Seed Pre-Soaking with Melatonin Improves Wheat Yield by Delaying Leaf Senescence and Promoting Root Development

: The e ﬀ ects of exogenous application of melatonin (MEL) on promoting plant growth and alleviating environmental stresses are already known, but the potential value in crop production is still poorly understood. In this study, the e ﬀ ects of seed pre-soaking with MEL on winter wheat ( Triticum aestivum L.) growth and yield were investigated in a continuous two-year pot experiment and another year of ﬁeld experimentation. Results showed that seed pre-soaking with di ﬀ erent concentrations of MEL (10, 100 and 500 µ M) for 24 h increased grain yields per plant from 29% to 80% in pot experiment and increased grain yield per area from 4–19% in ﬁeld experiment, compared with the controls. Further analysis showed that the beneﬁcial e ﬀ ects of MEL on improving wheat grain yield can be ascribed to: (1) increased spike number by enhancing tiller number; (2) enhanced carbon assimilation capacity by maintaining large leaf area, high photosynthetic rate and delaying leaf senescence; (3) promoted growth in root system. The result of this study suggests that MEL could be considered as an e ﬀ ective plant growth regulator for improving grain production in winter wheat.


Introduction
Food security is one of the major policy concerns in the world because of its large population that needs to be fed, while arable land resources are limited and decreasing. That is also the case in many developing countries where food shortage is still a critical problem [1]. Therefore, any strategies for enhancing the production of these basic foodstuffs would have massive ramifications. Plant growth regulators are widely used in modern agricultural production, and their application plays an important role in increasing and securing crop yield [2]. Therefore, finding new plant growth regulators are an effective approach to improving crop yield in agriculture production [3]. water (as control) for 24 h under dark condition. Four treatments were applied: (1) Control (CK), (2) 10 µmol/L MEL solution (MEL 10), (3) 100 µmol/L MEL solution (MEL 100) and (4) 500 µmol/L MEL solution (MEL 500). The MEL solutions were prepared as follows: 0.23g MEL was dissolved in 10 mL ethyl alcohol as a stock solution (100 mmol/L). The stock solution was diluted with distilled water to yield 10, 100 and 500 µmol/L. The treated seeds were sown on 25 October and harvested on 25 May. Fifteen seeds were sown per pot at a depth of 3 cm, and 6 uniform plants remained in each pot when the fourth leaf appeared. All pots were placed under a rain shed in the field. During the growth period, the soil water contents were maintained between the 75-95% of maximum pot capacity [26]. The average air temperature and sunshine hours in the wheat growing season during 2013-2016 (2013-2015: pot experiment; 2015-2016: field experiment) are shown in Figure 1. The MEL solutions were prepared as follows: 0.23g MEL was dissolved in 10 mL ethyl alcohol as a stock solution (100 mmol/L). The stock solution was diluted with distilled water to yield 10, 100 and 500 μmol/L. The treated seeds were sown on 25 October and harvested on 25 May. Fifteen seeds were sown per pot at a depth of 3 cm, and 6 uniform plants remained in each pot when the fourth leaf appeared. All pots were placed under a rain shed in the field. During the growth period, the soil water contents were maintained between the 75-95% of maximum pot capacity [26]. The average air temperature and sunshine hours in the wheat growing season during 2013-2016 (2013-2015: pot experiment; 2015-2016: field experiment) are shown in Figure 1.

Grain Yield and Dry Matter Accumulation
The grain yield (calculated based on per plant) in the pot experiment was investigated in two continuous seasons (2013)(2014)(2014)(2015). After maturity, the plants were harvested. Yield components, including spike number plant −1 , grain number spike −1 , thousand-grain weight, and harvest index, were investigated. Harvest index was calculated as: Harvest index = total grain weight (dry weight)/total aboveground biomass (dry weight). The growth (dry matter accumulation) was investigated at four growth stages (elongation, flowering, grain filling and physiological maturity) during 2014-2015 seasons. Plants were harvested and separated into shoot and root. The roots of each pot were carefully rinsed with water. All plant components were dried at 80 °C to constant weight and weighed, and the root/shoot ratio was calculated. The tiller number was investigated during the elongation stage and spike number was investigated when harvest was carried out.

Leaf Area and Photosynthetic Characters
Leaf area was measured at the beginning of grain-filling stage: leaf area = leaf length × leaf width × 0.835 [27]; Photosynthetic traits were measured in 10-day intervals from flowering to grain filling stage both for 2013-2014 and 2014-2015 growing seasons. The photosynthetic rate of flag leaf was measured between 09:00 and 11:00 by a portable photosynthesis system (LI-6400XT; LI-COR

Grain Yield and Dry Matter Accumulation
The grain yield (calculated based on per plant) in the pot experiment was investigated in two continuous seasons (2013)(2014)(2014)(2015). After maturity, the plants were harvested. Yield components, including spike number plant −1 , grain number spike −1 , thousand-grain weight, and harvest index, were investigated. Harvest index was calculated as: Harvest index = total grain weight (dry weight)/total aboveground biomass (dry weight). The growth (dry matter accumulation) was investigated at four growth stages (elongation, flowering, grain filling and physiological maturity) during 2014-2015 seasons. Plants were harvested and separated into shoot and root. The roots of each pot were carefully rinsed with water. All plant components were dried at 80 • C to constant weight and weighed, and the root/shoot ratio was calculated. The tiller number was investigated during the elongation stage and spike number was investigated when harvest was carried out.

Leaf Area and Photosynthetic Characters
Leaf area was measured at the beginning of grain-filling stage: leaf area = leaf length × leaf width × 0.835 [27]; Photosynthetic traits were measured in 10-day intervals from flowering to grain filling stage both for 2013-2014 and 2014-2015 growing seasons. The photosynthetic rate of flag leaf was measured between 09:00 and 11:00 by a portable photosynthesis system (LI-6400XT; LI-COR Biosciences, Lincoln, NE, USA). The chlorophyll concentration was determined through measuring the SPAD value by a SPAD meter (SPAD-502, Konica-Minolta, Tokyo, Japan). Chlorophyll fluorescence parameter was measured with a pulse amplitude modulated chlorophyll fluorescence system (Imaging PAM, Walz, Effeltrich, Germany). Maximal quantum yield of PSII photochemistry (Fv/Fm) was obtained using Imaging Win software (Version 2.40, Walz, Effeltrich, Germany).

Experiment 2: Effects of Seed Pre-Soaking with MEL on Wheat Yield in Field Experiment
In order to further confirm the effects of the MEL on improving wheat grain yield, a field experiment was conducted during the 2015-2016 growing season. The seed pre-soaking with MEL was the same as that in the pot experiment. The random complete block design was applied with three replications, and each plot was 5 × 10 m with row space of 0.15 m. The fertilizers were supplied with 180 Kg N ha −1 , 140 kg P 2 O 5 ha −1 and 75 Kg K 2 O ha −1 (recommended dosage in this area) before sowing. The sowing rate was 150 kg ha −1 . Wheat seeds were sowed on 15 October 2015 and harvested on 25 May 2016. The grain yield was measured from an area of 4 m −2 . The spike number ha −1 , grain number per spike −1 , thousand-grain weight and harvest index were also measured.

Statistical Analysis
The effects of treatment, year and the interactions between the treatments and years were calculated by combined ANOVA in SPSS 19.0 software (IBM Company, Chicago, IL). The comparisons among different treatments were made by Duncan's multiple range tests. Statistical comparisons were significant when p < 0.05. Path coefficient analysis was carried out to partition the correlation coefficients of yield components into direct and indirect effects on grain yield [28].

The Effects of MEL Application on Grain Yield in Pot Experiment
Seed pre-soaking with MEL significantly improved wheat grain yield (calculated based per plant) in both two growing seasons in the pot experiment ( Table 1). The MEL treatment increased yield from 29% to 40% in the 2013-2014 growing season and from 45% to 79% in the 2014-2015 growing season, compared with control. In the two-year pot experiments, the highest yield was found at 100 µM MEL treatments. The analysis of variance for yield in the first two growing seasons showed that year and MEL treatment had significant effects on grain yield, whereas interaction effects between year and MEL treatment were not significant. Path coefficient analysis was conducted to describe the effects of yield components ( Table 2). Spikes number per plant had the highest positive direct effect on yield (1.247), followed by grains number, spike −1 (1.056) and thousand-grain weight (0.414). The correlation coefficient between the grain yield and the spike number, grain yield and grain number spike −1 , the grain yield and grain weight were 0.335, 0.37 and 0.28, respectively.

The Effects of MEL Application on Tiller Number and Dry Matter Accumulation During the Growth Process in the Pot Experiment
MEL treatment increased the tiller number during the elongation stage, and increased the spike number per plant regarding harvest in the pot experiment (Table 1). Tiller number plant −1 was increased by 25% and 56% under 100 µM MEL treatments in two growing seasons, and the spike number was increased by 17% and 60% under 100 µM MEL treatments in two growing seasons. In pot experiment, both shoot and root dry weights were enhanced in MEL-treated plants during growth stage ( Figure 2). The higher dry matter accumulation was found at 100 µM MEL treatments. The shoot and root dry matter accumulations were increased by 57% and 62% during the mature period, respectively. The different extents of increase between shoot and root dry matters by MEL also lead to increased root/shoot ratio. Compared with the control plants, the ratio of root/shoot increased by 18% during the jointing stage, 20% during the flowering stage and 21% during the filling stage by 100 µM MEL treatments. Values are means ± SEs of twelve replicates. Significant differences between different treatments are indicated by different letters (p < 0.05).

The Effects of MEL Application on Leaf Area and Photosynthetic Characters During the Grain Filling Stage in the Pot Experiment
There were significant differences in leaf area plant -1 among treatments ( Figure 3). MEL-treated plants showed a tendency of increased leaf area under pot cultural conditions during both two growing seasons. The larger leaf area was found at under 100 μM MEL treatments, especially during the 2014-2015 season. The leaf area plant −1 under 100 μM MEL treatments was 32% and 53% larger Values are means ± SEs of twelve replicates. Significant differences between different treatments are indicated by different letters (p < 0.05).

The Effects of MEL Application on Leaf Area and Photosynthetic Characters During the Grain Filling Stage in the Pot Experiment
There were significant differences in leaf area plant −1 among treatments (  The photosynthetic rate of flag leaf begins to decrease after initial flowering, but the MEL-treated plants maintained higher photosynthetic rate compared with control plants (Figure 4). The higher photosynthetic rate was found in 100 μM MEL-treated plants. The photosynthetic rate in flag and the top third leaf under 100 μM MEL treatments were 25% and 42% higher than that of control plants after 40    The photosynthetic rate of flag leaf begins to decrease after initial flowering, but the MEL-treated plants maintained higher photosynthetic rate compared with control plants (Figure 4). The higher photosynthetic rate was found in 100 µM MEL-treated plants. The photosynthetic rate in flag and the top third leaf under 100 µM MEL treatments were 25% and 42% higher than that of control plants after 40   The photosynthetic rate of flag leaf begins to decrease after initial flowering, but the MEL-treated plants maintained higher photosynthetic rate compared with control plants (Figure 4). The higher photosynthetic rate was found in 100 μM MEL-treated plants. The photosynthetic rate in flag and the top third leaf under 100 μM MEL treatments were 25% and 42% higher than that of control plants after 40

The Effects of MEL Application on Grain Yield in Field Experiment
In order to further confirm the effects of MEL application on wheat yield under field conditions, a field experiment was conducted during the 2015-2016 growing season. As shown in Table 3, grain yields of MEL-treated plants were significantly higher than those of the control. The grain yield increased by 4-19% compared with control in different concentrations of MEL application. The highest grain yield was shown at 500 μM MEL treatment. The further investigation showed that MEL-treatment increased the spike number, grain number per spike and grain weight, and increased the yield.

The Effects of MEL Application on Grain Yield in Field Experiment
In order to further confirm the effects of MEL application on wheat yield under field conditions, a field experiment was conducted during the 2015-2016 growing season. As shown in Table 3, grain yields of MEL-treated plants were significantly higher than those of the control. The grain yield increased by 4-19% compared with control in different concentrations of MEL application. The highest grain yield was shown at 500 µM MEL treatment. The further investigation showed that MEL-treatment increased the spike number, grain number per spike and grain weight, and increased the yield.

Discussion
Seed pre-soaking with MEL achieved high grain yield during both two growing seasons in the pot experiment. The effect of MEL application on wheat grain yield also showed a dosage effect. The high grain yield could be ascribed to the high spike number per plant, high grain number per spike and high grain weight (Tables 1 and 2). In the pot experiment, MEL treatment increased the grain yield by 29-80%, compared with no MEL treatment. In the field experiment, MEL treatment increased the grain yield by 4-19%, compared with controls. Similar extents of yield increase have been reported in corn (20%) and mung bean (30%) by previous studies [7,10,17]. Taken together, these results show that there is a great value MEL has, for improving the wheat production.
In agriculture production, crop yield depends on both the dry matter accumulation during the whole season and the dry matter distribution (parting) when harvesting [29]. In this study, MEL-treated plants maintained the large dry matter accumulation; these results are supported by the observations that MEL promotes vegetative growth [30]. Dry matter distribution, presented by harvest index, was also enhanced by Mel treatment. The yield components analysis showed that MEL increased the yield by increasing both spike number and grain number (Table 1). Spike number depends on the tiller number and its survival after elongation [31]. In this study, the increased number of spikes could be ascribed to the MEL-treated induced increase in the tiller number, but not survival rate. And grain number is another reason that contributes to the increased yield in this study ( Table 1).
Seeds of plants contain a large amount of MEL, and seeds' endogenous MEL concentration could be significantly increased by exogenous MEL treatment [32,33]. In this study, MEL treatment increased the tilling number, which resulted in increased spike number. It has been reported that promoting shoot growth in rapidly developing tissues, particularly during germination and seedling development, is one of typical characteristics of MEL [20]. Increased photosynthetic rate could result in sufficient carbohydrate supplementation, which would be beneficial to tiller growth. Although how MEL regulates the tiller emergence is ambiguous now, the underlying mechanisms of MEL promoting seed germination and seedling growth are clearer. During seed germination, the endogenous indoleacetic acid (IAA) was found to be increased by exogenous MEL [16]. More evidence showed that there is significant crosstalk between MEL and other plant growth regulators, including cytokinin, salicylic acid, jasmonic acid, gibberellins (GA), abscisic acid (ABA) and ethylene [7,34,35]. As such, during the early stage of germination, MEL down-regulated the ABA biosynthesis gene and up-regulated the ABA catabolism gene; meanwhile, MEL up-regulated the GA biosynthesis gene, which resulted in a rapid decrease in ABA contents and increase in GA contents [36]. Our current study supports that MEL prompting the shoot growth during the germination and seedling growth stage, which could increase the spike number when harvesting. In addition to increasing the tiller number, increasing the grain number and grain weight could be other main factors that influence grain yield when harvesting. In wheat production, maintaining high photosynthetic rate during the flowing and grain filling stages contributes to enhanced grain number and grain weight [37]. In addition, in the later filling stage, delay the leaf senesces is prominent for achieving high yield [38]. In this study, the MEL treated plants maintained high photosynthetic rate and delayed the leaf senescence during grain filling stage. Although the function of MEL in enhancing the photosynthetic rate is largely unclear, the available evidence suggests that MEL is important not only in helping to combat the excess reactive oxygen species (ROS) produced in actively photosynthesizing tissues, but that MEL is involved in perception and response to different intensities and wavelengths of light [34,35]. In addition to directly enhanced photosynthesis rate, the facts that MEL application delays leaf senesces and protects the chlorophyll from degradation have also been widely reported in different studies, and one of the important reasons is that MEL enhances the plant antioxidant ability either directly or by activating the plant enzymatic or non-enzymatic antioxidant system, decreased the H 2 O 2 induced cell death and protein degradation [20,39]. Recently, MEL was found to up-regulate ABA catabolic genes and down-regulate ABA biosynthetic genes, resulting in a rapid reduction in ABA, which was involved in MEL-regulated leaf senescence [36]. Under drought and heat stress, MEL suppressed leaf senescence through block ABA signal pathway and promote the cytokinin pathway [19,40]. Although in this study, there was not drought nor heat treatment, the plant will unavoidably suffer these stresses during the growing season in natural conditions. In addition to enhancing photoassimilate, photoassimilate partitioning also affects the grain yield. Although there are no specific studies on the effect of MEL on the photoassimilate partitioning, the exogenous MEL regulated the carbon assimilation and degradation were found at transcript level. Such sugar metabolism-related genes were altered by MEL and contributed the plant growth and yield in soybeans [10]. The increase in harvest index by MEL application was also found in this study, suggesting that MEL is involved in alteration of photoassimilate partitioning.
In addition to the directly effects of MEL on shoot growth, MEL regulating the root development has also been confirmed in early 2000s [41]. MEL has been found to promote root branching, particularly the adventitious root formation in several species [5]. In crop production, one beneficial function of MEL on improving crop production is to promote the development of strong root system, which could lead to a greater absorptive capacity of nutrients and water from the soil [5,21]. For cucumbers, MEL stimulated the root germination and vitality and increased the root/shoot ratio; therefore, MEL may have an effect on strengthening cucumber root growth [42]. In this study, MEL-treated plants maintained high root/shoot ratio during the jointing, flowing and filling stages ( Figure 2). The high root/shoot ratio could contribute to achieving high yield by increasing the nutrients and water uptake, which could also be involved in regulating the wheat growth and yield. Recently, more evidence has supported that auxin participates in the MEL modulation of root growth [5,43,44]. Exogenously applied MEL stimulates root growth and raises IAA in root of etiolated seedling of Brassiea jucea (reported recently) [45]. In addition, Chen et al. [46] found that H 2 O 2 acts downstream of MEL to induce lateral root formation.
In this study, MEL treatment increased the grain yield by 29-80% in pot experiment, but only by 4-19% in field experiment. Further analysis showed that the maximum spike number per plant was increased by 15% and 57% in continuous two years pot experiment, but maximum spike number per area just increased by 5.5% in field experiment. Spike number has a maximum contribution to increase the yield among three yield components that were analyzed in this study (Table 2). Therefore, a lesser increase in spike number in a field experiment than in a pot experiment may lead to lesser increase in wheat yield in field experiment. In the pot experiment, the increase in tiller number was not limited by spaces because of the big edge effect; it was limited in field experiment because of the population effect. Thus, when yield was presented base on per plant in pot, it could enlarge the effect of MEL on promoting the grain yield. In addition, although the alleviating stress tolerance was not considered in this study, the wheat could also suffer stresses during the growing season; the contribution of MEL through alleviating the stresses cannot be ignored. The difference in environment and stress during the two years could also lead to different effects of MEL on enhancing yield in two-year experiments (Table 1). Besides, the differences in climate in three years may also present other reasons for the inconsistency of the yield increase between the pot and filed studies.
Seed pre-soaking with MEL largely increased the grain yield of winter wheat both in the pot and field experiments. Wheat is the largest staple crop in the world, yielding more than 7.5 million tons per year [47]. Therefore, a small increase in wheat yield will result in a big effect on food security in the word. Based on our current study and previous research about the mechanism of the MEL on plant development [29,[34][35][36][37]40,45], a potential mechanistic diagram of MEL on improving wheat grain yield is proposed for future research reference. (1) During the germination and seedling growth stages, the MEL could crosstalk with IAA and GA, increasing the tiller number and seedling growth, which leads to an increase in spike number. (2) During the flowering and grain filling stages, MEL may crosstalk with ABA and cytokinin, eliminate ROS and delay leaf senescence, and enhance carbon assimilation capacity, which leads to increased grain weight and grain number. (3) In the root, MEL may crosstalk with auxin and H 2 O 2 and promote root/shoot ratio (Figure 7). It is worth noticing that in the current study, when the seeds were pre-soaked with MEL, the priming effect could last the whole growing season, which was finally reflected by improved yield, suggesting that MEL has great prospects in agricultural production through exogenous application or even cultivated the plant with high MEL content.

Conflicts and interests:
Authors declare no conflict of interest.

Conflicts of Interest:
Authors declare no conflict of interest.