance of the atmiox1 line. As shown in Fig 5B, after prolonged exposure to alkaline stress, the proline content was increased in all lines with respect to normally grown seedlings. However, GsMIOX1a OX plants exhibited increased proline levels compared with WT plants (p<0.01), whereas atmiox1 mutant exhibited decreased total free proline (p<0.05). POD activity was increased in GsMIOX1a OX lines grown in stress conditions compared with WT plants (p<0.01), and POD activity in atmiox1 mutant plants was lower than WT plants (p<0.01). POD activity appeared to be less inhibited in the GsMIOX1a OX lines under alkaline stress conditions; therefore, GsMIOX1a OX lines retained more activity than the WT or atmiox1 mutant lines (Fig 5C).
In addition, previous studies have proposed a role for the MIOX pathway in the synthesis of AsA [33]. To determine if the insensitivity of the GsMIOX1a OX lines was due to a change in AsA levels, we also measured the AsA content in the WT, atmiox1 and OX plants. However, no significant difference was observed in AsA levels among WT, atmiox1 and OX plants under normal and alkaline stress conditions (Fig 5D and 5E).

Alkaline response and stress-inducible marker genes are affected by GsMIOX1a
To evaluate the mechanism of the improved alkaline tolerance conferred by GsMIOX1a OX, the transcript levels of various stress-related marker genes, including H + -Ppase, NADP-ME, KIN1 and RD29B, were analyzed under normal and alkaline stress conditions in the WT, atmiox1 and OX plants. H + -Ppase excludes H + ions via H + pumps located in the plasmalemma and tonoplast of plant cells [34], and NADP-ME balances malic acid levels [35]. Thus, both of these proteins function in adjusting the pH of the cytoplasm (pHc) and aid the mitigation of the high pH caused by alkaline stress. KIN1 and RD29B have long been proposed to play roles in various stress responses [36,37], including alkaline stress. Quantitative real-time PCR analysis indicated that all of the selected marker genes were significantly induced by NaHCO 3 in the GsMIOX1a OX lines but significantly decreased in the atmiox1 line. In all lines, H + -Ppase and  KIN1 expression levels peaked at 6 h, whereas NADP-ME and RD29B peaked at 3 h (Fig 6). These results also imply that GsMIOX1a OX promotes the accumulation of stress-responsive genes transcripts, which might be helpful for alkaline tolerance.

Discussion
Inositol is ubiquitous across the biological kingdom and is an essential component of eukaryotic cells [38]. Plants maintain an inositol pool at a basal level throughout their life cycle, and MIOX is used to control the metabolite level of myo-inositol in plants. In a few stress-tolerant plants, MIOX plays a role in the response to environmental stresses. These stress-tolerant properties are suggested to arise due to adaptive changes in their coding sequences, depending on the environment in which the plants live. For example, upland rice, which is normally grown in well-drained soil with low water-holding capacity and no surface water accumulation, is reported to encode drought-tolerant MIOX, whereas lowland rice, which is very susceptible to drought stress, encodes drought-sensitive MIOX [22]. In this study, we used G. soja 07256 because it is a relatively alkaline-tolerant plant that can germinate and set seeds in sodic soil at pH 9.02 and can survive in nutrient solutions containing 50 mM NaHCO 3 [8]. Because of this adaptation to harsh environmental conditions, G. soja 07256 is rich in resistance genes for alkaline stress tolerance.
Our lab recently analyzed RNA-seq data from roots of G. soja 07256 subjected to alkaline stress sampled at different time points (0, 1, 3, 6, 12 and 24 h). We observed that GsMIOX1a was strongly up-regulated among the five members of the MIOX family ( Figure A in S1 File). Although the GsMIOX2a gene appears to be induced more dramatically than GsMIOX1a, functional verification is ongoing. As a result, we focused on GsMIOX1a. In this study, the experimental data suggested that GsMIOX1a is an early alkaline stress-responsive gene in G. soja 07256 (Fig 2A), consistent with the RNA-seq data. The response of GsMIOX1a to alkaline stress was further confirmed in different soybean varieties ( Fig 2B). As expected, the highest expression change was observed in G. soja 07256. The different expressions of MIOX1a in different soybean varieties suggest the importance of GsMIOX1a in the plant response to alkaline stress. Similarly, in upland rice (Oryza sativa L. cv. IRAT109) not in lowland rice, OsMIOX is induced by drought, H 2 O 2 , salt, cold and abscisic acid [22]. No MIOX transcript analyses were conducted under abiotic stresses in stress-sensitive plant species, such as Arabidopsis thaliana, but AtMIOX4 OX Arabidopsis lines did not exhibit a significant phenotype under different concentrations of NaCl and sorbitol, thus suggesting a null function for MIOX4 under stress conditions [20]. These reports indicate that stress-induced expression of the MIOX gene appears to be species specific as well as dependent on the stress-responsive nature of the species.
Earlier studies have demonstrated that MIOX genes are differentially expressed in different organs of certain plant species, such as Arabidopsis thaliana [39] and rice [22]. In this study, GsMIOX1a transcripts were observed in all examined tissues, and the most abundant levels were noted in flowers (Fig 2C). This result suggests that the GsMIOX1a gene is differentially regulated in different organs to coordinate inositol metabolism. Subsequent experimental data suggested that the GsMIOX1a gene was functional in the Arabidopsis plant. In addition, GsMIOX1a gene expression in Arabidopsis plants improved the tolerance to alkaline stress in the germination and mature stages, whereas atmiox1 impaired tolerance at both stages (Figs 4 and 5A). Alkalinity triggers osmotic stress; therefore, many plants accumulate proline as an contents were measured in the leaves of 5-week-old GsMIOX1a OX Arabidopsis, atmiox1 and WT plants under normal and alkaline stress conditions in a growth chamber. Leaves of the same age and size were selected to minimize experimental error. Error bars represent the ± S.E. (n = 3). Significant differences were analyses using the T-test method. Significant differences from WT are denoted by one or two stars, corresponding to p<0.05 and p<0.01, respectively. doi:10.1371/journal.pone.0129998.g005 Alkaline Stress Tolerance Gene Assay osmoprotectant to resist detrimental effects on cellular components [40]. Proline protects plants from stress through different mechanisms, including contribution to detoxification of reactive oxygen species, cell turgor pressure maintenance, stabilization of enzymes/proteins and scavenging free radicals, to counteract the effects of osmotic stress [41]. Moreover, alkaline conditions also promote a rapid accumulation of ROS, which can cause cell damage and even death. Scavenging excessive ROS can avoid or alleviate the damage induced by stress to plant metabolism, enhancing the tolerance for alkalinity [42]. Our results suggested that the stressinduced growth inhibition of atmiox1 plants was associated with a reduced level of proline and POD activity, whereas the OX lines that accumulated more proline and had higher POD activity displayed enhanced growth responses under alkaline stress conditions. However, when plants were grown in a normal environment, we did not observe any apparent advantages for the OX lines (Fig 5B and 5C). These findings corroborate the results of J. Duan [22], who reported enhanced drought tolerance in rice (Oryza sativa L.) overexpressing OsMIOX (MIOX coding gene from upland rice). This tolerance was attributed to increased ROS scavenging enzyme activity and proline content in OX plants. Based on our results, we hypothesize that the GsMIOX1a OX plants exhibit increased alkaline stress tolerance by regulating downstream metabolism and accumulating more proline with improved POD activity, thereby counterbalancing excessive ROS, which may account for increased alkaline tolerance. In addition, proline Alkaline Stress Tolerance Gene Assay can scavenge ROS via an unclear mechanism [43], indicating that the higher levels of proline in OX plants may also scavenge ROS.
The main product of the MIOX pathway is UDP-GlcA. UDP-GlcA provides approximately 50% of the cell wall biomass and represents a key metabolite for the nucleotide sugar inter-conversion pathways [44][45][46]. UDP-GlcA is important for Arabidopsis root hair elongation [47], and the down-regulation of UDP-GlcA biosynthesis leads to swollen plant cell walls and severe developmental defects [48]. We observed enhanced tolerance towards alkaline stress in GsMIOX1a OX plants, and we hypothesized that GsMIOX1a OX improves UDP-GlcA levels, enhances root hair elongation and results in enhanced adaption to alkaline stress. However, further analysis is required to understand the MIOX-mediated cellular processes underlying alkaline stress tolerance.
To illustrate the mechanisms of altered alkaline stress tolerance conferred by the GsMIOX1a OX lines and atmiox1, we assessed the effects of GsMIOX1a OX and atmiox1 on the transcript levels of various abiotic stress-inducible marker genes (H + -Ppase, NADP-ME, KIN1 and RD29B). Under alkaline stress, all selected marker genes exhibited increased expression levels in the OX plants compared with WT plants, whereas the expression of these genes was decreased in knockout mutant (Fig 6). KIN1 and RD29B play important roles in adjusting the physiological conditions in plant cells [49]. The increased expression of these stress-inducible genes under alkaline stress conditions might contribute to improved alkaline tolerance in OX plants. H + -Ppase and NADP-ME, which function in intracellular pH regulation and acidification of the cytoplasm under stress, are induced by alkaline stress [50,51]. Therefore, the enhanced expression of H + -Ppase and NADP-ME may counterbalance the high pH and increase alkaline tolerance. However, we also should note that NADP-ME and RD29B were also induced in atmiox1. This induction may be attributed to the functional redundancy of the four MIOX genes in Arabidopsis. Therefore, the null function of a gene due to its mutation may be complemented by contributions from the other genes, and thus the induction is continuous.
The major plant AsA synthesis pathway is thought to occur via guanosine diphosphatemannose (GDP-Man) [52,53]. However, the presence of other AsA pathways in plants is controversial. Agius et al. [54] provided molecular evidence of the use of D-GlcUA as a precursor in AsA biosynthesis in a study in which the strawberry gene for D-GalUA reductase was overexpressed in Arabidopsis. Two studies have also reported the additional role of D-GlcUA in plants as a potential precursor for AsA [19,33]. Based on these studies, an animal-like alternative biosynthesis route for AsA mediated by MIOX has been proposed. To determine if AsA biosynthesis was affected by GsMIOX1a OX or atmiox1 mutant, AsA levels in WT, atmiox1 and OX lines were measured. No differences in AsA levels were observed among WT, atmiox1 and OX plants under normal and alkaline stress conditions (Fig 5F and 5G), consistent with the results of previous studies [20,21]. Therefore, a similar role of MIOX in plants and animals is questionable, and GsMIOX1a may not participate in AsA synthesis.
Taken together, these results suggest that the overexpression of GsMIOX1a, a novel gene identified in G. soja 07256, can endow Arabidopsis with alkaline tolerance, likely via the activation of proline biosynthesis, ROS scavenging and increased expression levels of stress-inducible marker genes. This study is the first to demonstrate that ectopic expression of MIOX improves alkaline tolerance in plants and will provide insight into salt-alkaline soil management.