OXIDATIVE STRESS 3 regulates drought-induced flowering through APETALA 1

Biochemical and Biophysical Research Communications 519 (2019) 585e590

Contents lists available at ScienceDirect

Biochemical and Biophysical Research Communications journal homepage: www.elsevier.com/locate/ybbrc

OXIDATIVE STRESS 3 regulates drought-induced flowering through APETALA 1 Minting Liang a, Shimin Xiao a, b, Jiajia Cai a, b, David W. Ow a, * a b

Plant Gene Engineering Center, South China Botanical Garden, Chinese Academy of Sciences, Guangzhou, China University of Chinese Academy of Sciences, Beijing, China

a r t i c l e i n f o

a b s t r a c t

Article history: Received 27 August 2019 Accepted 29 August 2019 Available online 17 September 2019

Stress-induced regulation of flowering time insures evolutionary fitness. Stress-induced late flowering is thought to result from a plant evoking tolerance mechanism to wait out the stress before initiating reproduction. Stress-induced early flowering, on the other hand, is thought to be a stress-escape response. By shortening their life cycle to produce seeds before severe stress leads to death, this insures survival of the species at the cost of lower seed yield. Previously, we reported that overexpression of OXS3 (OXIDATIVE STRESS 3) could enhance tolerance to cadmium and oxidizing agents in Arabidopsis whereas an oxs3 null mutant was slightly more sensitive to these chemicals. In this study, we found that the absence of OXS3 also causes early flowering under a mild drought stress treatment. This contrasts with the behavior of wild type Ws4 and Col ecotypes that responded to the same condition by delaying flowering time. We tested the hypothesis that OXS3 might ordinarily exert a negative regulatory role on flowering during drought stress, which in its absence, would lead to stress-induced early flowering. In a search of whether OXS3 could interfere with regulators that activate flowering, we found that OXS3 could bind SOC1 in vitro and in vivo. Overexpression of OXS3 in a transient expression assay was found to repress the AP1 promoter, and the full repression effect required SOC1. It is possible that the OXS3/SOC1 interaction serves to prevent precocious flower development and prevent low seed set from a premature stress-induced flowering response. © 2019 Elsevier Inc. All rights reserved.

Keywords: OXS3 Stress response Flowering regulation SOC1 AP1

1. Introduction Flowering is a milestone event for plant adaptive evolution on land. It enables more species to survive in larger and wider biotope with life cycles ranging from weeks to hundreds of years [1,2]. The plasticity of flowering offers strong evidence that plants integrate information from both changing environmental stimuli and endogenous development to time floral transition and longevity. The information perceived is transformed into molecular signals and centralized to the dynamic regulatory network to determine the reproductive strategy by the plant [3,4]. Drought tolerance is an important agronomic trait and different plants may respond by flowering early or late depending on the intensity and duration of the drought [5e8]. Late flowering is

* Corresponding author. Plant Gene Engineering Center, South China Botanical Garden, Chinese Academy of Sciences, Xingke Road 723, 510650, Guangzhou, China.: E-mail address: [email protected] (D.W. Ow). https://doi.org/10.1016/j.bbrc.2019.08.154 0006-291X/© 2019 Elsevier Inc. All rights reserved.

thought to result from a plant evoking tolerance mechanism to wait out the stress before initiating reproduction, while early flowering is thought to be a stress-escape response. By shortening their life cycle to produce seeds before severe stress leads to death, this process triggers earlier up-regulation of flowering genes such as GI (GIGANTEA), TSF (TWIN SISTER OF FT), FT (FLOWERING LOCUS T) and SOC1 (SUPPRESSOR OF OVEREXPRESSION OF CONSTANS 1) [9,10]. The regulation of flowering has been well studies in Arabidopsis. SOC1 is a MADS-box transcription factor regulated by several pathways. The SOC1 promoter is repressed by FLC (FLOWERING LOCUS C) of the vernalization and autonomous pathway [11e13], but is activated by CO (CONSTANS) along with FT during long days. SOC1 is also regulated by the age-dependent and gibberellic acid pathways, in which SPL (SQUAMOSA PROMOTER BINDING LIKE) and microRNA156 are involved [14,15]. During severe stress, SOC1 transcription can also be activated by OXS2 (OXIDATIVE STRESS 2) for stress-escape early flowering [9]. In Arabidopsis, floral meristem identity genes AP1 and LFY are required for the transition of an inflorescence meristem into a floral meristem [16]. SOC1 produced at the shoot apex interacts with AGL24 (AGAMOUS-LIKE24) and

586

M. Liang et al. / Biochemical and Biophysical Research Communications 519 (2019) 585e590

activates the floral meristem identity genes LFY and AP1 [17e20]. Arabidopsis OXS3 (OXIDATIVE STRESS 3) and some family members were previously described to be involved in tolerance to heavy metals and oxidative stress [21] because expression of OXS3 or OXS3-Like (O3L) cDNAs enhanced tolerance to various heavy metals and oxidizing chemicals in the fission yeast Schizosaccharomyces pombe. Overexpressing OXS3 in Arabidopsis also conferred slightly higher tolerance to cadmium and diamide, while an oxs3 null mutant was slightly more sensitive to tert-Butyl hydroperoxide. Since neither effects were dramatic, it was hypothesized that there is gene function redundancy of family members. Aside from finding that OXS3 co-localized with histone H4 to subnuclear speckles, how OXS3 functions in plant stress was unknown. Here, we report that while wild type Arabidopsis responded to a mild drought treatment with a late flowering phenotype, a null mutation in OXS3 showed early flowering reminiscent of a stress escape response. This led us to find that OXS3 could bind SOC1 in the nucleus and repress the activation of the AP1 promoter. Since AP1 controls sepal and petal development and has been used as a marker for floral transition [22e24], the repression of AP1 gene expression could mean that OXS3 can attenuate flowering. Since its absence leads to a more robust stress-escape type of early flowering, we hypothesize that OXS3 may act as a monitor to insure that mild stress-induced flowering do not lead to precocious flowering that would sacrifice seed yield. 2. Materials and methods 2.1. Plant materials and growth conditions Col (Col-0, cv. Columbia), Ws4 (Wassilewskija 4), Ws4(oxs3) [FLAG_076G08], Col(o3l4) [SALK_053249C], Col(o3l6) [CS858490] and Col(elf12) [CS877578] were from Arabidopsis Biological Centre for Col and Versailles Genetics and Plant Breeding Laboratory for Ws4. The soc1-2 null mutant was from Xingliang Hou (South China Botanical Garden, Chinese Academy of Sciences, Guangzhou, China). Combination mutants were generated by crossing single mutants, resulting in Col(o3l4;o3l6), Ws4(oxs3);Col(o3l4), Ws4(oxs3);Col(o3l6), Ws4(oxs3);Col(o3l4;o3l6), Ws4(oxs3);Col(o3l4;erf12), and Ws4(oxs3);Col(o3l6;erf12). Plants were grown in growth room under long day length of 16 h light/8 h dark at 22
C. Flowering time was measured as number of rosette leaf by bolting when the inflorescence reaches 1 cm, while size of rosette leaves set was quantified via ImageJ on the scaled photos. 2.2. Drought treatment Modified from Riboni et al. [10], plants were germinated on the agar plate with half strength of MS (Murashige and Skoog) medium until showing 2 fully open cotyledon leaves. Then they were transferred to a 10  10 cm square pots filled with a blend (1:1, v/v) of soil and vermiculite. The soil water capacity was calculated as follows: pots were filled with the blend and air dried for 48 h in an oven at 65
C to determine dry weight. Pots were subsequently soaked in water and measured for wet weight (100%). The relative soil water content (RSWC) reaching and maintaining ca. 30% were considered as drought condition. Each genotype was assayed in two parallel experiments: normal watering (80% ~ 90% RSWC) and drought (30% RSWC) conditions. RSWC was kept constant by application of water according to the pot weight every day. 2.3. Subcellular localization and BiFC Through use of the in-fusion cloning technique (Clontech), the GFP-encoding gene was fused to the C-terminus of OXS3 coding

region with expression of the gene fusion driven by the CaMV 35S RNA promoter. Separately, OXS3 and SOC1 were cloned into the pBiFC vectors [25] or pCambia3300 vector (http://www.cambia. org) and purified by the Genopure Plasmid Midi Kit (Roche). Protoplast isolation from 2- to 3-week old Arabidopsis Col WT leaves from long day grown plants and protoplast transformation with purified plasmids were performed essentially as described by Yoo et al. [26]. After incubating for 16 h at 22
C in the dark, protoplasts were examined under a fluorescent microscope (Leica DMI6000B). 2.4. Yeast two-hybrid assay OXS3 and 10 flowering regulator genes (Fig. 3) were fused to Cterminal or N-terminal halves of ubiquitin in bait vector pDHB1 or prey vector pPR3. TFL1 and FT were placed in pPR3 to prevent autoactivation. After transformation into yeast strain NMY51 by the DUAL hunter system (Dualsystems Biotech, P02305), the yeast was grown on vector selective SD media without Leu and Trp (SD-2DO) and incubated for 3 days at 28
C. Each of 3 independent clones from each plate was re-suspended in 20 mL ddH2O and 3 mL of suspension spotted on interaction selective medium without Leu, Trp, His and Ade (SD-4DO) for 3 days at 28
C. 2.5. Transactivation assay The cDNA of OXS3 was inserted into pCambia3300 (http://www. cambia.org) to yield pOE-OXS3, in which OXS3 is expressed from the CaMV 35S RNA promoter. The firefly luciferase and Renilla luciferase coding regions were PCR cloned into pCambia3300 to yield pFLuc, and pRLuc, respectively. Promoters of CO, SOC1, LFY and AP1 were PCR amplified from Arabidopsis genomic DNA as 2k bpfragments upstream from the transcription start site and inserted into pFLuc to yield pCO:FLuc, pSOC1:FLuc, pLFY:FLuc and pAP1:FLuc, respectively. Each plasmid was introduced into A. tumefaciens GV3101, except that pRluc was introduce into A. tumefaciens EHA105 as internal control. A. tumefaciens colonies carrying corresponding vectors were inoculated into 4 ml liquid LB (LuriaeBertani) medium containing 10 mg rifampin/L, 50 mg kanamycin/L and grown at 28
C for 24 h on a rotary shaker at 220 rpm. 0.2 ml cultures were used to inoculate 6 ml fresh LB medium containing 50 mg kanamycin/L and 20 mM acetosyringone. After cultivation at 28
C for 8 h, cells were collected by centrifugation at 12,000g for 2 min, re-suspended in infiltration buffer (10 mM MES, 10 mM MgSO4, 200 mM acetosyringone) and adjusted to an OD600 z 0.8 with the same buffer. Cell suspensions were then incubated at room temperature for 3 h without shaking. For Fig. 4A, tobacco plants (Nicotiana benthamiana) were grown in soil in a growth room at 26
C with a 14-h light/10 h dark photoperiod. Using a 1-ml syringe without needle, infiltration by Agrobacterium was carried out with healthy tobacco leaves of 5e6 week-old plants. Infiltration medium contained 1 ml of transcription activator strain with pOE-OXS3, 0.1 ml of strain containing pCO:FLuc, pSOC1:FLuc, pLFY:FLuc or pAP1:FLuc, and 5 ml of pRLuc strain for reference. After infiltration, plants were placed in a dark chamber with high humidity overnight, and then returned to the growth room at 26
C for another 2 days. Luciferase values were measured by the Dual-Luciferase® Reporter Assay System (Promega, E1910). For Fig. 4B, the AP1 promoter was fused to the firefly luciferase coding region, and subsequently inserted into vector pRluc to yield pAP1:Fluc-Rluc, in which the pAP1-Fluc gene fragment is the reporter while the Rluc fragment serves as the internal control. In this way, protoplast transformation used only 2 plasmids (transcription factor, reporter-control) instead of 3 (transcription factor, reporter and control). Protoplasts were isolated from 2 to 3-week old

M. Liang et al. / Biochemical and Biophysical Research Communications 519 (2019) 585e590

587

Fig. 1. Flowering time under long day growth. (A, B) Flowering time (Average ± SD) of mutant versus WT plants under drought. Data from an experiment with n number of plants indicated. Different letters indicate significant differences among plant groups (p < 0.05, Mann Whitney test). Results from (A) and (B) conducted in different experiments. (C) Representative photo of Ws4(oxs3) grown under normal watering condition or drought after bolting. Size of rosette leaves set from oxs3 plants grown in long days under well watered or drought conditions.

Fig. 2. Subcellular localization of OXS3 and interaction with SOC1. (A) Representative fluorescence microscopy photos of intracellular allocation of signals from Arabidopsis protoplasts showing OXS3-GFP, or interaction between nYFP-OXS3 and cYFP-SOC1 via BiFC assay after 16 h of recovery in the dark post PEG-mediated transformation. (B) Graphical representation of percentage of cells with fluorescence in nuclear-only, or nuclear and cytoplasmic. Mean and SD of distribution of cell types from at least three reproducible experiments. n ¼ total protoplast cells counted in all experiments.

588

M. Liang et al. / Biochemical and Biophysical Research Communications 519 (2019) 585e590

Fig. 3. Interaction of OXS3 and flowering regulation partners in vitro. (A) Flowering regulatory network in Arabidopsis [28]. Pathways that control plant development (autonomous, hormones, aging, sugars) and the response to environmental changes (ambient temperature, circadian clock, photoperiod, vernalization) converge on the floral integrator genes SOC1 and FT, which then activate floral meristem identity genes to initiate floral transition. Arrows indicate positive regulation. (B) Spilt-ubiquitin based yeast two-hybrid assays for OXS3 and 10 selected flowering regulators. EV, empty vector negative control. Genes in red were cloned into bait vector, and genes in black were cloned into prey vector. SD-2DO, SD medium without Leu and Trp; SD-4DO, SD medium without Leu, Trp, His and Ade. Representative results from 3 independent colonies from each co-transformation plate. (For interpretation of the references to colour in this figure legend, the reader is referred to the Web version of this article.)

Arabidopsis leaves of long-day grown plants and transformed with 15 mg of each plasmid as described by Yoo et al. [26]. After incubating for 16 h at 22
C in the dark, luciferase activities were measured by the Dual-Luciferase® Reporter Assay System (Promega, E1910). 3. Results & discussion 3.1. Precocious drought-mediated flowering from loss of OXS3 function OXS3 belongs to a 7-member gene family that shares a putative N-acetyltransferase domain [21]. Our previous study showed that a null oxs3 mutant was only mildly sensitive to stress, and this has led us to hypothesize redundant function among the 7 members of the OXS3 gene family. In an attempt to uncover redundant function among family members, we generated combination mutants to test under stress conditions. However, the oxs3 null mutant was only available in the Ws4 background while other null mutants were from the Col ecotype. Hence, both Ws4 and Col wild type (WT) controls were used for reference. We tested a drought treatment in which we observed late flowering in WT plants. Hence, we considered this treatment condition to be mild rather than a severe treatment in which we would expect stress-escape early flowering. As shown in Fig. 1A, the mild stress condition delayed Ws4 flowering from the 7 leaf to the 11 leaf stage, and Col flowering from the 9 leaf to the 11 leaf stage. This late flowering is likely due to a tolerance response to wait out the stress before initiating reproduction. However, in three triple mutants Ws4(oxs3);Col(o3l4;o3l6), Ws4(oxs3);Col(o3l4;erf12) and Ws4(oxs3);Col(o3l6;erf12), they not only lacked the delayed flowering response, but also showed early flowering (5-leaf stage; Fig. 1A) that is reminiscent of a stress-escape response. Since prior studies on erf12 which encodes ERF12 (ethylene response factor 12) had not reported early flowering [27], we suspected that this early flowering was likely due to the absence of OXS3, O3L4 and/or O3L6. To narrow down which of these proteins were involved, we tested single and double mutant combinations (Fig. 1B). The Col(o3l4) and

Col(o3l6) single mutants, as well as the Col(o3l4;o3l6) doublemutant behaved like the WT controls. Early flowering was found only when the Ws4(oxs3) allele was present, as in the Ws4(oxs3) single mutant or in the Ws4(oxs3);Col(o3l4) and Ws4(oxs3);Col(o3l6) double mutants (6e7 leaf stage; Fig. 1B). This suggested that loss of OXS3 function alone must be the primary cause responsible for early flowering under drought stress. Aside from this drought-responsive precocious flowering, the diameter of rosette leaf set in drought-treated Ws4(oxs3) plants were only half of those of well-watered Ws4 plants (Fig. 1C). The leaves were smaller, roundish, thick and hard under drought stress compared to the soft, stretched and big leaves of well-watered plants. Since an oxs3 null mutant is mildly sensitive to oxidative stress [21], it seemed reasonable to assume that the poorer growth of Ws4(oxs3) plants during drought stress could be due to a deficiency in stress tolerance. However, to invoke the same reasoning to account the early flowering seemed less convincing. A plant shortening its life cycle to produce seeds would be a stress response of last resort, when tolerance mechanisms cannot alleviate the severity of stress leading to death. Since the oxs3 mutant is only mildly sensitive to stress, it was unlikely to have been experiencing a life-threatening stress level. If so, this would imply that OXS3 must ordinarily exert a negative regulatory role on flowering during drought stress, such that its absence, the plant flowers earlier. 3.2. OXS3 interact with SOC1 in the nucleus OXS3 was previously reported to co-localize to nuclear speckles [21], and this suggested a possible role in regulating chromatin in transcription. The online NLS prediction tool (http://nls-mapper. iab.keio.ac.jp/cgi-bin/NLS_Mapper_form.cgi) [29] identified the bipartite NLS of OXS3 to be from 122 to 148aa (Figs. S1A and B). In a transient assay using Arabidopsis protoplasts, constructs expressing OXS3-GFP yielded GFP fluorescence solely from the nucleus in 88% of the protoplasts (Fig. 2A and B), while the remaining protoplasts showed GFP fluorescence in both the nucleus and the cytoplasm. None of the protoplasts showed GFP fluorescence from only the cytoplasm. These results are consistent with and extends previous

M. Liang et al. / Biochemical and Biophysical Research Communications 519 (2019) 585e590

589

Fig. 4. OXS3 can repress AP1 promoter activity. (A) Transient expression assay in tobacco leaves testing OXS3 effect on promoter expression. Firefly luciferase activity from cotransformation of 3 Agrobacterium strains: Strain 1: construct of firefly luciferase gene fused to promoter from CO (pCO), SOC1 (pSOC1), LFY (pLFY) or AP1 (pAP1); Strain 2: construct of CaMV 35S RNA promoter fused to Renilla luciferase gene; Strain 3: construct of CaMV 35S RNA promoter fused OXS3 (35S-OXS3), or no 35S-OXS3 in empty vector (EV). Firefly Luciferase activity normalized to Renilla luciferase activity and then to activity shown by cotransformation with empty vector. P value of Student's t-test, two-tailed: ***P  0.001; *P  0.05. Error bars show ± SD from 3 biological replicates. (B) Transient expression assay in Arabidopsis soc1-2 mutant protoplasts testing OXS3 effect on AP1 promoter expression. Error bars show ± SD from 4 biological replicates. (C) Model of OXS3 interacting with SOC1 to repress AP1 expression to prevent precocious flowering.

observations that OXS3 is mainly located in the nucleus [21]. With the hypothesis that OXS3 is necessary to prevent an earlier than WT rate of drought-responsive flowering, we reasoned that it must act either at the DNA, on chromatin factors, or on the floral regulators. Since most of the genes described for the flowering regulatory network encode regulatory proteins, we tested whether OXS3 could interact with these proteins in a yeast two-hybrid (Y2H) experiment. Fig. 3A lists the 10 flowering regulators tested: CO that monitors the circadian clock and photoperiod [30]; SOC1, FT, FD, and FLC that integrate environmental cues; LFY (LEAFY), AP1 (APETALA 1), FUL (FRUITFULL) and SEP3 (SEPALLATA 3), that initiate floral meristem development; and TFL1 (TERMINAL FLOWER 1) that helps maintain meristem indeterminacy [31]. The data show that OXS3 interacted only with SOC1, and not to other proteins tested (Fig. 3B). To see whether the interaction in yeast could also be found in plant cells, the bimolecular fluorescence complementation (BiFC) assay was used on Arabidopsis protoplasts. Positive interaction was found between SOC1 and OXS3 solely in the nucleus (Fig. 2A and B), which suggests that the OXS3/SOC1 complex might possibly regulate SOC1 targets. 3.3. OXS3 may repress flowering through preventing AP1 expression Since SOC1 is a key floral integrator that determines when to flower, its interaction with OXS3 may indicate a role at modulating SOC1 activity. At the shoot apex, SOC1 together with AGL24 activate floral meristem identity genes LFY and AP1 to establish and maintain flower meristem identity [20]. Therefore, we tested whether OXS3 has the ability to affect AP1 and/or LFY promoter expression. We also tested the SOC1 promoter, which could be a target of SOC1 auto-activation [32]. As a negative control, we used the CO promoter because CO is an upstream regulator, and the CO promoter is not known to be a target of SOC1. In a transient transactivation assay in tobacco, the data showed that the inclusion of OXS3 has statistically insignificant effects on the promoters of CO and LFY, but a nearly 3-fold increase of SOC1 promoter expression and a ~13-fold reduction of AP1 promoter expression (Fig. 4A). It is difficult to assess whether the nearly 3-fold increase in SOC1 promoter expression is biologically significant, but the high statistical significance of the ~13-fold reduction of AP1 promoter activity is a good indication that OXS3 could play a role in repressing AP1 expression. Since OXS3 interacts with SOC1, we also tested AP1 promoter expression in protoplasts from soc1-2 null mutant plants. The transient expression of OXS3 produced less of an effect on AP1 promoter repression, from an 8-fold reduction in WT protoplasts to

a 4-fold reduction in soc1 protoplasts (Fig. 4B). This suggests that OXS3-mediated repression of AP1 promoter activity is enhanced by the presence of SOC1. 4. Concluding remarks Ironically, while we started out with the premise that it would be necessary to use combination mutants of the OXS3 gene family to uncover noticeable stress-induced phenotypes, this study led us back to the early flowering response in an oxs3-only mutant. O3L4 and O3L6 do not appear to play a major role, and since we have not yet tested the 4 other members (O3L1, O3L2, O3L3, and O3L5), it is premature to conclude that OXS3 is the only family member involved in stress-induced flowering. Nonetheless, a single oxs3 null mutation is sufficient to cause a noticeable phenotype in stress-induced flowering. Since OXS3 can interact with SOC1 in the nucleus and has the capacity to repress AP1 expression, a likely possibility is that OXS3 can exert a negative regulatory role on stress-induced flowering. The notion that a protein can be involved in both ameliorating stress as well as regulating reproduction is not without precedence. In response to oxidative stress, the OXS2 transcription factor ameliorates stress tolerance during mild stress, as well as promotes stress-escape early flowering during severe stress [9,33]. Interestingly, both OXS2 and OXS3 operate through SOC1, although their effects are opposite. Whereas OXS2 binds the SOC1 promoter to enhance SOC1 transcription to expedite flowering, OXS3 interacts with SOC1 to repress AP1 expression to delay flowering. This speculation on the dual role of OXS3 raises questions on why and how. Why OXS3 regulates reproduction may be a result of evolutionary fitness. While OXS3 is doing its part to counter the stress, it may be advantageous to also monitor the stress level to prevent precocious early flowering. In the absence of this ‘safety check’ function, precocious early flowering would only lead to lower progeny yield, translating to decreased reproductive fitness. How can this be done mechanistically may be explained by protein availability. When OXS3 is heavily involved in stress tolerance, there is less protein available to delay flowering, but this is also a time when the plant is experiencing stress that might possibly need to activate the stress escape response. As the encountered stress is alleviated, OXS3 would become available to resume its role to repress flower development. Author contributions M.L. and D.O. designed the experiments. M.L. performed the

590

M. Liang et al. / Biochemical and Biophysical Research Communications 519 (2019) 585e590

experiments assisted by S.X. and J.C. for data shown in Figs. 1B and 4. M.L. and D.O. analyzed the data, discussed the results and wrote the manuscript.

[13] [14]

Conflicts of interest [15]

The authors declare that they have no conflict of interest. Acknowledgments Funding was provided by the Ministry of Science and Technology of China (2016YFD0101904) to D.O. and a Chinese Academy of Sciences Postdoctoral Fellowship to M.L.

[16]

[17] [18] [19]

Appendix A. Supplementary data Supplementary data to this article can be found online at https://doi.org/10.1016/j.bbrc.2019.08.154.

[20] [21]

[22]

References [1] O. Leyser, Regulation of shoot branching by auxin, Trends Plant Sci. 8 (2003) 541e545. [2] S. Kalisz, E.M. Kramer, Variation and constraint in plant evolution and development, Heredity 100 (2008) 171e177. [3] A. Srikanth, M. Schmid, Regulation of flowering time: all roads lead to Rome, Cell. Mol. Life Sci. 68 (2011) (2011) 2013e2037. [4] J.E. Purvis, G. Lahav, Encoding and decoding cellular information through signaling dynamics, Cell 152 (2013) 945e956. [5] S. Fukai, G. Pantuwan, B. Jongdee, et al., Screening for drought resistance in rainfed lowland rice, Field Crop. Res. 64 (1999) 61e74. [6] K. Takeno, Stress-induced flowering: the third category of flowering response, J. Exp. Bot. 67 (2016) 4925e4934. [7] P.E. Verslues, T.E. Juenger, Drought, metabolites, and Arabidopsis natural variation: a promising combination for understanding adaptation to waterlimited environments, Curr. Opin. Plant Biol. 14 (2011) 240e245. [8] A. Pajoro, S. Biewers, E. Dougali, et al., The (r)evolution of gene regulatory networks controlling Arabidopsis plant reproduction: a two-decade history, J. Exp. Bot. 65 (2014) 4731e4745. [9] R. Blanvillain, S. Wei, P.C. Wei, et al., Stress tolerance to stress escape in plants: role of the OXS2 zinc-finger transcription factor family, EMBO J. 30 (2011) 3812e3822. [10] M. Riboni, M. Galbiati, C. Tonelli, et al., GIGANTEA enables drought escape response via abscisic acid-dependent activation of the florigens and SUPPRESSOR OF OVEREXPRESSION OF CONSTANS1, Plant Physiol. 162 (2013) 1706e1719. [11] S.R. Hepworth, F. Valverde, D. Ravenscroft, et al., Antagonistic regulation of flowering-time gene SOC1 by CONSTANS and FLC separate promoter motifs, EMBO J. 21 (2002) 4327e4337. [12] C.A. Helliwell, C.C. Wood, M. Robertson, et al., The Arabidopsis FLC protein

[23]

[24]

[25] [26]

[27]

[28] [29]

[30]

[31] [32]

[33]

interacts directly in vivo with SOC1 and FT chromatin and is part of a highmolecular-weight protein complex, Plant J. 46 (2006) 183e192.
s, G. Coupland, The genetic basis of flowering responses to seasonal F. Andre cues, Nat. Rev. Genet. 13 (2012) 627e639. A. Yamaguchi, M.F. Wu, L. Yang, et al., The microRNA-regulated SBP-box transcription factor SPL3 is a direct upstream activator of LEAFY, FRUITFULL, and APETALA1, Dev. Cell 17 (2009) 268e278. J. Jung, Y. Ju, P.J. Seo, et al., The SOC1-SPL module integrates photoperiod and gibberellic acid signals to control flowering time in Arabidopsis, Plant J. 69 (2010) 577e588. V. Gregis, A. Sessa, L. Colombo, et al., AGAMOUS-LIKE24 and SHORT VEGETATIVE PHASE determine floral meristem identity in Arabidopsis, Plant J. 56 (2008) 891e902. D. Weigel, J. Alvarez, D.R. Smyth, et al., LEAFY controls floral meristem identity in Arabidopsis, Cell 69 (1992) 843e859. J. Lee, M. Oh, H. Park, et al., SOC1 translocated to the nucleus by interaction with AGL24 directly regulates LEAFY, Plant J. 55 (2008) 832e843. C. Liu, H. Chen, H.L. Er, et al., Direct interaction of AGL24 and SOC1 integrates flowering signals in Arabidopsis, Development 135 (2008) 1481e1492. J. Lee, I. Lee, Regulation and function of SOC1, a flowering pathway integrator, J. Exp. Bot. 61 (2010) 2247e2254. R. Blanvillain, J.H. Kim, S. Wu, et al., OXIDATIVE STRESS 3 is a chromatinassociated factor involved in tolerance to heavy metals and oxidative stress, Plant J. 57 (2009) 654e665. M.A. Mandel, C. Gustafson-Brown, B. Savidge, et al., Molecular characterization of the Arabidopsis floral homeotic gene APETALA1, Nature 360 (1992) 273e277. J.L. Bowman, J. Alvarez, D. Weigel, et al., Control of flower development in Arabidopsis thaliana by APETALA1 and interacting genes, Development 119 (1993) 721e743. J. Moon, S. Suh, H. Lee, et al., The SOC1 MADS-box gene integrates vernalization and gibberellin signals for flowering in Arabidopsis, Plant J. 35 (2003) 613e623. C. Liu, W. Xi, L. Shen, et al., Regulation of floral patterning by flowering time genes, Dev. Cell 16 (2009) 711e722. S.D. Yoo, Y.H. Cho, J. Sheen, Arabidopsis mesophyll protoplasts: a versatile cell system for transient gene expression analysis, Nat. Protoc. 2 (2007) 1565e1575. ~ ate-Sa
nchez, K.B. Singh, Identification of Arabidopsis ethyleneL. On responsive element binding factors with distinct induction kinetics after pathogen infection, Plant Physiol. 128 (2002) 1313e1322. F. Fornara, A. de Montaigu, G. Coupland, SnapShot: control of flowering in Arabidopsis, Cell 3 (2010) 550. S. Kosugi, M. Hasebe, M. Tomita, et al., Systematic identification of yeast cell cycle-dependent nucleocytoplasmic shuttling proteins by prediction of composite motifs, Proc. Natl. Acad. Sci. U.S.A. 106 (2009) 10171e10176. S.K. Yoo, K.S. Chung, J. Kim, et al., CONSTANS activates SUPPRESSOR OF OVEREXPRESSION OF CONSTANS 1 through FLOWERING LOCUS T to promote flowering in Arabidopsis, Plant Physiol. 139 (2005) 770e778. S. Shannon, D.R. Meeks-Wagner, A mutation in the Arabidopsis TFL1 gene affects inflorescence meristem development, Plant Cell 3 (1991) 877e892.
, S. Ferrario, et al., Characterization of SOC1's central R.G.H. Immink, D. Pose role in flowering by the identification of its upstream and downstream regulators, Plant Physiol. 160 (2012) 433e449. M.T. Liang, D.W. Ow, Nucleocytoplasmic OXIDATIVE STRESS 2 can relocate FLOWERING LOCUS T, Biochem. Biophys. Res. Commun. 517 (2019) 735e740.