Comparative transcriptome analysis of two rice genotypes differing in their tolerance to saline-alkaline stress

Saline-alkaline stress is an abiotic stress that suppresses rice plant growth and reduces yield. However, few studies have investigated the mechanism by which rice plants respond to saline-alkaline stress at a global transcriptional level. Dongdao-4 and Jigeng-88, which differ in their tolerance to saline-alkaline stress, were used to explore gene expression differences under saline-alkaline stress by RNA-seq technology. In seedlings of Dongdao-4 and Jigeng-88, 3523 and 4066 genes with differential levels of expression were detected, respectively. A total of 799 genes were upregulated in the shoots of both Dongdao-4 and Jigeng-88, while 411 genes were upregulated in the roots of both genotypes. Among the downregulated genes in Dongdao-4 and Jigeng-88, a total of 453 and 372 genes were found in shoots and roots, respectively. Gene ontology (GO) analysis showed that upregulated genes were enriched in several GO terms such as response to stress, response to jasmonic acid, organic acid metabolic process, nicotianamine biosynthetic process, and iron homeostasis. The downregulated genes were enriched in several GO terms, such as photosynthesis and response to reactive oxygen species. Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway analysis revealed that Dongdao-4 seedlings were specifically enriched in the biosynthesis of secondary metabolites such as diterpenoids and phenylpropanoids. The upregulated genes that were involved in secondary metabolite biosynthesis, amino acid biosynthesis, betalain biosynthesis, organic acid metabolic process, and iron homeostasis pathways may be central to saline-alkaline tolerance in both rice genotypes. In contrast, the genes involved in the diterpenoid and phenylpropanoid biosynthesis pathways may contribute to the greater tolerance to saline-alkaline stress in Dongdao-4 seedlings than in Jigeng-88. These results suggest that Dongdao-4 was equipped with a more efficient mechanism involved in multiple biological processes to adapt to saline-alkaline stress.

Introduction Soil salinization-alkalization threatens agricultural production worldwide, causing considerable damage to crop growth and loss of crop yield [1,2]. Based on data from the Food and Agriculture Organization/UNESCO, 831 million ha of soils globally are affected by salinealkaline stress [3]. Due to climate change and unsustainable cultivation practices, increased salinization of arable land is expected which could result in 50% land loss by the year 2050 [4]. Rice is more sensitive to saline-alkaline stress than barley and wheat, especially in the seedling and reproductive stages [5]. Therefore, elucidation of the physiological and molecular mechanisms of the rice plant response and adaptation to saline-alkaline stress will provide important clues for the selection and breeding of crop genotypes capable of growth in saline-alkaline soils.
To adapt to salt stress, plants are equipped with several mechanisms such as compartmentation of Na + into vacuoles, increased activities of antioxidant enzymes and accumulation of compatible solutes (e.g., proline and soluble sugars) [6,7]. Transcriptome changes related to these physiological processes under salt stress have been reported [8][9][10]. Under alkaline conditions, plants are exposed to high soil pH, thereby suffering from iron (Fe) deficiency. Fe deficiency responsive genes are reported to be involved in plant adaptation to alkaline conditions [11,12]. Saline-alkaline stress is more complex than individual salt and alkaline stresses due to the presence of high levels of toxic Na + , osmotic stress, and high soil pH [1]. To adapt to saline-alkaline stress, plants have evolved numerous strategies to cope with ionic stress, osmotic stress, oxidative stress, and high soil pH. Under saline-alkaline stress conditions, maintaining cellular ion homeostasis is an important mechanism that confers tolerance in plants. Several studies have revealed that maintaining a low Na + concentration in shoots contributes to the greater tolerance of saline-alkaline stress in plants [13,14]. Osmotic homeostasis and reactive oxygen species (ROS) scavenging have been reported to play important roles in saline-alkaline stress tolerance. Li et al. (2020) [15] reported that overexpression of VaERF3 induces saline-alkaline tolerance in transgenic Arabidopsis by enhancing the accumulation of proline and the activity of antioxidant enzymes. In salinealkaline soil, plants have not only evolved mechanisms to adapt to ionic stress, osmotic stress, and oxidative stress, but have also evolved strategies to cope with nutrient stress due to the high soil pH that results from the hydrolysis of NaHCO 3 and Na 2 CO 3 . Iron acquisition, nitrate assimilation, and calcium metabolism have been reported to be critical mechanisms in tolerance to saline-alkaline stress [16][17][18]. Moreover, the accumulation of organic acids (such as citric acid) and secretion of H + was higher in saline-alkaline-tolerant plants than in saline-alkaline-sensitive plants [16,19]. In addition, the integrity of chloroplast structure and the activation of photosynthesis have been reported as crucial for saline-alkaline resistance [20].
RNA-Seq technology is a powerful tool for revealing gene networks associated with the response of plants to stress and numerous genes have been shown to be regulated by salinealkaline stress in alfalfa [21,22], alkaligrass [23], and peach [20]. However, there are few reports that have evaluated the effect of saline-alkaline stress in rice at a global transcriptional level. In our previous studies, we evaluated the effect of saline-alkaline stress on the growth and germination in rice genotypes. We found that Dongdao-4 is a saline-alkaline tolerant stress genotype due to the greater efficiency in acquiring iron under saline-alkaline conditions at the seeding stage, while Jigeng-88 is a relatively saline-alkaline-sensitive genotype [16,24,25]. In this study, these two rice genotypes differing in their tolerance to saline-alkaline stress were compared at the global transcriptional level by RNA-Seq.

Plant growth and treatment
Dongdao-4 and Jigeng-88 seeds were germinated in tap water at 37˚C for 2 days in the dark and placed on moist tissue paper for 2 days at 30˚C. The germinated seedlings were then transferred to nutrient solution containing 1.425 mM NH 4 NO 3 , 0.42 mM NaH 2 PO 4 , 0.510 mM K 2 SO 4 , 0.998 mM CaCl 2 , 1.643 mM MgSO 4 , 0.168 mM Na 2 SiO 3 , 0.125 mM Fe-EDTA, 0.019 mM H 3 BO 3 , 0.009 mM MnCl 2 , 0.155 mM CuSO 4 , 0.152 mM ZnSO 4 , and 0.075 mM Na 2 MoO 4 . The seedlings were then placed in a growth chamber maintained at 30˚C/22˚C (day/night) with a 14-h photoperiod and relative humidity maintained at approximately 70%.
To compare the effect of saline-alkaline stress on rice seedlings at the global transcriptional level, Dongdao-4 and Jigeng-88 seedlings were grown in normal culture solution for 3 weeks. Half of the seedlings were transferred to culture solution supplemented with 40 mM Na + (10 mM Na 2 CO 3 and 20 mM NaCl) for 1 day and then exposed to 60 mM Na + (10 mM Na 2 CO 3 and 40 mM NaCl) for 1 week. The remaining seedlings kept in the normal culture solution were used as control plants. The solution was refreshed daily. The pH of the control hydroponic solution was adjusted to 6.0, while the solution used for saline-alkaline treatment was adjusted to pH 8.5.

RNA isolation, cDNA library construct and Illumina deep sequencing
The methods used for RNA isolation are described in   [16]. Two biological replicates were used for each genotype. The total RNA of shoots and roots was extracted from Dongdao-4 and Jigeng-88 seedlings using RNAiso reagent (TaKaRa Biotechnology Co., Ltd., Kyoto, Japan). RNA integrity and quality were assayed with agarose gel electrophoresis and a Bioanalyzer using an Agilent RNA 6000 Nano Chip (Agilent Technologies, Santa Clara, CA, USA). Samples with RNA integrity number values >8 were used for reverse-transcription into first-strand cDNA. cDNA libraries were prepared according to the manufacturer's protocol (Illumina, San Diego, CA, USA) and assayed using the BioAnalyzer 2100 system and qPCR (Kapa Biosystems, Woburn, MA, USA). Subsequently, the libraries were sequenced on an Illumina HiSeq 2000 instrument.

Sequencing data analysis
Raw reads were filtered to remove adaptors, tags of reads and low-quality reads. Then, the remaining high-quality reads were used for mapping analysis against the reference genome sequences (ftp://ftp.ensemblgenomes.org/pub/plants/release-30, last accessed November 30, 2015) using Tophat [26]. The empirical criterion of a greater than two fold change and a significant q value (false discovery rate-adjusted P value) of <0.05 based on two independent biological replicates were used to estimate differentially expressed genes (DEGs). GO enrichment was performed using Cytoscape software (http://www.cytoscape.org, version 2.5.2) with the Bingo plugin (http://www.psb.ugent.be/cbd/papers/BiNGO/, version 2.3). Hypergeometric tests with a Benjamini Hochberg false discovery rate (FDR) were performed using the default parameters to adjust the P-value. KEGG enrichment was performed using BLAST comparisons against the KEGG GENES database KAAS (KEGG Automatic Annotation Server). RNAseq data have been deposited in the NCBI sequencing read archive (https://www.ncbi.nlm.nih. gov/bioproject/PRJNA665812; Accession: PRJNA665812).

Real-time RT-PCR
The total RNA of shoots and roots was extracted from Dongdao-4 and Jigeng-88 seedlings using RNAiso reagent and reverse-transcribed into first-strand cDNA with a PrimeScript 1 RT Reagent kit (TaKaRa Biotechnology Co., Ltd., Kyoto, Japan). Three biological replicates were used for each genotype. Real-time RT-PCR was performed following the protocols described in   [16]. All the primers used for quantitative RT-PCR are listed in S1 Table in S1 File. The relative expression level was analyzed by the comparative Ct method.

RNA-seq analysis
To gain a better understanding of the mechanism underlying tolerance to saline-alkaline stress at a global transcriptional level, transcriptomic changes in seedlings of Dongdao-4 and Jigeng-88 were investigated by RNA-Seq. A total of 905 million reads were generated. More than 869 million clean reads were retained for assembly and further analysis. Among all the reads, more than 86% had Phred-like quality scores at the Q30 level (an error probability of 0.1%) ( Table 1). The sequencing reads were quality checked, and~fifty million of the remaining high-quality reads in each transcriptome sample were mapped to unique positions on the rice reference genome (Table 1).
Furthermore, principal component analysis (PCA) was used to evaluate the relationships among transcriptome samples. As shown in Fig 1A, the results revealed that the relationships among the two biological replicates of each genotype combination were close. In addition, real-time PCR analyses were performed to validate the data from RNA-sequencing. We randomly selected 19 DEGs to verify RNA-sequencing data by real-time PCR analysis. The ratios of stress to control are presented in Fig 1B. Pearson correlation analysis showed that the results generated by the two methods were highly correlated (Pearson correlation coefficients

PLOS ONE
Comparative transcriptome analysis of Dongdao-4 and Jigeng-88 R 2 = 0.92; Fig 1B), which suggested that the RNA-sequencing data were reliable and reproducible. Both analyses indicated that the throughput and sequencing quality were appropriate for further analysis.

Identification of DEGs in response to saline-alkaline stress
Genes were deemed to be DEGs if their transcript levels were 2-fold greater (P < 0.05 in Student's t-tests) under saline-alkaline stress than under control conditions. According to this criterion, there were 2413 DEGs in the shoot of Dongdao-4, of which 1252 were upregulated and 1161 were downregulated (Fig 2A). In the shoot of Jigeng-88 seedlings, 2157 DEGs were detected, with 1302 genes upregulated and 855 downregulated genes (Fig 2A). In the case of the Dongdao-4 root, there were 1110 upregulated and 1199 downregulated genes ( Fig 2B). In the root of Jigeng-88 seedlings, 953 were upregulated and 956 genes were downregulated ( Fig 2B). A total of 473 and 523 genes were specifically upregulated in the shoots of Dongdao-4 and Jigeng-88 seedlings, respectively (Fig 3A, S2 and S3 Tables in S1 File). There were 799 genes that were induced in shoots of both rice genotypes, with 142 of these genes showing greater transcript levels in Dongdao-4 than in Jigeng-88 ( Fig 3A, S4 Table in S1 File). In addition, the transcript levels of 657 and 500 genes were specifically reduced in the shoots of Dongdao-4 and Jigeng-88 seedlings, respectively (Fig 3B, S5 and S6 Tables in S1 File). A total of 453 genes were downregulated in the shoots of both rice genotypes, with 121 of them having greater transcript levels in Jigeng-88 than Dongdao-4 (Fig 3B, S7 Table in S1 File). In the roots of Dongdao-4 and Jigeng-88, 750 and 444 genes were specifically upregulated, respectively (Fig 3C, S8 and S9 Tables in S1 File). There were 411 genes that were induced in the shoots of both rice genotypes, with 97 genes having higher transcriptional levels in Dongdao-4 than Jigeng-88 (Fig 3C, S10 Table in S1 File). Of the downregulated genes in the roots of Dongdao-4 and in Jigeng-88 seedlings, the transcription levels of 827 and 584 genes were specifically reduced, respectively (Fig 3D, S11 and S12 Tables in S1 File). A total of 372 genes were downregulated in the roots of both rice genotypes, with 83 of these genes having greater transcriptional levels in Jigeng-88 than Dongdao-4 (Fig 3D, S13 Table in S1 File).

Gene ontology (GO) categorization of DEGs
Gene Ontology (GO) analysis was conducted to obtain an overview of the functional information of DEGs. For the common upregulated DEGs, 5 and 6 GO terms associated with biological process were significantly enriched in shoots and roots, respectively ( Table 2). The higher number of enriched terms in the roots than in the shoots indicated that the roots were more sensitive to saline-alkaline stress. There were 5 GO terms in shoots which were involved in response to stress, response to jasmonic acid and molecule catabolic process (Table 2). In contrast, the 8 GO terms in roots were involved in metabolic processes such as organonitrogen compound metabolic processes, organic acid metabolic processes, the nicotianamine biosynthetic process, iron homeostasis, etc. (Table 2). Among the downregulated genes in shoots and roots, there were 3 and 2 GO terms, respectively, that were significantly enriched in biological processes ( Table 3). The 3 GO terms in shoots were involved in photosynthesis, while the 2 GO terms enriched in roots were involved in the response to reactive oxygen species.
We also conducted GO analysis of specifically upregulated genes in the shoots of Dongdao-4 and Jigeng-88 seedlings, and found that no GO terms were enriched. Among the specifically downregulated genes, a total of 3 and 9 GO terms related to biological processes were significantly enriched in the shoots of Dongdao-4 and Jigeng-88 seedlings, respectively (Fig 4A). The downregulated genes in the shoots of Dongdao-4 were involved in the cell cycle and cytokinesis, and those in Jigeng-88 were involved in photosynthesis, the pentose-phosphate shunt, and translation ( Fig 4A). In the roots of Dongdao-4 seedlings, one GO term (response to organic substance) was enriched among the specifically upregulated genes. For the 500 specifically upregulated DEGs in roots of Jigeng-88 seedlings, 8 GO terms related to biological process were significantly enriched and were involved in response to stress and hormones (Fig 4B). For the 827 specifically downregulated genes in the roots of Dongdao-4 seedlings, there were 3 GO terms that were all involved in the response to reactive oxygen species (Fig 4C). There were no enriched GO terms among the specifically downregulated genes in Jigeng-88 roots.

Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway annotation
We carried out a Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway analysis to uncover the signaling pathways involved in saline-alkaline stress in Dongdao-4 and Jigeng-88 seedlings. As shown in Fig 5A, 473 upregulated genes specifically in Dongdao-4 shoots were significantly enriched in 3 KEGG terms, i.e., diterpenoid biosynthesis, biosynthesis of secondary metabolites, and phenylpropanoid biosynthesis (Fig 5A). In contrast, all 523 upregulated

PLOS ONE
Comparative transcriptome analysis of Dongdao-4 and Jigeng-88 genes specifically only in the shoots of Jigeng-88 seedlings were significantly enriched in only one KEGG term, i.e., fatty acid degradation (Fig 5A). Among the 799 upregulated genes in the shoots of Dongdao-4 and Jigeng-88, 14 KEGG terms such as valine, leucine and isoleucine degradation, propanoate metabolism, and fatty acid degradation were significantly enriched ( Fig 5A). No KEGG term was significantly enriched for 657 specific downregulated genes in Dongdao-4 shoot. However, the 500 genes specific downregulated in Jigeng-88 shoot were significantly enriched in 3 KEGG terms, namely ribosomes, photosynthesis and porphyrin, and chlorophyll metabolism (Fig 5B). For the 453 common downregulated genes in shoots of both genotypes, 8 KEGG terms were significantly enriched with photosynthesis-antenna proteins, photosynthesis, and fatty acid elongation being the most enriched KEGG terms. (Fig 5B). The 750 root-specific upregulated genes of Dongdao-4 were assigned to 2 KEGG terms, i.e., genes related to biosynthesis of secondary metabolites and diterpenoid biosynthesis (Fig 5C), and the 444 root-specific upregulated genes of Jigeng-88 were significantly enriched in only one KEGG term, namely glutathione metabolism (Fig 5C). For the 411 common upregulated genes in roots of Dongdao-4 and Jigeng-88 seedlings, 14 KEGG terms were significantly enriched, including betalain biosynthesis, secondary metabolites, and amino acids biosynthesis (Fig 5C). For the 827 specific downregulated genes in roots of Dongdao-4 seedlings, 4 KEGG terms were significantly enriched for phenylpropanoid biosynthesis, cutin, suberin and wax biosynthesis, metabolic pathways, and biosynthesis of secondary metabolites (Fig 5D).

PLOS ONE
Comparative transcriptome analysis of Dongdao-4 and Jigeng-88 However, no KEGG terms were significantly enriched for the 584 specifically downregulated genes in root of Jigeng-88 seedlings (Fig 5D). The 372 downregulated genes in roots of both genotypes were significantly enriched in 3 KEGG terms, i.e., genes related to metabolic pathways, phenylpropanoid biosynthesis, and nitrogen metabolism (Fig 5D).

Differential expression of genes involved in iron homeostasis
Maintaining iron homeostasis is essential for rice growth under saline-alkaline conditions. Among upregulated genes in both rice genotypes, the functional category 'nicotianamine biosynthetic process' was enriched. Nicotianamine biosynthesis is critical for phytosiderophore mediated iron acquisition. We detected genes involved in the nicotianamine biosynthetic process and other processes related to iron acquisition and translocation. A total of 12 and 18 DEGs involved in iron acquisition and translocation were found in shoot and root of Dongdao-4, respectively (Table 4). In shoot and root of Jigeng-88, 9 and 18 DEGs involved in iron acquisition and translocation were identified, respectively (Table 4). In shoots of both rice genotypes, genes involved in iron homeostasis, including OsYSL2, OsYSL15, OsIRO2, OsNAAT3, OsNAS1, and OsNAS2, were significantly increased, particularly in Dongdao-4 compared to Jigeng-88 (Table 4). Genes involved in iron acquisition and translocation, such as OsIRT1, OsIRT2, OsYSL2, OsYSL15, OsIRO2, OsNAAT3, OsNAS1, and OsNAS2, were significantly increased in the roots of both genotypes, but the expression was higher in Dongdao-4 than in Jigeng-88 (Table 4).

Differential expression of genes involved in Gibberellin (GA) and phenylpropanoid biosynthesis
GA is known as an important hormone in plants during responses to abiotic stresses such as cold stress, salt stress, and drought stress. Among the genes specifically upregulated in Dongdao-4, one KEGG term, diterpenoid biosynthesis, which is involved in GA biosynthesis, was significantly enriched. A gene involved in GA biosynthesis (Os03g0856700) was upregulated by saline-alkaline stress (Table 5). In addition, among the genes specifically upregulated in Dongdao-4, genes involved in phenylpropanoid metabolism were also upregulated by salinealkaline stress (Table 5).

Discussion
In recent years, numerous studies have been conducted to elucidate the mechanisms by which plants adapt to saline-alkaline stress. However, there have been few reports that have evaluated the effect of saline-alkaline stress on rice seedlings at the global transcriptional level. We reported in our earlier study that an efficient Fe acquisition system was involved in tolerance to saline-alkaline stress in Dongdao-4 plants [16,24,25]. In the present study, the mechanisms underlying this greater saline-alkaline tolerance in Dongdao-4 were investigated at the global transcriptional level. Among these DEGs, genes involved in the response to stress, response to jasmonic acid, organic acid metabolic, iron homeostasis and betalain biosynthesis were upregulated in both Dongdao-4 and Jigeng-88. These results indicated that these common upregulated genes may be implicated in tolerance to saline-alkaline stress in both rice genotypes (Tables 2 and 4 and Fig 5). In addition, we determined that numerous regulated genes in Dongdao-4 were involved in GA and phenylpropanoid biosynthesis (Fig 5 and Table 5). The genes involved in diterpenoid and phenylpropanoid biosynthesis may underlie the greater tolerance of Dongdao-4 seedlings to saline-alkaline stress. Among the 799 common upregulated genes in shoots of both genotypes, 142 genes had magnitudes of induction that were higher in Dongdao-4 than in Jigeng-88. Among these genes, we found several transcription factors such as MYB, WRKY and MADs-box transcription factors that are involved in abiotic stress. For example, Os04g0304400 encodes OsMADS25 and imparts salinity tolerance in rice through elevated activity of antioxidant enzymes, increased accumulation of the osmoprotective solute proline, and reductions in the number of open stomata [27]. OsWRKY8, which is encoded by Os05g0583000, is involved in osmotic stress tolerance. The overexpression of OsWRKY8 increased tolerance to osmotic stress in Arabidopsis by modulating ABA-independent responsive genes [28]. Os11g0684000 encodes OsJAMyb and is a JA-inducible gene. Yokotani et al. (2013) [29] reported that OsJA-Myb not only functioned in resistance to blast infection but was also involved in osmotic adjustment, ROS removal, and ion homeostasis. In addition to transcription factors, other abiotic stress-responsive genes were observed, such as Os12g0150200, Os12g0569700, and Os02g0644000. Os12g0150200 encodes CYP94C2b and belongs to the CytP450 gene family. CYP94C2b delays stress-induced leaf senescence and induces salt tolerance in rice by inactivating JA [30]. Os12g0569700, also known as OsHsp17.0, may enhance tolerance to salt and drought stresses via production of higher proline and lower malondialdehyde levels in rice [31,32]. Os02g0466400 encodes inositol 1,3,4-trisphosphate 5/6-kinase and is essential for drought and salt stress responses in rice [33]. For the common upregulated genes in roots of both rice genotypes, there were 97 genes with expression levels higher in Dongdao-4 than in Jigeng-88. Among these genes, Os03g0230300, encoding OsSRO1c, confers greater tolerance to drought stress in overexpressing lines through interaction with SNAC1 to promote stomatal closure and reduce water loss [34][35][36].
Among the downregulated genes in both genotypes, 121 and 83 genes showed greater levels of inhibition in the shoots and roots of Dongdao-4, respectively, than in Jigeng-88 under saline-alkaline treatment. Among these genes, several genes were involved in the response to osmotic stress and phytohormone inactivation. Os09g0457100 encodes the ABA 8-hydroxylase gene OsABA8ox3, and compared to the wild type, OsABA8ox3 RNAi transgenic rice seedlings demonstrated tolerance to abiotic stress via control of ABA concentrations [37]. Os01g0182600 encodes the OsGIGANTEA (OsGI) protein, which negatively regulates osmotic stress in rice [38]. The study revealed that mutation of OsGI led to tolerance to osmotic stress due to increased proline and sucrose contents and accelerated stomatal movement [36]. These results indicated that decreases in the transcript levels of genes involved in the responses to osmotic stress and phytohormone inactivation were mechanisms that reduced the effects of the stress, thus contributing to the greater tolerance of Dongdao-4 plants.
Among the upregulated genes in shoots of both rice genotypes, a total of 88 genes were assigned to metabolic pathways. Among these genes, Os07g0154100 and Os08g0445700, have been reported to be involved in responses to abiotic stress [39][40][41][42], with the former encoding 9-cis-epoxycarotenoid dioxygenase, which is a key gene in gibberellic acid biosynthesis, and the latter encoding a stress responsive gene, trehalose-6-phosphate synthase. Expression of OsNCED4 in Arabidopsis increased ABA levels and enhanced tolerance to drought and moderate salt stress [39], and transgenic rice overexpressing OsTPSs accumulated trehalose, which could confer greater tolerance to cold, high salinity, and drought stress [40][41][42]. For the common upregulated genes in roots of both rice genotypes, genes were significantly enriched in metabolic pathways and betalain biosynthesis. However, these genes may be indirectly involved in the response to abiotic stress. Os08g0140300 and Os08g0140500 were assigned to betalain biosynthesis and are involved in serotonin biosynthesis, which plays an important role in adaptation to environmental changes [43,44]. Indeed, the overexpression of OsTDC in rice plants led to accumulation of higher levels of serotonin and delayed senescence in leaves relative to wild-type plants [43,44]. These results indicated that genes assigned to betalain biosynthesis may facilitate tolerance to saline-alkaline stress.