Overexpression of AmDUF1517 enhanced tolerance to salinity,drought, and cold stress in transgenic cotton

HAO Yu-qiong, LU Guo-qing, WANG Li-hua, WANG Chun-ling, GUO Hui-ming, Ll Yi-fei, CHENG Hongmei

Abstract As abiotic stresses become more severe as a result of global climate changes, the growth and development of plants are restricted. In the development of agricultural crops with greater stress tolerance, AmDUF1517 had been isolated from the highly stress-tolerant shrub Ammopiptanthus mongolicus, and can significantly enhance stress tolerance when inserted in Arabidopsis thaliana. In this study, we inserted this gene into cotton to analyze its potential for conferring stress tolerance.Two independent transgenic cotton lines were used. Southern blot analyses indicated that AmDUF1517 was integrated into the cotton genome. Physiological analysis demonstrated that AmDUF1517-transgenic cotton had stronger resistance than the control when treated with salt, drought, and cold stresses. Further analysis showed that trans-AmDUF1517 cotton displayed significantly higher antioxidant enzyme (superoxide dismutase (SOD), peroxidase (POD), catalase (CAT), and glutathione S-transferase (GST)) activity and less reactive oxygen species (ROS) accumulation, which suggests that overexpression of AmDUF1517 can improve cotton resistance to stress by maintaining ROS homeostasis, as well as by alleviating cell membrane injury. These results imply that AmDUF1517 is a candidate gene in improving cotton resistance to abiotic stress.

Keywords: transgenic cotton, stress tolerance, AmDUF1517, Ammopiptanthus mongolicus, reactive oxygen species

1. lntroduction

Cotton (Gossypium hirsutum), an important fiber and cash crop worldwide (Lee 1984), has higher salt, drought, and cold stress tolerance than other crops, but these stresses are still major problems for cotton growth and productivity(Ashraf 2002). As global climate changes cause more severe environmental stresses, improving stress tolerance in cotton and other crops is an increasingly critical objective(Ahuja et al. 2010).

Genetic engineering is an effective method to realize this goal (Ding et al. 2014; Zeng et al. 2015). Through this method, many stress-related genes have been introduced into cotton or other plants to enhance stress tolerance.For example, overexpression of mustard annexin AnnBJ1 in transgenic cotton can significantly improve abioticstress tolerance and fiber quality (Divya and Kirti 2008).Transgenic cotton which overexpressed AvDH1 was more resistant to salt than wild type (WT) plants (Chen et al. 2016).Overexpression of ABP9, a maize bZIP transcription factor,can confer salt and drought stress tolerance in transgenic cotton plants (Wang et al. 2017). Some stress-related genes,such as dehydration-responsive element-binding proteins(DREBs), play an important role in signaling pathways and interact with cold- and dehydration-responsive elements(DREs) in plants to confer tolerance to freezing, drought,and salinity, by regulating the related signaling pathways(Kizis et al. 2001; Bouaziz et al. 2012). Many transgenic plants harboring the DREB1 were significantly more tolerant to cold, salt, and drought (Gilmour and Thomashow 2000; Wang et al. 2008; Zong et al. 2016). Transgenic Nicotiana tabacum L. overexpressing GhDREB1, which isstrongly induced by low temperature in cotton, had higher cold tolerance than WT (Shan et al. 2007). Some other important stress-related genes that edit ethylene-responsive factor (ERF) proteins are important regulators of many pathogenesis-related (PR) genes, and play important roles in stress responses (Ohmetakagi and Shinshi 1995).Some members of this family have been isolated, such as NtERF2, NtERF3, and NtERF4 from tobacco and AtERF1,AtERF2, and AtERF3 from Arabidopsis (Ohmetakagi and Shinshi 1995; Fujimoto et al. 2000; Ohta et al. 2000).Transgenic tobacco overexpressing SodERF3, isolated from sugarcane, showed higher salt and drought tolerance(Trujillo et al. 2008). The NAC gene family encodes plantspecific transcription factors that play an important role in stress response (Shah et al. 2013). In addition, some quality-improvement genes have also been introduced into cotton (Li et al. 2004; Shangguan et al. 2008). So far, many potential stress-related genes for genetic transformation have been isolated, and used for cultivating various crop varieties (Zhou et al. 2015; Wang et al. 2016).

Although many stress-resistant genes from different plant species have been identified, it is still crucial to exploit other novel genes. Ammopiptanthus mongolicus,an evergreen broadleaf shrub from the northeastern desert of China that has evolved over a long period in severe environments, can survive extreme conditions including cold, high salinity, and drought (Guo et al. 2010; Liu et al.2013; Hu et al. 2016). This strong stress tolerance to harsh environments has attracted much attention, but little research has focused on the mechanism for this stress resistance. Some temperature-related transcript-derived fragments (TDFs) from Ammopiptanthus mongolicus were isolated in our previous research. Using 5′- and 3′-RACE PCR, we cloned three full-length cDNA sequences including AmRubisco Activase, AmDUF1517, and class II small AmHSP. Transgenic A. thaliana plants that overexpress AmDUF1517, driven by a constitutive Cauliflower mosaic virus-35S promoter, were much more resistant to cold,drought and salinity when treated with cold, PEG, and NaCl in a growth chamber. Mutations of DUF1517 in Arabidopsis led to greater sensitivity to stress treatments (Gu and Cheng 2014). Other studies have shown that proteins with DUF domains can regulate sexual reproduction or polar growth of plant cells in Arabidopsis (Jonesrhoades et al. 2007; Cao et al. 2010). DUF proteins also take part in plant growth and development (Hansen 2012).

Overexpression of AmDUF1517 significantly improved tolerance to various abiotic stresses in Arabidopsis (Gu and Cheng 2014), but it is yet unknown if the same strategy could be applied to a crop species. Thus, it is important to study whether AmDUF1517 can act as a potential target gene to confer stress tolerance in transgenic cotton. In this study, we generated AmDUF1517-overexpressing cotton lines. The function of AmDUF1517 under different stress conditions was analyzed. In the present study, overexpression AmDUF1517 in cotton can improve the cotton resistance to salt, drought, and cold stress by maintaining ROS homeostasis, as well as alleviating cell membrane injury.

2. Materials and methods

2.1. Construction of plasmid, cotton transformation,and molecular analysis

Forward primer M15-F1 and reverse primer M15-R1 were used to amplify the full-length AmDUF1517 cDNA sequence containing the BamHI and SacI digestion sites. The PCR product was cloned into vector PBI121 double-digested with BamHI and SacI restriction enzymes under the control of the Cauliflower mosaic virus 35S promoter (CaMV35S)(Fig. 1-A). Plasmid PBI121-AmDUF1517 was then transferred into Agrobacterium tumefaciens strain LBA4404.The cotton transformation was conducted according to a modified Agrobacterium-mediated protocol (Zhang 2013;Fig. 1-B): Hypocotyl segments from sterile seedlings were immersed in the Agrobacterium solution (OD600=0.4) for 5 min, then co-cultured on MSB1 medium (MS basal salts plus B5vitamins medium added with 30 g L–1glucose, 0.1 mg L–1kinetin, 0.1 mg L–12,4-D, and 3 g L–1Phytagel (Sigma,St. Louis, USA) in the dark for 2 d. The explants were then transferred onto inductive medium (MSB1 medium with 50 mg L–1kanamycin and 500 mg L–1cephalosporin) for callus formation (Fig. 1-B-(1) and (2)). Resistant calluses were cultured on MS basal salts plus B5vitamins supplemented with 1.9 g L–1KNO3, 3% (m/v) glucose, and 2.8 g L–1phytagel for 2 mon until the embryonic callus was acquired (Fig. 1-B-(3)). Regenerated plants were transplanted into soil (1:1 vermiculite to humus). AmDUF1517-transgenic cotton was designated as trans-AmDUF1517. Plants were determined to be transgenic using kanamycin resistance and PCR analysis. The PCR was conducted as follows: 95°C for 5 min; 95°C for 45 s, 52°C for 45 s, 72°C for 45 s, 30 cycles;72°C for 10 min; the predicated PCR product was 906 bp.Genomic DNA was isolated from transgenic cotton and R15 using a cetyl trimethyl ammonium bromide (CTAB)extraction buffer (Clarke 2009). For Southern blot, the cotton genomic DNA of transgenic cotton (L86, L1153) and R15 plants was separately digested with BamHI and SacI digestion enzymes. A 558-bp fragment of AmDUF1517,was amplified with the M15-F2/R2 primers, and was labeled with α32P as a hybridization probe. The Southern blot was conducted as previously described (Gebbie 2014). We also analyzed the expression level of AmDUF1517 in T1trans-AmDUF1517 cotton lines (L86, L1153). Total RNA of T1trans-AmDUF1517 cotton was extracted using a plant RNA kit (Yuanpinghao Biotech, China), then reverse transcribed into cDNA using a TransScript One-Step gDNA Removal and cDNA Synthesis SuperMix Kit (TransGen Biotech, China)according to the manufacturer’s instructions. The cDNAs were used as templates to investigate the expression of AmDUF1517 in cotton by semi-quantitative RT-PCR and real time quantitative RT-PCR (qRT-PCR).

Fig. 1 Generation of transgenic cotton and molecular analyses of trans-AmDUF1517 cotton. A, T-DNA region of the plasmid PBI121-AmDUF1517 for transformation. AmDUF1517 gene was controlled by Cauliflower mosaic virus 35S (CaMV35S) promoter.NPTII, used as resistance selection. RB, T-DNA right border; LB, T-DNA left border. B, T-DNA region of the empty vector PBI121 for transformation. (1), beginning of induction of resistant calluses; (2), calluses were developed on the selective medium for 2 mon for the induction; (3), embryonic calluses were obtained; (4)–(5), embryonic calluses were developed on regeneration medium for 2–3 mon; (6), regenerated transgenic cotton 2 wk after transferring to soil. C, genetic transformation of cotton. D-(1),Southern blot analysis in trans-AmDUF1517 cotton lines (L86, L1153). Genomic DNA was digested with BamHI and SacI. D-(2),expression level of AmDUF1517 in T1 trans-AmDUF1517 cotton lines was detected by semi-quantitative RT-PCR (1) and real-time quantitative RT-PCR (2).

2.2. Stress tolerance tests in seedlings

Four-wk-old plants of the T2trans-AmDUF1517 and R15 cotton strains were used in stress treatments. Seedlings were planted in pots (diameter, 22 cm; height, 16 cm)containing soil mix (1:1 vermiculite to humus) with 16 h light/8 h dark at (27±1)°C. Plants were divided into four groups when they had 4–6 true leaves: a control group(regular watering), a salt stress (NaCl treatment) group, a drought stress (PEG treatment) group, and a cold stress(4°C) group. Each cotton line of each stress treatment consisted of three replicates (pots) (≥16 plants). For the salt and drought tolerance tests, 4-wk-old cotton plants were irrigated twice weekly with 2 000 mL of 250 mmol L–1NaCl solution or with 20% PEG solution respectively, and after 20 days, the physiological variables were measured. For the cold stress tests, 4-wk-old cotton plants were treated with 4°C for 2 d, then physiological variables were measured. For antioxidant tests, ROS content of the cotton plants in different treatments was measured, and the activities of superoxide dismutase (SOD), peroxidase (POD), catalase (CAT), and glutathione S-transferase (GST), and the relative expression levels of GhSOD, GhPOD, GhCAT, and GhGST, were also measured. The activities of four antioxidant enzymes were determined using the total superoxide dismutase (T-SOD),peroxidase, CAT, and glutathione S-transferase (GSH-ST)assay kits (Jiancheng Bioengineering Institute, Nanjing,China). The expression of ROS scavenging enzyme genes, including GhSOD, GhPOD, GhCAT, and GhGST,was measured using qRT-PCR. Cotton small-subunit rRNA(SSU) was used as a standard control. The leaves were collected for physiological measurements, frozen in liquid N2, and stored at -70°C until further use.

2.3. Expression analysis of stress-related genes in cotton plants

2.4. Physiological measurements

After the stress treatment, fresh weight (FW), turgid weight(TW), and dry weight (DW) were measured, and relative water content (RWC) was calculated by the formula RWC(%)=(FW–DW)/(TW–DW)×100 (González and González-Vilar 2001). Chlorophyll was extracted with aqueous acetone(80%) and measured spectrophotometrically using the established methods (Arnon 1949). Soluble sugar content was determined using the anthrone method with sucrose as the standard (Li and Li 2013). Free proline was extracted with toluene, and absorbance of the reaction product at 520 nm was measured colorimetrically using toluene as a blank(Abrahám et al. 2010). Malondialdehyde (MDA) content was determined according to standard methods (Schmedes and H?lmer 1989). The content of ROS (primarily HO and

22in leaves was measured according to a reported method with minor modifications (Esim et al. 2014).

2.5. Statistical analysis

All the assays above were repeated at least three times with three biological replicates. Data were counted as the means of three independent experiments±SD, and were analyzedstatistically by Duncan’s multiple range tests using SPSS software (IBM, New York, USA). Results were consideredstatistically significant when P＜0.05.

3. Results

3.1. Transformation and molecular analysis of transgenic cotton overexpressing AmDUF1517

Through the Agrobacterium-mediated transformation, we generated five AmDUF1517-overexpressing cotton lines,and because of the difference in the number of seeds,we chose two independent transgenic cotton lines for further study. Southern blot analysis showed that the gene AmDUF1517 was integrated into cotton genome in transgenic cotton lines L86 and L1153 (Fig. 1-C). We also detected the expression level of AmDUF1517 in trans-AmDUF1517 cotton lines (L86 and L1153). Semiquantitative RT-PCR result and qRT-PCR showed that both cotton lines contained the gene AmDUF1517, and the expression level of AmDUF1517 in transgenic cotton line 1153 was slighter higher than that in line 86 (Fig. 1-D).

3.2. Overexpression of AmDUF1517 in cotton enhances tolerance to high salinity, drought, and cold

Under stress conditions, trans-AmDUF1517 and R15 showed dramatic differences. RWC in trans-AmDUF1517 cotton lines was 90.5% (L86), 94.1% (L1153), and 92.7%(R15) in control conditions. With NaCl exposure, RWC in trans-AmDUF1517 cotton lines was reduced to 84.5% (L86),83% (L1153), and 73.6% in R15. For the PEG treatment,RWC in trans-AmDUF1517 cotton lines was reduced to 87.4% (L86), 83.5% (L1153), and 83.5% (R15) as compared to the control treatment. In the 4°C treatment, RWC in trans-AmDUF1517 cotton lines was reduced to 85.2% (L86),79.8% (L1153), and 72.7% (R15) (Fig. 2-A).

Total chlorophyll, soluble sugar, and free proline contents in trans-AmDUF1517 cotton lines were 29.1,34.6, and 68.2% (L86) and 31.6, 30.4, and 58.8% (L1153),respectively, higher than those in R15 under salt stress.After PEG exposure, total chlorophyll, soluble sugar, and free proline contents in trans-AmDUF1517 cotton lines were 66.1, 92.3, and 34.6% (L86), and 65.3, 80.4, and 36.4% (L1153), respectively, higher than those in R15.When treated in 4°C, the total chlorophyll, soluble sugar,and free proline levels in trans-AmDUF1517 cotton lines were 51.3, 28.7, and 37.9% (L86), and 45, 21, and 39.3%(L1153), respectively, higher than those in R15 (Fig. 2-B–D).There were no differences in relative water content, total chlorophyll, soluble sugars, and free proline content between trans-AmDUF1517 and R15 under control conditions.

When height of trans-AmDUF1517 and R15 cotton plants was measured after being exposed to salt stress for 1 wk, growth was retarded compared with the respective control lines. After 20 days, leaves of cotton R15 and trans-AmDUF1517 lines L86 and L1153 became yellow, wilted,and defoliated, whereas leaves of trans-AmDUF1517 werestill green. Plant height of trans-AmDUF1517 cotton lines was significantly greater than that of R15 at the end of the treatment, 34.5% (L86) and 29.0% (L1153) higher than that R15 (Appendix B). In drought stress, where cotton plants were treated with PEG for 20 days, the growth was also retarded compared with the untreated plants. The leaves of trans-AmDUF1517 cotton lines were greener than those of R15. Height of trans-AmDUF1517 cotton lines was 18.2% (L86) and 22.3% (L1153) higher than that of R15(Appendix B). When in the 4°C treatment, the growth of all cotton lines was retarded, and after 2 days, leaves became yellow and defoliated. R15 had more detached leaves than trans-AmDUF1517 cotton. However, plant height did not differ significantly, which may be due to the short period of the 4°C treatment.

Imprisoned as they were, they never regretted their actions. Determined to do good for the people forever, they turned themselves into four rivers, which flowed past high mountains and deep valleys, crossing the land from the west to the east and finally emptying into the sea. And so China's four great rivers were formed -- the Heilongjian (Black Dragon) in the far north, the Huanghe (Yellow River) in central China, the Changjiang (Yangtze, or Long River) farther south, and the Zhujiang (Pearl) in the very far south.

Fig. 2 Physiological assays. A, relative water content. B, chlorophyll content. C, soluble sugars content. D, proline content.Cotton seedlings were grown with regular watering, 250 mmol L–1 NaCl, 20% PEG or at 4°C. FW, fresh weight. Different letters above the columns between different cotton lines in CK, NaCl, PEG, or cold treatment show significant differences (P＜0.05). Values are means of three independent experiments; bars indicate SD.

Overall, overexpression of AmDUF1517 in transgenic cotton led to greater relative water content and total chlorophyll content, and higher accumulation of osmoprotectants understress conditions.

3.3. Oxidation resistance was significantly improved in transgenic cotton overexpressing AmDUF1517

MDA content, an indicator of oxidative stress, was 47.7%(L86) and 59.4% (L1153) lower in trans-AmDUF1517 cotton lines than those in R15 under salt stress. Under the PEG treatment, MDA content in trans-AmDUF1517 cotton lines was 30.3% (L86) and 27.1% (L1153) lower than that in R15,and was 24.6% (L86) and 18.2% (L1153), lower than that in R15 after 2-d 4°C exposure (Fig. 3-C).

Overexpression of AmDUF1517 in cotton enhanced the activities of four antioxidant enzymes (SOD, POD, CAT, and GST) in cotton, and decreased accumulation of MDA understress treatments.

3.4. Expression analysis of stress-related genes

The expression levels of several stress-related genes,including GhDREB1A, GhDREB1B, GhDREB1C, GhERF2,GhNAC3, and GhRD22 were analyzed. After abioticstress treatments, the expression levels of GhDREB1A,GhDREB1B, and GhDREB1C in trans-AmDUF1517 cotton lines were significantly higher than those in R15. There was no obvious difference in the expression levels of GhDREB1A, GhDREB1B, and GhDREB1C between trans-AmDUF1517 cotton lines and R15 in normal condition.GhERF2 expression was also significantly different between trans-AmDUF1517 cotton lines and R15 when under stress treatments. Expression was higher in trans-AmDUF1517 cotton lines than that in R15. In addition, the expression of GhRD22 was also dramatically upregulated compared with that in R15 under abiotic stress, especially after salt and PEG exposure. Under NaCl exposure, the expression of GhRD22 in trans-AmDUF1517 cotton lines was 1.79-fold (L86) and 0.56-fold (L1153) higher than that in R15.After PEG treatment, the expression of GhRD22 in trans-AmDUF1517 cotton lines was 2.84-fold (L86) and 1.48-fold(L1153) higher than that in R15. However, there was no significant difference for the expression of GhNAC3 among trans-AmDUF1517 cotton L86, L1153, and R15 under saltstress. When treated with cold stress, expression was increased in all cotton plants, and was significantly higher in trans-AmDUF1517 cotton lines than that in R15. The expression in L86 and L1153 was 43.7 and 60.0% higher than that in R15, respectively (Fig. 4).

4. Discussion

4.1. Trans-AmDUF1517 cotton behaved high tolerance to salinity, drought, and cold

With the development of cotton genetic engineering technologies, transformation has played an important role in developing new cotton cultivars. In the presentstudy, R15 was used as an acceptor for the transformation as a preponderant strain differentiated in the process of tissue culture and regeneration of Coker 312 (Wang et al. 2017). Through Agrobacterium-mediated genetic transformation, we acquired five trans-AmDUF1517 cotton lines. In the functional analysis of transgenic cotton plants overexpressing AmDUF1517, we found that trans-AmDUF1517 cotton performed better than R15 under the salt, drought, and cold stresses. After NaCl and PEG exposure, plant height of the trans-AmDUF1517 cotton lines was significantly higher than that of R15. Relative water content was substantially lower in R15 than that in trans-AmDUF1517 cotton lines, which showed higher water conservation. This advantage may confer a higher survival rate in an environment with water shortage.

High salinity, drought, and cold can cause an imbalance in cellular ions, thus inducing both ion toxicity and osmoticstress (Greenway and Munns 1980). Osmotic adjustments within cells, a fundamental regulatory mechanism, can be achieved by the accumulation of osmoprotectants. The accumulation of osmolytes, such as soluble carbohydrates and free proline, can provide an effective index to assess osmotic stress resistance in crops (Gil et al. 2011; Ogawa and Yamauchi 2015). The presence of osmolytes can enhance water retention capacity and maintain intracellularstability, reduce oxidative damage, and improve the adaptability of plants to abiotic stresses (Ashraf and Foolad 2007). Proline can also protect plant cells against increased levels of ROS (Miller et al. 2010). Here, when cotton plants were exposed to salt, drought, and cold, soluble sugars and free proline greatly increased in trans-AmDUF1517 cotton lines compared with levels in R15. The results implied that trans-AmDUF1517 cotton plants had enhanced tolerance to osmotic stress, which could reduce the injuries caused by water deficit, ion toxicity, and freezing after salt, drought,or cold treatment.

Fig. 3 Antioxidant tolerance was significantly improved in trans-AmDUF1517 cotton compared with R15. Plants were treated with water (CK), NaCl, PEG, or 4°C.content. content. C, malondialdehyde (MDA) content. D, the activities of antioxidant enzymes. (1), superoxide dismutase (SOD) activity; (2), peroxidase (POD) activity; (3), catalase (CAT) activity; (4), glutathione S-transferase (GST) activity. E, expression of the antioxidant enzymes. (1), GhSOD (DQ120514); (2), GhPOD (FJ793812); (3),GhCAT (X52135); (4), GhGST (EU074792). Different letters above the columns between different cotton lines in CK or stress treatment show significant differences (P＜0.05). Values are means of three independent experiments; bars indicate SD.

Fig. 4 Quantitative RT-PCR analysis of expression levels of stress-related genes in R15 and trans-AmDUF1517 cotton lines.A, GhDREB1A (AY321150). B, GhDREB1B (AY422828). C, GhDREB1C (GQ848094). D, GhERF2 (AY781117). E, GhRD22(AY464056). F, GhNAC3 (EU706341). Plants were treated with water (CK), NaCl, PEG, or 4°C. Total RNA was extracted from cotton leaves and then reverse transcribed into cDNA. Cotton small-subunit rRNA (SSU) was used as the control to normalize the amount of template in the reaction. Different letters above the columns between different cotton lines in CK or stress treatment show significant differences (P＜0.05). Values are means of three independent experiments; bars indicate SD.

Numerous studies have shown that abiotic stress can lead to the accumulation of ROS, primarily, andAs important signal transduction molecules, they can enhance the level of cytosolic free Ca2+, which can act as a second messenger to transmit signals (Jiang et al. 2013).However, excess levels ofandduring stress can lead to accumulation of MDA and to membrane damage(Neill et al. 2002). Membrane damage and cell membrane lipid peroxidation have been used as criteria to assess the level of abiotic stress tolerance in plants (Neill et al. 2002).SOD, POD, CAT, and GST, as important ROS scavengers,can protect plants from osmotic stress (Giannopolitis et al.1977; Azarabadi et al. 2017). In this study, ROS and MDA contents in trans-AmDUF1517 cotton lines were significantly lower than those in R15 when exposed to salt, drought, and cold. SOD, POD, CAT, and GST activities were much higher in trans-AmDUF1517 cotton lines than those in R15 after thestress exposures. Corresponding to the enzyme activity, the expression levels of their encoding genes including GhSOD,GhPOD, GhCAT, and GhGST were also significantly higher in the transgenic lines (L86, L1153) than that in R15 afterstress treatments. We speculated that trans-AmDUF1517 cotton may be able to protect plants from osmotic stress by maintaining ROS homeostasis. In addition, excess ROS and membrane damage can result in the degradation of chlorophyll (Kurepa et al. 1998). In this study, the retention of chlorophyll in trans-AmDUF1517 cotton lines was higher than that in R15, which reflected enhanced stability of the cell membrane in trans-AmDUF1517 cotton following stress treatment. The results suggested that AmDUF1517 may be involved in maintaining homeostasis of ROS through modulating the activity of the ROS-scavenging enzymes.Overexpression of AmDUF1517 in cotton significantly enhanced its ability to scavenge ROS, alleviated cell membrane injury, and improved oxidation resistance to extreme environments such as high salinity, drought, and cold.

4.2. Expression regulatory mechanism of trans-AmDUF1517 cotton in high salinity, drought, and cold

Expression of the stress-induced DREB1 genes can confer stress tolerance to transgenic plants (Shan et al.2007). Constitutive expression of either CBF1 (DREB1b)or CBF3 (DREB1a) in transgenic Arabidopsis has been shown to induce the expression of COR genes, which are cold-regulated and enhance resistance to low temperatures(Gilmour and Thomashow 2000). Transgenic rice plants overexpressing OsDREB1F enhanced tolerance to salt,drought, and cold (Wang et al. 2008). Overexpression of GhDREB1A and GhDREB1B in transgenic tobacco conferred higher tolerance to low-temperature and high-saltstress (Shan et al. 2004). In this work, overexpression of AmDUF1517 in transgenic cotton significantly improved the stress tolerance to cold, drought, and high salinity.Expression levels of GhDREB1A, GhDREB1B, and GhDREB1C were significantly higher in trans-AmDUF1517 cotton lines than those in R15 after stress treatments. The results suggested that the expressions of AmDUF1517,GhDREB1A, GhDREB1B, and GhDREB1C may play a role in the regulatory mechanism in response to abioticstresses. AmDUF1517 may take part in the modulation of the stress-related signal pathways. In addition,GhDREB1A and GhDREB1C expression was significantly upregulated in trans-AmDUF1517 cotton especially after cold stress, which suggested that AmDUF1517 may play an important role in response to cold stress. GhERF2,a positive trans-acting factor, could be significantly induced by stress treatments and regulate multiple stress responses in cotton (Jin et al. 2010). In this study, the high levels of GhERF2 in trans-AmDUF1517 cotton lines suggested AmDUF1517 may have multiple regulatory mechanisms in response to abiotic stresses. GhNAC3, astress-response gene, can be induced by cold and ABA treatments (Meng et al. 2009). In this study, expression was significantly upregulated in cold-treated cotton plants,particularly in trans-AmDUF1517 cotton. Our result was consistent with the previous finding that GhNAC3 can be induced by cold treatment, and indicated that AmDUF1517 plays an important role in cold signal pathways. The expression of GhRD22 can rise to a high level, when plants are treated with salt and drought stresses (Yue et al. 2012).In this study, the expression of GhRD22 was significantly upregulated in trans-AmDUF1517 cotton under salt and drought stresses, which reflected that there may be some conjunct signal pathways to response to salt and droughtstresses in trans-AmDUF1517 cotton. In conclusion, we speculated that AmDUF1517 may take part in the stress response pathways through regulating the expressions of these stress-related genes either directly or indirectly.There were some interrelated regulatory mechanisms in trans-AmDUF1517 cotton in the salt, drought, and cold response pathways.

4.3. The prospective

In many cases, crops may confront multiple abioticstresses at the same time, and it is thus necessary for cotton plants to display multiple stress tolerance. In ourstudy, overexpression of AmDUF1517 in cotton can greatly improve stress resistance to high salinity, drought, and cold simultaneously. Our results provide a new germplasm resource for the improvement of stress tolerance in cotton.Enhanced stress tolerance is triggered by complex signal regulatory networks involved in many related genes, and the precise mechanisms of the enhanced stress resistance in AmDUF1517-transgenic cotton are not yet clear, and experimental evidence on the function of AmDUF1517 in response to multiple stresses is lacking. Further studies should thus explore signal regulatory networks and the functions of AmDUF1517 in greater depth.

5. Conclusion

Overexpression of AmDUF1517 in cotton plants can significantly enhance its stress tolerance to salinity, drought,and cold by maintaining ROS homeostasis, decreasing the accumulation of MDA, and alleviating cell membrane injury under stress conditions. AmDUF1517 seems to be a candidate gene in improving cotton resistance to abioticstress.

Acknowledgements

This work was financially supported by the Key Project for Breeding Genetic Modified Organisms, China(2016ZX08005004) and the Agricultural Science and Technology Innovation Program of Chinese Academy of Agricultural Sciences.

Appendicesassociated with this paper can be available on http://www.ChinaAgriSci.com/V2/En/appendix.htm