Introduction

Plants have evolved various light receptors, such as phytochromes, cryptochromes, and phototropins, to use light as a cue for growth and development (Franklin and Quail 2010; Chen and Chory 2011; Suetsugu and Wasda 2013). LOV KELCH PROTEIN 2 (LKP2), ZEITLUPE (ZTL) or LOV KELCH PROTEIN 1 (LKP1), and FLAVIN-BINDING KELCH REPEAT F-BOX 1 (FKF1) form a small family of photoreceptors in Arabidopsis (Kiyosue and Wada 2000; Nelson et al. 2000; Somers et al. 2000; Jarillo et al. 2001; Schultz et al. 2001) and are involved in the regulation of the circadian clock by ubiquitin-mediated degradation of clock components such as TIMING OF CAB EXPRESSION 1 (TOC1) and PSEUDO-RESPONSE REGULATOR 5 (PRR5) (Más et al. 2003; Kiba et al. 2007; Baudry et al. 2010; Wang et al. 2010). ztl mutants show a long-period phenotype (approximately 27 h; Somers et al. 2004). Both ztl fkf1 and ztl fkf1 lkp2 mutants also exhibit a long-period phenotype (approximately 29 h), although the amplitude of the circadian rhythm is weaker in the latter mutant (Baudry et al. 2010). Overexpression of LKP2 or ZTL shortens the period and causes arrhythmicity (Schultz et al. 2001; Somers et al. 2004).

Protein of this family also regulates flowering time. Overexpression of LKP2 or ZTL in Arabidopsis leads to downregulation of the expression of CONSTANS (CO) and FLOWERING LOCUS T (FT), leading to delayed flowering under long-day conditions (Somers et al. 2004; Miyazaki et al. 2011), whereas ztl and lkp2 ztl mutants exhibit early-flowering phenotypes, although the lkp2 mutant has a wild-type phenotype (Takase et al. 2011). fkf1 mutants exhibit a late-flowering phenotype (Nelson et al. 2000; Imaizumi et al. 2003).

The function of LKP2 is postulated to overlap mostly with that of ZTL because they are structurally more closely related to each other than to FKF1, and because overexpression of either gene in Arabidopsis leads to common phenotypes, i.e., arrhythmicity, late flowering, and elongated hypocotyls (Kiyosue and Wada 2000; Somers et al. 2000; Schultz et al. 2001). Furthermore, ZTL promoter-driven LKP2 expression can complement the ztl mutant phenotype (Baudry et al. 2010). However, as mentioned above, the phenotypes of ztl and lkp2 knockout mutants are different—the ztl mutant shows a longer circadian rhythm and an early-flowering phenotype, whereas the lkp2 mutant shows wild-type phenotype both in the circadian rhythm and flowering (Somers et al. 2000; Baudry et al. 2010; Takase et al. 2011). Such differences may reflect the differences in the relative expression levels of LKP2 and ZTL. The expression level of ZTL is approximately eight times that of LKP2 under 16 h light/8 h dark and 8 h light/16 h dark conditions (Mockler et al. 2007; Michael et al. 2008), and LKP2 expression is only 4 % of the level observed for ZTL under constant light conditions (Baudry et al. 2010). Overexpression of ZTL conferred both strong and weak late-flowering phenotypes, whereas that of LKP2 conferred only a strong late-flowering phenotype (Kiyosue and Wada 2000, Schultz et al. 2001). Light-dependent signals affect multiple aspects of plant growth, development, and responses to environmental stresses (González et al. 2012; Oh et al. 2014). Thus, LKP2 may have functions beyond its well-studied roles in regulation of the circadian clock and flowering time.

Limited water availability is a major environmental factor affecting productivity of crop plants; recent global warming is predicted leading to further water shortages in most agricultural regions of the world (Xoconostle-Cázares et al. 2010). Many transgenic studies have attempted to improve the drought tolerance of plants by overexpressing genes for dehydration-inducible transcription factors such as dehydration-responsive element-binding factors (DREBs)/C-repeat binding factors (CBFs) or their downstream targets such as responsive to dehydration (RD)/cold-regulated (COR) genes (Yamaguchi-Shinozaki and Shinozaki 2006; Saibo et al. 2009; Qin et al. 2011).

In this study, we performed microarray analysis to elucidate unidentified effects of LKP2 overexpression in plants. We found that LKP2 overexpression confers dehydration tolerance in Arabidopsis, accompanied by the activation of DREB/CBF and their downstream target genes.

Materials and methods

Plant materials and growth conditions

Wild-type Arabidopsis thaliana Columbia (Col) accession and the transgenic plants (T5 generation) 35S:GFP-LKP2 line 1 (GFP-LKP2-1), 35S:GFP-LKP2 line 2 (GFP-LKP2-2), and 35S:GFP (GFP) (Yasuhara et al. 2004; Miyazaki et al. 2011) were used in this study. Plants were grown on plates with 0.8 % (w/v) agar containing 1/2 basal salt Murashige and Skoog medium and 1 % sucrose (GM medium) or in pots with vermiculite supplemented with 0.1 % Hyponex (Hyponex, Tokyo, Japan) under long-day conditions at 22 °C (16 h light/8 h dark, 80–100 μmol m−2 s−1).

Microarray analysis

Thirty two-day-old non-stressed GFP-LKP2-1 and GFP plants grown on GM agar plates were used for analysis. Leaves were harvested at zeitgeber time 0 (ZT0). Leaves from three seedlings were pooled before extraction. RNA was prepared using an RNeasy Plant Mini Kit (QIAGEN, Valencia, CA, USA). The microarray experiment was conducted by Miltenyi Biotec K.K. (Tokyo, Japan) using Agilent Whole Arabidopsis Genome Oligo Microarrays 4 × 44 K V4 (G2519F) (Agilent Technologies, Böblingen, Germany). RNA samples (100 ng) were used for the linear T7-based amplification step. To produce Cy3-labeled cRNA, amplified RNA was labeled using an Agilent Low Input Quick Amp Labeling Kit (Agilent Technologies) following the manufacturer’s protocol. Yields of cRNA and the dye incorporation rate were measured using an ND-1000 spectrophotometer (NanoDrop Technologies, Wilmington, DE, USA). Hybridization was performed according to the Agilent 60-mer oligo microarray processing protocol using an Agilent Gene Expression Hybridization Kit (Agilent Technologies). Cy3-labeled fragmented cRNA (1.65 μg) in hybridization buffer was hybridized overnight (17 h, 65 °C) in Agilent’s recommended hybridization chamber and oven. The microarrays were then washed once with Agilent Gene Expression Wash Buffer 1 for 1 min at room temperature followed by a second wash with Agilent Gene Expression Wash Buffer 2 (preheated at 37 °C) for 1 min. The last washing step was performed with acetonitrile. Fluorescence signals were detected using Agilent Microarray Scanner System (Agilent Technologies). Agilent Feature Extraction Software (Agilent Technologies) was used to read out and process the microarray image files. Microarray data are accessible through GEO series accession number GSE37197 (http://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE37197). Candidate genes with a fold change >2 and p < 0.01 were used for pathway analysis with the Gene Functional Classification Tool in DAVID Bioinformatics Resources 6.7 (http://david.abcc.ncifcrf.gov/home.jsp). Gene annotations were retrieved from TAIR (http://www.arabidopsis.org/).

Dehydration stress treatment

Eight-day-old seedlings grown on GM agar plates were harvested at ZT4, placed on paper towels (Nippon Paper Crecia, Tokyo, Japan) and allowed to air-dry at 22 °C and 40 % humidity under white light (25 μmol m−2 s−1) for various time periods. For bioassay, the seedlings were allowed to air-dry for 75 min, and then returned to GM agar plates for rehydration and growth. Leaf and root numbers of each plant were counted 8 days after the treatment. The plants were then transferred into pots and plant survival rates were determined 3 weeks after the treatment (i.e., 13 days in vermiculite); 20 plants per group were analyzed. The survival experiment was repeated three times.

RT-PCR analysis

Eight-day-old seedlings grown on GM agar plates were harvested or subjected to the dehydration stress, and then immediately frozen in liquid N2. Total RNA was isolated using Sepasol-RNAI (Nacalai Tesque, Kyoto, Japan) according to the manufacturer’s instructions. To synthesize cDNA, total RNA (1 μg) was primed with the Oligo d(T) primer and reverse transcribed using a High-Capacity cDNA Reverse Transcription Kit (Applied Biosystems, Foster City, CA, USA). PCR amplification was performed with AmpliTaq Gold (Applied Biosystems) and gene-specific primer sets listed in Table S1. PCR reactions were carried out in a thermal cycler with denaturation at 94 °C for 5 min, followed by a variable number of cycles (Table S1) at 94 °C for 30 s, 50 °C for 20 s, and 72 °C for 1 min (GeneAmp PCR System 2700, Applied Biosystems). PCR products were separated on a 2.0 % (w/v) agarose gel, stained with ethidium bromide, and visualized under UV light. Gel images were captured with a Printgraph video capture device (AE-6911FX; Atto, Tokyo, Japan). Each PCR fragment was sequenced directly to confirm its identity.

Stomatal aperture

Stomatal apertures in the abaxial epidermis of two-week-old rosette plants were measured according to the method described by Kinoshita et al. (2011) using images taken with a BX51 microscope (Olympus, Tokyo, Japan) and Scion Image software (Scion Corporation; http://scion-image.software.informer.com).

Statistical analyses

Data are presented as the mean ± SE. Student–Newman–Keuls test was performed with IBM SPSS Statistics for Windows, Version 20.0 (IBM Corp., Armonk, NY, USA).

Results

Gene expression profile of GFP-LKP2 plants

Using microarray analysis, we compared the expression profiles of RNAs isolated from leaves of adult GFP-LKP2-1 and GFP (control) plants in the absence of dehydration stress. Only genes with a > twofold change in GFP-LKP2-1 plants were considered (accession numbers GSM913367 and GSM913368; Table S2). A total of 544 genes were upregulated (p < 0.01), including the transposable element gene AT2G27375 (fold change of 23) and LKP2 (fold change of 21). A total of 739 genes were downregulated (p < 0.01), including ANTHOCYANIN 5-AROMATIC ACYLTRANSFERASE1 (fold change of −100) and FT (fold change of −4). The upregulated and downregulated genes were classified into several groups (Table S3). We found that, in GFP-LKP2-1 plants, genes involved in external encapsulating structure organization, response to light stimulus, response to abiotic stimulus, chromatin organization, and secondary metabolic process were upregulated, whereas genes involved in response to organic substances, regulation of transcription, and immune response were downregulated.

RT-PCR analysis of dehydration–related genes detected by microarray analysis

Microarray analysis suggested that many biological processes were altered in GFP-LKP2 plants. We focused on several genes involved in response to dehydration (Cluster 2, response to abiotic stimulus, in Table S3). Among the abiotic stimulus-responsive Cluster 2 genes (Table 1), we selected five dehydration-inducible genes, namely LTP3, DREB1A/CBF3, RD29A, DEHYDRIN, and LTI30, and performed RT-PCR to evaluate their upregulation in GFP-LKP2-1 and GFP-LKP2-2 plants in comparison with control plants (Col and GFP) using RNA isolated from 8-day-old seedlings (Fig. 1). This experiment confirmed that the expression of LTP3, DREB1A/CBF3, RD29A, and DEHYDRIN was higher in both GFP-LKP2-1 and GFP-LKP2-2 plants in comparison with both controls; an ambiguous result was obtained for LTI30.

Table 1 Abiotic stress-responsive upregulated genes in GFP-LKP2 plants in comparison with GFP plants identified by the Gene Functional Classification Tool within DAVID Bioinformatics Resources 6.7
Full size table
Fig. 1
figure 1

Confirmation of microarray data using RT-PCR. Expression of genes identified by microarray analysis as upregulated in LKP2-GFP plants and annotated as dehydration-inducible genes in 8-day-old GFP-LKP2-1, GFP-LKP2-2, and control (Col and GFP) seedlings in the absence of stress. Actin gene (ACT2) was used as a control

Full size image

After dehydration stress, GFP-LKP2 plants develop more leaves and roots and have a higher survival rate than control plants

The upregulation of several dehydration-inducible genes in GFP-LKP2 plants suggested that these plants might be more dehydration-tolerant than GFP plants. To assess this possibility, we examined the effect of dehydration stress on leaf and root growth in GFP-LKP2 and control (Col and GFP) plants. Seedlings were dehydrated for 75 min and then grown on GM agar plates. Eight days after release from dehydration stress, GFP-LKP2 plants developed more leaves and roots than treated control plants (Fig. 2a, b). By contrast, in the absence of stress, the number of roots and leaves in GFP-LKP2 plants did not exceed that in control plants (Fig. S1).

Fig. 2
figure 2

Dehydration tolerance of GFP-LKP2 plants. a GFP-LKP2 (lines 1 and 2) and control (Col and GFP) seedlings on GM agar plates before and after dehydration stress. b Leaf and root numbers of GFP-LKP2 and control plants grown for 8 days on GM agar medium after dehydration stress for 75 min (n = 20; mean ± SE). c GFP-LKP2 and control plants grown for 3 weeks in pots with vermiculite after exposure to dehydration stress. d Survival rates of the GFP-LKP2 and control plants 3 weeks after dehydration stress. Twenty plants from each group were used in one experiment (n = 3; mean ± SE). Different letters indicate statistical difference (p < 0.05, Student–Newman–Keuls test)

Full size image

Three weeks after release from dehydration stress, almost all GFP-LKP2 plants grew well in pots, while only 12 % of control plants survived (Fig. 2c, d).

Precocious accumulation of DREB1A–C/CBF1–3 and RD29A mRNA in GFP-LKP2 plants under normal conditions is followed by their further increase upon dehydration

Next we examined the expression of four genes, DREB1A-C/CBF1-3, DREB2A, B, and RD29A, three of which have been identified by our microarray analysis (Table 1) and three other are known as dehydration-responsive genes involved in stress tolerance (Yamaguchi-Shinozaki and Shinozaki 2006).

In the absence of stress, the levels of DREB1A–C/CBF1–3 and RD29A mRNAs were substantially higher in GFP-LKP2 plants than in control plants (Col and GFP; Fig. 3a). After the onset of dehydration stress, the mRNA levels of all four genes were robustly induced in both GFP-LKP2 and control plants. In control plants, DREB2B and RD29A peaked at 2 h. However, DREB2B peaked at 1 h in both GFP-LKP2 lines, whereas RD29A peaked at 40 min in GFP-LKP2-1 and at 1 h in GFP-LKP2-2.

Fig. 3
figure 3

RT-PCR analysis of stress-responsive transcripts during and after dehydration stress in GFP-LKP2 (lines 1 and 2) and control (Col and GFP) 8-day-old seedlings. a DREB1A/CBF3, DREB1B/CBF1, DREB1C/CBF2, DREB2A, DREB2B, and RD29A. b MYB2, MYC2, RD22, AREB1, AREB2, ABF3, RD29B, ANAC019, ANAC055, ANAC072, ZFHD1, ERD1, and STZ. ACT2 was used as a control. Time 0 = untreated

Full size image

Besides the DREB/CBF pathway, three other signaling networks are postulated to be involved in the induction of RD or Early Responsive to Dehydration (ERD) during the response to dehydration (Yamaguchi-Shinozaki and Shinozaki 2006; Saibo et al. 2009), namely (1) myeloblastosis (MYB)–myelocytomatosis (MYC) pathway, (2) the abscisic acid (ABA) pathway, and (3) NAM (no apical meristem)/ATAF (Arabidopsis transcription activation factor)/CUC (cup-shaped cotyledon) (NAC) and zinc finger homeodomain (ZFHD) pathway. We performed RT-PCR to evaluate the contribution of these three pathways to the enhanced dehydration stress tolerance of GFP-LKP2 plants. No obvious upregulation of 13 selected genes was detected in unstressed GFP-LKP2 compared to control plants (Fig. 3b, time 0). After dehydration stress, the expression of only one gene, ANAC055, was higher in GFP-LKP2 plants than in both Col and GFP plants (Fig. 3b).

GFP-LKP2 plants exhibit a closed stomata phenotype

We previously reported that a florigen gene FT is downregulated in GFP-LKP2 plants, leading to late flowering (Takase et al. 2011). FT positively regulates not only flowering time but also stomatal opening (Kinoshita et al. 2011). Therefore, we speculated that FT downregulation might confer a “less open” stomata phenotype in GFP-LKP2 plants. To evaluate this possibility, we examined stomatal apertures microscopically (Fig. 4a). Stomata opened more in the daytime (Fig. 4b) and less at night (Fig. S2) in control plants (Col and GFP), in line with a previous report (Dodd et al. 2005). Stomata were less open in the daytime in GFP-LKP2 plants than in control plants (Fig. 4b). There was no difference in stomata opening at night between the control and GFP-LKP2 plants (Fig. S2). Stomatal density was the same in the control and GFP-LKP2 plants (Fig. 4c). Thus, the “less open” stomata phenotype, which limits water loss by transpiration, may explain the enhanced dehydration stress tolerance of GFP-LKP2 plants.

Fig. 4
figure 4

Stomatal aperture and density in GFP-LKP2 (lines 1 and 2) and control (Col and GFP) plants. a Stomata at zeitgeber time 4 (ZT4). Bar 10 μm. b Stomatal aperture under long-day (16 h light/8 h dark) conditions (n = 20; mean ± SE). c Stomatal density (n = 20; mean ± SE). Different letters indicate statistical difference (p < 0.05, Student–Newman–Keuls test)

Full size image

Discussion

In the present study, using microarray analysis, we characterized the gene expression profile of GFP-LKP2 plants in comparison with GFP plants. AT2G27375, a transposable element gene, was most strongly upregulated; however, the biological significance of this upregulation remains unclear. The fact that LKP2 was the second most upregulated gene validates this microarray analysis. The late-flowering phenotype of LKP2-overexpressing plants is due to downregulation of the florigen gene FT (Takase et al. 2011). The fact that FT was downregulated in GFP-LKP2 plants also validates this microarray analysis.

Functional classification placed the differentially expressed genes into several groups, such as those involved in cell wall organization, response to light, cell division and growth, and responses to abiotic stimuli. We examined upregulation of several dehydration responsive genes, as this process is one of the most fundamental in both basic and applied plant biology. The enhanced dehydration tolerance of the GFP-LKP2 plants is underscored by the following observations. First, the leaf and root numbers in the GFP-LKP2 plants were higher than in the control plants after the recovery from dehydration. Second, after dehydration stress the survival rate of GFP-LKP2 plants was higher than that of controls.

DREB1A–C/CBF1–3 play a central role in cold stress responses in Arabidopsis (Yamaguchi-Shinozaki and Shinozaki 2006). These genes encode transcription factors that belong to a small group within the AP2/ERF superfamily, whereas DREB2A and DREB2B encode transcription factors that belong to another subgroup of the same superfamily (Liu et al. 1998; Qin et al. 2011). DREB1A–C/CBF1–3 are induced by cold stress, whereas DREB2A and DREB2B are induced following dehydration (Qin et al. 2011). Since DREB1A–C/CBF1–3, DREB2A, and DREB2B recognize the DRE cis-acting element, which contains the core motif A/GCCGAC, they activate many drought- and cold-inducible genes (including RD29A) whose promoters contain the DRE element (Fowler and Thomashow 2002; Sakuma et al. 2006; Qin et al. 2011). The cold-inducible genes function not only in cold stress, but also in dehydration tolerance (Yamaguchi-Shinozaki and Shinozaki 2006). Indeed, transgenic plants overexpressing DREB1A and those overexpressing constitutively active DREB2A exhibit dehydration tolerance (Liu et al. 1998; Sakuma et al. 2006). Thus, both the enhanced expression of DREB1A–C/CBF1–3 in the absence of stress and accelerated accumulation of DREB2B mRNA in response to dehydration stress could explain the enhanced dehydration tolerance of GFP-LKP2 plants.

The transcript levels of approximately 1/3 of expressed A. thaliana genes, including genes responsive to abiotic or biotic stresses, are regulated in a circadian manner (Covington et al. 2008). DREB1A–C/CBF1–3 have circadian expression that peaks at midday (Dong et al. 2011). This higher DREB1A–C/CBF1–3 expression before night may prepare plants for the daily temperature changes by providing a certain level of preformed cold tolerance (Nakamichi et al. 2009). PRR5 (PSEUDO-RESPONSE REGULATOR 5), a clock component, represses DREB1A–C/CBF1–3 expression via binding to their promoter regions (Nakamichi et al. 2009, 2012). LKP2 is reported to degrade PRR5 (Baudry et al. 2010). Thus, the reduction in the PRR5 level due to LKP2 overexpression may engender higher expression of DREB1A–C/CBF1–3 and their target genes (direct, indirect, or both) in GFP-LKP2 plants.

The MYB–MYC pathway, ABA pathway, and NAC and ZFHD pathway are postulated to be involved in dehydration tolerance in Arabidopsis (Yamaguchi-Shinozaki and Shinozaki 2006; Saibo et al. 2009). Using RT-PCR, we analyzed the expression of seven genes (MYB2, MYC2, RD22, AREB1, AREB2, ABF3, and RD29B) that belong to these pathways. Under dehydration stress, ABA accumulation induces ABA-responsive genes that encode proteins required for dehydration tolerance (Saibo et al. 2009). The expression of MYB2 and MYC2 genes is induced by ABA (Urao et al. 1993; Abe et al. 1997) and in its turn induces RD22 (Abe et al. 2003). Most ABA-responsive genes contain cis-acting elements called ABRE (ABA-responsive elements), which are recognized by a group of bZIP-type transcription factors named AREBs (ABRE-binding proteins), or ABRE-binding factors (ABFs) (Yamaguchi-Shinozaki and Shinozaki 2006). Expression of AREB1, AREB2, and ABF3 is also upregulated by ABA (Uno et al. 2000; Fujita et al. 2005; Yoshida et al. 2010). RD29B is a downstream target of AREB1, AREB2, and ABF3 (Yoshida et al. 2010). As the expression levels of MYB2, MYC2, RD22, AREB1, AREB2, ABF3, and RD29B were not apparently different in GFP-LKP2 and control plants under both conditions (Fig. 3b), we infer that the MYB2–MYC2 and ABA pathways are unlikely to be involved in dehydration tolerance of LKP2-overexpressing plants.

In the NAC and ZFHD pathway, NAC and ZFHD transcription factors modulate the expression of early response genes upon dehydration stress (Fujita et al. 2004; Tran et al. 2004). Upon dehydration, only the level of ANAC055 was apparently higher in the GFP-LKP2 plants than in the controls (Fig. 3b). Transgenic plants overexpressing ANAC055 have increased drought tolerance (Tran et al. 2004); thus, ANAC055 protein may protect LKP2-overexpressing plants against dehydration stress.

We do not know how overproduction of GFP-LKP2 enhances accumulation of DREB2B and ANAC055 transcripts. Overexpression of LKP2 causes circadian disorder and late flowering partially by degradation of PRR5 and inhibition of FKF1 by its exclusion from the nuclei; it probably also causes polyubiquitination and subsequent degradation of unidentified LKP2 target proteins by 26S proteasome (Schultz et al. 2001; Takase et al. 2011; Miyazaki et al. 2011). One or more of these or some unidentified mechanisms may underlie the elevated sensitivity of GFP-LKP2 plants to dehydration stress.

FT is a florigen gene expressed in vascular tissues; in the apical meristem, FT interacts with the bZIP transcription factor FD (Abe et al. 2005). In the guard cells, FT positively regulates stomatal opening. Ectopically expressed FT in guard cells confers a continuously wide-open stomata phenotype, whereas a loss-of-function ft mutant exhibits a continuously narrowly opened stomata phenotype (Kinoshita et al. 2011). LKP2 overexpression reduces FT expression in leaves (Takase et al. 2011). This may explain the continuously narrowly opened stomata phenotype in GFP-LKP2 plants (Fig. 4, S2). Although in the guard cells, CO and GIGANTEA are involved in FT expression (Ando et al. 2013), no such data are available for the LKP2/ZTL/FKF1 family photoreceptors.

The loss-of-function lkp2 mutants do not exhibit a dehydration-tolerant phenotype (data not shown). This is not surprising, because the lkp2 mutants also retain normal circadian rhythm and flowering time (Baudry et al. 2010; Takase et al. 2011). The reason why lkp2 mutant seedlings do not show enhanced dehydration tolerance may be that the expression level of LKP2 is much lower than that of ZTL. ZTL expression level is about eight times that of LKP2 under 16 h light/8 h dark conditions (Mockler et al. 2007; Michael et al. 2008). Thus, ZTL, FKF1, or both might be sufficient in the absence of LKP2, thereby compensating for any defects in response to dehydration.

We cannot exclude the possibility that the GFP-fusion protein has an artificial effect, though overexpression of either GFP-LKP2 or LKP2 conferred elongated hypocotyls, a circadian rhythm defect, and late flowering in Arabidopsis (Schultz et al. 2001; Miyazaki et al. 2011). Regardless of whether or not GFP-LKP2 has an artificial effect, it would be useful for biotechnological applications since overproduction of GFP-LKP2 enhances dehydration tolerance.

Plants that overexpress LKP2 show pleiotropic phenotypes, such as late flowering (Schultz et al. 2001; Takase et al. 2011), hypocotyl elongation (Schultz et al. 2001; Miyazaki et al. 2011), circadian rhythm defects (Schultz et al. 2001; Miyazaki et al. 2011), and dehydration tolerance (this study). This suggests the involvement of LKP2 in the basic processes of plant development, growth, and stress responses. Further studies of LKP2 will increase our understanding of the functions of LKP2/FKF1/ZTL photoreceptor family. Furthermore, our results suggest that, in addition to its well-documented activity in the regulation of flowering time, this photoreceptor family gene can be exploited for future biotechnological applications, such as the enhancement of dehydration tolerance.

Author contribution statement

Conceived and designed the experiments: YM, HA, TK. Performed the experiments: YM, HA. Analyzed the data: YM, HA, TT. Contributed reagents, materials, and analysis tools: YM, HA, MK. Wrote the paper: YM, TK.