Drought is one of the most serious abiotic stresses limiting crop productivity. Drought affects plant growth in many ways. Plant wilting, leaf area and biomass reduction are the most common manifestations of drought stress (Ibrahim et al., 1998), as well as photosynthesis (Ye, 2010; Zhang et al., 2011), oxidative stress (Zhao et al., 2013), and osmoregulation substances accumulation (Liu et al., 2011). In order to alleviate the adverse effects of drought stress on plants, researchers have conducted a large number of experiments, such as breeding drought-tolerant varieties and adopting modified cultivation methods, as well as adding exogenous substances. It is found that silicon not only promotes plant growth but also improves plant mineral absorption, and plays a positive role in plant cold tolerance, drought tolerance, salt tolerance, alkali tolerance, resistance to pests and diseases, and resistance to heavy metals (Li et al., 2017; Etesami, 2018), indicating the potential application of silicon fertilizer in crop production in arid regions. So far, many studies have focused on the mechanism of silicon-mediated drought resistance, including physical, physiological and biochemical aspects. Under drought stress, adding silicon can increase the size of a single cell, mainly by promoting cell water absorption and forming high expansion pressure to promote leaf expansion and growth (Liu et al., 2017). Silicon can increase stomatal conductance and net photosynthetic rate (Zhang et al., 2019), improve the performance of antioxidant system, and reduce the damage to the cell membrane of the cornea (Yang et al., 2017). Silicon may affect light energy absorption by optimizing the components of thylakoid membrane proteins in rice seedlings under drought stress and play a role in transformation and transfer (Wang et al., 2019). Silicon can be used as osmotic regulators to response to drought stress and have the function of replacing mineral nutrients as well in plants (Kang et al., 2016). In addition, silicon increased crop yield and improved grain quality (Tayyab et al., 2018). The molecular mechanism of exogenous silicon improving drought resistance is still unclear, which needs further study. High-throughput technologies, such as proteomics and RNA-Seq methods, have been widely used to identify genes or proteins differentially expressed between different tissues, varieties and treatments (Zhu et al., 2019). Several effects of silicon on gene expression under drought stress have been carried out by using these techniques. For example, silicon can enhance the expression of P5CS gene in wheat under drought stress. P5CS is a stress-inducible gene, which has the potential to improve the tolerance of wheat to drought stress (Maghsoudi et al., 2018).

In this study, RNA-Seq transcriptome sequencing was performed on RNA samples obtained from control, drought stress and silicon + drought stress treatments. Differentially expressed genes (DEG) were identified by comparing the transcription levels between the control and drought samples and between drought and silicon + drought samples. The results showed that silicon may be an inducement to induce drought tolerance of oat roots.

1 Results and Analysis

1.1 Quality of sequencing data

Illumina HiSeq 2000 system was used to sequence the RNA of oat roots under different drought stress treatments. After removing the contaminated, unknown and low-quality bases from the joints, 905 800 122 clean high-quality sequences were obtained from 915 946 044 original reads, and 710 341 844 clean reads could be mapped to the assembled transcripts. After filtering, the proportion of the Q-score not less than 20 was 97.95%~98.25%, and the proportion of Q-score not less than 30 was 93.96%~94.69%. The GC ratio (the proportion of base G and C in total base number after filtration) ranged from 54.99% to 55.62%, indicating that the sequencing results could be used for subsequent assembly and transcriptome analysis (Table 1).

1.2 Analysis of differentially expressed genes

RSEM software was used for transcriptome quantification, and the comparison of transcriptome and assembly results was completed. After the number of reads on the genes was obtained by gene expression analysis, DESeq2 software was used for statistical analysis to obtain the differentially expressed genes between groups. The differential expression genes between samples were detected with the screening condition (p-adjust <0.05 & |log₂FC| >=1). Compared with CK_R, 8 177 genes were up-regulated and 7 556 genes were down-regulated in DS_R, 5 965 genes were up-regulated and 4 483 genes were down-regulated in DS_Si_R. Compared with DS_R, 150 genes were up-regulated and 84 genes were down-regulated in DS_Si_R(Figure 1).

Analyzing the Venn diagram for the number and expression level of common differentially expressed genes and unique differentially expressed genes among the three treatments, we found that the number of common genes among the groups was 52 (Figure 2). Cluster analysis of these common genes showed that the expression patterns of common differentially expressed genes were divided into four types (Figure 3). Among the three treatments, the expression of gene category A (21 genes) was significantly up-regulated, the expression of gene category B (15 genes) was significantly up-regulated and then down-regulated, the expression of gene category C (12 genes) was significantly down-regulated and then up-regulated, and the expression of gene category D (4 genes) was significantly up-regulated and then down-regulated. The expression patterns of the four genes are shown in Figure 4.

1.3 Functional analysis of differentially expressed genes

Functional annotation was performed on the differentially expressed genes and GO Terms corresponding to the differentially expressed genes were counted. The results showed that the GO Terms of up-regulated genes and down-regulated genes were the same under drought stress, which was 57. Down-regulated differentially expressed genes include biological processes: cell wall organization (GO:0071555, 125 genes), cell wall organization or biogenesis (GO:0071554, 146 genes), external encapsulating structure organization (GO:0045229, 125 genes), hydrogen peroxide catabolic process (GO:0042744, 142 genes), hydrogen peroxide metabolic process (GO:0042743, 142 genes), reactive oxygen species metabolic process (GO:0072593, 143 genes), response to oxidative stress (GO:0006979, 154 genes); Cell components: cell wall (GO:0005618, 124 genes), external encapsulating structure (GO:0030312, 124 genes), extracellular region (GO:0005576, 324 genes); Molecular function: antioxidant activity (GO:0016209, 155 genes), carbohydrate binding (GO:0030246, 165 genes), hydrolase activity (GO:0004553, 244 genes), monooxygenase activity (GO:0004497, 166 genes), oxidoreductase activity (GO:0016684, GO:0016705, GO:0016709, 395 genes), peroxidase activity (GO:0004601, 146 genes) (Figure 5). Up-regulated differentially expressed genes include biological processes: establishment of localization in cell (GO:0051649, 40 genes), macromolecule localization (GO:0033036, 46 genes), response to oxidative stress (GO:0006979, 51 genes); Cell components: extracellular region (GO:0005576, 73 genes), intrinsic component of plasma membrane (GO:0031226, 40 genes), membrane protein complex (GO:0098796, 38 genes); Molecular function: antioxidant activity (GO:0016209, 42 genes), carbohydrate binding (GO:0030246, 104 genes), hydrolase activity (GO:0004553, 79 genes), monooxygenase activity (GO:0004497, 105 genes), oxidoreductase activity (GO:0016684, GO:0016705, GO:0016709, 188 genes), peroxidase activity (GO:0004601, 36 genes) (Figure 6). The GO Terms enrichment statistics showed that the expression of genes related to hydrogen peroxide catabolic process, hydrogen peroxide metabolic process, reactive oxygen species metabolic process, response to oxidative stress, oxidoreductase activity and peroxidase activity were mainly down-regulated.

Under silicon + drought stress, 42 GO Terms were down-regulated and 116 GO Terms were up-regulated. There were many down-regulated differentially expressed genes in terms of biological processes: cellular macromolecule metabolic process (GO:0044260, 8 genes), cellular nitrogen compound metabolic process (GO:0034641, 5 genes), cellular process (GO:0009987, 13 genes), generation of precursor metabolites and energy (GO:0006091, 5 genes), macromolecule metabolic process (GO:0043170, 9 genes), nitrogen compound metabolic process (GO:0006807, 5 genes), organic substance metabolic process (GO:0071704, 9 genes), primary metabolic process (GO:0044238, 9 genes); Cell components: chloroplast (GO:0009507, 7 genes), plastid (GO:0009536, 7 genes); Molecular function: nucleic acid binding (GO:0003676, 5 genes) (Figure 7). Up-regulated differentially expressed genes include biological processes: biological regulation (GO:0065007, 8 genes), cellular aromatic compound metabolic process (GO:0006725, 8 genes), cellular macromolecule metabolic process (GO:0044260, 7 genes), cellular nitrogen compound metabolic process (GO:0034641, 8 genes), cellular process (GO:0009987, 34 genes), heterocycle metabolic process (GO:0046483, 6 genes), macromolecule metabolic process (GO:0043170, 7 genes), nitrogen compound metabolic process (GO:0006807, 10 genes), nucleobase-containing compound metabolic process (GO:0006139, 6 genes), organic cyclic compound metabolic process (GO:1901360, 7 genes), organic substance metabolic process (GO:0071704, 25 genes), primary metabolic process (23 genes), regulation of biological process (GO:0050789, 7 genes), regulation of cellular process (GO:0050794, 7 genes), transmembrane transport (GO:0055085, 7 genes); Cell components: membrane (GO:0016020, 19 genes); Molecular function: cofactor binding (GO:0048037, 12 genes), FMN binding (GO:0010181, 6 genes), nucleic acid binding (GO:0003676, 9 genes), oxidoreductase activity (GO:0016491, 20 genes), oxidoreductase activity, acting on single donors with incorporation of molecular oxygen, acting on the aldehyde or oxo group of donors, NAD or NADP as acceptor (GO:0016701, GO:0016903, GO:0016620, 9 genes) (Figure 8). The expression of genes related to oxidoreductase activity was mainly up-regulated, and the expression of genes related to cellular nitrogen compound metabolic process and cellular macromolecule metabolic process was similar.

KEGG metabolic pathways can be divided into seven categories: metabolism, genetic information processing environmental information processing, cellular processes, organismal systems, human diseases, and drug development. The KEGG metabolic pathway analysis of differentially expressed genes in oat roots showed that 1 966 genes were annotated by metabolic pathway under drought treatment, 550 genes were annotated by genetic information processing pathway, 257 genes were annotated by environmental information processing pathway, 157 genes were annotated by cellular process pathway, 145 genes were annotated by organismal systems pathway, and 8 genes were annotated by human disease pathway. The number of genes annotated in the metabolic pathway is the largest, and the metabolic pathways include carbohydrate metabolism, like glycolysis/gluconeogenesis (ko00010, 116 genes); Amino acid metabolism, like cysteine and methionine metabolism (ko00270, 73 genes); Energy metabolism), like oxidative phosphorylation (ko00190, 67 genes); Biosynthesis of other secondary metabolites, like phenylpropanoid biosynthesis (ko00940, 256 genes); Metabolism of cofactors and vitamins, like ubiquinone and other terpenoid-quinone biosynthesis (ko00130, 26 genes) (Figure 9). Under silicon + drought treatment, 39 genes were annotated in metabolic pathway, 19 genes were annotated in genetic information processing pathway, 6 genes were annotated in environmental information processing pathway, 8 genes were annotated in cell process pathway, and 4 genes were annotated in organismal systems pathway. The metabolic pathways included the Metabolism of terpenoids and polyketides, like diterpenoid biosynthesis (ko00904, 1 gene); Metabolism of other amino acids, like cyanoamino acid metabolism (ko00460, 1 gene); Energy metabolism like oxidative phosphorylation (ko00190, 5 genes); Carbohydrate metabolism, like starch and sucrose metabolism (ko00500, 4 genes); Biosynthesis of other secondary metabolites, like phenylpropanoid biosynthesis (ko00940, 4 genes); Amino acid metabolism, like arginine and proline metabolism (ko00330, 3 genes) (Figure 10).

2 Discussion

Under drought stress, the integrity of plant membrane and the spatial interval of intracellular enzymes were destroyed, a variety of metabolic processes were affected, and the dynamic balance of various protective enzyme systems was broken (Liu et al., 2019). Based on RNA-Seq, 176 555 single genes were identified in oat roots. A total of 15 733 genes were differentially expressed between DS_R and CK_R, of which 8 177 genes were up-regulated and 7 556 genes were down-regulated. The results showed that these genes were involved in the drought resistance process at the molecular level and played a very important role. This work laid a molecular foundation for further study of the specific functions of related genes in oat under drought stress and other abiotic stress factors. In addition, GO enrichment analysis showed that some related genes such as hydrogen peroxide catabolic process, hydrogen peroxide metabolic process, reactive oxygen species metabolic process, response to oxidative stress, oxidoreductase activity and peroxidase activity were inhibited under drought stress, indicating that the antioxidant system may be destroyed under 25% PEG-6000 drought stress, which is similar to the results of alfalfa and rice under drought stress (Li et al., 2019; Lian et al., 2019), the expression of antioxidant enzyme genes were induced by drought stress.

Under drought stress after silicon application, the activities of H⁺-ATPase in roots, and pyrophosphatase, superoxide dismutase, peroxidase, catalase and glutathione reductase in plasma membrane and vacuole membrane were stimulated, and the concentration of glutathione was increased (Ming et al., 2012; Manivannan and Ahn, 2017). The up-regulation of oxidoreductase activity-related genes under silicon + drought stress indicated that silicon could up-regulate the expression of oxidoreductase gene, which was conducive to the increase of antioxidant enzyme activity and the generation and elimination of reactive oxygen species in plant cells. This is consistent with the effect of silicon in improving drought resistance of Glycine max, Phyllostachys violascens, Elaeagnus angustifolia (Li et al., 2004; Pan et al., 2013), indicating that silicon may increase drought resistance of plants by participating in antioxidant stress in plants under drought stress.

3 Materials and Methods

3.1 Materials

This test material ‘Sweety Ⅰ’ was introduced from Canada, which was provided by Beijing Baiqingyuan Animal Husbandry Technology Development Co. Ltd, and widely cultivated in Inner Mongolia. Na₂SiO₃·9H₂O (purchased from macklin) was used as the silicon source, and PEG-6000 (purchased from Solarbio) was used to simulate the drought stress environment.

3.2 Preparation of plant materials

Oat seeds were washed thoroughly in distilled water and germinated on wet filter paper at 25°C for 3 d in dark incubator. Transplanted the germinated seeds into a 96-hole plate, opened the bottom of each plate and inserted the oat root into the nutrient solution through the opening. Adjusted greenhouse lighting time to 16 h/8 h (day/night), with the relative humidity in the greenhouse of 40%~60%, the greenhouse temperature of 25°C. 1/2 Hoagland solution was used. The composition of nutrient solution was shown in Table 1. Three different seedling culture methods were as follows: (1)1/2Hoagland nutrient solution for 14 d; (2)1/2Hoagland nutrient solution for 10 d, and 25%PEG-6000+1/2Hoagland nutrient solution for 4 d; (3) 10 mmol/L Na₂SiO₃·9H₂O+1/2Hoagland nutrient solution for 10 d, 25%PEG-6000 + 1/2Hoagland nutrient solution for 4 d. There were three treatments: control (CK_R), drought stress (DS_R), and silicon + drought stress (DS_Si_R). Each treatment was repeated three times, and the nutrient solution was updated every 2 days. After the culture period (14 d), oat root samples were frozen in liquid nitrogen, and then stored at −80°C until analysis.

3.3 Preparation and sequencing of RNA-Seq library

Construction and sequencing of RNA-Seq library were performed by Shanghai Majorbio Med Tech. Co. Ltd. Total RNA was extracted from oat root samples. The mRNA molecules containing poly (A) were purified from the total RNA using IMB with poly-T oligonucleotide. The extracted mRNA fragment was reduced to 300 bp by RNA fragment buffer. Random primers were used to reversely transcribed the cut mRNA fragment into the first strand cDNA, and then the second strand cDNA was synthesized. Then, the cDNA fragment underwent a terminal repair process, adding a single 'A' base and connecting the linker sequence. Illumina paired-end sequencing of 2×150 bp library was performed.

3.4 Sequencing data processing and transcript splicing

Software SeqPrep and Sickle were used for statistics (Liu et al., 2019).

3.5 Differentially expressed genes analysis, Venn analysis and cluster analysis

Statistical analysis was performed using DESeq2 software. Venn analysis showed the number of genes in each gene set and the overlap of genes between gene sets to identify common and unique genes between gene sets. The expression pattern clustering analysis was performed on genes with similar expression patterns.

3.6 GO enrichment analysis and KEGG pathway analysis of differentially expressed genes

GO enrichment analysis was carried out with Goatools software. And functional classification was performed by KEGG database (Chen and Hou, 2019).

Authors’ Contributions

ZJ was the experimental designers and executor of this study, completing the experimental results and data analysis, paper writing and revision. YQ, YZJ, and WYQ participated in some of the experiments. WYQ was the project designer and director, guiding experimental design. All authors read and approved the final manuscript.

Acknowledgments

This study was jointly funded by the Ordos Comprehensive Experimental Station of National Forage Industry Technology System (CARS-34), the National Key Research and Development Program (2016YFC0500605), and the Special Fund for Basic Scientific Research Expenses of Central-level Public Welfare Research Institutes (Institute of Grassland Research of CAAS, 1610332018002).

References

Chen J., and Hou X., 2019, Bioinformatics analysis of differential genes between cotyledons and true leaves in Arabidopsis thaliana, Fenzi Zhiwu Yuzhong (Molecular Plant Breeding), 17(12): 3894-3901

Etesami H., 2018, Can interaction between silicon and plant growth promoting rhizobacteria benefit in alleviating abiotic and biotic stresses in crop plants?Agriculture，Ecosystems & Environment, 253: 98- 112
https://doi.org/10.1016/j.agee.2017.11.007

Ibrahim L., Proe M.F., and Cameron A.D., 1998, Interactive effects of nitrogen and water availabilities on gas exchange and whole-plant carbon allocation in poplar, Tree Physiol., 18(7): 481-487
https://doi.org/10.1093/treephys/18.7.481
PMid:12651359

Jiu C.F., Shi G.R., Yu R.G., and Zhang Z., 2017, Eco-physiological mechanisms of silicon-induced alleviation of cadmium toxicity in plants: A review, Shengtai Xuebao (Acta Ecologica Sinica), 37(23): 7799-7810

Jiu Y., Chen G.L., Cai G.F., Zhang Z.X., and Yue X., 2011, Growth and osmoregulation substances accumulation of Glycyrrhiza uralensis seedling under drought stress, Xibei Zhiwu Xuebao (Acta Botanica Boreali-Occidentalia Sinica), 31(11): 2259-2264

Kang J., Zhao W., and Zhu X., 2016, Silicon improves photosynthesis and strengthens enzyme activities in the C₃ succulent xerophyte Zygophyllum xanthoxylum under drought stress, J. Plant Physiol., 199: 76-86
https://doi.org/10.1016/j.jplph.2016.05.009
PMid:27302008

Lian L., Xu H.B., He W., Zhu Y.S., Pan L.Y., Wei Y.D., Zheng Y.M., Luo X., Xie H.A., and Zhang J.F., 2019, Expression of antioxidant enzyme genes in rice under PEG-simulated drought stress, Fujian Nongye Xuebao (Fujian Journal of Agricultural Sciences), 34(3): 255-263

Li D.D., Liang Z.S., Yang Z.Q., Xu X.X., and Han R.L., 2019, Effect of exogenous 5-aminolevulinic acid on physiological characteristics and secondary metabolite contents of alfalfa seeding under drought stress, Xibei Zhiwu Xuebao (Acta Botanica Boreali-Occidentalia Sinica), 39(10): 1827-1834

Liu J.X., Ou X.B., and Wang J.C., 2019, Effects of exogenous hydrogen peroxide (H₂O₂) on the physiological characteristics in leaves of Avena nuda L. seedlings under drought stress, Ganhanqu Nongye Yanjiu (Agricultural Research in the Arid Areas), 37(4): 146-153

Liu T., Wei W.Y., Liu J.X., Yang M., Xie H., He S.Y., Yang Q., and Wang K.Y., 2019, Chain specificity transcriptome analysis in different temperatures conditioned of Yersinia ruckeri，Shuisheng Shengwu Xuebao (Acta Hydrobiologica Sinica), 43(5): 969-976

Li P., Zhao C.Z., Zhang Y.Q., Wang X.M., Wang J.F., Wang F., and Bi Y.R., 2017, Silicon enhances the tolerance of Poa annua to cadmium by inhibiting its absorption and oxidative stress, Biol. Plant., 61(4): 741-750
https://doi.org/10.1007/s10535-017-0731-x

Li Q.F., Ma C.C., Li H.P., Xiao Y.L., and Liu X.Y., 2004, Effects of soil available silicon on growth, development and physiological functions of soybean, Yingyong Shengtai Xuebao (Chinese Journal of Applied Ecology), (1): 73-76

Maghsoudi K., Emam Y., Niazi A., Pessarakli M., and Arvin M.J., 2018, P5CS expression level and proline accumulation in the sensitive and tolerant wheat cultivars under control and drought stress conditions in the presence/absence of silicon and salicylic acid, J. Plant Interact., 13(1): 461-471
https://doi.org/10.1080/17429145.2018.1506516

Ming D.F., Yuan H.M., Wang Y.H., Gong H.J., and Zhou W.J., 2012, Effects of silicon on the physiological and biochemical characteristics of roots of rice seedlings under water stress, Zhongguo Nongye Kexue (Scientia Agricultura Sinica), 45(12): 2510-2519

Manivannan A., and Ahn Y.K., 2017, Silicon regulates potential genes involved in major physiological processes in plants to combat stress, Front. Plant Sci., 8: 1346
https://doi.org/10.3389/fpls.2017.01346
PMid:28824681 PMCid:PMC5541085

Pan Y., Rong J.Q., Sheng W.X., and Gui R.Y., 2013, Silicon and drought resistance of Phyllostachys violascens, Zhejiang Nonglin Daxue Xuebao (Journal of Zhejiang A & F University), 30(6): 852-857

Tayyab M., Islam W., and Zhang H., 2018, Promising role of silicon to enhance drought resistance in wheat, Communications in Soil Science & Plant Analysis, 49(22): 2932-2941
https://doi.org/10.1080/00103624.2018.1547394

Wang Y., Zhang B., Jiang D., and Chen G., 2019, Silicon improves photosynthetic performance by optimizing thylakoid membrane protein components in rice under drought stress, Environ. Exp. Bot., 158: 2941
https://doi.org/10.1080/00103624.2018.1547394

Yang H., Deng Y., Xu L., and Yan J., 2017, Effects of silicon on the antioxidant system and membrane stability of Saponaria officinalis under drought stress, Acta Prataculturae Sinica, 26(10): 77-86

Ye Z.P., 2010, A review on modeling of responses of photosynthesis to light and CO₂, Zhiwu Shengtai Xuebao (Chinese Journal of Plant Ecology), 34(6): 727-740

Zhang H.W., Chen B., Wen X.Y., Zhang J., Wang X.D., Li J.W., Xu Z.C., and Huang W.X., 2019, Effects of exogenous silicon on growth, leaf photosynthesis and physiological indexes of tobacco seedlings under drought stress, Shengwu Jishu Tongbao (Biotechnology Bulletin), 35(1): 17-26

Zhang R.H., Zheng Y.J., Ma G.S., Zhang X.H., Lu H.D., Shi J.T., and Xue J.Q., 2011, Effects of drought stress on photosynthetic traits and protective enzyme activity in maize seeding, Shengtai Xuebao (Acta Ecologica Sinica), 31(5): 1303-1311

Zhao B., Acharya S.N., Liu J., Zhang X., Li L., and Chen Q., 2013, Effect of water stress on protective enzyme system and yield of naked oat (Avena nuda L.), Research on Crops, 14(3): 702-710

Zhu Y., Yin J., Liang Y., and Liu J., 2019, Transcriptomic dynamics provide an insight into the mechanism for silicon-mediated alleviation of salt stress in cucumber plants, Ecotoxicol. Environ. Saf., 174: 245-254
https://doi.org/10.1016/j.ecoenv.2019.02.075
PMid:30831473