Exogenous Melatonin Delays Dark-Induced Grape Leaf Senescence by Regulation of Antioxidant System and Senescence Associated Genes (SAGs)

Leaf senescence is a developmentally programmed and degenerative process which comprises the last stage of the life cycle of leaves. In order to understand the melatonin effect on grapevine leaf senescence, the dark treatment on detached leaves of Vitis vinifera L. cv. Red Globe was performed to induce leaf senescence at short period of time. Then, a series of physiological and molecular changes in response to exogenous melatonin were measured. Results showed that 100 μM of melatonin treatment could significantly delay the dark induced leaf senescence, which is accompanied by the decreased production of reactive oxygen species (ROS). Meanwhile, melatonin treatment could increase the scavenging activity of antioxidant enzymes, such as peroxidase (POD), superoxide dismutase (SOD), and catalase (CAT). Simultaneously, ascorbate (AsA) and glutathione (GSH) contents, the activities of ascorbate peroxidase (APX), and glutathione reductase (GR) were significantly higher than control treatment in samples treated with melatonin. Furthermore, melatonin treatment showed to suppress the expression of leaf senescence-associated genes (SAGs). All these results demonstrated that melatonin could activate the antioxidant and Ascorbate-Glutathione (AsA-GSH) cycle system and repress the expression of SAGs that lead to delay the dark induced grape leaf senescence.


Introduction
Leaf development plays vital roles in the growth and development of plant, especially when photosynthesis takes place in this main area [1]. The environmental conditions such as insufficient light, low CO 2 concentration, high temperature, and humidity could affect the life cycle of leaves. For grapevines, these stress factors could not only influence their growth, but also accelerate the leaf senescence and affect the rate of photosynthesis that resulted in a reduction of yield and fruit quality [2,3]. Therefore, delaying leaf senescence of grapevines could be of vital importance to increase the level of fruit quality and production. transferred to the laboratory. Leaves were rinsed with distilled water and soaked in the different concentrations of MT including 50, 100, 200, 500 µM with 0.1% (v/v) Tween-80 for 60 seconds. For each treatment, 50 leaves were selected and dried at room temperature. Finally, each leaf was transferred and incubated in a growth chamber at constant temperature (28 • C) and 80%~90% relative humidity (RH) without light. Leaf samples were collected after 0, 4,8,12,16, and 20 days of each treatment, quickly frozen in liquid nitrogen, and stored at −80 • C, then 0.5 g samples were taken from the three mixed leaves to do following examinations, but not for chlorophyll content and electrolyte leakage measurements, the fresh leaves were used for these two tests. For RNA extraction, leaf samples were collected in control (0 µM) and 100 µM MT treatment at different times including 0, 4 and 8 days after dark-induced senescence.

Measurement of Chlorophyll Content
To measure the chlorophyll content, leaf samples (0.5 g) were extracted with 50 mL acetone and alcohol (2:1 v/v) solution at different times of MT exposure. Then, the absorbance of chlorophyll extracts was determined at 649 nm and 665 nm by a UV-Visible spectrophotometer (UV759CRT, Yoke, Shanghai, China), and chlorophyll content was calculated according to the method described by Lichtenthaler and Wellburn [65].

Determination of MDA Content and Electrolyte Leakage
For the measurement of malondialdehyde (MDA) content, 0.5 g of leaf powder was transferred in a chilled solution that contains 5 mL of Trichloroacetic acid solution (100 g·L −1 ). Each mixture was centrifuged at 10 000 g for 20 min at 4 • C, and the supernatant was kept for measuring the MDA content. The mixture of 2.0 mL supernatant and 2.0 mL of 0.67% Thiobarbituric acid boiled for 20 min. After cooling, the mixture was centrifuged again and the absorbance value of the supernatant was determined at 450 nm, 532 nm, and 600 nm. The MDA content was determined according to the method described by Cao et al. [66] and expressed in mM·g −1 FW Electrolyte leakage was measured by the method of Dionisio-Sese et al. [67]. In order to test electrolyte leakage, fresh leaf samples (0.1 g) were cut into pieces and transferred into the tubes containing 10 mL deionized water. The tubes were placed in a water bath at a constant temperature of 32 • C for 120 min. Then, the initial electrical conductivity (R1) was tested by an electrical conductivity meter (DDS-307, Rex, Shanghai, China). Next, the tubes boiled for 20 min, cells completely were killed, and all electrolytes released. When temperature cooled down to 25 • C, the final electrical conductivity (R2) was determined. The electrolyte leakage was expressed following the formula: electrolyte leakage (%) = R1/R2 × 100.

Extraction and Antioxidant Enzymes Assay
To prepare the crude enzyme extraction, leaf powder (0.5 g) was transferred in a chilled extracting solution with 9 mL of 0.1 M sodium phosphate buffer (pH 7.8) containing 0.1 mM EDTA-Na 2 and 1 % polyvinylpyrrolidone. Each mixture was centrifuged at 12 000 g for 20 min at 4 • C, and the supernatant was kept for measuring the antioxidant enzymes activity. SOD activity was measured by the nitro blue tetrazolium (NBT) illumination method [68]. Accordingly, 3.3 mL of reaction mixtures was formed of 1.5 mL of 50 mM sodium phosphate buffer (pH 7.8), 0.3 mL of 130 mM methionine, 0.3 mL of 750 µM NBT, 0.3 mL of 100 µM EDTA-Na 2 , 0.3 mL of 20 µM riboflavin, 0.1 mL of the enzyme extract, and 0.5 mL of distilled water. Then, the color reaction of mixtures was at a light intensity of 4000 lx for 20 min. After the reaction finished, we used the black cloth to terminate the color reaction. Finally, we monitored the SOD activity at 560 nm according to the inhibition of the photochemical reduction of NBT. One unit of SOD activity was defined as the amount of enzyme needed to contain NBT photochemical reduction of 50%.
The activity of POD was determined at 470 nm by a UV-Visible spectrophotometer [69]. 10.0 mL of reaction mixtures contained 1.0 mL of the enzyme extract, 1.0 mL of 0.1% (m/v) guaiacol, 7.0 mL distilled CAT activity was detected by recording the decrease in absorbance at 240 nm [70], as a result of the decomposition of H 2 O 2. The reaction mixture (3.0 mL) contained 2.9 mL of 20 mM H 2 O 2 and 0.1 mL of the enzyme extracts. This reaction was also initiated by adding H 2 O 2. After 15 seconds of reaction, the absorbance was recorded every 30 seconds. The CAT activity of the enzyme extracts was expressed in U·min −1 ·g −1 FW.
The activity of APX was measured at 290 nm by the decrease in absorbance because the reduced ascorbate was oxidized [71]. 3.0 mL of reaction mixtures consisted of 2.6 mL of reaction buffer with 0.1 mM EDTA-Na 2 and 0.5 mM ascorbate, 0.1 mL of enzyme extract, and 0.3 mL of 2 mM H 2 O 2 . This reaction was initiated by adding H 2 O 2 . After 15 seconds of reaction, the absorbance was monitored every 30 seconds. The APX activity of the enzyme extracts was expressed in U·min −1 ·g −1 FW.
For the GR activity, it was measured according to the method of Carlberg and Mannervik [72]. 3.0 mL of reaction mixtures consists of 2.7 mL of 0.1 M sodium phosphate buffer (pH 7.5) with 1.0 mM EDTA-Na 2 , 0.1 mL 5.0 mM oxidized glutathione, and 0.2 mL of enzyme extract. This reaction was initiated by adding 0.04 mL 4.0 mM nicotinamide adenine dinucleotide phosphate (NADPH). After 15 seconds of reaction, the absorbance was monitored every 30 seconds. The GR activity of the enzyme extracts was expressed in U·min −1 ·g −1 FW.

Extraction and Analysis of Antioxidant Substances
Briefly, 0.5 g of leaf tissues was crashed using liquid nitrogen and transferred into 8 mL cold 5% (m/v) sulfosalicylic acid and blended well. The mixture was centrifuged at 16 000 g for 25 min at 4 • C, and the supernatant was removed for the following tests.
The AsA was measured by the phenanthroline colorimetric method [73]. 5 mL of reaction mixtures included 1.0 mL of supernatant, 1.0 mL of 50 g·L −1 (m/v) TCA, 1 mL of absolute ethanol, 0.5 mL of 0.4% (v/v) phosphoric acid-alcohol solution, 1.0 mL of 5 g·L −1 (m/v) phenanthroline-alcohol solution, and 0.5 mL of 0.3 g·L −1 FeCl 3 -alcohol solution. The mixture was incubated at 30 • C for 60 min. The absorbance of the reaction system was determined at 534 nm. The content of the AsA was expressed in mg·g −1 FW. GSH was determined according to the methods of Griffith [74]. The 2.5 mL of the reaction system included 1.0 mL of the supernatant, 1.0 mL of 0.1 mM sodium phosphate buffer (pH 7.7), and 0.5 mL 4 mM 5,5'-dithio-bis-2-nitrobenzoic acid (DTNB) (dissolved in potassium phosphate buffer pH 6.8. The reaction was incubated at 25 • C for 10 min. Finally, the absorbance was recorded at 412 nm. The content of GSH was expressed in nM·g −1 FW.

Quantifications of O 2 − and H 2 O 2
The O 2 − production rate was measured according to method described by Zhang et al. [75].
Briefly, 0.5 g leaf tissue was ground using liquid nitrogen, homogenized in 5.0 mL 0.05 mM sodium phosphate buffer (pH 7.8) containing 1 mM EDTA, 0.3% Triton X-100, and 2% PVP and centrifuged at 12 000 g for 20 min at 4 • C. Successively, 1.0 mL supernatant, 1.0 mL PBS (pH 7.8), and 1.0 mL 1 mM hydroxylamine hydrochloride were mixed and incubated at 25 • C for 60 min. Then, 1.0 mL 17 mM p-aminobenzene sulfonic acid and 1.0 mL 7.0 mM naphthylamine were added to the PBS and hydroxylamine hydrochloride mixture and incubated at 25 • C for 20 min. The absorbance was measured at 530 nm using a spectrophotometer (UV759CRT, Yoke, Shanghai, China). The production rate was expressed in nM·min −1 ·g −1 FW. The content of H 2 O 2 was measured according to the method described by Yahmed et al. [76]. Briefly, 0.5 g leaf tissue was crashed using liquid nitrogen and homogenized in 5.0 mL cold 0.1% (m/v) TCA. The homogenate was centrifuged at 12 000 g for 20 min and 0.5 mL of the supernatant was added to 0.5 mL of 0.1 mM sodium phosphate buffer (pH 7.0) and 1.0 mL 1.0 M of potassium iodide. The mixture was incubated in dark at 28 • C for 15 min. The absorbance was measured at 390 nm. The content of H 2 O 2 was based on a standard curve generated with known H 2 O 2 concentrations.

RNA Isolation and Quantitative Real-time PCR
Total RNA of leaves was extracted with RNA extraction kit (Real-Times Biotechnology, Beijing, China) according to the manufacturer's instructions. The quality of RNA was examined by 1% (m/v) agarose gel and further assessed by Nanodrop TM 2000 Spectrophotometer (Thermo Fisher, New York, NY, USA).
The qRT-PCR was performed by Real-Time fluorescence quantitative PCR instrument (LightCyclery 96-Real-Time PCR system, Roche, Switzerland) with SYBR Green PCR Master Mix (Takara, Kusatsu, Japan). The thermal profile was used: 95 • C for 15 min, followed by 95 • C for 10 s, 60 • C for 30 seconds, and 72 • C for 30 seconds for 40 cycles. The expression levels of three senescence-associated genes (SAG12CysProt and SAG13) were analyzed by the comparative 2 −∆∆CT method. The qRT-PCR experiments were performed for three biological replications. The primer sequences of SAGs [77] with modifications for qRT-PCR were shown in Table 1.

Statistical Analysis
All data were analyzed by one-way ANOVA method, followed by Duncan's multiple range tests. The data in all figures were presented as 'means ± standard deviation (S.D.)' with three replications.

MT Could Monitor Dark Yellowing in a Dose-Dependent Behavior
To validate whether exogenous MT application has effect on the dark-induced leaf senescence in grapevine, the different concentrations of MT (0, 50, 100, 200, and 500 µM) were used, and the leaf appearance was investigated in different time courses (0, 4, 8, 12, 16, and 20 d). Since leaf yellowing is one of the most direct senescence symptoms, we observed a difference between samples treated with and without MT. As shown in Figure  All the leaves were incubated in a dark growth chamber where temperature and RH were held at 28 °C and 80%-90% without light, respectively. For each treatment, 50 leaves were considered.
Since leaf senescence has a relationship with chlorophyll content, electrolyte leakage, and MDA content, we evaluated these parameters in grapevine leaves treated with MT and compared them to controls. As shown in Figure 2A, although dark condition in grapevine leaves decreased the chlorophyll content in a time-dependent manner, there is no significant difference among the samples treated with and without MT at time points 0, 4, and 8 days after exposure. The changes of chlorophyll content start to appear between different concentrations of MT from the 12 th to the 20 th day after exposure. For example, on the 12 th day, the chlorophyll content with 100 μM pre-treatment was 1.51 mg·g -1 FW, whereas it was 1.04 mg·g -1 FW in the control treatment, which was statistically significant. Results also indicated that 100 μM MT treatment could preserve the chlorophyll content at a higher level than the other treatments. Since leaf senescence has a relationship with chlorophyll content, electrolyte leakage, and MDA content, we evaluated these parameters in grapevine leaves treated with MT and compared them to controls. As shown in Figure 2A, although dark condition in grapevine leaves decreased the chlorophyll content in a time-dependent manner, there is no significant difference among the samples treated with and without MT at time points 0, 4, and 8 days after exposure. The changes of chlorophyll content start to appear between different concentrations of MT from the 12th to the 20th day after exposure. For example, on the 12th day, the chlorophyll content with 100 µM pre-treatment was 1.51 mg·g −1 FW, whereas it was 1.04 mg·g −1 FW in the control treatment, which was statistically significant. Results also indicated that 100 µM MT treatment could preserve the chlorophyll content at a higher level than the other treatments. The electrolyte leakage measurements indicated that MT had different effects on the treated and non-treated samples ( Figure 2B). The ion leakage decreased in initial stages at the onset of senescence in grapevine leaves (four days) in response to different concentrations of MT and decrement continued during time series after MT exposure. Among different concentrations of MT, 100 μM concentration could significantly decrease the electrolyte leakage in comparison to other treatments, especially to normal conditions. Analysis of MDA also exhibited that MT in different concentrations began to decrease MDA content during time series after exposure compared to control. However, MDA content increased with prolonged MT treatment, but the increment was lower in different concentrations of MT than in control treatment from days 4 to 20 ( Figure 2C). Among different concentrations of MT, the concentration of 100 μM showed a significant decrease in all-time series after exposure.
According to obtained results of the chlorophyll content, electrolyte leakage, and MDA content, we observed that 100 μM MT treatment had the best effect on inhibiting the dark induced leaf senescence. Therefore, we preferentially chose this concentration to perform the next analysis.

MT Could Decline the Accumulation of Oxidizing Agents
Since senescence can involve oxidative damage, we examined the effect of MT on the content of reactive oxygen species (ROS) like superoxide (O2 − ) and hydrogen peroxide (H2O2) as strong oxidizing agents in grapevine leaves during dark-induced senescence. As shown in Figure 3A and B, the accumulation of H2O2 and O2 − increased gradually during time series in control samples. Although the changes pattern in the content of oxidizing agents was similar, the concentration of H2O2 and O2 − were significantly lower after treatment with 100 μM MT during times 8-20 day when compared to the controls, with exception O2 − concentration at the 16 th day ( Figure 3B). Since O2 − can Analysis of MDA also exhibited that MT in different concentrations began to decrease MDA content during time series after exposure compared to control. However, MDA content increased with prolonged MT treatment, but the increment was lower in different concentrations of MT than in control treatment from days 4 to 20 ( Figure 2C). Among different concentrations of MT, the concentration of 100 µM showed a significant decrease in all-time series after exposure.
According to obtained results of the chlorophyll content, electrolyte leakage, and MDA content, we observed that 100 µM MT treatment had the best effect on inhibiting the dark induced leaf senescence. Therefore, we preferentially chose this concentration to perform the next analysis.

MT Could Decline the Accumulation of Oxidizing Agents
Since senescence can involve oxidative damage, we examined the effect of MT on the content of reactive oxygen species (ROS) like superoxide (O 2 − ) and hydrogen peroxide (H 2 O 2 ) as strong oxidizing agents in grapevine leaves during dark-induced senescence. As shown in Figure 3A be as substrate in the H2O2 synthesis reaction, this result suggests that 100 μM MT treatment could significantly suppress H2O2 and O2 − production in detached grapevine leaves.

Antioxidant Enzyme Responses to MT Treatment
Oxidant agents require the antioxidant system, especially antioxidant enzymes including POD, CAT, and SOD, which can effectively scavenge the reactive oxygen species in a defense system. In order to test whether MT treatment could promote the activity of these enzymes, the activity of POD, CAT, and SOD was evaluated. As shown in Figure 4, the activity of these three antioxidant enzymes changed and showed the similar trends during leaf senescence. With prolonged MT exposure, the activities of POD, CAT, and SOD in the detached leaves indicated the similar patterns of changes that enhanced at first, and then reduced in both treatments. Moreover, the rate of decrement of POD and CAT activity ( Figure 4A and B) is greater than that of SOD ( Figure 4C). The activity of POD and CAT reached a peak at days 4 and 12, respectively, while SOD showed a high level of activity during days 8 to 12 after MT exposure. Moreover, the MT pretreatment could significantly increase the activities of POD, CAT, and SOD when compared to control treatment ( Figure 4). It demonstrated that MT treatment could increase the activity of antioxidant enzymes to scavenge ROS production and protect leaves from senescence.

Antioxidant Enzyme Responses to MT Treatment
Oxidant agents require the antioxidant system, especially antioxidant enzymes including POD, CAT, and SOD, which can effectively scavenge the reactive oxygen species in a defense system. In order to test whether MT treatment could promote the activity of these enzymes, the activity of POD, CAT, and SOD was evaluated. As shown in Figure 4, the activity of these three antioxidant enzymes changed and showed the similar trends during leaf senescence. With prolonged MT exposure, the activities of POD, CAT, and SOD in the detached leaves indicated the similar patterns of changes that enhanced at first, and then reduced in both treatments. Moreover, the rate of decrement of POD and CAT activity ( Figure 4A,B) is greater than that of SOD ( Figure 4C). The activity of POD and CAT reached a peak at days 4 and 12, respectively, while SOD showed a high level of activity during days 8 to 12 after MT exposure. Moreover, the MT pretreatment could significantly increase the activities of POD, CAT, and SOD when compared to control treatment ( Figure 4). It demonstrated that MT treatment could increase the activity of antioxidant enzymes to scavenge ROS production and protect leaves from senescence.

Effect of MT on the Ascorbate-Glutathione Cycle
To continue the survival of leaves in senescence condition, a balance between oxidantantioxidant systems is required. AsA-GSH is an important antioxidant system in plants, which can synergize with other ROS scavenging systems to remove excessive accumulation of oxidizing agents.

Effect of MT on the Ascorbate-Glutathione Cycle
To continue the survival of leaves in senescence condition, a balance between oxidant-antioxidant systems is required. AsA-GSH is an important antioxidant system in plants, which can synergize with other ROS scavenging systems to remove excessive accumulation of oxidizing agents. This cycle involves the key antioxidants compounds including AsA and GSH, and the main enzymes like APX, GR [78,79]. For example, AsA could function directly to detoxify H 2 O 2 . Thus, to investigate the relationship between MT treatment and AsA-GSH cycle, we evaluated the changes of AsA and GSH contents as antioxidant metabolites, and the activities of APX and GR linking these metabolites. As shown in Figure 5, MT treatment significantly increased the content of AsA (during times 4 to 20 day) and GSH (during times 8 to 20 day) in comparison to control treatment ( Figure 5B,D). Moreover, MT treatment could significantly hike the activities of APX and GR compared to control treatment. The behavior of the two enzymes was similar during the time series after MT exposure because their activity was raised from days 0 to 12 and then fell from days 12 to 20 ( Figure 5A,C). These results elucidated that AsA-GSH cycle plays an important role with MT treatment to delay or suppress leaf senescence.

MT Could Inhibit the Expression Levels of SAGs
Leaf senescence involves regulatory pathways related to gene expression in the senescence program, especially senescence-associated genes (SAGs). In order to verify that MT treatment could delay or inhibit leaf senescence at the molecular level, we quantified transcript levels of some senescence-associated genes or senescence-up-regulated genes (SAGs) in grapevine leaves. The relative expression levels of two genes SAG12 and SAG13 were dramatically upregulated at days 4 and 8 under treatment and reached their maximum levels after 8 days ( Figure 6). Meanwhile, pretreatment with 100 μM MT indicated that the levels of gene expression follow a similar trend in response to MT and decrease during days 4 and 8 after exposure (Figure 6). After 8 days, the expression levels of SAG12 and SAG13 were 67.75-folds and 36.73-folds higher in the control than MT treatment, respectively ( Figure 6A and B). These results showed that MT treatment could inhibit the expression of SAGs such as SAG12 and SAG13.

MT Could Inhibit the Expression Levels of SAGs
Leaf senescence involves regulatory pathways related to gene expression in the senescence program, especially senescence-associated genes (SAGs). In order to verify that MT treatment could delay or inhibit leaf senescence at the molecular level, we quantified transcript levels of some senescence-associated genes or senescence-up-regulated genes (SAGs) in grapevine leaves. The relative expression levels of two genes SAG12 and SAG13 were dramatically upregulated at days 4 and 8 under treatment and reached their maximum levels after 8 days ( Figure 6). Meanwhile, pretreatment with 100 µM MT indicated that the levels of gene expression follow a similar trend in response to MT and decrease during days 4 and 8 after exposure (Figure 6). After 8 days, the expression levels of SAG12 and SAG13 were 67.75-folds and 36.73-folds higher in the control than MT treatment, respectively ( Figure 6A,B). These results showed that MT treatment could inhibit the expression of SAGs such as SAG12 and SAG13. senescence-associated genes or senescence-up-regulated genes (SAGs) in grapevine leaves. The relative expression levels of two genes SAG12 and SAG13 were dramatically upregulated at days 4 and 8 under treatment and reached their maximum levels after 8 days ( Figure 6). Meanwhile, pretreatment with 100 μM MT indicated that the levels of gene expression follow a similar trend in response to MT and decrease during days 4 and 8 after exposure (Figure 6). After 8 days, the expression levels of SAG12 and SAG13 were 67.75-folds and 36.73-folds higher in the control than MT treatment, respectively ( Figure 6A and B). These results showed that MT treatment could inhibit the expression of SAGs such as SAG12 and SAG13.

Discussion
Leaf senescence plays an important role in plant life cycle and is mainly affected by low temperature, uncomfortable light, drought, pathogens attack, and hormones. We induced leaf senescence by dark treatment and examined the function of MT on grapevine leaf senescence. Many studies reported that darkness induce senescence when individual leaves are detached, but not when whole plants are darkened [80][81][82]. Therefore, in the current study, the treatment of darkness on detached grapevine leaves was performed. This system could contribute to study mechanisms of leaf senescence on other species. In the senescence program, chloroplast disintegration and chlorophyll degradation happen at the cellular level and leaf color changes from green to yellow, which is the most remarkable phenotype of leaf senescence [83]. Therefore, the change of chlorophyll content is one of the most important and typical indicators to evaluate leaf senescence. MT is involved during plant growth and development. It also could inhibit leaf senescence that has been widely studied in perennial ryegrass [56], kiwifruit [57], apples [58,62], and adzuki bean [84]. In this work, exogenous MT application on detached grapevine leaves could inhibit yellowing in the dark (Figure 1) through slowing down the chlorophyll degradation rate (Figure 2A) that led to delay dark-induced leaf senescence. These results strongly suggested that exogenous application of MT had a positive effect on delaying grapevine leaf senescence.
During leaf senescence, reactive oxygen species (ROS) are produced and the activities of antioxidant enzymes including CAT, SOD and POD decrease, which resulted in the imbalance of ROS metabolism [8,85]. In addition, excessive ROS could oxidize cell membrane lipid, which directly destroys the biological membrane system and leads to electrolyte leakage and MDA production. In this study, we found that electrolyte leakage ( Figure 2B Figure 3B), and also inhibits electrolyte leakage ( Figure 2B).
This indicated that MT could modulate the production of ROS to slow down the dark-induced leaf senescence. Moreover, MT was an antioxidant substance that could directly reduce ROS level in organisms. Consequently, it would alleviate damage to the membrane system [86][87][88]. In addition, some researchers suggested that MT treatments could also improve the activities of SOD, CAT and POD in the process of leaf senescence [56][57][58]62]. This is consistent with our study that the three enzymes activities increased firstly and then decreased with the aging process within MT treatment ( Figure 4). Overall, MT could scavenge of ROS by activating the antioxidant enzymes to delay grapevine leaf senescence.
In the AsA-GSH cycle system, APX and GR are two key enzymes, which eliminate ROS accompanied with SOD, POD, and CAT enzymes to maintain the balance between oxidant (ROS) and antioxidant systems and keep the stability of cell membrane [89,90]. APX is a key enzyme in chloroplast to detoxify H 2 O 2 through AsA. GR is one of the pivotal enzymes to maintain the effective function of AsA-GSH cycle through reducing oxidized glutathione (GSSG) to GSH by an NADPH-dependent pathway [91,92]. Here, the AsA and GSH concentrations increased ( Figure 5B and D) with the reduced production of O 2 − and H 2 O 2 (Figure 3), meanwhile, the APX and GR activities were enhanced ( Figure 5A,C) during grapevine leaf senescence. Furthermore, AsA and GSH contents, and the activities of APX and GR ( Figure 5) were significantly higher in the treated samples with MT than in the controls, demonstrating that MT could modulate the AsA-GSH cycle system to remove ROS. This was consistent with the studies on apples and ryegrass in which exogenous MT treatment could regulate the AsA-GSH cycle to delay leaves senescence with the enhanced APX and GR activities as scavengers of ROS, as well as the higher concentration of AsA [56,58]. Moreover, during the senescence program, most of the genes involved in leaf senescence were upregulated, such as SAG12 and SAG13 [11,93]. SAG12 in Arabidopsis, a gene encoding a cysteine protease, is highly specific in aging and is often used as a marker gene for aging [94]. A previous study showed that SAG12 was definitely activated by developmental senescence, but not triggered by the regulation of hormone or in response to stresses [95]. For example, during natural senescence in grapevine, the SAG12 also was activated [77]. In addition, PeSAG12-1 was highly induced with the increase of ROS production during dark-induced senescence of Pelargonium cuttings [96]. Furthermore, the expression of many other SAGs was also induced by ROS [97,98], indicating that there is a close relationship between ROS and SAG genes during leaf senescence. Present results indicate that pretreatment of MT dramatically suppressed SAGs expression at 8 days, while the controls had higher expression levels ( Figure 6). Therefore, MT could repress the expression of SAG genes to delay grapevine leaf senescence.
According to the above discussion, we came to a model where the dark induced grapevine leaf senescence triggered the high production of ROS, which could further activate the expression of SAGs (Figure 7). Pre-treatment of MT could activate the antioxidant enzymes and AsA-GSH cycle system to reduce the production of ROS, which finally prevent the expression of SAGs to delay dark induced grapevine leaf senescence (Figure 7). protease, is highly specific in aging and is often used as a marker gene for aging [94]. A previous study showed that SAG12 was definitely activated by developmental senescence, but not triggered by the regulation of hormone or in response to stresses [95]. For example, during natural senescence in grapevine, the SAG12 also was activated [77]. In addition, PeSAG12-1 was highly induced with the increase of ROS production during dark-induced senescence of Pelargonium cuttings [96]. Furthermore, the expression of many other SAGs was also induced by ROS [97,98], indicating that there is a close relationship between ROS and SAG genes during leaf senescence. Present results indicate that pretreatment of MT dramatically suppressed SAGs expression at 8 days, while the controls had higher expression levels ( Figure 6). Therefore, MT could repress the expression of SAG genes to delay grapevine leaf senescence. According to the above discussion, we came to a model where the dark induced grapevine leaf senesce

Conclusions
Senescence or biological aging is a time-dependent process. During darkness, pre-treatment of MT could significantly decrease the production of ROS and increase the activity of antioxidant enzymes including SOD, CAT, and POD. In addition, MT activates the AsA-GSH cycle system during senescence program, raises the content of AsA and GSH, and induces activities of APX and GR.

Conclusions
Senescence or biological aging is a time-dependent process. During darkness, pre-treatment of MT could significantly decrease the production of ROS and increase the activity of antioxidant enzymes including SOD, CAT, and POD. In addition, MT activates the AsA-GSH cycle system during senescence program, raises the content of AsA and GSH, and induces activities of APX and GR. Meanwhile, MT treatment could suppress the expression of leaf senescence related genes (SAGs). These results suggest that MT can be involved in cellular homeostasis to preserve a balance between oxidant-antioxidant systems that led to a delay in the dark-induced leaf senescence.

Conflicts of Interest:
The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.