lable at ScienceDirect

Plant Physiology and Biochemistry 85 (2014) 31e40

Contents lists avai

Plant Physiology and Biochemistry

journal homepage: www.elsevier .com/locate/plaphy

Research article

Overexpression of codA gene confers enhanced tolerance to abioticstresses in alfalfa

Hongbing Li a, b, 1, Zhi Wang a, c, 1, Qingbo Ke a, d, Chang Yoon Ji a, d, Jae Cheol Jeong a,Haeng-Soon Lee a, d, Yong Pyo Lim c, Bingcheng Xu b, Xi-Ping Deng b, Sang-Soo Kwak a, d, *

a Plant Systems Engineering Research Center, Korea Research Institute of Bioscience and Biotechnology (KRIBB), Daejeon 305-806, Republic of Koreab State Key Laboratory of Soil Erosion and Dryland Farming on the Loess Plateau, Institute of Soil and Water Conservation, Chinese Academy of Science andMinistry of Water Resources, Northwest A & F University, Yangling 712100, Shaanxi, PR Chinac Department of Horticulture, Chunnam National University, Daejeon, Republic of Koread Department of Green Chemistry and Environmental Biotechnology, Korea University of Science & Technology, Daejeon, Republic of Korea

a r t i c l e i n f o

Article history:Received 5 September 2014Accepted 16 October 2014Available online 18 October 2014

Keywords:codAGlycinebetaineAlfalfaSWAP2 promoterSalt stressDrought stressOxidative stressBiomass

Abbreviations: codA, choline oxidase; GB, glycinebperoxidase; IAA, indole acetic acid; MS, Murashige aogen; TP, transit peptide; T-nos, nos terminator;malondialdehyde; PCR, polymerase chain reaction;time PCR; ROS, reactive oxygen species; RWC, relatibarbituric acid; TCA, trichloroacetic acid.* Corresponding author. Plant Systems Enginee

Gwahak-ro, Yuseong-gu, Daejeon 305-806, Republic oE-mail address: sskwak@kribb.re.kr (S.-S. Kwak).

1 These authors contributed to the work equally.

http://dx.doi.org/10.1016/j.plaphy.2014.10.0100981-9428/© 2014 Elsevier Masson SAS. All rights re

a b s t r a c t

We generated transgenic alfalfa plants (Medicago sativa L. cv. Xinjiang Daye) expressing a bacterial codAgene in chloroplasts under the control of the SWPA2 promoter (referred to as SC plants) and evaluatedthe plants under various abiotic stress conditions. Three transgenic plants (SC7, SC8, and SC9) wereselected for further characterization based on the strong expression levels of codA in response to methylviologen (MV)-mediated oxidative stress. SC plants showed enhanced tolerance to NaCl and droughtstress on the whole plant level due to induced expression of codA. When plants were subjected to250 mM NaCl treatment for 2 weeks, SC7 and SC8 plants maintained higher chlorophyll contents andlower malondialdehyde levels than non-transgenic (NT) plants. Under drought stress conditions, all SCplants showed enhanced tolerance to drought stress through maintaining high relative water contentsand increased levels of glycinebetaine and proline compared to NT plants. Under normal conditions, SCplants exhibited increased growth due to increased expression of auxin-related IAA genes compared toNT plants. These results suggest that the SC plants generated in this study will be useful for enhancedbiomass production on global marginal lands, such as high salinity and arid lands, yielding a sustainableagricultural product.

© 2014 Elsevier Masson SAS. All rights reserved.

1. Introduction

Environmental stress, including drought and high salinity, is oneof the most important agricultural problems affecting plant growthand crop yield. It is estimated that arid and semi-arid regions ac-count for approximately 30% of the total worldwide area and thatnearly half of all irrigated lands are salt-affected (Sivakumar et al.,2005; Mali et al., 2012). As the world population is increasingfastly and is expected to reach 9 billion by 2050 (Varshney et al.,

etaine; SWPA2, sweet potatond Skoog; MV, methyl viol-NT, non-transgenic; MDA,

q-RT-PCR, quantitative real-ve water content; TBA, thio-

ring Research Center, 125f Korea.

served.

2011), the amount of available agricultural land is shrinking dueto urbanization, industrialization, desertification, climatic change,and/or habitat use. Thus, there is a need to utilize salt- and drought-affected land to meet world food and energy requirements(Reguera et al., 2012).

Plants have evolved built-in mechanisms to cope with differenttypes of abiotic and biotic stresses imposed by unfavorable envi-ronments (Vij and Tyagi, 2007; Kathuria et al., 2009). One of thebasic strategies adopted by most plants is the accumulation of lowmolecular weight water-soluble compounds known as “compatiblesolutes” (Yancey et al., 1982; Bohnert et al., 1995; Chen and Murata,2002). In general, these compounds accumulate to high concen-trations under water or salt stress and protect plants from stressthrough different modes of action, including contributing tocellular osmotic adjustment, detoxifying reactive oxygen species(ROS), protecting membrane integrity, and stabilizing enzymes/proteins (Yancey et al., 1982; Bohnert and Jensen, 1996).

Among compatible solutes, glycinebetaine (GB) is a particularlyeffective protectant against abiotic stresses (Chen and Murata,

H. Li et al. / Plant Physiology and Biochemistry 85 (2014) 31e4032

2008). GB is a quaternary amine that accumulates in a variety ofplants, animals, and microorganisms in response to abiotic stress(Rhodes and Hanson, 1993; Chen and Murata, 2008). Levels ofaccumulated GB are generally correlated with the extent of stresstolerance (Rhodes and Hanson, 1993). GB effectively stabilizes thequaternary structures of enzymes and complex proteins, and itmaintains the highly ordered state of membranes at non-physiological temperatures and salt concentrations (Papageorgiouand Murata, 1995). In photosynthetic systems, GB stabilizes theoxygen-evolving photosystem II proteinepigment complex underhigh NaCl concentrations (Murata et al., 1992; Papageorgiou andMurata, 1995). GB can accumulate rapidly in GB-synthesizingplants in response to environmental stresses, including salinity,drought, and low temperature (Rhodes and Hanson, 1993). Exoge-nous application of GB improves the growth and survival rate ofplants under a variety of stress conditions (Park et al., 2004; Chenand Murata, 2008). Studies of GB have focused on GB-mediatedtolerance to various types of stress and at various stages of theplant lifecycle (Sulpice et al., 2003; Park et al., 2004).

Increasing evidence from a series of in vivo and in vitro studies ofthe physiology, biochemistry, genetics, and molecular biology ofplants strongly suggests that GB performs an important function inplants subjected to environmental stresses (Sakamoto and Murata,2000, 2002). Therefore, pathways for enhanced production of GBhave been introduced into transgenic plants to impart stresstolerance (Sakamoto and Murata, 2000; Chen and Murata, 2002,2008, 2011). The most successful genetic engineering of GBbiosynthesis has been achieved by expressing choline oxidase inplants, which is encoded by a soil bacterial (Arthrobacter globi-formis) codA gene and can directly convert choline into GB and H2O2(Ikuta et al., 1977; Deshnium et al., 1995). This gene has beenintroduced into various plants, such as Arabidopsis (Hayashi et al.,1997; Alia et al., 1998; Sulpice et al., 2003), Brassica (Prasad et al.,2000), tomato (Park et al., 2004; Goel et al., 2011), rice (Mohantyet al., 2002; Su et al., 2006; Kathuria et al., 2009), maize (Quanet al., 2004), and potato (Ahmad et al., 2008, 2010). Most trans-genic plants with the ability to synthesize GB show enhancedtolerance to various types of environmental stress at all stages oftheir lifecycle (Chen and Murata, 2008, 2011).

Alfalfa (Medicago sativa L.) is a perennial forage legume of greatagronomic importance worldwide, which provides abundantforage for animals and can improve soil fertility. The deep rootsystem of alfalfa can help prevent soil and water loss in dry lands(Osborn et al., 1997). As a perennial forage crop, alfalfa is a fairlyhardy species and has a relatively high level of drought and salttolerance compared with many food crops. Although alfalfa can begrown under slightly saline-alkaline conditions and in semi-aridareas, its productivity is significantly reduced under high salinityand drought stress conditions. Developing alfalfa germplasm withincreased salt and drought tolerance would provide a long-termsolution to this problem. Genetic engineering has proven to be arevolutionary technique for generating transgenic plants withenhanced tolerance to various abiotic stresses, as one can transfer adesired gene from any genetic resource and alter the expression ofexisting gene(s). Since the first report of genetic transformation inalfalfa (Deak et al., 1986), many transgenic alfalfa plants have beenobtained by gene transfer methods (Avraham et al., 2005; Zhanget al., 2005; Ram�on et al., 2009; Tang et al., 2013). However, thereare no reports on the improvement of stress tolerance in alfalfa byexpressing a bacterial codA gene, despite the fact that alfalfa is a GBaccumulating plant (Wood et al., 1991).

A robust expression system with an appropriate promoter is animportant prerequisite for efficient expression of foreign genes inplant cells. We previously isolated an oxidative stress-induciblesweet potato peroxidase (SWPA2) promoter from cell cultures of

sweet potato (Kim et al., 2003). The SWPA2 promoter exhibitshigher expression after various stress treatments than the 35Spromoter of the Cauliflower mosaic virus (CaMV 35S promoter) intransgenic tobacco and potato plants (Kim et al., 2003; Tang et al.,2006). In previous studies, we used the SWPA2 promoter togenerate various transgenic plants with enhanced tolerance tovarious environmental stresses, including sweet potato, potato, tallfescue, and poplar (Tang et al., 2006; Lee et al., 2007; Lim et al.,2007; Kim et al., 2011).

In this study, we successfully targeted choline oxidase (codA)cDNA derived from A. globiformis to alfalfa (cv. Xinjiang Daye)chloroplasts under the control of the stress-inducible SWPA2 pro-moter. The overexpression of the codA transgene resulted inenhanced tolerance to NaCl and drought stress on the whole plantlevel. To the best of our knowledge, this is the first report regardingthe engineering of a GB biosynthesis pathway in alfalfa plants.

2. Materials and methods

2.1. Plant material and plant expression vector

The Xinjiang Daye cultivar of alfalfa (Medicago sativa L.), referredto high stress resistance (Wang et al., 2009a, 2009b), was used forthe generation of transgenic alfalfa plants. Alfalfa seeds (cv. Xin-jiang Daye) were provided by Prof. Bo Zhang from Xinjiang Agri-culture University, China. The codA gene expression vector wasconstructed according to a previous description (Ahmad et al.,2008). The codA gene was kindly provided by Prof. Norio Murata,National Institute for Basic Science, Japan. In brief, the plantexpression vector pGAH/codA containing the transit peptide (TP)and the nos terminator (T-nos) under the control of the CaMV 35Spromoter (Hayashi et al., 1997) was constructed. The TP, codA, andnos terminator fragments were excised from pGAH/codA andligated to the oxidative stress-inducible SWPA2 promoter (Kimet al., 2003), after which the cassette was cloned into the HindIIIand EcoRI sites of the pCAMBIA2300 binary vector, with the NPTIIgene serving as a selectable marker. The resulting vector wastransformed into Agrobacterium tumefaciens (EHA105 strain) foralfalfa transformation.

2.2. Plant transformation and regeneration

The alfalfa seeds were surface-sterilized with 0.5% sodiumhypochloride solution for 5 min, thoroughly rinsed 7e8 times withdistilled water and germinated on Murashige and Skoog (MS)medium (pH 5.7) (Murashige and Skoog, 1962) with half-strengthof vitamins and salt under a 16/8 h light/dark cycle, with a lightintensity of 350 mmol m�2 s�1 and a relative humidity of 65% at25 ± 2 �C. After 5 days of germination, the hypocotyls were utilizedfor plant transformation via A. tumefaciens (EHA105 strain) con-taining the SWPA2 promoter-codA cassette. The transformed cellswere selected on Schenk and Hildebrandt (SH) medium (Schenkand Hildebrandt, 1972) containing 2.0 mg L�1 2, 4-dichlorophenoxyacetic acid (2, 4-D), 0.2 mg L�1 kinetin,250 mg L�1 cefotaxime and 50 mg L�1 kanamycin. Shoots wereregenerated from the calli by transferring to MS medium contain-ing 1.0 mg L�1 benzylaminopurine (BAP), 0.3 mg L�1 1-naphthylaceticacid (NAA), 250 mg L�1 cefotaxime, and 50 mg L�1

kanamycin. During the experiments, the cultures were maintainedin a culture room at 25 ± 2 �C under a 16 h photoperiod. Regen-erated shoots were transferred to MS medium for rooting. Therooted plantlets were then acclimated in pots in the growthchamber for 1 week before they were transferred to soil.

H. Li et al. / Plant Physiology and Biochemistry 85 (2014) 31e40 33

2.3. Methyl viologen treatment and ion leakage analysis

The oxidative stress tolerance assay was conducted according tothe method of Kwon et al. (2002). Six leaf discs (8 mm in diameter)collected from alfalfa plants at the same position were floated on asolution containing 0.4% (w/v) sorbitol and 5 mM MV, incubated inthe dark for 12 h to allow diffusion of the MV into the leaves andthen subjected to continuous light treatment (150 mmol m�2 s�1) at25 �C. The loss of cytoplasmic solutes following MV treatment(based on the electrical conductance of the solution) was measuredwith an ion conductivity meter (model 455C, Istek Co., Seoul, Ko-rea) over a time period ranging from 0 to 24 h and compared withthe total conductivity of the solution following tissue destruction.The extent of cellular damage was quantified by ion leakage, whichis a measure of membrane disruption.

2.4. Salt and drought stress treatments

Plants were propagated under sterile conditions in Petri dishescontaining MS medium supplemented 100 mg L�1 kanamycin. Forstress tolerance evaluations, plantlets rooted in MS medium weretransferred to pots filledwith equal quantities of soil and grown in agrowth chamber under a 16-h photoperiod at a light intensity of100 mmol m�2 s�1 and 60% (w/v) relative humidity at 25 �C. For saltand drought stress analysis on the whole plant level, 4-week-oldsimilarly sized alfalfa plants grown at 25 �C in a growth chamber in10-cm diameter pots were selected for analysis. The selected plantswere irrigated with 250 mM NaCl solution once every 3 days for 2weeks and then with tap water for 1 week. The tolerance oftransgenic plants to salt stress was estimated based on the contentsof glycinebetaine (GB), chlorophyll, and malondialdehyde (MDA) inthe leaves after treatment.

Drought stress was imposed by withholding the water supply tothe plants, which were maintained at 25 �C under a light intensityof 100 mmol photons m�2 s�1. Before withholding the water supply,the plants were irrigated with similar quantities of water from traysplaced underneath the pots for 1 week. After 1 week of waterwithholding, the plants were again irrigated and permitted torecover from the drought conditions. The degree of drought stresswas assessed by measuring the free proline content and relativewater content (RWC) in leaves after 5 days of water withholding.RWC (%) was measured using the procedures described by Ahmadet al. (2008). The following formula was utilized: RWC (%) ¼ [(FW-DW)/(TW-DW)] � 100, in which FW¼ immediate weight of freshlycollected leaves, TW ¼ turgid weight of leaves after incubation inwater for 6 h at 20 �C in the light, and DW¼ dry weight of the sameleaves after drying at 80 �C for 48 h. RWC% was measured in thefourth fully expanded leaf from the shoot apical meristem.

2.5. PCR and RT-PCR analysis

Genomic DNA was extracted from leaves of alfalfa plants usingthe protocol described by Kim and Hamada (2005). The PCR was

Table 1Gene-specific primers used for genomic and RT-PCR analysis.

cDNA Forward primer

codA (Genomic PCR) GCTGCTGGAATCGGGCTAcodA (RT-PCR) CACAACTCCTGCATCGCCTTMsIAA3 (KJ996098) GCAATTCGCCTGGACTTCCTMsIAA5 (KJ955634) TCTGTGCCAGTGGAACAAGGTMsIAA6 (KJ955635) TGCTCATCAAACCCAATCATCTMsIAA7 (KJ955636) AGCAGTAACAGCAACCATAATCMsIAA16 (KJ955637) CACAGTTCCATTGAATGTACGAActin (JQ028730.1) TCCTAGGGCTGTGTTTCCAAGT

conducted with purified genomic DNA in PCR premix (Cat. no. K-2012, Bioneer, Korea) using two convergent primers complemen-tary to the codA gene (Table 1). The amplification reactions con-sisted of 94 �C for 5 min (1 cycle), followed by 30 cycles (94 �C for30 s, 62 �C for 30 s, and 72 �C for 1 min) and a final extension cycleof 7 min at 72 �C. The PCR products were separated on a 1% agarosegel, stained with ethidium bromide, and visualized under UV. Allsubsequent experiments were conducted on cloned plants from theT0-generation of transgenic plants.

Leaves from 4-week-old plants grown in soil were treated with5 mMMV, 250 mM NaCl solution, or withdrawing water for 1 weekto activate the SWPA2 promoter and induce codA expression. TotalRNA was extracted using the TRIzol reagent (Invitrogen, Carlsbad,CA) and treated with RNase-free DNase I to remove any contami-nating genomic DNA. For semi-quantitative and quantitativeexpression analysis of codA, the actin gene and various IAA genes(MsIAA3,MsIAA5,MsIAA6,MsIAA7, andMsIAA16) (Wang et al., 2014)in alfalfa plants, total RNA (2 mg) was used to generate first-strandcDNA using an RT-PCR kit (TOPscript™ RT DryMIX) according to themanufacturer's instructions. For the Semi-quantitative RT-PCRanalysis, the PCR conditions were identical for both codA and actinas described in the genomic PCR analysis, except for the extensiontime 30-s/cycle. And the PCR products were separated and visual-ized as above description of genomic PCR analysis. Quantitativereal-time PCR (q-RT-PCR) was performed in a fluorometric thermalcycler (DNA Engine Opticon 2, MJ Research, USA) using the fluo-rescent dye EverGreen according to the manufacturer's in-structions. Transcript levels were calculated relative to the controls.Data represent means and standard errors of three biological rep-licates. The expression levels of the codA, actin, and various IAAgenes were analyzed by q-RT-PCR using the gene-specific primerslisted in Table 1. The PCR product of each alfalfa IAA gene wasconfirmed by sequencing analysis and further analyzed throughalignment with other known plant IAAs (data not shown).

2.6. Relative chlorophyll content

Chlorophyll contents were measured with a portable chloro-phyll meter (SPAD-502, Konica Minolta, Japan) from intact fullyexpanded fifth leaves (from the shoot apical meristem) of indi-vidual plants. The relative chlorophyll content after salt stress(250 mM NaCl) was determined via comparison with the chloro-phyll content under growth chamber conditions with 16-h photo-period at a light intensity of 100 mmol m�2 s�1 and 60% (w/v)relative humidity at 25 �C.

2.7. Glycinebetaine analysis

Four-week-old plants grown in soil were irrigated with 250 mMNaCl solution or subjected to water withholding for 1 week toinduce the expression of codA. The samples were collected after 3days and 5 days of salt and drought treatment, respectively. GBextraction was conducted as described by Park et al. (2004) and GB

Reverses primer Amplicon size (bp)

TGGGCTTATCGCGGAAGT 1000GTTGGTTTCCAGCCGCTTGTA 122CGTCGCCTACAAGCATCCAAT 125TGGTGACTCAGAGCCTGGTAA 119TTCTGTTTCTGCTGTTGCTACT 106CCTTGAAATTGATGAGCCAACA 143TGAGGTGGCAATTAGATGAAGA 149TGGGTGCTCTTCAGGAGCAA 200

H. Li et al. / Plant Physiology and Biochemistry 85 (2014) 31e4034

was measured via UV-VIS spectrophotometry according to Zhouet al. (2008).

2.8. Malondialdehyde analysis

Lipid peroxidation was estimated by measuring malondialde-hyde (MDA) contents according to a modified thiobarbituric acid(TBA) method (Sunarpi et al., 2005). Approximately 0.1 g of leaftissue was ground in 0.5 ml 10% trichloroacetic acid (TCA) using amortar and pestle. The homogenate was centrifuged at 10,000 rpmfor 20 min. The reaction mixture (containing 0.4 ml extract and0.6 ml TBA) was heated at 100 �C for 30 min, quickly cooled on ice,and centrifuged again at 10,000 rpm for 20 min. The absorbances at450, 532, and 600 nm were determined using an ultraviolet spec-trophotometer (Spectronic, Genesys™2, USA). Three biologicalreplicates were performed.

2.9. Free proline analysis

The free proline content of drought-treated plants wasmeasured using a spectrophotometer according to the method ofBates et al. (1973). Leaf tissue (0.1 g) was homogenized in 3 ml ofsulfosalicylic acid (3%) and centrifuged at 10,000 g for 15 min. Then,1 ml of the supernatant was added to a test tube, along with 1 mlglacial acetic acid and 1ml ninhydrin reagent. The reaction mixturewas boiled in a water bath at 100 �C for 30 min. After cooling, 2 mltoluene was added to the reaction mixture, which was then vor-texed for 30 s. The upper phase (containing proline) was measuredwith a spectrophotometer (UV-2550, Shimadzu, Japan) at 520 nmusing toluene as the blank. The proline content (mg/g FW) wasquantified by the ninhydrin acid reagent method using proline asthe standard (Bates et al., 1973).

Fig. 1. Development of transgenic alfalfa plants expressing the codA gene in chloro-plasts. (A) Schematic representation of the T-DNA regions of pScodA used for alfalfatransformation. SWPA2 pro, sweet potato peroxidase anionic 2 promoter; TP, transitpeptide from the sequence of the small subunit of Rubisco (tobacco); codA, cholineoxidase cDNA; T-nos, termination sequence from the nopaline synthase gene. (B)Genomic DNA PCR analysis of the codA gene from transgenic plants. M, size markers;PC, positive control; NT, non-transgenic plant; Numbers (1e12), independent trans-genic lines. (C) Semi-quantitative RT-PCR analysis of 12 lines expressing stable codAgene integration in transgenic plants following MV treatment.

2.10. Statistical analysis

Data were statistically analyzed with Statistical Package for theSocial Sciences (SPSS 16). Means were separated using Duncan'smultiple range test at p ¼ 0.05.

3. Results

3.1. Generation of transgenic alfalfa plants

Transgenic alfalfa plants overexpressing codA in chloroplastsunder the control of an oxidative stress-inducible SWPA2 promoter(referred to as SC plants) were successfully generated via Agro-bacterium-mediated transformation. The plant expressionconstruct included a gene encoding a transit peptide (TP), whichfacilitates the targeting of choline oxidase to the chloroplast(Fig. 1A). The putative regenerated alfalfa plants were selected onSH medium containing kanamycin. An initial screening of the pu-tative transgenic alfalfa plants was implemented by genomic PCRanalysis using codA-specific primers (Table 1). The results revealedthat the recombinant codA gene had been integrated into the ge-nomes of transgenic plants in each of the 12 independent lines,

Fig. 2. Plant growth and expression of auxin-related IAA genes in non-transgenic (NT)and transgenic (SC) plants under normal growth conditions for 1 month. (A) Plantphenotype. (B) Expression patterns of the various alfalfa IAA genes in leaves. Total RNAwas extracted and purified following the manufacturer's instructions. First-strandcDNA synthesis and PCR were conducted in accordance with the manufacturer's in-structions. The expression levels of the IAA genes were normalized to that of the alfalfaactin gene as the internal control. Data are expressed as the mean ± SD of threereplicates. Asterisks indicate a significant difference between NT and SC plants at*p < 0.05 or **p < 0.01 by t-test.

H. Li et al. / Plant Physiology and Biochemistry 85 (2014) 31e40 35

whereas no integration was detected in the NT line (Fig. 1B). Wepropagated the transgenic plant lines in a plant growth chamberand subjected them to semi-quantitative RT-PCR analysis with acodA gene-specific primer using RNA from MV-treated leaf discs.After stress induction by MV, induced expression of codA wasdetected in the leaf disc of all 12 transgenic lines, but not in the NT

Fig. 3. Effect of MV-mediated oxidative stress on the ion leakage of non-transgenic(NT) and transgenic (SC) plants at 0 and 24 h after 5 mM MV treatment. (A) Differ-ential visible damage of leaf discs. (B) Relative membrane permeability. Percentages ofrelative membrane permeability were calculated using 100% to represent values ob-tained after autoclaving. (C) Transcript levels of codA gene expression. Total RNA wasextracted after 0 and 24 h MV treatment. The expression level of codAwas normalizedto that of the alfalfa actin gene as the internal control. Data are expressed as themean ± SD of three replicates. Asterisks indicate a significant difference between NTand SC plants at *p < 0.05 or **p < 0.01 by t-test.

plants (Fig. 1C). The strongest induction of codA expression wasdetected in lines SC7, SC8, and SC9 after MV treatment. Plants fromthese three transgenic lines (referred to as SC7, SC8, and SC9 plants)were selected for further characterization.

3.2. Enhanced growth and increased expression of auxin-relatedIAA genes in SC plants

When acclimated NT and SC plants were grown in soil for 1month in the growth chamber, SC plants exhibited 1.26e1.37-foldhigher plant height than NT plants (Fig. 2A). To determinewhether the increased growth of SC plants was related to auxinbiosynthesis, we investigated the expression of auxin-responsiveIAA genes, including MsIAA3, MsIAA5, MsIAA6, MsIAA7, andMsIAA16. Under normal conditions, the SC plants showed increasedtranscript levels of five auxin-responsive IAA genes (Fig. 2B), indi-cating that overexpression of codA may play a positive role in anauxin transport-related signaling pathway.

3.3. Improved tolerance to MV-mediated oxidative stress

To evaluate the methyl viologen (MV) mediated-oxidative stresstolerance of SC plants, we treated leaf discs of 1 month old SC andNT plants with 5 mM MV solution for 24 h. MV is a non-selectiveherbicide that produces massive bursts of ROS, which disruptmembrane integrity and lead to cell death (Bowler et al., 1991).Different extents of visible damagewere observed in leaf discs fromSC and NT plants (Fig. 3A). Severe necrosis was observed in NT leafdiscs, whereas SC plants exhibited less necrosis. The ion leakagelevel of transgenic SC7, SC8, and SC9 plants was 25%, 21%, and 27%,respectively, whereas the NT plants exhibited 43% ion leakage after24 h of MV treatment (Fig. 3B). SC plants displayed significantly(p ¼ 0.05) less ion leakage than NT plants under MV-stress treat-ment. Since the heterologous gene was driven by the stress-inducible SWPA2 promoter, we examined the expression patternsof codA after MV-triggered oxidative stress by q-RT-PCR. Thetranscript levels of codA were significantly higher in transgeniclines SC7, SC8, and SC9 than in NT under MV treatment (Fig. 3C),indicating that overexpression of codA increases the tolerance toMV-mediated oxidative stress.

3.4. Enhanced tolerance of SC plants to high salinity stress

To evaluate the tolerance of SC plants to salt stress, we subjected1-month-old SC and NT plants to 250 mM NaCl treatment for 14days. This treatment had a significantly smaller impact on thegrowth of SC plants versus NT plants. Before salt treatment, thehealth status of SC and NT plants were indistinguishable (Fig. 4A).However, NT plants exhibited severely inhibited growth withyellowish leaf coloration after salt treatment, and they were dead 2weeks later, in contrast to the vigorous growth of SC plants (Fig. 4A).Since the heterologous gene was driven by the stress-inducibleSWPA2 promoter, we examined the expression patterns of codAafter salt treatment by q-RT-PCR. The transcript levels of codAsignificantly increased in transgenic SC7, SC8, and SC9 plants aftersalt treatment (Fig. 4B). To determine whether overexpression ofcodA increased the glycinebetaine (GB) contents after salt treat-ment, we measured the GB contents in leaves of alfalfa plants(Fig. 4C). Before salt treatment, the GB content in the NT leaves was7.89 mmol g�1, indicating that alfalfa is a natural accumulator of GB,whereas that in SC7, SC8, and SC9 plants was 8.8, 9.0, and7.9 mmol g�1 fresh weight, respectively. After salt treatment, SC7,SC8 and SC9 plants exhibited 1.76-, 1.17-, and 1.21-fold higher GBlevels than NT plants, respectively.

Fig. 4. Salt stress analysis of non-transgenic (NT) and transgenic (SC) plants. (A) Plant growth before salt treatment (upper panel) and after 250 mM NaCl treatment for 14 days(lower panel). (B) q-RT-PCR analysis of codA expression in leaves. Total RNA was extracted before salt stress and after 250 m NaCl treatment for 1 day. The expression level of codAwas normalized to that of the alfalfa actin gene as the internal control. (C) Glycinebetaine (GB) content of plant leaves before salt treatment and after 250 mM NaCl treatment for 3days. (D) Chlorophyll contents of plant leaves before salt treatment and after 250 mM NaCl treatment for 4 days. (E) Malondialdehyde (MDA) contents in leaves before salt treatmentand after 250 mM NaCl treatment for 1 day. Data are expressed as the mean ± SD of three replicates. Asterisks indicate a significant difference between NT and SC plants at *p < 0.05or **p < 0.01 by t-test.

H. Li et al. / Plant Physiology and Biochemistry 85 (2014) 31e4036

We then investigated the chlorophyll and malondialdehyde(MDA) contents in plants before and after salt stress treatment, asthese values represent important stress-related physiological andbiochemical parameters. Before salt stress, there was no significantdifference in chlorophyll content in the leaves of NT versus SC lines.However, after salt treatment, SC7 and SC8 plants maintained highlevels of chlorophyll (3.5%e5.2% reduction) compared with NTplants (Fig. 4D), indicating that the photosystem of SC plants is lessaffected than that of NT plants during salt stress. To test themembrane stability of the plant lines, which represents the degreeof cell membrane damage under stress, we determined the MDAcontents in these plants. Before salt stress, the MDA contents in SCtransgenic lines were similar to those of NT plants. Under 250 mMNaCl treatment for 24 h, the MDA contents of NT plants weregreater than those of SC plants overexpressing codA (Fig. 4E).

3.5. Enhanced tolerance of SC plants to drought stress

To determine whether codA is involved in the drought toleranceof transgenic alfalfa, we investigated the performance of the codA-overexpressing alfalfa plants under drought stress. Under normalgrowth conditions, there were no significant differences in plantgrowth between NT plants and the three SC transgenic lines(Fig. 5A). After drought treatment for 7 days, SC plants grew welland only some leaves turned yellow, whereas NT plants showedsevere wilting and curling. Following 3 days of rehydration afterdrought treatment, the NT plants were nearly dead, with no re-covery observed, whereas SC8 and SC9 plants fully recovered from

the dehydration conditions. The survival rates of transgenic plantsafter drought treatment were greater than those of NT plants(Fig. 5A). The transcript levels of codA in SC7, SC8, and SC9 linessignificantly increased after drought treatment (Fig. 5B). The GBcontents in the leaves of SC plants also significantly increased afterdrought stress (Fig. 5C). After drought stress, the GB contents in SCplants exhibited 1.21-, 1.39-, and 1.24-fold increases compared withthose under normal conditions, whereas the GB contents in NTplants were unchanged.

We analyzed the degree of water loss in plants after droughtstress. SC plants exhibited higher relative water content (RWC)values than NT plants under drought stress (Fig. 5D), suggestingthat SC plants exhibited less water loss than NT plants during thedehydration process. We also measured the contents of proline, atype of compatible osmolyte, in SC and NT plants. Under controlconditions, the free proline contents were not significantlydifferent between NT and SC plants. However, after withholdingwater for 5 days, SC plants accumulated higher free proline levelsthan NT plants (Fig. 5E).

4. Discussion

This study demonstrates the successful transformation of anelite alfalfa cultivar, producing plants harboring the chloroplast-targeted codA gene under the control of the oxidative stress-inducible SWPA2 promoter. The transgenic alfalfa plants (SCplants) exhibited increased tolerance to oxidative stress, as well assalt and drought stress, via induced expression of codA. This

Fig. 5. Drought stress analysis of non-transgenic (NT) and transgenic (SC) plants. (A) Plant growth before drought stress (upper panel) and after water withholding for 1 week andrehydration for 3 days (lower panel). (B) q-RT-PCR analysis of codA expression before drought stress and after water withholding for 5 days. The expression level of codA wasnormalized to that of the alfalfa actin gene as the internal control. (C) GB contents of NT and SC plants before drought stress and after water withholding for 5 days. (D) Relativewater content of NT and SC plants before drought stress and after water withholding for 5 days. (E) Proline contents in the leaves of SC and NT plants before drought stress and afterwater withholding for 5 days. Data are expressed as the mean ± SD of three replicates. Asterisks indicate a significant difference between NT and SC plants at *p < 0.05 or **p < 0.01by t-test.

H. Li et al. / Plant Physiology and Biochemistry 85 (2014) 31e40 37

observation is consistent with the results of previous studies of theheterologous expression of bacterial codA in various transgenicplants (Hayashi et al., 1997; Alia et al., 1998; Sakamoto and Murata,2000; Park et al., 2004, 2007; Ahmad et al., 2008, 2010; Kathuriaet al., 2009; Goel et al., 2011). These results indicate that engi-neering increased GB biosynthesis in alfalfa is a feasible approach toimproving its tolerance to multiple abiotic stresses.

Many approaches have been used to engineer stress tolerance inplants, with limited success (Park and Chen, 2006). One majorproblem that this type of engineering can lead to undesirable orunfavorable phenotypic changes, such as growth retardation anddecreased yields. To date, only positive effects of GB accumulationin transgenic plants have been reported, even in plants grownunder non-stress conditions (Sulpice et al., 2003; Chen and Murata,2011). In addition, exogenous treatment of GB can increase plantgrowth and yield (Xing and Rajashekar, 1999). Moreover, thecodAetransgenic plants exhibited increased production of flowersand seeds, even the size of flower buds and fruit compared with

wild-type plants (Park et al., 2007b; Chen andMurata, 2008). Understressful conditions, the performance of many types of transgenicplants is better than that of untransformed controls, suggestingthat this technique has great potential for agricultural applications(Park et al., 2007a; Chen and Murata, 2008, 2011). The plant hor-mone auxin regulates a number of cellular and developmentalprocesses, including cell division, cell growth, and differentiation(Friml, 2003). At the molecular level, auxin exerts its function byregulating the expression of numerous auxin-responsive genes,including AUX⁄IAA genes. To confirmwhether the increased growthof SC plants was related to the expression of auxin-responsivegenes, we investigate the expression of MsIAA genes in theseplants. Interestingly, the expression of auxin-responsive MsIAAgenes, including MsIAA3, MsIAA5, MsIAA6, MsIAA7, and MsIAA16,was significantly up-regulated at the transcriptional level undernormal growth conditions (Fig. 2B). However, the relationship ofcodA to auxin biosynthesis remains to be clarified to elucidate thepossible mode of action of choline oxidase.

H. Li et al. / Plant Physiology and Biochemistry 85 (2014) 31e4038

Transgenic lines harboring codA exhibited enhanced toleranceto a variety of stresses. As the codA gene was driven by a stress-inducible promoter, the induction patterns of codA underdifferent stress conditions, including MV, salt and drought, verifythe stable integration and transcription of foreign genes in SCplants (Figs. 3C, 4B and 5B).

Abiotic stresses, such as salinity and drought, which disturbredox homeostasis in plant cells, can induce oxidative stress byinducing a burst of ROS. ROS have been implicated in all types ofstresses, as they can damage lipids, proteins, DNA, and carbohy-drates. If ROS are not scavenged sufficiently in the proper period oftime, senescence and programmed cell death might occur (Apeland Hirt, 2004; Miller et al., 2010). On the other hand, scavengingtoo many ROS may negatively affect signal transduction. Therefore,balancing ROS generation and scavenging is essential for the stressresponse in plants. Despite the occurrence of GB-mediatedenhancement of stress tolerance in many plants, its mechanisticinfluence on stress signaling and perception remains unclear due toa lack of direct evidence supporting the participation of GB inantioxidative defense systems. Studies have demonstrated that GBalone does not have antioxidative activity in vitro (Smirnoff andCumbes, 1989). Thus, its ROS-scavenging function must be indi-rect. The current results, which demonstrate that the transformantshad enhanced stress tolerance, as evidenced by the lower extent ofvisible damage in leaf discs and lower levels of ion leakage underchallenging conditions imposed by 5 mM MV-mediated oxidativestress (Fig. 3), are consistent with previous findings.

Salt stress causes ROS production, which results in the accu-mulation of MDA in plants due to membrane lipid peroxidation(Gill and Tuteja, 2010). Oxidative stress-induced membrane dam-age and cell membrane stability have been used as efficient criteriato assess the degree of salt tolerance in plants (Meloni et al., 2003;Sairam et al., 2005). Our results show that the MDA contents werelower in transgenic alfalfa than in NT plants under salt stress con-ditions (Fig. 4E). These results imply that the degree of membraneinjury in transgenic plants was less than that of NT plants, which isconsistent with the enhanced salt tolerance phenotype of trans-genic alfalfa.

Under stress conditions, alfalfa plants instinctively synthesizeand accumulate more GB, since alfalfa is a natural GB accumulator.Both the SC and NT plants accumulated GB to similar levels beforestress treatment (Figs. 4C and 5C). The GB content in SC plantssignificantly increased under salt and drought stress due to theanticipated induction of codA transcriptional expression under thecontrol of stress-inducible SWPA2 promoter. The increased GBcontents in the SC plants contributed to their enhanced tolerance tooxidative, salt, and drought stresses. A series of studies usingtransgenic plants have demonstrated that GB, even at low levels,can confer transgenic plants with increased tolerance to cold,freezing, drought, and salt stress (Chen andMurata, 2008, 2011). GBlevels as low as 0.09 mmol g�1 FW can impart chilling stress toler-ance in tomato plants (Park et al., 2004), and 0.035 mmol g�1 FWGBis sufficient to impart salinity tolerance in transgenic tobacco plants(Holmstrom et al., 2000). Many studies have demonstrated that GBsynthesized in chloroplasts is more effective at imparting stresstolerance in transgenic plants that GB synthesized elsewhere (Parket al., 2007a). We assume that in the current study, codA, whichfunctions in GB synthesis, was successfully targeted to the alfalfachloroplasts, because the same transit peptide was utilized as thatreported by Hayashi et al. (1997) and Ahmad et al. (2008). Mech-anistically, GB stabilizes macromolecular activity and membraneintegrity (Sakamoto and Murata, 2002). Indeed, we detectedreduced ion leakage and MDA content in the leaf cells of codAtransgenic alfalfa (Figs. 3B and 4E) in this study, suggestingimproved cell membrane homeostasis in SC plants. Therefore, it can

be assumed that GB protects the cell membrane from stress-induced injuries, likely through alleviating oxidative stress.

Transgenic tomato plants that synthesize GB in the chloroplastexhibit higher levels of photosynthetic activity than wild-typeplants under salt stress conditions (Park et al., 2007a). Chloro-plasts are the major sites of photo oxidative reactions, whichabundantly produce ROS. The protective effect conferred by GBguards the machinery required for the degradation and synthesis ofD1 protein under stress exerted by high salt conditions (Ohnishiand Murata, 2006). In the current study, SC transgenic plantsmaintained higher levels of chlorophyll and salinity toleranceduring the entire period of salt stress than NT plants (Fig. 4).Therefore, we can conclude that the higher photosynthetic activitymaintained by SC plants versus NT plants was the result of GBsynthesis in the chloroplasts, indicating that GB synthesized inchloroplasts protects not only the photosynthetic machinery butalso the antioxidant defense mechanisms of plants.

There is evidence that high levels of salt cause an imbalance ofcellular ions, leading to both ion toxicity and osmotic stress(Greenway and Munns, 1980). High salt concentrations make itmore difficult for roots to take up water. Additionally, high saltconcentrations within the plant can be toxic due to the disturbanceof essential cellular metabolic pathways, such as protein synthesisand enzyme activity. Osmotic adjustments by accumulatingosmoprotectants inside the cell are essential for reducing thecellular osmotic potential against an osmotic gradient between rootcells and the outside saline solution, which eventually restoreswater uptake into roots during salinity stress (Greenway andMunns, 1980). When suffering from abiotic stresses, many plantsaccumulate compatible osmolytes, such as free proline and solublesugar (Liu and Zhu, 1997; Armengaud et al., 2004). These osmolytesfunction as osmoprotectants, allowing the plants to tolerate stress.In this study, we examined the contents of the osmoprotectantproline in plants. The results showed that the contents of freeproline were greater under drought stress conditions, and the freeproline levels in codA-overexpressing transgenic plants weregreater than those in NT plants (Fig. 5E). Thus, the increasedaccumulation of proline contributes to the increased droughttolerance of transgenic alfalfa.

GB is a compatible solute, also known as an osmolyte, whichperforms an important function in plants, as evidenced by theirenhanced tolerance to osmotic stress. However, the increased levelof GB in transgenic plants indicates that GB plays a role other thanosmoprotection. A variety of studies have demonstrated that GBexerts protective effects and stabilizes macromolecules, enzymeactivity, and membranes under stressful conditions (Papageorgiouand Murata, 1995; Chen and Murata, 2002; Sakamoto andMurata, 2002). Quan et al. (2004) reported that GB-synthesizingtransgenic maize withstood drought conditions better than wild-type (WT) plants, even though the GB contents in the transgenicplants were significantly lower than those of natural accumulators.Moreover, transgenic tobacco plants evidencing GB-synthesizingability also exhibit enhanced tolerance to polyethyleneglycol(PEG)-mediated osmotic stress conditions (Shen et al., 2002). Ac-cording to our data, GB-synthesizing alfalfa plants evidencedenhanced tolerance to drought stress via the maintenance of higherrelative water contents and higher free proline contents than theNT plants, and the transgenic plants recovered normal growth afterirrigation (Fig. 5).

In conclusion, GB accumulation in transgenic alfalfa plants bythe heterologous overexpression of codA significantly improvedplant tolerance to oxidative, salt, and drought stress, which wasmediated by the induction or activation of GB synthesis, increasedproline accumulation, and improved protection of membraneintegrity. Thus, this study has increased our understanding of the

H. Li et al. / Plant Physiology and Biochemistry 85 (2014) 31e40 39

molecular mechanisms of GB function in plants, although someaspects are still unclear, such as how ROS affect the biosynthesis ofGB and proline at the molecular level. The relationship between GBbiosynthesis and the expression of auxin-related IAA genes remainsto be studied. Nevertheless, the enhanced performance of thesetransgenic alfalfa lines without phenotypic defects demonstratesthat engineering the GB biosynthetic pathway is not only a feasibleapproach to improving the economically important crop alfalfa, butit also holds great promise for breeding programs of other crops. Inthis study, transgenic alfalfa overexpressing codA was significantlymore drought and salt tolerant than NT plants. We hope that thetransgenic alfalfa plants generated in this study can be used notonly as a forage crop but also as a crop for covering marginal landssuch as salt-affected or dry lands.

Author contributions

S.S. Kwak, H. Li and Z. Wang: conceived and designed the ex-periments. Z. Wang, H. Li, Q. Ke, and J.C. Jeong: performed the ex-periments. J.C. Jeong, H.S. Lee, B. Xu, and Y.P. Lim: analyzed the data.X. Deng and H.S. Lee: contributed reagents/materials/analysis tools.H. Li, Z. Wang and S.S. Kwak: wrote the paper.

Acknowledgments

This work was supported by grants from the Korea-China In-ternational Collaboration Project, the National Research Foundation(NRF) of Korea, the National Natural Science Foundation of China(no.51479189), the 111 project of the Ministry of Education, China(No. B12007) and the National Center for GM Crops (PJ008097),Biogreen21 Project for Next Generation, Rural DevelopmentAdministration, Korea.

References

Ahmad, R., Kim, M.D., Back, K.H., Kim, H.S., Lee, H.S., Kwon, S.Y., Murata, N.,Chung, W.I., Kwak, S.S., 2008. Stress-induced expression of choline oxidase inpotato plant chloroplasts confers enhanced tolerance to oxidative, salt, anddrought stresses. Plant Cell. Rep. 27, 687e698.

Ahmad, R., Kim, Y.H., Kim, M.D., Kwon, S.Y., Cho, K., Lee, H.S., Kwak, S.S., 2010.Simultaneous expression of choline oxidase, superoxide dismutase and ascor-bate peroxidase in potato plant chloroplasts provides synergistically enhancedprotection against various abiotic stresses. Physiol. Plant 138, 520e533.

Alia, Hayashi H., Chen, T.H.H., Murata, N., 1998. Transformation with a gene forcholine oxidase enhances the cold tolerance of Arabidopsis during germinationand early growth. Plant Cell. Environ. 21, 232e239.

Apel, K., Hirt, H., 2004. Reactive oxygen species: metabolism, oxidative stress, andsignal transduction. Annu. Rev. Plant Biol. 55, 373e399.

Armengaud, P., Thiery, L., Buhot, N., Grenier-de March, G., Savour�e, A., 2004. Tran-scriptional regulation of proline biosynthesis in Medicago truncatula revealsdevelopmental and environmental specific features. Physiol. Plant 120,442e450.

Avraham, T., Badani, H., Galili, S., Amir, R., 2005. Enhanced levels of methionine andcysteine in transgenic alfalfa (Medicago sativa L.) plants over-expressing theArabidopsis cystathionine g-synthase gene. Plant Biotechnol. J. 3, 71e79.

Bates, L., Waldren, R., Teare, I., 1973. Rapid determination of free proline for waterstress studies. Plant Soil 39, 205e207.

Bohnert, H.J., Jensen, R.G., 1996. Strategies for engineering waterstress tolerance inplants. Trends Biotechnol. 14, 89e97.

Bohnert, H.J., Nelson, D.E., Jensen, R.G., 1995. Adaptations to environmental stresses.Plant Cell. 7, 1099e1111.

Bowler, C., Slooten, L., Vandenbranden, S., De, R., Botterman, J.R., Synesma, C.J.,Montagu, M.V., Inze, D., 1991. Manganese superoxide dismutase can reducesellular damage mediated by oxygen radicals in transgenic plants. EMBO J. 10,1723e1732.

Chen, T.H.H., Murata, N., 2002. Enhancement of tolerance to abiotic stress bymetabolic engineering of betaines and other compatible solutes. Curr. Opin.Plant Biol. 5, 250e257.

Chen, T.H.H., Murata, N., 2008. Glycinebetaine: an effective protectant againstabiotic stress in plants. Trends Plant Sci. 13, 499e505.

Chen, T.H.H., Murata, N., 2011. Glycinebetaine protects plants against abiotic stress:mechanisms and biotechnological applications. Plant Cell. Environ. 34, 1e20.

Deak, M., Kiss, G., Koncz, C., Dudits, D., 1986. Transformation of Medicago byAgrobacterium-mediated gene transfer. Plant Cell. Rep. 5, 97e100.

Deshnium, P., Los, D.A., Hayashi, H., Mustardy, L., Murata, N., 1995. Transformationof Synechococcus with a gene for choline oxidase enhances tolerance to saltstress. Plant Mol. Biol. 29, 897e907.

Friml, J., 2003. Auxin transport-shaping the plant. Curr. Opin. Plant Biol. 6, 7e12.Gill, S.S., Tuteja, N., 2010. Reactive oxygen species and antioxidant machinery in

abiotic stress tolerance in crop plants. Plant Physiol. Biochem. 48, 909e930.Goel, D., Singh, A.K., Yadav, V., Babbar, S.B., Murata, N., Bansal, K.C., 2011. Trans-

formation of tomato with a bacterial codA gene enhances tolerance to salt andwater stresses. J. Plant Physiol. 168, 1286e1294.

Greenway, H., Munns, R., 1980. Mechanisms of salt tolerance in nonhalophytes. Ann.Rev. Plant Physiol. 31, 149e190.

Hayashi, H., Alia, Mustardy L., Deshnium, P., Ida, M., Murata, N., 1997. Trans-formation of Arabidopsis thaliana with the codA gene for choline oxidase;accumulation of glycinebetaine and enhanced tolerance to salt and cold stress.Plant J. 12, 133e142.

Holmstrom, K.O., Somersalo, S., Mandal, A., Palva, T.E., Welin, B., 2000. Improvedtolerance to salinity and low temperature in transgenic tobacco producingglycine betaine. J. Exp. Bot. 51, 177e185.

Ikuta, S., Imamura, S., Misaki, H., Horiuti, Y., 1977. Purification and characterizationof choline oxidase from Arthrobacter globiformis. J. Biochem. 82, 1741e1749.

Kathuria, H., Giri, J., Nataraja, K.N., Murata, N., Udayakumar, M., Tyagi, A.K., 2009.Glycinebetaine-induced water-stress tolerance in codA-expressing transgenicindicarice is associated with up-regulation of several stress responsive genes.Plant Biotechnol. J. 7, 512e526.

Kim, K.Y., Kwon, S.Y., Lee, H.S., Hur, Y., Bang, J.W., Kwak, S.S., 2003. A novel oxidativestress-inducible peroxidase promoter from sweetpotato: molecular cloning andcharacterization in transgenic tobacco plants and cultured cells. Plant Mol. Biol.51, 831e838.

Kim, S.H., Hamada, T., 2005. Rapid and reliable method of extracting DNA and RNAfrom sweetpotato, Ipomoea batatas (L) Lam. Biotechnol. Lett. 27, 1841e1845.

Kim, Y.H., Kim, M.D., Choi, Y.I., Park, S.C., Yun, D.J., Noh, E.W., Lee, H.S., Kwak, S.S.,2011. Transgenic poplar expressing Arabidopsis NDPK2 enhances growth as wellas oxidative stress tolerance. Plant Biotechnol. J. 9, 334e347.

Kwon, S.Y., Jeong, Y.J., Lee, H.S., Kim, J.S., Cho, K.Y., Allen, R.D., Kwak, S.S., 2002.Enhanced tolerance of transgenic tobacco plants expressing both superoxidedismutase and ascorbate peroxidase in chloroplasts against methyl viologen-mediated oxidative stress. Plant Cell. Environ. 25, 873e882.

Lee, S.H., Ahsani, N., Lee, K.W., Kim, D.H., Lee, D.G., Kwak, S.S., Kwon, S.Y., Kim, T.H.,Lee, B.H., 2007. Simultaneous overexpression of both CuZn superoxide dis-mutase and ascorbate peroxidase in transgenic tall fescue plants confersincreased tolerance to a wide range of abiotic stresses. J. Plant Physiol. 164,1626e1638.

Lim, S., Kim, Y.H., Kim, S.H., Kwon, S.Y., Lee, H.S., Kim, J.G., Cho, C.Y., Paek, K.Y.,Kwak, S.S., 2007. Enhanced tolerance of transgenic sweetpotato plantsexpressing both superoxide dismutase and ascorbate peroxidase in chloroplastsagainst methyl viologen-mediated oxidative stress. Mol. Breed. 19, 227e239.

Liu, J., Zhu, J.K., 1997. Proline accumulation and salt-stress-induced gene expressionin a salt-hypersensitive mutant of Arabidopsis. Plant Physiol. 114, 591e596.

Mali, B.S., Thengal, S.S., Pate, P.N., 2012. Physico-chemical characteristics of saltaffected soil from Barhanpur. India Ann. Biol. Res. 3, 4091e4093.

Meloni, D.A., Oliva, M.A., Martinez, C.A., Cambraia, J., 2003. Photosynthesis andactivity of superoxide dismutase, peroxidase and glutathione reductase incotton under salt stress. Environ. Exp. Bot. 49, 69e76.

Miller, G., Suzuki, N., Ciftci-Yilmaz, S., Mtttler, R., 2010. Reactive oxygen specieshomeostasis and signalling during drought and salinity stresses. Plant Cell.Environ. 33, 453e467.

Mohanty, A., Kathuria, H., Ferjani, A., Sakamoto, A., Mohanty, P., Murata, N.,Tyagi, A.K., 2002. Transgenics of an elite indica rice variety Pusa Basmati 1harbouring the codA gene are highly tolerant to salt stress. Theor. Appl. Genet.106, 51e57.

Murata, N., Mohanty, P.S., Hayashi, H., Papageorgiou, G.C., 1992. Glycinebetainestabilizes the association of extrinsic proteins with the photosynthetic oxygenevolving complex. FEBS Lett. 296, 187e189.

Murashige, T., Skoog, F., 1962. A revised medium for rapid growth and bioassayswith tobacco cultures. Physiol. Plant 15, 473e479.

Ohnishi, N., Murata, N., 2006. Glycinebetaine counteracts the inhibitory effects ofsalt stress on the degradation and synthesis of D1 protein during photo-inhibition in Synechococcus sp. PCC 7942. Plant Physiol. 141, 758e765.

Osborn, T., Brouwer, D., McCoy, T., 1997. Molecular marker analysis of alfalfa. In:McKersie, B., Brown, D. (Eds.), Biotechnology and the Improvement of ForageLegumes. CABI Publishing, Wallingford, UK, pp. 91e109.

Park, E.J., Chen, T.H.H., 2006. Improvement of cold tolerance in horticultural cropsby genetic engineering. Crop Improv. 17, 69e120.

Park, E.J., Jeknic, Z., Pino, M.T., Murata, N., Chen, T.H.H., 2007a. Glycinebetaineaccumulation is more effective in chloroplasts than in the cytosol for protectingtransgenic tomato plants against abiotic stress. Plant Cell. Environ. 30,994e1005.

Park, E.J., Jeknic, Z., Sakamota, A., Denoma, J., Yuwansiri, R., Murata, N., Chen, T.H.H.,2004. Genetic engineering of glycinebetaine synthesis in tomato protects seeds,plants, and flowers from chilling damage. Plant J. 40, 474e487.

Park, E.J., Jekni�c, Z., Chen, T.H.H., Murata, N., 2007b. The codA transgene for glyci-nebetaine synthesis increases the size of flowers and fruits in tomato. PlantBiotechnol. J. 5, 422e430.

H. Li et al. / Plant Physiology and Biochemistry 85 (2014) 31e4040

Papageorgiou, G.C., Murata, N., 1995. The unusually strong stabilizing effects ofglycinebetaine on the structure and function of the oxygen-evolving photo-system II complex. Photosynth. Res. 44, 243e252.

Prasad, K.V.S.K., Sharmila, P., Kumar, P.A., Pardha, S.P., 2000. Transformation ofBrassica juncea (L.) Czern with bacterial codA gene enhances its tolerance to saltstress. Mol. Breed. 6, 489e499.

Quan, R., Shang, M., Zhang, H., Zhao, Y., Zhang, J., 2004. Engineering of enhancedglycinebetaine synthesis improves drought tolerance in maize. Plant Bio-technol. J. 2, 477e486.

Ram�on, S., Cecilia, C., Gabriel, I., 2009. Enhanced tolerance to multiple abioticstresses in transgenic alfalfa accumulating trehalose. Crop Sci. 49, 1791e1799.

Reguera, M., Peleg, Z., Blumward, E., 2012. Targeting metabolic pathways for geneticengineering abiotic stress-tolerance in crops. Biochim. Biophys. Acta 1819,186e194.

Rhodes, D., Hanson, A.D., 1993. Quaternary ammonium and tertiary sulfoniumcompounds in higher-plants. Annu. Rev. Plant Physiol. Plant Mol. Biol. 44,357e384.

Sairam, R.K., Srivastava, G.C., Agarwal, S., Meena, R.C., 2005. Differences in antiox-idant activity in response to salinity stress in tolerant and susceptible wheatgenotypes. Biol. Plant 49, 85e91.

Sakamoto, A., Murata, N., 2000. Genetic engineering of GB synthesis in plants:current status and implications for enhancement of stress tolerance. Exp. Bot.51, 81e88.

Sakamoto, A., Murata, N., 2002. The role of glycine betaine in the protection ofplants from stress: clues from transgenic plants. Plant Cell. Environ. 25,163e171.

Schenk, R.U., Hildebrandt, A.C., 1972. Medium and techniques for induction andgrowth of monocotyledonous and dicotyledonous plant cell cultures. Can. J. Bot.50, 199e204.

Shen, Y.G., Du, B.X., Zhang, W.K., Zhand, J.S., Chen, S.Y., 2002. AhCMO regulated bystresses in Atriples hortensis, can improve drought tolerance in transgenic to-bacco. Theor. Appl. Genet. 105, 815e821.

Sivakumar, M.V.K., Das, H.P., Brunini, O., 2005. Impacts of present and future climatevariability and change of agriculture and forestry in the arid and semi-aridtropics. Clim. Change 70, 31e72.

Smirnoff, N., Cumbes, Q.J., 1989. Hydroxyl radical scavenging activity of compatiblesolutes. Phytochemistry 28, 1057e2060.

Sulpice, R., Tsukaya, H., Nonaka, H., Mustardy, L., Chen, T.H.H., Murata, N., 2003.Enhanced formation of flowers in salt-stressed Arabidopsis after genetic engi-neering of the synthesis of glycinebetaine. Plant J. 36, 165e176.

Su, J., Hirji, R., Zhang, L., He, C.K., Selvaraj, G., Wu, R., 2006. Evaluation of thestressinducible production of choline oxidase in transgenic rice as a strategy forproducing the stress-protectant glycine betaine. J. Exp. Bot. 57, 1129e1135.

Sunarpi, T., Horie, J., Motoda, M., Kubo, H., Yang, K., Yoda, R., Horie, W.Y., Chan, H.Y.,Leung, K., Hattori, M., Konomi, M., Osumi, M., Yamagami, J.I., Schroeder, N.U.,2005. Enhanced salt tolerance mediated by AtHKT1 transporter-induced Naþunloading from xylem vessels to xylem parenchyma cells. Plant J. 44, 928e938.

Tang, L., Cai, H., Ji, W., Luo, X., Wang, Z., Wu, J., Wang, X., Cui, L., Wang, Y., Zhu, Y.,Bai, X., 2013. Overexpression of GsZFP1 enhances salt and drought tolerance intransgenic alfalfa (Medicago sativa L.). Plant Physiol. Biochem. 71, 22e30.

Tang, L., Kwon, S.Y., Kim, J.S., Choi, J.S., Cho, K.Y., Sung, C.K., Kwak, S.S., Lee, H.S.,2006. Enhanced tolerance of transgenic potato plants expressing both super-oxide dismutase and ascorbate peroxidase in chloroplasts against oxidativestress and high temperature. Plant Cell. Rep. 25, 1380e1386.

Varshney, R.K., Bansal, K.C., Aggarwal, P.K., Datta, S.K., Craufurd, P.Q., 2011. Agri-cultural biotechnology for crop improvement in a variable climate: hope orhype? Trends Plant Sci. 16, 363e371.

Vij, S., Tyagi, A.K., 2007. Emerging trends in the functional genomics of the abioticstress response in crop plants. Plant Biotechnol. J. 5, 361e380.

Wang, W.B., Kim, Y.H., Lee, H.S., Deng, X.P., Kwak, S.S., 2009a. Differential anti-oxidation activities in two alfalfa cultivars under chilling stress. Plant Bio-technol. Rep. 3, 301e307.

Wang, W.B., Kim, Y.H., Lee, H.S., Kim, K.Y., Deng, X.P., Kwak, S.S., 2009b. Analysis ofantioxidant enzyme activity during germination of alfalfa under salt anddrought stresses. Plant Physiol. Biochem. 47, 570e577.

Wang, Z., Li, H.B., Ke, Q.B., Jeong, J.C., Lee, H.S., Xu, B.C., Deng, X.P., Lim, Y.P.,Kwak, S.S., 2014. Transgenic alfalfa plants expressing AtNDPK2 exhibit increasedgrowth and tolerance to abiotic stresses. Plant Physiol. Biochem. 84, 67e77.

Wood, K.V., Stringham, K.J., Smith, D.L., Volenec, J.J., Hendershot, K.L., Jackson, K.A.,Rich, P.J., Yang, W.J., Rhodes, D., 1991. Betaines of alfalfa. Characterization by fastatom bombardment and desorption chemical ionization mass spectrometry.Plant Physiol. 96, 892e897.

Xing, W., Rajashekar, C.B., 1999. Alleviation of water stress in beans by exogenousglycine betaine. Plant Sci. 148, 185e192.

Yancey, P.H., Clark, M.E., Hand, S.C., Bowlus, R.D., Somero, G.N., 1982. Living withwater stress: evolution of osmolyte systems. Science 217, 1214e1222.

Zhang, J.Y., Broeckling, C.D., Blancaflor, E.B., Sledge, M.K., Sumner, L.W., Wang, Z.Y.,2005. Overexpression of WXP1, a putative Medicago truncatula AP2 domain-containing transcription factor gene, increases cuticular wax accumulationand enhances drought tolerance in transgenic alfalfa (Medicago sativa). Plant J.42, 689e707.

Zhou, Q., Wu, Y.M., Hu, X.H., Ni, H.T., 2008. Determination of betaine content in beetroot by colorimetric method sugar crops of China. Sugar Crops China 4, 27e30.