SlGAD2 is the target of SlTHM27, positively regulates cold tolerance by mediating anthocyanin biosynthesis in tomato

Abstract Cold stress significantly limits the yield and quality of tomato. Deciphering the key genes related to cold tolerance is important for selecting and breeding superior cold-tolerant varieties. γ-aminobutyric acid (GABA) responds to various types of stress by rapidly accumulating in plant. In this study, glutamic acid decarboxylase (GAD2) was a positive regulator to enhance cold stress tolerance of tomato. Overexpression of SlGAD2 decreased the extent of cytoplasmic membrane damage and increased the endogenous GABA content, antioxidant enzyme activities, and reactive oxygen species (ROS) scavenging capacity in response to cold stress, whereas Slgad2 mutant plants showed the opposite trend. In addition, SlGAD2 induced anthocyanin biosynthesis in response to cold stress by increasing the content of endogenous GABA. Further study revealed that SlGAD2 expression was negatively regulated by the transcription factor SlTHM27. However, the transcript levels of SlTHM27 were repressed under cold stress. Antioxidant enzyme activities, SlGAD2 transcript levels, GABA and anthocyanin contents were significantly increased in Slthm27 mutant plants. Further, our study demonstrated that SlTHM27 decreases SlGAD2-promoted cold resistance in tomato by repressing SlGAD2 transcription. Overall, our results showed that the SlTHM27-SlGAD2 model regulates the cold tolerance in tomato by regulating GABA and anthocyanin.


Introduction
Abiotic stresses such as salt, heat, cold, and drought, are among the major factors contributing to the decline in global crop yields and quality [1,2].Although plants have evolved with the ability to resist environmental stresses, the frequency and intensity of stresses encountered by plants have increased in recent years due to climate change [3,4].Among these stresses, low temperature is an unavoidable environmental factor that limits agricultural productivity [4].Below 12 • C, high levels of oxidative metabolites accumulate in the plant, affecting the protein and DNA structure, damaging the biofilm and plant tissues, and consequently inhibit the plant growth [5,6].Various researches have indicated that plants can scavenge ROS generated by enzymatic antioxidant systems (SOD, POD, CAT, APX, etc.) and non-enzymatic antioxidant systems (ASH, GSH, carotenoids, and f lavonoids, etc.) under cold stress [6,7].Flavonoids are a class of highly biologically active plant secondary metabolites that have surpassed the performance of some common antioxidants [7].As active oxygen scavengers, f lavonoids reduce free radical damage to plant cells under unfavorable conditions by localizing and neutralizing free radicals [8].
Anthocyanins are a class of f lavonoids.They not only impart vibrant colors to nutritive tissues such as f lowers, leaves, and fruits of the plants, but also act as strong antioxidants for ROS scavenging and against microorganisms in defense reactions [9,10].Anthocyanin biosynthesis includes a series of enzymes such as chalcone isomerase (CHI), chalcone synthase (CHS), f lavonoid 3-hydroxylase (F3H), dihydrof lavonol 4-reductase (DFR), and UDPglycosidic f lavonoid transferase (UFGT) [9,11].Several transcription factors (TFs) have also been found to regulate the expression of these anthocyanin-synthesizing genes, such as SlANT1 and SlAN2 of the MYB family, SlGL3 and SlTT8 of the bHLH class, HY5 and BBX20 [12][13][14][15][16].More and more evidence suggested that low temperature induces the expression of anthocyanin synthesizing genes, which in turn boosts the production of anthocyanins, and at the same time, the anthocyanin accumulation can also improve low temperature tolerance of the plants [17,18].Crifò et al. also discovered that low temperatures promoted anthocyanin accumulation in blood oranges [19].MdMYB308L improved the cold stress tolerance of apple through anthocyanin accumulation [20].
Gamma-aminobutyric acid (GABA) acts a key factor in the regulation of plant growth, carbon/nitrogen balance, gene expression, ion homeostasis, and oxidative homeostasis under abiotic stresses [21][22][23].Pretreatment with GABA has increased the cold tolerance of tomato and peach fruits [24].Exogenous GABA significantly upregulated the expression of WRKY75 and MYB13, and improved the tolerance of Agrostis stolonifera L. to drought [25].Liu et al. discovered that GABA is an effective osmotic agent to reduce reactive oxygen species production in tobacco (Nictiana tabacum L.) under water stress [26].In addition, GABA can also alleviate plant damage caused by stresses such as high temperature [22,25], low temperature [27], salt [28], and heavy metals [29] through rapid accumulation.Of course, GABA is also a signaling molecule that activates the phenylalanine pathway and enriches f lavonoids, including anthocyanins [30].
In plants, glutamic acid decarboxylase (GAD) is the ratelimiting enzyme for GABA synthesis by catalyzing the irreversible synthesis of GABA from glutamic acid (Glu) [28].The expression of GAD1 in mulberry leaves is induced by NaCl which promotes the synthesis of GABA, and consequently enhances the salt tolerance of mulberry leaves [31].The increased transcript levels of CiGAD1 and CiGAD2 promoted the accumulation of GABA, and improved the salt stress resistance in mallow [2].Cold stress significantly increased the GABA content in quinoa [32].These studies demonstrated that GAD is the most sensitive gene for GABA synthesis under abiotic stress.Globally, tomato (Solanum lycopersicum L.) is a widely grown economic crop with high nutritional value [33].Because tomato originates from the tropics, the low temperatures negatively affect its growth, yield, and quality [33].Although a lot of studies have been done on how low temperatures alter anthocyanin biosynthesis, the regulation network of GABA content and anthocyanin in cold-stressed tomato remains unclear.
Our study revealed that exogenous GABA (55 mM) significantly improved the low temperature tolerance of tomato.In addition, SlGAD2 was significantly induced by analysing the transcript levels of GABA synthesis-related genes (SlGAD1-5) at low temperature.We also identified SlTHM27, a R2R3 MYB-like TF, responds to cold stress by repressing the expression of SlGAD2.Interestingly, endogenous GABA increased anthocyanin accumulation under cold stress.Taken together, we revealed a novel pathway that is SlTHM27-SlGAD2 to regulate cold stress, which might have potential applications in molecular breeding.

Cold induces GABA accumulation and exogenous GABA enhances cold tolerance in tomato
Due to the lack of knowledge about GABA accumulation in tomato seedlings under cold stress, we measured endogenous GABA levels in tomato seedlings at 4 • C. GABA levels accumulated significantly with the duration of cold treatment and peaked at 48 h (Fig. S1, see online supplementary material).To investigate the role of GABA in cold response, the different concentrations of GABA were applied to wild-type (WT) tomato seedlings.Under normal environmental conditions, there was no significant difference in plant height, fresh weight and dry weight of tomato seedlings by exogenous GABA supply compared to the G0 (0 mM GABA) treatment (Fig. 1a, c-f).However, under low temperature treatment, spraying 55 mM GABA resulted in better seedling status compared to the other concentration treatments (Fig. 1a).Further analysis showed that exogenous spraying of 55 mM GABA (G55) significantly increased cold-stressed tomato seedings height, stem thickness, fresh and dry weight (Fig. 1c-f).Ion leakage ref lects the extent of stress induced damage to plasma membrane [34].Compared with the control, low temperature treatment significantly increased the ion leakage and MDA content in tomato seedling.Exogenous spraying of 55 mM GABA resulted in a significant decrease in ion leakage and MDA levels of the seedlings at low temperature (Fig. 1g and h).At the same time, we found that under low temperature stress, tomato seedlings accumulated a large amount of H 2 O 2 whereas its accumulation in the GABA (55 mM) treated seedlings was significantly reduced (Fig. 1b and i).In conclusion, cold promoted the accumulation of GABA in tomato, while exogenous 55 mM GABA attenuated the cold-induced injury.

SlGAD2 is induced by cold stress in tomato
To reveal the genes involved in cold-induced GABA accumulation, we cloned five GAD genes with 86.28% sequence alignment identity and high homology in the conserved regions (Fig. S2, see online supplementary material).Under normal environmental conditions, the expression of five GAD homologs was analysed in different tissues of WT.There were significant differences in the transcript levels of the five GADs in different tomato tissue.Among them, SlGAD1 was heavily induced in leaves, f lowers and fruits (Fig. 2a); SlGAD5 was more highly expressed in f lowers than in other tissues (Fig. 2e).SlGAD3 and SlGAD4 were significantly induced in leaves and f lowers (Fig. 2a-d).SlGAD2 was highly expressed in all tissues (Fig. 2b).
Four GADs (SlGAD1, SlGAD2, SlGAD3, and SlGAD4) with high transcript levels in tomato leaves were explored in response to cold stress.RT-qPCR results showed that all four selected GADs were induced under low temperature treatment, but compared with the other three SlGADs, only SlGAD2 was most significantly induced and continuously upregulated (Fig. 2f).

SlGAD2 positively regulates anthocyanin synthesis and antioxidant enzyme activities to enhance tomato cold tolerance
To explore the role of SlGAD2 in low temperature tolerance, we obtained two SlGAD2 overexpressing transgenic lines (SlGAD2 OE#4 and SlGAD2 OE#5) in the 'Ailsa Craig' tomato background.It was also confirmed that SlGAD2 expression was significantly increased in both transgenic lines (Fig. S3, see online supplementary material).After 4 days of exposure to low temperature (4 • C), the SlGAD2 OE#4 and SlGAD2 OE#5 exhibited a cold tolerance phenotype as compared to WT (Fig. S4a, see online supplementary material).Meanwhile, SlGAD2 overexpression plants had higher GABA level than WT under normal conditions, whereas low temperature increased GABA accumulation, especially in the SlGAD2 overexpression plants (Fig. S4b, see online supplementary material).Under low temperature treatment, the SlGAD2 overexpressed lines had lower ion leakage, MDA content, H 2 O 2 and O 2 − accumulation as compared to WT (Fig. S4c-f, see online supplementary material).In addition, SlGAD2-overexpressing lines also have higher SOD, POD, and CAT activities than WT (Fig. S4g-i, see online supplementary material), which is consistent with their phenotype of higher cold tolerance.Surprisingly, under low-temperature stress we found pigmentation near the veins in the leaves of SlGAD2 overexpressed plants (Fig. S5a, see online supplementary material).Based on quantitative analysis of anthocyanin levels, anthocyanin levels in the leaves of overexpressed SlGAD2 plants were higher than those of WT under normal conditions and were especially more pronounced under cold stress (Fig. S5b, c, see online supplementary material).The transcript levels of genes involved in anthocyanin synthesis were also analysed.The RT-qPCR results showed that low temperature treatment induced the transcription of SlCHS, SlF3H, SlDFR, and SlUFGT, and it was more so in the SlGAD2 overexpressed plants (Fig. S5d-g, see online supplementary material).In summary, To further verify the relationship between GABA levels and anthocyanin accumulation, we examined the endogenous GABA and anthocyanin levels in WT tomato leaves after exogenous application of GABA (55 mM).The results showed exogenous spraying of GABA significantly increased the anthocyanin content compared with the control, and this difference was more pronounced under cold stress (Fig. S6a, see online supplementary material).In addition, exogenous sprayed GABA significantly increased the anthocyanin content of tomato leaves (Fig. S6b, see online supplementary material).Thus, the accumulation of endogenous GABA helped to promote the increase of anthocyanins in tomato leaves.
To further confirm that the cold-tolerant phenotype of SlGAD2 OE is caused by enhanced SlGAD2 function, we constructed SlGAD2 mutants using CRISPR-Cas9 mediated targeting mutagenesis in the 'AC' background and selected two mutants without the CRISPR-Cas9 transgene for low temperature treatment (Fig. S7, see online supplementary material).Under cold stress, Slgad2 mutant plants exhibited a cold-sensitive phenotype compared to WT (Fig. 3a).Meanwhile, the GABA content of Slgad2 mutant plants was much lower than that of WT both under normal culture conditions and cold treatment (Fig. 3b).Compared with WT, Slgad2 mutant plants had higher ion leakage level, MDA, and H 2 O 2 content and lower SOD, POD, and CAT activities under cold stress (Fig. 3c-i).
Anthocyanin level of Slgad2 mutant plants were reduced compared to the WT.This difference was more pronounced under cold stress (Fig. 3j).Similarly, the transcript levels of SlCHS, SlF3H, SlDFR, and SlUFGT were reduced in Slgad2 mutant plants, especially more significantly under low temperature conditions (Fig. 3k-n).Analysis of these data further supports that SlGAD2 plays a positive role in tomato cold tolerance and anthocyanin accumulation.

SlTHM27 is a transcription factor regulating SlGAD2
To determine the upstream transcription factor of SlGAD2, a tomato cDNA library was screened using Y1H with the SlGAD2 promoter fragment as bait.SlTHM27 (Solyc10g055410.1), a protein of the R2R3 MYB family, has been screened and is a homologue of MdMYB16 with 66.67% sequence similarity (Fig. 4a; Fig. S8, see online supplementary material).SlTHM27 was only present in the nucleus according to subcellular localization results (Fig. 4b).
By searching the PlantTFDB database (http://planttfdb.gao-lab.org/prediction.php),we found that the SlGAD2 promoter sequence was present with the SlTHM27 putative binding elements TTAGGT and TTTGGT motifs (Fig. 4c).Using an electrophoretic mobility shift assay (EMSA), we found that the SlTHM27-MBP fusion protein was able to bind to the TTAGGT element site but not the TTTGGT element in the SlGAD2 promoter region (Fig. 4d; Fig. S9, see online supplementary material).Addition of a competing probe resulted in a decrease in binding strength.Mutation of 'TTAGGT' to 'AAAAA' significantly reduced the binding capacity (Fig. 4d).Thus, SlTHM27 binds directly to the promoter of SlGAD2.
To further characterize the role of SlTHM27 in gene activation, we performed a dual luciferase assay on tobacco leaves.A 1.5 kb SlGAD2 promoter fragment driving a firef ly LUC reporter construct was used to generate an effector construct using SlTHM27 (Fig. 4e).At 25 • C, SlTHM27 significantly repressed the expression of SlGAD2, but this repression was significantly alleviated after low temperature treatment (Fig. 4e).The qRT-PCR results showed that the transcript levels of SlTHM27 were all significantly higher than 0 h under normal conditions, while the transcript levels of SlTHM27 were significantly suppressed under low temperature conditions, especially with the lowest expression level of SlTHM27 at 3 h of low temperature treatment (Fig. 4f).Furthermore, we performed sequence comparison with homologous genes in Arabidopsis, apple and tomato and found that SlTHM27 has an EAR motif at its C-terminal end (Fig. S10, see online supplementary material).The EAR motif is a key feature of EAR-type transcriptional inhibitors [35].

SlTHM27 negatively regulates anthocyanin biosynthesis and antioxidant enzyme activities to reduce tomato cold tolerance
Because SlTHM27 has a high sequence similarity (66.67%) to MdMYB16, which was previously reported to negatively regulate anthocyanin synthesis in apple [35], we constructed the SlTHM27 mutants using CRISPR-Cas9 mediated target mutagenesis (Fig. S11, see online supplementary material).We evaluated anthocyanin accumulation in tomato leaves of Slthm27 mutant plants.Anthocyanin content measurements showed that Slthm27 mutant plants accumulated significantly higher levels of anthocyanins than WT under both normal and low temperature treatments conditions (Fig. 5a).RT-qPCR results showed that SlCHS, SlF3H, SlDFR, and SlUFGT expression levels were higher in Slthm27 mutant plants, and this difference was more significant under cold treatments (Fig. 5b-e).Studies have proven a positive correlation between anthocyanins and cold tolerance in plants [17].We would like to further analyse the role of SlTHM27 under cold stress in tomato.We found no morphological differences between WT and Slthm27 mutant plants (Slthm27 #1 and Slthm27 #2) under normal conditions.However, the Slthm27 mutant plants showed a more cold-tolerant phenotype compared to WT under low temperature (Fig. 5f).Meanwhile, Slthm27 mutant plants showed significantly lower ion leakage and MDA level than control, indicating less damage to membrane lipids compared to the WT under low temperature (Fig. 5h and i).Furthermore, the Slthm27 mutant lines showed lower accumulation of H 2 O 2 , O 2 − and higher SOD, POD, and CAT activities than WT under cold treatment (Fig. 5j-n).Slthm27 mutant plants displayed significantly higher levels of SlGAD2 transcripts and GABA contents under either normal or cold culture conditions (Fig. 5g; Fig. S12, see online supplementary material).These results suggested that silencing SlTHM27 promotes GABA accumulation, reduces ROS levels, and improves cold tolerance in tomato seedlings.In conclusion, the SlTHM27 negatively regulates anthocyanin synthesis and tolerance to low temperature in tomato.

SlTHM27 decreases SlGAD2-promoted cold tolerance in tomato by repressing SlGAD2 transcription
To further verify that SlTHM27 regulates tomato cold tolerance by regulating SlGAD2, we silenced SlGAD2 in Slthm27 mutant plants.The results indicated that under cold stress, there was a significant increase in the transcript level of SlGAD2 in the Slthm27 mutant compared to the WT with pTRV.In the background of the Slthm27 mutant, where SlGAD2 was silenced, the expression of SlGAD2 was significantly suppressed, although it remained higher than the expression observed in the WT background with pTRV-SlGAD2 (Fig. S13, see online supplementary material).We found that compared with pTRV in the WT background, the cold tolerance of pTRV in the Slthm27 mutant background was significantly enhanced, with reduced ion leakage, MDA and H 2 O 2 contents, and significantly increased antioxidant enzyme activities (SOD, POD, and CAT), along with significantly higher GABA and anthocyanin contents (Fig. 6a-j).However, in the Slthm27 mutant background, knockdown of SlGAD2 partially impaired cold tolerance in Slthm27 mutant plants due to increased ion leakage level, MDA and H 2 O 2 levels, and reduced antioxidant enzyme activity (Fig. 6a-i).Furthermore, under low temperature treatment, the GABA content of pTRV-SlGAD2 in the Slthm27 mutant background was more similar to that of pTRV in the WT background (Fig. 6b).It suggests that SlTHM27 regulates SlGAD2 to affect GABA synthesis.Taken together, our results suggest that SlTHM27 negatively regulates cold tolerance in tomato by inhibiting SlGAD2-promoted GABA accumulation and anthocyanin biosynthesis.

Exogenous spraying of 55 mM GABA effectively improved tomato cold tolerance
Tomato, originally from the tropics, is highly sensitive to lowtemperature stress [33].Therefore, elucidating the molecular mechanism of tomato cold sensitivity is important for crop breeding and improvement.GABA is not only a metabolic substance but also a signaling molecule that plays a critical role in mitigating cold injury in various species through accumulation [2,36,37].In this study, tomato seedlings showed typical symptoms of cold injury under cold stress, including slow growth and reduction in fresh and dry weight, which were effectively alleviated by exogenous spraying of 55 mM GABA (Fig. 1).Moreover, we found that exogenous GABA maintained the integrity of cellular structure by improving the activity of antioxidant enzymes to scavenge ROS (Fig. 1).It was further demonstrated that exogenous GABA could regulate the activation capacity of plant antioxidant defense system under cold stress [37,38].Therefore, exogenous spraying of GABA is an effective way to improve the cold tolerance of tomato seedlings during tomato cultivation in facilities.

SlGAD2 is a positive regulator to improve the cold tolerance of tomato
GAD is a key enzyme in GABA production [28].Our results showed that the tissue expression of the five GAD genes varied greatly, with all four GAD genes except SlGAD5 being transcribed at higher levels in leaves (Fig. 2a-e); SlGAD1 and SlGAD2 being more highly expressed in ripe tomato fruits (Fig. 2a and b).Some studies have shown that the expression of SlGAD1-3 was essential for the synthesis of GABA in tomato fruits [39].The only difference was that the present study found lower expression of SlGAD3 in ripe fruits, possibly due to the fact that the mRNA level of SlGAD3 is highest early in fruit development and decreases with fruit ripening [40].In addition, this study investigated the transcript levels of SlGADs in tomato leaves under cold stress for the first time (Fig. 2).Under cold stress, the transcript levels of SlGAD1 and SlGAD3 decreased and then increased, whereas the transcript level of SlGAD4 showed a concurrent trend of increase and then decrease; only the relative expression of SlGAD2 increased significantly with increasing stress time (Fig. 2f).It is noteworthy that the relative expression of SlGAD2 was basically consistent with the dynamic changes of GABA content under cold stress.This implied that SlGAD2 plays a crucial role in change of the GABA content of tomato seedlings under cold stress.

SlGAD2 positively regulates the cold tolerance in tomatoes by scavenging ROS and increasing anthocyanin content
We further investigated the mechanism of action of SlGAD2, given that SlGAD2 was the most sensitive to cold stress.We found that overexpression of SlGAD2 increased the cold tolerance of tomato seedlings by increasing GABA content and antioxidant capacity (Fig. S4, see online supplementary material).On the contrary, Slgad2 mutants showed the opposite trend (Fig. 3).This was consistent with the conclusion that exogenous GABA improves cold tolerance by increasing antioxidant capacity in banana fruits [38].However, it is not clear whether GABA improves cold resistance in tomato through other pathways.In this study, we found that anthocyanin content was significantly accumulated in SlGAD2overexpressing plants (Fig. S5, see online supplementary material).Anthocyanins are a major class of f lavonoids whose synthesis and accumulation are induced by low temperature [17,18].The anthocyanin content and the expression of anthocyaninrelated genes were significantly increased in SlGAD2 overexpressing plants under low-temperature stress, whereas the anthocyanin content was significantly suppressed in Slgad2 mutant plants (Fig. 3j-n).Anthocyanins can effectively scavenge excessive ROS to maintain normal cellular redox homeostasis under abiotic stress [10].Therefore, anthocyanins accumulated in the overexpressed SlGAD2 transgenic tomato attenuated low temperature induced oxidative damage.At the same time, we found that spraying 55 mM GABA increased the content of endogenous GABA, which also promoted the anthocyanin accumulation and improved cold tolerance in tomato (Fig. S6, see online supplementary material).These results further demonstrated that GABA accumulation could efficiently scavenge ROS through both enzymatic and non-enzymatic antioxidant systems under cold stress, and SlGAD2 played an essential role in this process.In addition, it will be interesting to study how GABA triggers the expression of anthocyanin synthesis genes.

SlTHM27 negatively regulated cold by inhibiting SlGAD2 transcription
Most of the MYB class TFs can respond positively to plant tolerance to abiotic stress [41], only a few MYB class TFs were negative regulators of abiotic stress response [42,43].For example, VcMYB4a, which has an EAR repressor domain of structure, was down-regulated by low-temperature treatment, and blueberry healing tissues exhibited a cold-sensitive phenotype after VcMYB4a overexpression [44].Similarly, we found that the SlTHM27 protein sequence contained a C-terminal EAR repressor motif (Fig. S10, see online supplementary material), and its mutant plants exhibited enhanced cold tolerance (Fig. 5).In addition, MYB class transcriptional activators and repressors have  been widely explored in the regulation of anthocyanin biosynthesis [35,45].AtMYB4, which contains an EAR motif, represses C4H expression [46].MdMYB16 inhibits anthocyanin synthesis in apple healing tissues by repressing MdANS and MdUFGT expression [35].In this study, we found that SlTHM27 is an MdMYB16 homolog with 66.67% protein sequence similarity.SlTHM27 was found to repress SlGAD2 transcription by qRT-PCR and Dual-LUC assays (Fig. 4e; Fig. S12, see online supplementary material).Furthermore, our results convincingly demonstrated that SlGAD2 acts downstream of SlTHM27 (Figs 4 and 6).Thus, the enhanced cold tolerance of Slthm27 mutant plants depends on the increased transcript level of SlGAD2, which in turn promotes GABA accumulation and improves anthocyanin content and ROS scavenging (Figs 5 and 6).Taken together, our work has revealed for the first time the mechanism of the SlTHM27-SlGAD2 regulatory module responds to cold stress by regulating GABA levels (Fig. 7).Cold stress inhibited the mRNA level of the negative regulator SlTHM27 to weaken the transcriptional repression of SlGAD2 and induced the synthesis of GABA to improve tomato resistance through enzymatic and non-enzymatic antioxidant systems.Our study provides valuable insights for improving cold tolerance in tomato.

Vector construction and genetic transformation
The CDS amplicon of SlGAD2 gene (ID: Solyc11g011920.1)was inserted into pHellsgate2 vector, and the fusion-expressing vector was introduced into the tomato 'Ailsa Craig' (WT) to achieve genetic transformation.The SlGAD2 overexpression plants were produced and identified by PCR.The primers used in this research are listed in Table S1 (see online supplementary material).
To edit the SlTHM27 and SlGAD2 genes using the CRISPR/ Cas9 system.Targets were designed and selected using Cas-Designer (http://www.rgenome.net/cas-designer/).The guide RNA sequence was constructed and inserted into pBSE402 [47].Tomato transformation was performed according to previously described methods.Positively transformed plants were identified by extracting genomic DNA from stable transgenic lines, cloning potential editing fragments of SlTHM27 and SlGAD2, respectively, and sequencing them.
For gene silencing, primers listed in Table S2 (see online supplementary material) were used to amplify cDNA fragment of SlGAD2, and the PCR amplification product were then transformed into the TRV2 vector.The fusion expression vector was transfected into wild-type tomato cotyledons to generate SlGAD2 gene-silencing plants.For details of the method, please refer to our group's previous study [48].

Plant materials and treatment
We germinated both WT and transgenic tomato seeds at 28 • C, after which they were planted separately in cavity trays and cultured in a light incubator under culture conditions: 25 • C/16 • C (day/night).Seedlings were transplanted to nurseries with four fully expanded true leaves, and continued to be grown until the fifth true leaf was fully expanded under environmental conditions as described previously.For the cold treatment, tomato seedlings were incubated in a cold chamber at 4 • C for 4 days.

RT-qPCR analysis
Total RNA was extracted with tomato leaves, and reverse transcription did with PrimeScriptTM RT Kit (Takara Bio, Shiga, Japan).RT-qPCR analysis was performed with ChamQ SYBR qPCR Master Mix (Vazyme, Nanjing, China).SlACTIN was employed as the parameter.

Bioinformatics analysis
DNAMAN was used for multiple amino acid comparisons of protein sequences, and the SMART program was used to analyse conserved protein structural domains [49].The online tool PlantTFDB (http://planttfdb.gao-lab.org/prediction.php)was used to analyse the TF binding sites in the SlGAD2 promoter.

SlTHM27-GFP subcellular localization
The CDS sequences of SlTHM27 was inserted into pAC402-GFP.The fusion proteins were transferred to one-month-old tobacco leaves.GFP signal was detected by laser scanning confocal microscopy (TCS-SP8 SR; Leica, Wetzlar, Germany).NLS-mCherry was used to as the nucleus marker.

Yeast one-hybrid (Y1H) assay
The SlGAD2 promoter was introduced into the pAbAi vector and then digested with BbsI (NEB, Ipswich, MA, USA) as a bait.The CDS sequences of SlTHM27 was cloned into pGADT7 as a prey vector.Y1H experiments were performed according to the Matchmaker Gold Y1H manufacturer's instructions.Table S1 (see online supplementary material) lists the primers used for amplification.

Dual-luciferase reporter assay system
For the LUC assay, the fusion reporter gene (proSlGAD2) and effector (SlTHM27) plasmids were inserted into Agrobacterium tumefaciens GV3101, respectively.One-month-old tobacco leaves were transiently transformed as described previously [50].A dual luciferase reporter assay system (Promega) was used to detect luciferase activity.In low temperature treatment, tobacco plants were subjected to treatment at 4 • C for 3 h and then proteins were extracted.Ten independent biological samples were used.

EMSA
The truncated SlTHM27 was cloned into pMAL-c5X.The resulting plasmid was converted into Escherichia coli strain BL21 (DE3) and amplified for 8 h at 28 • C. MBP-SlTHM27 fusion protein was purified using straight-chain starch resin (NEB, E8201S, USA).The EMSA assays were carried out with the Light Shift Chemiluminescent EMSA Kit (ThermoFisher Scientific).Table S1 (see online supplementary material) lists the primers used for amplification.

Determination of GABA content
GABA was extracted using 0.1 g of tomato leaves, and GABA content was measured by LC-MS [51].GABA content was analysed using three independent experiments each having three replicates.

Determination of total anthocyanin content
Total anthocyanins were extracted from tomato leaves using a solution of methanol and HCl (0.1%, v/v) at 4 • C overnight.Absorbance was detected at 530, 657 nm using a UV-Vis spectrophotometer (Shimadzu UV-1780) [10].

H 2 O 2 , ion leakage, enzymes activities, and MDA content measurements
H 2 O 2 were determined as previously described by Xie et al. [50].Ion leakage was measured according to the method of Jiang et al. [52].SOD, POD, CAT activities, and MDA levels were calculated as previously described [29].

DAB and NBT staining of the tomato leaves
Tomato leaves were analysed for H 2 O 2 by placing them in a 1 mg mL −1 solution of 3,3 -diaminobenzidine (DAB) (pH 3.8) in the light for 8 h.For O 2 − analysis, the tomato leaves were immersed in a 0.5 mg mL −1 solution of nitroblue tetrazolium (NBT) in the dark for 8 h.

Determination of growth
The height of tomato seedlings was measured in cm using a meterstick and stem thickness in mm using a Vernier caliper.Fresh and dry weights of tomato seedlings were measured using a balance with accuracy to one thousandth of a millimeter.

Statistical analysis
Data are presented as mean ± SD.Statistical analysis was performed with SPSS 23.0.One-way ANOVA was used to consider differences significant at P < 0.05 or P < 0.01 (Tukey's test).

4 Figure 1 .
Figure 1.Exogenous γ-aminobutyric acid (GABA) on the cold tolerance of tomato seedlings.(a) Phenotypic changes in cold tolerance of tomato seedlings treated with different concentrations of GABA.Bar, 2.5 cm.G0, G40, G55, G60, and G70 mean exogenous sprays of 0 mM, 40 mM, 55 mM, 60 mM, and 70 mM of GABA, respectively.(b) Nitroblue tetrazolium (NBT) and diaminobenzidine (DAB) staining of leaves with water or GABA-treated plants under control and cold treatment (4 • C) for 4 days.(c-f) Changes in plant height (c), stem thickness (d), fresh (e), and dry weight (f) of water or GABA-treated plants under control and cold treatment (4 • C) for 4 days.(g-i) ion leakage (g), MDA (h), and H 2 O 2 (i) of water or GABA-treated plants measured before or after cold treatment.Values represent the average of six (c-f) or four (g-i) independent measurements, and error bars represent standard errors.Different letters of the columns indicate significant differences (P < 0.05).

Figure 2 .
Figure 2. SlGAD2 was significantly induced by cold stress.(a-e) Changes in transcript levels of the GABA synthesis-related genes SlGAD1 (a), SlGAD2 (b), SlGAD3 (c), SlGAD4 (d), and SlGAD5 (e) in roots, stems, leaves, f lowers, and ripe fruits of tomato.(f) Changes in relative expression of SlGAD1-5 in leaves of tomato under cold treatment.0 h in all genes be set to 1. Different letters of the columns indicate significant differences (P < 0.05).

4 Figure 3 .
Figure 3. Slgad2 mutant plants are more sensitive to cold stress.(a) Phenotypic changes in WT and Slgad2 mutant plants treated at low temperature (4 • C) for 4 days.Bar, 2.5 cm.(b) Endogenous GABA content tomato leaves shown in (a).(c-e) Electrolyte leakage (c), MDA content (d) and H 2 O 2 content (e) of Slgad2 mutant plants under control and cold stress (4 • C).(f) NBT and DAB staining of leaves from WT and Slgad2 mutant lines under control and cold stress (4 • C). (g-i) SOD activity (g), POD activity (h) and CAT activity (i) in Slgad2 mutant plants under control and cold stress.Plants were exposed to cold stress for 0 and 4 days, and leaves were collected for the measurements.(j) Anthocyanin content of Slgad2 mutant plants under control and cold stress (4 • C). (k-n) Relative expression levels of SlCHS, SlF3H, SlDFR, and SlUFGT in Slgad2 mutant plants under cold treatment for 4 days.SlACTIN was used as an internal control.Different letters of the columns indicate significant differences (P < 0.05).

Figure 4 .
Figure 4. Interaction of SlTHM27 with SlGAD2 promoter.(a) An interaction between SlTHM27 and promoters of SlGAD2 by Y1H assays.Empty-AD as a control.(b) Subcellular localization of SlTHM27-GFP and NLS-mCherry in tobacco (Nicotiana benthamiana) leaves.Bar, 50 μM.(c) Schematic representation of the SlTHM27 binding element position on the SlGAD2 promoter predicted by PlantTFDB.Black triangles represents the predicted positions of the binding elements.p1 represents the 'TTAGGT' binding element, p2 and p3 represent the 'TTTGGT' binding element.(d) SlTHM27-MBP was able to bind to the p1 site of the promoter in SlGAD2 by EMSA analysis.(e) The inhibition of SlGAD2 transcription by SlTHM27 was verified by assaying relative LUC/REN activity.The control used was the empty vector pGreen II 62-SK (EV).In cold treatment, infested tobacco was grown under normal conditions for 69 h and then in a growth chamber at 4 • C for another 3 h, after which the leaves were collected.(f) SlTHM27 relative expression changes before and after cold treatment.Different letters of the columns indicate significant differences (P < 0.05).

4 Figure 5 .
Figure 5. SlTHM27 negatively regulates cold stress in tomato.(a) Anthocyanin content of Slthm27 mutant plants under control and cold stress (4 • C). (b-e) Relative expression levels of SlCHS, SlF3H, SlDFR, and SlUFGT under cold treatment in Slthm27 mutant plants for 4 days.(f) Phenotypic changes in WT and Slthm27 mutant plants treated at low temperature (4 • C) for 4 days.Bar, 2.5 cm.(g) Endogenous GABA content tomato leaves shown in (f).(h-j) Electrolyte leakage (h), MDA content (i) and H 2 O 2 content (j) of Slthm27 mutant plants under control and cold stress (4 • C).Plants were exposed to cold stress for 4 days, and leaves were collected for an ion leakage assay, MDA and H 2 O 2 content measurement.(k) NBT and DAB staining of leaves from WT and SlTHM27-silenced lines under control and cold stress (4 • C). (l-n) SOD activity (l), POD activity (m), and CAT activity (n) in Slthm27 mutant plants under control and cold stress.Plants were exposed to cold stress for 0 and 4 days, and leaves were collected for the measurements.Different letters of the columns indicate significant differences (P < 0.05).

4 Figure 6 .
Figure 6.SlTHM27 negatively regulates cold tolerance in tomato by repressing SlGAD2 transcription and GABA accumulation.(a) Phenotypic changes of silenced (pTRV-SlGAD2) or non-silenced SlGAD2 (pTRV) in WT and Slthm27 mutant plants before and after treatment at low temperature (4 • C) for 4 days.Bar, 2.5 cm.(b) Endogenous GABA content tomato leaves shown in (a).(c-e) Electrolyte leakage (c), MDA content (d), and H 2 O 2 content (e) of silenced (pTRV-SlGAD2) or non-silenced SlGAD2 (pTRV) in WT and Slthm27 mutant plants under control and cold stress (4 • C).Plants were exposed to cold stress for 4 days, and leaves were collected for an ion leakage assay, MDA and H 2 O 2 content measurement.(f) NBT and DAB staining of leaves of silenced (pTRV -SlGAD2) or non-silenced SlGAD2 (pTRV) in WT and Slthm27 mutant plants under control and cold stress (4 • C). (g-i) SOD activity (g), POD activity (h), and CAT activity (i) of silenced (pTRV -SlGAD2) or non-silenced SlGAD2 (pTRV) in WT and Slthm27 mutant plants under control and cold stress.Plants were exposed to cold stress for 0 and 4 d and leaves were collected for the measurements.(j) Anthocyanin content of silenced (pTRV-SlGAD2) or non-silenced SlGAD2 (pTRV) in WT and Slthm27 mutant plants under control and cold stress (4 • C).The different letters of the columns indicate significant differences (P < 0.05).

Figure 7 .
Figure 7.A working model for SlTHM27-SlGAD2 in response to cold stress.Under low-temperature stress, the transcript level of SlTHM27 was repressed, thereby increasing the expression level of SlGAD2.SlGAD2 promoted the accumulation of GABA, which increased antioxidant enzyme activities and anthocyanin levels.This helped to scavenge excess reactive oxygen species (ROS) and improved cold tolerance in tomato.