OsLUX Confers Rice Cold Tolerance as a Positive Regulatory Factor

During the early seedling stage, rice (Oryza sativa L.) must overcome low-temperature stress. While a few cold-tolerance genes have been characterized, further excavation of cold-resistance genes is still needed. In this study, we identified a cold-induced transcription factor—LUX ARRHYTHMO (LUX)—in rice. OsLUX was found to be specifically expressed in leaf blades and upregulated by both cold stress and circadian rhythm. The full-length OsLUX showed autoactivation activity, and the OsLUX protein localized throughout the entire onion cell. Overexpressing OsLUX resulted in increased cold tolerance and reduced ion leakage under cold-stress conditions during the seedling stage. In contrast, the knockout of OsLUX decreased seedling cold tolerance and showed higher ion leakage compared to the wild type. Furthermore, overexpressing OsLUX upregulated the expression levels of oxidative stress-responsive genes, which improved reactive oxygen species (ROS) scavenging ability and enhanced tolerance to chilling stress. Promoter analysis showed that the OsLUX promoter contains two dehydration-responsive element binding (DREB) motifs at positions −510/−505 (GTCGGa) and −162/−170 (cCACCGccc), which indicated that OsDREB1s and OsDREB2s probably regulate OsLUX expression by binding to the motif to respond to cold stress. Thus, OsLUX may act as a downstream gene of the DREB pathway. These results demonstrate that OsLUX serves as a positive regulatory factor of cold stress and that overexpressing OsLUX could be used in rice breeding programs to enhance abiotic stress tolerance.


Introduction
Rice (Oryza sativa L.) is an important staple food worldwide, especially in Asia. However, cold stress is a major abiotic stress that can adversely affect plant growth and development defects, and reduced crop yields [1][2][3][4]. Under low-temperature stress, rice seedlings grow slowly, leaves turn yellow or even wither, and plants grow stunted or even die. To maintain normal development under cold stress, plants have gradually evolved a response mechanism that specifically responds to stress, in which the expression regulation of stress-related genes is a key regulatory mechanism [5][6][7]. Therefore, it is particularly important to further explore cold-tolerant transcription factors to prevent low-temperature damage to rice.
Transcription factors are critical proteins involved in regulating gene expression at the transcriptional level under cold stress [8,9]. The DREB1/CBF (DRE-binding/C-repeat binding factor) protein belongs to the APETALA2 (AP2) family of transcription factors and is considered to be one of the key components of cold acclimatization [10][11][12][13]. The functions of DREB1/CBFs orthologous genes in rice, namely OsDREB1A, OsDREB1B, and OsDREB1C, have been identified to regulate cold tolerance [13][14][15][16][17]. The MYB transcription factor family tolerance in seedlings, while the knockout of OsLUX reduces seedling cold tolerance and increases ion leakage. Furthermore, overexpression of OsLUX improves ROS scavenging ability and enhances tolerance to chilling stress by upregulating the expression levels of OsAPX1, OsAPX2, and OsPOX1. The OsLUX promoter contains two DREB-binding elements motifs at positions −510/−505 (GTCGGa) and −162/−170 (cCACCGccc), which indicates that the OsDREB1s and OsDREB2s probably regulate OsLUX expression. Taken together, our findings demonstrate that OsLUX serves as a positive regulatory factor of cold stress and plays a crucial role in integrating cold-stress responses and circadian rhythms in rice. These results suggest that LUX plays an essential role in regulating the plant's circadian rhythm and integrating it with abiotic stress signaling pathways, which could have significant implications for crop breeding and agricultural productivity in the face of climate change.

Expression Pattern of the OsLUX Gene
To understand the expression pattern of the OsLUX gene, we examined it through reverse-transcription quantitative PCR (RT-qPCR) analysis in different organs of 14-weekold wild-type rice (O. sativa L. japonica. cv. Zhonghua11, ZH11) grown in Yoshida's nutrient solution for hydroponic culture. The results showed that OsLUX expression exhibited a rhythmic pattern consistent with previous studies [33][34][35] (Figure 1A Figure S1). We also investigated the response of OsLUX expression to cold stress and ABA treatment. The RT-qPCR assay revealed that OsLUX expression was induced by cold stress, with transcript levels peaking at 12 h after treatment ( Figure 1B). Furthermore, OsLUX expression was significantly upregulated after one hour of exogenous ABA treatment, but then rapidly decreased ( Figure 1C). Interestingly, the expression pattern of OsLUX under cold stress differed from that under ABA treatment. Additionally, the OsLUX transcript was found to be widely distributed in all organs, with high expression levels particularly in leaf blades ( Figure 1D).

Subcellular Localization of OsLUX Protein
To determine the subcellular localization of OsLUX, we cloned the OsLUX coding sequence in the pA7-GFP vector fused to the N-terminus of the gene-encoding enhanced green fluorescent protein (GFP) under the control of the CaMV 35S promoter. We then transiently expressed the construct expressing the OsLUX-GFP (OsLUX:GFP) fusion protein or the empty vector pA7-GFP (GFP) in isolated onion (Allium cepa L.) epidermal cells by particle bombardment and detected the fluorescence by confocal laser scanning microscopy. In the onion cells expressing OsLUX:GFP, the green fluorescent signals were observed throughout the cell, including the nucleus and the cytoplasm, similar to the green fluorescent signals of GFP alone ( Figure 2).

OsLUX Functions as a Transcription Factor
To test whether OsLUX has transcription activity, OsLUX was fused in-frame to the GAL4 DNA-binding domain (BD) in the pGBKT7 (pBD) vector and the fusion constructs pBD-OsLUX were transformed into the yeast cell (Saccharomyces cerevisiae). As shown in Figure 3A, the transformants containing pBD-OsLUX and the positive control harboring pAD-T + pBD-53 could grow normally on SD/-Trp/-His/-Ade medium exclusively and exhibited the activity of β-galactosidase reporter gene upon the addition of X-α-gal ( Figure 3A). On the other hand, we performed reporter-effector transient expression assays in rice protoplasts. Effector constructs carry OsLUX driven by the CaMV 35S promoter as a detection group or do not carry OsLUX as a control. The reporter plasmid contains a firefly luciferase (Ff-LUC) reporter gene driven by the dehydration-response element (DRE), the GAL4 DNA-binding domain (GAL4DBD), and the minimal CaMV 35S promoter. Renilla luciferase (Rn-LUC) was used to normalize transfection efficiency ( Figure 3B). Comparing the LUC activity with the control group, transfection with the OsLUX showed higher relative LUC activity, confirming that OsLUX is a transcription factor with autoactivations ( Figure 3C).

Subcellular Localization of OsLUX Protein
To determine the subcellular localization of OsLUX, we cloned the OsLUX coding sequence in the pA7-GFP vector fused to the N-terminus of the gene-encoding enhanced green fluorescent protein (GFP) under the control of the CaMV 35S promoter. We then transiently expressed the construct expressing the OsLUX-GFP (OsLUX:GFP) fusion protein or the empty vector pA7-GFP (GFP) in isolated onion (Allium cepa L.) epidermal cells by particle bombardment and detected the fluorescence by confocal laser scanning microscopy. In the onion cells expressing OsLUX:GFP, the green fluorescent signals were observed throughout the cell, including the nucleus and the cytoplasm, similar to the green fluorescent signals of GFP alone ( Figure 2).

OsLUX Functions as a Transcription Factor
To test whether OsLUX has transcription activity, OsLUX was fused in-frame to the GAL4 DNA-binding domain (BD) in the pGBKT7 (pBD) vector and the fusion constructs pBD-OsLUX were transformed into the yeast cell (Saccharomyces cerevisiae). As shown in Figure 3A, the transformants containing pBD-OsLUX and the positive control harboring pAD-T+pBD-53 could grow normally on SD/-Trp/-His/-Ade medium exclusively and exhibited the activity of β-galactosidase reporter gene upon the addition of X-α-gal ( Figure 3A). On the other hand, we performed reporter-effector transient expression assays in rice protoplasts. Effector constructs carry OsLUX driven by the CaMV 35S promoter as a detection group or do not carry OsLUX as a control. The reporter plasmid

Knockout of OsLUX Enhances Tolerance to Cold Stress in Rice
To investigate the biological function of OsLUX, we obtained a T-DNA insertion mutant line of OsLUX in the ZH11 background from the Rice Mutant Database (RMD) in Wuhan, China. We isolated homozygous mutant plants from segregating progeny plants using T-DNA-and gene-specific primers through genomic DNA PCR ( Figure 4A,B). We further tested the transcript level of OsLUX using RT-PCR and RT-qPCR assays, which showed that the relative expression of OsLUX was abolished in homozygous mutant seedlings with the control ( Figure 4C,D). In addition, we constructed an overexpression vector under the control of the CaMV 35S promoter and transformed it into wild-type ZH11, resulting in several transgenic lines ( Figure 4E). We confirmed the transgenic rice lines through RT-qPCR and found that the OsLUX transcript abundance was higher in the OsLUX-overexpressing lines (OE3, OE9, OE10, OE11, and OE13) compared to the wild-type rice. We selected three transgenic lines with higher expression levels (OE10, OE11, and OE13) for the subsequent experiments ( Figure 4F). , driven by three copies of DREs, five copies of the GAL4 and a mini 35S promoter. The effector containing DREB1A on a reporter plasmid. Other effector plasmids containing the GAL4DBD bound the GAL4 sites. Rn-LUC serves as the internal control. (C) Relative LUC activity assay in rice protoplast that had been transfected with both effecter plasmid GOsLUX and reporter plasmids. The effecter plasmid GAL4DBD and reporter plasmid co-transfection used as the control. Data are the means and SD of three replicates, *** p < 0.001, Student t-test.

Knockout of OsLUX Enhances Tolerance to Cold Stress in Rice
To investigate the biological function of OsLUX, we obtained a T-DNA insertion mutant line of OsLUX in the ZH11 background from the Rice Mutant Database (RMD) in Wuhan, China. We isolated homozygous mutant plants from segregating progeny plants using T-DNA-and gene-specific primers through genomic DNA PCR ( Figure 4A,B). We further tested the transcript level of OsLUX using RT-PCR and RT-qPCR assays, which showed that the relative expression of OsLUX was abolished in homozygous mutant seedlings with the control (Figure 4C,D). In addition, we constructed an overexpression vector under the control of the CaMV 35S promoter and transformed it into wild-type ZH11, resulting in several transgenic lines ( Figure 4E). We confirmed the transgenic rice lines through RT-qPCR and found that the OsLUX transcript abundance was higher in the OsLUX-overexpressing lines (OE3, OE9, OE10, OE11, and OE13) compared to the wild-type rice. We selected three transgenic lines with higher expression levels (OE10, OE11, and OE13) for the subsequent experiments ( Figure 4F). Under normal non-stressed conditions in a hydroponic solution, there was no difference observed between transgenic and wild-type plants ( Figure 5A upper lane). However, following exposure to cold stress (4 °C for 5 d) and subsequent recovery under normal conditions for 7 days, OsLUX-overexpressing line seedlings exhibited continuous growth and remained mostly green, whereas the oslux mutant seedlings showed severe leaf rolling and wilting compared to the wild-type ( Figure 5A bottom lane). The survival rate of OsLUX-overexpressing plants was significantly higher than that of wild-type seedlings, while the oslux mutant showed a much lower survival rate than wild-type seedlings ( Figure 5B). We also measured ion leakage and found that it was significantly increased in the oslux mutant after cold treatment, while ion leakage was weaker in OsLUX-overexpressing plants than in WT plants ( Figure 5C). Abiotic stress can cause an increase in ROS, leading to further damage to plants and even cell death. We observed no significant difference in H2O2 content between all lines before low-temperature treatment. However, after cold treatment, the H2O2 content of the oslux mutant was significantly Under normal non-stressed conditions in a hydroponic solution, there was no difference observed between transgenic and wild-type plants ( Figure 5A upper lane). However, following exposure to cold stress (4 • C for 5 d) and subsequent recovery under normal conditions for 7 days, OsLUX-overexpressing line seedlings exhibited continuous growth and remained mostly green, whereas the oslux mutant seedlings showed severe leaf rolling and wilting compared to the wild-type ( Figure 5A bottom lane). The survival rate of OsLUX-overexpressing plants was significantly higher than that of wild-type seedlings, while the oslux mutant showed a much lower survival rate than wild-type seedlings (Figure 5B). We also measured ion leakage and found that it was significantly increased in the oslux mutant after cold treatment, while ion leakage was weaker in OsLUX-overexpressing plants than in WT plants ( Figure 5C). Abiotic stress can cause an increase in ROS, leading to further damage to plants and even cell death. We observed no significant difference in H 2 O 2 content between all lines before low-temperature treatment. However, after cold treatment, the H 2 O 2 content of the oslux mutant was significantly higher than that of WT, whereas the H 2 O 2 content of OsLUX-overexpressing plants was significantly lower than that of WT ( Figure 5D).
higher than that of WT, whereas the H2O2 content of OsLUX-overexpressing plants was significantly lower than that of WT ( Figure 5D).

OsLUX-Overexpressing May Activate ROS-Scavenging Ability
In order to explore the mechanism of OsLUX in regulating plants' ability to scavenge H2O2 in rice, we detected the expression of ROS scavenging-related genes [45][46][47][48]. It was discovered that OsAPX1, OsAPX2, and OsPOX1 were not significantly changed in all plants before cold treatment. However, after cold treatment, the expression of these genes was significantly upregulated in OsLUX-overexpressing lines and generally repressed in oslux mutants ( Figure 6A-C). The DREB-/CBF-dependent transcriptional regulatory pathway plays an essential role in chilling responses in plants. Therefore, the expression levels of the DREB1A and DREB1B genes were detected. It was found that there was no significant difference in the expression levels of the two genes in all lines before or after low-temperature treatment ( Figure 6D,E). The expression of DREB1A and DREB1B was activated after low-temperature treatment, indicating that they may be independent or upstream of the regulation of OsLUX. Nonetheless, OsLUX overexpression improved ROS scavenging ability and enhanced tolerance to chilling stress.

OsLUX-Overexpressing May Activate ROS-Scavenging Ability
In order to explore the mechanism of OsLUX in regulating plants' ability to scavenge H 2 O 2 in rice, we detected the expression of ROS scavenging-related genes [45][46][47][48]. It was discovered that OsAPX1, OsAPX2, and OsPOX1 were not significantly changed in all plants before cold treatment. However, after cold treatment, the expression of these genes was significantly upregulated in OsLUX-overexpressing lines and generally repressed in oslux mutants ( Figure 6A-C). The DREB-/CBF-dependent transcriptional regulatory pathway plays an essential role in chilling responses in plants. Therefore, the expression levels of the DREB1A and DREB1B genes were detected. It was found that there was no significant difference in the expression levels of the two genes in all lines before or after low-temperature treatment ( Figure 6D,E). The expression of DREB1A and DREB1B was activated after low-temperature treatment, indicating that they may be independent or upstream of the regulation of OsLUX. Nonetheless, OsLUX overexpression improved ROS scavenging ability and enhanced tolerance to chilling stress.

OsLUX May Act as a Downstream Gene of the DREB Pathway
We found that there was no significant difference in the expression levels of DREB1A and DREB1B genes. Therefore, we wondered if OsLUX might be a downstream gene of the DREB/CBF pathway that plays a role in the cold-stress response. To investigate this, we analyzed the 1500 base pairs upstream of the OsLUX initiation codon for promoter elements using PlantPAN3.0 [49]. The analysis suggested that the promoter contains two DREB-binding element motifs at positions −510/−505 (GTCGGa) and −162/−170 (cCAC-CGccc). Both OsDREB1s and OsDREB2s could potentially regulate OsLUX expression by binding to the motif at position −510/−505 (GTCGGa) (Figure 7). In addition, OsDREB1H, OsDREB1I, OsDREB2B, and OsDREB2C could also bind to the motif at position −162/−170 (cCACCGccc) to regulate OsLUX expression in response to cold stress. Furthermore, OsLUX may also autoregulate its own expression level through the motif at position −485/−495 (tccGAATCttg) (Figure 7). Taken together, these findings suggest that OsLUX may be both autoregulated and mediated by OsDREB1s and OsDREB2s as a downstream gene of the DREB/CBF pathway in response to cold stress.

OsLUX May Act as a Downstream Gene of the DREB Pathway
We found that there was no significant difference in the expression levels of DREB1A and DREB1B genes. Therefore, we wondered if OsLUX might be a downstream gene of the DREB/CBF pathway that plays a role in the cold-stress response. To investigate this, we analyzed the 1500 base pairs upstream of the OsLUX initiation codon for promoter elements using PlantPAN3.0 [49]. The analysis suggested that the promoter contains two DREB-binding element motifs at positions −510/−505 (GTCGGa) and −162/−170 (cCAC-CGccc). Both OsDREB1s and OsDREB2s could potentially regulate OsLUX expression by binding to the motif at position −510/−505 (GTCGGa) (Figure 7). In addition, OsDREB1H, OsDREB1I, OsDREB2B, and OsDREB2C could also bind to the motif at position −162/−170 (cCACCGccc) to regulate OsLUX expression in response to cold stress. Furthermore, OsLUX may also autoregulate its own expression level through the motif at position −485/−495 (tccGAATCttg) (Figure 7). Taken together, these findings suggest that OsLUX may be both autoregulated and mediated by OsDREB1s and OsDREB2s as a downstream gene of the DREB/CBF pathway in response to cold stress.

Discussion
Previous studies have shown that lots of MYB family members regulate rice cold tolerance [3]. In our study, we identified OsLUX, a cold-induced transcription factor that encodes an MYB family protein. OsLUX expression was highly expressed in leaves and induced by cold treatment (Figure 1B,D). Moreover, transcriptional autoactivation and reporter-effector transient expression assays showed that the OsLUX protein has autoactivation activity, indicating that OsLUX can act as a transcriptional regulator to regulate the expression of downstream target genes in response to cold stress (Figure 3). Furthermore, we obtained knockout and overexpression transgenic lines of OsLUX to verify its role in cold stress (Figure 4). The survival rate of the oslux mutant was significantly lower than that of wild-type seedlings, whereas the OsLUX-overexpressing lines showed much higher survival rates than the wild-type seedlings ( Figure 5A,B). The ion leakage and H2O2 content data showed that the oslux mutant exhibited significantly increased levels after cold-stress treatment, while the ion leakage and H2O2 content of OsLUX-overexpressing plants were lower than the WT plants ( Figure 5C,D). OsLUX overexpression induced the expression of OsAPX1, OsAPX2, and OsPOX1, which improved ROS scavenging ability and enhanced tolerance to chilling stress. Therefore, our results suggest that OsLUX plays a positive regulatory role in cold-stress response in rice.

Discussion
Previous studies have shown that lots of MYB family members regulate rice cold tolerance [3]. In our study, we identified OsLUX, a cold-induced transcription factor that encodes an MYB family protein. OsLUX expression was highly expressed in leaves and induced by cold treatment (Figure 1B,D). Moreover, transcriptional autoactivation and reporter-effector transient expression assays showed that the OsLUX protein has autoactivation activity, indicating that OsLUX can act as a transcriptional regulator to regulate the expression of downstream target genes in response to cold stress ( Figure 3). Furthermore, we obtained knockout and overexpression transgenic lines of OsLUX to verify its role in cold stress (Figure 4). The survival rate of the oslux mutant was significantly lower than that of wild-type seedlings, whereas the OsLUX-overexpressing lines showed much higher survival rates than the wild-type seedlings ( Figure 5A,B). The ion leakage and H 2 O 2 content data showed that the oslux mutant exhibited significantly increased levels after cold-stress treatment, while the ion leakage and H 2 O 2 content of OsLUX-overexpressing plants were lower than the WT plants ( Figure 5C,D). OsLUX overexpression induced the expression of OsAPX1, OsAPX2, and OsPOX1, which improved ROS scavenging ability and enhanced tolerance to chilling stress. Therefore, our results suggest that OsLUX plays a positive regulatory role in cold-stress response in rice.
The DREB1/CBF transcription factor has proved to be one of the key components of cold acclimatization [10][11][12]. However, our results show that there was no significant difference in DREB1A and DREB1B expression levels in OsLUX-overexpressing plants and oslux mutants, before or after low-temperature treatment ( Figure 6D,E). Promoter analysis showed that the OsLUX promoter contains two DREB-binding elements motifs at positions −510/−505 (GTCGGa) and −162/−170 (cCACCGccc), indicating that OsDREB1s and OsDREB2s may regulate OsLUX expression by binding to these motifs to respond to cold stress ( Figure 7). Thus, OsLUX may act as a downstream gene of the DREB pathway. The transcriptional regulation could be further validated using the available DREB transgenic rice RNA-seq database. However, so far, we have not found a usable RNA-seq database to prove our point. In addition, OsLUX may mediate LUX autoregulation through the LBS (LUX binding site) at position −485/−495 (tccGAATCttg) (Figure 7).
Besides its role in regulating the circadian clock, OsLUX has been shown to be involved in abiotic stress tolerance. Previous studies had demonstrated that loss-of-function mutants of OsLUX exhibited lower survival rates under salt stress [35]. In our study, we found that the knockout of OsLUX decreased cold tolerance, while the overexpression of OsLUX increased cold tolerance at the seedling stage compared to wild-type under cold-stress conditions. Therefore, the OsLUX gene in rice plays a critical role in regulating the plant's circadian rhythm and stress tolerance. By regulating the expression of clock-controlled genes and stress-responsive genes, OsLUX allows the plant to synchronize its physiological processes with the day-night cycle and respond to environmental stresses. Further research on OsLUX and its involvement in other physiological processes could provide valuable insights into the molecular mechanisms of plant growth and development, as well as strategies for improving crop productivity and stress tolerance [50][51][52].

Plant Material and Treatments
Rice cultivar Zhonghua11 (ZH11, Oryza sativa L. subsp. japonica) was used in this research article. Seeds were sterilized in 0.3% NaClO and soaked in fresh water at 28 • C for 3 d (with daily water changes). The germinated seeds were sowed on 96-well plastic plates and cultured in Yoshida's nutrition solution in the growth chamber, with a circadian cycle of 12 h light/28 • C and 12 h dark/22 • C at 500-600 µmol m −2 s −1 light intensity and 60% humidity until the three-leaf stage. In the field experiments, ZH11 wild-type rice and OsLUX transgenic rice were grown in natural experiment field under a conventional cultivation environment in the Jiangpu Experiment Station of Nanjing Agricultural University, China.
For the tissue-specific expression profile assay, some three-leaf stage seedlings were transferred to the paddy field until the booting stage, and the rice tissues (including stem, node, leaf blade, leaf sheath, immature panicle, mature panicle, and anther) were harvested at the proper time.
To explore the expression pattern under cold conditions, the three-leaf-stage seedlings were transferred to a growth chamber with a circadian cycle of 12 h light/4 • C and 12 h dark/4 • C. For ABA treatment, seedlings were transferred to Yoshida's nutrition solution containing 100 µM ABA. Samples were collected at a series of time points after each treatment, immediately frozen in liquid nitrogen, and were used for further RNA isolation. For diurnal expression analysis, plants were grown for 14 d in the greenhouse under natural day-length conditions of 12 h light/12 h dark, 28 • C, and entrained for 3 d.
For seedling cold-stress treatment, three-week-old wild-type OsLUX and OsLUXoverexpression seedlings cultured in Yoshida's nutrition solution were transferred into a growth chamber under 4 • C conditions for 5 days, with a circadian cycle of 12 h/12 h (light/dark). The survival rate was estimated after recovering for a week, and plants with green leaves were calculated as survivors.

RNA Extraction, cDNA Preparation, and RT-qPCR
Total RNA was extracted using TRIzol reagent (Invitrogen, Waltham, MA, USA) according to the manufacturer's instructions. Genomic DNA elimination and reverse transcription were performed using the HiScript II Q RT SuperMix for qPCR kit with gDNA wiper (Vazyme, Nanjing, China). Real-time quantitative PCR (RT-qPCR) was performed with the AceQ qPCR SYBR Green Master Mix (Vazyme, China) on the ABI QuantStudio One Real-Time PCR System (Applied Biosystems, Waltham, MA, USA). Reactions contained 10 µL 2× SYBR Green Master Mix reagent, 1 µL of 1:4 diluted reverse-transcription reaction, and 150 nM of each gene-specific primer in a final volume of 20 µL. The following standard thermal profile was used for all PCRs: 50 • C for 2 min; 95 • C for 2 min; 40 cycles of 95 • C for 15 s, 60 • C for 30 s, and 72 • C for 30 s. Relative expression levels were normalized to that of OsActin1 (LOC_Os03g50885) and presented as 2 −∆∆CT . The qPCR primers for the OsLUX, OsAPX1, OsAPX2 and OsPOX1 genes are listed in Table S1.

Subcellular Localization of OsLUX in Onion Epidermal Cells
The coding sequence of OsLUX was cloned into SalI/SpeI sites of the pA7-GFP vector to obtain a transcriptional fusion of OsLUX and GFP under the control of the CaMV 35S promoter. The fusion (CaMV35S:OsLUX-GFP) and control (CaMV35S:GFP) plasmids were delivered into onion epidermal cells by particle-bombardment. After bombardment, the bombarded tissues were incubated on 1/2 Murashige and Skoog agar medium in darkness for 24-36 h. The GFP signal was observed with confocal laser scanning microscopy (Nikon ECLIPSE 80i, Tokyo, Japan).

Trans-Activation Analysis of OsLUX
A yeast two-hybrid assay was carried out by using a Clontech system. ORF fragments of OsLUX were inserted into BamHI/EcoRI sites of the bait vector pGBKT7 (pBD) to generate the pBD-OsLUX construct. According to the protocol provided by the manufacturer, pBD-OsLUX, the positive control (pAD-T + pBD-53), and the negative control (empty pBD) were used to transform the yeast strain AH109, containing HIS3, ADE2, LacZ and MEL1 reporter genes. The transformed strains were streaked onto SD/-Trp or SD/-Trp/-His/-Ade/X-α-gal plates for 3 d.

Genotyping of OsLUX Mutant
The T-DNA insertion of mutant line OsLUX with a ZH11 background was obtained from the Rice Mutant Database (RMD), maintained by the National Center of Plant Gene Research (Wuhan) at Huazhong Agricultural University. The genotyping of homozygote seedlings was identified by PCR with gene-specific primers (FP, RP) and a T-DNA border primer (BP). Homozygous mutant plants were then used in semi-quantitative RT-PCR and real-time quantitative RT-PCR assays to test the transcript levels of the corresponding gene normalized to OsActin1. The primer sequences are listed in Table S1.

Generation of OsLUX Overexpression Plants
To generate overexpression transgenic rice, the coding sequence of OsLUX was amplified and cloned into SacI/KpnI sites linearized by binary expression vector pCAMBIA1300, in which transferred gene expression is under the control of the CaMV 35S promoter. The resulting construct was introduced into Agrobacterium tumefaciens EHA105 by the freeze-and-thaw method, and then transformed into the rice cultivar ZH11 using the Agrobacterium-mediated method. The T1 transgenic plants were selected by Yoshida's nutrition solution containing 30 mg/L hygromycin. Seeds from each T1 plant were individually collected. Selected T2 plants were propagated, and homozygous lines of overexpressing plants were confirmed by RT-qPCR analysis.

Measurement of Ion Leakage
Ion leakage was measured using the top leaf of rice seedlings, which were harvested from each line before and under cold-stress treatment. The samples were incubated in 1 mL of distilled water for 60 min, with 60 rpm shaking. The conductivity was determined using a B-173 conductivity meter (Horiba, Kyoto, Japan). Ion leakage was calculated as the ratio of conductivity values between measurements before and after autoclaving.