Transcript
Page 1: The RING-Finger Ubiquitin Ligase HAF1 Mediates Heading date 1 …croplab.hzau.edu.cn/__local/8/44/09/7F493FBF4BE3C2A36DFF... · 2019-04-22 · The RING-Finger Ubiquitin Ligase HAF1

The RING-Finger Ubiquitin Ligase HAF1 MediatesHeading date 1 Degradation during PhotoperiodicFlowering in RiceOPEN

Ying Yang,1 Debao Fu,1 Chunmei Zhu, Yizhou He, Huijun Zhang, Tao Liu, Xianghua Li, and Changyin Wu2

National Key Laboratory of Crop Genetic Improvement, National Center of Plant Gene Research (Wuhan), Huazhong AgriculturalUniversity, Wuhan 430070, China

The photoperiodic response is one of the most important factors determining heading date in rice (Oryza sativa). Althoughrhythmic expression patterns of flowering time genes have been reported to fine-tune the photoperiodic response,posttranslational regulation of key flowering regulators has seldom been elucidated in rice. Heading date 1 (Hd1) encodesa zinc finger transcription factor that plays a crucial role in the photoperiodic response, which determines rice regionaladaptability. However, little is known about the molecular mechanisms of Hd1 accumulation during the photoperiodresponse. Here, we identify a C3HC4 RING domain-containing E3 ubiquitin ligase, Heading date Associated Factor 1 (HAF1),which physically interacts with Hd1. HAF1 mediates ubiquitination and targets Hd1 for degradation via the 26S proteasome-dependent pathway. The haf1mutant exhibits a later flowering heading date under both short-day and long-day conditions. Inaddition, the haf1 hd1 double mutant headed as late as hd1 plants under short-day conditions but exhibited a heading datesimilar to haf1 under long-day conditions, thus indicating that HAF1 may determine heading date mainly through Hd1 undershort-day conditions. Moreover, high levels of Hd1 accumulate in haf1. Our results suggest that HAF1 is essential to precisemodulation of the timing of Hd1 accumulation during the photoperiod response in rice.

INTRODUCTION

Rice (Oryza sativa) is one of themost important cereal crops in theworld. Heading date in rice is a critical determinant of regionaladaptability and grain production. Although rice is regarded asashort-day (SD)plant, artificial selectionduringdomesticationhasdeveloped cultivars adapted to different photoperiods in a broadrange of latitudes (Izawa, 2007; Song et al., 2015). The molecularbasis for regulation of flowering has been extensively studiedusingArabidopsis thaliana and rice,which represent long-day (LD)and SD plants (Song et al., 2015), respectively. Photoperiodicflowering in rice is regulated through two distinct pathways: theGI-Hd1-Hd3a pathway for adaption in SD conditions, a counter-part of the GI-CO-FT pathway in Arabidopsis (Turck et al., 2008);and the uniqueGhd7-Ehd1-Hd3a/RFT1pathway for adaptation inLDconditions (Xueet al., 2008;Tsuji et al., 2011;Songet al., 2015).In contrast to the GI-CO-FT pathway in Arabidopsis, the riceGI-Hd1-Hd3a pathway has evolved a distinct regulation patternamong orthologous pathways in rice. A previous study showedthat overexpression of rice GI suppressed Hd3a expression, re-sulting in late flowering under bothSDandLDconditions (Hayamaet al., 2003). This result indicated that riceGI functions oppositelytoGI in Arabidopsis and suppresses flowering in rice (Fowler et al.,

1999; Hayama et al., 2003). Further investigation showed thatincreased expression ofHd1 coincideswith overexpression ofGI,suggesting that GI-promoted Hd1 homolog expression is con-served between rice and Arabidopsis (Hayama et al., 2003).Hd1 is a major quantitative trait locus associated with photo-

period sensitivity among different rice cultivars (Yano et al., 2000).Recently, a genome-wide association study revealed that Hd1gene variation plays an important role in rice adaptation, whichdetermines thedistributionof varieties across low to high latitudes(Zhao et al., 2011; Huang et al., 2012). Allelic variation of Hd1 isinvolved in SD flowering in rice, suggesting that variation in Hd1proteins contributes to thediversity offlowering times incultivatedrice (Takahashi et al., 2009). Molecular analyses have demon-strated that Hd1 is a crucial regulator of the photoperiodic flow-ering pathway in rice (Izawa et al., 2002; Hayama et al., 2003).Although the function of Hd1 in rice is highly similar to that of COin Arabidopsis, the direct regulation ofHd3a byHd1 has not beenconfirmed. Overexpression ofHd1 has been reported to suppressHd3a and delay flowering under SD conditions (Ishikawa et al.,2011). By contrast, constitutive expression of CO promotesflowering by increasing FT transcription in Arabidopsis (Onouchiet al., 2000). Gene expression analysis has revealed that Hd1promotes FT-like genes in the inductive SD conditions and re-presses FT-like gene expression in noninductive LD conditions(Izawaet al., 2002). This antagonistic action ofHd1 activitymayberegulated posttranscriptionally because the expression pattern ofHd1 is almost the same in different photoperiods (Izawa et al., 2002;Ishikawaet al., 2011).Hd1expression is under circadian control, andit peaks at dusk under LD conditions (Yano et al., 2000; Izawa et al.,2002). In the presence of light, the Hd1 protein complex suppressesHd3a under LD conditions (Izawa et al., 2002; Ishikawa et al., 2011).

1 These authors contributed equally to this work.2 Address correspondence to [email protected] author responsible for distribution of materials integral to the findingspresented in this article in accordance with the policy described in theInstructions for Authors (www.plantcell.org) is: Changyin Wu ([email protected]).OPENArticles can be viewed online without a subscription.www.plantcell.org/cgi/doi/10.1105/tpc.15.00320

The Plant Cell, Vol. 27: 2455–2468, September 2015, www.plantcell.org ã 2015 American Society of Plant Biologists. All rights reserved.

Page 2: The RING-Finger Ubiquitin Ligase HAF1 Mediates Heading date 1 …croplab.hzau.edu.cn/__local/8/44/09/7F493FBF4BE3C2A36DFF... · 2019-04-22 · The RING-Finger Ubiquitin Ligase HAF1

Further investigation showed that Hd1 protein levels in Hd1-overexpressing plants are not altered in the presence of light(Ishikawaetal., 2011).SuppressionofHd3aanddelayedfloweringby overexpression of Hd1 under SD conditions depend on phyB,indicating that light may modulate Hd1 protein control of Hd3atranscription (Ishikawa et al., 2011). In Arabidopsis, posttranslationalregulation of CO protein is crucial for the photoperiodic induction ofFT expression (Jang et al., 2008). Various light signals modulate COprotein stability throughout the day (Valverde et al., 2004; Jang et al.,2008). Blue light hasbeen shown to stabilizeCO, but red light acts toreduceCOabundance in aphyB-dependentmanner (Valverdeet al.,2004; Jang et al., 2008; Liu et al., 2008). Although previous studiessuggest Hd1 promotes flowering under SD conditions and sup-presses it under LD conditions (Yano et al., 2000; Izawa et al., 2002;Hayama et al., 2003; Ogiso et al., 2010), the mechanism un-derlying how Hd1 abundance is regulated has not previouslybeen reported.

The ubiquitin 26S proteasome system (UPS) is critical in en-abling plants to alter their proteome to integrate internal andexternal signals for developmental plasticity and environmentaladaptation (Chen and Hellmann, 2013; Wang and Shi, 2014). TheUPS pathway, including ubiquitin-activating enzymes (E1s),ubiquitin-conjugating enzymes (E2s), and ubiquitin ligases (E3s),catalyzes the ubiquitination of various protein substrates fortargeteddegradation via the26Sproteasome (ChenandHellmann,2013). Substrate specificity mainly depends on the type of E3 li-gase,which includes threemainclasses:HECT,RING,andU-box inplants (ChenandHellmann, 2013). InArabidopsis, twoRING-fingerE3 ubiquitin ligases, CONSTITUTIVELYPHOTOMORPHOGENIC1(COP1)andHIGHEXPRESSIONOFOSMOTICALLYRESPONSIVEGENES1 (HOS1), influence the photoperiodic flowering pathway(Valverde etal., 2004; Jangetal., 2008; Liu et al., 2008; Lazaroetal.,2012). High CO transcript levels have been detected during thedark phase of both LD and SD conditions, but they do not correlatewith CO protein accumulation (Jang et al., 2008). In darkness,COP1 acts as an E3 ubiquitin ligase to ubiquitinateCO, promotingits degradation and consequently repressing flowering (Janget al., 2008). In the daylight period, CRY-mediated signal maytranslocate COP1 from the nucleus to the cytoplasm, therebystabilizing CO, activating FT transcription, and inducing flowering(Liu et al., 2008). In addition, during the daytime, HOS1 is requiredto degrade CO, and it prevents CO accumulation and FT expres-sion during the early daylight hours (Lazaro et al., 2012). Thus, bydirectly targeting CO through UPS, tight regulation of CO levels isachieved in response to different daylengths in Arabidopsis.

Although several transcriptional factors involved in photoperi-odic flowering have been characterized in rice (Tsuji et al., 2011),their direct interaction relationship and posttranscriptional regu-lation remain largely unknown. To elucidate the mechanism un-derlying Hd1 accumulation during the photoperiodic flowering inrice, we characterize an Hd1-interacting protein, Heading dateAssociated Factor 1 (HAF1), which is required for modulation ofheading date. HAF1 encodes a C3HC4 RING domain-containingE3 ubiquitin ligase. Moreover, we show that HAF1 physically in-teracts with Hd1 in vitro and in planta and regulates Hd1 proteinabundance during the photoperiod in rice. Mutation of HAF1significantly delays heading date under both SD and LD con-ditions. Thus, we propose that HAF1 is required to precisely

modulate the timing of Hd1 accumulation and to ensure an ap-propriate photoperiodic response in rice.

RESULTS

HAF1 Interacts with Hd1 in Yeast Cells

Previousstudiesshowed thatHd1playsacrucial role inpromotingflowering under SD conditions (Yano et al., 2000; Izawa et al.,2002; Hayama et al., 2003). The Hd1 protein contains two func-tional domains: two tandem repeats of zinc finger domains at theN terminus and the CCT domain at the C terminus (Yano et al.,2000). To elucidate themolecularmechanismofHd1mediation offlowering in rice, we performed a yeast two-hybrid assay usingHd1 as bait to identify interacting factors of Hd1. We first in-vestigated the transcriptional activation activity of Hd1 in yeast(Saccharomyces cerevisiae). Autoactivation was detected in thefull-length Hd1 and truncated Hd1 fragment constructs (aminoacids 78 to 395, 117 to 395, and 326 to 395) (Figure 1A). Allfragments contained theCCTdomainofHd1,which indicated thatthe CCT domain might have a functional transcription activationactivity. However, no transcriptional activation of the truncatedHd1 protein (amino acids 1 to 325) was detected (Figure 1A), andthis fragment was used as a bait to screen a yeast prey cDNAlibrary prepared from penultimate leaf blades of Zhonghua 11(ZH11) (O. sativa ssp japonica). Four independent clones con-taining the complete open reading frame of a gene encodinga finger proteinwere confirmed to interactwithHd1 (amino acids 1to 325), but not Hd1 (amino acids 1 to 116 or 1 to 77) (Figure 1B). ABLAST search of the open reading frame against the TIGR RiceGenome Annotation database (http://rice.plantbiology.msu.edu/index.shtml) retrievedasequence thatwasannotated asaC3HC4zinc finger gene (LOC_Os04g55510), which we named Headingdate Associated Factor 1 (HAF1).To investigate whether the C3HC4 zinc finger of HAF1 is re-

quired for the interaction with Hd1, a series of truncated HAF1cDNAs were cloned into the bait constructs, and their tran-scriptional activation activity was assayed in yeast. All the trun-cated fragments of HAF1 (amino acids 1 to 112, 99 to 619, 288 to619, 288 to 667, and 620 to 667) showed no transcriptional ac-tivation activity (Supplemental Figure 1). Yeast two-hybrid assaydemonstrated that constructs containing the N terminus (aminoacids 1 to 112) or C terminus (amino acids 620 to 667) of HAF1were sufficient for the interactionwithHd1 (Figure 1C). This findingsuggests that the C-terminal region of HAF1, which includes theC3HC4 domain, may be involved in the interaction with Hd1.Subsequently, we transformed yeast cells with baits encodingtwo regions of HAF1 (amino acids 99 to 619 and 288 to 667) andprey vectors with a series of truncated Hd1 cDNA fragments toexamine their interaction (Figure 1D). The results indicated that thesecondB-box zinc finger domain of Hd1 is required for interactionwith HAF1, but the CCT domain of Hd1 is dispensable.

HAF1 Physically Interacts with Hd1 in Vivo and in Vitro

To examine whether HAF1 interacts with Hd1 in plant cells, weused bimolecular fluorescence complementation assays to inves-tigate the interaction between the HAF1 and Hd1 in the abaxial

2456 The Plant Cell

Page 3: The RING-Finger Ubiquitin Ligase HAF1 Mediates Heading date 1 …croplab.hzau.edu.cn/__local/8/44/09/7F493FBF4BE3C2A36DFF... · 2019-04-22 · The RING-Finger Ubiquitin Ligase HAF1

epidermal cells of tobacco (Nicotiana benthamiana) leaf (Waadtet al., 2008).Hd1was fused to thesplitCFPCterminus (SCC:Hd1),whileHAF1wasattached to the split CFPN terminus (HAF1:SCN).Interaction of the two proteins would reconstitute the cyan fluores-cent chrome, allowing for detection of cyan fluorescence. Leaf epi-dermal cells were analyzed 5 d after infiltration with Agrobacteriumtumefaciens harboring the constructs. As anticipated, strong cyanfluorescence was observed in the nuclei of leaf cells when the full-length HAF1 and Hd1 proteins were coexpressed (Figure 2A), sug-gesting that HAF1 interacts with Hd1 in nuclei. When SCC:Hd1and HAF1:SCN were expressed in leaf cells separately, no fluo-rescencewasdetected(Figures2Band2C).Thesedataindicatethatthe interaction of HAF1 with Hd1 occurs in the nuclei in planta.

Consistent with the yeast two-hybrid and bimolecular fluores-cence complementation results, pull-down assays also dem-onstrated that HAF1 interactswith Hd1 in vitro.Maltose bindingprotein-tagged HAF1 (MBP-HAF1) andGST-tagged Hd1 (GST-Hd1) fusion proteins were expressed separately. As shown inFigure 2D, MBP-HAF1 bound to GST-Hd1, while MBP-HAF1 didnot bind to the negative control (GST). These results suggestthat HAF1 physically interacts with Hd1 in vitro.

Identification of the haf1 Gene in Rice

To characterize the function of HAF1 in rice, we obtained an haf1mutant from our rice T-DNA insertional library in variety ZH11 (Wuet al., 2003; Zhang et al., 2007). Analysis of the genomic sequenceflanking the T-DNA insertion site in haf1 showed that it is located1660 bp upstream from the ATG start codon of HAF1 on chro-mosome 4 (Figure 3A). RT-PCR analysis revealed that HAF1 wassuppressed in leaves from plants homozygous for the T-DNAinsertion, whereas the expression levels of the other genes sur-rounding the insertion site were not altered in haf1 compared withthe wild type (Supplemental Figure 2A). To investigate the de-velopmental defects of haf1, 23 T1 plants were planted in the fieldduring the 2010 rice-growing season in Wuhan, China, which hasnatural LD conditions. A pair of gene-specific primers (P1/P2)coupled with a T-DNA-specific primer (P3) were used to test thegenotype of the T-DNA insertion site (Figure 3A). Among 23 T1plants, five plants homozygous for the T-DNA insertion showeda late heading date, while the other plants, either heterozygous orhomozygous for the wild type, had a normal heading date (Figure3B). The phenotype segregation ratio in the T1 family of 23 plants

Figure 1. Hd1 Interacts with HAF1 in Yeast.

(A) Schematic diagram of full-length and a series of truncated Hd1 fusions to DNA binding domain (BD) and transcriptional activity assays of the indicatedproteins.(B)Yeast two-hybrid assay showing interaction between theHd1 (amino acids 1 to 325)withHAF1. The pGBK-T7/pGAD-HAF1 andpGBK-LAX2(156-394)/pGAD-LAX1 combinations were separately used as negative and positive controls.(C) Yeast two-hybrid assay showing interaction between amino acids 1 to 112 and 620 to 667 (C3HC4 domain) of HAF1 with Hd1.(D) Yeast two-hybrid assay showing that zinc finger domain-dependent Hd1-HAF1 interaction.

Hd1 Ubiquitination by HAF1 2457

Page 4: The RING-Finger Ubiquitin Ligase HAF1 Mediates Heading date 1 …croplab.hzau.edu.cn/__local/8/44/09/7F493FBF4BE3C2A36DFF... · 2019-04-22 · The RING-Finger Ubiquitin Ligase HAF1

(wild type:mutant = 18:5, x2 = 0.71 < 3.84 for 3:1) suggested thatthe late heading date was caused by a recessive mutation ofa single Mendelian locus. All progeny of T2 plants either het-erozygous or homozygous for the wild type exhibited normaltiming of heading, and all T-DNA insertion homozygotes showedlate heading. These results strongly suggested that mutation ofthe gene HAF1 results in a late heading date.

To further verify that the late heading date was caused by themutationofHAF1, an11.51-kbgenomicDNA fragment containingthe entire HAF1 coding region and the 2877-bp upstream and2952-bp downstream sequences was introduced into a haf1mutant background (Supplemental Figure 2B). Of 173 plants re-generated, 80 headed at least 2 weeks earlier than the 14 plantstransformed with the empty vector (the negative control) (Figure3C). DNA gel blot hybridization analysis of 38 randomly chosenT0HAF1-transgenicplants revealed that16plantscontainedonecopy of the transgene, and the others carried two ormore copies(Supplemental Figure 2C). All T1 progenies produced by self-pollinationof the single-copy transformants showedsegregationof headingdate under natural growthconditions. Asexpected, alltransgene-positive segregants restored normal heading date(Supplemental Figure 2D). We therefore concluded that HAF1modulates heading date in rice.

Heading Date Characterization of haf1

To investigate the heading date of haf1 under different photo-periodic conditions, we grew haf1 and corresponding wild typeunder twoartificialdaylength treatments:SDconditions (10h light/14 h dark) and LD conditions (14 h light/10 h dark). Under LDconditions, theheadingdateofhaf1 (83.4065.64)wasdelayed21d compared with the wild type (62.546 1.40) (Figures 3D and 3F).

Under the SD conditions, the heading date of haf1 (70.626 4.33)was delayed 17 d compared with the wild type (53.73 6 3.61)(Figures 3E and 3F). Thus, the haf1mutants showed late headingcompared with the wild-type plants under both photoperiodicconditions.To examine whether reduced growth rate caused late flowering

in haf1, we next compared the leaf emergence rate of haf1 andwild-type plants until heading. The leaf emergence rate of the haf1mutants was almost indistinguishable from that of the wild-typeplants prior to heading under both conditions (Figures 3Gand3H).Heading occurred in the wild-type plants after they produced12.696 0.75 leaves under LD and 10.926 0.67 leaves under SD(Figures 3Gand3H). Thehaf1plants exhibited a later headingdatecompared with the wild type and producedmore leaves under LD(13.8060.94 leaves) andSD (12.1460.53 leaves) (Figures3Gand3H). These results suggested that the delayed heading datesunder both photoperiodic conditions were due to a prolongedfloral transition in haf1.With regard to other agronomic traits, such as plant height, tiller

number per plant, 1000-grain weight, panicle length, and spikeletnumber per panicle, haf1 had no obvious differences comparedwith the wild type (Supplemental Table 1). These results dem-onstrated that the mutation in HAF1 does not have any obviouseffects other than causing a later heading date.

Expression Pattern of HAF1 and Subcellular Localization

To examine the temporal and spatial expression pattern ofHAF1,we performed RT-qPCR to analyze the transcription levels ofHAF1 under natural LD in selected tissues from the wild type:root and leaf blade at the vegetative stage and the root, leafblade, leaf sheath, culm, and young panicles at the reproductive

Figure 2. Hd1 Interacts with HAF1 in Vivo and in Vitro.

(A) to (C) Bimolecular fluorescence complementation assays showing the interaction between Hd1 and HAF1 in N. benthamiana leaf epidermal cells.N. benthamiana leaves were cotransfected with Agrobacterium carrying constructs for expression of full-length Hd1 andHAF1 that were tagged with SCCand SCN, respectively. The split CFP fragments were used in negative controls. Bars = 10 µm.(D) In vitro pull-down assay demonstrating the direct interaction between Hd1 and HAF1. MBP-HAF1 was pulled down (PD) by GST-Hd1 immobilized onglutathione-conjugated agarose beads and analyzed by immunoblotting (IB) using an anti-MBP antibody. Each input (IN) lanewas immunoblotted using ananti-GST or anti-MBP antibody.

2458 The Plant Cell

Page 5: The RING-Finger Ubiquitin Ligase HAF1 Mediates Heading date 1 …croplab.hzau.edu.cn/__local/8/44/09/7F493FBF4BE3C2A36DFF... · 2019-04-22 · The RING-Finger Ubiquitin Ligase HAF1

stage. HAF1 transcripts were detected in all examined tissues,but they were relatively abundant in leaf blade (SupplementalFigure 3A). The preferential expression of HAF1 in leaf bladeis consistent with its involvement in photoperiodic floweringin rice.

Next,weexamined the transcript levels ofHAF1 in leaf blades atvarious developmental stages. The leaf blades were collected atdusk every 5 d from the 16th day after germination under LDconditions. The RT-qPCR results showed that HAF1 was con-sistently expressed in leaf blade from the vegetative to the re-productive stage (Supplemental Figure 3B). This expression

pattern may indicate that HAF1 is required for the normal growthof rice.We also assayed the subcellular localization of the HAF1 pro-

tein. A construct, 35S:HAF1:GFP, was made to include the HAF1cDNA fused in frame to GFP under the control of cauliflowermosaic virus 35S promoter. Hd1 was fused to CFP as a nuclearprotein marker. Rice protoplasts from etiolated seedlings werecotransfected with 35S:HAF1:GFP and 35S:Hd1:CFP by poly-ethylene glycol treatment. The GFP-fused full-length HAF1 co-localized with CFP-fused Hd1 in the nucleus (SupplementalFigure 3C), suggesting that HAF1 is a nuclear protein.

Figure 3. Identification of HAF1 and Its Mutant.

(A) The HAF1 gene structure and T-DNA insertion site. Filled boxes represent exons and lines represent introns.(B) Genotyping of HAF1 segregants by PCR. W, wild-type; H, heterozygous; M, homozygous for T-DNA insertion.(C) The phenotype-rescued haf1 plant generated by genetic complementation (positive) and haf1 mutant transformed by empty pCAMBIA2301 vector(control).(D) and (E) Phenotypes of the wild type (left) and haf1 (right) at heading stage under LD (D) and SD (E) conditions. Photographs were taken at 65 d aftersoaking.(F) Days to heading in wild-type and haf1 mutant plants under LD and SD conditions. Error bars indicate SD; n = 15 plants.(G) and (H) Comparison of leaf emergence rates between the wild-type plant and haf1 mutant under LD (G) and SD (H) conditions during development(means 6 SD, n = 15).

Hd1 Ubiquitination by HAF1 2459

Page 6: The RING-Finger Ubiquitin Ligase HAF1 Mediates Heading date 1 …croplab.hzau.edu.cn/__local/8/44/09/7F493FBF4BE3C2A36DFF... · 2019-04-22 · The RING-Finger Ubiquitin Ligase HAF1

Effects of HAF1 on Gene Expression inPhotoperiod Pathway

To further investigate the role of HAF1 in photoperiod-controlledflowering in rice, we compared the diurnal expression patterns ofmain flowering time regulators in the haf1 and wild-type plantsunder both LD (16 h light/8 h dark) and SD (8 h light/16 h dark)conditions. We monitored transcripts day and night by RT-qPCRat 4-h intervals duringa48-hperiod.Comparedwith theexaminedgenes involved in photoperiod flowering, theHAF1 transcript levelwasvery low in thewild-typeplant andshoweddiurnal or circadianpatterns under either LDorSDconditions (Figures 4Aand4B). Theexpression of HAF1 was much more abundant during the darkperiod than the light period (Figures 4A and 4B). In the haf1 mu-tants, the HAF1 transcript level was dramatically reduced underboth conditions, especially during the dark period (Figures 4A and4B). HAF1 had no effect on the expression of GI, which showedidentical expression patterns in haf1 and the wild type under bothconditions (Figures 4C and 4D). However, HAF1 had a consider-able effect on the expression of Hd1 and Ehd1. HAF1 mutationsuppressed the expression peak of Hd1 during the light periodunder LD and SD conditions (Figures 4E and 4F). The expressionof Ehd1 was almost completely suppressed in haf1 under bothconditions (Figures 4G and 4H). Ultimately, the expression of twoflorigen genes, RFT1 in LD and Hd3a in SD (Tamaki et al., 2007;Komiya et al., 2008), was significantly reduced in haf1 (Figures4I and 4J). Taken together, these results suggested that HAF1functions upstream of Ehd1, Hd1, RFT1, and Hd3a in the LD andSD photoperiodic flowering pathway. Thus, HAF1 might be in-volved inbothHd1-dependentandEhd1-dependentphotoperiodicflowering pathways in rice (Tsuji et al., 2011; Song et al., 2015).

HAF1 Is a RING Finger E3 Ubiquitin Protein Ligase

HAF1 is predicted to encode a 667-amino acid protein containingaC3HC4-typeRING finger at theC terminus (Figure 5A). TheRINGdomain in HAF1 contains the conserved consensus sequenceC-X2-C-X(9-39)-C-X(1-3)-H-X(2-3)-C-X2-C-X(4-48)-C-X2-C (Figure 5A),in which X is any amino acid (Borden and Freemont, 1996). E3ubiquitin ligases are characterized by conserved RING fingerdomain function (Stone et al., 2005), prompting us to test whetherHAF1 also has E3 ubiquitin ligase activity.

We first tested whether HAF1 had enzyme activity for self-ubiquitination. We expressed HAF1 in Escherichia coli fusedwith MBP and purified MAP-HAF1 from the soluble fraction(Supplemental Figure 4A). Self-ubiquitination of HAF1 was as-sayed in the presence of E1-activating enzyme, E2-conjugatingenzyme, and His-ubiquitin. A polyubiquitination signal was ob-served by immunoblot using an anti-His antibody (Figure 5B,second lane from the right). The anti-MBP blot analysis alsoshowed that MBP-HAF1 was ubiquitinated (Figure 5B, secondlane from the right). As thenegative controls, nopolyubiquitinationof HAF1 was detected when E1, E2, His-ubiquitin, or MBP-HAF1was absent in the reaction (Figure 5B). This result suggested thatHAF1 has E3 ubiquitin ligase activity.

The RING domain has been reported to usually form a stabledimer, which is required for ubiquitin transfer (Liew et al., 2010).Given that the molecular weight of the ubiquitinated MBP-HAF1(Figure 5B, second lane from the right) exceeds the theoretical

Figure 4. Diurnal Expression Patterns of Flowering Regulators in the WildType and haf1.

Diurnal expression patterns ofHAF1 ([A] and [B]),GI ([C] and [D]),Hd1 ([E]and [F]), Ehd1 ([G] and [H]), RFT1 (I), and Hd3a (J) in wild-type and haf1plants in SD and LD conditions. The expression levels measured by RT-qPCR are shown relative to theUBQmRNA. In all panels, themean of eachpoint is based on the average of three biological repeats calculated usingthe relative quantification method. ZT, Zeitgeber time. The open and filledbars at the bottom represent the light and dark periods, respectively. Errorbars indicate SE.

2460 The Plant Cell

Page 7: The RING-Finger Ubiquitin Ligase HAF1 Mediates Heading date 1 …croplab.hzau.edu.cn/__local/8/44/09/7F493FBF4BE3C2A36DFF... · 2019-04-22 · The RING-Finger Ubiquitin Ligase HAF1

calculation for the monomer, we speculated that HAF1 may forma homodimer to be ubiquitinated. The truncated fragments ofHAF1 (amino acids 99 to 619, 288 to 667, and 620 to 667) thatserved as bait were able to interact with full-length HAF1 in theyeast two-hybrid assay (Figure 5C). Homodimerization of HAF1was further confirmedbypull-downexperiment (Figure5D). Takentogether, HAF1, as a RING finger protein, may form a dimer tofunction as an E3 ubiquitin ligase.

HAF1 Mediates Ubiquitination of Hd1

SinceHAF1 interactswithHd1andHAF1hasE3 ligaseactivity,weconsidered the possibility that HAF1 mediates ubiquitination of

Hd1. To examine this possibility, we purified the tagged proteinsMBP-HAF1 and GST-Hd1 (Supplemental Figure 4B) for ubiq-uitinationassays invitro, andHd1was found tobeubiquitinatedbyHAF1 in the presence of E1, E2, and His-Ub (Figure 6A). Withincreased amounts of GST-Hd1 protein, the intensity of ubiq-uitinated GST-Hd1 bands was stronger (as shown in Figure 6A,asterisk from the fifth to eighth lanes from the left), whereas thesignals from self-ubiquitination of HAF1 declined (SupplementalFigure 5, top panel). These results indicated that HAF1 mediatesubiquitination of Hd1 in vitro.To confirm the ubiquitination of Hd1 by HAF1, we employed an

efficient in vivo ubiquitination assay by agroinfiltration expressionof both substrates and E3 ligases in N. benthamiana (Liu et al.,

Figure 5. Molecular Characterization of HAF1.

(A)Aschematic representationof theHAF1protein and amino acid alignments of theHAF1RING finger domainwithRING-like domains fromother species.Zm,Zeamays; Si,Setaria italica; Tu,Triticumurartu; Sb,Sorghumbicolor; Bd,Brachypodiumdistachyon; Eg,Elaeis guineensis; Pt,Populus trichocarpa; Pd,Phoenix dactylifera; Tc, Theobroma cacao; Jc, Jatropha curcas; Nt, Nicotiana tomentosiformis; At, Arabidopsis thaliana.(B) E3 ubiquitin ligase activity of HAF1. MBP-HAF1 fusion proteins were assayed for E3 ubiquitin ligase activity in the presence of E1, E2, and His-ubiquitin(His-Ub). Ubiquitinated proteins were detected by immunoblotting (IB) with anti-HIS antibody and anti-MBP antibody, respectively.(C)HAF1homodimersweredetected in yeast two-hybrid assays. The emptypGBK-T7/pGAD-HAF1andpGBK-LAX2(156-394)/pGAD-LAX1combinationswere separately used as negative control and positive controls, respectively.(D)HAF1homodimersweredetectedvia in vitropull-downassays.MBP-HAF1waspulleddown (PD) byGST-HAF1 immobilizedonglutathione-conjugatedagarose beads and analyzed by immunoblotting (IB) using an anti-MBP antibody. Each input (IN) lane was immunoblotted using an anti-GST or anti-MBPantibody.

Hd1 Ubiquitination by HAF1 2461

Page 8: The RING-Finger Ubiquitin Ligase HAF1 Mediates Heading date 1 …croplab.hzau.edu.cn/__local/8/44/09/7F493FBF4BE3C2A36DFF... · 2019-04-22 · The RING-Finger Ubiquitin Ligase HAF1

2010). We transiently expressed HA-tagged Hd1 (HA-Hd1) viaagroinfiltration, HA-Hd1 protein was detected, and Hd1 andGAPHD mRNA expression levels were analyzed by RT-PCR(Figure 6B). When HA-Hd1 and Flag-HAF1 were coexpressed inthe same leaf area, HA-Hd1was rapidly degraded (Figure 6B, firstlane from the left), but HA-GFP, an internal control in this ex-periment, was not (Figure 6B). However, when the 26S protea-some inhibitor MG132 was infiltrated into the same region,HA-Hd1 protein could be detected using the HA antibody (Figure6B, second lane from the left). Together, these results suggestedthatHAF1 targetsHd1 for ubiquitin-dependent degradationby the26S proteasome.

HAF1 Regulates Circadian Accumulation of Hd1

Previous studies speculated that posttranslational regulation ofHd1 protein level might play an important role in determining theheading date of rice (Luan et al., 2009; Ishikawa et al., 2011).Because HAF1 mediates Hd1 degradation in vitro, we examinedwhether the accumulation of Hd1 depends on HAF1 in vivo. Weseparately harvested 14-d-old seedling leaves from the wild typeand haf1 under SD conditions at Zeitgeber time = 0, whenHAF1 ishighly expressed (Figure 4B). Total protein extractswere prepared

from the leaf tissues and incubated with MBP-Hd1 for cell-freedegradationassays.We found thatMBP-Hd1wasdegradedmorerapidly during incubation with protein extracts from the wild typecompared with those from haf1 mutants. In the presence ofMG132, the rate of MBP-Hd1 degradation was lower when thesameprotein extracts from thewild typewereused (SupplementalFigure 6A). By contrast, no obvious degradation of the controlMBP protein and rice ACTIN was observed when using the sameplant extracts (Supplemental Figures 5B and 5C), indicating thatHAF1 is specifically responsible for Hd1 protein degradation.Next, we further examined Hd1 rhythmic accumulation in wild-

type and haf1 plants entrained in both SD and LD conditions.Consistent with the previous examinations, Hd1 expression wasrhythmic in the wild type and haf1 (Figures 4C and 4D), with highexpressionduring thedarkperiods under both conditions (Figures4C and 4D). Although expression of Hd1 protein in rice was low(Ishikawa et al., 2011), immunoprotein gel blot analysis using anti-Hd1 antibody showed that Hd1 remained relatively abundant inthe light (Figures 6C and 6D). In haf1, Hd1 accumulation showedno clear diurnal rhythm but rather almost constant levels for thetime points studied (Figures 6C and 6D). The greater Hd1 accu-mulation inhaf1 than in thewild typesuggested thatHAF1controlsHd1 stability in vivo.

Figure 6. HAF1 Ubiquitinates Hd1 and Promotes Degradation of Hd1.

(A) HAF1 ubiquitinates Hd1 in vitro. GST-Hd1 ubiquitination assays were performed using MBP-HAF1, E1, E2, and His-Ub.(B)HAF1 and the proteasome regulateHd1 stability in vivo. Immunoblot analysis of protein extracts corresponding to agroinfiltratedN. benthamiana leaveswith the indicated plasmids in the presence or absenceofMG132.HA-Hd1 andHA-GFPwere detected using anti-HAantibody andFLAG-HAF1using anti-FLAG antibody. FLAG-HAF1 (HAF1), HA-Hd1 (Hd1), and GAPDH mRNA expression levels were analyzed by RT-PCR.(C)and (D)Circadian accumulationofHd1protein inwild-type andhaf1plants grownunderSDandLDconditions. Total protein extracts from35-d-oldplantleaves were loaded into each lane.Hd1 andGAPDHmRNA expression levels were analyzed by competitive RT-PCR. ACTIN abundance was detected asloading control. Anti-Hd1 antibody was used to detect Hd1 abundance. CBB, Coomassie blue staining.

2462 The Plant Cell

Page 9: The RING-Finger Ubiquitin Ligase HAF1 Mediates Heading date 1 …croplab.hzau.edu.cn/__local/8/44/09/7F493FBF4BE3C2A36DFF... · 2019-04-22 · The RING-Finger Ubiquitin Ligase HAF1

Genetic Interaction of HAF1 and Hd1

To investigate the genetic relationship between HAF1 and Hd1,we generated a homozygous haf1 hd1 doublemutant in the samebackground of ZH11. Previous studies demonstrated that Hd1promotes flowering under SD conditions, whereas it suppressesflowering under LD conditions (Izawa et al., 2002; Hayama et al.,2003; Ishikawa et al., 2005). To analyze the genetic effect ofHAF1on Hd1, we obtained the hd1 mutant in ZH11 (Luan et al., 2009).Compared with the wild type, hd1 headed later under SD con-ditions (Supplemental Figure 7) and earlier under LD conditions(Supplemental Figure 8). Under SD conditions, the haf1 hd1doublemutant headed as late as hd1plants, indicating thatHd1 isgenetically epistatic to HAF1 (Supplemental Figure 7). However,under LD conditions, the haf1 hd1 double mutant exhibiteda heading date similar to that of haf1 (Supplemental Figure 8).Taken together, these data indicate that HAF1 may influenceheading date mainly through Hd1 under SD conditions. Thisfindingagreeswithother studies that showHd1playsacrucial rolein promoting heading under SD conditions (Yano et al., 2000;Izawa et al., 2002; Hayama et al., 2003). Since the major headingregulators Ehd1 and RFT1may function independently from Hd1under LD conditions (Doi et al., 2004; Komiya et al., 2008; Komiyaet al., 2009), we propose that HAF1 may participate in headingdate regulation byubiquitination of flowering regulators other thanHd1 under LD conditions.

DISCUSSION

Although ubiquitin-mediated degradation of proteins in thephotoperiodic flowering pathway has been well characterized inArabidopsis, little is known about the molecular mechanism re-sponsible for circadian accumulation of flowering regulators inrice. Here, we identified a C3HC4-type RING-finger E3 ubiquitinligase, HAF1, as an essential factor for Hd1 abundance in headingdatedeterminationof rice.Hd1 interactswithHAF1, andHd1 is thedirect substrate of HAF1 for ubiquitination in vitro and in vivo.Mutation of HAF1 resulted in Hd1 accumulation and delayedflowering, especially under SD conditions. Our results demon-strate that HAF1 has a pivotal role in promoting flowering throughregulation of Hd1 accumulation during the photoperiod responsein rice.

HAF1 Is a RING Finger E3 Ubiquitin Ligase

HAF1 encodes a C3HC4-type RING-finger E3 ubiquitin ligase,which contains a zinc finger domain at the C terminus. A BLASTpsearch of HAF1 against the NCBI reference protein database(http://blast.ncbi.nlm.nih.gov/) retrievedhighly similar orthologs inArabidopsis, indicating that HAF1might represent an E3 ubiquitinligase generally present in plants. In Arabidopsis, COP1 is a well-characterized C3HC4-type RING-finger E3 ubiquitin ligase that isalso involved in regulating flowering time by directly targetingtranscriptional activatorCO for degradation in thedark (Janget al.,2008; Liu et al., 2008). The COP1 protein comprises three rec-ognizable domains: a RING-finger domain, a coiled-coil domain,and sevenWD40 repeats (Deng et al., 1992;McNellis et al., 1994),all of which have been implicated in mediating the interaction of

COP1 with other proteins (Deng et al., 1992; Wang et al., 2001;Yang et al., 2001; Yu et al., 2008). Although HAF1 contains theconserved C3HC4-type RING-finger domain, none of the othercharacterized domains were found in HAF1. Our investigationsuggested that theC3HC4RINGdomainofHAF1 is involved in theinteraction with Hd1 (Figure 1C). The RINGdomain has previouslybeen shown to be required for dimer formation of C3HC4 RINGfinger E3 ubiquitin ligase (Liew et al., 2010; Feltham et al., 2011;Johnson et al., 2012). C3HC4 RING E3 ligases, such as RAG1(Rodgers et al., 1996), RNF4 (Plechanovová et al., 2011), and cIAP(Mace et al., 2008), can self-interact to form homodimers. Due totwo RING domains possibly being required to spatially accom-modate E2 (Plechanovová et al., 2011), homodimerization waslinked to the ubiquitination function of RING E3 ligases. Based onyeast two-hybrid analysis and GST pull-down assay, HAF1 wasconfirmed to physically interact with itself to form a homodimer(Figures 5C and 5D). Homodimerization of HAF1 might be pre-requisite for the ubiquitination of Hd1.Given that Hd1 and its ortholog CO were shown to be subject

to ubiquitin-based degradation mediated by HAF1 and COP1,respectively, HAF1might exert a corresponding function asCOP1in rice.Previousstudiesshowed thatCOP1 repressesfloweringbypromotingdegradationofCOduring thenight inArabidopsis (Janget al., 2008; Liu et al., 2008). Our investigation revealed that HAF1promotesflowering through regulationof theubiquitinationofHd1in rice. haf1 showed later flowering, but other agronomic traits didnot differ compared with the wild type (Supplemental Table 1),suggesting that the sole function of HAF1 may center on theheading date, without any other effects in rice. However, COP1 isapleiotropic effector that regulatesnot only thefloral transitionbutalso photomorphogenesis in seedlings (McNellis et al., 1996;Schwechheimer and Deng, 2000; Luo et al., 2014). A cop1mutantdisplayed constitutive photomorphogenic development in dark-ness (Dengetal., 1991).A recentstudysuggested thatHAF1mightinteractwithHAL3andbe involved in a newly discovered pathwayin the light-regulated growth of rice (Sun et al., 2009). To in-vestigatewhether photomorphogenesis in rice requiresHAF1, wecompared the length of coleoptiles and the first leaf between haf1and the wild type under various light conditions, including con-tinuous irradiation with red, far-red, and blue (Bc) for 9 d (Takanoet al., 2005). We also performed the same examination in con-tinuous darkness. No apparent differences were found betweenthe wild type and haf1mutants grown in the dark and various lightconditions (Supplemental Table 2). Rice PPS has recently beenreported to be an ortholog of Arabidopsis COP1. The ppsmutantshows a prolonged juvenile phase and also exhibits photomor-phogenesis in the dark. However, COP1, unlike PPS, is not in-volved in the juvenile-adult phase change (Tanaka et al., 2011).These results suggest thatmechanismsdifferent fromthose in ricehave developed to ensure developmental progress.

HAF1 Interacts with Hd1 and Promotes Its Degradation

HAF1 showed obvious diurnal expression patterns with a peak atthebeginningof daytimeanda troughat dusk (Figures 4Aand4B).The transcripts ofHd1 also displayed diurnal expression patterns,with high expression levels during thenight andapeakatmidnight(Figures 4E and 4F). The phase of expression pattern in HAF1 is

Hd1 Ubiquitination by HAF1 2463

Page 10: The RING-Finger Ubiquitin Ligase HAF1 Mediates Heading date 1 …croplab.hzau.edu.cn/__local/8/44/09/7F493FBF4BE3C2A36DFF... · 2019-04-22 · The RING-Finger Ubiquitin Ligase HAF1

opposite to that in Hd1, indicating that diurnal expression ofHAF1 might be required for degradation of Hd1 remnants by themorning. However, how is HAF1 itself regulated in post-transcriptional level? COP1 in Arabidopsis is known to be regu-lated by its suppressor CSU1, which negatively regulates COP1abundance by maintaining COP1 homeostasis in dark-grownseedlings (Xu et al., 2014). The necessary components for HAF1degradationneed tobe investigated in future studies.Additionally,given the various interactions between proteins andCOP1with orwithout light (Jang et al., 2008; Liu et al., 2008), we speculate thatthe abundant accumulation of HAF1 itself and its physical in-teraction with other proteins may synergistically regulate HAF1function toward Hd1 accumulation in heading date control in rice.

Light was shown to play a pivotal role in the Hd1-mediatedregulation of flowering in rice (Ishikawa et al., 2005, 2011). At dusk,Hd1 expression increases, and the Hd1 protein is expected tobe present in the dark. Under SD conditions, Hd1 accumulatesat night and activates Hd3a expression to promote flowering(Ishikawa et al., 2011). When days become longer, Hd1 protein isexposed to light at dusk and becomes an inhibitor to suppressflowering (Ishikawa et al., 2011). In Hd1-overexpressing plants,high levelsofHd1mayexist in the light atdusk, resulting indelayedflowering under SD conditions (Ishikawa et al., 2011). Previousstudies suggested that Hd1 mediates regulation of Hd3a ex-pression mainly through PhyB (Ishikawa et al., 2005, 2011). SinceHd1 is the direct target of HAF1 for ubiquitination, the increasedlevels of Hd1 protein appear in the haf1mutant during the daytime(Figure 6C). In the presence of afternoon light, modified Hd1 actsas a repressor of Hd3a expression, resulting in delayed floweringof haf1 under SD conditions.

HAF1 May Participate in Heading Date Control underLD Conditions

Our investigation suggested that HAF1 influences the Hd1- andEhd1-dependent pathways (Figure 4), which induce flowering ofriceunderSDandLDconditions, respectively. Inhaf1mutants, thetranscript levelsofHd3aandRFT1weredecreased (Figures 4I and4J), suggesting that HAF1 is involved in the photoperiodic reg-ulation of flowering in rice. Under both photoperiodic conditions,the expression of Ehd1 was almost eliminated in haf1 mutants(Figures 4G and 4H), suggesting that HAF1 is also required forEhd1 expression under LD conditions. Ehd1 has been demon-strated to be a unique activator to induce flowering in rice in-dependently of Hd1 (Doi et al., 2004). Considering that thetranscript level ofEhd1 is downregulated in plants overexpressingHd1 (Ishikawaetal., 2011), theaccumulationofHd1protein inhaf1might repress theexpressionofEhd1underSDandLDconditions.Furthermore, under LDs, the haf1 hd1 double mutant flowers aslate as haf1 (Supplemental Figure 8), indicating that HAF1 mayhave a crucial role in the photoperiod pathway under LD con-ditions. Recently, a photoperiodic flowering pathway, ELF3-Ghd7-Ehd1-RFT1 (Tsuji et al., 2011; Zhao et al., 2012; Yang et al.,2013), was proposed for the induction of rice flowering underLD conditions. We speculate that additional componentsmay interact with HAF1 and be regulated by HAF1 at theposttranscriptional level to mediate flowering progress in LDconditions.

METHODS

Plant Materials and Growth Conditions

T-DNA-tagged mutant haf1 was identified from our T-DNA insertionalmutant library (http://rmd.ncpgr.cn/;Wu et al., 2003). Screening of headingdatemutantswasperformedbyplanting themutantmaterialsundernaturalLD in the experimental field of Huazhong Agriculture University at Wuhan(30.4°N, 114.2°E), China, in the summer of 2010. We also planted the haf1and wild-type plants under artificial SD (10 h light/14 h dark, 28°C) and LD(14 h light/10 h dark, 28°C) conditions in a growth control room (B8;Eshengtaihe Ctrl Tech). Heading datewas recorded as the number of daysfrom germination to the day of panicle emergence from the leaf sheath.

We grew the seedlings at 28°C in a Four Color LED Absolute Cold Lightplant incubator (BiotronLH-55LED-SS;NipponNKSystem) under differentmonochromatic light sources, far red (735 nm), red (670 nm), and blue(470 nm), to examine the role of HAF1 in photomorphogenesis.

Genotyping of Mutant Plants

Genotyping of the haf1-segregating population was performed by PCRusing the following primers: P1, P2, and P3. PCR was conducted with aninitial step of 94°C incubation for 5min; a second step of 30 cycles of 94°Cfor45s,58°C for45s, and72°C for1minandfinal productsof 463and1289bp. Genotyping of the hd1 was performed as described by Luan et al.(2009). Leaf emergence rate was calculated according to the measure-ments described by Itoh et al. (1998).

Yeast Two-Hybrid Assay

Total RNA for construction of the yeast AD-fused cDNA library was pre-pared from penultimate leaf blades of ZH11 using Matchmaker LibraryConstruction and Screening Kits (Clontech). The library was screenedusing truncatedHd1 protein (amino acids 1 to 325) as bait by yeastmating.Plasmids from positive clones were verified for interaction by retrans-formation into yeast (Saccharomyces cerevisiae) AH109 cells containingpGBKT7-Hd1 (amino acids 1 to 325). pGBK-LAX2(156-394)/pGAD-LAX1was the positive control (Tabuchi et al., 2011).

To test Hd1 and HAF1 transcriptional activity, the full-length or trun-cated CDS of Hd1 and HAF1 were amplified and cloned into the pGBKT7(Clontech), respectively. The assays for transactivation activity were thenperformed following the manual of Matchmaker Gold Yeast Two-HybridSystem (Clontech). Transformants were plated on either tryptophan-negative synthetic dropout medium (SD/Trp2) or tryptophan-negative andadenine-negative synthetic dropout medium (SD/Trp2/Ade2).

Bimolecular Fluorescence Complementation Assays

The cDNAs of Hd1 and HAF1 were individually cloned into pSCYCE-R andpSCYNE vectors that contained either N- or C-terminal CFP fragments(Waadt et al., 2008), designated SCC:Hd1 and HAF1:SCN, respectively.The bimolecular fluorescence complementation assays using Nicotianabenthamiana (tobacco) were performed as described by Yuan et al. (2012).

Vector Construction and Rice Transformation

An 11.51-kbBglII-BglII genomic DNA fragment containing the entireHAF1coding region and the 2877-bp upstream and 2952-bp downstream se-quences was isolated by digestion of the Clemson BAC clone OSJN-Ba0010D21 (kindlyprovidedbyR.Wing,University ofArizona)and insertedinto the binary vector pCAMBIA2301 (Cambia). An empty pCAMBIA2301vector was used as a control. Constructs were introduced into Agro-bacterium tumefaciens EHA105 and transformed into homozygous callifrom haf1 by Agrobacterium-mediated transformation as previously

2464 The Plant Cell

Page 11: The RING-Finger Ubiquitin Ligase HAF1 Mediates Heading date 1 …croplab.hzau.edu.cn/__local/8/44/09/7F493FBF4BE3C2A36DFF... · 2019-04-22 · The RING-Finger Ubiquitin Ligase HAF1

described by Wu et al. (2003). G418 was used for the selection of trans-formant plants.

Subcellular Localization of HAF1

To construct the subcellular localization plasmids, the full-length CDS ofHAF1 was cloned into a pM999-GFP vector that contained GFP reportergene (primers are described in Supplemental Data Set 1). Protoplasts wereisolated from 10- to 15-d-old rice (Oryza sativa) etiolated seedlings andtransformed with the pM999-CFP-Hd1. Fluorescence in the transformedprotoplasts was captured using a confocal laser scanning microscope(Leica TCS SP2) after incubation at 28°C for 12 to 16 h.

DNA Extraction and DNA Gel Blot Hybridization

Fresh leaves were harvested from field-grown plants, and genomic DNAwas extracted using the CTAB method described by Murray andThompson (1980). The probes for DNA gel blot hybridizationwere the PCRproducts amplified with T-primers (primers are described in SupplementalData Set 1). The experimental procedures for DNA gel blot analysis fol-lowed methods previously described by Liu et al. (1997).

Gene Diurnal Expression Analysis

Rice seedlings were grown for 30 d in the greenhouse under naturaldaylength conditions. For LD samples, the plants were transferred toa growth chamber set for LD (16 h light/8 h dark, 28°C and 50% humidity)and entrained for 5 d. Then, penultimate leaves were harvested every 4 hduringa48-hperiod fromwild-typeandhaf1plants. ForSDsamples,plantswere transferred to a growth chamber set for SD (8 h light/16 h dark, 28°Cand 50% humidity) and entrained for 5 d. Samples were collected in thesame way as for LD conditions.

Total RNA was extracted from various tissues using the TRIzol reagent(Invitrogen) according to the manufacturer’s instructions, and first-strandcDNA was synthesized from 2.5 mg total RNA with SuperScript III reversetranscriptase (Invitrogen). RT-qPCR was performed with the AppliedBiosystems 7500 real-time PCR detection system using SYBR GreenMasterMix (AppliedBiosystems). Themeasurementswere obtained usingthe relative quantification method (Livak and Schmittgen, 2001).

Purification of Recombinant Proteins

Full-length Hd1 cDNA was inserted into a GST fusion vector pGEX-4T-1(Pharmacia). The resulting GST-Hd1 fusion was expressed in BL21-CodonPlus (Stratagene) Escherichia coli cells and purified using gluta-thione Sepharose 4B (Pharmacia). MBP-tagged HAF1 protein and MBPalonewereexpressed inBL21-CodonPlusandpurifiedusingamylose resinbeads (New England Biolabs).

Pull-Down Assays

The coding sequences of HAF1 and Hd1were cloned into the pGEX-4T-1vector to generate the constructs to express GST-HAF1 and GST-Hd1,respectively. The method for in vitro pull-down assays was modified fromXia et al. (2013). Bacterial lysates containing ;15 mg GST-Hd1 or GST-HAF1 fusion proteins were mixed with lysates containing ;30 mg MBP-HAF1 fusionproteins.GlutathioneSepharose (30mL;GELifeSciences)wasadded with rocking at 4°C for 60 min. Beads were washed four times withtheTGHbuffer (50mMHEPES, pH7.5, 150mMNaCl, 1.5mMMgCl2, 1mMEGTA, pH 8.0, 1% Triton X-100, 10% glycerol, 1 mM PMSF, and 13Complete protease inhibitor cocktail [Roche]), and the isolated proteinswere further separated on two 10% SDS-PAGE gels and detected by im-munoblot analysis with anti-GST antibody (Abmart M20007L; 1:1000 di-lution)andanti-MBPantibody (NEBE8032S;1:10,000dilution), respectively.

Ubiquitination Assays in Vitro

Reaction mixtures (30 mL) contained 110 ng E1 (Boston Biochem), 170 ngE2 (UBCh5a; Boston Biochem), 1 mg His-ubiquitin (Boston Biochem), and2 mg MBP-HAF1 fusion protein and various concentrations of GST-Hd1fusion protein in reaction buffer containing 50mM Tris, pH 7.5, 3 mMDTT,5 mMMgCl2, and 2 mM ATP. After 6 h incubation at 30°C, reactions werestopped by adding sample buffer, and half of the mixtures were separatedonto two 10% SDS-PAGE gels. Ubiquitinated GST-Hd1 was detectedusing anti-GST antibody (Abmart M20007L; 1:1000 dilution) and anti-HISantibody (Abmart M20001L; 1:1000 dilution). Images were visualized onTanon-5200 Chemiluminescent Imaging System (Tanon Science andTechnology).

Ubiquitination Assay in Vivo

In vivo ubiquitination assay was performed according to the protocoldescribed by Liu et al. (2010). We coinfiltrated the Agrobacterium strainscarrying the Flag-HAF1 andHA-Hd1 plasmids intoN. benthamiana leaves.The corresponding empty vectors were used as the controls, and the HA-GFP plasmid was added as an internal control. For proteasome inhibition,leaves were infiltrated with 10 mM MgCl2 and 50 mM MG132 (Sigma-Aldrich) solution for 12 h before sample collection. Three days after in-filtration, samples were collected for protein and RNA extraction. Forprotein-level analysis, the extracts were analyzed using anti-Flag antibody(Sigma-Aldrich F3165; 1:1000 dilution) and anti-HA antibody (Abcamab9110; 1:1000 dilution). For RNA level expression analysis, RT-PCR wasperformed.

Hd1 Polyclonal Antibody Preparation

For preparation of Hd1 polyclonal antibody, a 960-bp DNA fragmentencoding a 320-amino acid peptide of Hd1 (residues 1 to 320) was clonedinto pET32a vector (Novagen). The recombinant protein was expressed inE. coliDE3 (Transgen) and purified using nickel nitrilotriacetic acid agarose(Qiagen) to produce rabbit polyclonal antibodies (prepared by Abclonal ofChina). The antibody (1:200 dilution) was tested by immunoblot analysisusing purified GST-Hd1 and MBP-Hd1 from E. coli.

Detection of Hd1 Abundance in Rice

Rice leaves from35-d-old seedlingswasground into a finepowder in liquidnitrogen.A200-mLaliquotofextractionbuffer (50mMTris-HCl,pH7.5,150mMNaCl, 0.1% SDS, and 13 Complete protease inhibitor cocktail [Roche])was added to each 100-mg powder sample. The mixture was vortexedand then chilled on ice for 5 min. Samples were centrifuged at 16,000g for20min at 4°C, and the supernatant was collected and stored at270°C. Anequal amount of rice proteinwas loaded and separated by SDS-PAGE andthen stained by Coomassie Brilliant Blue or detected by immunoblotanalysis with anti-ACTIN antibody (Abmart M20009L; 1:1000 dilution).For detection of Hd1 abundance, an equal amount of rice protein wasseparatedona10%SDS-PAGEanddetectedby immunoblot analysiswithanti-Hd1 antibody (1:200 dilution).

Cell-Free Protein Degradation Assay

For cell-free protein degradation assays, seedlings were grown in green-house for 10 to 14 d under SD conditions. Total proteins were extracted inthedegradationbuffer (25mMTris-HCl, pH7.5, 10mMNaCl, 10mMMgCl2,5 mM DTT, and 10 mM ATP). To monitor the degradation of the expressedrecombinant MBP-Hd1 and MBP proteins, 200 ng purified MBP-Hd1 andMBP protein was added to 75mL seedling extract for individual assays. Thereaction mixtures were incubated at 30°C for the indicated time points.Reactions were blocked by adding the sample buffer. An equal amount of

Hd1 Ubiquitination by HAF1 2465

Page 12: The RING-Finger Ubiquitin Ligase HAF1 Mediates Heading date 1 …croplab.hzau.edu.cn/__local/8/44/09/7F493FBF4BE3C2A36DFF... · 2019-04-22 · The RING-Finger Ubiquitin Ligase HAF1

reactions separated on a 10% SDS-PAGE and detected by immunoblotanalysis with anti-MBP antibody (NEB E8032S; 1:10,000 dilution). As aloading control, plant ACTIN abundance was detected with anti-ACTINantibody (Abmart M20009L;1:1000 dilution).

Accession Numbers

Sequence data from this article can be found in the GenBank databasesunder the following accession numbers: HAF1, LOC_Os04g55510; Hd1,LOC_Os06g16370; OsELF3, LOC_Os06g05060; Ehd1, AB092506; OsGI,AJ133787; Hd3a, AB052944; and RFT1, AB062676.

Supplemental Data

Supplemental Figure 1. Diagram of HAF1 truncations and transcrip-tional activation activity assays in yeast.

Supplemental Figure 2. Complementation test of haf1.

Supplemental Figure 3. Expression profiles and subcellular localiza-tion of HAF1.

Supplemental Figure 4. Purified MBP-HAF1, GST-HAF1, and GST-Hd1 fusion protein expressed in E. coli.

Supplemental Figure 5. HAF1 ubiquitinates Hd1 in vitro.

Supplemental Figure 6. Cell-free degradation assay of MBP-Hd1.

Supplemental Figure 7. Phenotype of haf1 hd1 double mutants underSD conditions.

Supplemental Figure 8. Phenotype of haf1 hd1 double mutants underLD conditions.

Supplemental Table 1. Comparison of agronomic traits between haf1mutants and wild-type plants under natural LD field conditions inWuhan (2010).

Supplemental Table 2. Comparison of photomorphogenesis betweenthe wild type and haf1 under red, far-red, blue, and dark conditions.

Supplemental Data Set 1. Primers for genotyping, RT-PCR, qRT-PCR, and plasmid construction.

ACKNOWLEDGMENTS

This work was supported by National Program on the Development ofBasic Research (2013CB126900), National 863 Project (2012AA10A303),National Natural Science Foundation of China (31370222 and 31425018),Specialized Research Fund for the Doctoral Program of Higher Education(20130146110012), and Fundamental Research Funds for the CentralUniversities. We thank Yunhai Li and Ran Xu (Chinese Academy of Sci-ences) for technical assistancewith the ubiquitination assays andWeijiangLuan (Tianjin Normal University) for providing the hd1 mutant.

AUTHOR CONTRIBUTIONS

C.W., Y.Y., andD.F. conceived this project anddesigned the research. Y.Y.identified the haf1mutant and performed the genetic transformation. Y.Y.,D. F., C.Z., Y.Z., H.Z., and T.L. performed the genetic analysis. D.F., C.Z.,and X. L. performed the ubiquitin activities analyses. C.Y. supervised thestudy. C.W., Y.Y., and D.F. analyzed the data and wrote the article.

Received April 14, 2015; revised July 2, 2015; accepted July 31, 2015;published August 21, 2015.

REFERENCES

Borden, K.L., and Freemont, P.S. (1996). The RING finger domain:a recent example of a sequence-structure family. Curr. Opin. Struct.Biol. 6: 395–401.

Chen, L., and Hellmann, H. (2013). Plant E3 ligases: flexible enzymesin a sessile world. Mol. Plant 6: 1388–1404.

Deng, X.W., Caspar, T., and Quail, P.H. (1991). cop1: a regulatorylocus involved in light-controlled development and gene expressionin Arabidopsis. Genes Dev. 5: 1172–1182.

Deng, X.W., Matsui, M., Wei, N., Wagner, D., Chu, A.M., Feldmann,K.A., and Quail, P.H. (1992). COP1, an Arabidopsis regulatorygene, encodes a protein with both a zinc-binding motif and a G betahomologous domain. Cell 71: 791–801.

Doi, K., Izawa, T., Fuse, T., Yamanouchi, U., Kubo, T., Shimatani,Z., Yano, M., and Yoshimura, A. (2004). Ehd1, a B-type responseregulator in rice, confers short-day promotion of flowering andcontrols FT-like gene expression independently of Hd1. Genes Dev.18: 926–936.

Feltham, R., Bettjeman, B., Budhidarmo, R., Mace, P.D., Shirley,S., Condon, S.M., Chunduru, S.K., McKinlay, M.A., Vaux, D.L.,Silke, J., and Day, C.L. (2011). Smac mimetics activate the E3 li-gase activity of cIAP1 protein by promoting RING domain di-merization. J. Biol. Chem. 286: 17015–17028.

Fowler, S., Lee, K., Onouchi, H., Samach, A., Richardson, K.,Morris, B., Coupland, G., and Putterill, J. (1999). GIGANTEA:a circadian clock-controlled gene that regulates photoperiodicflowering in Arabidopsis and encodes a protein with several pos-sible membrane-spanning domains. EMBO J. 18: 4679–4688.

Hayama, R., Yokoi, S., Tamaki, S., Yano, M., and Shimamoto, K.(2003). Adaptation of photoperiodic control pathways producesshort-day flowering in rice. Nature 422: 719–722.

Huang, X., et al. (2012). Genome-wide association study of floweringtime and grain yield traits in a worldwide collection of rice germ-plasm. Nat. Genet. 44: 32–39.

Ishikawa, R., Tamaki, S., Yokoi, S., Inagaki, N., Shinomura, T.,Takano, M., and Shimamoto, K. (2005). Suppression of the floralactivator Hd3a is the principal cause of the night break effect in rice.Plant Cell 17: 3326–3336.

Ishikawa, R., Aoki, M., Kurotani, K., Yokoi, S., Shinomura, T.,Takano, M., and Shimamoto, K. (2011). Phytochrome B regulatesHeading date 1 (Hd1)-mediated expression of rice florigen Hd3a andcritical day length in rice. Mol. Genet. Genomics 285: 461–470.

Itoh, J.I., Hasegawa, A., Kitano, H., and Nagato, Y. (1998). A re-cessive heterochronic mutation, plastochron1, shortens the plas-tochron and elongates the vegetative phase in rice. Plant Cell 10:1511–1522.

Izawa, T. (2007). Adaptation of flowering-time by natural and artificialselection in Arabidopsis and rice. J. Exp. Bot. 58: 3091–3097.

Izawa, T., Oikawa, T., Sugiyama, N., Tanisaka, T., Yano, M., andShimamoto, K. (2002). Phytochrome mediates the external lightsignal to repress FT orthologs in photoperiodic flowering of rice.Genes Dev. 16: 2006–2020.

Jang, S., Marchal, V., Panigrahi, K.C., Wenkel, S., Soppe, W.,Deng, X.W., Valverde, F., and Coupland, G. (2008). ArabidopsisCOP1 shapes the temporal pattern of CO accumulation conferringa photoperiodic flowering response. EMBO J. 27: 1277–1288.

Johnson, A.E., Collier, S.E., Ohi, M.D., and Gould, K.L. (2012).Fission yeast Dma1 requires RING domain dimerization for itsubiquitin ligase activity and mitotic checkpoint function. J. Biol.Chem. 287: 25741–25748.

Komiya, R., Ikegami, A., Tamaki, S., Yokoi, S., and Shimamoto, K.(2008). Hd3a and RFT1 are essential for flowering in rice. De-velopment 135: 767–774.

2466 The Plant Cell

Page 13: The RING-Finger Ubiquitin Ligase HAF1 Mediates Heading date 1 …croplab.hzau.edu.cn/__local/8/44/09/7F493FBF4BE3C2A36DFF... · 2019-04-22 · The RING-Finger Ubiquitin Ligase HAF1

Komiya, R., Yokoi, S., and Shimamoto, K. (2009). A gene network forlong-day flowering activates RFT1 encoding a mobile floweringsignal in rice. Development 136: 3443–3450.

Lazaro, A., Valverde, F., Piñeiro, M., and Jarillo, J.A. (2012). TheArabidopsis E3 ubiquitin ligase HOS1 negatively regulates CON-STANS abundance in the photoperiodic control of flowering. PlantCell 24: 982–999.

Liew, C.W., Sun, H., Hunter, T., and Day, C.L. (2010). RING domaindimerization is essential for RNF4 function. Biochem. J. 431: 23–29.

Liu, K., Wang, J., Li, H., Xu, C., Liu, A., Li, X., and Zhang, Q. (1997). Agenome-wide analysis of wide compatibility in rice and the preciselocation of the S5 locus in the molecular map. Theor. Appl. Genet.95: 809–814.

Liu, L., Zhang, Y., Tang, S., Zhao, Q., Zhang, Z., Zhang, H., Dong,L., Guo, H., and Xie, Q. (2010). An efficient system to detect proteinubiquitination by agroinfiltration in Nicotiana benthamiana. Plant J.61: 893–903.

Liu, L.J., Zhang, Y.C., Li, Q.H., Sang, Y., Mao, J., Lian, H.L., Wang,L., and Yang, H.Q. (2008). COP1-mediated ubiquitination ofCONSTANS is implicated in cryptochrome regulation of flowering inArabidopsis. Plant Cell 20: 292–306.

Livak, K.J., and Schmittgen, T.D. (2001). Analysis of relative geneexpression data using real-time quantitative PCR and the 2(-D D C(T))

Method. Methods 25: 402–408.Luan, W., Chen, H., Fu, Y., Si, H., Peng, W., Song, S., Liu, W., Hu, G.,

Sun, Z., Xie, D., and Sun, C. (2009). The effect of the crosstalkbetween photoperiod and temperature on the heading-date in rice.PLoS One 4: e5891.

Luo, Q., Lian, H.L., He, S.B., Li, L., Jia, K.P., and Yang, H.Q. (2014).COP1 and phyB physically interact with PIL1 to regulate its stabilityand photomorphogenic development in Arabidopsis. Plant Cell 26:2441–2456.

Mace, P.D., Linke, K., Feltham, R., Schumacher, F.R., Smith, C.A.,Vaux, D.L., Silke, J., and Day, C.L. (2008). Structures of the cIAP2RING domain reveal conformational changes associated with ubiquitin-conjugating enzyme (E2) recruitment. J. Biol. Chem. 283: 31633–31640.

McNellis, T.W., von Arnim, A.G., Araki, T., Komeda, Y., Miséra, S.,and Deng, X.W. (1994). Genetic and molecular analysis of an allelicseries of cop1 mutants suggests functional roles for the multiple proteindomains. Plant Cell 6: 487–500.

McNellis, T.W., Torii, K.U., and Deng, X.W. (1996). Expression of anN-terminal fragment of COP1 confers a dominant-negative effect onlight-regulated seedling development in Arabidopsis. Plant Cell 8:1491–1503.

Murray, M.G., and Thompson, W.F. (1980). Rapid isolation of highmolecular weight plant DNA. Nucleic Acids Res. 8: 4321–4325.

Ogiso, E., Takahashi, Y., Sasaki, T., Yano, M., and Izawa, T. (2010).The role of casein kinase II in flowering time regulation has di-versified during evolution. Plant Physiol. 152: 808–820.

Onouchi, H., Igeño, M.I., Périlleux, C., Graves, K., and Coupland, G.(2000). Mutagenesis of plants overexpressing CONSTANS demon-strates novel interactions among Arabidopsis flowering-time genes.Plant Cell 12: 885–900.

Plechanovová, A., Jaffray, E.G., McMahon, S.A., Johnson, K.A.,Navrátilová, I., Naismith, J.H., and Hay, R.T. (2011). Mechanism ofubiquitylation by dimeric RING ligase RNF4. Nat. Struct. Mol. Biol.18: 1052–1059.

Rodgers, K.K., Bu, Z., Fleming, K.G., Schatz, D.G., Engelman, D.M., and Coleman, J.E. (1996). A zinc-binding domain involved inthe dimerization of RAG1. J. Mol. Biol. 260: 70–84.

Schwechheimer, C., and Deng, X.W. (2000). The COP/DET/FUSproteins-regulators of eukaryotic growth and development. Semin.Cell Dev. Biol. 11: 495–503.

Song, Y.H., Shim, J.S., Kinmonth-Schultz, H.A., and Imaizumi, T.(2015). Photoperiodic flowering: time measurement mechanisms inleaves. Annu. Rev. Plant Biol. 66: 10.1–10.24.

Stone, S.L., Hauksdóttir, H., Troy, A., Herschleb, J., Kraft, E., andCallis, J. (2005). Functional analysis of the RING-type ubiquitin li-gase family of Arabidopsis. Plant Physiol. 137: 13–30.

Sun, S.Y., Chao, D.Y., Li, X.M., Shi, M., Gao, J.P., Zhu, M.Z., Yang,H.Q., Luan, S., and Lin, H.X. (2009). OsHAL3 mediates a newpathway in the light-regulated growth of rice. Nat. Cell Biol. 11: 845–851.

Tabuchi, H., et al. (2011). LAX PANICLE2 of rice encodes a novelnuclear protein and regulates the formation of axillary meristems.Plant Cell 23: 3276–3287.

Takahashi, Y., Teshima, K.M., Yokoi, S., Innan, H., andShimamoto, K. (2009). Variations in Hd1 proteins, Hd3a promoters,and Ehd1 expression levels contribute to diversity of flowering time incultivated rice. Proc. Natl. Acad. Sci. USA 106: 4555–4560.

Takano, H., Obitsu, S., Beppu, T., and Ueda, K. (2005). Light-induced carotenogenesis in Streptomyces coelicolor A3(2): identi-fication of an extracytoplasmic function sigma factor that directsphotodependent transcription of the carotenoid biosynthesis genecluster. J. Bacteriol. 187: 1825–1832.

Tamaki, S., Matsuo, S., Wong, H.L., Yokoi, S., and Shimamoto, K.(2007). Hd3a protein is a mobile flowering signal in rice. Science316: 1033–1036.

Tanaka, N., Itoh, H., Sentoku, N., Kojima, M., Sakakibara, H.,Izawa, T., Itoh, J., and Nagato, Y. (2011). The COP1 ortholog PPSregulates the juvenile-adult and vegetative-reproductive phasechanges in rice. Plant Cell 23: 2143–2154.

Tsuji, H., Taoka, K., and Shimamoto, K. (2011). Regulation of flow-ering in rice: two florigen genes, a complex gene network, andnatural variation. Curr. Opin. Plant Biol. 14: 45–52.

Turck, F., Fornara, F., and Coupland, G. (2008). Regulation andidentity of florigen: FLOWERING LOCUS T moves center stage.Annu. Rev. Plant Biol. 59: 573–594.

Valverde, F., Mouradov, A., Soppe, W., Ravenscroft, D., Samach,A., and Coupland, G. (2004). Photoreceptor regulation of CON-STANS protein in photoperiodic flowering. Science 303: 1003–1006.

Waadt, R., Schmidt, L.K., Lohse, M., Hashimoto, K., Bock, R., andKudla, J. (2008). Multicolor bimolecular fluorescence complemen-tation reveals simultaneous formation of alternative CBL/CIPKcomplexes in planta. Plant J. 56: 505–516.

Wang, F., and Shi, Y. (2014). Progress in structural biology of 26Sproteasome. Scientia Sinica Vitae. 44: 965–974.

Wang, H., Ma, L.G., Li, J.M., Zhao, H.Y., and Deng, X.W. (2001).Direct interaction of Arabidopsis cryptochromes with COP1 in lightcontrol development. Science 294: 154–158.

Wu, C., Li, X., Yuan, W., Chen, G., Kilian, A., Li, J., Xu, C., Li, X.,Zhou, D.X., Wang, S., and Zhang, Q. (2003). Development of en-hancer trap lines for functional analysis of the rice genome. Plant J.35: 418–427.

Xia, T., Li, N., Dumenil, J., Li, J., Kamenski, A., Bevan, M.W., Gao,F., and Li, Y. (2013). The ubiquitin receptor DA1 interacts with theE3 ubiquitin ligase DA2 to regulate seed and organ size in Arabi-dopsis. Plant Cell 25: 3347–3359.

Xu, D., Lin, F., Jiang, Y., Huang, X., Li, J., Ling, J., Hettiarachchi, C.,Tellgren-Roth, C., Holm, M., and Deng, X.W. (2014). The RING-finger E3 ubiquitin ligase COP1 SUPPRESSOR1 negatively regulatesCOP1 abundance in maintaining COP1 homeostasis in dark-grownArabidopsis seedlings. Plant Cell 26: 1981–1991.

Xue, W., Xing, Y., Weng, X., Zhao, Y., Tang, W., Wang, L., Zhou, H.,Yu, S., Xu, C., Li, X., and Zhang, Q. (2008). Natural variation in

Hd1 Ubiquitination by HAF1 2467

Page 14: The RING-Finger Ubiquitin Ligase HAF1 Mediates Heading date 1 …croplab.hzau.edu.cn/__local/8/44/09/7F493FBF4BE3C2A36DFF... · 2019-04-22 · The RING-Finger Ubiquitin Ligase HAF1

Ghd7 is an important regulator of heading date and yield potential inrice. Nat. Genet. 40: 761–767.

Yang, H.Q., Tang, R.H., and Cashmore, A.R. (2001). The signalingmechanism of Arabidopsis CRY1 involves direct interaction withCOP1. Plant Cell 13: 2573–2587.

Yang, Y., Peng, Q., Chen, G.X., Li, X.H., and Wu, C.Y. (2013).OsELF3 is involved in circadian clock regulation for promotingflowering under long-day conditions in rice. Mol. Plant 6: 202–215.

Yano, M., Katayose, Y., Ashikari, M., Yamanouchi, U., Monna, L.,Fuse, T., Baba, T., Yamamoto, K., Umehara, Y., Nagamura, Y.,and Sasaki, T. (2000). Hd1, a major photoperiod sensitivity quan-titative trait locus in rice, is closely related to the Arabidopsisflowering time gene CONSTANS. Plant Cell 12: 2473–2484.

Yu, J.W., et al. (2008). COP1 and ELF3 control circadian function andphotoperiodic flowering by regulating GI stability. Mol. Cell 32: 617–630.

Yuan, S., Zhu, H., Gou, H., Fu, W., Liu, L., Chen, T., Ke, D., Kang, H.,Xie, Q., Hong, Z., and Zhang, Z. (2012). A ubiquitin ligase ofsymbiosis receptor kinase involved in nodule organogenesis. PlantPhysiol. 160: 106–117.

Zhang, J., et al. (2007). Non-random distribution of T-DNA insertionsat various levels of the genome hierarchy as revealed by analyzing13 804 T-DNA flanking sequences from an enhancer-trap mutantlibrary. Plant J. 49: 947–959.

Zhao, J., Huang, X., Ouyang, X., Chen, W., Du, A., Zhu, L., Wang, S.,Deng, X.W., and Li, S. (2012). OsELF3-1, an ortholog of Arabi-dopsis early flowering 3, regulates rice circadian rhythm and pho-toperiodic flowering. PLoS One 7: e43705.

Zhao, K., et al. (2011). Genome-wide association mapping revealsa rich genetic architecture of complex traits in Oryza sativa. Nat.Commun. 2: 467.

2468 The Plant Cell

Page 15: The RING-Finger Ubiquitin Ligase HAF1 Mediates Heading date 1 …croplab.hzau.edu.cn/__local/8/44/09/7F493FBF4BE3C2A36DFF... · 2019-04-22 · The RING-Finger Ubiquitin Ligase HAF1

1

Supplemental Data. Yang et al. (2015). Plant Cell. 10.1105/tpc.15.00320.

Supplemental Figure 1. Diagram of HAF1 truncations and transcriptional activation activity assays in yeast (A) Schematic representation of the truncated HAF1 protein. Black box represent RING

finger domain. (B) The transcriptional activation activity assays of the truncated proteins

from (A). pGBK-LAX1 (Yoshida et al., 2013) and pGBK-T7 were separately used as

positive and negative controls.

Page 16: The RING-Finger Ubiquitin Ligase HAF1 Mediates Heading date 1 …croplab.hzau.edu.cn/__local/8/44/09/7F493FBF4BE3C2A36DFF... · 2019-04-22 · The RING-Finger Ubiquitin Ligase HAF1

2

Supplemental Data. Yang et al. (2015). Plant Cell. 10.1105/tpc.15.00320.

Supplemental Figure 2. Complementation test of haf1

(A) RT-PCR analyses of the expression levels of genes surrounding the insertion site in

haf1 and wild type (WT). RNA was extracted from leaves under LDs. GAPDH was used

as a control. (B) Schematic representation of the construct for genetic complementation

test. (C) Copy number of 38 T0 complementary transgenic lines through DNA gel

blotting. The single-copy transgenic lines was indicated by arrows. (D) Days to heading

in the T1 generation of pC2301-HAF1 (#2, #6, #9, #14) and pCAMBIA2301 transgenic

positive plants under LDs and SDs. Error bars indicate standard deviations; n = 40

plants.

Page 17: The RING-Finger Ubiquitin Ligase HAF1 Mediates Heading date 1 …croplab.hzau.edu.cn/__local/8/44/09/7F493FBF4BE3C2A36DFF... · 2019-04-22 · The RING-Finger Ubiquitin Ligase HAF1

3

Supplemental Data. Yang et al. (2015). Plant Cell. 10.1105/tpc.15.00320.

Supplemental Figure 3. Expression profiles and subcellular localization of HAF1 (A) RT-qPCR analysis of HAF1 expression levels in roots (1), leaves (2) of

seedling, roots (3), leaves (4), sheathes (5), culms (6), and panicles of 1 cm (7), 2

cm (8), 4.5 cm (9), 8 cm (10), 16.5 cm (11), and 19 cm (12) from mature plants from

ZH11. Ubiquitin was used as a control. The data shown are the means ± SE of three

independent experiments. (B) Expression patterns of HAF1 at various developmental

stages of leaf. The numbers below the x axis indicate the days after soaking. Ubiquitin was used as a control. The data shown are the means ± SE of three independent

experiments. (C) Subcellular localization of HAF1. Hd1-fused cyan fluorescent protein

(CFP) was used as a nuclear marker. Scale bars = 10 µm.

Page 18: The RING-Finger Ubiquitin Ligase HAF1 Mediates Heading date 1 …croplab.hzau.edu.cn/__local/8/44/09/7F493FBF4BE3C2A36DFF... · 2019-04-22 · The RING-Finger Ubiquitin Ligase HAF1

4

Supplemental Data. Yang et al. (2015). Plant Cell. 10.1105/tpc.15.00320.

Supplemental Figure 4. Purified MBP-HAF1, GST-HAF1, and GST-Hd1 fusion protein expressed in E. coli (A) Purified MBP-HAF1 fusion protein expressed in E. coli. Lane 1 indicates MBP protein,

lane 2 indicates GST protein, and lane 3 indicates MBP-HAF1 fusion protein. (B) Purified

GST-HAF1 and GST-Hd1 fusion protein expressed in E. coli. Lane 1 indicates GST

protein, lane 2 indicates GST-HAF1 fusion protein, and lane 3 indicates GST-Hd1 fusion

protein.

Page 19: The RING-Finger Ubiquitin Ligase HAF1 Mediates Heading date 1 …croplab.hzau.edu.cn/__local/8/44/09/7F493FBF4BE3C2A36DFF... · 2019-04-22 · The RING-Finger Ubiquitin Ligase HAF1

5

Supplemental Data. Yang et al. (2015). Plant Cell. 10.1105/tpc.15.00320.

Supplemental Figure 5. HAF1 ubiquitinates Hd1 in vitro

GST-Hd1 ubiquitination assays were performed using MBP-HAF1, E1, E2, and His-Ub,

immunoblot was performed with anti-His (top panel) and anti-MBP (bottom panel)

antibodies. Asterisk in top panel indicates the ubiquitinated GST-Hd1.

Page 20: The RING-Finger Ubiquitin Ligase HAF1 Mediates Heading date 1 …croplab.hzau.edu.cn/__local/8/44/09/7F493FBF4BE3C2A36DFF... · 2019-04-22 · The RING-Finger Ubiquitin Ligase HAF1

6

Supplemental Data. Yang et al. (2015). Plant Cell. 10.1105/tpc.15.00320.

Supplemental Figure 6. Cell-free degradation assay of MBP-Hd1 In vitro cell-free degradation assay showing the delayed degradation of expressed

recombinant MBP-Hd1(A), but not MBP protein (C), in haf1 and in the presence of 50 μM

MG132 of ZH11(Wild type). ACTIN abundance was detected as loading control (B).

The protein levels at different time points were detected by immunoblotting.

Page 21: The RING-Finger Ubiquitin Ligase HAF1 Mediates Heading date 1 …croplab.hzau.edu.cn/__local/8/44/09/7F493FBF4BE3C2A36DFF... · 2019-04-22 · The RING-Finger Ubiquitin Ligase HAF1

7

Supplemental Data. Yang et al. (2015). Plant Cell. 10.1105/tpc.15.00320.

Supplemental Figure 7. Phenotype of haf1 hd1 double mutants under SD conditions

(A) Phenotypes of WT, haf1, hd1, and haf1 hd1 at heading stage. (B) Genotyping of HAF1 and Hd1 segregates with primers. (C) Days to heading in WT, haf1, hd1, and haf1 hd1 plants under SD conditions. Error bars indicate standard deviations; n = 10 plants.

Page 22: The RING-Finger Ubiquitin Ligase HAF1 Mediates Heading date 1 …croplab.hzau.edu.cn/__local/8/44/09/7F493FBF4BE3C2A36DFF... · 2019-04-22 · The RING-Finger Ubiquitin Ligase HAF1

8

Supplemental Data. Yang et al. (2015). Plant Cell. 10.1105/tpc.15.00320.

Supplemental Figure 8. Phenotype of haf1 hd1 double mutants under LD conditions

(A) Phenotypes of WT, haf1, hd1, and haf1 hd1 at heading stage under LD conditions. (B)

Genotyping of HAF1 and Hd1 segregates with primers. (C) Days to heading in WT, haf1, hd1, and haf1 hd1 plants under LD conditions. Error bars indicate standard deviations; n =

10 plants.

Page 23: The RING-Finger Ubiquitin Ligase HAF1 Mediates Heading date 1 …croplab.hzau.edu.cn/__local/8/44/09/7F493FBF4BE3C2A36DFF... · 2019-04-22 · The RING-Finger Ubiquitin Ligase HAF1

9

Supplemental Data. Yang et al. (2015). Plant Cell. 10.1105/tpc.15.00320.

Supplemental Table 1. Comparison of agronomic traits between haf1 mutants and wild-type plants under natural LD field conditions in Wuhan (2010).

The data corresponded to the mean ± standard error from 40 individuals.

*t-test showed p-value > 0.05.

Supplemental Table 2. Comparison of photomorphogenesis between wild-type and haf1 under red (R), far-red (FR), blue (B), and dark conditions.

Light Treatments Tissues Wild type

(mm)

haf1

(mm)

p value*

Rc Coleoptile lengths 3.62 ± 0.87 3.06 ± 0.67 0.09

Lengths of first leaves 13.27 ± 1.62 12.84 ± 1.60 0.07

FRc Coleoptile lengths 5.91 ± 1.66 5.93 ± 0.92 0.96

Lengths of first leaves 16.29 ± 4.21 16.87 ± 3.97 0.07

Bc Coleoptile lengths 2.69 ± 0.62 3.08 ± 0.95 0.20

Lengths of first leaves 7.86 ± 1.38 8.05 ± 0.68 0.54

Dark Coleoptile lengths 20.95 ± 2.66 18.77 ± 2.68 0.15

Lengths of first leaves 34.97 ± 3.77 32.70 ± 2.54 0.47

The data corresponded to the mean ± standard error from 10 to 20 individuals.

*t-test showed p-value > 0.05.

Traits Wild type haf1 p value*

Plant height (cm) 86.15 ± 4.13 87.22 ± 3.52 0.68

Tillers number per plant 8.80 ± 3.56 9.42 ± 3.39 0.35

1,000-grain weight (g) 30.33 ± 6.16 28.69 ± 1.28 0.82

Panicle length (cm) 17.04 ± 0.17 18.26 ± 1.03 0.45

Spikelets number per

panicle 125.12 ± 7.7 125.6 ± 35.80 0.92

Page 24: The RING-Finger Ubiquitin Ligase HAF1 Mediates Heading date 1 …croplab.hzau.edu.cn/__local/8/44/09/7F493FBF4BE3C2A36DFF... · 2019-04-22 · The RING-Finger Ubiquitin Ligase HAF1

10

Supplemental Data. Yang et al. (2015). Plant Cell. 10.1105/tpc.15.00320.

Supplemental References 1. Yoshida, A., Sasao, M., Yasuno, N., Takagi, K., Daimon, Y., Chen, R., Yamazaki,

R., Tokunaga, H., Kitaguchi, Y., Sato, Y., Nagamura, Y., Ushijima, T., Kumamaru,T., Iida, S., Maekawa, M., and Kyozuka J. (2013). TAWAWA1, a regulator of rice

inflorescence architecture, functions through the suppression of meristem phase

transition. Proc. Natl. Acad. Sci. U.S.A 110: 767–772.


Top Related