The Arabidopsis E3 Ubiquitin Ligase HOS1 Negatively Regulates CONSTANS Abundance in the Photoperiodic Control of Flowering

1 The E3 ubiquitin ligase HOS1 participates in the control of photoperiodic flowering in Arabidopsis negatively regulating CONSTANS abundance Ana Lazaro1, Federico Valverde2, Manuel Piñeiro1, Jose A. Jarillo1,3 1 Centro de Biotecnología y Genómica de Plantas (CBGP), INIA-UPM. Instituto Nacional de Investigación y Tecnología Agraria y Alimentaria, Campus de Montegancedo, 28223 Madrid, Spain. 2 Instituto de Bioquímica Vegetal y Fotosíntesis, Consejo Superior de Investigaciones Científicas y Universidad de Sevilla. Américo Vespucio 49, 41092 Sevilla, Spain. 3 To whom correspondence should be addressed. e-mail: [email protected] Telephone: 34-91-3364576; Fax 34-91-7157721 Running title: HOS1 regulates CO abundance Key words: HOS1, E3 ubiquitin ligase, CONSTANS, flowering time, floral repression, Arabidopsis. Estimated article length: 16,6 pages 2 SINOPSIS (400 characters long) The regulation of CONSTANS (CO) gene expression is crucial to measure accurately the changes in daylength that influence flowering time in Arabidopsis. CO is under both transcriptional and post-translational control. Here we demonstrate that ESD6/HOS1 is required to modulate precisely the timing of CO accumulation. This regulation is essential to maintain low levels of FT expression during the first part of the day and a correct photoperiodic response in Arabidopsis. 3 ABSTRACT The early in short days 6 (esd6) mutant was isolated in a screening devoted to isolate mutations that accelerate flowering time in Arabidopsis. Among other developmental alterations, esd6 displays early flowering in both long and short day conditions. Fine mapping of the mutation showed that esd6 was affected in the HIGH EXPRESSION OF OSMOTICALLY RESPONSIVE GENES 1 (HOS1) locus, which encodes a RING finger-containing protein that works as an E3 ubiquitin ligase. esd6/hos1 mutation causes decreased expression of the FLC gene and shows a strong requirement of a functional CO protein for its early flowering phenotype under long days. Besides, CO and HOS1 physically interact in vitro and in planta, and HOS1 is regulating CO abundance, particularly during the daylight period. Accordingly, hos1 mutation causes a shift in the regular long day pattern of expression of FT transcript, starting to rise four hours after dawn. In addition, HOS1 interacts synergistically with COP1, another regulator of CO protein stability, in the control of flowering time. Taken together, these results indicate that HOS1 is involved in the control of CO abundance, ensuring that CO activation of FT occurs only when the light period reaches a certain length and preventing precocious flowering in Arabidopsis. 4 INTRODUCTION The integration of complex signals from environmental and endogenous cues is necessary to enable plants to time the floral transition at the most advantageous moment (Michaels, 2009; de Montaigu et al., 2010; Imaizumi, 2010). Plants growing at northern latitudes adapt their developmental program to the varying daylengths and temperatures that occur along the year (Jackson, 2009). Arabidopsis is a facultative long-day (LD) plant in which flowering time is controlled by a network of six major pathways: information about daylength, low winter temperatures and growth temperature are mediated by the photoperiod, the vernalization and the ambient temperature pathway, respectively. In contrast, the aging, the autonomous and the gibberellin pathways act more independently of ambient conditions (Fornara et al., 2010). A potent repressor of flowering, FLOWERING LOCUS C (FLC), integrates signals coming from both the vernalization and the autonomous pathway (Amasino, 2010). Eventually, the whole network converges in the regulation of the floral integrators: FLOWERING LOCUS T (FT), TWIN SISTER OF FT (TSF) and SUPPRESSOR OF OVEREXPRESSION OF CONSTANS 1 (SOC1) (Fornara et al., 2010). The photoperiod pathway comprises several genes, including GIGANTEA (GI), CONSTANS (CO), and FT (Kobayashi and Weigel, 2007; Turck et al., 2008). Mutations in any of these genes cause a delay in flowering mainly under LDs, whereas their overexpression accelerates flowering independently of daylength (Turck et al., 2008). CO is a B-box-type protein that acts in the vascular tissue of the leaves to activate FT and TSF transcription (An et al., 2004; Jackson, 2009; Tiwari et al., 2010). CO may induce FT expression by forming a DNA binding complex with NUCLEAR FACTOR Y (NF-Y)/HEME ACTIVATOR PROTEIN (HAP) proteins (Wenkel et al., 2006; Kumimoto et al., 2010) or ASYMMETRIC LEAVES 1 (AS1) (Song et al., 2011), and 5 by binding the FT promoter directly at CO-responsive elements (Tiwari et al., 2010), which are functional cis-elements required for FT expression (Adrian et al., 2010). FT protein, and possibly TSF, are part of the florigen that moves to the shoot apical meristem to induce flowering in response to LDs (Corbesier et al., 2007; Jaeger and Wigge, 2007; Jang et al., 2009). Plants have developed a sophisticated molecular mechanism to measure daylength based on the coincidence of an internal rhythm, set by the circadian clock, with an external cue, such as light. The ability to distinguish LDs from short days (SDs) is largely the result of the complex regulation of CO, both at the transcriptional and posttranslational level and may have arisen very early in plant evolution (Serrano et al., 2009; Valverde, 2011). Under LDs, CO mRNA shows two peaks of expression, the first following the expression of GI at the end of a LD, when plants are still exposed to light; and the second during the night. Under SDs, only the night peak of CO expression takes place (Suarez-Lopez et al., 2001). The precise timing of CO also requires the degradation of a family of repressors, the CYCLING DOF FACTORS (CDFs), by the F-Box proteins FLAVIN-BINDING, KELCH REPEAT, F-BOX 1 (FKF1), ZEITLUPE (ZTL) and LOV KELCH PROTEIN 2 (LKP2), in conjunction with GI (Imaizumi et al., 2005; Sawa et al., 2007; Fornara et al., 2009). The increased expression of CO in the light under LDs but not SDs is crucial for the promotion of flowering, because exposure to light is required for stabilization of CO protein (Valverde et al., 2004; Jang et al., 2008). The high CO transcript levels detected during the dark phase of both LD and SDs do not correlate with CO protein accumulation because the RING finger protein CONSTITUTIVE PHOTOMORPHOGENIC 1 (COP1) promotes CO degradation in the dark (Jang et al., 2008; Liu et al., 2008b). Mutations in COP1, encoding a component of an ubiquitin 6 ligase complex, cause extreme early flowering under SDs. This early flowering phenotype is largely dependent on CO activity and correlates with an increase in FT transcription in the cop1 mutant (Jang et al., 2008; Liu et al., 2008b). COP1 and CO interact both in vivo and in vitro, and it has been proposed that COP1 contributes to daylength perception by reducing the abundance of the CO protein during the night (Jang et al., 2008; Liu et al., 2008b; Chen et al., 2010). However, in the morning, CO degradation occurs independently of COP1 (Jang et al., 2008). Therefore, it has been suggested that an unidentified E3 ubiquitin ligase must collaborate in CO degradation during the early part of the day to ensure that CO induction of FT only takes place in LDs (Jang et al., 2008). In addition to the length of the daily light/dark periods, plants also perceive light quality. Blue and far-red light promote flowering, while red light (RL) delays it (Valverde et al., 2004). Far-red light can increase CO protein levels independently of transcription (Kim et al., 2008). Blue light mediates photoperiodic control of the floral initiation at least by three different mechanisms: first, it promotes the interaction of FKF1 and GI necessary for the CDFs degradation (Sawa et al., 2007); second, the blue light receptor Cryptochrome 2 (Cry2) prevents GI and CO proteolysis by COP1 (Liu et al., 2008b; Yu et al., 2008) and a blue light-dependent interaction of CRY2 with SPA1 regulates COP1 activity (Zuo et al., 2011) ; and third, Cry2 modulates FT transcription directly (Liu et al., 2008a). On the other hand, the red light photoreceptor Phytochrome B (PhyB) has been implicated in the degradation of CO during the first part of the day (Valverde et al., 2004; Jang et al., 2008). Screenings devoted to the isolation of early flowering mutants have revealed the existence of genes that repress the floral transition (Pouteau et al., 2004). Floral repressors are essential to safeguard against premature flowering, and knowledge of 7 how these repressors interact with the floral promotion pathways is just emerging (Pouteau et al., 2004; Roux et al., 2006). Here we demonstrate that the esd6 early flowering mutant is affected in the HOS1 gene. HOS1 encodes a protein with E3 ubiquitin ligase activity, previously described as a regulator of cold acclimation responses (Lee et al., 2001; Dong et al., 2006a). The early flowering phenotype of hos1 is completely suppressed by mutations in CO gene in Landsberg erecta (Ler) background and notably delayed by co mutations in Columbia (Col) background. Moreover, we show that HOS1 physically interacts with CO in vitro and in planta and regulates CO protein abundance during the daylight period, indicating the participation of another RING finger-containing protein, in addition to COP1, in the photoperiodic control of flowering time in Arabidopsis. Thus, we propose that HOS1 is required to modulate precisely the timing of CO accumulation, and that this regulation is essential to maintain low levels of FT during the first part of the day and, subsequently, a correct photoperiodic response in Arabidopsis. 8 METHODS 1. Genetic stocks and growth conditions Arabidopsis thaliana (L.) mutant seed stocks used were in Ler, Col and C24 genetic backgrounds and were obtained from the Arabidopsis Biological Resource Centre (ABRC) of Ohio State University (Columbus, USA), the Nottingham Arabidopsis Centre (NASC) in the UK, and personal donations. C24 accession and mutant hos1-1 seeds were kindly donated by Dr. J.K. Zhu (Lee et al., 2001). The monogenic mutants used in this work were described previously: fca-1, ft-1, co-2 and gi-3 (Koornneef, 1991); fve-3 (Ausin et al., 2004); flc-3 (Michaels and Amasino, 1999); phyB-1 (Reed et al., 1993); fha-1 (Guo et al., 1998); vrn1-2 fca-1 (Levy et al., 2002); vin3-4 FRI Sf-2 (Sung and Amasino, 2004); fld-1 (He et al., 2003); siz1-2 (Miura et al., 2005); fkf1-1 (Nelson et al., 2000); cop1-4 (Deng et al., 1991); co-10 (Laubinger et al., 2006); soc1-1 (Samach et al., 2000); and the Col FRI Sf-2 line was described by Lee and Amasino (Lee and Amasino, 1995). Plants were grown in plastic pots containing a mixture of substrate and vermiculite (3:1) or in MS (Murashige and Skoog) medium supplemented with 1% (w/v) sucrose and 0.8% (w/v) agar for in vitro culture. Controlled environmental conditions were provided in growth chambers at 22°C and 70% relative humidity. Plants were illuminated with cool-white fluorescent lights (approximately 120 μmol m–2 s–1). LD conditions consisted of 16 h of light followed by 8 h of darkness; SD conditions consisted of 8 h of light followed by 16 h of darkness. 2. Phenotypic analysis Total leaf number was scored as the number of leaves in the rosette (excluding cotyledons) plus the number of leaves in the inflorescence at the time of opening of the first flower (Koornneef, 1991). Cauline, adult and juvenile leaves were scored 9 independently. Rosette leaves lacking abaxial trichomes were considered as juvenile leaves (Telfer et al., 1997). Data are shown as mean ± s.e.m. Root length was measured at different developmental stages in seedlings grown in MS medium supplemented with 1% (w/v) sucrose and 1% (w/v) plant agar in Petri dishes placed vertically. Total chlorophyll content (Ct) was calculated as described by Moran (Moran, 1982). 3. Map-based cloning of esd6 mutation and molecular characterization of the hos1 alleles A mapping population was generated from the crossing of the esd6 mutant, in Ler background, and a Col wild-type plant. The analysis of 550 early flowering plants with several polymorphic molecular markers (Supplemental Table 1) located the esd6 mutation to the upper arm of chromosome 2, between markers C005 and T5I7. Mutations hos1-1, in C24 background, hos1-2, in Ler background, and hos1-4, in Col background, generate premature stop codons in the seventh, fifth and first exon of the HOS1 locus respectively. The T-DNA insertion mutant hos1-3, isolated in Col background, was obtained from NASC (SALK_069312). 4. Genetic analysis Double mutants were constructed by crossing the monogenic hos1 mutants with lines carrying the mutations flc-3, fca-1, fve-3, fld-1, siz1-2, fha-1, gi-3, co-2, co-10, cop1-4, fkf1-1, ft-1, soc1-1, vrn1-2 or vin3-4. Double mutants were isolated from selfed F2 progeny using molecular markers. A dCAP marker was designed for the hos1-2 mutation (PCR amplification using 5´-TTTTTACATGGCCGGTTCAGATC-3´ and 5´- GCAATGTAATGTGAAACTAGGCGA-3´ primers followed by BglII digestion). For the hos1-3 mutation we used 5´-GGTTTCTGGACCGCATATTTC-3´, 5´- GGCTTCTGACCAGAGAGTGTT-3´ and the SALK LB1 primer. hos1-3 was also 16 identified the line SALK_069312, which carried an insertion in the fifth exon of HOS1 gene (Figure 2B). This T-DNA mutant allele was named hos1-3 and, an additional allele, hos1-4 was obtained later on in our laboratory during the screening of an EMS- mutagenized population of Col plants. hos1-4 mutation created a single nucleotide deletion in the position 88 of the HOS1 genomic annotation, which generated a premature stop codon in the first exon of the gene (Figure 2B). All hos1 alleles analysed display an early flowering phenotype both in LD and SD photoperiods, but the fact that they flower earlier under inductive photoperiods, indicates that the mutation does not completely abolish the plant photoperiod response (Figure 2C and Table 1). The AtHOS1 gene is around 5.5 Kb long, bears 9 exons and encodes a protein of 915 amino acids that contains a non-canonical RING finger domain in the N-terminal region and a putative nuclear localization signal in the C-terminal part (Figure 2B and Supplemental Figure 2). AtHOS1 is a unique gene in Arabidopsis and putative orthologues have only been found in plants. The cysteine residues present in the RING finger domain are totally conserved between all AtHOS1 orthologues (Supplemental Figure 2). RING finger domains are found in proteins with E3 ubiquitin ligase activity that participate in the ubiquitin/26S proteasome pathway (Deshaies and Joazeiro, 2009; Vierstra, 2009). Previously, it has been described that AtHOS1 can function as an E3 ubiquitin ligase in ubiquitination assays (Dong et al., 2006a). hos1 mutations affect FLC expression and have an FLC-independent effect in the control of flowering time The early flowering phenotype of hos1 mutants suggested that HOS1 could be a negative regulator of the floral transition in Arabidopsis. To test this hypothesis, we analyzed the phenotype of double mutants carrying hos1 and different mutations 17 affecting flowering time. It had been previously described that FLC expression levels were reduced in the hos1-1 mutant compared to C24 accession (Lee et al., 2001). In order to check if the hos1 early flowering phenotype was fully dependent on FLC, the effect of hos1-3 mutation in an flc null genetic background (flc-3) (Michaels and Amasino, 1999) was tested. Both hos1-3 and hos1-3 flc-3 double mutant plants flowered with the same number of leaves, irrespectively of photoperiodic conditions, although the hos1-3 flc-3 plants bolted consistently earlier than hos1-3 under LD (Figure 3A and Table 1). This result may indicate that there is no additional effect of flc null mutation on the acceleration of flowering time caused by hos1. Besides, both hos1- 3 and hos1-3 flc-3 plants flowered clearly earlier than flc-3 plants under both LD and SD conditions (Figure 3A and Table 1), indicating that the effect of the hos1 mutation on flowering time could not be exclusively dependent on FLC activity, and that there is an FLC-independent effect responsible for the early flowering phenotype of hos1. To find out whether FLC expression was altered in the hos1 mutant alleles isolated in different backgrounds, semiquantitative RT-PCR analyses were performed in hos1-1, hos1-2, and hos1-3 mutants and the corresponding wild-type genotypes. Consistently with previous results (Lee et al., 2001), in all hos1 mutants assayed, FLC transcript was clearly down-regulated (Figure 3B), and therefore it cannot be ruled out that this change in FLC expression has an effect on the early flowering time of the hos1 alleles. Dominant alleles of the FRIGIDA (FRI) gene confer a vernalization requirement that delays flowering through the up-regulation of FLC (Johanson et al., 2000). In order to find out the genetic relationship between HOS1 and FRI, the mutant hos1-3 was crossed with a Col plant bearing an active FRI allele introgressed from the San Feliu-2 (Sf-2) accession (Lee and Amasino, 1995). Under LD conditions, the hos1-3 FRI Sf-2 line showed an additive phenotype, the FRI late-flowering phenotype being only 18 partially suppressed by hos1-3 (Figure 3D upper panel and Table 1). This suggests that HOS1 and FRI do not regulate FLC expression through the same pathway in LDs. In contrast, the hos1-3 FRI Sf-2 plant flowered with approximately the same number of leaves as the Col FRI Sf-2 plants under SD conditions, abolishing the effect of the hos1 mutation (Figure 3D lower panel and Table 1). Because HOS1 locus is involved in cold signal transduction (Lee et al., 2001) and vernalization regulates FLC expression (Amasino, 2010), we hypothesized that HOS1 could be controlling FLC transcript levels through the vernalization pathway. To analyze this, we generated combinations between hos1 and two other mutants impaired in the vernalization response such as vernalization 1 (vrn1) and vernalizationinsensitive 3 (vin3) (Levy et al., 2002; Sung and Amasino, 2004), both in late flowering backgrounds that allowed us to observe the acceleration of flowering due to the vernalization treatment. The hos1 mutation did not impair the acceleration of flowering caused by vernalization when combined with the late flowering fca-1 or FRI Sf-2 plants (Supplemental Table 2). Besides, no difference in flowering time was found for the hos1-2 vrn1-2 fca-1 triple mutant grown after either 1 or 4 weeks of vernalization treatment (Supplemental Table 2). The same result was observed for hos1-3 vin3-4 carrying an active FRI allele, as both 1 and 4 week-vernalized plants flowered with approximately the same number of leaves (Supplemental Table 2). Thus, we concluded that HOS1 does not regulate FLC expression through the vernalization pathway. Considering that the autonomous pathway also converges on the regulation of FLC expression, the flowering phenotype of double mutants combining hos1 and mutations in representative autonomous pathway genes was analysed, in particular the hos1-3 fve- 3, hos1-2 fca-1 and hos1-3 fld-1 double mutants. Under LD, these double mutant plants showed an additive phenotype because the late-flowering phenotype of autonomous 19 pathway mutants was only partially suppressed by hos1 (Figure 3C upper panel and Table 1). In contrast, under SDs, flowering time of these double mutants was very similar to the one displayed by the autonomous pathway mutants, as they produced only a few leaves less than fve-3, fca-1 and fld-1 respectively (Figure 3C lower panel and Table 1). Altogether, these results indicate that the hos1 mutation cannot accelerate flowering in SD when combined with genetic backgrounds that have very high FLC expression levels, such as mutations of the autonomous pathway or active alleles of FRI. In contrast, under LD the repressive effect of HOS1 on flowering time may be mediated by additional pathways that remain inactive in SD. The E3 SUMO ligase SIZ1 promotes FLC expression by repressing the autonomous pathway gene FLD (Jin et al., 2008). Besides, SIZ1 stabilizes the ICE1 protein, which has been implicated in the regulation of freezing tolerance in Arabidopsis (Miura et al., 2007). Because it had been described that ICE1 was also targeted by HOS1 (Dong et al., 2006a), we checked the genetic relationship existing between hos1 and siz1 mutants. Flowering time of siz1 plants relative to wild-type was slightly earlier under LDs, and substantially earlier under SDs (Jin et al., 2008) (Table 1). When we combined siz1-2 with the hos1-2 mutation, the double mutant flowering time resembled that of hos1-2 in LDs but was earlier than any of the parental lines in SDs, suggesting a synergistic genetic interaction between both loci (Table 1). The early flowering phenotype of hos1-2 requires a functional CO gene We also analyzed the phenotype of double mutants carrying hos1 and mutations in genes representative of the photoperiod pathway, such as CRY2/FHA, GI and CO, that delay flowering mainly under LDs (Koornneef et al., 1998). While hos1-2 fha-1 and 20 hos1-2 gi-3 double mutants showed an additive flowering phenotype between hos1-2 and fha-1 and gi-3 late flowering mutants, the genetic interaction observed between hos1-2 and co-2 was completely different (Figure 4 and Table 1). Under LDs, the hos1- 2 mutation did not accelerate flowering time when it was combined with co-2 (Figure 4C and Table 1); indeed, hos1-2 co-2 plants flowered with the same number of leaves as co-2 mutant, indicating a strong requirement of a functional CO gene for the early flowering phenotype of hos1-2. However, under SD conditions hos1-2 co-2 flowered as early as hos1-2 (Table 1), given that co mutations do not delay flowering under this photoperiodic condition. These genetic results suggest that HOS1 is involved in the photoperiodic control of flowering time as a negative regulator of CO under LDs. FKF1 is an F-Box protein (Imaizumi et al., 2005) that mediates the cyclic degradation of CDF proteins, which are repressors of CO expression (Imaizumi et al., 2005; Fornara et al., 2009). To study if there was any genetic interaction between FKF1 and HOS1, the double mutant hos1-3 fkf1-1was analyzed and it showed an additive phenotype between the late flowering time of fkf1-1 and the early flowering phenotype of hos1-3 in LDs (Table 1). This result indicates that HOS1 does not participate in the FKF1 transcriptional regulatory pathway that controls CO expression. hos1 mutants show an altered pattern of FT expression FLC acts repressing the expression of the floral integrators FT and SOC1, while the photoperiod pathway activate FT and SOC1 expression through CO (Yoo et al., 2005; Searle et al., 2006; Turck et al., 2008). Because hos1 mutations showed downregulation of FLC expression (Figure 3B) and the co-2 mutation was epistatic to hos1-2 under LDs (Figure 4C), we decided to check the genetic relationship between HOS1 and the floral integrators FT and SOC1. The hos1-2 ft-1 double mutant showed a similar flowering 21 phenotype to ft-1 under LD conditions, suggesting a strong requirement of FT by the hos1 early flowering phenotype (Figure 5A and Table 1). In contrast, hos1-2 soc1-1 double mutant was additive between both parental lines in both LD and SD conditions (Figure 5A and Table 1). This result is in accordance with the epistasis observed between co-2 and hos1-2, considering that FT is the main target of CO under LDs (Yoo et al., 2005). In order to check whether the whole effect of HOS1 on flowering time was through FT and SOC1, we generated the triple mutant hos1-2 ft-1 soc1-1. Flowering time analysis showed that the triple mutant was slightly earlier than ft-1soc1-1 (Figure 5A and Table 1). This result indicates that the early flowering phenotype of hos1 mutation requires functional FT and SOC1 genes, although we cannot rule out that HOS1 could regulate other protein(s) involved in the control of flowering time. In order to deepen into the genetic relationship observed between HOS1 and the photoperiod pathway, we performed a time course expression analysis over a 24h period in Ler and hos1-2 seedlings both in LD and SD. First of all, we demonstrated that HOS1 transcript did not show a diurnal oscillation in Ler background (Figure 5B). Subsequently, we analyzed the temporal expression pattern of CO, FT and SOC1 genes using semiquantitative RT-PCR and quantitative real-time PCR (Q-PCR) approaches (Fig 5C-E). As previously reported, CO transcript level in Ler background was high at dawn and dusk, and during the night, remaining low for the rest of the light period of the day (Suarez-Lopez et al., 2001) (Fig 5C-E). In hos1-2, the same pattern of CO expression was observed, indicating that hos1 mutation did not affect significantly the levels or the expression profile of the CO transcript (Figure 5C-E). However, the expression pattern of the floral integrator FT was clearly altered in the hos1-2 mutant compared to the wild-type both in LD and SD conditions (Fig 5C-E). FT transcript usually shows a peak of expression at dusk in LDs (around ZT16), following the 22 evening increase observed in CO mRNA. In the hos1-2 mutant, a peak of FT expression in the subjective morning, mainly at ZT4, but also at ZT8 was observed, when the CO transcript levels are barely detectable in the mutant (Figure 5C and D). To check whether this alteration was due to a specific developmental stage of the plant or if it relied on the genetic background, FT transcript levels were analysed at ZT4 and ZT8 in Col and hos1-3 plants harvested 8, 10, 12 and 15 days after germination (DAG). In every single stage tested, FT expression was higher in hos1-3 in relation to Col in the first part of the day (Supplemental Figure 3). In SD, an increased FT expression in the hos1-2 mutant, starting to rise at ZT8 and peaking at ZT12 was observed, which may explain the early flowering phenotype displayed by the mutant under non-inductive photoperiods (Figure 5C and E). A small but consistent increase in SOC1 expression was also detected in hos1-2 plants grown in LD photoperiods in comparison to the wildtype (Figure 5C and D). Thus, we conclude that CO transcript levels are not modified substantially by the hos1 mutation and that HOS1 is required to repress the expression of FT in the first part of the day in LD. Considering that HOS1 has an E3 ubiquitin ligase activity, we speculated that it may be involved in the degradation of protein(s) that regulate FT expression. Both the genetic analysis and the expression assays suggested that this protein could be CO. HOS1 interacts in vitro and in vivo with CO and regulates its abundance Given the proposed epistatic interaction between co-2 and hos1-2 mutants and considering that CO transcript levels were not affected in the hos1-2 mutant, we decided to analyze whether there was a physical interaction between CO and HOS1. For this purpose, in vitro pull-down experiments using MBP-HOS1 and in vitro-translated CO protein were performed. As shown in Figure 6A, MBP-HOS1, but not MBP alone, was 23 able to interact with CO protein. Whether the interaction between CO and HOS1 also occurred in vivo was tested using the bimolecular fluorescence complementation (BiFC) technique. For that, the N terminus of YFP was cloned upstream of HOS1 (YFN-HOS1) and the C terminus of YFP was fused to CO (YFC-CO). By Agrobacterium tumefaciens coinfiltration, these constructs were transiently expressed in abaxial epidermal cells of tobacco and Arabidopsis leaves (Voinnet et al., 2003). Reconstitution of YFP fluorescence was examined by confocal microscopy two days after transient coexpression of the protein pairs. Yellow fluorescence in the nucleus was detected for coexpression of YFC-CO and YFN-HOS1, while no yellow fluorescence was observed when YFC-CO was coexpressed with YFN-AKIN β, or when YFC-AKIN10 was coexpressed with YFN-HOS1, as negative controls (Figure 6B). As a positive control, the interaction between amino and carboxy parts of AKINβ and AKIN10 SnRKs proteins (Ferrando et al., 2001), was tested (Figure 6B). CO-HOS1 interaction could be observed in conspicuous nuclear speckles, which have been often associated with foci of proteasome degradation, as previously described for the interaction between CO and COP1 (Jang et al., 2008). These results demonstrate that CO and HOS1 colocalize and physically interact in the nuclei of plant cells. The interaction between HOS1 and CO proteins was further explored by in vivo co-immunoprecipitation analysis. For this experiment, we generated 35S::HOS1-GFP transgenic plants in Col background and we also introduced this construct in 35S::CO plants. Nuclear proteins isolated from two-week-old 35S::HOS1-GFP 35S::CO and 35S::HOS1-GFP plants, grown in LD and harvested at ZT4 in the presence of MG132 proteasome inhibitor, were immunoprecipitated after incubation with an anti-GFP antibody (Figure 6C). The eluted purified proteins were blotted and detected with anti- CO antibody. As shown in Figure 6C, CO antibody recognized a clear signal in nuclear 24 extracts from 35S::HOS1-GFP 35S::CO (Elution HC, marked by an arrow) plants but not in control extracts from the 35S::HOS1-GFP (Elution H) plants. Given their apparent molecular mass and the comigration with the input CO protein (marked with an arrow), which included 1/50 protein amount of the immunoprecipitated fraction (input HC), the recognized bands must correspond to CO protein. In this blot, a higher molecular mass CO form (asterisk, upper panel) could also be detected. Because HOS1 has been described as a RING-finger E3 ligase (Dong et al., 2006a), the same blot was probed with anti–Ubiquitin antibody, resulting in only the upper immunoprecipitated band being recognized (asterisk, lower panel). This result demonstrates that a fraction of the immunoprecipitated CO protein was ubiquitinated in vivo, further supporting the association between both proteins and pointing to a role of HOS1 in the proteasomedependent degradation of CO. It has been reported that HOS1 has auto-ubiquitination ligase activity in vitro and that it can also mediate the ubiquitination and degradation of ICE1 transcription factor (Dong et al., 2006a). To further analyze whether HOS1 may also regulate CO degradation in vivo, a construct constitutively expressing CO fused to LUC was transformed into wild-type Arabidopsis plants. One representative line, 35S::CO–LUC 6-2, displaying an early flowering phenotype, was introduced into hos1-3 by genetic crossing. The 35S::CO–LUC 6-2/hos1-3 plants flowered earlier than either the hos1-3 mutant or the 35S::CO–LUC 6-2 plants, indicating that CO–LUC construct was fully functional (Supplemental Table 3). Using luciferase fluorescence in vivo imaging, we found that under LDs the CO protein levels were significantly lower in the wild-type than in the hos1 mutant background three hours after dawn (ZT3) (Figure 6D), suggesting that the degradation of CO that occurs in the wild-type is impaired in the hos1 mutant. Quantification of luciferase activity corroborated that CO protein 25 accumulated to higher levels in the hos1 mutant than in the wild-type plants (Figure 6E). This accumulation of CO protein observed at ZT3 correlates with the early peak of FT expression present in the hos1 mutant (Figure 5C and D and Supplemental Figure 3). To further assess the role of HOS1 in CO regulation, we performed western blot assays to detect CO protein in nuclear extracts from wild-type and hos1-3 plants grown under LD photoperiods. In these immunoblots, CO protein was present at lower abundance in the wild-type than in the hos1-3 mutant plants, particularly during the daylight period (Figure 6F). From these data, we conclude that HOS1 is involved in the photoperiodic regulation of flowering through the modulation of CO protein levels in vivo. HOS1 is a nuclear-localized protein that acts in the phloem to regulate photoperiodic flowering HOS1 protein is ubiquitously expressed in all plant tissues (Lee et al., 2001). Computer analysis of the HOS1 amino acid sequence predicted a nuclear localization signal in the C-terminus of the protein (Supplemental Figure 2). Previous reports localized HOS1 into the cytoplasm of transgenic Arabidopsis seedlings overexpressing a HOS1-GFP construct, grown under dark conditions at normal growth temperature. However, in response to low temperature treatments, HOS1-GFP accumulated in the nucleus (Lee et al., 2001). To determine whether the subcellular localization of HOS1 was altered by light/dark conditions, we overexpressed a HOS1-GFP construct in the hos1-3 mutant. The homozygous line 35S::HOS1-GFP/hos1-3 4-1-4 showed a delay in flowering time when compared with hos1-3, indicating that the fusion protein was functional in the repression of flowering (Figure 7 E). Subsequently, we grew 35S::HOS1-GFP/hos1-3 4-1-4 transgenic plants at 22ºC under both continuous light and dark, and analyzed GFP fluorescence in root cells by confocal microscopy. As 32 genes independently of CBFs, and both LOV1 and SFR6 control flowering time through the photoperiod pathway. Other mutants, such as low expression of osmotically responsive genes 4 (los4) and atnup160, display an early flowering phenotype and altered CBF expression levels (Gong et al., 2005; Dong et al., 2006b). On the other hand, co and gi photoperiod pathway mutants show altered tolerance to freezing temperatures (Cao et al., 2005; Yoo et al., 2007), and fve mutant flowers late and shows increased expression of FLC and CBF genes (Kim et al., 2004). It has been proposed very recently that in warm late spring SOC1 downregulates CBFs expression and promotes flowering, while in cold early spring or fall, induction of FLC expression by the CBFs delays flowering and confers cold resistance to the plant (Seo et al., 2009). Besides regulating CO stability, HOS1 controls FLC expression (Figure 3B) (Lee et al., 2001), which is also repressed by prolonged exposure to cold temperatures (Amasino, 2010). The positive effect of HOS1 on FLC expression appears to be independent of the vernalization pathway (Bond et al., 2011) (Supplemental Table 2) and is still uncharacterised. Taken together, these results suggest that HOS1, among other genes, may participate in the photoperiod and temperature signal crosstalk, integrating information coming from both pathways and facilitating a proper response to changing environmental conditions. To our knowledge this is the first E3 ubiquitin ligase proposed to integrate both environmental signals, specifically targeting for degradation key factors involved in the regulation of each response. Further studies will be necessary for an in-depth understanding of how these pathways modulate each other´s activity to optimize plant adaptation. 33 BIBLIOGRAPHY Adrian, J., Farrona, S., Reimer, J.J., Albani, M.C., Coupland, G., and Turck, F. (2010). cis-Regulatory elements and chromatin state coordinately control temporal and spatial expression of FLOWERING LOCUS T in Arabidopsis. Plant Cell 22, 1425-1440. Amasino, R. (2010). Seasonal and developmental timing of flowering. Plant J 61, 1001- 1013. An, H., Roussot, C., Suarez-Lopez, P., Corbesier, L., Vincent, C., Pineiro, M., Hepworth, S., Mouradov, A., Justin, S., Turnbull, C., and Coupland, G. (2004). CONSTANS acts in the phloem to regulate a systemic signal that induces photoperiodic flowering of Arabidopsis. Development 131, 3615-3626. Ausin, I., Alonso-Blanco, C., Jarillo, J.A., Ruiz-Garcia, L., and Martinez-Zapater, J.M. (2004). Regulation of flowering time by FVE, a retinoblastoma-associated protein. Nat Genet 36, 162-166. Bond, D.M., Dennis, E.S., and Finnegan, E.J. (2011). The low temperature response pathways for cold acclimation and vernalization are independent. Plant Cell Environ 34, 1737-1748. Cao, S., Ye, M., and Jiang, S. (2005). Involvement of GIGANTEA gene in the regulation of the cold stress response in Arabidopsis. Plant Cell Rep 24, 683- 690. Cao, Y., Dai, Y., Cui, S., and Ma, L. (2008). Histone H2B monoubiquitination in the chromatin of FLOWERING LOCUS C regulates flowering time in Arabidopsis. Plant Cell 20, 2586-2602. Corbesier, L., Vincent, C., Jang, S., Fornara, F., Fan, Q., Searle, I., Giakountis, A., Farrona, S., Gissot, L., Turnbull, C., and Coupland, G. (2007). FT protein movement contributes to long-distance signaling in floral induction of Arabidopsis. Science 316, 1030-1033. Chen, H., Huang, X., Gusmaroli, G., Terzaghi, W., Lau, O.S., Yanagawa, Y., Zhang, Y., Li, J., Lee, J.H., Zhu, D., and Deng, X.W. (2010). Arabidopsis CULLIN4-Damaged DNA Binding Protein 1 Interacts with CONSTITUTIVELY PHOTOMORPHOGENIC1-SUPPRESSOR OF PHYA Complexes to Regulate Photomorphogenesis and Flowering Time. Plant Cell 22, 108-123. Chiang, G.C., Barua, D., Kramer, E.M., Amasino, R.M., and Donohue, K. (2009). Major flowering time gene, FLOWERING LOCUS C, regulates seed germination in Arabidopsis thaliana. Proc Natl Acad Sci U S A 106, 11661- 11666. de Montaigu, A., Toth, R., and Coupland, G. (2010). Plant development goes like clockwork. Trends Genet 26, 296-306. del Olmo, I., Lopez-Gonzalez, L., Martin-Trillo, M.M., Martinez-Zapater, J.M., Pineiro, M., and Jarillo, J.A. (2010). EARLY IN SHORT DAYS 7 (ESD7) encodes the catalytic subunit of DNA polymerase epsilon and is required for flowering repression through a mechanism involving epigenetic gene silencing. Plant J 61, 623-636. Deng, X.W., Caspar, T., and Quail, P.H. (1991). cop1: a regulatory locus involved in light-controlled development and gene expression in Arabidopsis. Genes Dev 5, 1172-1182. Deshaies, R.J., and Joazeiro, C.A. (2009). RING domain E3 ubiquitin ligases. Annu Rev Biochem 78, 399-434. 34 Dong, C.H., Agarwal, M., Zhang, Y., Xie, Q., and Zhu, J.K. (2006a). The negative regulator of plant cold responses, HOS1, is a RING E3 ligase that mediates the ubiquitination and degradation of ICE1. Proc Natl Acad Sci U S A 103, 8281- 8286. Dong, C.H., Hu, X., Tang, W., Zheng, X., Kim, Y.S., Lee, B.H., and Zhu, J.K. (2006b). A putative Arabidopsis nucleoporin, AtNUP160, is critical for RNA export and required for plant tolerance to cold stress. Mol Cell Biol 26, 9533- 9543. Ferrando, A., Koncz-Kalman, Z., Farras, R., Tiburcio, A., Schell, J., and Koncz, C. (2001). Detection of in vivo protein interactions between Snf1-related kinase subunits with intron-tagged epitope-labelling in plants cells. Nucleic Acids Res 29, 3685-3693. Fornara, F., de Montaigu, A., and Coupland, G. (2010). SnapShot: Control of flowering in Arabidopsis. Cell 141, 550 e551-552. Fornara, F., Panigrahi, K.C., Gissot, L., Sauerbrunn, N., Ruhl, M., Jarillo, J.A., and Coupland, G. (2009). Arabidopsis DOF transcription factors act redundantly to reduce CONSTANS expression and are essential for a photoperiodic flowering response. Dev Cell 17, 75-86. Gong, Z., Dong, C.H., Lee, H., Zhu, J., Xiong, L., Gong, D., Stevenson, B., and Zhu, J.K. (2005). A DEAD box RNA helicase is essential for mRNA export and important for development and stress responses in Arabidopsis. Plant Cell 17, 256-267. Guo, H., Yang, H., Mockler, T.C., and Lin, C. (1998). Regulation of flowering time by Arabidopsis photoreceptors. Science 279, 1360-1363. He, Y., Michaels, S.D., and Amasino, R.M. (2003). Regulation of flowering time by histone acetylation in Arabidopsis. Science 302, 1751-1754. Imaizumi, T. (2010). Arabidopsis circadian clock and photoperiodism: time to think about location. Curr Opin Plant Biol 13, 83-89. Imaizumi, T., Schultz, T.F., Harmon, F.G., Ho, L.A., and Kay, S.A. (2005). FKF1 F-box protein mediates cyclic degradation of a repressor of CONSTANS in Arabidopsis. Science 309, 293-297. Imlau, A., Truernit, E., and Sauer, N. (1999). Cell-to-cell and long-distance trafficking of the green fluorescent protein in the phloem and symplastic unloading of the protein into sink tissues. Plant Cell 11, 309-322. Jackson, S.D. (2009). Plant responses to photoperiod. New Phytol 181, 517-531. Jaeger, K.E., and Wigge, P.A. (2007). FT protein acts as a long-range signal in Arabidopsis. Curr Biol 17, 1050-1054. Jang, S., Torti, S., and Coupland, G. (2009). Genetic and spatial interactions between FT, TSF and SVP during the early stages of floral induction in Arabidopsis. Plant J 60, 614-625. Jang, S., Marchal, V., Panigrahi, K.C., Wenkel, S., Soppe, W., Deng, X.W., Valverde, F., and Coupland, G. (2008). Arabidopsis COP1 shapes the temporal pattern of CO accumulation conferring a photoperiodic flowering response. EMBO J 27, 1277-1288. Jin, J.B., Jin, Y.H., Lee, J., Miura, K., Yoo, C.Y., Kim, W.Y., Van Oosten, M., Hyun, Y., Somers, D.E., Lee, I., Yun, D.J., Bressan, R.A., and Hasegawa, P.M. (2008). The SUMO E3 ligase, AtSIZ1, regulates flowering by controlling a salicylic acid-mediated floral promotion pathway and through affects on FLC chromatin structure. Plant J 53, 530-540. 35 Johanson, U., West, J., Lister, C., Michaels, S., Amasino, R., and Dean, C. (2000). Molecular analysis of FRIGIDA, a major determinant of natural variation in Arabidopsis flowering time. Science 290, 344-347. Kim, H.J., Hyun, Y., Park, J.Y., Park, M.J., Park, M.K., Kim, M.D., Lee, M.H., Moon, J., Lee, I., and Kim, J. (2004). A genetic link between cold responses and flowering time through FVE in Arabidopsis thaliana. Nat Genet 36, 167- 171. Kim, S.Y., Yu, X., and Michaels, S.D. (2008). Regulation of CONSTANS and FLOWERING LOCUS T expression in response to changing light quality. Plant Physiol 148, 269-279. Knight, H., Thomson, A.J., and McWatters, H.G. (2008). SENSITIVE TO FREEZING 6 integrates cellular and environmental inputs to the plant circadian clock. Plant Physiol 148, 293-303. Kobayashi, Y., and Weigel, D. (2007). Move on up, it's time for change mobile signals controlling photoperiod-dependent flowering. Genes Dev 21, 2371-2384. Koornneef, M. (1991). Isolation of higher plant developmental mutants. Symp Soc Exp Biol 45, 1-19. Koornneef, M., Alonso-Blanco, C., Peeters, A.J., and Soppe, W. (1998). Genetic Control of Flowering Time in Arabidopsis. Annu Rev Plant Physiol Plant Mol Biol 49, 345-370. Kumimoto, R.W., Zhang, Y., Siefers, N., and Holt, B.F., 3rd. (2010). NF-YC3, NFYC4 and NF-YC9 are required for CONSTANS-mediated, photoperioddependent flowering in Arabidopsis thaliana. Plant J 63, 379–391. Laubinger, S., Marchal, V., Le Gourrierec, J., Wenkel, S., Adrian, J., Jang, S., Kulajta, C., Braun, H., Coupland, G., and Hoecker, U. (2006). Arabidopsis SPA proteins regulate photoperiodic flowering and interact with the floral inducer CONSTANS to regulate its stability. Development 133, 3213-3222. Lazaro, A., Gomez-Zambrano, A., Lopez-Gonzalez, L., Pineiro, M., and Jarillo, J.A. (2008). Mutations in the Arabidopsis SWC6 gene, encoding a component of the SWR1 chromatin remodelling complex, accelerate flowering time and alter leaf and flower development. J Exp Bot 59, 653-666. Lee, H., Xiong, L., Gong, Z., Ishitani, M., Stevenson, B., and Zhu, J.K. (2001). The Arabidopsis HOS1 gene negatively regulates cold signal transduction and encodes a RING finger protein that displays cold-regulated nucleo-cytoplasmic partitioning. Genes Dev 15, 912-924. Lee, I., and Amasino, R.M. (1995). Effect of Vernalization, Photoperiod, and Light Quality on the Flowering Phenotype of Arabidopsis Plants Containing the FRIGIDA Gene. Plant Physiol 108, 157-162. Levy, Y.Y., Mesnage, S., Mylne, J.S., Gendall, A.R., and Dean, C. (2002). Multiple roles of Arabidopsis VRN1 in vernalization and flowering time control. Science 297, 243-246. Lincoln, C., Long, J., Yamaguchi, J., Serikawa, K., and Hake, S. (1994). A knotted1-like homeobox gene in Arabidopsis is expressed in the vegetative meristem and dramatically alters leaf morphology when overexpressed in transgenic plants. Plant Cell 6, 1859-1876. Liu, H., Yu, X., Li, K., Klejnot, J., Yang, H., Lisiero, D., and Lin, C. (2008a). Photoexcited CRY2 interacts with CIB1 to regulate transcription and floral initiation in Arabidopsis. Science 322, 1535-1539. Liu, L.J., Zhang, Y.C., Li, Q.H., Sang, Y., Mao, J., Lian, H.L., Wang, L., and Yang, H.Q. (2008b). COP1-Mediated Ubiquitination of CONSTANS Is 36 Implicated in Cryptochrome Regulation of Flowering in Arabidopsis. Plant Cell 20, 292-306. Martin-Trillo, M., Lazaro, A., Poethig, R.S., Gomez-Mena, C., Pineiro, M.A., Martinez-Zapater, J.M., Jarillo, J.A., (Print), -., Article, J., and Research Support, N.-U.S.G.t. (2006). EARLY IN SHORT DAYS 1 (ESD1) encodes ACTIN-RELATED PROTEIN 6 (AtARP6), a putative component of chromatin remodelling complexes that positively regulates FLC accumulation in Arabidopsis. Development 133, 1241-1252. Michaels, S.D. (2009). Flowering time regulation produces much fruit. Curr Opin Plant Biol 12, 75-80. Michaels, S.D., and Amasino, R.M. (1999). FLOWERING LOCUS C encodes a novel MADS domain protein that acts as a repressor of flowering. Plant Cell 11, 949- 956. Miura, K., Jin, J.B., Lee, J., Yoo, C.Y., Stirm, V., Miura, T., Ashworth, E.N., Bressan, R.A., Yun, D.J., and Hasegawa, P.M. (2007). SIZ1-Mediated Sumoylation of ICE1 Controls CBF3/DREB1A Expression and Freezing Tolerance in Arabidopsis. Plant Cell 19, 1403-1414. Miura, K., Rus, A., Sharkhuu, A., Yokoi, S., Karthikeyan, A.S., Raghothama, K.G., Baek, D., Koo, Y.D., Jin, J.B., Bressan, R.A., Yun, D.J., and Hasegawa, P.M. (2005). The Arabidopsis SUMO E3 ligase SIZ1 controls phosphate deficiency responses. Proc Natl Acad Sci U S A 102, 7760-7765. Moran, R. (1982). Formulae for determination of chlorophyllous pigments extracted with N,N-dimethylformamide. Plant Physiol 69, 1376-1381. Morris, K., Thornber, S., Codrai, L., Richardson, C., Craig, A., Sadanandom, A., Thomas, B., and Jackson, S. (2010). DAY NEUTRAL FLOWERING represses CONSTANS to prevent Arabidopsis flowering early in short days. Plant Cell 22, 1118-1128. Nelson, D.C., Lasswell, J., Rogg, L.E., Cohen, M.A., and Bartel, B. (2000). FKF1, a clock-controlled gene that regulates the transition to flowering in Arabidopsis. Cell 101, 331-340. Park, B.S., Eo, H.J., Jang, I.C., Kang, H.G., Song, J.T., and Seo, H.S. (2010). Ubiquitination of LHY by SINAT5 regulates flowering time and is inhibited by DET1. Biochem Biophys Res Commun 398, 242-246. Pineiro, M., Gomez-Mena, C., Schaffer, R., Martinez-Zapater, J.M., and Coupland, G. (2003). EARLY BOLTING IN SHORT DAYS is related to chromatin remodeling factors and regulates flowering in Arabidopsis by repressing FT. Plant Cell 15, 1552-1562. Pouteau, S., Ferret, V., Gaudin, V., Lefebvre, D., Sabar, M., Zhao, G., and Prunus, F. (2004). Extensive phenotypic variation in early flowering mutants of Arabidopsis. Plant Physiol 135, 201-211. Reed, J.W., Nagpal, P., Poole, D.S., Furuya, M., and Chory, J. (1993). Mutations in the gene for the red/far-red light receptor phytochrome B alter cell elongation and physiological responses throughout Arabidopsis development. Plant Cell 5, 147-157. Roux, F., Touzet, P., Cuguen, J., and Le Corre, V. (2006). How to be early flowering: an evolutionary perspective. Trends Plant Sci 11, 375-381. Samach, A., Onouchi, H., Gold, S.E., Ditta, G.S., Schwarz-Sommer, Z., Yanofsky, M.F., and Coupland, G. (2000). Distinct roles of CONSTANS target genes in reproductive development of Arabidopsis. Science 288, 1613-1616. 37 Sawa, M., Nusinow, D.A., Kay, S.A., and Imaizumi, T. (2007). FKF1 and GIGANTEA Complex Formation is Required for Day-Length Measurement in Arabidopsis. Science 318, 261-265. Searle, I., He, Y., Turck, F., Vincent, C., Fornara, F., Krober, S., Amasino, R.A., and Coupland, G. (2006). The transcription factor FLC confers a flowering response to vernalization by repressing meristem competence and systemic signaling in Arabidopsis. Genes Dev 20, 898-912. Seo, E., Lee, H., Jeon, J., Park, H., Kim, J., Noh, Y.S., and Lee, I. (2009). Crosstalk between Cold Response and Flowering in Arabidopsis Is Mediated through the Flowering-Time Gene SOC1 and Its Upstream Negative Regulator FLC. Plant Cell 21, 3185-3197. Serrano, G., Herrera-Palau, R., Romero, J.M., Serrano, A., Coupland, G., and Valverde, F. (2009). Chlamydomonas CONSTANS and the Evolution of Plant Photoperiodic Signaling. Curr Biol. Simon, R., Igeno, M.I., and Coupland, G. (1996). Activation of floral meristem identity genes in Arabidopsis. Nature 384, 59-62. Song, Y.H., Lee, I., Lee, S.Y., Imaizumi, T., and Hong, J.C. (2011). CONSTANS and ASYMMETRIC LEAVES 1 complex is involved in the induction of FLOWERING LOCUS T in photoperiodic flowering in Arabidopsis. Plant J 69, 332-342. Suarez-Lopez, P., Wheatley, K., Robson, F., Onouchi, H., Valverde, F., and Coupland, G. (2001). CONSTANS mediates between the circadian clock and the control of flowering in Arabidopsis. Nature 410, 1116-1120. Sung, S., and Amasino, R.M. (2004). Vernalization in Arabidopsis thaliana is mediated by the PHD finger protein VIN3. Nature 427, 159-164. Tamura, K., Fukao, Y., Iwamoto, M., Haraguchi, T., and Hara-Nishimura, I. (2010). Identification and characterization of nuclear pore complex components in Arabidopsis thaliana. Plant Cell 22, 4084-4097. Telfer, A., Bollman, K.M., and Poethig, R.S. (1997). Phase change and the regulation of trichome distribution in Arabidopsis thaliana. Development 124, 645-654. Tiwari, S.B., Shen, Y., Chang, H.C., Hou, Y., Harris, A., Ma, S.F., McPartland, M., Hymus, G.J., Adam, L., Marion, C., Belachew, A., Repetti, P.P., Reuber, T.L., and Ratcliffe, O.J. (2010). The flowering time regulator CONSTANS is recruited to the FLOWERING LOCUS T promoter via a unique cis-element. New Phytol 187, 57-66. Turck, F., Fornara, F., and Coupland, G. (2008). Regulation and identity of florigen: FLOWERING LOCUS T moves center stage. Annu Rev Plant Biol 59, 573-594. Valverde, F. (2011). CONSTANS and the evolutionary origin of photoperiodic timing of flowering. J Exp Bot 62, 2453-2463. Valverde, F., Mouradov, A., Soppe, W., Ravenscroft, D., Samach, A., and Coupland, G. (2004). Photoreceptor regulation of CONSTANS protein in photoperiodic flowering. Science 303, 1003-1006. Vega-Sanchez, M.E., Zeng, L., Chen, S., Leung, H., and Wang, G.L. (2008). SPIN1, a K homology domain protein negatively regulated and ubiquitinated by the E3 ubiquitin ligase SPL11, is involved in flowering time control in rice. Plant Cell 20, 1456-1469. Vierstra, R.D. (2009). The ubiquitin-26S proteasome system at the nexus of plant biology. Nat Rev Mol Cell Biol 10, 385-397. 38 Voinnet, O., Rivas, S., Mestre, P., and Baulcombe, D. (2003). An enhanced transient expression system in plants based on suppression of gene silencing by the p19 protein of tomato bushy stunt virus. Plant J 33, 949-956. Wenkel, S., Turck, F., Singer, K., Gissot, L., Le Gourrierec, J., Samach, A., and Coupland, G. (2006). CONSTANS and the CCAAT box binding complex share a functionally important domain and interact to regulate flowering of Arabidopsis. Plant Cell 18, 2971-2984. Wilczek, A.M., Roe, J.L., Knapp, M.C., Cooper, M.D., Lopez-Gallego, C., Martin, L.J., Muir, C.D., Sim, S., Walker, A., Anderson, J., Egan, J.F., Moyers, B.T., Petipas, R., Giakountis, A., Charbit, E., Coupland, G., Welch, S.M., and Schmitt, J. (2009). Effects of genetic perturbation on seasonal life history plasticity. Science 323, 930-934. Yoo, S.K., Chung, K.S., Kim, J., Lee, J.H., Hong, S.M., Yoo, S.J., Yoo, S.Y., Lee, J.S., and Ahn, J.H. (2005). CONSTANS activates SUPPRESSOR OF OVEREXPRESSION OF CONSTANS 1 through FLOWERING LOCUS T to promote flowering in Arabidopsis. Plant Physiol 139, 770-778. Yoo, S.Y., Kim, Y., Kim, S.Y., Lee, J.S., and Ahn, J.H. (2007). Control of flowering time and cold response by a NAC-domain protein in Arabidopsis. PLoS ONE 2, e642. Yu, J.W., Rubio, V., Lee, N.Y., Bai, S., Lee, S.Y., Kim, S.S., Liu, L., Zhang, Y., Irigoyen, M.L., Sullivan, J.A., Zhang, Y., Lee, I., Xie, Q., Paek, N.C., and Deng, X.W. (2008). COP1 and ELF3 control circadian function and photoperiodic flowering by regulating GI stability. Mol Cell 32, 617-630. Zhu, J., Shi, H., Lee, B.H., Damsz, B., Cheng, S., Stirm, V., Zhu, J.K., Hasegawa, P.M., and Bressan, R.A. (2004). An Arabidopsis homeodomain transcription factor gene, HOS9, mediates cold tolerance through a CBF-independent pathway. Proc Natl Acad Sci U S A 101, 9873-9878. Zuo, Z., Liu, H., Liu, B., Liu, X., and Lin, C. (2011). Blue Light-Dependent Interaction of CRY2 with SPA1 Regulates COP1 activity and Floral Initiation in Arabidopsis. Curr Biol 21, 841-847. 39 ACKNOWLEDGEMENTS We thank Dr. I. Ausín (University of California, Los Angeles, USA) for isolating the esd6 mutant, Dr. R. Catalá (CIB, CSIC, Madrid, Spain) for providing a Gatewaycompatible HOS1 full length cDNA, Dr. J. Salinas (CIB, CSIC, Madrid, Spain) for granting the access to the luciferase imaging system, Dr. P. González-Melendi (CBGP, UPM-INIA, Madrid, Spain) for technical support with confocal microscope and Dr. JC. del Pozo and members of his lab (CBGP, UPM-INIA, Madrid, Spain) for helpful discussions. pYFN43 and pYFC43 vectors, and SNF1 protein kinase constructs, used as positive controls in the BiFC assays, were kindly provided by Dr. A. Ferrando (IBMCP, Valencia, Spain). C24 and hos1-1 mutant seeds, as well as a HOS1 expression construct in the pMAL vector, were kindly provided by Dr. J.K Zhu (University of California, Riverside, California, USA). The 35S::CO construct, the plasmids containing the SUC2 and KNAT1 promoters and co-10 cop1-4 seeds were kindly donated by Dr. G. Coupland (Max Planck Institute, Cologne, Germany), while the 35S::CO-LUC plasmid was kindly provided by Dr. H-Q Yang (Shanghai Institute for Biological Sciences, Chinese Academy of Science, Shanghai, China). This work was supported by grants BIO2008- 00351 to JAJ, BIO2007-61215 to MP, CSD2007-00057 and BIO2010-15589 to JAJ and MP, and BIO2007-61837 and BIO2010-16027 to FV from the Spanish Ministerio de Ciencia e Innovación. 40 FIGURE LEGENDS Figure 1. Phenotypic characterization of the esd6 mutant. (A,B) Flowering time phenotype of Ler and esd6 plants grown in LD conditions for 23 days (A) or in SD conditions for 54 days (B). (C) Phenotype of Ler and esd6 plants grown in LD conditions for 35 days (D) Histograms comparing the number of juvenile, adult and cauline leaves in Ler and esd6 plants grown under both LD and SD photoperiods. (E) Rosette and cauline leaves of Ler and esd6 plants grown in LDs. (F) Detached Ler and esd6 flowers and siliques from plants grown under LD conditions. (G) Root elongation in 11 day-old Ler and esd6 seedlings. (H) Total chlorophyll content (Ct) in Ler and esd6 and phyB-1 mutant seedlings. Figure 2. Identification of ESD6. (A) Map-based cloning of ESD6. The genetic interval and the bacterial artificial chromosome (BAC) clones in the genomic region surrounding ESD6 are shown. The number of recombinant events between molecular markers is given in parentheses. The position of HOS1 ORF in T5I7 BAC clone is indicated. (B) Scheme of the ESD6/HOS1 gene structure showing the polymorphisms associated to the different hos1 mutant alleles isolated. Exons are represented by squared boxes, while introns are drawn by a line. (C) Pictures illustrating the flowering time of hos1 mutants and their respective wild-type genotypes in LD and SD conditions. Plants were grown for 23 days under LD conditions (upper panel). SD pictures (lower panel) were taken after 64 days for C24 and hos1-1, 58 days for Ler and hos1-2 and 60 days for Col and hos1-3. Figure 3. hos1 mutations downregulate FLC expression and have an FLC- independent effect in the control of flowering time. (A) Flowering time phenotype of hos1-3 flc-3 double mutant in LD (upper panel) and SD (lower panel) conditions. (B) Analysis of the expression of FLC in 14 day-old hos1 mutant seedlings and their corresponding wild-type genotypes. FLC expression was monitored by semiquantitative RT-PCR analysis over 22 cycles for C24 and hos1-1 and over 28 cycles for Ler, hos1-2, Col and hos1-3. For the UBQ10 control 22 cycles were used. (C) Flowering time phenotype of hos1-3 fve-3 double mutant plants grown in LD (upper panel) and SD (lower panel) conditions. (D) Flowering time phenotype of hos1-3 plants bearing an active allele of FRI in LDs (upper panel) and SDs (lower panel). 41 Figure 4. Genetic analyses between hos1 and mutations in photoperiod pathway genes. Flowering time phenotype of hos1-2 fha-1 (A), hos1-2 gi-3 (B) and hos1-2 co-2 (C) double mutants grown in LD conditions. Figure 5. The early flowering phenotype of hos1 depends on FT and SOC1 functional proteins and hos1 mutation alters the pattern of expression of FT. (A) Flowering time phenotype of hos1-2 soc1-1, hos1-2 ft-1 and hos1-2 ft-1 soc1-1 triple mutant plants grown in LD conditions. (B) HOS1 expression pattern over a 24h time course in Ler seedlings grown for 8 days in LDs and for 16 days in SDs. Samples were harvested every 4h after dawn. HOS1 expression was monitored by semiquantitative RT-PCR analysis over 20 cycles. (C, D, E) Expression analysis of different flowering time genes over a 24h time course in Ler and hos1-2 seedlings grown for 8 days in LDs and 16 days in SDs. Samples were harvested every 4h after dawn (C) Semiquantitative RT-PCR analysis comparing CO (22 cycles), FT (28 cycles) and SOC1 (24 cycles) expression (D) Quantitative realtime PCR (Q-PCR) analysis of CO, FT and SOC1 expression in LD conditions (E) Same as (D) but Ler and hos1-2 seedlings grown in SD conditions. Relative expression levels were normalized to β - ACTIN expression. Figure 6. HOS1 interacts with CO and regulates its abundance. (A) HOS1 and CO interact in vitro. A Coomassie stained SDS-PAGE showing MBP (42KDa) and MBP-HOS1 fusion protein (147KDa) expressed in Escherichia coli BL21 Rosetta strain and purified on amylose resin is shown in the upper panel. The lower panel shows the result of a pull-down assay with MBP and MBP-HOS1 proteins incubated with 35S-Met-labelled CO protein. Retained CO protein was visualized after autoradiography of the dried gel. (B) HOS1 interacts with CO in planta. BiFC assay coexpressing the C terminus of YFP fused to CO (CO-YFC) and the N terminus of YFP to HOS1 (YFN-HOS1) in tobacco and Arabidopsis leaves. Yellow fluorescence in the nucleus was indicative of interaction. Negative and positive controls were included in the assay. (C) In vivo coimmunoprecipitation between HOS1 and CO. GFP-HOS1 protein was immunoprecipitated from 35S::CO 35S::GFP-HOS1 (HC) and 35S::GFPHOS1 (H) plants employing GFP antibody-agarose columns and detected with anti-CO A Col hos1-3 flc-3 hos1-3 flc-3 C Col hos1-3 fve-3 hos1-3 fve-3 UBQ10 FLC B Col hos1-3 Col FRI hos1-3 FRI D LD SD SD LD SD LD Figure 3. hos1 mutations downregulate FLC expression and have an FLC-independent effect in the control of flowering time. (A) Flowering time phenotype of hos1-3 flc-3 double mutant in LD (upper panel) and SD (lower panel) conditions. (B) Analysis of the expression of FLC in 14 day-old hos1 mutant seedlings and their corresponding wild-type genotypes. FLC expression was monitored by semiquantitative RT-PCR analysis over 22 cycles for C24 and hos1-1 and over 28 cycles for Ler, hos1-2, Col and hos1-3. For the UBQ10 control 22 cycles were used. (C) Flowering time phenotype of hos1-3 fve-3 double mutant plants grown in LD (upper panel) and SD (lower panel) conditions. (D) Flowering time phenotype of hos1-3 plants bearing an active allele of FRI in LDs (upper panel) and SDs (lower panel). Ler hos1-2 fha-1 hos1-2 fha-1 A Ler hos1-2 gi-3 hos1-2 gi-3 B Ler hos1-2 co-2 hos1-2 co-2 C LD LD LD Figure 4. Genetic analyses between hos1 and mutations in photoperiod pathway genes. Flowering time phenotype of hos1-2 fha-1 (A), hos1-2 gi-3 (B) and hos1-2 co-2 (C) double mutants grown in LD conditions. Figure 5. The early flowering phenotype of hos1 depends on FT and SOC1 functional proteins and hos1 mutation alters the pattern of expression of FT. (A) Flowering time phenotype of hos1-2 soc1-1, hos1-2 ft-1 and hos1-2 ft-1 soc1-1 triple mutant plants grown in LD conditions. (B) HOS1 expression pattern over a 24h time course in Ler seedlings grown for 8 days in LDs and for 16 days in SDs. Samples were harvested every 4h after dawn. HOS1 expression was monitored by semiquantitative RT-PCR analysis over 20 cycles. (C, D, E) Expression analysis of different flowering time genes over a 24h time course in Ler and hos1-2 seedlings grown for 8 days in LDs and 16 days in SDs. Samples were harvested every 4h after dawn (C) Semiquantitative RT-PCR analysis comparing CO (22 cycles), FT (28 cycles) and SOC1 (24 cycles) expression (D) Quantitative realtime PCR (Q-PCR) analysis of CO, FT and SOC1 expression in LD conditions (E) Same as (D) but Ler and hos1-2 seedlings grown in SD conditions. Relative expression levels were normalized to β -ACTIN expression. C Ler hos1-2 Ler hos1-2 FT UBQ10 CO SOC1 hos1-2 ft-1 soc1-1 HOS1 B UBQ10 HOS1 UBQ10 Ler Ler hos1-2 ft-1 hos1-2 ft-1 soc1-1 hos1-2 soc1-1 ft-1 soc1-1 hos1-2 ft-1 soc1-1 ALD D E 0,0 0,2 0,4 0,6 0,8 1,0 1,2 FT/ACT 0,0 0,2 0,4 0,6 0,8 1,0 1,2 SOC1/ACT 0,0 0,2 0,4 0,6 0,8 1,0 1,2 SOC1/ACT Ler hos1-2 CO/ACT 0,0 0,2 0,4 0,6 0,8 1,0 1,2 CO/ACT FT/ACT 0,0 0,2 0,4 0,6 0,8 1,0 1,2 1,4 0,0 0,2 0,4 0,6 0,8 1,0 1,2 1,4 Figure 6. HOS1 interacts with CO and regulates its abundance. (A) HOS1 and CO interact in vitro. A Coomassie stained SDS-PAGE showing MBP (42KDa) and MBP-HOS1 fusion protein (147KDa) expressed in Escherichia coli BL21 Rosetta strain and purified on amylose resin is shown in the upper panel. The lower panel shows the result of a pull-down assay with MBP and MBP-HOS1 proteins incubated with 35S-Met-labelled CO protein. Retained CO protein was visualized after autoradiography of the dried gel. (B) HOS1 interacts with CO in planta. BiFC assay coexpressingthe C terminus of YFP fused to CO (CO-YFC) and the N terminus of YFP to HOS1 (YFN-HOS1) in tobacco and Arabidopsis leaves. Yellow fluorescence in the nucleus was indicative of interaction. Negative and positive controls were included in the assay. (C) In vivo coimmunoprecipitationbetween HOS1 and CO. GFPHOS1 protein was immunoprecipitatedfrom 35S::CO 35S::GFP-HOS1 (HC) and 35S::GFP-HOS1(H) plants employing GFP antibody-agarosecolumns and detected with anti-CO antibodies (Elution HC, upper panel). The arrow marks both the co-immunoprecipitatedCO protein and input. Antibodies against ubiquitinrecognized, in the same blot, a higher molecular mass band (asterisk) corresponding to an ubiquitinatedCO form (Elution HC, lower panel). Controls included input (1/50) from HC and H samples as well as column flowthrough and washes. Protein markers are shown on the right. (D) Non invasive in vivo luciferase imaging of 35S::CO–LUC 6-2 and 35S::CO–LUC 6-2/hos1-3seedlings. Pictures show 7 dayold seedlings grown in LDs 3 h after the lights are on. At this time, more CO-LUC protein accumulates in hos1 mutant (below right) than in the wild-type (below left). (E) Quantification of the luciferase activity in 35S::CO–LUC 6- 2 (blue bars) and 35S::CO–LUC 6-2/hos1-3 (grey bars) seedlings expressed as luciferase counts per second (LCPS) in serial dilutions of fresh tissue in Steadylite Plus Reagent (mg/ml). (F) Immunoblot showing CO protein levels during a 24 h time course in nuclear extracts from Col and hos1-3 plants grown under LDs. Numbers above each lane represent hours after dawn that the sample was harvested. Histone H3 was used as a loading control. Relative quantification of each band compared to the control is expressed below the upper panel ( α -CO). YFC –CO YFN-HOS1 YFC-CO YFN-HOS1 Arabidopsis YFC –CO YFN-AKIN β YFC –AKIN 10 YFN-HOS1 YFC –AKIN 10 YFN-AKIN β A 0 20 40 60 80 100 120 140 50mg/ml 25mg/ml 12,5mg/ml LCPS (x103) 35S::CO–LUC 6-2 35S::CO–LUC 6-2/ hos1-3 E F αH3 αCO hos1-3 Col 0.5 3 6 9 12 22 0.5 3 6 9 12 22 ZT (h) 7,1 5,6 2,7 2,5 2,1 1 0.2 1,3 0,3 0,7 5,2 0,9 BTransmitted light YFP Chlorophyl fluorescence Merged 35S::CO–LUC 6-2/hos1-3 35S::CO–LUC 6-2 Col D Tobacco 147 MBP MBPHOS1 42 input MBP MBP-HOS1 35S-Met CO KDa C α CO *58 46 Input Flowthrough Wash Elution MWM HC H HC H HC H HC H KDa *58 46 α Ub Figure 7. HOS1 is a nuclear-localized protein that exerts an effect in the phloem to regulate photoperiodic flowering Localization of HOS1-GFP in the root cells of 10-day old 35S::HOS1-GFP/hos1-3plants analysed under confocal microscopy. (A) Plants grown under continuous light. (B) Plants grown in darkness. (C) A representative nuclear image of a light-grown seedling showing DAPI staining (upper panel), GFP fluorescence (middle panel) and the merge of both images (lower panel). Tissue specific expression of HOS1 in phloem companion cells and in the shoot apical meristem of hos1 transgenic plants. (D) Flowering time phenotype of Col, hos1-3, SUC2::HOS1/hos1-3 and KNAT1::HOS1/hos1-3 plants in LD. (E) Quantification of flowering time of Col, hos1-3, 35S::HOS1-GFP/hos1-3, SUC2::HOS1/hos1-3 and KNAT1::HOS1/hos1-3 plants grown in LD conditions. Total number of leaves was scored in approximately 10 plants and is represented as mean ± s.e.m. ABC DAPI GFP channel Merged Col hos1-3 pSUC2-HOS1/ hos1-3 pKNAT1-HOS1/ hos1-3 D 0 4 8 12 16 n leaves E A Col hos1-3 cop1-4 hos1-3 cop1-4 Col hos1-3 Col 35S::CO C Col hos1-3 cop1-4 hos1-3 cop1-4 co-10 hos1-3 co-10 hos1-3 cop1-4 co-10 cop1-4 co-10 B hos1-3 35S::CO SD LD LD LD Figure 8. Genetic interaction between HOS1 and COP1 in the control of flowering time. (A) Flowering time phenotype of hos1-3 cop1-4 double mutant in LD (upper panel) and SD (lower panel) conditions. (B) Flowering time phenotype of double and triple mutant combinations between hos1-3 co-10 cop1-4 mutants grown in LD conditions. (C) Comparison of flowering time phenotype between LD-grown Col and hos1-3 plants bearing a 35S::CO transgene. Figure 9. Model for HOS1 function in the photoperiodic control of flowering time. The transcription of CO gene depends primarily on the circadian clock (thick black line). In the evening, the degradation of CDFs by the GI/FKF1 complex allows CO transcript levels to increase, and CO protein accumulates due to a photoreceptor-mediated repression of COP1. At this time CO can promote FT expression and induce flowering. During the night, COP1 activity causes rapid degradation of CO protein by the ubiquitination/26S proteasome system. In the daylight period HOS1 is required to degrade CO. Additional data will be necessary to establish the possible involvement of HOS1 in the mechanism of CO degradation mediated by PhyB that has been proposed to operate in the morning. CO mRNA PhyB Cry2 HOS1 COP1 CO CO CO CO CO CO CO CO CO CO CO COP1 Supplemental Figure 1. Flower, silique and root length measurement in Ler and esd6. Data were scored on an average of 10 plants grown under LD conditions. Ler esd6 0,2 0,6 1,0 1,4 Silique length (cm) 1 3 5 5 DAG 10DAG 15DAG Ler esd6 Root length (cm) 0 1 2 3 4 5 Ler esd6 Flower length (mm) Supplemental Figure 2. Sequence comparison of Arabidopsis HOS1 (AtHOS1) with the HOS1 orthologues from Ricinus communis (RcHOS1), Populus trichocarpa (PtHOS1), Citrus trifoliate (CtHOS1), Vitis vinifera (VvHOS1), Oryza sativa (OsHOS1) and Physcomytrella patents (PpHOS1). Amino acid residues in black are identical, and those in grey are functionally similar in the sequences. The conserved cysteine residues in the RING finger domain are indicated with black triangles. The putative nuclear localization signal present in the C terminal region of HOS1 is underlined. Genbank accession numbers are NP_181511 for AtHOS1, XP_002531460 for RcHOS1, XP_002304293 for PtHOS1, ACY92092 for CtHOS1, XP_002264751 for VvHOS1, AAR07079 for OsHOS1 and XP_00178023 5 for PpHOS1. ZT 48 FT UBQ10 48 48 48 810 12 15 hos1-3 Col hos1-3 Col hos1-3 Col hos1-3 Col hos1-3 Col hos1-3 Col hos1-3 Col hos1-3 Col Supplemental Figure 3. FT expression analysis in LD grown Col and hos1-3 seedlings harvested 4 and 8 h after dawn at 8, 10, 12 and 15 DAG.