Inflorescence Meristem Identity in Rice Is Specified byOverlapping Functions of Three AP1/FUL-Like MADS BoxGenes and PAP2, a SEPALLATA MADS Box Gene C W

Kaoru Kobayashi,a Naoko Yasuno,a Yutaka Sato,b Masahiro Yoda,a Ryo Yamazaki,a Mayumi Kimizu,c

Hitoshi Yoshida,d Yoshiaki Nagamura,b and Junko Kyozukaa,1

aGraduate School of Agriculture and Life Sciences, University of Tokyo, Yayoi, Bunkyo, Tokyo 113-8657, JapanbNational Institute of Agrobiological Sciences, Tsukuba, Ibaraki 305-8602, JapancCrop Development Division, National Agricultural Research Center, National Agriculture and Food Research Organization,Jo-etsu, Niigata 943-0193, JapandRice Research Division, National Institute of Crop Science, National Agriculture and Food Research Organization,Tsukuba 305-8518, Japan

In plants, the transition to reproductive growth is of particular importance for successful seed production. Transformation ofthe shoot apical meristem (SAM) to the inflorescence meristem (IM) is the crucial first step in this transition. Using lasermicrodissection and microarrays, we found that expression of PANICLE PHYTOMER2 (PAP2) and three APETALA1 (AP1)/FRUITFULL (FUL)-like genes (MADS14,MADS15, andMADS18) is induced in the SAM during meristem phase transition in rice(Oryza sativa). PAP2 is a MADS box gene belonging to a grass-specific subclade of the SEPALLATA subfamily. Suppression ofthese three AP1/FUL-like genes by RNA interference caused a slight delay in reproductive transition. Further depletion ofPAP2 function from these triple knockdown plants inhibited the transition of the meristem to the IM. In the quadrupleknockdown lines, the meristem continued to generate leaves, rather than becoming an IM. Consequently, multiple shootswere formed instead of an inflorescence. PAP2 physically interacts with MAD14 and MADS15 in vivo. Furthermore, theprecocious flowering phenotype caused by the overexpression of Hd3a, a rice florigen gene, was weakened in pap2-1mutants. Based on these results, we propose that PAP2 and the three AP1/FUL-like genes coordinately act in the meristem tospecify the identity of the IM downstream of the florigen signal.

INTRODUCTION

Plants progress through distinct developmental phases in theirlife cycles (reviewed in Huijser and Schmid, 2011). The transitionfrom vegetative to reproductive growth, generally called flow-ering, is vital for reproductive success. Flowering occurs in re-sponse to environmental and developmental signals, such asdaylength, plant age, and temperature (reviewed in Liu et al.,2009; Wellmer and Riechmann, 2010). Multiple flowering signalsare eventually transmitted to the shoot apical meristem (SAM),a group of stem cells in the shoot apex, where they are con-verted to local cues that evoke the developmental phasetransition. In Arabidopsis thaliana, SUPPRESSOR OF OVER-EXPRESSION OF CO1 (SOC1) and FLOWERING LOCUS T (FT)function as the main floral integrators that transmit multipleflowering signals to the meristem. SOC1 is one of the first genesto be induced in the SAM at this transition (Lee et al., 2000). FT is

a major component of florigen, a long-distance flowering signal(Abe et al., 2005; Wigge et al., 2005). The FT protein is producedin leaves and then transported to the SAM, where it interactswith FLOWERING LOCUS D (FD). In the SAM, the SOC1 and FT-FD complex promote expression of several downstream genes,including LEAFY, APETALA1 (AP1), and FRUITFULL (FUL),which specify floral meristem identity (Abe et al., 2005; Wiggeet al., 2005; Corbesier et al., 2007).The aerial part of a plant is generated by the SAM, which

produces leaves during vegetative development. Upon transi-tion to reproductive development, the SAM converts into aninflorescence meristem (IM) and starts to generate componentsof the inflorescence. Development of the IM is variable de-pending on species, leading to formation of species-specificinflorescence architecture. The conversion from the vegetativeSAM to the IM is less well understood at the molecular level thanevents such as flowering time control, floral meristem specifi-cation, and flower development. Several genes show a dramaticchange in their expression patterns in the meristem at phasetransition (Cardon et al., 1997; Hempel et al., 1997; Lee et al.,2000; Samach et al., 2000; Torti et al., 2012). This reflectsa global change in meristem gene expression that is associatedwith the establishment of IM identity (Schmid et al., 2003; Parket al., 2012).In grasses, the shift of the meristem phase from the SAM

to the IM is accompanied by clear changes in a number of

1 Address correspondence to akyozuka@mail.ecc.u-tokyo.ac.jp.The 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: Junko Kyozuka(akyozuka@mail.ecc.u-tokyo.ac.jp).C Some figures in this article are displayed in color online but in black andwhite in the print edition.WOnline version contains Web-only data.www.plantcell.org/cgi/doi/10.1105/tpc.112.097105

The Plant Cell, Vol. 24: 1848–1859, May 2012, www.plantcell.org ã 2012 American Society of Plant Biologists. All rights reserved.

developmental and morphological traits. In rice (Oryza sativa),the first morphological change is a rapid increase in meristemsize (Hoshikawa, 1989). The IM starts to initiate inflorescencebranch meristem (BM) primordia. Heading date 3a (Hd3a) andRICE FLOWERING LOCUS T1 (RFT1) function as major com-ponents of florigen in rice (Tamaki et al., 2007; Komiya et al.,2008, 2009). Recent findings show that Hd3a protein generatedin leaves is transported to the cytoplasm of shoot apex cellswhere it interacts with 14-3-3 proteins (Taoka et al., 2011). TheHd3a/14-3-3 complex then enters the nucleus and activates therice ortholog of FD through phosphorylation. The absence ofboth Hd3a and RFT1 functions completely blocks transition toreproductive growth (Komiya et al., 2009). This implies that, inthe meristem, FD and possibly other downstream targets ofHd3a and RFT1 induce genes that promote meristem phasetransition. Expression of both MADS14 and MADS15, whichencode AP1/FUL-like MADS domain proteins, was suppressedin Hd3a and RFT1 double knockdown plants, suggesting thatthese genes function downstream of the florigen signal (Komiyaet al., 2008, 2009). Moreover, it has been demonstrated thatMADS15 transcription is induced by phosphorylated FD in thepresence of Hd3a (Taoka et al., 2011). These reports suggestthat MADS14 and MADS15 may be regulators of phase transi-tion in the meristem; however, neither their functions nor theirexpression patterns in the meristem at phase transition havebeen demonstrated.

In this study, we demonstrate that four MADS box genes,PANICLE PHYTOMER2 (PAP2) (Kobayashi et al., 2010), MADS14,MADS15, and MADS18 are activated in the meristem at phasetransition and function redundantly in specifying IM identity in rice.

RESULTS

PAP2 Expression Is Induced in the Meristem uponReproductive Transition

Rice PAP2 encodes MADS34, a grass-specific SEPALLATA(SEP) subfamily MADS domain protein (see Supplemental Figure1 online). PAP2 plays multiple roles, including the determinationof spikelet meristem (SM) identity and the suppression of theextra growth of glumes (Gao et al., 2010; Kobayashi et al., 2010).To understand PAP2 function further, we analyzed its spatialand temporal expression throughout inflorescence devel-opment. A 2.2-kb fragment of the PAP2 promoter region wasfused with the b-glucuronidase (GUS) gene and transformed intowild-type rice plants. T1 seedlings were grown for 7 d underlong-day (LD) conditions and then transferred to short-day (SD)conditions, which promote the transition to the reproductivephase. GUS activity was not detected in the SAM during thevegetative phase (Figures 1A and 1B). Upon transition to re-productive development, clear induction of PAP2:GUS activitythroughout the SAM was observed (Figure 1C). GUS activitybecame localized to the incipient SM at the spikelet initiationstage (Figures 1D and 1E). During spikelet organ development,GUS activity was observed in initiating glumes (Figure 1F). PAP2expression in the SM and glumes was consistent with the knownfunctions of PAP2, namely, promotion of spikelet identity and

suppression of extra growth of glumes (Kobayashi et al., 2010).Localization of PAP2 mRNA in the meristem at transition stageand the SMs was confirmed by in situ hybridization analysis(Figures 1G and 1H). To determine the timing of PAP2 inductionprecisely, the relationship between the meristem size and theonset of PAP2:GUS activity was examined (Figure 1I). Elonga-tion of the SAM, a sign of the transition to reproductive growth,was accompanied by a dramatic increase in GUS activity. Theseresults suggest that PAP2 is involved in reproductive transitionand/or the establishment of the identity of the IM.

Figure 1. Spatial and Temporal Distribution of PAP2 Promoter Activity.

(A) to (F) GUS staining in the shoot apex of PAP2:GUS plants at thevegetative stage (A), slightly before reproductive transition (B), at thetransition stage (C), and at the spikelet initiation stage ([D] and [E]). Aclose-up view of the region enclosed with a square in (D) is shown in (E).GUS activity in the developing spikelet is shown in (F).(G) and (H) In situ hybridization analysis of PAP2 expression in the shootapex at the transition stage (G) and in the inflorescence at spikelet ini-tiation stage. Arrowhead, SAM; BM, branch meristem; SM, spikeletmeristem; FM, floral meristem.(I) The relationship between meristem size and GUS activity is shown.

Inflorescence Meristem Identity of Rice 1849

A null allele of the PAP2 locus, pap2-1, does not show anysignificant changes in the very early stages of inflorescencedevelopment (Kobayashi et al., 2010). Because the SEP MADSdomain proteins usually function in complexes with other, re-lated MADS domain proteins (reviewed in Gramzow andTheissen, 2010), we postulated that the effects of the loss ofPAP2 activity during reproductive phase transition may bemasked by redundancy with other gene(s). To test this idea, wesearched for MADS box genes that show overlapping expres-sion patterns with PAP2 in the SAM during the reproductivetransition. For this purpose, we adopted a laser microdissection(LM) and microarray technique and specifically sampled SAMsat various developmental stages (Figure 2A). All samples werecollected from plants grown in the field in 2008. SAMs from theVegetative 1 (V1) stage were harvested on June 24th before theinduction of RFT1, a primary determinant of rice flowering undernatural LD conditions (Komiya et al., 2009). V2 SAMs weresampled on July 8th, several days after RFT1 induction was firstobserved in the leaves. Because slight variations in size wereobserved in the SAMs collected on July 15th, each SAM wasidentified as being at either the transition (V/R) or elongation (R1)stage, depending on its size as measured under the microscope.In situ hybridization analysis of ABERRANT PANICLE ORGANI-ZATION1 (APO1) expression in V/R stage SAMs revealed a mixtureof APO1-expressing and nonexpressing SAMs (see SupplementalFigure 2 online). Because induction of APO1 is concurrent withreproductive transition (Ikeda-Kawakatsu et al., 2009), this indicatesthat the V/R stage is a critical time in the transition of the SAM tothe IM. Inflorescences at the primary branch initiation, secondarybranch initiation, and spikelet initiation stages were collected as R2,R3, and R4 stages (Figures 2A and 2B).

Microarray analysis was performed with RNAs isolated fromdissected meristems and very young inflorescences. The overallprofile of gene expression during meristem phase transition isshown in Supplemental Figure 3 online. The levels of PAP2 RNAincreased from the V2 to the V/R stage, and this was consistentwith changes in PAP2:GUS activity (Figures 2B and 2C). In ad-dition, expression of three MADS box genes (out of 34 MADS boxgenes on the microarray) showed increases in expression be-tween the V1 and the V/R stages (Figures 2B and 2C). Intriguingly,these three genes are MADS14, MADS15, and MADS18, whichbelong to the AP1/FUL-like subfamily of MADS box genes (Littand Irish 2003; Fornara et al., 2004; Preston and Kellogg, 2006;see Supplemental Figure 1 online). The only other member of thissubfamily in rice, MADS20, showed no detectable expression inmeristems at any of the stages examined in this study. Two of thethree MADS box genes,MADS14 and MADS18, displayed similarexpression patterns in which the mRNA level increased betweenthe V1 and V2 stages (Figure 2C). Accumulation of MADS14mRNA further increased from V2 to V/R, whereas the expressionof MADS18 remained relatively constant after the V2 stage.MADS15 showed an expression pattern similar to PAP2; that is,its mRNA level increased from the V2 to the V/R stage, accom-panying the transition to reproduction.

The SEP subfamily of rice consists of five genes, includingPAP2 (Malcomber and Kellogg, 2005; Zahn et al., 2005). Aspreviously reported (Arora et al., 2007; Kobayashi et al., 2010),none of the other four SEP genes, MADS1, MADS5, MADS7,

Figure 2. LM Analysis of Genes Expressed in Shoot Meristems duringReproductive Transition.

(A) Micrographs of the meristem stages that were analyzed. V1 and V2are meristems collected before transition. V/R and R1 are meristems atthe transition stage and elongation stage, respectively. R2 and R3 are theprimary branch initiation and secondary branch initiation stages of theinflorescences, respectively.(B) Heat map representing the expression of rice MADS box genes duringearly stages of inflorescence development. Genes that show a significantlevel of expression in at least one of the 24 arrays performed are included.The intensity of the red color represents absolute expression of each gene.(C) Expression patterns of four MADS box genes that showed a signifi-cant increase in expression level at reproductive transition. The valuesrepresent mean expression levels 6 SD (n = 3 to 5) obtained from themicroarray analysis of microdissected meristems.

1850 The Plant Cell

and MADS8, showed detectable expression at the early stagesof inflorescence development studied here except that a lowlevel of MADS5 expression was observed in the spikelet initia-tion stage (R4) (Figure 2B).

Inflorescence Development Was Severely Perturbed in theMADS Quadruple Knockdown Plants

To understand the role of PAP2 and the three AP1/FUL-likegenes during reproductive transition and early stages of inflo-rescence development, we generated transgenic rice plants inwhich the expression levels of MADS14, MADS15, or MADS18were reduced by RNA interference (RNAi). Despite a significantreduction in the levels of mRNA from each of these genes,neither inflorescence development nor flowering time was al-tered in single RNAi plants (see Supplemental Figure 4 online;Fornara et al., 2004). We then produced pMADS14;15;18i, anRNAi construct that simultaneously suppresses expression ofMADS14, MADS15, and MADS18. This construct was in-troduced into wild-type, pap2-1/+, and pap2-1 plants. Plantstransformed with the empty RNAi vector were used as controls.Specificity of the RNAi target for the three genes was confirmedby quantitative real-time RT-PCR (see Supplemental Figure 5online). The expression level ofMADS20, another member of theAP1/FUL-like genes of rice, was not clearly affected in trans-genic lines containing pMADS14;15;18i.

Growth of all transgenic plants was normal during the vege-tative phase, whereas inflorescence development was severelyperturbed in the homozygous mutant pap2-1 plants transformedwith pMADS14;15;18i (Figure 3). In normal rice development,reproductive transition is accompanied by the initiation of stemelongation. The stem gradually elongates during the early stagesof inflorescence development, with rapid elongation startinga few days before heading. This results in the exposure of theinflorescence as the flowers open (Figures 3A and 3B). Althoughextensively delayed, stem elongation finally occurred and pro-ceeded very slowly in the MADS14;15;18i pap2-1 quadrupleknockdown lines, indicating that reproductive transition beganin these plants. However, leaf initiation continued even after thestart of stem elongation (Figure 3C). The leaves produced duringthe later stages showed various abnormalities, including serra-tion and altered phyllotaxy (Figure 3C).

The severity of defects in inflorescence development variedamong transgenic lines. In weakly affected lines, althoughdelayed and reduced in size, the inflorescence was eventuallyformed. In moderate lines, the vegetative SAM was observed inthe growing inflorescence, indicating incomplete reproductivetransition (see Supplemental Figure 6 online). In strong lines, theinflorescence was not formed (see below). The severity roughlycorrelates with the effectiveness of knockdown of the three AP1/FUL-like genes (see Supplemental Figure 5 online). We chosetwo severe quadruple knockdown lines (q8 and q10) and ana-lyzed their reproductive development by scanning electron mi-croscopy (Figures 4A to 4J), sectioning (Figures 4K to 4R), in situhybridization (Figures 4S and 4T), and observation under themicroscope (Figures 4U to 4Z). In wild-type rice, primary in-florescence branches arise in the axils of bracts (extremely re-duced leaves) in a spiral phyllotaxy (Figures 4A to 4C, 4L, 4U,

and 4V). Each primary branch forms secondary branches andspikelets (Figures 4D, 4E, 4M, and 4N). Hair-like structures ariseon the bracts (Figure 4C), and the developing inflorescence iscovered with the hair-like structure by the spikelet initiationstage (Figures 4D, 4E, and 4V). Change to spiral phyllotaxy wasobserved in both q8 and q10, indicating the start of reproductivetransition and consistent with the start of stem elongation (Fig-ures 4F to 4I). However, unlike wild-type inflorescence de-velopment in which primary branches are formed, initiation ofvegetative leaf primordia continued (Figures 4G to 4J, 4P, 4Q,and 4R). Axillary meristems were formed in the axils of theseleaves and grew as shoots (Figures 4G, 4H, and 4W). Thisprocess was reiterated and led to the formation of multiple shootson one stem (Figures 4I, 4J, 4Q, 4X, 4Y, and 4Z). Spikelets wereproduced in the wild-type inflorescence (Figures 4M and 4N),whereas shoot formation was repeated and each meristem con-tinued to generate leaf primordia in the quadruple knockdownplants (Figures 4Q and 4R).To determine further the identity of meristems in the quadruple

knockdown plants, we performed in situ hybridization analysis ofAPO1, which is normally expressed in the IM and the BM but not inthe vegetative SAM (Ikeda-Kawakatsu et al., 2009). Absence of theAPO1 signal in the quadruple knockdown plants confirmed that theaxillary meristems retain the characteristics of the vegetative SAM(Figures 4S and 4T). Our combined data suggest that progressionto reproductive transition was severely blocked in MADS14;15;18ipap2-1 plants; thus, IM identity was not established. We proposethat PAP2 and three AP1/FUL-like genes act redundantly to pro-mote IM identity and are thus IM identity genes.

AP1/FUL-Like MADS Box Genes Are Involved in theRegulation of Flowering

In contrast with the severe defects observed in the quadrupleknockdown plants, few abnormalities was observed in in-florescence development of the MADS14;15;18i wild-type and

Figure 3. Inflorescence Development of MADS14;15;18i/pap2-1 Qua-druple Knockdown Plants.

(A) Nontransgenic wild-type (left) and MADS14;15;18i pap2-1 quadrupleknockdown (right) plants.(B) Inflorescences on a wild-type plant after heading.(C) Heading never occurred in the quadruple knockdown lines. Instead,leaf production continued after heading had occurred in the wild-typeplants. A close-up view of the youngest leaf showing a serration (ar-rowhead) is shown in the inset.[See online article for color version of this figure.]

Inflorescence Meristem Identity of Rice 1851

Figure 4. Impaired Inflorescence Development in MADS14;15;18i pap2-1 Quadruple Knockdown Plants.

(A) to (J) Scanning electron microscopy analysis of wild-type ([A] to [E]) and MADS14;15;18i pap2-1 ([F] to [J]) apices. Each shoot and leaf likestructure are indicated with an arrow and yellow dot, respectively ([F] and [G]).(K) to (N) Sections of wild-type inflorescences immediately after transition (K), at the inflorescence branch initiation stage (L), and at the spikeletinitiation stage (M). A close-up view of a developing spikelet from (M) is shown in (N).(O) to (R) Sectioning of an MADS14;15;18i pap2-1 apex. Shoot apexes at an early stage (O) and the later stage ([P] to [Q]) are shown. The apicalmeristem continued to initiate leaf primordia. Multiple shoots derived from axillary meristems of the leaves are observed (Q), and each meristemappears as a vegetative SAM producing leaf primordia. Each shoot is indicated with a star ([P] and [Q]). A close-up view of a meristem initiating leafprimordia in (Q) is shown in (R).(S) and (T) In situ hybridization showing APO1 expression (dark stain) in a developing inflorescence at the secondary branch initiation stage in the wild-type apex (S). APO1 expression is not observed in the apex of the MADS14;15;18i pap2 plant (T).(U) and (V) Developing inflorescences in wild-type plants at the primary branch initiation stage (U) and secondary branch initiation to spikelet initiationstage (V).(W) to (Z) Continuous shoot production in MADS14;15;18i pap2-1 quadruple knockdown plants. Shoots and axillary buds are indicated by circles andarrowheads, respectively. Leaves were removed to show the shoot apices in (U) to (Z).B, bract; FM, floral meristem; LP, leaf primordia; PB, primary branch; PH, panicle hair; S, spikelet. Bars = 200 mm in (K) to (T), 100 mm in (U) to (W), 1 mmin (X) and (Y), and 1 cm in (Z).

1852 The Plant Cell

MADS14;15;18i pap2-1/+ plants (see Supplemental Figure 7online). This implies that, among the four genes, PAP2 functionalone is sufficient for normal inflorescence development. How-ever, because upregulation of PAP2 expression was observed insome of the triple knockdown lines (see Supplemental Figure 5online), we cannot rule out a possibility that the increased PAP2activity compensated for the reduction of AP1/FUL expressionin triple knockdown lines.

We observed a slight delay in flowering in the MADS14;15;18iwild-type and MADS14;15;18i pap2-1/+ plants (Figure 5A). Toquantify the delay, transgenic plants of the T0 generation weregrown under LD conditions in a growth chamber until emergence ofthe eighth leaf and then transferred to SD conditions. We countedthe number of leaves produced after transfer to SD conditions in sixand seven independent transgenic plants of MADS14;15;18i wildtype and MADS14;15;18i pap2-1/+, respectively. The number ofleaves in a single plant from each line and the average leaf numberin each genetic background are shown in Figure 5A. Although theeffects varied among independent transgenic lines, a noticeableincrease in leaf number was observed in plants from transgeniclines containingMADS14;15;18i. These results indicate a possibilitythat the three AP1/FUL-like genes of rice are involved in the reg-ulation of flowering time.

PAP2 and AP1/FUL-Like Genes Function Upstream of Hd3aand RFT1 in Leaves

To reveal the cause of the delay in flowering time, the expressionlevels of the rice flowering genes Hd3a and RFT1 were examined(Figures 5B and 5C). Quadruple knockdown and pap2-1 plantswere grown under natural LD conditions and then transferred toa growth chamber under SD conditions. Expression of Hd3a andRFT1 were induced after the transfer to SD conditions in wild-

type control plants. By contrast, extremely low levels of Hd3ainduction were observed in the quadruple knockdown lines. Aslight decrease in Hd3a induction rate was observed in pap2-1.Expression levels of RFT1 under the SD conditions were alsolowered in pap2-1 and the quadruple lines compared with thatobserved in wild-type plants.

Interaction of MADS Domain Proteins

MADS domain proteins dimerize and may form multimericcomplexes to function (reviewed in Gramzow and Theissen,2010). Therefore, we examined the interactions among PAP2,MADS14, MADS15, and MADS18 using the yeast two-hybridmethod (Figure 6A). The entire coding regions of the cDNAswere fused to the activation domain (pGAD) and binding domain(pGBD) and tested for interactions. As shown in Figure 6A,a strong interaction was observed between MADS14 andMADS15. Weak but significant interactions also occurred be-tween PAP2 and PAP2, PAP2 and MADS14, and PAP2 andMADS15. This indicates that PAP2 is able to form heterodimerswith MADS14 and MADS15, in addition to forming a homodimer.To pursue further the possibility that PAP2 interacts withMADS14 and MADS15, we performed transient coexpressionassays of epitope-tagged PAP2 and MADS14 or MADS15 inNicotiana benthamiana leaves. The presence of MADS14-FLAG,MADS15-FLAG, and PAP2-Myc proteins in total protein sam-ples was confirmed by immunoblotting using anti-Myc anti-bodies (Figures 6B and 6C). After coimmunoprecipitation usinganti-FLAG antibodies, interactions between MADS14 and PAP2(Figure 6B) and MADS15 and PAP2 (Figure 6C) were detectedby immunoblotting analysis with an anti-Myc antibody. Thesebiochemical data indicate that PAP2 has the ability to interactwith MADS14 and MADS15 in vivo.

Figure 5. Delay of Flowering Time in Triple Knockdown Plants.

(A) The numbers of leaves produced by individual plants between transfer to SD conditions and heading. Plants were transferred to SD conditions at theeight-leaf stage, when the eighth leaf was fully expanded. Data are shown for wild-type (wt) and pap2-1/+ and pap2-1 mutant lines transformed witheither the empty vector or the MADS14;15;18i construct. Each bar in the left panel indicates an independent transgenic line. The average number ofleaves in each genotype is shown in the right panel. Bars and asterisks in the right panel represent SD and significant difference from wild-typedetermined by the Student’s t test at P < 0.05, respectively.(B) and (C) Relative expression levels of Hd3a (B) and RFT1 (C) in leaves of pap2-1 and MADS14;15;18i pap2-1 (quad) plants. RNA was isolated fromplants under LD conditions and 3 d after transfer to SD conditions. Bars represent SD for three biological replications.

Inflorescence Meristem Identity of Rice 1853

Genetic Interaction between Hd3a and PAP2

Although it has been suggested that MADS14 and MADS15 actdownstream of Hd3a/RFT1 in the meristem, their functions inthis pathway have not been clearly elucidated (Komiya et al.,2008, 2009; Taoka et al., 2011). Misexpression of Hd3a causesextremely early flowering in rice and a number of other plantspecies (Izawa et al., 2002; Tamaki et al., 2007). We took ad-vantage of this phenomenon to test the hypothesis that PAP2functions downstream of Hd3a in the meristem to establish IMidentity. Reproductive transition was remarkably accelerated byintroduction of the rolC-Hd3a:GFP (for green fluorescent protein)gene (Tamaki et al., 2007) in wild-type plants (Figures 7A and

7B). In severe cases, spikelets developed directly from callus(data not shown). By contrast, when the same rolC-Hd3a:GFPconstruct was introduced into a pap2-1 background, the plantsexhibited a novel phenotype (Figures 7C and 7D). Precociousstem elongation was observed, suggesting transition to repro-ductive development; however, inflorescence development didnot initiate. Instead, vegetative shoots were repeatedly gen-erated resulting in a bushy appearance. Roots initiated fromeach tiller shoot, and tillers were connected with very thin string-like stems (Figure 7C). Inflorescences eventually initiated in therolC-Hd3a:GFP pap2-1 lines. Spikelets in these lines containedincreased numbers of glumes (specialized leaves subtendingflowers), indicating retarded specification of SM identity. Theseresults indicate that PAP2 function is required to initiate theprecocious inflorescence development induced by Hd3a mis-expression, supporting our hypothesis that PAP2 and the threeAP1/FUL-like genes act downstream of Hd3a to promote IMidentity.

DISCUSSION

Regulation of IM Identity in Rice

Our results show that expression of PAP2, a MADS box genebelonging to a grass-specific subclade of the SEP subfamily,and three AP1/FUL-like genes (MADS14, MADS15, andMADS18) is activated in the meristem at the transition to thereproductive phase. The knockdown of the three AP1/FUL-likegenes did not significantly affect inflorescence development. Onthe other hand, the elimination of PAP2 function in the tripleknockdown plants severely impeded transition of the SAM to theIM. Because IM identity is established normally in the pap2-1mutant (Kobayashi et al., 2010), the most likely interpretation ofthe results with the quadruple knockdown plants is that IMidentity in rice is determined by the combined actions of thethree AP1/FUL-like genes with PAP2. The negative correlationbetween total expression levels of three AP1/FUL-like genesand the severity observed in inflorescence development prob-ably indicates the redundancy in the functions of the threegenes. However, currently it is unknown whether all four genescontribute equally in specifying IM identity. The individual con-tributions of each gene to the regulation of IM identity needs tobe elucidated.It has been proposed that MADS14 and MADS15 function

downstream of Hd3a and RFT1 to transduce the florigen signal(Komiya et al., 2008, 2009). More recently, transient assays incultured cells were used to show that the transcription ofMADS15 is indeed induced by a complex containing Hd3a,a 14-3-3 protein, and FD (Taoka et al., 2011). Here, we show thatnot only MADS14 and MADS15 but also MADS18 and PAP2exhibit clear induction in the SAM before any visible change inmeristem morphology. We also confirmed that PAP2, MADS14,and MADS15 interact with one another in vivo. The alleviation ofthe precocious flowering phenotype caused by ectopic over-expression of Hd3a in the pap2-1 background also supports thenotion that PAP2 functions downstream of Hd3a in the meri-stem. On the basis of these results, we conclude that the fourgenes analyzed in this study act in the SAM as downstream

Figure 6. Interaction among Four MADS Box Proteins.

(A) Yeast two-hybrid interaction assays. Full-length coding regions ofPAP2, MADS14, MADS15, and MADS18 were cloned in pGBKT7. PAP2or MADS15 were also cloned in pGADT7. After cotransformation ofpGBKT7 and pGAD17 vectors, yeast was grown on nonselective (-LT),weak selective (-LTH), and strong selective (-LTHA) media. The emptyactivation domain vector (empty) was included as a negative control.(B) and (C) Interaction assay in N. benthamiana leaves using MADS14-FLAG, PAP2-Myc, and MADS15-Myc (B) and MADS15-FLAG, PAP2-Myc,and MADS14-Myc (C). Proteins were coexpressed in N. benthamianaleaves, and the total protein extracts were subjected to immunoprecipi-tation (IP) using an anti-FLAG antibody. The resultant immune complexeswere analyzed on immunoblots with anti-Myc antibodies.

1854 The Plant Cell

regulators of the florigen signal and give rise to the phasechange of the SAM to the IM (Figure 8).

Meanwhile, the extensive decline of the Hd3a and RFT1 ex-pression levels in leaves of the knockdown plants under SDconditions suggests the possibility that the AP1/FUL-like genesof rice and PAP2 function upstream of Hd3a and RFT1 in leaves.An attractive scenario to explain these results is that AP1/FUL-like genes work in distinct ways in leaves and the SAM to pro-mote reproductive transition (Figure 8). First, AP1/FUL-likegenes induce expression of Hd3a and RFT1 in leaves. Once theflorigen is transported from the leaves to the SAM, transcriptionof the AP1/FUL-like genes and PAP2 is induced in the SAM

itself, resulting in the transition of the meristem phase to the IM.This two-step regulation might help to accelerate meristemphase change by amplifying the florigen signal.Currently, it is not clear whether the delay in flowering ob-

served in the triple and quadruple knockdown plants is causedby the loss of AP1/FUL-like activities in leaves, in the meristem,or both. Similarly, we cannot rule out a possibility that the failurein establishing IM identity in the quadruple knockdown plants isa consequence of the reduced activity of Hd3a and RFT1, flo-rigen genes, in leaves. However, we favor an interpretation thatthe meristem defects were caused by the reduced function ofthe three AP1/FUL-like and PAP2 genes in the meristem basedon the clear induction of these genes in the SAM upon thetransition. Results of the genetic analysis between PAP2 andHd3a also support a model in which Hd3a works upstream ofthe four MADS box genes in the control of meristem phasetransition. To obtain a conclusive view, roles of the four MADSbox genes in leaves and the meristem should be experimentallydistinguished by more detailed analysis in the near future.Our findings that AP1/FUL-like genes work upstream of Hd3a

and RFT1 in leaves are consistent with the report that MADS14and MADS15 work downstream of Early heading date1, a posi-tive regulator of rice flowering, in leaves (Doi et al., 2004).However, these results are inconsistent with a previous reportdescribing that the expression of MADS14 and MADS15 inleaves is regulated by RFT1 (Komiya et al., 2008). A possibleexplanation for this discrepancy is that RFT1 and AP1/FUL-likegenes mutually activate each other’s expression. Temperategrasses, including wheat (Triticum aestivum) and barley (Hordeumvulgare), require a period of cold temperature to initiate re-productive transition. In these species, VERNALIZATION1(VRN1), encoding an AP1/FUL-like subfamily MADS domainprotein, is a critical regulator of the vernalization response andflowering (Danyluk et al., 2003; Trevaskis et al., 2003; Yan et al.,2003; Shitsukawa et al., 2007). In wheat leaves, VRN1 functionsupstream of FT (Shimada et al., 2009). It has been proposedthat VRN1 triggers flowering through activation of the daylengthresponse in leaves and acceleration of phase transition in themeristem, although the function of VRN1 in the meristem has notbeen clearly demonstrated (Greenup et al., 2009). This suggeststhat the functions of the AP1/FUL-like genes might be con-served in grass species irrespective of any vernalization re-quirement for flowering.

Contribution of Grass-Specific Genes to the Evolution ofGrass-Specific Forms

Gene duplication is a source of novel genes. Duplicated genesacquire new functions by changes in their coding regions and/ortheir regulatory sequences. The SEP and AP1/FUL subfamiliesof the MADS box genes probably originated and underwentmultiple duplication events before the diversification of the ex-tant angiosperms (Litt and Irish, 2003; Malcomber and Kellogg,2005; Malcomber et al., 2006; Preston and Kellogg 2006; Zahnet al., 2005). The evolution of the SEP and AP1/FUL homologs isconceivably associated with the subsequent dramatic successof angiosperms. Further duplication events occurred in the an-cestors of grasses, resulting in grass-specific SEP and AP1/FUL

Figure 7. Genetic Interaction between PAP2 and Hd3a.

(A) A wild-type plant regenerated from callus with no inflorescences.(B) A wild-type plant transformed with prolC-Hd3a:GFP, of the same ageas that shown in (A), with two inflorescences (yellow arrowheads).(C) A pap2-1 plant transformed with rolC-Hd3a:GFP, of the same age asthose shown in (A) and (B). Stem elongation is observed but no in-florescences have formed. The thin, elongated stems are marked withwhite arrowheads.(D) Frequencies of phenotypes observed in rolC-Hd3a:GFP/wild type(wt) and rolC-Hd3a:GFP/pap2-1 transgenic plants. Twenty and 16 in-dependent lines for rolC-Hd3a:GFP/wild type and rolC-Hd3a:GFP/pap2-1, respectively, were analyzed.(E) An inflorescence on a rolC:Hd3a/wild-type plant. A few normalspikelets are present.(F) An inflorescence on a rolC:Hd3a/pap2-1 plant. Each spikelet containsan increased number of elongated glumes (white arrowheads).

Inflorescence Meristem Identity of Rice 1855

genes. Among the five SEP genes in rice, PAP2 is unique inbeing transcribed in the SAM upon phase transition. This is inaddition to its expression in the SM and glumes, in which ab-normalities are observed in pap2-1 mutants. The importance ofthe SEP proteins to mediate multimerization of MADS proteinshas been demonstrated. In Arabidopsis, SEP genes influencefloral meristem identity and floral organ development throughformation of multimeric complexes with other MADS domainproteins that are expressed in a spatially and temporally regu-lated manner (de Folter et al., 2005; Immink et al., 2009). Inaddition to biochemical analysis, the interaction between SEPand AP1/FUL MADS box genes has also been demonstrated bygenetic analysis in a variety of species (Ditta et al., 2004; Pelazet al., 2001). This indicates that SEP and AP1/FUL proteinspotentially interact when they exist in the same cells. Thus, oncePAP2 acquired specialized machinery allowing it to be tran-scribed in the SAM upon reproductive transition, its interactionwith SAM-expressed AP1/FUL MADS box genes, such asMADS14 and MADS15, is expected. This enables PAP2 to beinvolved in the regulation of IM identity. This is a clear ex-ample of gene duplication leading to diversification in the lo-cation and timing of expression, resulting in the creation ofa novel function.

The Role of PAP2 in Rice Inflorescence Development

The timing of the meristem phase change is crucial for plantdevelopment, and subtle modifications of this timing readilyresult in the emergence of novel morphologies (Prusinkiewiczet al., 2007). During rice inflorescence development, newlyformed meristems acquire BM or SM identity depending uponthe timing and position of their occurrence. We showed pre-viously that PAP2 functions as a positive regulator of SM identity

(Kobayashi et al., 2010). PAP2 also functions as a suppressor ofthe extra elongation of glumes (Gao et al., 2010; Kobayashiet al., 2010). In general, leaf growth is extensively suppressedduring reproductive growth. Assuming that suppression ofglumes is likely to be essential to secure the SM identity in grassinflorescence development, an attractive interpretation of PAP2function is that it ensures transition to the SM in two ways,namely, induction of SM identity and suppression of glumes.The new role of PAP2 as a positive regulator of IM identity re-vealed in this study further expands the importance of PAP2 inthe regulation of rice inflorescence development. In all threeroles, PAP2 positively regulates the progression of the steps ininflorescence development. Generation of the PAP2 gene ingrasses may be a molecular innovation in inflorescence de-velopment. Although the functions of PAP2 orthologs in othergrass species remain to be analyzed, the maize (Zea mays) andwheat orthologs MADS4 and AGLG, respectively, exhibit ex-pression patterns similar to that of PAP2, suggesting that theirfunction may be conserved in grass species (Yan et al., 2003;Danilevskaya et al., 2008).

METHODS

Sample Collection for LM

Rice (Oryza sativa ssp japonica cv Nipponbare) plants grown in a paddyfield in Tsukuba, Japan (;36°N) under field conditions during the 2008cultivation season were used. Shoot apices were collected at differentgrowth stages. The samples were fixed in 75% ethanol and 25% acetateand embedded in paraffin using a microwave processor (Energy BeamSciences) according to Takahashi et al. (2010). The paraffin-embeddedsamples were cut into 10-µm-thick sections using a microtome (LeicaRM2255). LM was performed using the Veritas Laser MicrodissectionSystem LCC1704 (Arcturus).

Figure 8. A Model of the Genetic Interaction between MADS Box Genes Regulating the Specification of IM Identity in Rice.

AP1/FUL-like genes and probably PAP2 induce expression of Hd3a and RFT1 in leaves. Once the florigen is transported from the leaves to the SAM,transcription of the AP1/FUL-like genes and PAP2 is induced in the SAM. This results in the transition of the vegetative meristem phase to the IM.

1856 The Plant Cell

Microarray Analysis

Total RNA was extracted from the collected cells with a Pico-Pure RNAisolation kit (Arcturus) according to themanufacturer’s protocol. The quantityand quality of RNAs were checked using the Agilent 2100 Bioanalyzer(Agilent Technologies). One-color spikemixwas added to the total RNA priorto the labeling reaction, and labeling was performed using a Quick AmpLabeling Kit, One-Color (Agilent Technologies). The cyanine-3 (Cy3)-labeledcRNA was purified using the RNeasy mini kit (Qiagen), and the quantity waschecked using a NanoDrop ND-1000 UV-VIS spectrophotomer (NanoDropTechnologies). The Cy3-labeled cRNA sample was fragmented and hy-bridized with 43 44kmicroarray of rice (Rice Annotation Project-Data Base)gene expressionmicroarray slides (Agilent; G2519F#15241) at 65°C for 17 h.The hybridization and washing were performed according to the manu-facturer’s instructions. The slides were scanned on an Agilent G2505B DNAmicroarray scanner, and background correction of the Cy3 raw signals wasperformed using Agilent’s Feature Extraction software (version 9.5.3.1).

We detected 36,100 independent signals corresponding to 27,201annotated loci and 340 microRNA precursors annotated in the RiceAnnotation Project-Data Base. We selected 19,881 probes that each hada raw signal intensity above 50 in at least one of the 24 microarray data.The numbers of microarrays examined are three for the V1, V2, V/R, R3,and R4 stages, four for the R1 stage, and five for the R2 stage. Theselected raw signal intensities were subjected to quantile normalizationfor interarray comparison and then transformed to a log2 scale using theR program (http://www.r-project.org/) and the limma package (Smyth,2005). For redundant probes with the same locus ID, we calculated theaverage normalized values and thereby obtained the expression profile for15,193 genes. The median expression values across the seven stageswere subtracted for each gene, and the gene-normalized values wereassigned as relative expression values.

Vector Construction

A 520-bp fragment from MADS14 containing significantly high homologywith MADS15 and MADS18 was used to generate the RNAi trigger. Thisfragment was amplified using the primers OM14i-110XbaF and OM14i-630XbaR (see Supplemental Table 1 online) and then cloned into thepZH2Bi KXB SpeBam vector (Kuroda et al., 2010) in a tail-to-tail orien-tation flanking the intron from a rice aspartic protease gene, resulting inthe vector pOsMADS14;15;18i. Similarly, 67-, 77-, and 64-bp fragmentswere used as RNAi triggers to generate single knockdown lines ofMADS14, MADS15, and MADS18, respectively. These fragments wereamplified using the primer sets OM14iXba-F/OM14iXba-R (MADS14),OM15iXba-F/OM15iXba-R (MADS15), and OM18iXba-F/OM18iXba-R(MADS18) (see Supplemental Table 1 online) and then cloned into pZH2BiKXB SpeBam in a tail-to-tail orientation flanking the intron from a riceaspartic protease gene, resulting in the vectors pMADS14i, pMADS15i,and pMADS18i, respectively. To construct pPAP2:GUS, a 2.2-kb PAP2promoter region was amplified using the primers shown in SupplementalTable 1 online. This fragment was cloned into the pENTR/D-TOPO vector(Invitrogen). The resulting vector was inserted pGWB3 by the LR re-combination reaction (Invitrogen) (Nakagawa et al., 2007).

Rice Transformation

Rice transformation was performed as described by Hiei et al. (1994).Wild-type (cv Nipponbare) and pap2-1 (Kobayashi et al., 2010) seeds wereused to induce callus for transformation. Hygromycin-selected T1 gen-eration transgenic plants were grown in a growth chamber.

GUS Staining

Transgenic plants containing the PAP2:GUS gene were grown under LDconditions (14 h light at 28°C, 10 h dark at 24°C) in a chamber for 7 d and

subsequently transferred to SD conditions (12 h light at 28°C, 12 h dark at24°C). Meristem samples were treated with acetone on ice for 15 min.After washing with 100 mM NaPO4 buffer, pH7.0, the samples were in-cubated in GUS staining solution (100 mM NaPO4 buffer, pH 7.0, 10 mMEDTA, 0.5 mM potassium ferricyanide, 0.5 mM potassium ferrocyanide,0.1% Triton X-100, and 0.5 mg/mL 5-bromo-4-chrolo-3-indolyl-b-D-glucuronide) at 37°C in the dark.

Scanning Electron Microscopy

Samples were fixed in 2.5% glutaraldehyde overnight at 4°C and de-hydrated in a series of ethanol solutions, and then the final ethanol so-lution was substituted with 3-methylbutyl acetate. The samples were thendried at critical point, sputter-coated with platinum, and observed undera scanning electron microscope (S-4000; Hitachi) at an acceleratingvoltage of 10 kV.

Real-Time RT-PCR Analysis

Total RNA was extracted using the Plant RNA Isolation Mini Kit (AgilentTechnologies). After DNase I treatment, first-strand cDNA was synthe-sized by SuperScript III reverse transcriptase (Invitrogen). RT-PCR wasperformed using the primer sets shown in Supplemental Table 1 online.The relative abundance of mRNAs were measured with a Light Cycler 480system (Roche Applied Science) and normalized against the expression ofubiquitin. For real-time PCR, three replicates were used for each sample.

In Situ Hybridization Analysis

In situ hybridization of the APO1 and PAP2 genes was performed aspreviously described (Ikeda-Kawakatsu et al., 2009; Kobayashi et al., 2010).

Yeast Two-Hybrid Analysis

The coding regions of PAP2, MADS14, MADS15, and MADS18 wereamplified using the primer sets shown in Supplemental Table 1 online,cloned into pTA2 (TOYOBO) byTAcloning, and then transferred to the yeasttwo-hybrid vector pGBKT7 containing the GAL4 DNA binding domain andthe TRP1 gene. PAP2 and MADS15 were cloned into pGADT7 containingthe GAL4 activation domain and LEU2. Plasmids were cotransformed intothe AH109 strain of Saccharomyces cerevisiae with the Matchmaker two-hybrid system (Clontech). Two-hybrid interactions were assayed on se-lective synthetic definedmedia, lacking Leu andTRP (-LT), Leu, Trp, andHis(-LTH), or Leu, TRP, His, and adenine (-LTHA).

Transient Expression in Nicotiana benthamiana andCoimmunoprecipitation Assays

To express the tag-fused proteins, the coding regions of PAP2,MADS14,andMADS15were amplified using the primer sets shown in SupplementalTable 1 online and cloned into pENTR/D-TOPO (Invitrogen), then clonedinto pGWB11 (35S promoter, C-FLAG) or pGWB17 (35S promoter,C-4xMyc) using LR clonase mix (Invitrogen) (Nakagawa et al., 2007). Theexpression constructs were introduced into Agrobacterium tumefaciensstrain EHA105. For transient expression, Agrobacterium strains carryingeach construct were infiltrated into leaves of 3- to 4-week-old N. ben-thamiana as previously described (Kinoshita et al., 2010). Protein wasextracted from N. benthamiana leaves 3 d after infiltration. Total proteinwas extracted according to Gregis et al. (2009). Both total and immuno-precipitated protein were analyzed by SDS-PAGE and immunoblotting usinganti-FLAG (anti-ECS [DDDDK]) (Bethyl; A190-101A) and anti c-Myc (Bethyl;A190-104A) antibodies. Goat IgG horseradish peroxidase–conjugatedsecondary antibody was used (R and D systems; HAF017).

Inflorescence Meristem Identity of Rice 1857

Accession Numbers

Sequence data from this article can be found in the GenBank/EMBLdatabases under the following accession numbers: Os-MADS34,Q6Q9H6; Os-MADS18, Q0D4T4; Os-MADS20, Q2QQA3; Os-MADS14,P0C5B1; and Os-MADS15, Q6Q912.

Supplemental Data

The following materials are available in the online version of this article.

Supplemental Figure 1. Phylogenetic Analysis of the FUL/AP1,AGL6-Like, and SEP Subfamilies of MADS Box Proteins in Grasses.

Supplemental Figure 2. In Situ Hybridization Analysis of APO1Expression in Meristems at Transition (V/R) Stage.

Supplemental Figure 3. The Overall Profile of Gene Expression duringMeristem Phase Transition.

Supplemental Figure 4. Phenotype of MADS14, MADS15, orMADS18 Single Knockdown Plants.

Supplemental Figure 5. Relative Expression Levels of MADS14,MADS15, MADS18, MADS20, and PAP2 in Transgenic Plants Con-taining an OsMADS14;15;18iRNAi Construct.

Supplemental Figure 6. Correlation between the Reduction ofExpression Levels and the Severity of the Phenotype.

Supplemental Figure 7. Inflorescence Development Is Not Affected inthe Triple Knockdown Plants.

Supplemental Table 1. Primers Used in This Study.

Supplemental Data Set 1. Text File of the Alignment Used for thePhylogenetic Analysis Shown in Supplemental Figure 1.

ACKNOWLEDGMENTS

We thank Ko Shimamoto for prolC-Hd3a:GFP, Yasufumi Daimon forphylogenetic analysis, and Rebecca Harcourt for critical reading of thearticle. This work was supported in part by Grants-in-Aid from the Ministry ofEducation, Culture, Sports, Science and Technology, Japan (22119008,22247004, and 21027012 to J.K. and 23580012 to H.Y.), by the Ministry ofAgriculture, Forestry and Fisheries (Genomics for Agricultural Innovation; IPG-0001 to J.K. and N.Y.), by Genomics for Agricultural Innovation (RTR0002 toY.S. and Y.N.), by Genomics for Agricultural Innovation (GAM-207 to H.Y.),and by the Program for Promotion of Basic Research Activities for InnovationBioscience (PROBRAIN; to J.K.).

AUTHOR CONTRIBUTIONS

K.K., N.Y., Y.S., M.Y., R.Y., M.K., H.Y., Y.N., and J.K. designed andperformed the research. K.K., Y.S., and J.K. analyzed data. J.K. wrotethe article.

Received February 15, 2012; revised April 10, 2012; accepted April 21,2012; published May 8, 2012.

REFERENCES

Abe, M., Kobayashi, Y., Yamamoto, S., Daimon, Y., Yamaguchi, A.,Ikeda, Y., Ichinoki, H., Notaguchi, M., Goto, K., and Araki, T.(2005). FD, a bZIP protein mediating signals from the floral pathwayintegrator FT at the shoot apex. Science 309: 1052–1056.

Arora, R., Agarwal, P., Ray, S., Singh, A.K., Singh, V.P., Tyagi, A.K.,and Kapoor, S. (2007). MADS-box gene family in rice: Genome-wide identification, organization and expression profiling duringreproductive development and stress. BMC Genomics 8: 242.

Cardon, G.H., Höhmann, S., Nettesheim, K., Saedler, H., andHuijser, P. (1997). Functional analysis of the Arabidopsis thalianaSBP-box gene SPL3: A novel gene involved in the floral transition.Plant J. 12: 367–377.

Corbesier, L., Vincent, C., Jang, S., Fornara, F., Fan, Q., Searle, I.,Giakountis, A., Farrona, S., Gissot, L., Turnbull, C., andCoupland, G. (2007). FT protein movement contributes to long-distance signaling in floral induction of Arabidopsis. Science 316:1030–1033.

Danilevskaya, O.N., Meng, X., Selinger, D.A., Deschamps, S.,Hermon, P., Vansant, G., Gupta, R., Ananiev, E.V., andMuszynski, M.G. (2008). Involvement of the MADS-box geneZMM4 in floral induction and inflorescence development in maize.Plant Physiol. 147: 2054–2069.

Danyluk, J., Kane, N.A., Breton, G., Limin, A.E., Fowler, D.B., andSarhan, F. (2003). TaVRT-1, a putative transcription factor associ-ated with vegetative to reproductive transition in cereals. PlantPhysiol. 132: 1849–1860.

de Folter, S., Immink, R.G.H., Kieffer, M., Parenicová, L., Henz,S.R., Weigel, D., Busscher, M., Kooiker, M., Colombo, L., Kater,M.M., Davies, B., and Angenent, G.C. (2005). Comprehensiveinteraction map of the Arabidopsis MADS box transcription factors.Plant Cell 17: 1424–1433.

Ditta, G., Pinyopich, A., Robles, P., Pelaz, S., and Yanofsky, M.F.(2004). The SEP4 gene of Arabidopsis thaliana functions in floralorgan and meristem identity. Curr. Biol. 14: 1935–1940.

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.

Fornara, F., Parenicová, L., Falasca, G., Pelucchi, N., Masiero, S.,Ciannamea, S., Lopez-Dee, Z., Altamura, M.M., Colombo, L.,and Kater, M.M. (2004). Functional characterization of OsMADS18,a member of the AP1/SQUA subfamily of MADS box genes. PlantPhysiol. 135: 2207–2219.

Gao, X., Liang, W., Yin, C., Ji, S., Wang, H., Su, X., Guo, C., Kong,H., Xue, H., and Zhang, D. (2010). The SEPALLATA-like gene Os-MADS34 is required for rice inflorescence and spikelet de-velopment. Plant Physiol. 153: 728–740.

Gramzow, L., and Theissen, G. (2010). A hitchhiker’s guide to theMADS world of plants. Genome Biol. 11: 214.

Greenup, A., Peacock, W.J., Dennis, E.S., and Trevaskis, B. (2009).The molecular biology of seasonal flowering-responses in Arabi-dopsis and the cereals. Ann. Bot. (Lond.) 103: 1165–1172.

Gregis, V., Sessa, A., Dorca-Fornell, C., and Kater, M.M. (2009).The Arabidopsis floral meristem identity genes AP1, AGL24 andSVP directly repress class B and C floral homeotic genes. Plant J.60: 626–637.

Hempel, F.D., Weigel, D., Mandel, M.A., Ditta, G., Zambryski, P.C.,Feldman, L.J., and Yanofsky, M.F. (1997). Floral determinationand expression of floral regulatory genes in Arabidopsis. De-velopment 124: 3845–3853.

Hiei, Y., Ohta, S., Komari, T., and Kumashiro, T. (1994). Efficienttransformation of rice (Oryza sativa L.) mediated by Agrobacteriumand sequence analysis of the boundaries of the T-DNA. Plant J. 6:271–282.

Hoshikawa, K. (1989). The Growing Rice Plant. (Tokyo, Japan: NosanGyoson Bunka Kyokai).

1858 The Plant Cell

Huijser, P., and Schmid, M. (2011). The control of developmentalphase transitions in plants. Development 138: 4117–4129.

Ikeda-Kawakatsu, K., Yasuno, N., Oikawa, T., Iida, S., Nagato, Y.,Maekawa, M., and Kyozuka, J. (2009). Expression level of AB-ERRANT PANICLE ORGANIZATION1 determines rice inflorescenceform through control of cell proliferation in the meristem. PlantPhysiol. 150: 736–747.

Immink, R.G.H., Tonaco, I.A.N., de Folter, S., Shchennikova, A.,van Dijk, A.D.J., Busscher-Lange, J., Borst, J.W., and Angenent,G.C. (2009). SEPALLATA3: The ‘glue’ for MADS box transcriptionfactor complex formation. Genome Biol. 10: R24.

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.

Kinoshita, A., Betsuyaku, S., Osakabe, Y., Mizuno, S., Nagawa, S.,Stahl, Y., Simon, R., Yamaguchi-Shinozaki, K., Fukuda, H., andSawa, S. (2010). RPK2 is an essential receptor-like kinase thattransmits the CLV3 signal in Arabidopsis. Development 137: 3911–3920.

Kobayashi, K., Maekawa, M., Miyao, A., Hirochika, H., andKyozuka, J. (2010). PANICLE PHYTOMER2 (PAP2), encodinga SEPALLATA subfamily MADS-box protein, positively controlsspikelet meristem identity in rice. Plant Cell Physiol. 51: 47–57.

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.

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.

Kuroda, M., Kimizu, M., and Mikami, C. (2010). A simple set ofplasmids for the production of transgenic plants. Biosci. Biotechnol.Biochem. 74: 2348–2351.

Lee, H., Suh, S.S., Park, E., Cho, E., Ahn, J.H., Kim, S.G., Lee, J.S.,Kwon, Y.M., and Lee, I. (2000). The AGAMOUS-LIKE 20 MADSdomain protein integrates floral inductive pathways in Arabidopsis.Genes Dev. 14: 2366–2376.

Litt, A., and Irish, V.F. (2003). Duplication and diversification in theAPETALA1/FRUITFULL floral homeotic gene lineage: Implicationsfor the evolution of floral development. Genetics 165: 821–833.

Liu, C., Thong, Z., and Yu, H. (2009). Coming into bloom: Thespecification of floral meristems. Development 136: 3379–3391.

Malcomber, S.T., and Kellogg, E.A. (2005). SEPALLATA gene di-versification: Brave new whorls. Trends Plant Sci. 10: 427–435.

Malcomber, S.T., Preston, J.C., Reinheimer, R., Kossuth, J., andKellogg, E. (2006). Developmental gene evolution and the origin ofgrass inflorescence diversity. Adv. Bot. Res. 44: 425–481.

Nakagawa, T., Kurose, T., Hino, T., Tanaka, K., Kawamukai, M.,Niwa, Y., Toyooka, K., Matsuoka, K., Jinbo, T., and Kimura, T.(2007). Development of series of gateway binary vectors, pGWBs,for realizing efficient construction of fusion genes for plant trans-formation. Biosci. Biotechnol. Biochem. 71: 2095–2100.

Park, S.J., Jiang, K., Schatz, M.C., and Lippman, Z.B. (2012). Rateof meristem maturation determines inflorescence architecture intomato. Proc. Natl. Acad. Sci. USA 109: 639–644.

Pelaz, S., Gustafson-Brown, C., Kohalmi, S.E., Crosby, W.L., andYanofsky, M.F. (2001). APETALA1 and SEPALLATA3 interact topromote flower development. Plant J. 26: 385–394.

Preston, J.C., and Kellogg, E.A. (2006). Reconstructing the evolu-tionary history of paralogous APETALA1/FRUITFULL-like genes ingrasses (Poaceae). Genetics 174: 421–437.

Prusinkiewicz, P., Erasmus, Y., Lane, B., Harder, L.D., and Coen,E. (2007). Evolution and development of inflorescence archi-tectures. Science 316: 1452–1456.

Samach, A., Onouchi, H., Gold, S.E., Ditta, G.S., Schwarz-Sommer,Z., Yanofsky, M.F., and Coupland, G. (2000). Distinct roles ofCONSTANS target genes in reproductive development of Arabi-dopsis. Science 288: 1613–1616.

Schmid, M., Uhlenhaut, N.H., Godard, F., Demar, M., Bressan, R.,Weigel, D., and Lohmann, J.U. (2003). Dissection of floral in-duction pathways using global expression analysis. Development130: 6001–6012.

Shimada, S., Ogawa, T., Kitagawa, S., Suzuki, T., Ikari, C.,Shitsukawa, N., Abe, T., Kawahigashi, H., Kikuchi, R., Handa, H.,and Murai, K. (2009). A genetic network of flowering-time genes inwheat leaves, in which an APETALA1/FRUITFULL-like gene, VRN1,is upstream of FLOWERING LOCUS T. Plant J. 58: 668–681.

Shitsukawa, N., Ikari, C., Shimada, S., Kitagawa, S., Sakamoto, K.,Saito, H., Ryuto, H., Fukunishi, N., Abe, T., Takumi, S., Nasuda,S., and Murai, K. (2007). The einkorn wheat (Triticum monococcum)mutant, maintained vegetative phase, is caused by a deletion in theVRN1 gene. Genes Genet. Syst. 82: 167–170.

Smyth, G.K. (2005). Limma: Linear models for microarray data. InBioinformatics and Computational Biology Solutions Using R andBioconductor, R. Gentlemen, V. Carey, S. Dudoit, R. Irizarry, and W.Huber, eds (New York: Springer), pp. 397–420.

Takahashi, H., Kamakura, H., Sato, Y., Shiono, K., Abiko, T.,Tsutsumi, N., Nagamura, Y., Nishizawa, N.K., and Nakazono, M.(2010). A method for obtaining high quality RNA from paraffin sec-tions of plant tissues by laser microdissection. J. Plant Res. 123:807–813.

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.

Taoka, K. et al. (2011). 14-3-3 proteins act as intracellular receptorsfor rice Hd3a florigen. Nature 476: 332–335.

Torti, S., Fornara, F., Vincent, C., Andrés, F., Nordström, K., Göbel,U., Knoll, D., Schoof, H., and Coupland, G. (2012). Analysis of theArabidopsis shoot meristem transcriptome during floral transitionidentifies distinct regulatory patterns and a leucine-rich repeatprotein that promotes flowering. Plant Cell 24: 444–462.

Trevaskis, B., Bagnall, D.J., Ellis, M.H., Peacock, W.J., and Dennis,E.S. (2003). MADS box genes control vernalization-induced flow-ering in cereals. Proc. Natl. Acad. Sci. USA 100: 13099–13104.

Wellmer, F., and Riechmann, J.L. (2010). Gene networks controllingthe initiation of flower development. Trends Genet. 26: 519–527.

Wigge, P.A., Kim, M.C., Jaeger, K.E., Busch, W., Schmid, M.,Lohmann, J.U., and Weigel, D. (2005). Integration of spatial andtemporal information during floral induction in Arabidopsis. Science309: 1056–1059.

Yan, L., Loukoianov, A., Tranquilli, G., Helguera, M., Fahima, T.,and Dubcovsky, J. (2003). Positional cloning of the wheat vernal-ization gene VRN1. Proc. Natl. Acad. Sci. USA 100: 6263–6268.

Zahn, L.M., Kong, H., Leebens-Mack, J.H., Kim, S., Soltis, P.S.,Landherr, L.L., Soltis, D.E., Depamphilis, C.W., and Ma, H.(2005). The evolution of the SEPALLATA subfamily of MADS-boxgenes: A preangiosperm origin with multiple duplications through-out angiosperm history. Genetics 169: 2209–2223.

Inflorescence Meristem Identity of Rice 1859

DOI 10.1105/tpc.112.097105; originally published online May 8, 2012; 2012;24;1848-1859Plant Cell

Hitoshi Yoshida, Yoshiaki Nagamura and Junko KyozukaKaoru Kobayashi, Naoko Yasuno, Yutaka Sato, Masahiro Yoda, Ryo Yamazaki, Mayumi Kimizu,

MADS Box GeneSEPALLATA, a PAP2-Like MADS Box Genes and FUL/AP1Inflorescence Meristem Identity in Rice Is Specified by Overlapping Functions of Three

 This information is current as of July 20, 2020

 

Supplemental Data /content/suppl/2012/05/02/tpc.112.097105.DC1.html

References /content/24/5/1848.full.html#ref-list-1

This article cites 48 articles, 26 of which can be accessed free at:

Permissions https://www.copyright.com/ccc/openurl.do?sid=pd_hw1532298X&issn=1532298X&WT.mc_id=pd_hw1532298X

eTOCs http://www.plantcell.org/cgi/alerts/ctmain

Sign up for eTOCs at:

CiteTrack Alerts http://www.plantcell.org/cgi/alerts/ctmain

Sign up for CiteTrack Alerts at:

Subscription Information http://www.aspb.org/publications/subscriptions.cfm

is available at:Plant Physiology and The Plant CellSubscription Information for

ADVANCING THE SCIENCE OF PLANT BIOLOGY © American Society of Plant Biologists