Introduction

Lycium barbarum L. (also named Fructus lycii or wolfberry, family Solanaceae) is a perennial bush common to most areas of China, Europe, and the Mediterranean region. Its fruits have been used for centuries in China as a traditional herbal medicine and as a valuable nourishing tonic (Committee of Chinese Pharmacopoeia 1990). Recently, medical research has indicated that these fruits have many pharmacological functions such as improving vision, tonifying the liver and kidneys, nourishing blood function, resisting cancer and senescence, moisturizing the lungs, etc. (Chang and So 2007; Xie et al. 2001; Xu et al. 2000).

The worldwide increase in soil salinity and the scarcity of fresh water are two major abiotic stresses facing today’s agriculture (Mahajan and Tuteja 2005). Drought and high salinity significantly reduce the yield and fruit quality of L. barbarum (Lin et al. 2007), but improving resistance to abiotic stress by traditional breeding technology in this perennial woody plant is a long-term endeavor. Modern transgenic approaches provide a more rapid means for overcoming drought and high salinity problems. Many effective protection systems exist in plants that enable them to perceive, respond to, and properly adapt to various stress signals (Bartels and Sunkar 2005; Bray 2004; Ma and Bohnert 2007; Maggio et al. 2006; Tran et al. 2007a; Yamaguchi-Shinozaki and Shinozaki 2006; Zhu 2002), and a variety of genes and gene products have been identified that involve responses to drought and high salinity stress (Ingram and Bartels 1996; Ma and Bohnert 2007; Shinozaki and Yamaguchi-Shinozaki 2007; Tayal et al. 2004). Genes expressed under conditions of abiotic stress are anticipated to promote cellular tolerance through protective functions in the cytoplasm, such as alteration of cellular water potential to promote water uptake, control of cell membrane ion accumulation and retention, reduced production of reactive oxygen species (ROS) inside cells, and strengthening of ROS-scavenging enzymatic systems (RoyChoudhury et al. 2007; Yang et al. 2008).

Among these genes, ATHK1 (At2g17820) has been identified as coding for a protein involved in osmotic stress perception under water deficit and salt stress, and its proposed product, a histidine containing phosphotransfer protein AHP2, has also been isolated in Arabidopsis thaliana (Urao et al. 2000, 2001; Wohlbach et al. 2008). ATHK1 knockout mutants caused by T-DNA insertion in Arabidopsis indicate that ATHK1 plays an important role upstream from MAPK in the osmotic stress signal transduction pathway (Hao et al. 2004). Moreover, gain-of-function and loss-of-function approaches show ATHK1 to be a positive regulator of drought, salt stress, and abscisic acid (ABA) responses. Its transcripts accumulate rapidly when A. thaliana is exposed to external osmotic stress, suggesting an important function of ATHK1 for the efficient sensing of environmental signals (Tran et al. 2007b).

Many characteristics of L. barbarum, such as its pest and disease resistance, growth habit, and fruit quality, could be manipulated by means of genetic transformation in a short period of time. Gene transfer technology through the Agrobacterium-mediated transformation of L. barbarum leaf explants and the subsequent regeneration of transgenic plants via somatic embryogenesis has been described (Du et al. 2006; Hu et al. 2002). However, few studies have focused on increasing the tolerance of L. barbarum to high soil salinity and water deficit through transgenic approaches. In this study, we generated transgenic plants expressing ATHK1 to investigate the effect of this gene on drought and salt stress tolerance in L. barbarum.

The expression of ATHK1 in L. barbarum resulted in enhanced tolerance to high salinity and water deficit. The ATHK1 transgenic lines revealed several physiological and biochemical strategies for adaptation to drought stress conditions and salt-induced injury, which might be further characterized for better understanding of osmosensing in plants.

Material and Methods

Plant Material and Growth Conditions

A. thaliana and L. barbarum seeds were disinfected in 0.1% (w/v) HgCl2 for 10 min, then rinsed five times with sterile distilled water. Surface-sterilized Arabidopsis seeds were grown on Murashige and Skoog’s (MS) agar medium at 22°C with a 16-h day/8-h night cycle for RNA extraction.

Surface-sterilized L. barbarum seeds were germinated on 1/2 MS medium (pH 5.8) under a 16-h light/8-h dark photoperiod at 22°C for transformation. For various physiological experiments, the seeds were sown aseptically on 1/2 MS medium supplemented with 100 mg/L kanamycin. After 20 days, they were transplanted to plastic pots containing a mixture of vermiculite and turf soil (1:1, by volume) in a greenhouse under 16-h light/8-h dark cycles at 22°C. The plants were watered every 2 days.

Generation and Screening of Transgenic L. barbarum Plants

Total RNAs were extracted from Arabidopsis as described by the Arabidopsis Laboratory Manual (2004). The full-length coding sequences (CDS) of ATHK1 was amplified by PrimeScript™ RT-PCR kit (TaKaRa) using primers H1 (5′CCGCTCGAGATGCGAGGAGATAGCTTCT CA3′) and H2 (5′CCGCTCGAGTCAAGCGGACAATGAAGTTTG3′). The resulting fragment was then cloned into pKYL71 at the XhoI site to yield 2×P 35S -ATHK1, and the correct inserted direction was confirmed by sequencing.

The pKYL71 plasmid containing correct direction 2×P 35S -ATHK1 and the empty pKYL71 plasmid was transformed into L. barbarum by Agrobacterium tumefaciens LBA4404 with the leaf-disc method (Liu et al. 2007). Transgenic plants were transferred into soil.

Southern hybridization and reverse transcription polymerase chain reaction (RT-PCR) were then used to further screen transgenic plants containing ATHK1 gene. RT-PCR was carried out by using the two primers (H3: 5′CAGAGG AAACGGAAGTGATG 3′, H4: 5′AGATAGACTTGTAGTTGAGGGA 3′). The extracted DNA was digested with HindIII for 2 h, then separated in 1% agarose gel and transferred onto nylon membrane for southern hybridization. Radioactive P32-labeled probe was generated by PCR from ATHK1 cDNA with the same pair of RT-PCR primers. Hybridization and exposure were carried out according to standard methods (Sambrook and Russell 2000).

Drought, Osmotic and Salt Stress Treatments

For drought and osmotic stress treatment tests, healthy wild-type (WT) and transgenic L. barbarum plants, grown under a normal watering regime for 8 weeks, were divided into three groups, each containing ten individual WT and transgenic plants of similar size and age. The first group of plants was treated with 30% (w/w) polyethylene glycol (PEG) 6000 for 24, 48, and 72 h to simulate osmotic stress then was photographed. The second group was subjected to drought stress by omitting watering, followed by rehydration after 30 days. Observation and photos were taken at 15 and 30 days, then again 10 days after re-watering. The third group served as a control and was watered every 2 days.

For salinity resistance experiments, healthy and green 8 weeks WT and transgenic plants of similar size and age were divided into six groups, which were respectively treated with different concentration neutral salt solutions containing 0, 0.3%, 0.6%, 0.9%, 1.2%, 1.5% NaCl (NaCl weight: dry soil weight) for salt tolerance observations.

Evaluation of detached leaf water loss ratio, relative water content (RWC), and ion leakage ratio of cell membranes

The rate of water loss from excised leaves was measured as described by Jenks et al. (1994) with minor modifications. The same sized leaf sections were excised from transgenic and control lines, then submerged in distilled water, and soaked for 2 h. Excess water was then removed with filter paper, and the leaf sections were then weighed (g 0). After drying in an oven at 70°C, they were weighed again each 1 h (g t). Leaf water loss was calculated as follows: fresh weight loss (%) = (g 0 − g t)/g 0 × 100. Relative water content was carried out according to the method of Brini et al. (2007). Membrane damage was measured and was assayed by measuring ion leakage from leaf discs according to Rizhsky et al. (2002).

Measurement of MDA, H2O2, O2 −, CAT, SOD, and POD

The comparative rates of lipid peroxidation were assayed from the leaves of control and stress-treated L. barbarum seedlings by determining the levels of malondialdehyde (MDA; Velikova et al. 2000). H2O2 levels from leaf tissues of control and salt- or drought-treated L. barbarum seedlings were determined by the method of Velikova et al. (2000). O2 − production rate was measured using the hydroxylamine reaction of Elstner and Heupel (1976) with some modifications. Fresh seedling tissues were ground with 600 μl 65 mmol/L phosphate buffer (PB) and filtered. Filtrate was centrifuged at 5,000 × g for 10 min and then 300 μl of the upper phase was transferred to a new centrifuge tube, followed by the addition of 270 μl PB and 30 μl 10 mmol/L hydroxylamine chloride and incubation for 20 min. Then, 600 μl 17 mmol/L p-aminobenzene sulfonic acid and 600 μl 7 mmol/L а-naphthylamine were added to each tube. After 20 min of incubation at 25°C, an equal volume of n-butyl alcohol was added and the mixture was centrifuged again. The absorbance of the supernatant layer was read at 580 nm against a PB blank.

Catalase (CAT; EC1.11.1.6) activity was assayed by monitoring the decomposition of H2O2 at 240 nm (E = 40 mM−1 cm−1) according to the method of Aebi et al. (1974). The reaction mixture contained the appropriate extract in 50 mM potassium PB (pH 7.0), and 10 mM H2O2 was added to initiate the reaction. One unit of CAT activity was defined as the amount of protein that catalyzed the decomposition of 1 μmol H2O2 per minute. The activity of superoxide dismutase (SOD; EC1.15.1.1) was measured according to the protocol of Giannopolitis and Ries (1977) and was monitored as the inhibition of the photochemical reduction of nitro blue tetrazolium (NBT). A unit of SOD activity was defined as the amount of enzyme required to cause 50% inhibition in the reduction of NBT at 560 nm and was expressed per milligram protein. Peroxidase (POD; EC1.11.1.7) activity was measured according to the method of Velikova et al. (2000) with some modifications. The oxidation of guaiacol was measured by the increase in absorbance at 470 nm over 1 min. The reaction mixture contained 600 μl guaiacol 1% (w/v) aqueous solution, 2.8 ml of 10 mM PBS (pH 7.0), and 0.1 ml enzyme extract. The reaction was initiated with 150 μl of 100 mM H2O2.

Determination of Free Proline, Total Chlorophyll, and Soluble Protein

Determination of free proline content was performed as described by Bates et al. (1973) in leaves of similar age and size, as proline content can vary depending on leaf size and plant age. The total chlorophyll content of healthy and fully expanded leaves (of similar age and size) from untreated or 0.9% NaCl and drought-treated plants was determined spectrophotometrically according to the method described by Arnon (1949) with some modifications. Chlorophyll was extracted by soaking leaves in dimethyl sulfoxide solution for 48 h. Absorbance of extracts was measured at 635 and 645 nm with a spectrophotometer. Soluble protein was determined by the Bradford method using bovine serum albumin as a standard (Bradford 1976).

ATHK1 Gene Expression Analysis of Transgenic L. barbarum Under Different Stresses

Eight-week-old transgenic L. barbarum seedlings were treated with either 0.9% NaCl for 5 days, drought for 15 days, PEG 6000 for 48 h, or 100 μM ABA. For each plant, semiquantitative RT was performed using 2 μg total RNA per reaction (Weigel and Glazebrook 2004).

Measurement of Endogenous ABA

Endogenous hormones was extracted according to Walker-Simmons (2005), purified using Sep-Park C18 cartridges (Waters Co., USA), and stored at −20°C before detection. ABA was detected with an ELISA kit (Nanjing Agricultural University) according to the manufacturer’s instructions.

Statistical Analysis

All experiments were performed in triplicate with ten plants of similar age and size per treatment. Data represent the means of three experiments (±SD).

Results

Stable Expression of ATHK1 Full-Length CDS in L. barbarum

To improve L. barbarum responses to water deficit and salt stress, full-length CDS of ATHK1 (Fig. 1a) was cloned by RT-PCR and expressed in plasmid pKYL71 with the 2×CaMV 35S promoter. The empty pKYL71 plasmid and the pKYL71 plasmid with the resulting fragment were then transformed into L. barbarum by A. tumefaciens with the leaf-disk method. The pKYL71 vector contains the NPT-II gene (conferring resistance to kanamycin) as a selecting marker (Fig. 1b), and more than ten transgenic plants were obtained. Kanamycin-resistant transformants were confirmed as having the ATHK1 CDS integrated successfully by Southern hybridization (Fig. 1d) and that it could be stably expressed by RT-PCR (Fig. 1c).

Fig. 1
figure 1

Confirmation and analysis of transgenic L. barbarum expressing ATHK1. a Diagram of ATHK1 full-length CDS. H1 and H2 primers used for cloning of full-length CDS of ATHK1; H3 and H4 primers used for Southern blotting hybridization and RT-PCR analyses. b Schematic map of the pKYL71 construct used for L. barbarum transformation. ATHK1 full-length CDS was inserted between the 2× CaMV 35S promoter and terminator rbcs polyA at XhoI site. c RT-PCR analysis showing the expression of ATHK1 in WT, transgenic lines, and transgenic lines bearing empty vector using specific primers. ~430-bp fragment indicated that ATHK1 was constitutively expressed in transgenic L. barbarum. WT non-transformed WT plants, E transgenic plants bearing empty vector. Lanes 1 and 2 independently transformed ATHK1 plants. ACTIN: RNA was used as an internal template control. d Southern blotting analysis to confirm the presence of ATHK1. ~1,870-bp hybridization band showed that ATHK1 were successfully integrated into the L. barbarum genome. DNA was digested with HindIII, separated, and hybridized to H3, H4 probe. WT non-transformed WT lines used as control. E transgenic plants bearing empty vector. Lanes 1 and 2 DNA from different ATHK1 transformed plants

Full size image

In this experiment, WT plants and transgenic plants bearing empty vectors had no apparent differences in their morphology after 30 days dehydration stress or treatment with 0.9% NaCl, so WT plants were used as controls for the remainder of the experiments.

Enhanced Tolerance of Transgenic Plants to Drought and Osmotic Stress

Transgenic Lycium plants showed no apparent differences in their morphology compared to WT (Fig. 2a).

Fig. 2
figure 2

WT (untransformed) and transgenic L. barbarum L. plants under drought and salt stresses. a WT plants and transgenic lines were grown on soil in standard conditions. Plants were watered by capillarity every 2 days. b After 15 days of water deficit stress. c After 30 days of water deficit stress. d Ten days after re-watering, following 30 days drought treatment. e After 30% PEG6000 for 72 h. f After 10 days of 0.6% NaCl treatment. g After 10 days of 0.9% NaCl treatment. h After 10 days of 1.5% NaCl treatment

Full size image

In order to observe drought stress tolerance of transgenic plants, we compared WT lines with the ATHK1 transgenic lines over a prolonged period of drought treatment. The leaves of the transgenic plants retained their green color during the stress period, while the leaves of WT plants lost their green color and began to crinkle within 15 days of water deprivation (Fig. 2b). After 30 days of deprivation, the leaves of transgenic lines showed a slight wilting, while WT leaves became totally dried up and substantially faded (Fig. 2c). After 30 days dehydration stress, the transgenic plants recovered their vigor within 10 days of re-watering, while the WT plants started to die (Fig. 2d).

To study the response to osmotic stress in ATHK1 transformed plants, we also treated transgenic and WT plants with 30% PEG6000, a commonly used inducer of water deficit or osmotic stress (Behera et al. 2003). None of the plants showed any apparent differences after 24 h, but the leaves of the WT plants began to wither after 48 h and were obviously faded by 72 h compared to the transgenic plants (Fig. 2e). This result showed that WT plants were more sensitive to osmotic stress than were transgenic lines.

Increased Tolerance to Salt Stress in Transgenic Plants

To test whether ATHK1 expression could enhance salt tolerance, 8-week seedlings were subjected to 0, 0.3%, 0.6%, 0.9%, 1.2%, and 1.5% NaCl and plant survival was monitored daily. No obvious difference was observed between the transformed and WT seedlings when treated with low salt concentration (less than 0.6% NaCl; Fig. 2f). The toxic effects of higher salt concentration (above 0.9%) were delayed and obviously attenuated in ATHK1 transgenic lines. After 10 days at 0.9% and 1.2% NaCl, the ATHK1 transgenic lines continued to grow, albeit at a slower rate, whereas control lines (non-transgenic) exhibited chlorosis and died (Fig. 2g). When transformed and WT lines were subjected to 1.5% NaCl for 10 days, transformed plants began to wither and etiolate, whereas the WT plants died (Fig. 2h).

RWC and Water Loss Ratio of Detached Leaves After Drought and Salt Treatments

No noticeable differences between the WT and transgenic lines were seen in the well-watered normal growth condition. However, after 15 days of withholding water, differences in plant water status between the transgenic and WT lines became evident. RWC in the stressed condition declined from the initial value of 80.7% to 45.9% in the WT plants, but was maintained at about 82.6% in the transgenic lines (Fig. 3a). The RWC in leaf tissues of WT plants declined from the initial 80.5% value to 37.3% after 10 days at 0.9% NaCl. In the transgenic plants, RWC was unaltered (83.4%) after 10 days of salt treatment (Fig. 3b). The rate of water loss of detached WT leaves over 12 h was faster than that for transgenic lines under standard water and soil conditions (Fig. 3c).

Fig. 3
figure 3

RWC of WT and transgenic L. barbarum after 15 days of water deficit (a) or after 10 days salt stress of 0.9% NaCl (b) and the water loss rate of excised leaves of ATHK1 transgenic and WT plants under normal condition (c). Results are the mean ± SD of three individual measurements

Full size image

Ion Leakage Ratio of Cell Membrane after Drought and Salt Treatments

In WT leaves, the ion leakage ratio was 50% higher than control values after 15 days of water deficit and 43% higher after 10 days of 0.9% NaCl treatment. In contrast, transgenic plants showed no apparent changes in ion leakage from the untreated condition (Fig. 4a).

Fig. 4
figure 4

Effects of drought and salinity stresses on L. barbarum. Cell membrane ion leakage ratio of leaves between ATHK1 transgenic and WT plants after 15 days water deficit or 10 days of 0.9% NaCl treatment (a). Proline content of WT and ATHK1 transgenic plants after 15 days drought or 30% PEG6000 for 72 h (b) or after increasing NaCl concentrations for 10 days (c). Chlorophyll contents were determined for leaf disks of WT and transgenic plants under 0.9% NaCl for 5 or 10 days and after water deficit for 15 or 30 days (d). Untreated L. barbarum seedlings served as experimental control. Values are the mean ± SD (bar) for three observations (n = 3)

Full size image

Effect of Water Deficit, Osmotic, and Salt Stresses on Proline Content

Under well-watered conditions, the concentration of free proline in leaves of transgenic plants was similar to that of WT plants. After 15 days of drought stress, proline content in transgenic lines was four times higher than WT plants. After 72 h of 30% PEG6000 treatment, the free proline content of transgenic plants was substantially higher than that of WT plants (Fig. 4b).

When exposed to increasing NaCl concentration, the free proline content of transgenic plants showed marked increases compared to the WT plants, with the greatest increase noted at 0.9% NaCl (Fig. 4c).

Chlorophyll Loss after Water Deficit, Osmotic, and Salt Stresses

Total chlorophyll content in leaf discs was significantly decreased by 15 and 30 days of drought. However, at the same water stress, the chlorophyll loss in transgenic plants was significantly less noticeable than that in the WT plants. Salinity (0.9% NaCl) reduced the chlorophyll content by 29.5% after 5 days in WT plants and by 46.5% after 10 days. The transgenic plants suffered a much lower chlorophyll loss, with the maximum loss after 5 days of salt stress being 10% and 13% after 10 days. (Fig. 4d)

Lipid Peroxidation During Water Deficit, Osmotic, and Salt Stresses

MDA concentration, used as a measure of lipid peroxidation, was similar in WT and transgenic plants grown in the standard growth conditions. Under saline conditions, the WT plants showed maximum rise in MDA concentration (almost 2.1 times above the basal level) after 10 days of salt treatment. In contrast, the transgenic plants showed only a slight increase in MDA levels (0.2 times) over basal levels (Fig. 5a).

Fig. 5
figure 5

MDA concentration and ROS (H2O2 and O2−) activity as an index of lipid peroxidation and oxidative injury in WT and ATHK1 transgenic L. barbarum lines in response to salinity and drought stresses. MDA content in response to 0.9% NaCl salinity (a) and 15 days drought (b). H2O2 content (c) and O2− production rate (d) under 15-day drought stress conditions. Effect of increasing NaCl concentrations on H2O2 content (e) and rate of O2− production (f). The data represent the means of three observations (n = 3). The vertical bar at the top represents standard error in each case

Full size image

After 15 days of water deficit, or after 30% PEG6000 treatment for 72 h, a relatively strong increase in MDA production was observed in WT plants, but only a slight increase was seen in the transgenic lines (Fig. 5b).

H2O2 and O2 − Levels Under Drought and Salt Stress

H2O2 levels in WT plants increased 4.2 times after 15 days of water deficit compared to well-watered counterparts. In contrast, the transgenic plants showed only a 1.18-fold increase in H2O2 levels over untreated plants. In general, the overall increases in H2O2 after water deficit stress were comparatively lower in transgenic lines than in the WT lines (Fig. 5c). The endogenous O2 − production rate was the same in both transgenic and non-transgenic plants under normal conditions. In response to drought stress, O2 − production rate markedly increased (4.9-fold) in WT plants but was only slightly increased in transgenic plants after 15 days or water deficit (Fig. 5d).

Salt stress caused a marked increase in the level of H2O2 and production rate of O2 − of WT plants with increases in NaCl concentration, while these increased much more slowly in transgenic lines (Figs. 5e, f). The increases in H2O2 content and O2 − production rate were most evident at 0.9% NaCl.

Antioxidative Defense Systems in L. barbarum Under Water Deficit and NaCl Stress

There were almost no differences in the activities of these three enzymes among transgenic and WT lines under normal conditions. However, when plants were treated with different concentrations of NaCl, the activities of CAT, SOD, and POD in WT seedlings declined rapidly, except at 0.3% NaCl where a slight increase in SOD or CAT activities was seen. At the 0.9% NaCl concentration, almost no CAT enzyme activity was detected, while SOD and POD activities were considerably lowered in WT plants. In transgenic lines, on the other hand, the activities of these three enzymes were increased in response to various concentrations of salt (Table 1).

Table 1 The effect of increasing NaCl concentrations on the activities of CAT, SOD, and POD in WT and transgenic L. barbarum L plants
Full size table

SOD, POD, and CAT activity in WT plants declined after 15 days of drought stress, and all enzyme activity had almost disappeared after 30 days of drought. However, at the same water conditions, SOD, POD, and CAT activities in the ATHK1 transgenic lines had increased and were significantly higher throughout the water deficit period (Table 2).

Table 2 Changes in protective enzymes activities of transgenic and WT L. barbarum L plants under drought stress
Full size table

Effect of Drought and Salinity Stress on Soluble Protein Levels

Protein content decreased markedly in response to both water and salt stress in leaves of WT lines compared to unstressed controls. In transformed lines, soluble protein decreases were less than those in WT under the same stress conditions (Fig. 6a).

Fig. 6
figure 6

Effects of drought and salinity stress on soluble protein content in leaf discs of ATHK1 transgenic and WT L. barbarum plants (a). Changes in ABA content in transgenic L. barbarum under different days of drought treatment (b) and increasing NaCl concentrations (c). Data represent the means ± SD of three replicates

Full size image

Increased ATHK1 gene expression in transgenic L. barbarum plants under drought, salt, and osmotic stresses

To examine the levels of ATHK1 gene expression in transgenic L. barbarum plants under various stresses and hormone treatments, ten independent transgenic lines were subjected to semiquantitative RT-PCR. As shown in Fig. 7, transcript level of the ATHK1 gene was higher under 0.9% NaCl after 15 days drought or following PEG6000 compared to levels from transgenic plants growing in normal watering and soil conditions. Expression of the ATHK1 gene induced by ABA was not increased obviously in transgenic lines.

Fig. 7
figure 7

Expression level of ATHK1 in transgenic plants under various stress treatments. Two micrograms of RNA from 8-week-old plants that had been treated with 0.9% NaCl, 100 μM ABA, 15 days drought or PEG6000 were subjected to semiquantitative RT-PCR. ACTIN: RNA was used as an internal template control

Full size image

Endogenous ABA Content of Transgenic L. barbarum Plants Under Drought and Salt Stress

ABA content increased sharply during a 30-day drought treatment and then began to decrease (Fig. 6b). Under salt stress, endogenous ABA content initially increased reached a maximum at 0.9% NaCl. ABA levels declined at higher NaCl concentrations (1.2% and 1.5%; Fig. 6c).

Discussion

Plants have developed multiple physiological and biochemical systems which enable them to tolerate abiotic environmental stresses (Ma et al. 2006; Amtmann et al. 2005). Osmotic stress is one of the major factors reducing plant growth and productivity (Boyer 1982), and it occurs under drought as well as high salinity and low temperature conditions. The change in osmotic potential in cells caused by water loss and high salinity triggers various molecular responses in plants (Ma and Bohnert 2007; Xiong and Zhu 2002). Overexpression of ATHK1 using its native promoter increased ATHK1 expression under dehydration or salinity stress and consequently improved the drought and salt tolerance of transgenic A. thaliana plants (Tran et al. 2007b). In this paper, a full-length CDS of ATHK1 in Arabidopsis, with a major open reading frame of 3,621 bp and that encoded a polypeptide of 1,207 amino acids with a predicted molecular weight of 135 kDa, was isolated by RT-PCR. This cDNA was successfully transformed into L. barbarum by A. tumefaciens, confirmed by Southern hybridization. RT-PCR showed that ATHK1 could be stably expressed in transgenic L. barbarum. No constitutive transcript accumulation in hybridization signals in Southern blotting and RT-PCR were detected in the WT, indicating that the gene is not endogenously present in the L. barbarum genome. The selected independent transgenic lines showed normal WT growth and morphology under normal conditions over a 6-month observation period.

All of our data indicated that the transgenic L. barbarum plants were more tolerant than WT to water deficit and salt stress. Transgenic plants showed normal growth even under extreme drought and high salt conditions, while WT plants became wrinkled and lost color, often dying following extreme stress. In addition, when transferred to standard water and soil conditions after a stress treatment, the transgenic lines showed faster recovery than WT plants. WT L. barbarum seedlings could not survive when the maximum neutral salt concentration exceeded 0.9% NaCl. However, neutral salt concentration lower than 1.5% NaCl could not influence the transgenic plant seedlings.

High water level maintenance in plants is an indication of high drought and salt tolerance and can be expressed as relative water content (Flower and Ludlow 1986). RWC is frequently used as screening tool for drought and salinity tolerance, for example in cereals and sunflower (Talamè et al. 2007; Rampino et al. 2006; Liu and Baird 2003; Teulat et al. 2003; Malatrasi et al. 2002). Maintenance of high RWC under dehydration and salinity stress can be attributed to ATHK1 in transgenic L. barbarum. The water loss ratio of excised leaves was used as an indicator of osmotic resistance (Hao et al. 2004) and was substantially higher in transgenic plants than in WT plants. Water loss rate and RWC results both suggested the possibility of a respectively lower transpiration rate and an increased water uptake in the transgenic L. barbarum plants.

In general, osmotically resistant plants have better adaptability to drought and high salinity conditions than do less resistant plants due to reduced alteration of cell membrane permeability during osmotic adjustment in water deficit and high salinity environment. The cell membrane permeability is reflected by its conductance value, which is proportional to increased ion leakage ratio, a serious impairment in WT wheat cell membranes (Bajji et al. 2004). In our experiments, cellular membrane stability in the transgenic lines seemed to be unaffected by water deficit and high salinity conditions, whereas in WT plants, it is badly damaged. This result suggested an increased ability of the transgenic plants to counteract the damaging effects of salinity and drought stress on endogenous cell membranes, which allowed the transgenic plants to survive under the stress conditions imposed in this study.

Proline has also been reported to contribute to cellular membrane stability (Ashraf and Harris 2004; Hanson and Burnet 1994). Its accumulation is considered to enable proline to act as an important osmolyte (Behera et al. 2003), allowing osmotic adjustment under water deficit and salt stress conditions (Nanjo et al. 1999; Verslues and Bray 2004). In Arabidopsis, proline-deficient mutants are hypersensitive to osmotic stress (Brini et al. 2007). Our results confirmed again that increased endogenous proline levels in transgenic L. barbarum under salinity and drought treatments were associated with enhanced ability of plants to tolerate osmotic stress.

Chlorophyll changes are fundamental to understanding a plant’s responses to the environment in which it resides (Schlemmer et al. 2005). The chlorophyll content increased in Amaranthus tricolor treated with 300 mM NaCl for 7 days (Wang and Nii 2000). Graminaceous chlorophyllic cells subjected to osmotic stress developed substantially higher amount of chlorophyll (García-Valenzuela et al. 2005). On the other hand, it has been reported that many different proteins are synthesized and accumulated in response to osmotic stress, including dehydrins, heat shock proteins, and detoxification proteins (Campalans et al. 1999; Smirnoff 1998). PEG treatment was found to decrease the soluble protein content in leaf discs of gerbera (Lai et al. 2007). Our results showed that after salt and drought treatments, the degradation of chlorophyll and soluble protein was much lower in leaves of transgenic Lycium plants than in WT plants.

ROS at high concentrations are highly cytotoxic, inducing DNA damage, lipid peroxidation, and protein degradation (Sun 1990), as well as membrane damage resulting in electrolyte leakage (Apse and Blumwald, 2002). Plants generate abundant ROS in form of O2 − and H2O2 under salt and drought stresses (He et al. 2007; Luna et al. 2005). Lipid hydroperoxidation is another effective indicator of cellular oxidative damage and can be measured by changes in MDA content in leaves (Parvanova et al. 2004). In our study, there were no differences in O2 − production rate, MDA, and H2O2 content in transgenic and non-transgenic plants before stresses were applied. However, a relatively strong increase of O2 − production rate, MDA production, and H2O2 content was observed in WT plants, while only a slight increase was seen in transgenic plants expressing ATHK1 under both drought and NaCl stress. The increased O2 −, MDA, and H2O2 level further confirmed that WT L. barbarum produced excessive ROS under water deficit and salt stress. This slow increase in H2O2 and O2− may have been responsible for the much stronger tolerance of transgenic lines treated under different concentrations of NaCl and water deficit.

Plants also possess systems to scavenge ROS to prevent the destructive oxidative reactions (Foyer et al. 1994). These consist of a number of antioxidant defense systems found in aerobic cells to counteract the damaging effects of ROS (Tsai et al. 2005; Lee et al. 2001; Orendi et al. 2001). The induction of ROS-scavenging enzymes such as SOD, POD, and CAT is the most common response to ROS synthesized under environmental stress (Lai et al. 2007). SOD and POD activities have been found to be enhanced in osmotically stressed leaf discs of rape plants compared to unstressed control plants (Aziz and Larher 1998). In our study, under normal conditions, enzyme activities of CAT, SOD, and POD remained unchanged in both transgenic and WT lines. When subjected to drought and salt stresses, a rapid decrease in CAT, SOD, and POD activity was observed in WT L. barbarum, while the activities of SOD, CAT, and POD were greatly increased in transgenic plants by drought and high NaCl stresses. The transgenic lines in our study showed less accumulation of O2 −, MDA, H2O2 and higher activity of SOD, CAT, and POD, which were believed to protect plants from salinity and water deficit injuries. These results suggest that expression of ATHK1 may play a crucial role in preventing the over-accumulation of ROS by ROS-scavenging activities, thus protecting cell membrane integrity and enhancing salinity and drought tolerance in transgenic L. barbarum plants.

ATHK1 mRNA was expressed in transgenic L. barbarum and increased after drought, salt, and osmotic stress treatments, suggesting that it has a function in transgenic lines. ATHK1 positively regulates many ABA-inducible genes and controls stress responses in Arabidopsis through both ABA-dependent and independent signaling pathways (Wohlbach et al. 2008; Tran et al. 2007a). It appears that dehydration triggers the production of ABA, which in turn induces the expression of different genes. Several genes that are induced by water stress are not responsive to exogenous ABA treatment (Shinozaki and Yamaguchi-Shinozaki 2007). Our findings suggest that ABA did not induce increased expression of ATHK1 gene in transgenic lines, although endogenous ABA was accumulated rapidly under drought and salinity stress.

In this study, we observed that ATHK1 under CaMV 35S promoter may function as a plant osmosensor and play an important role in an osmotic stress signaling pathway in transgenic L. barbarum. The ATHK1 protein is likely involved in regulating the transcription of other related stress genes in transgenic L. barbarum, but the altered expression of a large number of stress-responsive genes involved in osmotic stress pathway still needs to be investigated. The regulatory role of ATHK1 in transgenic L. barbarum also needs further study.

Statistical evaluation of our data showed that all the transgenic plants performed better than did the non-transgenic plants under water deficit and high salinity conditions. This indicated that transgenic L. barbarum had a more enhanced salt and drought tolerance than did WT plants. To date, this is the first report on the expression of ATHK1 in L. barbarum by means of an A. tumefaciens-mediated gene transfer system. The activation of several stress-inducible pathways and accumulation of several metabolites or antioxidants in transgenic plants may represent a new way to develop plants that are able to survive stress conditions through osmotic regulation. Our next task will be to assess the growth performance of these transgenic plants in the field when subjected to a combination of drought and salinity stresses.