Methyl jasmonate mediates melatonin-induced cold tolerance of grafted watermelon plants

Root–shoot communication has a critical role in plant adaptation to environmental stress. Grafting is widely applied to enhance the abiotic stress tolerance of many horticultural crop species; however, the signal transduction mechanism involved in this tolerance remains unknown. Here, we show that pumpkin- or figleaf gourd rootstock-enhanced cold tolerance of watermelon shoots is accompanied by increases in the accumulation of melatonin, methyl jasmonate (MeJA), and hydrogen peroxide (H2O2). Increased melatonin levels in leaves were associated with both increased melatonin in rootstocks and MeJA-induced melatonin biosynthesis in leaves of plants under cold stress. Exogenous melatonin increased the accumulation of MeJA and H2O2 and enhanced cold tolerance, while inhibition of melatonin accumulation attenuated rootstock-induced MeJA and H2O2 accumulation and cold tolerance. MeJA application induced H2O2 accumulation and cold tolerance, but inhibition of JA biosynthesis abolished rootstock- or melatonin-induced H2O2 accumulation and cold tolerance. Additionally, inhibition of H2O2 production attenuated MeJA-induced tolerance to cold stress. Taken together, our results suggest that melatonin is involved in grafting-induced cold tolerance by inducing the accumulation of MeJA and H2O2. MeJA subsequently increases melatonin accumulation, forming a self-amplifying feedback loop that leads to increased H2O2 accumulation and cold tolerance. This study reveals a novel regulatory mechanism of rootstock-induced cold tolerance.


Introduction
As sessile organisms, plants frequently face challenges from various environmental factors throughout their life cycle. In particular, cold stress is one of the most destructive abiotic stresses due to its adverse effects on plant growth and development and subsequent negative impacts on crop productivity 1 . To adapt and survive cold exposure, plants have evolved sophisticated defense mechanisms. When a plant senses that temperature decreases via molecular sensors, the production of secondary messengers is triggered, and a set of transcriptional regulators are subsequently activated to regulate plant tolerance 2 . At the whole-organism level, root-to-shoot communication is crucial for the increased survival of plants under environmental stress. Under drought conditions, roots produce more abscisic acid (ABA), which is then transported to the leaves to reduce water loss via stomatal closure 3 . In many horticultural crop species, grafting is widely applied to enhance plant tolerance to various environmental stresses, such as soil-borne pathogens, salt, and low temperature. Grafting-induced plant tolerance is associated with the inherent resistance of rootstocks and some rootstock-sourced signals that are transported to shoots and subsequently regulate shoot responses 4 . Therefore, grafting is also a useful tool to reveal the signaling mechanisms related to root-shoot communication. For instance, by using cucumber scions grafted onto heat-tolerant luffa rootstock, Li et al. 5 revealed that rootproduced ABA as a long-distance signal could alter the expression of heat shock protein (HSP) 70 and subsequently improve heat tolerance of the shoots.
Melatonin (N-acetyl-5-methoxytryptamine) was initially identified as an essential animal hormone that has regulatory roles in various biological processes 6 . In 1995, melatonin was identified in vascular plants for the first time 7,8 . A number of subsequent studies have shown phytomelatonin to be an essential regulator in plant growth and development; postharvest physiology; and defense against various environmental stresses, such as pathogen infection, drought, salinity, nutrient deficiency, and heavy metals [9][10][11] . In particular, the recent identification of the first phytomelatonin receptor (CAND2/ PMTR1) in Arabidopsis has opened the door to consider melatonin a new phytohormone 12,13 . Increasing numbers of studies have indicated that melatonin enhances the cold tolerance of various plant species, including melon, tomato, watermelon, and Arabidopsis 14 . Moreover, some evidence has shown melatonin to be a novel long-distance signal that can be transported from roots to shoots [15][16][17] . Therefore, it would be interesting to investigate the involvement of melatonin in grafting-induced cold tolerance.
In plants, jasmonates (JAs) such as jasmonic acid (JA) and its methyl ester (methyl jasmonate, MeJA) act as important phytohormones that regulate multiple plant processes, such as seed germination, root growth, flowering, leaf senescence, and defense responses to various biotic and abiotic stresses 18,19 . Accumulating amounts of data have shown that JAs have a positive role in inducing plant tolerance to cold stress 19 . Cold exposure rapidly induces JA biosynthesis-related genes and subsequent JA accumulation 20 . Moreover, exogenous application of MeJA enhances Arabidopsis tolerance to cold, while mutants defective in JA biosynthesis or signaling exhibit hypersensitivity to cold stress 19 . Recent studies involving grafting experiments have shown that JA is involved in root and shoot communication to fine-tune plant responses to shoot wounding and root-knot nematodes 21,22 . However, little is known about the involvement of JA in root-shoot communication under cold stress.
Watermelon is a widely cultivated vegetable crop species worldwide and is very sensitive to cold stress 23 . Grafting onto pumpkin can induce watermelon tolerance to cold 24,25 . Our recent study suggests that melatonin application to the roots can confer cold tolerance to the shoot via xylem transport, and such induction involves the expression of several genes involved in JA signaling and the production of H 2 O 2 17 , an important secondary messenger in the cold response 26 . This raises the possibility that melatonin has an important role in rootstock-induced shoot tolerance against the cold by interacting with JA and H 2 O 2 . To test this assumption, the present study evaluated the roles of melatonin, MeJA, and H 2 O 2 and their interaction in rootstock-induced cold tolerance. Our results provide convincing evidence that melatonin-induced cold tolerance of grafted watermelon plants essentially involves MeJA and H 2 O 2 signaling. This study provides novel insight into the mechanism of rootstock-induced cold tolerance in cucurbits.

Plant materials
Three cucurbit species, namely, watermelon (Citrullus lanatus (Thunb.) Matsum. & Nakai cv. Nongkeda No. 5, Cl), pumpkin (Cucurbita moschata Duch cv. Weizhen No. 1, Cm), and figleaf gourd (Cucurbita ficifolia Bouché, Cf), were used in the current study. Germinated seeds of Cl, Cm, and Cf as rootstocks were sown into plastic pots (7 × 7 × 7.5 cm (length × width × height, respectively)) filled with 210 cm 3 of commercial peat-based compost, and 7 d later, germinated seeds of Cl (as scions) were sown. Top insertion grafts were performed when the cotyledons of watermelon (as scions) had fully expanded 27 . The resulting three groups of grafted seedlings were designated Cl/Cl, Cl/Cm, and Cl/Cf; Cl/Cl plants were used as control. The plants were grown in growth chambers with a temperature of 25/18°C (day/night), a 12-h photoperiod, and photosynthetic photon flux density (PPFD) of 400 μmol m −2 s −1 . The seedlings were watered every 2 d and supplied with Hoagland's nutrient solution every 3 d.
Experimental design Experiment 1. To evaluate the effects of different rootstocks on scion tolerance to cold stress, Cl/Cl, Cl/Cm, and Cl/Cf plants with four true leaves were transferred into growth chambers maintained at 25°C for the control treatment or 4°C for cold treatment. At 0, 6, 12, 18, and 24 h after cold treatment, the relative expression levels of C-REPEAT BINDING FACTOR 1 (ClCBF1) and ClCBF2 were measured. At 36 h after cold treatment, the maximum photochemical efficiency of PSII (Fv/Fm) and the relative electrical conductivity (REC) were measured. At 12 h after cold treatment, root or leaf samples were harvested for biochemical assays. Experiment 2. To determine the effects of exogenous melatonin, MeJA, and H 2 O 2 on the cold tolerance of Cl/Cl seedlings, the leaves were first sprayed with distilled water (as a control), melatonin at 150 μM 17 , MeJA at 200 μM 19 , or H 2 O 2 at 2 mM 5 . Melatonin (Sigma-Aldrich, St. Louis, MO, USA) or MeJA (Sigma-Aldrich) was dissolved in ethanol followed by dilution with distilled water at a ratio of 1/10,000 (v/v). Each plant was sprayed with 20 mL of the respective chemical solution or distilled water (as a control). At 12 h after pretreatment with melatonin, MeJA, or H 2 O 2 , the plants were subjected to 4°C temperature. To block the synthesis or accumulation of melatonin, JA, or H 2 O 2 in Cl/Cl plants, the leaves were pretreated with 100 μM p-chlorophenylalanine (CPA, a melatonin synthesis inhibitor) 28,29 , 5 mM diethyldithiocarbamic acid (DIECA, a JA biosynthesis inhibitor) 30 , or 20 μM diphenylene iodonium (DPI, an inhibitor of NADPH oxidase, which produces H 2 O 2 ) 31 . Eight hours later, the plants were sprayed with melatonin or MeJA, and 12 h later, they were exposed to 4°C. Experiment 3. To assess the involvement of melatonin, MeJA, and H 2 O 2 and their interactions in rootstockinduced cold tolerance, the leaves of Cl/Cm and Cl/Cf seedlings were sprayed with CPA, DIECA, or DPI 8 h prior to cold exposure. The leaves were sampled at 12 h, whereas the Fv/Fm and REC were measured at 36 h after cold exposure.

Cold tolerance assays
The Fv/Fm was measured using an imaging pulseamplitude-modulated (PAM) chlorophyll fluorometer (Heinz Walz, GmbH, Effeltrich, Germany) according to the method described by Li et al. 5 . The REC was determined and calculated as described by Hong et al. 32 .

Melatonin measurements
Melatonin was extracted and measured as described previously 33 . Frozen samples (0.5 g) were homogenized in 5 mL of acetone-methanol buffer (acetone/methanol/water = 89/10/1) on ice. After centrifugation (5,000 g, 4°C) for 10 min, 0.5 mL of trichloric acid (1%) was added to the supernatant for protein precipitation. After centrifugation at 10,000×g for 10 min at 4°C, the supernatants were used to quantify melatonin levels using an ELISA kit (Shanghai Lanpai Biotechnology Co., Ltd., Shanghai, China) according to the manufacturer's instructions.

Quantification of MeJA
MeJA was extracted as described by Pan et al. 34 . For the extraction of MeJA, 0.5 g of frozen leaf samples was homogenized in 5 mL of 1-propanol/H 2 O/concentrated HCl (2/1/0.002, v/v/v) and subsequently incubated overnight. The extracts and 5 mL of dichloromethane were then mixed together, after which the mixture was shaken at 4°C for 30 min. After centrifugation, the obtained lower phase was dried under a stream of N 2 gas. The residue was then dissolved in methanol. MeJA concentrations were analyzed using an ELISA kit (China Agricultural University, Beijing, China) following the manufacturer's instructions.

Analysis of H 2 O 2
H 2 O 2 in the leaf samples was measured as described by Bellincampi et al. 35 . Briefly, the leaf material (0.3 g) was homogenized in 2 mL of 0.2 M HClO 4 . After centrifugation (10,000×g, 4°C) for 15 min, an aliquot of supernatant (0.5 mL) was added to 0.5 mL of assay reagent consisting of 500 μM ammonium ferrous sulfate, 200 μM xylenol orange, 50 mM H 2 SO 4 , and 200 mM . The data from a to d were measured at 36 h after cold treatment. e Changes in the relative expression of ClCBF1 and ClCBF2 at 0, 6, 12, 18, and 24 h, respectively, after cold treatment. The data are the means of three replicates (±SDs). The means denoted with different letters differed significantly at P < 0.05 sorbitol. After incubation for 1 h, the absorbance at OD 560 was recorded. DAB staining of H 2 O 2 was performed following the protocol described by Thordal-Christensen et al. 36 .

qRT-PCR analysis
Total RNA was extracted from the leaves using an RNA extraction kit (AxGen, Union City, CA, USA). Residual DNA was removed using a DNase Mini Kit (Qiagen, Hilden, Germany). The isolated total RNA (1 μg) was reverse transcribed using a ReverTra Ace qPCR RT Kit (Toyobo, Osaka, Japan). qRT-PCR was then performed using an iCycler iQ Multicolor PCR Detection System (Bio-Rad, Hercules, CA, USA) as described by Li et al. 17 . The primers used for qRT-PCR are listed in Supplementary Table S1. The relative expression levels were standardized to those of watermelon β-ACTIN and were calculated as described by Livak and Schmittgen 37,38 .

Statistical analysis
The experiment was carried out in accordance with a completely randomized design, with three independent biological replicates. Each replicate included 15 plants. For statistical analysis, the data were analyzed using variance (ANOVA), and P values < 0.05 were considered statistically significant according to Tukey's test.

Pumpkin and figleaf gourd rootstocks induced cold tolerance in watermelon shoots
We first compared the leaf phenotypes, Fv/Fm values, and REC of watermelon plants grafted onto the rootstocks of watermelon (Cl/Cl), pumpkin (Cl/Cm), and figleaf gourd (Cl/Cf). Under normal conditions at 25°C, no significant differences in plant growth or Fv/Fm were observed among the different grafted plants, although the REC of the Cl/Cm and Cl/Cf plants was slightly higher than that of the Cl/Cl plants (Fig. 1a-d). Cold stress caused severe leaf wilting, a significant decrease in Fv/Fm, and an increase in REC of the Cl/Cl plants. However, pumpkin and especially figleaf gourd as rootstocks alleviated the cold-induced wilting, Fv/Fm decrease, and REC increase in watermelon leaves. For example, the Fv/Fm of Cl/Cm and Cl/Cf plants were 0.46 and 0.55, respectively, which was much higher than that (0.31) of Cl/Cl plants at 36 h after cold exposure.
The relative expression of ClCBF1 and ClCBF2 was upregulated after cold treatment, and the highest peak occurred at 12 h in nearly all of the grafted plants; however, the highest expression of ClCBF1 in the Cl/Cl plants occurred at 24 h after cold exposure (Fig. 1e). Importantly, Cl/Cm and Cl/Cf plants showed more significant increases in transcript levels of ClCBF1 and ClCBF2 than did Cl/Cl plants after cold exposure. For example, ClCBF1 transcript levels were upregulated 98.1-, 361.6-, and 266.4-fold in Cl/Cl, Cl/Cm, and Cl/Cf plants, respectively, while the ClCBF2 transcripts were upregulated 5.3-, 13.3-, and 28.4-fold, respectively, at 12 h after cold stress.

Different rootstocks induced melatonin accumulation differently in leaves of watermelon plants under cold stress
At the optimum growth temperature, the melatonin contents in pumpkin and figleaf gourd roots were similar to and lower than those in watermelon roots, respectively (Fig. 2). However, both pumpkin and figleaf gourd as Interestingly, the melatonin contents in the roots of Cl/Cm and especially Cl/Cf were obviously higher than those in the Cl/Cl roots after cold exposure. Moreover, the levels of melatonin and ClASMT transcripts in the Cl/Cm and Cl/Cf leaves were also higher than those in the Cl/Cl leaves.

Pumpkin and figleaf gourd rootstocks induced MeJA and H 2 O 2 accumulation in leaves of watermelon plants under cold stress
Under normal conditions at 25°C, pumpkin or figleaf gourd as rootstocks decreased the MeJA contents but increased the relative expression of JASMONATE ZIM-DOMAIN 1 (ClJAZ1, a repressor gene involving in JA signaling) in watermelon leaves (Fig. 3a). The MeJA content in Cl/Cl leaves was almost unchanged after exposure to cold stress; however, that in Cl/Cm and Cl/Cf leaves significantly increased. The MeJA contents in the Cl/Cm and Cl/Cf leaves were 40.4% and 19.7% higher than those in the Cl/Cl leaves, respectively, at 12 h after cold stress. The relative expression of ClJAZ1 was upregulated in the Cl/Cl leaves but downregulated in the Cl/ Cm and Cl/Cf leaves of plants under cold stress.
Unexpectedly, no significant differences were observed in root MeJA content among the different grafted plants under cold stress ( Supplementary Fig. S1).
The H 2 O 2 generated by RESPIRATORY BURST OXI-DASE HOMOLOG (RBOH) functions as an important signaling molecule in regulating plant tolerance to cold stress. Thus, we investigated whether H 2 O 2 is involved in rootstock-induced cold tolerance. At optimum growth temperatures, pumpkin as a rootstock induced the accumulation of ClRBOHD transcripts in watermelon leaves compared to watermelon rootstocks (Fig. 3b). However, there were no significant differences in ClRBOHF transcripts or H 2 O 2 content among any of the grafted plants (Fig. 3b, c). Exposure to 4°C increased H 2 O 2 accumulation and ClRBOHD and ClRBOHF transcript levels in all grafted plants. The leaves of Cl/Cm and especially Cl/Cf plants showed higher H 2 O 2 levels and ClRBOHD/F transcripts than did those of Cl/Cm under cold stress. For example, the H 2 O 2 contents in the Cl/Cm and Cl/Cf leaves were 41.8% and 91.3% higher than those in the Cl/ Cl leaves, respectively, after cold exposure.

Role of melatonin in rootstock-induced accumulation of MeJA and H 2 O 2 and cold tolerance
To evaluate the role of melatonin in rootstockinduced cold tolerance, we first analyzed the effect of exogenous melatonin on the cold tolerance of Cl/Cl plants. Melatonin (150 μM) pretreatment alleviated the cold-induced Fv/Fm decrease and REC increase  (Fig. 4a). In melatonin-pretreated plants, the Fv/Fm was 14.8% higher while the REC was 35.0% lower after exposure to cold stress than in the control plants. As CPA application can prevent melatonin biosynthesis 29 , we thus evaluated the effect of CPA on the cold tolerance of Cl/Cm and Cl/Cf plants. Pretreatment with CPA attenuated the pumpkin-and figleaf gourd rootstockinduced alleviation of cold stress, as reflected by the Fv/Fm decrease and REC increase (Fig. 4b).
At the optimum growth temperatures, exogenous melatonin application did not induce significant changes in MeJA and H 2 O 2 levels in Cl/Cl leaves (Fig. 4c, d). However, the contents of MeJA and H 2 O 2 increased by 42.9% and 44.1%, respectively, in response to melatonin under cold stress. More importantly, pretreatment with CPA attenuated or completely blocked pumpkin-and figleaf gourd rootstock-induced increases in MeJA and H 2 O 2 accumulation in the leaves (Fig. 4e, f). These data suggest that melatonin has an important role in pumpkin-or figleaf gourd rootstock-induced accumulation of MeJA and H 2 O 2 and cold tolerance.

Roles of MeJA and H 2 O 2 in melatonin-or rootstockinduced cold tolerance
To evaluate the roles of MeJA and H 2 O 2 in cold tolerance acquired by grafting and melatonin, we first evaluated the effects of foliar applications of MeJA or H 2 O 2 on the cold tolerance of Cl/Cl plants. Both MeJA (200 μM) and H 2 O 2 (2 mM) application alleviated the cold-induced decrease in Fv/Fm and increase in REC of Cl/Cl plants (Fig. 5a). Pretreatment with DIECA (an inhibitor of JA biosynthesis, 5 mM) or DPI (an inhibitor of H 2 O 2 production, 20 μM) attenuated or abolished the pumpkin-or figleaf gourd rootstock-induced alleviation of cold stress, as reflected by a decline in Fv/Fm and an increase in REC (Fig. 5b). Furthermore, DIECA or DPI application also prevented the melatonin-induced Fv/Fm increase and REC decrease after exposure to cold stress (Fig. 5c). Taken together, these results suggest that MeJA and H 2 O 2 are involved in melatonin-or rootstockinduced cold tolerance.

Role of MeJA in rootstock-induced accumulation of melatonin and H 2 O 2
To evaluate whether MeJA regulates melatonin biosynthesis in a feedback manner, we analyzed the response of melatonin to MeJA. The application of MeJA significantly increased melatonin accumulations in the Cl/Cl leaves after cold stress (Fig. 6a), while DIECA attenuated the pumpkin-or figleaf gourd rootstock-induced increase in melatonin contents (Fig. 6b). Moreover, CPA pretreatment attenuated the MeJA-induced Fv/Fm increase c Changes in the Fv/Fm and REC of Cl/Cl plants. The plant leaves were sprayed with DIECA or DPI 8 h prior to melatonin treatment. Twelve hours later, the plants were exposed to cold at 4°C. The data are presented as the means of three replicates (±SDs). The means denoted with different letters differ significantly at P < 0.05 and REC decrease of Cl/Cl leaves of plants under cold stress (Fig. 6c).
To further characterize the role of H 2 O 2 in MeJAenhanced cold tolerance, we first evaluated the effects of MeJA on H 2 O 2 accumulation in Cl/Cl plants. MeJA application induced an increase in H 2 O 2 accumulation under cold stress (Fig. 7a), while DIECA pretreatment completely abolished the melatonin-induced increase in H 2 O 2 . Similarly, DIECA also completely blocked the pumpkin-and figleaf gourd rootstock-induced increase in H 2 O 2 accumulation under cold stress (Fig. 7b). Furthermore, DPI pretreatment attenuated the MeJA-induced alleviation of cold stress, as reflected by the Fv/Fm decrease and REC increase (Fig. 7c).

Melatonin is involved in rootstock-induced cold tolerance
Grafting onto tolerant rootstocks is well known to enhance plant tolerance to various environmental stresses, such as soil-borne pathogens, cold, and salinity.
Consistent with the findings of previous studies 24, 25 , we found that the rootstocks of pumpkin or figleaf gourd enhanced watermelon tolerance to cold stress (Fig. 1). By using RNA-seq analysis, Xu et al. 39 found that pumpkin rootstocks could alter the expression of COLD-RESPONSIVE (COR) genes in shoots of watermelon plants under chilling stress. CBFs, the major activators of a subset of COR genes, have essential roles in modulating plant responses to cold stress 40 . In the present study, pumpkin or figleaf gourd as rootstocks significantly increased the transcripts of ClCBF1 and ClCBF2 in watermelon leaves after cold exposure (Fig. 1e). Therefore, it is plausible that some long-distance signal(s) originating from rootstocks are involved in inducing the shoot response to cold stress.
Increasing amounts of evidence have demonstrated that melatonin can induce cold tolerance in various plant species, and such induction is associated with the upregulated expression of CBFs 41,42 . Here, we observed that pumpkin or figleaf gourd rootstocks increased the  (Fig. 2). Furthermore, exogenous melatonin improved the cold tolerance of Cl/Cl plants, while inhibition of melatonin biosynthesis by CPA attenuated pumpkin-or figleaf gourd rootstockinduced cold tolerance (Fig. 4a, b). These results suggest that melatonin has an important role in pumpkin-or figleaf gourd rootstock-induced cold tolerance. It is worth noting that the increased melatonin contents in Cl/Cm or Cl/Cm leaves were accompanied by greater increases in melatonin contents in pumpkin or figleaf gourd rootstocks, respectively, after exposure to cold stress. Our previous study provided evidence that, as a long-distance signal, melatonin can be transported from roots to shoots via the xylem, subsequently inducing shoot tolerance to cold stress 17 . These data thus suggest that under cold stress, pumpkin-or figleaf gourd rootstock-sourced melatonin may act as a long-distance signal that induces melatonin accumulation and cold tolerance in watermelon leaves.
Melatonin and MeJA function together in a self-amplifying feedback loop in rootstock-induced cold tolerance Like melatonin, JAs have important roles in regulating plant tolerance to cold stress 43 . Under cold stress, JAs trigger the degradation of JAZ proteins, which releases INDUCER OF CBF EXPRESSION (ICE) from repression and then activates CBF-mediated transcriptional regulatory cascades 20 . In the current study, we observed that pumpkin and figleaf gourd rootstocks led to increased MeJA accumulation but decreased ClJAZ1 transcript levels in leaves of watermelon plants under cold stress (Fig. 3a). Moreover, exogenous MeJA improved the cold tolerance of Cl/Cl plants (Fig. 5a), while inhibition of JA biosynthesis by DIECA decreased the cold tolerance of Cl/Cm and Cl/Cf plants (Fig. 5b), suggesting that MeJA is involved in rootstock-induced cold tolerance.
Recently, crosstalk between melatonin and MeJA in plant defense against biotic stress has been revealed in some studies. For instance, Liu et al. 44 revealed that melatonin induces tomato fruit resistance to Botrytis cinerea by activating the JA signaling pathway. However, the crosstalk between melatonin and MeJA in abiotic stress responses remains largely unknown. Our results showed that melatonin and MeJA increased the accumulation of each other in the leaves of Cl/Cl plants under cold stress (Figs. 4c and 6a), while pretreatment with CPA or DIECA attenuated the pumpkin-or figleaf gourd rootstock-induced increase in MeJA or melatonin, respectively (Figs. 4e and 6b). Furthermore, melatonin-and MeJA-induced cold tolerance was attenuated or completely blocked by DIECA and CPA, respectively, in Cl/Cl plants (Figs. 5c and 6c). Taken together, these results suggest that melatonin and MeJA function together in a self-amplifying feedback loop, in which melatonin induces MeJA accumulation and MeJA subsequently increases melatonin accumulation during the cold response of Cl/Cm and Cl/Cf plants. Such crosstalk between melatonin and MeJA during the cold response is similar to that between ABA and H 2 O 2 in plant responses to heat and oxidative stresses 45 .

MeJA mediates melatonin-and rootstock-induced H 2 O 2 accumulation and subsequent cold tolerance
As an important secondary messenger, the H 2 O 2 generated by RBOH has an essential role in regulating plant tolerance against various abiotic stresses, including cold stress 26,46 . In our study, pumpkin or figleaf gourd as rootstocks increased the accumulation of H 2 O 2 , as well as the levels of ClRBOHD and ClRBOHF transcripts under cold stress (Fig. 3b, c). Additional experiments showed that exogenous application of H 2 O 2 enhanced the cold tolerance of Cl/Cl plants, while inhibition of H 2 O 2 generation by DPI prevented pumpkin-or figleaf gourd rootstock-induced cold tolerance (Fig. 5a, b). These results indicate that H 2 O 2 is involved in rootstockinduced cold tolerance, and these findings are consistent with previous findings that H 2 O 2 is involved in rootstockinduced heat tolerance of cucumber plants 5 .
Melatonin is a well-documented antioxidant that can effectively remove reactive oxygen species (ROS) and consequently alleviate oxidative stress 47 . However, accumulating studies have shown that melatonin can increase H 2 O 2 accumulation in plant tissues, and as a key signaling molecule, H 2 O 2 mediates melatonin-induced lateral root formation, stomatal closure, and tolerance to environmental stresses 12,48,49 . In agreement with these findings, our findings also showed that melatonin increased the H 2 O 2 accumulation in Cl/Cl leaves, while pretreatment with CPA completely abolished the pumpkin-or figleaf gourd rootstock-induced increase in H 2 O 2 under cold stress (Fig. 4d, f). Moreover, DPI pretreatment completely blocked the melatonin-induced cold tolerance of Cl/Cl plants (Fig. 5c). Thus, it is apparent that H 2 O 2 mediates melatonin-induced cold tolerance of Cl/Cm and Cl/Cf plants.
It has been well established that JA and H 2 O 2 function synergistically to regulate multiple physiological processes, such as leaf senescence, drought tolerance, stomatal closure, and wound-induced responses, and JA functions upstream of H 2 O 2 50,51 . Consistent with the data of these studies, our data show that MeJA mediates the melatonin-induced cold tolerance of Cl/Cm and Cl/Cf plants by inducing H 2 O 2 accumulation. This conclusion is based on the following evidence: (1) foliar application of MeJA increased H 2 O 2 accumulation in Cl/Cl plants in response to cold (Fig. 7a), while H 2 O 2 application failed to induce an increase in MeJA accumulation ( Supplementary  Fig. S2); (2) pretreatment with DIECA prevented melatonin-or rootstock-induced H 2 O 2 accumulation (Fig. 7a, b); and (3) application of DPI attenuated the MeJAinduced cold tolerance of Cl/Cl plants (Fig. 7c).
To date, the mechanisms underlying rootstock-induced shoot tolerance to cold stress are unclear; however, we have demonstrated in the present study that, as a potential rootto-shoot signal, melatonin is involved in rootstock-induced cold tolerance. The rootstock-promoted melatonin accumulation in leaves induces MeJA accumulation, which in turn increases melatonin accumulation, forming a selfamplifying feedback loop. MeJA further triggers H 2 O 2 generation and subsequently enhances cold tolerance (Fig. 8). Thus, melatonin-induced increases in MeJA and subsequent H 2 O 2 have an essential role in rootstock-scion communication in response to cold stress.