Alternative Pathway is Involved in Nitric Oxide-Enhanced Tolerance to Cadmium Stress in Barley Roots

Alternative pathway (AP) has been widely accepted to be involved in enhancing tolerance to various environmental stresses. In this study, the role of AP in response to cadmium (Cd) stress in two barley varieties, highland barley (Kunlun14) and barley (Ganpi6), was investigated. Results showed that the malondialdehyde (MDA) content and electrolyte leakage (EL) level under Cd stress increased in two barley varieties. The expressions of alternative oxidase (AOX) genes (mainly AOX1a), AP capacity (Valt), and AOX protein amount were clearly induced more in Kunlun14 under Cd stress, and these parameters were further enhanced by applying sodium nitroprussid (SNP, a NO donor). Moreover, H2O2 and O2− contents were raised in the Cd-treated roots of two barley varieties, but they were markedly relieved by exogenous SNP. However, this mitigating effect was aggravated by salicylhydroxamic acid (SHAM, an AOX inhibitor), suggesting that AP contributes to NO-enhanced Cd stress tolerance. Further study demonstrated that the effect of SHAM application on reactive oxygen species (ROS)-related scavenging enzymes and antioxidants was minimal. These observations showed that AP exerts an indispensable function in NO-enhanced Cd stress tolerance in two barley varieties. AP was mainly responsible for regulating the ROS accumulation to maintain the homeostasis of redox state.


Introduction
Cadmium (Cd), one of the most toxic heavy metals to the environment, has drawn great attention worldwide. It not only inhibits plant growth, but also pollutes the food chain, seriously threatening human health [1,2]. When Cd is accumulated excessively, at the physiological level, a suite of symptoms, such as chlorosis, underdevelopment, and programmed cell death, are induced in plants [3][4][5]. At the cellular level, over-accumulated Cd affects enzyme activity and changes protein structure [4,6,7]. More importantly, cellular redox homoeostasis is disturbed, and reactive oxygen species (ROS) burst, which further leads to oxidative stresses. To better deal with Cd toxicity, plants have evolved various defense strategies. Activating antioxidant enzymes and non-enzymatic antioxidants to counteract the oxidative stress has been widely accepted [8,9]. Alternative pathway (AP) also contributes to enhancing Cd tolerance [10]. Even though great effort has been made in the past, the protective mechanisms against Cd stress still need to be explored.

Plant Materials and Growth Conditions
Barley varieties Ganpi6 and Kunlun14 were used in the present study, and Ganpi6 was regarded as the control. Seeds were surface-disinfected with 2% hypochlorite for 8 min, and washed thoroughly with sterile water. Then, seeds were transferred onto a nylon net on top of 200 mL plastic beakers (20 seedlings per beaker), which were filled with 1/4-strengh modified culture solution [40]. Materials were allowed to grow in a chamber with 16 h light/8 h dark cycles. Culture solution was changed every 2 days. Different chemicals were added to the medium for various treatments after 6 days. Roots were utilized immediately for the following assays.

Root Elongation Measurement
After 6 day growth, different concentrations of Cd were added to 200 mL 1/4-strengh modified Johnsons nutrient solution for 48 h [40]. The roots were collected immediately for measurement of root length with Image J.

Root Electrolyte Leakage Determination
Electrolyte leakage (EL) was monitored according to the method described by Tang et al. (2014) [41]. First, 0.2 g roots were washed in the deionized water for more than three times. The roots were dipped in the deionized water for 2 h at 25 • C. The conductivity of the water bath (C 1 ) and the de-ionized water (C 0 ) was measured. Root samples were then boiled at 100 • C for 40 min, and the conductivity of the water bath (C 2 ) was measured when it cooled to room temperature.

Measurements of Respiration Rate
Respiration was measured on the basis of the method described by Wang et al. [43]. First, 0.03 g roots were cut into 2 mm small segments and put into 2 mL 50 mM phosphate buffer (pH 6.8). After reaction for 2 min at room temperature, the oxygen value slope was defined as the total respiration rate (V t ). Then, 2 mM KCN or 2 mM salicylhydroxamic acid (SHAM) was added and reacted for 2 min; the oxygen value slope was defined as the AP capacity (V alt ) or the CP capacity (V cyt ), independently.

Determination of NO Content
NO level was analyzed on the basis of the method described by Wang et al. [19]. Roots were bathed in the buffer containing [20 mM 4-amino-5-methylamino-2 7 -difluorofluorescein diacetate (DAF-FM-DA, NO fluorescent probe), 0.25 mM NaCl, 1 mM CaCl 2 , 10 mM Hepes-NaOH (pH 7.0)] for 30 min. Then, roots were washed more than three times with 10 mM Hepes-NaOH (pH 7.0)] buffer, and scanned in the Leica SM IRBE stereomicroscope. The images were analyzed by the Leica ZEN software.

Determination of Nitric Oxide Synthase (NOS) Activity
NOS activity was measured according to the method described by Lin et al. [45]. First, 0.25 g roots were homogenized with 2 mL of extraction buffer and centrifuged at 15,000× g for 15 min. Then, 100 µL of supernatant was added to the reaction solution, and 0.2 mM NADPH (reduced nicotinamide adenine dinucleotide phosphate) was added to initiate the reaction. The absorbance at 340 nm was read.

Cloning, Sequencing, and Bioinformatics Analyses
To obtain the intact open reading frame of HvAOX1, primers (Table 1) were designed on the basis of AOX sequence in highland barley roots. Genbank accession numbers were AK363239.1 (HvAOX1a), AK365405.1 (HvAOX1d1), and AK251266.1 (HvAOX1d2). A total of 15 µL reaction mixture was used for PCR. The PCR conditions were 2 min at 98 • C, then 40 cycles of 20 sec at 98 • C, 30 sec at 56.0-58.5 • C, 10 min at 72 • C, and 10 min at 25 • C. The clear single target band was cloned to the pBlunt vector (TransGen Biotech, Beijing, China). Table 1. Primer sequences.

RNA Isolation and qRT-PCR
About 100 mg samples were used for total RNA isolation from various plant tissues by using Trizol reagent. [cDNA synthesis kit (TRANS, Beijing, China)] was used for cDNA synthesis. The qRT-PCR mixture contained 5 µL SYBR Green I Master Mix, 0.5 µL forward primers, 0.5 µL reverse primers, and 4 µL deionized water. Each gene was amplified in three biological replicates. The results were analyzed by Rotor-Gene Real-Time Analysis Software 6.1. The specific primers are listed in Table 1.

Western Blot Analysis
Western blot analysis was carried out following the method described by Zhao et al. [46]. Proteins were separated on a 10% acrylamide gel. After electrophoresis, the proteins were transferred to a polyvinylidene difluoride membrane. The membrane was blocked for 3 h with 10% bovine serum albumin in buffer solution (150 mM NaCl, 0.05% Tween-20, 10 mM Tris-HCl (pH8.0)). Primary antibody against Arabidopsis AOX was added and incubated for 10 h. After rinsing three times, secondary antibody was added and incubated for visualization.

H 2 O 2 and O 2
− Staining H 2 O 2 and O 2 − staining was carried out according to the method described by Wang et al. [47].

Extraction and Estimation of Antioxidants
First, 0.25 g fresh roots were ground in 2 mL meta-phosphoric acid (5%) and centrifuged at 11,000× g for 20 min, and then 1.0 U ascorbate oxidase was added. The absorbance was read at 265 nm. Oxidized ascorbate (DHA, dehydroascorbic acid) content was equal to the total ascorbic acid (AsA) content minus reduced AsA. Oxidized glutathione (GSSG) and reduced glutathione (GSH) contents were measured according to the method described by Giraud et al. [48].

Antioxidant Enzyme Activity Assay
First, 0.25 g fresh roots were ground in 3 mL pre-cooled 25 mM Hepes buffers, and then were centrifuged at 4 • C for 20 min at 12,000× g. The supernatants were collected for examining the activity of antioxidant enzymes. SOD (Superoxide dismutase), CAT (Catalase), POD (Peroxidase) and APX (Ascorbate peroxidase) activities were measured on the basis of the method described by Jian et al. [36]. DHAR (Didehydroascorbic acid reductase), MDHAR (Monodehydroascorbic acid reductase), GR (Glutathione reductase), and GPX (Glutathione peroxidase) activities were measured following the method described by Zhang et al. [49].

Statistical Analysis
Each experiment was repeated at least three times with three replicates. The confidence coefficient was set at 0.05. Experiments that required an analysis of variance were analyzed using SPSS 17.0 analysis of variance (ANOVA) and Origin 8.

Effects of Cd Stress on Dry Weight, Root Elongation, MDA Content, and EL
Cd stress led to toxicity symptoms and inhibited the elongation of Ganpi6 and Kunlun14 roots in a dose-dependent manner. After exposure to various concentrations (0-200 µM) of Cd, the root growth of barley (Ganpi6) and highland barley (Kunlun14) was gradually inhibited ( Figure 1A,B). Under 150 µM Cd treatment, dry weight in Ganpi6 and Kunlun14 plants was reduced by 42.29% and 33.49%, respectively ( Figure 1C), whereas root elongation in Ganpi6 and Kunlun14 was decreased by 55.2% and 50.9%, respectively ( Figure 1D). To further explore the cellular membrane damage caused by Cd stress, two important indicators, MDA and EL levels, were examined. As shown in Figure 1E,F, the MDA content in Ganpi6 and Kunlun14 roots in the presence of 150 µM Cd was elevated 5.83-fold and 4.69-fold, respectively, whereas the EL was increased by 80.3% and 70.7%, respectively. Thus, 150 µM Cd was selected for the subsequent assays.

Effects of Cd Stress on the Respiratory Pathways
To explore the response of respiration to Cd stress, V t , V cyt , and V alt in Ganpi6 and Kunlun14 were examined under various Cd concentrations. As shown in Figure 2A, V t was rapidly increased when Cd concentration was lower than 150 µM, then it showed a declined trend. Compared with V cyt , V alt showed a similar pattern observed in V t under Cd stress ( Figure 2B,C). It was noteworthy that V alt , V alt /V cyt were higher in Kunlun14 than those in Ganpi6 (Figure 2), suggesting that V alt in Kunlun14 had a greater contribution to V t than Ganpi6 did under Cd stress. The time course of respiration changes under 150 µM Cd treatment were examined. The results showed that V t , V alt , and V cyt in both Ganpi6 and Kunlun14 reached their maximal values at 12 h, followed by decreases. Nevertheless, all of them were still markedly higher than that in control roots ( Figure 2E-G).

Cd Stress Induced NO Production from NR Pathway
More and more evidence indicates that NO is involved in enhancing plant tolerance to Cd [50,51]. To explore whether Cd stress excites NO release in Ganpi6 and Kunlun14, NO production was imaged by using DAF-FM, a fluorescent probe, together with NO donor (SNP) or scavenger (PTIO) treatment. In Ganpi6 and Kunlun14 roots, weak NO fluorescence signal was observed in control roots. However, Cd stress greatly enhanced the fluorescence signal, which was further intensified by SNP treatment ( Figure 3A,B). Notably, the NO fluorescence signal was stronger in Kunlun14 than that in Ganpi6. Applying PTIO almost eliminated Cd-induced NO fluorescence signal ( Figure 3A,B). To further explore the source of Cd-induced endogenous NO production in Ganpi6 and Kunlun14 roots, NR inhibitor (NaN 3 ) and NOS inhibitor (L-NNA) were used to detect NO original level. The results showed that NaN 3 in Ganpi6 and Kunlun14 roots nearly abolished Cd-induced NO accumulation, whereas applying L-NNA had almost no impact on Cd-induced NO accumulation ( Figure 3A), indicating that the Cd-induced NO production originated from the NR-dependent pathway. To further confirm this conclusion, NR and NOS activities were examined. Under Cd stress, NR activity in Ganpi6 and Kunlun14 roots was increased by 57.6% and 72.1%, respectively ( Figure 3C). Furthermore, the NR activity showed a similar pattern as that of NO production ( Figure 3B,C). However, the NOS activity had little change under Cd stress ( Figure 3D).
Some studies reported that mitochondria might be an important source for NO generation via single electron leak from the electron transfer chain (ETC) to nitrite [52,53]. Moreover, recent studies have confirmed that AOX functions to protect ETC components from over-reduction in plants, thus preventing single electron leak [54]. To explore the link between NO and AP under Cd stress in Ganpi6 and Kunlun14 roots, SHAM (AOX inhibitor) was used to block the function of AP. As shown in Figure 3, under Cd + SHAM stress, NO fluorescence signal was strongly enhanced, suggesting that dysfunctional AP leads to over-reduction of ETC components.

Expression Patterns of AOX Genes in Highland Barley
cDNA of AOX was cloned from highland barley using the candidate gene approach [55]. Comparison with the nucleotide sequences to barley AOXs confirmed that the highland barley AOXs sequences were AOX1 genes. These genes were named HvAOX1a, HvAOX1d1, and HvAOX1d2. The corresponding accession numbers were MK361118, MK361119, and MK361120. The transcript levels of HvAOX genes in different developmental stages of highland barley were analyzed using qRT-PCR. As shown in Figure 4, three HvAOX genes were detected in all tissues but with specific expression patterns. The expression of HvAOX1a was higher than that of HvAOX1d1 and HvAOX1d2 in all tissues except in stamens. It was worth mentioning that the transcript levels of HvAOXs showed similar increasing pattern with the leaf age (from first leaves to function leaves), implying a common developmental-related feature of all HvAOX genes ( Figure 4).

Exogenous NO Enhanced HvAOX Expression in Ganpi6 and Kunlun14 Roots under Cd Stress
AP usually runs at a low level, however, it could be observably induced when plants suffer from various environmental stresses [25]. To examine whether the increase of V alt in Cd-stressed roots is regulated at the transcriptional level, HvAOX expression was investigated. As shown in Figure 5, HvAOX1a, HvAOX1d1, and HvAOX1d2 in Ganpi6 and Kunlun14 were all expressed in roots. Compared with HvAOX1d1 and HvAOX1d2, the expression level of HvAOX1a was markedly increased under Cd stress. To further explore how the expression levels of HvAOXs were regulated in Cd-stressed roots, we manipulated the production of NO and monitored HvAOXs expression. Under Cd + SNP treatment, the expression of HvAOX1a was increased by 3.5 times and 4.4 times in Ganpi6 and Kunlun14 roots, respectively ( Figure 5A), whereas HvAOX1d1 and HvAOX1d2 were just slightly increased ( Figure 5B,C). Under Cd + SHAM treatment, the expression level of HvAOXs was dramatically reduced.

Exogenous NO Enhanced V alt and AOX Protein Level under Cd Stress
To further explore NO effects on respiration under Cd stress, V alt and V cyt were examined. As shown in Figure 6A, under Cd + SNP treatment, V alt in Ganpi6 and Kunlun14 roots was increased by 52.4% and 60.4%, respectively, whereas V cyt had nearly no change. When AP was inhibited by SHAM under Cd stress, V alt was decreased to nearly the control level, and V cyt still had little change ( Figure 6B). It is worth mentioning that when AP was blocked under Cd stress, applying SNP did not reverse the trend. To further decipher whether the increase of AP in Cd-stressed roots occurs at the protein level, the AOX protein level was detected. Western blotting results showed that Cd stress excited the AOX protein expression in two barley varieties, and the expression of AOX protein was higher in Kunlun14 than in Ganpi6. SNP treatment further enhanced the AOX protein levels in both Ganpi6 and Kunlun14 under Cd stress. However, AOX protein amount was markedly reduced when AP was inhibited by SHAM under Cd stress, and even exogenous SNP did not recover the reduction ( Figure 6C).

Exogenous NO Did Not Relieve Cd-Induced Oxidative Stress under SHAM Treatment
To explore whether AP is engaged in Cd-induced oxidative damage in two barley varieties, MDA and EL levels were examined. The MDA content in Ganpi6 and Kunlun14 was increased by 78.7% and 62.2%, respectively, whereas the EL level was increased by 60.7% and 50.2%, respectively. Applying SNP greatly relieved the oxidative damage caused by Cd. The MDA content was decreased by 60.3% and 57.2% in Ganpi6 and Kunlun14, respectively, whereas EL was decreased by 33.4% and 30.7%, respectively ( Figure 7C,D). Under Cd + SHAM treatment, the MDA content was increased by 38.6% and 33.5% in Ganpi6 and Kunlun14, respectively, whereas EL was increased by 29.5% and 26.3%, respectively, and the increases of MDA and EL were not reversed by SNP ( Figure 7C,D). Because Cd-induced oxidative damage might be triggered by ROS accumulation, H 2 O 2 and O 2 − levels were examined by histochemical staining. As shown in Figure 7A

Effects of Exogenous NO on AsA and GSH Levels in the Presence of SHAM under Cd Stress
To elucidate the role of AP in controlling ROS homeostasis, the levels of small antioxidant molecules were examined. In Ganpi6 and Kunlun14 roots, an increment in the AsA level (0.8 times and 1.3 times, respectively), a decrease in the DHA level (15.2% and 15.5%, respectively), and an increase in the AsA/DHA ratio (1.2 times and 1.6 times, respectively) were observed under Cd stress. Applying SNP under Cd stress elevated the AsA level by 23.0% and 28.2%, and AsA/DHA ratio by 11.8% and 39.4% in Ganpi6 and Kunlun14 roots, respectively. However, the presence of SHAM, Cd, or Cd + SNP treatment had almost no impact on AsA or DHA levels in Ganpi6 and Kunlun14 roots ( Figure 8A,B). The GSH level was elevated by 37.9% and 49.6% in Ganpi6 and Kunlun14 roots, respectively, whereas the GSSG level was decreased by 13.3% and 12.9%, respectively, and the GSH/GSSG ratio was increased by 49.6% and 72.0%, respectively. Applying SNP under Cd stress enhanced the GSH level by 10.9% and 12.1%, lowered the GSSG level by 33.3% and 38.7%, and elevated the GSH/GSSG ratio by 47.6% and 54.0% in Ganpi6 and Kunlun14 roots, respectively. Under Cd + SNP treatment, applying SHAM almost had no effects on GSH or GSSG levels in Ganpi6 and Kunlun14 roots ( Figure 8D,E).
To further explore the mechanism of redox balance maintenance, activities of ascorbate-glutathione cycle-related enzymes were examined. Under Cd stress, the activities of GR was increased by 78.6% and 86.8%, MDHAR was increased by 28.9% and 34.3%, GPX was increased by 61.3% and 64.3%, and DHAR was increased by 38.4% and 39.7% in Ganpi6 and Kunlun14 roots, respectively. Under Cd + SNP treatment, the activities of GR and MDHAR were further enhanced, whereas the activities of GPX and DHAR were reduced in Ganpi6 and Kunlun14 roots. Under Cd + SNP + SHAM treatment, the activities of these enzymes were nearly undisturbed ( Figure 9E-H). These results suggest that AP involvement in SNP-improved tolerance is not achieved by influencing the ascorbate-glutathione cycle.

Effects of Exogenous NO on Antioxidant Enzymes in the Presence of SHAM under Cd Stress
The antioxidant enzyme activities were enhanced under various stresses to reduce the level of ROS [36,42]. As shown in Figure 9A-D, under Cd stress, SOD activity was elevated by 28.6% and 35.8%, CAT activity by 50.4% and 56.7%, and POD activity by 1.5 times and 1.7 times in Ganpi6 and Kunlun14, respectively. Under Cd + SNP treatment, activities of all these enzymes (except POD) were further increased in Ganpi6 and Kunlun14 roots. Nevertheless, applying SHAM had nearly no distinct impact on these antioxidant enzymes under Cd + SNP treatment. These results suggest that AP involvement in SNP-improved Cd tolerance in Ganpi6 and Kunlun14 is not achieved by affecting the antioxidant enzyme system ( Figure 9A-D).

Discussion
In this study, we examined the responses of barley (Ganpi6) and highland barley (Kunlun14) to Cd stress, and further explored the possible function of the AOX pathway (AP) in NO-enhanced tolerance to Cd stress. The results suggest that AP is involved in NO-enhanced tolerance to Cd stress in barley roots by controlling ROS accumulation.
Oxidative stress is considered as one of important responses to Cd toxicity [56,57]. In our experiment, we observed that the dry weight and the root growth were markedly decreased. However, EL and MDA were obviously increased in two varieties with increased Cd concentration ( Figure 1). As a result, Cd stress induced serious oxidative damage in two varieties.
How Cd activates AP is largely unknown in plants. Recently, some reports suggest that NO could relieve heavy metal toxicity [14,15,51]. In this study, to explore the role of NO in regulating AP under Cd stress in two barley roots, NO donor (SNP) and scavenger (PTIO) were used. Results indicate that NO plays an important role in response to Cd stress. When AP was inhibited under Cd stress, NO fluorescence signal was enhanced (Figure 3), implying that NO and AP have a close relationship in response to Cd stress. Recently, research has reported that NO ameliorates arsenic toxicity by altering the alternative oxidase (Aox1) gene expression in Hordeum vulgare L [58]. Applying exogenous SNP relieved Cd-induced oxidative stress in Ganpi6 and Kunlun14 roots. However, when AP was inhibited by SHAM, Cd-induced oxidative stress was further aggravated and it was not reversible by application of SNP (Figure 7), suggesting AP plays an indispensable role in response to Cd stress and participates in the SNP-improved Cd stress tolerance. This is perhaps another crucial reason for the enhanced Cd tolerance in two barley roots. When AP was inhibited by SHAM under Cd stress, there was nearly no effect on Cd content in two barley shoots or roots ( Figure S1), suggesting that AP is not involved in enhancing Cd tolerance by decreasing Cd uptake or increasing discharge of Cd. In addition, Cd content in two barley shoots was minor compared with Cd content in roots. Therefore, inhibition of Cd transport from root to stem could be one of the reasons for the tolerance of barley to Cd stress. We further observed that Cd-induced NO mainly resulted from the NR pathway, which is consistent with some earlier reports [59][60][61]. Some studies have demonstrated that NOS-regulated NO production is engaged in salt stress and zinc tolerance [21,62], and another report revealed that NO production results from both NR-and NOS-depended pathways [60]. The different source of NO production might be due to various stresses and plant species.
Antioxidant molecules and antioxidant defense system are core elements in keeping ROS homeostasis in barley plants exposed to various stressors [63][64][65]. Applying SNP could stimulate the antioxidant defense system to enable plant tolerance to stressors [37,66,67]. Substantial evidence showed that AOX functions in maintaining ROS homeostasis [33]. A lack of AOX results in elevated activity of ROS-related scavenging enzymes in some plants [48,68]. In this study, we found that the activities of ROS-related scavenging enzymes and antioxidant molecules were enhanced in two barley roots under normal and stressed conditions in the presence of SNP, which is in line with the previous study in barley seedlings [64]. The results also showed that SHAM had nearly no effects on ROS-related scavenging enzymes and antioxidant molecules (Figures 8 and 9). Furthermore, the results showed that the correlation between AP and ROS-related scavenging enzymes or antioxidant molecules was minor (Table S1), suggesting that AP engagement in SNP-elevated tolerance is not achieved by affecting ROS-related scavenging enzymes and antioxidant molecules. This observation is different from some previous studies [33,34], possibly due to different stress types and intensity. These results indicated that the elevated ROS-related scavenging enzymes and antioxidant molecules might also be in charge of NO-enhanced tolerance to Cd stress in two barley varieties.

Conclusions
On the basis of these results, a signal pathway model was proposed (Figure 10). It illustrates the physiological and biochemical mechanisms of the specific role of AOX pathway involvement in the NO-mediated Cd tolerance in two barley varieties. NO, which is mainly derived from the NR pathway, is stimulated in response to Cd stress. NO acts as a signal molecule to affect the expression of HvAOX1a, HvAOX1d1, and HvAOX1d2; corresponding AOX proteins; and ultimately V alt . NO inhibited the ROS burst by promoting AP and ROS-related scavenging enzymes and antioxidant molecules. There was no correlation between AP and antioxidant enzymes or the ascorbate-glutathione cycle in response to Cd stress. Heavy metal tolerance requires an integrated physiological and biochemical processes. In our study, we prove that the AOX pathway plays an indispensable role in the SNP-elevated resistance to Cd stress in barley roots. However, it is worth noting that the capacity of AP is not the actual activity of AP [69,70]. Measurements of the AP activity in vivo should be done in future because it is crucial for determining the role of AOX in plants.

Conflicts of Interest:
The authors declare no conflict of interest.