Plant Mol Biol DOI 10.1007/s11103-015-0324-z

CYCLIN‑DEPENDENT KINASE G2 regulates salinity stress response and salt mediated flowering in Arabidopsis thaliana Xiaoyan Ma1 · Zhu Qiao1 · Donghua Chen1 · Weiguo Yang1 · Ruijia Zhou1 · Wei Zhang1 · Mei Wang1 

Received: 28 November 2014 / Accepted: 21 April 2015 © Springer Science+Business Media Dordrecht 2015

Abstract  Cyclin-dependent protein kinases are involved in many crucial cellular processes and aspects of plant growth and development, but their precise roles in abiotic stress responses are largely unknown. Here, Arabidopsis thaliana CYCLIN-DEPENDENT KINASE G2 (CDKG2) was shown to act as a negative regulator of the salinity stress response, as well as being involved in the control of flowering time. GUS expression experiments based on a pCDKG2::GUS transgene suggested that CDKG2 was expressed throughout plant development, with especially high expression levels recorded in the seed and in the flower. The loss-of-function of CDKG2 led to an increased tolerance of salinity stress and the up-regulation of the known stress-responsive genes SOS1, SOS2, SOS3, NHX3, RD29B, ABI2, ABI3, MYB15 and P5CS1. Flowering was accelerated in the cdkg2 mutants via the repression of FLC and the consequent up-regulation of FT, SOC1, AP1 and LFY. Transgenic lines constitutively expressing CDKG2 showed greater sensitivity to salinity stress and were delayed in flowering. Furthermore, the CDKG2 genotype affected the response of flowering time to salinity stress. Our data connect CDKG2 to

Xiaoyan Ma, Zhu Qiao and Donghua Chen have contributed equally to this work. Electronic supplementary material  The online version of this article (doi:10.1007/s11103-015-0324-z) contains supplementary material, which is available to authorized users. * Mei Wang [email protected] 1



Key Laboratory of Plant Cell Engineering and Germplasm Innovation, Ministry of Education, School of Life Science, Shandong University, Jinan 250100, China

undescribed functions related to salt stress tolerance and flowering time through the regulation of specific target genes. Keywords  CDKG2 · Salt stress · Flowering time · Transcriptional regulation · Arabidopsis

Introduction Cyclin-dependent protein kinases (CDKs) play important roles in the regulation of the cell cycle progression (Morgan 1997). In diverse plant species, the number and variety of CDKs is far beyond and their kinase activity needs to be induced by the binding of a regulatory cyclin partner (Mironov et al. 1999; Morgan 1997). Based on motifs present in their cyclin-binding domains, Arabidopsis thaliana CDKs have been organized into types CDKA through CDKG (Menges et al. 2005; Umeda et al. 2005). Recent research indicates that the members of this family are involved in many crucial cellular processes and plant growth process, such as G1/S and G2/M transitions (Hemerly et al. 1995), cell proliferation during the male gametophyte development (Iwakawa et al. 2006), cell fates determination in floral organs and cell expansion in leaves (Inze and De Veylder 2006; Wang and Chen 2004). CDKG class with conserved PLTSLRE motif comprises two members, CDKG1 and CDKG2 (Menges et al. 2005). The product of the former regulates pre-mRNA splicing during pollen wall formation (Huang et al. 2013), and recent evidence suggests that the CDKG1/CYCLINL complex is essential for synapsis and recombination during male gametogenesis (Zheng et al. 2014). Meanwhile, CDKG2 transcript is found predominantly in young seedlings and suspension/callus cells (Menges et al. 2005),

13



and its product appears to be required to induce organogenesis from in vitro cultured cells (Z˙ abicki et al. 2013). However, very little is known about the exact function of CDKG2 in the regulation of salinity stress response and plant flowering time. Soil salinity has a profound effect on plant development. In many species, the onset of flowering is delayed when the plant experiences salinity stress (Van Zandt and Mopper 2002). The transition from vegetative to reproductive growth is regulated by the interplay of both endogenous and exogenous signals (Baurle and Dean 2006; Boss et al. 2004). At least four independent signaling pathways are known to participate in this regulation, namely the photoperiod-, vernalization-, the gibberellin-dependent pathways and the autonomous pathway (Boss et al. 2004; Mouradov et al. 2002; Simpson and Dean 2002). In combination, these pathways interact to control the floral pathway integrators, FLOWERING LOCUS T (FT) and SUPPRESSOR OF OVEREXPRESSION OF CO1 (SOC1) (Lee et al. 2000; Moon et al. 2003, 2005; Samach et al. 2000). The SOC1 and FT genes are repressed by FLOWERING LOCUS C (FLC) which plays a center role both in vernalization and autonomous pathway, as well as activated by CONSTANS (CO) which is specific to photoperiod pathway (Borner et al. 2000; Lee et al. 2000; Michaels and Amasino 2001; Mouradov et al. 2002). Subsequently, the floral pathway integrators activate the transcription of floral meristem identity genes, APETALA1 (AP1) and LEAFY (LFY), resulting in the initiation of floral meristems (Kobayashi and Weigel 2007; Liu et al. 2009a, b; Michaels 2009). The autonomous pathway genes promote flowering by repressing FLC expression (Michaels and Amasino 1999, 2001; Sheldon et al. 1999). The known components of the autonomous pathway are the four RNA-binding proteins FCA (Macknight et al. 1997), FPA (Schomburg et al. 2001), FLK (Lim et al. 2004) and FY (Simpson et al. 2003), and the two chromatin remodeling proteins FVE (Ausin et al. 2004) and FLD (He et al. 2003). To date, the involvement of cell cycle regulation of FLC expression in the autonomous pathway has yet to be reported. Here, we show that CDKG2 acts as a negative regulator of the salinity stress response and plant flowering transition. The loss-of-function cdkg2 mutant proved to be insensitive to salinity stress and early flowering, while the constitutive expression of CDKG2 induced a highly salinity-sensitive phenotype and late flowering. Moreover, we showed that CDKG2 mediated salt regulation of flowering. Combined studies of its putative downstream target genes shed light on the molecular basis of the uncovered new roles of CDKG2 protein in salt stress response and flowering time control.

13

Plant Mol Biol

Materials and methods Plant materials and growing conditions All T-DNA insertion mutants (SALK_012428, SALK_090262) used in this research were in the A. thaliana Col-0 background. The mutants were validated using PCR-based assays directed at both genomic DNA and mRNA. Seeds were surface-sterilized by immersion in 75 % ethanol for 3 min, followed by 95 % ethanol for 1 min. They were then blotted dry and plated on Murashige and Skoog basal medium (MS) plates containing 1 % w/v sucrose and 0.7 % w/v agar and transferred to the darkness for 3 days at 4 °C. Arabidopsis plants were grown on nutrition soil with vermiculite (nutrition soil:vermiculite, 2:1) in 22 ± 2 °C growth room. Longday (LD) condition is referred to as 16 h of light and 8 h of darkness and short-day (SD) condition as 8 h of light and 16 h of darkness. Transgenic plasmid constructs of CDGK2 and plant transformation The CDGK2 (At1g67580) open reading frame was amplified by PCR using the primer pair CDKG2-OE-F/-R (Supplementary Table 1), and the amplicon was inserted into the pEASY-Blunt vector. After sequence confirmation, the CDKG2 was ligated into the pSTART vector via the BamHI/SacI cloning site to generate the construct p35S::CDKG2 used for the overexpression in A. thaliana. To obtain pCDKG2::GUS, the CDKG2 promoter sequence, amplified using the primer pair CDKG2-GUS-F/-R (Supplementary Table 1), was inserted into the HindIII/BamHI cloning site of the binary vector pCambia-Ubi’Gus. The binary plasmids were transferred into the Agrobacterium tumefaciens strain EHA105 and transformed into A. thaliana Col-0 plant by means of the floral dip method (Clough and Bent 1998). T1 seedlings were plated on half strength MS agar plates containing 50 mg/L kanamycin to select for transformants, which were then selfed to allow for the selection of transgene homozygous T3 lines. Salinity stress treatment A germination assay was based on plating ~80 surface-sterilized seeds per line on half strength solidified MS medium containing either 0, 100 or 150 mM NaCl. The seeds were put at 4 °C for 3 days before transferring to 22 °C for germination. The percentage of successfully germinated individuals (defined by the emergence of the radicle) was recorded over time. To measure primary root growth following NaCl treatment, at least 20 seedlings were grown on vertically orientated plates containing 0, 100 or 150 mM

Plant Mol Biol

NaCl and the distance from the root tip to the hypocotyl base was measured after 7 days. Flowering time measurement The effect of photoperiod on flowering time was obtained from plants grown first on non-salinized solidified MS medium for 7 days, then potted into soil under either LD or SD conditions and ~70 % relative humidity. To examine vernalization effects, plants were germinated and grown for 6 weeks at 4 °C on solidified MS medium, then transferred to 21 ± 2 °C. To determine the effect of exogenously supplied gibberellin (GA), plants grown under LD conditions were sprayed with 20 μM GA twice a week until flowering. The effect of salinity was monitored by irrigating soilbased plants with water for 2 weeks, followed by two irrigations with 200 mM NaCl. Flowering time was measured as the days from sowing to floral bud formation and the total rosette numbers at flowering for each treatment. Each treatment involved at least 30 plants. Quantitative RT‑PCR (qPCR) Total RNA was isolated from 4-week-old seedlings of Col-0 and cdkg2 mutants with the TRIzol reagent (DBI Bioscience) and treated with RNase-free DNase I (Promega). The digested RNA was then reverse transcribed into cDNA using oligo (dT) primers and SuperScript™ III Reverse Transcriptase according to the manufacturer’s instructions (Invitrogen). The 10 μL qPCRs were based on SYBR Premix Ex Taq mix (Takara) and the reference sequence was a fragment of AtActin2. The sequences of all of the primers used are given in Supplementary Table 1. GUS staining Homozygous T3 generation of transgenic pCDGK2::GUS plants sampled at different growth stages were stained in a GUS staining solution containing 2 mM X-Gluc, 2 mM K3Fe(CN)6, 2 mM K4Fe(CN)6, 0.1 % Triton X-100 and 10 mM EDTA in 50 mM sodium phosphate buffer (pH 7.2) and incubated for 12 h at 37 °C after vacuum infiltration. Subsequently, samples were destained in 50, 70, 100 % ethanol for 5 min consecutively, and then bleached by immersion in 100 % ethanol. The decolorized tissues were observed under a bright field through a microscope (Olympus SZX16) and photographed using a digital camera (Olympus E-620). Sub‑cellular localization of CDKG2 The p35S::CDKG2-GFP construct was obtained by amplifying the CDKG2 open reading frame using the

primer pair CDKG2-GFP-F/-R (Supplementary Table 1), then inserting the product into the BamHI/SalI cloning site of CaMV35S-GFP (Liu et al. 2014). After purifying both p35S::CDKG2-GFP and p35S::GFP using a NucleoBond® Xtra Midi kit (MN), they were separately transformed into A. thaliana Col-0 mesophyll protoplasts as described (Sheen 2001). Following an overnight incubation in the dark, both GFP signals emitted by the transformed protoplasts and chlorophyll autofluorescence were monitored by confocal laser-scanning microscopy (LSM 700; Carl Zeiss) using the excitation wavelengths 488 and 647 nm, respectively. Bimolecular fluorescence complementation (BiFC) assay BiFC assays were performed in vivo as described (Waadt et al. 2008; Walter et al. 2004). The coding sequences of CDKG2 amplified using the primer pair CDKG2-GFP-F/-R (Supplementary Table 1) was inserted into the BamHI/SalI cloning site of pSPYNE, thus resulting in the plasmid YNE-CDKG2. The RCY1 coding sequence was amplified using the primer pair RCY1-F/-R (Supplementary Table 1) and inserted into the pSPYCE(M) via the BamHI/SmaI cloning site to generate the plasmid YCE(M)-RCY1. The two constructs were co-transferred into A. thaliana mesophyll protoplasts, and the YFP signal was captured by a confocal laser-scanning microscope (LSM 700; Carl Zeiss), as above.

Results Transcription of CDKG2 and sub‑cellular localization of CDKG2 Strong GUS activity was observed in the dry seed of the pCDKG2::GUS transgenic (Fig. 1a), but during its germination, the level of GUS expression decreased, and was limited to the radical (Fig. 1b, c). No GUS was detected in the cotyledons of early seedlings (Fig. 1d), but was present in the rosette leaf vasculature, the shoot apex, the lateral root initiation site and the root tip (Fig. 1e–h). During the reproductive stage, GUS was abundant in the sepal, stigma, anther and filament (Fig. 1i–k). Older siliques expressed more GUS activity than younger ones (Fig.  1l). Thus, CDKG2 appeared to be ubiquitously expressed during development, but especially in the seeds and flowers. When the p35S::CDKG2-GFP was transiently expressed in mesophyll protoplasts, GFP expression was confined to the nucleus (Fig. 1m). Previously TasRO1 was shown to be targeted to the nucleus (Liu et al. 2014). To confirm the nucleus location of CDKG2,

13



Plant Mol Biol

Fig. 1  GUS staining of transgenic lines carrying pCDKG2::GUS and subcellular localization of CDKG2. Seed prior to germination (a), germinating seed after imbibition for 72 h (b), 84 h (c) and 96 h (d), leaf of 10-day-old seedling (e), shoot apical meristem (f), lateral root (g), primary root (h), flower of 35-day-old plant (i–k), young and ripening siliques (l) were stained with X-gluc. m The transient expression of CDKG2-GFP in Arabidopsis mesophyll protoplasts. Images

were captured by a laser scanning confocal microscope using the following wavelengths: GFP (excitation, 488 nm; emission, 509 nm), and chlorophyll auto fluorescence (excitation, 448 nm; emission, 647 nm). n Colocalization of GFP-tagged CDKG2 and RFP-tagged nucleus marker protein TaSRO1. The yellow signal indicates areas of overlap between the green and red signals. Bars 10 μm

the p35S::CDKG2-GFP was co-transformed into protoplasts along with p35S::TaSRO1-RFP. As the GFP and RFP signals substantially overlapped one another (Fig. 1n), the conclusion was that CDKG2 is deposited in the nucleus, consistent with the known transcription profile of CDKG2 (http://suba.plantenergy.uwa.edu.au/flatfile.php?id=AT1G67580).

The contribution of CDKG2 to the salinity stress response

13

The data from protein interaction database indicates that CDKG2 might interact with RCY1 (Arginine-Rich Cyclin 1). To further validate the interaction between CDKG2 and RCY1 in their natural cellular context, we used the BiFC

Plant Mol Biol

approach. This is based on the reconstitution of a fluorescent complex from two fragments of YFP fluorescent protein when brought into close proximity via the interaction of two proteins fused to the fragments (Waadt et al. 2008). Protoplasts containing both YNE-CDKG2 and YCE(M)RCY1 constructs revealed strong YFP signal in the nucleus, whereas those containing either YNE-CDKG2 and YCE(M) or YCE(M)-RCY1 and YNE produced no fluorescence (Supplemental Fig. 1). These data suggest that CDKG2 interacts physically with RCY1 in living plant cells.

Constitutive expression of RCY1 confers salt tolerance to yeast and transgenic Arabidopsis plants (Forment et al. 2002). To study whether CDKG2 also plays a role in salinity stress response, we first transformed CDKG2 into yeast cells and grew under high salt conditions. Yeast cells harboring CDKG2 showed salt sensitivity phenotypes in the presence of 300 or 700 mM NaCl (Supplemental Fig. 2). To further investigate the function of CDKG2 in plants, we obtained two independent CDKG2 T-DNA insertion mutants SALK_012428 and SALK_090262, referred

Fig. 2  Molecular identification of loss-of-function and overexpression of CDKG2 plants and seed germination. a Schematic diagram of the CDKG2 gene structure and T-DNA insertion site. ATG and TGA indicate start and stop codons. The T-DNA was inserted in the sixth intron of the CDKG2 genomic DNA. Black boxes, solid lines, and striped boxes denote exons, introns, and untranslated regions, respectively. b PCR validation of the T-DNA mutants. The PCR product was amplified using the following primers: a LP + RP; b LP + LBb1.3; c RP + LBb1.3. c Expression profile of CDKG2 in wild-type and cdkg2 mutants. d RT-PCR analysis of CDKG2 transcript in the empty vector control (VC) and over-expression lines (OE). AtActin2 was used as an internal control. The germination rate of Col-0, cdkg2 mutants, VC and CDKG2 OE lines (OE18 and OE22) at indicated time under normal conditions (e), 100 mM NaCl (f) and 150 mM NaCl (g)

13



Plant Mol Biol

to as cdkg2-1 and cdkg2-2 respectively (Fig. 2a–c) and two overexpression transgenic lines (OE-1 and OE-2) (Fig. 2d). The ability of both cdkg2-1 and cdkg2-2 to seed germination was less severely inhibited by the presence of either 100 mM or 150 mM NaCl than that of the wild type at different time points (Fig. 2f). The opposite trend

was observed for CDKG2 OE lines compared with transgenic line carrying the vector control (VC) in the presence of NaCl (Fig. 2g). Furthermore, NaCl inhibited seedling growth and the root elongation of the seedlings were less severe in cdkg2 mutants than those in the wild type, and more severe in CDKG2 OE lines than those in the VC

Fig. 3  Transcript abundance of genes associated with the salinity stress response in A. thaliana. The seeds of Col-0 and cdkg2 mutants were put at 4 °C for 3 days before transferring to 22 °C for germination. Seeds germinated under non-stressed conditions (CK) and 150 mM NaCl for 5 days before RNA isolation. Transcript levels relative to that of AtActin2 are presented. The relative expression levels

of Col-0 under normal conditions were set at 1.0. Error bars represent mean ± SD from three technical replicates. The experiment was repeated once with similar results. The single asterisk, double asterisk represent significant difference determined by the Student’s t test at P