Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Molecular Characterization of FLOWERING LOCUS T-Like Genes of Apple (Malus × domestica Borkh.)

Molecular Characterization of FLOWERING LOCUS T-Like Genes of Apple (Malus × domestica Borkh.) Abstract The two FLOWERING LOCUS T (FT)-like genes of apple (Malus × domestica Borkh.), MdFT1 and MdFT2, have been isolated and characterized. MdFT1 and MdFT2 were mapped, respectively, on distinct linkage groups (LGs) with partial homoeology, LG 12 and LG 4. The expression pattern of MdFT1 and MdFT2 differed in that MdFT1 was expressed mainly in apical buds of fruit-bearing shoots in the adult phase, with little expression in the juvenile tissues, whereas MdFT2 was expressed mainly in reproductive organs, including flower buds and young fruit. On the other hand, both genes had the potential to induce early flowering since transgenic Arabidopsis, which ectopically expressed MdFT1 or MdFT2, flowered earlier than wild-type plants. Furthermore, overexpression of MdFT1 conferred precocious flowering in apple, with altered expression of other endogenous genes, such as MdMADS12. These results suggest that MdFT1 could function to promote flowering by altering the expression of those genes and that, at least, other genes may play an important role as well in the regulation of flowering in apple. The long juvenile period of fruit trees prevents early cropping and efficient breeding. Our findings will be useful information to unveil the molecular mechanism of flowering and to develop methods to shorten the juvenile period in various fruit trees, including apple. Introduction Apple (Malus spp.), which belongs to the Rosaceae subfamily Maloideae, is one of the most important fruit tree crops in the world. However, the long juvenile phase characteristic of woody plants, including apple, makes their breeding cycle slower (Zimmerman 1972, Hackett 1985). In apple, the juvenile phase generally lasts from 4 to 8 years or more. For example, the ‘Fuji’ apple (Malus × domestica Borkh.) from ‘Ralls Janet’ × ‘Delicious’, one of the most popular apple cultivars in the world (O’Rourke et al. 2003), set fruit for the first time 12 years after sowing (Sadamori et al. 1963), whereas the ‘Sansa’ apple from ‘Gala’ × ‘Akane’, top-grafted on fruit-bearing trees for early selection, set fruit after 7 years (Yoshida et al. 1988). In general, selection efficiency is limited until the seedlings set fruit. In the case of fruit breeding, the most important traits for selection, such as texture, flavor, ripening time and shelf life, are related to the fruit itself. Therefore, various practical techniques to accelerate flowering and fruiting of seedlings in the juvenile phase have been considered for years. In apple, grafting the seedling onto fruit-bearing trees (top grafting) or onto dwarfing rootstocks, such as ‘Malling 9 (M. 9)’ (Pearl 1932) and ‘JM1’ (Soejima et al. 1998), results in earlier flowering by 1 or 2 years. After the end of the juvenile phase, flower induction occurs in late June, and flower initiation with morphological changes at the apical meristem starts to form floral primordia in mid July in Morioka, Japan (Kotoda et al. 2000). However, little is known about the physiological and genetic factors involving the transition to flowering in apple. On the other hand, studies on Arabidopsis (Arabidopsis thaliana) and snapdragon (Antirrhinum majus) have led to the identification of many genes involved in flowering. In Arabidopsis, the transition from the vegetative to the reproductive phase is initiated by four independent pathways of signal transduction, i.e. the autonomous, and gibberellin-, vernalization- and light-dependent pathways (for reviews, see Koornneef 1998, Levy and Dean 1998). These signals are transmitted via integrator genes, such as FLOWERING LOCUS T (FT), SUPPRESSOR OF OVEREXPRESSION OF CONSTANS1 (SOC1)/AGAMOUS LIKE20 and FLOWERING LOCUS C (FLC), to the floral meristem identity genes APETALA1 (AP1) and LEAFY (LFY) at the apical meristems (for reviews, see Araki 2001, Jack 2004, Michaels 2009). In addition to the integrator genes, TERMINAL FLOWER1 (TFL1; Bradley et al. 1997, Oshima et al. 1997) is also a key gene that represses flowering and maintains the inflorescence meristem by preventing the expression of AP1 and LFY (Ratcliffe et al. 1998, Ratcliffe et al. 1999). TFL1 was the first member of the TFL1/FT family to be identified as a homolog of an animal phosphatidylethanolamine-binding protein (PEBP), considered to be a Raf-1 kinase inhibitor (Yeung et al. 1999). Thereafter, the Arabidopsis genome was found to contain six members of the TFL1/FT family, including ARABIDOPSIS THALIANA CENTRORADIALIS (ATC), BROTHER OF FT AND TFL1 (BFT), FT, MOTHER OF FT AND TFL1 (MFT) and TWIN SISTER OF FT (TSF) (Kardailsky et al. 1999, Kobayashi et al. 1999, Mimida et al. 2001, Yoo et al. 2004, Yamaguchi et al. 2005). FT was identified as a causative gene for a typical late- flowering mutant ft, the phenotype of which was opposite to that of tfl1 (Kardailsky et al. 1999, Kobayashi et al. 1999). Hanzawa et al. (2005) reported that the antagonistic functions of a floral activator and a repressor encoded by FT and TFL1, respectively, could be converted by a single amino acid substitution. Recently, it was found that the FT protein moves from the leaf to the shoot apex, where FT interacts with the bZIP transcription factor, FD, to activate AP1 in Arabidopsis (Abe et al. 2005, Wigge et al. 2005, Corbesier et al. 2007). In rice (Oryza sativa), it was demonstrated that the protein encoded by Hd3a (Kojima et al. 2002), a rice equivalent of the FT gene, similarly travels from the leaf to the shoot apex (Tamaki et al. 2007), probably in the absence of GF14c (a G-box factor 14-3-3c protein), a direct inhibitor of Hd3a (Purwestri et al. 2009). Furthermore, grafting experiments in cucurbit species (Cucurbita maxima and C. moschata) demonstrated that the FT protein, but not FT mRNA crossed the graft union in the phloem translocation stream (Lin et al. 2007). Experiments in Arabidopsis also revealed that the FT protein is graft transmissible (Notaguchi et al. 2008). The results of these studies suggest that the FT protein acts as a mobile flower-inducing signal, such as a florigen, as noted by Chailahyan (for reviews, see Chailahyan 1968, Giakountis and Coupland 2008, Zeevaart 2008). Based on studies of Arabidopsis and other plant species, much effort has been devoted to unveiling the molecular mechanism of flowering and to manipulate the flowering time in horticultural trees since the last decade (for reviews, see Benlloch et al. 2007, Hanke et al. 2007, Wilkie et al. 2008). Apple genes related to flowering have been investigated by analyzing its expression patterns (Yao et al. 1999, Kotoda et al. 2000, van der Linden et al. 2002, Esumi et al. 2005, Hättasch et al. 2008) and heterologous transgenic plants (Sung et al. 1999, Kotoda et al. 2002, Wada et al. 2002). In this context, we have searched for the genes that maintain the juvenile/vegetative growth or induce the transition to flowering in apple. As a result, transgenic apple expressing the antisense RNA of MdTFL1 (an apple ortholog of TFL1) showed precocious flowering (Kotoda et al. 2003). Therefore, we further investigated the genes of the TFL1/FT family as candidates that can shorten the juvenile/vegetative growth phase and manipulate the flowering time in apple (Kotoda and Wada 2005, Kotoda et al. 2006, Mimida et al. 2009). In this study, we have isolated and characterized the two FT-like genes of apple and examined their function using transgenic Arabidopsis. Furthermore, we investigated transgenic apple overexpressing the FT-like gene to clarify its function in apple. Results Identification of FT-like genes from apple We previously isolated MdFT, an apple homolog of FT, and classified it as an FT-like gene by phylogenetic analysis (Kotoda and Wada 2005). However, the result of DNA blot analysis suggested the existence of another MdFT-like gene in apple (Fig. 1A). Therefore, we screened the cDNA library derived from flower buds of ‘Fuji’ apple to obtain FT-like genes homologous to MdFT. As a result, positive clones converged to two kinds of cDNA. One corresponded to MdFT, which we renamed MdFT1, but the other seemed to be a novel gene homologous to MdFT, which we designated MdFT2. The MdFT1 and MdFT2 genes consisted of four exons of 198, 62, 41 and 224 bp encoding a putative protein of 174 amino acids, which resembled the genomic structure of the FT gene (Fig. 1B). The coding sequence of MdFT1 cDNA exhibited 94.2% identity to that of MdFT2 cDNA at the nucleotide level, although the sequence of each exon showed slightly different identities for MdFT1 and MdFT2 (93.4, 100.0, 95.1 and 95.1% for the first, second, third and fourth exons, respectively). The genomic sequence including the coding region of MdFT2 (1,963 bp) was longer than that of MdFT1 (1,620 bp) due to the extended introns of MdFT2 (Fig. 1B). Fig. 1 View largeDownload slide (A) DNA blot analysis for FT-like genes of apple. The genomic DNA (15 μg) of the ‘Fuji’ apple was digested individually with EcoRI, HindIII or XbaI and then separated on a 0.8% (w/v) agarose gel. DNA bands were transferred to Hybond N+ and hybridized with a digoxigenin (DIG)-labeled MdFT1 cDNA. Hybridization was performed in DIG Easy Hyb (Roche Diagnostics) at 42°C for 16 h followed by two washes in 0.5 × SSC containing 1% (w/v) SDS at 68°C for 20 min. The molecular size markers are shown in kb on the left. (B) Schematic representation of the genomic organization of FT-like genes of the apple and Arabidopsis FT gene. Boxes represent exons and lines represent introns. Numbers represent the lengths (bp) of exons (above the boxes) and introns (below the lines). The scale bar on the map represents approximately 200 bp. H represents a restriction enzyme site of HindIII in MdFT1 and MdFT2 genes. (C) Comparison of the deduced protein sequence of MdFT1 and MdFT2 (accession Nos. AB161112 and AB458504, respectively) with those of the TFL1/FT family from apple, Arabidopsis, tomato, grapevine, Lombardy poplar and pea. The AGI (Arabidopsis Genome Initiative; http://www.arabidopsis.org) code or accession number of each gene, as pertinent, is as follows: FT (AGI code At1g65480), TSF (AGI code At4g20370), SFT (tomato, accession No. AY186735), VvFT (grapevine, accession No. DQ871590), PnFT1 (Lombardy poplar, accession No. AB106111), TFL1 (AGI code At5g03840), ATC (AGI code At2g27550), SP (tomato, accession No. U84140), PsTFL1a (pea, accession No. AY340579), VvTFL1A (VvTFL1) (grapevine, accession No. DQ871591), MdTFL1 (apple, accession No. AB052994) and MdCENa (apple, accession No. AB366641). Amino acids in black and in gray are identical and similar, respectively, in at least seven of the 14 members of the TFL1/FT family. Amino acids common to genes in the FT clade are shown in pink and those in the TFL1 clade are shown in green. The gaps indicated by dashes are attributed to the lack of amino acids. The asterisks indicate the amino acid positions related to antagonistic functions between TFL1 and FT (Hanzawa et al. 2005, Ahn et al. 2006). The first square in red, called segment B, represents the region of a potential ligand-binding pocket in TFL1/FT family proteins, and the second square in red represents a region more conserved in FT than in TFL1 (Ahn et al. 2006). The triangles show the intron positions, which are conserved in the TFL1/FT family proteins. (D) Phylogenetic analysis of TFL1/FT family proteins. The tree was constructed by the Neighbor–Joining (N–J) method for the deduced amino acid sequence of the members of the TFL1/FT family from apple (MdFT1, MdFT2, MdTFL1, MdTFL1a, accession No. AB366643; MdCENa and MdCENb, accession No. AB366642), Arabidopsis (ATC, BFT, FT, MFT, TFL1 and TSF), grapevine [VvFT and VvMFT, accession No. DQ871594; VvTFL1A (VvTFL1) and VvTFL1B, accession No. DQ871592; VvTFL1C, accession No. DQ871593], Lombardy poplar (PnFT1 and PnFT3, accession No. AB110612; PnFTL1, accession No. AB369067; PnFTL4, accession No. AB181241), morning glory (InFT1, accession No. ABW73562), orange (CsTFL, accession No. AY344244), pea (PsTFL1a and PsTFL1c, accession No. AY343326), Satsuma mandarin (CiFT1, accession No. AB027456), snapdragon [CENTRORADIALIS (CEN), accession No. S81193) and tomato (SP, SFT and SP9D, accession No. AY186738). The protein sequence data were obtained from the DNA database (DDBJ/EMBL/GenBank). The N–J unrooted dendrograms were generated from the alignment of deduced amino acids with the Clustal X program, and the phylogenetic tree was displayed using the N–J plot unrooted program (Perrière and Gouy, 1996). Bootstrap values for 1,000 resamplings are shown in each branch. The unit for the scale bar displays branch lengths (0.02 substitutions/site). Fig. 1 View largeDownload slide (A) DNA blot analysis for FT-like genes of apple. The genomic DNA (15 μg) of the ‘Fuji’ apple was digested individually with EcoRI, HindIII or XbaI and then separated on a 0.8% (w/v) agarose gel. DNA bands were transferred to Hybond N+ and hybridized with a digoxigenin (DIG)-labeled MdFT1 cDNA. Hybridization was performed in DIG Easy Hyb (Roche Diagnostics) at 42°C for 16 h followed by two washes in 0.5 × SSC containing 1% (w/v) SDS at 68°C for 20 min. The molecular size markers are shown in kb on the left. (B) Schematic representation of the genomic organization of FT-like genes of the apple and Arabidopsis FT gene. Boxes represent exons and lines represent introns. Numbers represent the lengths (bp) of exons (above the boxes) and introns (below the lines). The scale bar on the map represents approximately 200 bp. H represents a restriction enzyme site of HindIII in MdFT1 and MdFT2 genes. (C) Comparison of the deduced protein sequence of MdFT1 and MdFT2 (accession Nos. AB161112 and AB458504, respectively) with those of the TFL1/FT family from apple, Arabidopsis, tomato, grapevine, Lombardy poplar and pea. The AGI (Arabidopsis Genome Initiative; http://www.arabidopsis.org) code or accession number of each gene, as pertinent, is as follows: FT (AGI code At1g65480), TSF (AGI code At4g20370), SFT (tomato, accession No. AY186735), VvFT (grapevine, accession No. DQ871590), PnFT1 (Lombardy poplar, accession No. AB106111), TFL1 (AGI code At5g03840), ATC (AGI code At2g27550), SP (tomato, accession No. U84140), PsTFL1a (pea, accession No. AY340579), VvTFL1A (VvTFL1) (grapevine, accession No. DQ871591), MdTFL1 (apple, accession No. AB052994) and MdCENa (apple, accession No. AB366641). Amino acids in black and in gray are identical and similar, respectively, in at least seven of the 14 members of the TFL1/FT family. Amino acids common to genes in the FT clade are shown in pink and those in the TFL1 clade are shown in green. The gaps indicated by dashes are attributed to the lack of amino acids. The asterisks indicate the amino acid positions related to antagonistic functions between TFL1 and FT (Hanzawa et al. 2005, Ahn et al. 2006). The first square in red, called segment B, represents the region of a potential ligand-binding pocket in TFL1/FT family proteins, and the second square in red represents a region more conserved in FT than in TFL1 (Ahn et al. 2006). The triangles show the intron positions, which are conserved in the TFL1/FT family proteins. (D) Phylogenetic analysis of TFL1/FT family proteins. The tree was constructed by the Neighbor–Joining (N–J) method for the deduced amino acid sequence of the members of the TFL1/FT family from apple (MdFT1, MdFT2, MdTFL1, MdTFL1a, accession No. AB366643; MdCENa and MdCENb, accession No. AB366642), Arabidopsis (ATC, BFT, FT, MFT, TFL1 and TSF), grapevine [VvFT and VvMFT, accession No. DQ871594; VvTFL1A (VvTFL1) and VvTFL1B, accession No. DQ871592; VvTFL1C, accession No. DQ871593], Lombardy poplar (PnFT1 and PnFT3, accession No. AB110612; PnFTL1, accession No. AB369067; PnFTL4, accession No. AB181241), morning glory (InFT1, accession No. ABW73562), orange (CsTFL, accession No. AY344244), pea (PsTFL1a and PsTFL1c, accession No. AY343326), Satsuma mandarin (CiFT1, accession No. AB027456), snapdragon [CENTRORADIALIS (CEN), accession No. S81193) and tomato (SP, SFT and SP9D, accession No. AY186738). The protein sequence data were obtained from the DNA database (DDBJ/EMBL/GenBank). The N–J unrooted dendrograms were generated from the alignment of deduced amino acids with the Clustal X program, and the phylogenetic tree was displayed using the N–J plot unrooted program (Perrière and Gouy, 1996). Bootstrap values for 1,000 resamplings are shown in each branch. The unit for the scale bar displays branch lengths (0.02 substitutions/site). Sequence analysis using the deduced amino acid sequences of TFL1/FT homologs revealed that MdFT1 and MdFT2 were the most similar to each other (94.3% identity), and they also had a high identity with FT-like members, such as VvFT, PnFT1, PnFT3, SFT, CiFT1, TSF and FT (86.7, 85.5, 85.1, 84.5, 83.6, 76.3 and 74.6%, respectively) at the amino acid level (for accession numbers, see the legend of Fig. 1C). The alignment of the deduced amino acid sequences of the TFL1/FT family showed that there were 19 amino acid residues specific to FT-like members and 15 residues specific to TFL1/CEN-like members (Fig. 1C). In particular, both MdFT1 and MdFT2 had Tyr84 and Gln139 in the positions corresponding to Tyr85 and Gln140 of Arabidopsis FT (Fig. 1C, see asterisks). Tyr85/His88 and Gln140/Asp144 are likely to be the most critical residues for distinguishing FT and TFL1 in Arabidopsis (Hanzawa et al. 2005, Ahn et al. 2006). In order to clarify the relationships among the TFL1/FT family members, we constructed a phylogenetic tree by Neighbor–Joining (N–J) distance analysis of apple, Arabidopsis, grapevine (Vitis vinifera), morning glory (Ipomoea nil), orange (Citrus sinensis), pea (Pisum sativum), Lombardy poplar (Populus nigra), Satsuma mandarin (Citrus unshiu), snapdragon and tomato (Solanum esculentum) as shown in Fig. 1D. The phylogenetic tree was divided into four major clades, represented by BFT, FT, MFT and TFL1. As expected, MdFT2, together with MdFT1, was classified into the FT clade. FT-like members of apple were more closely related to those of deciduous woody plants, such as grapevine and poplar. Localization of MdFT1 and MdFT2 on a linkage map The Rosaceae subfamily Maloideae, which includes the genus Malus, has a higher haploid base chromosome number, x = 17, than other members of the family Rosaceae, probably due to the hypothesized polyploid origin (Sax 1933, Chevreau et al. 1985). Since the DNA blot and sequence analysis suggested that MdFT1 and MdFT2 were not allelic (Fig. 1A), we attempted to show clearly on which linkage group MdFT1 and MdFT2 were located and to consider the evolutionary origin of those genes. Therefore, genetic mapping was performed for the genes using two kinds of F1 mapping populations from crosses between ‘Delicious’ (De) and Mitsubakaido (Malus sieboldii Rehder) and between ‘Ralls Janet’ (Ra) and Mitsubakaido (Igarashi et al. 2008). As a result, MdFT1 was assigned to the south end of linkage group 12 (LG 12) of ‘Delicious’ (De12), with 70.7 cM from the markers rD05A20–0.9 and E35M48–350 at the north end, whereas MdFT2 was assigned to the locus close to the marker E38M61–395 at the south end of LG 4 of ‘Ralls Janet’ (Ra04) (Fig. 2). Fig. 2 View large Download slide Location of the FT-like genes on the linkage map of the apple cultivars ‘Ralls Janet’ (Ra) and ‘Delicious’ (De). MdFT1 and MdFT2 were located on LG 12 (De12) and LG 4 (Ra04), respectively. Linkage groups (LGs) of ‘Ralls Janet’ (Ra04) and ‘Delicious’ (De12) are derived from segregation data of ‘Ralls Janet’ and ‘Delicious’ populations, respectively. LG numbers are according to Maliepaard et al. (1998). Loci are listed on the right side of the linkage maps. The left side shows genetic distances (cM). The linkage phase information is provided as + or −, indicating on which of the homologous chromosomes the marker/allele is located. In the case of SSR markers, the sizes of SSR bands are shown by three-figure numbers or the relative sizes of the SSR bands are shown by the letter ‘L (long)’ or ‘S (short)’. CAPS markers are shown with the gene name, the restriction enzyme (Dr, DraI; Hf, HinfI) and the length of the scored fragment (kbp). For a detailed explanation of molecular markers, see Supplementary Table S2 of this paper and the materials and methods of Igarashi et al. (2008). Fig. 2 View large Download slide Location of the FT-like genes on the linkage map of the apple cultivars ‘Ralls Janet’ (Ra) and ‘Delicious’ (De). MdFT1 and MdFT2 were located on LG 12 (De12) and LG 4 (Ra04), respectively. Linkage groups (LGs) of ‘Ralls Janet’ (Ra04) and ‘Delicious’ (De12) are derived from segregation data of ‘Ralls Janet’ and ‘Delicious’ populations, respectively. LG numbers are according to Maliepaard et al. (1998). Loci are listed on the right side of the linkage maps. The left side shows genetic distances (cM). The linkage phase information is provided as + or −, indicating on which of the homologous chromosomes the marker/allele is located. In the case of SSR markers, the sizes of SSR bands are shown by three-figure numbers or the relative sizes of the SSR bands are shown by the letter ‘L (long)’ or ‘S (short)’. CAPS markers are shown with the gene name, the restriction enzyme (Dr, DraI; Hf, HinfI) and the length of the scored fragment (kbp). For a detailed explanation of molecular markers, see Supplementary Table S2 of this paper and the materials and methods of Igarashi et al. (2008). Expression patterns of MdFT1 and MdFT2 in apple To identify when and where MdFT1 and MdFT2 were expressed, quantitative real-time reverse transcription–PCR (qRT–PCR) was performed on various tissues of apple in the adult phase, juvenile phase or tissue culture using primers specific to each gene (Supplementary Table S1). The transcript of MdFT1 accumulated mainly in apical buds of fruit-bearing shoots (FBS) (AB1, on June 24; AB2, on August 7), flower buds (at the balloon stage), floral organs, such as stamens, and whole young fruits. The level of MdFT1 transcript was very low in the tissues of 1-month-old seedlings in the juvenile phase, such as roots, stems, mature leaves and apical buds of vegetative shoots, with little detection in seeds and cultured shoots, which included apical buds, stems and leaves. The level of MdFT1 transcript in the mature leaves was relatively low in both the juvenile and the adult phase as compared with that detected in the apical buds and reproductive organs in the adult phase, although the expression of MdFT1 in mature leaves was higher in the adult phase than in the juvenile phase. The MdFT1 transcript level in mature fruit and peel was below the detection limit (Fig. 3A). On the other hand, the transcript of MdFT2 was detected mainly in reproductive organs, such as flower buds, sepals, petals, stamens, carpels, receptacles, peduncles and whole young fruits, with some expression also detected in mature fruit (Fig. 3B). In apical buds of FBS, we also detected the expression of MdFT2, but the expression level of MdFT2 was relatively lower than that of MdFT1. In the tissues of 1-month-old seedlings, MdFT2 was expressed at relatively higher levels than MdFT1; however, the transcript of MdFT2 in mature leaves was low in both the juvenile and the adult phases, similarly to MdFT1. We could not detect the expression of MdFT2 in seeds and peel (Fig. 3B). The transcripts of MdFT2 accumulated in shoots of cultured tissues more than those of MdFT1, although the transcription level of both was much lower than that in the apical buds of FBS. Fig. 3 View large Download slide Expression patterns of MdFT1 (A) and MdFT2 (B) in various tissues, and seasonal expression patterns of MdFT1 (C), MdFT2 (D), MdTFL1 (E) and MdAP1 (F) in apple by quantitative real-time RT–PCR. The samples for (A) and (B) from left to right are as follows: seeds (Sd), roots (Rt), stems (Sm), mature leaves (MLJ) and apical buds (ABJ) of 1-month-old seedlings (from a cross of cv. ‘Fuji’ × cv. ‘Orin’) in the juvenile phase; mature leaves (ML, collected on June 24) in the adult phase; apical buds of FBS on June 24 (AB1) and on August 7 (AB2) in the adult vegetative/reproductive phase; flower buds at the balloon stage (FB), sepals (Se), petals (Pe), stamens (St), carpels (Ca), receptacles (Re), peduncles (Pd), young fruit on June 15 (YF), mature fruit on November 18 (MF) and peel (Pl) from ‘Fuji’ apple in the adult reproductive phase; and shoots of tissue culture (CS). The samples for (C) to (F) from left to right are as follows: apical buds of fruit-bearing shoots (FBS) on June 6, June 27, July 24, August 22, September 19, October 24, November 7, December 19, January 23, Feburuary 20, March 19 and April 7; or succulent shoots (SS) on June 6, June 27, July 24, August 22, September 19 and October 24. Levels of detected amplicons were normalized by reference to amplified products that corresponded to apple HistoneH3. Values are means ± SD of results from three replicates. Each symbol without a bar indicates that the standard deviation fell within the symbol. The primer sets used in quantitative real-time RT–PCR analysis and the PCR conditions are described in Supplementary Table S1. Fig. 3 View large Download slide Expression patterns of MdFT1 (A) and MdFT2 (B) in various tissues, and seasonal expression patterns of MdFT1 (C), MdFT2 (D), MdTFL1 (E) and MdAP1 (F) in apple by quantitative real-time RT–PCR. The samples for (A) and (B) from left to right are as follows: seeds (Sd), roots (Rt), stems (Sm), mature leaves (MLJ) and apical buds (ABJ) of 1-month-old seedlings (from a cross of cv. ‘Fuji’ × cv. ‘Orin’) in the juvenile phase; mature leaves (ML, collected on June 24) in the adult phase; apical buds of FBS on June 24 (AB1) and on August 7 (AB2) in the adult vegetative/reproductive phase; flower buds at the balloon stage (FB), sepals (Se), petals (Pe), stamens (St), carpels (Ca), receptacles (Re), peduncles (Pd), young fruit on June 15 (YF), mature fruit on November 18 (MF) and peel (Pl) from ‘Fuji’ apple in the adult reproductive phase; and shoots of tissue culture (CS). The samples for (C) to (F) from left to right are as follows: apical buds of fruit-bearing shoots (FBS) on June 6, June 27, July 24, August 22, September 19, October 24, November 7, December 19, January 23, Feburuary 20, March 19 and April 7; or succulent shoots (SS) on June 6, June 27, July 24, August 22, September 19 and October 24. Levels of detected amplicons were normalized by reference to amplified products that corresponded to apple HistoneH3. Values are means ± SD of results from three replicates. Each symbol without a bar indicates that the standard deviation fell within the symbol. The primer sets used in quantitative real-time RT–PCR analysis and the PCR conditions are described in Supplementary Table S1. Seasonal expression pattern in apical buds of apple in the adult phase showed that MdFT1 was expressed to higher levels than MdFT2 in FBS from late June to August. In addition, MdFT1 was expressed relatively more highly in FBS than in succulent shoots (SS), which are vegetative shoots that do not produce flower buds, while the expression level of MdFT2 was similar between FBS and SS, from early June to mid August (Fig. 3C, D). The expression of MdFT2 peaked in mid September and thereafter decreased during the dormant period with a rapid increase from late March (Fig. 3D). On the other hand, the transcript of MdTFL1 decreased rapidly toward mid July before the peak of MdFT1 expression in apical buds of FBS, with the signal in those of SS showing a delayed decrease at late September (Fig. 3E). For MdAP1, it started to increase from late August after the period of flower induction, consistent with the onset of the development of floral organ primordia (Fig. 3F). Ectopic expressions of the apple FT-like genes in Arabidopsis To determine the effects of apple FT-like genes on flowering time and inflorescence morphology, we generated two kinds of transgenic Arabidopsis plants with MdFT1 or MdFT2 under the control of the cauliflower mosaic virus (CaMV) 35S promoter fused with the Ω sequence (35SΩ) (Fig. 4A). Consequently, we obtained >20 independent transgenic lines for 35SΩ : MdFT1 and 35SΩ:MdFT2 (designated 35SΩ : MdFT1/wt and 35SΩ : MdFT2/wt, respectively). Among them, 10 independent lines per construct were used for further analysis. Eight to 10 plants in the T2 generation per line were grown under long-day (LD, 16 h light/8 h dark) conditions, and their phenotypes were examined. Fig. 4 View largeDownload slide Phenotypes of transgenic Arabidopsis with apple FT-like genes under long-day conditions. (A) Schematic representation of the transformation vectors, 35SΩ:MdFT1 and 35SΩ:MdFT2. LB, left border; nptII, neomycin phosphotransferase II gene; P35S, promoter region of cauliflower mosaic virus 35S; Pnos, nos promoter; RB, right border; Tnos, 3′ region of nos. (B) Appearance of wild-type Arabidopsis (ecotype Columbia, left), transgenic Arabidopsis with 35SΩ:MdFT1 (middle) and transgenic Arabidopsis with 35SΩ:MdFT2 (right) 26 d after transfer to the growth chamber. (C) Inflorescence and flower of wild-type Arabidopsis (left), transgenic Arabidopsis with 35SΩ:MdFT1 (middle) and transgenic Arabidopsis with 35SΩ:MdFT2 (right). The photographs were taken after flowering for each plant. For details on the flowering time of transgenic lines, see Table 1. The wild type (Col) was used for Arabidopsis transformation. The plants in (B) and (C) were grown under LD conditions (16 h light/8 h dark) at 22°C. Fig. 4 View largeDownload slide Phenotypes of transgenic Arabidopsis with apple FT-like genes under long-day conditions. (A) Schematic representation of the transformation vectors, 35SΩ:MdFT1 and 35SΩ:MdFT2. LB, left border; nptII, neomycin phosphotransferase II gene; P35S, promoter region of cauliflower mosaic virus 35S; Pnos, nos promoter; RB, right border; Tnos, 3′ region of nos. (B) Appearance of wild-type Arabidopsis (ecotype Columbia, left), transgenic Arabidopsis with 35SΩ:MdFT1 (middle) and transgenic Arabidopsis with 35SΩ:MdFT2 (right) 26 d after transfer to the growth chamber. (C) Inflorescence and flower of wild-type Arabidopsis (left), transgenic Arabidopsis with 35SΩ:MdFT1 (middle) and transgenic Arabidopsis with 35SΩ:MdFT2 (right). The photographs were taken after flowering for each plant. For details on the flowering time of transgenic lines, see Table 1. The wild type (Col) was used for Arabidopsis transformation. The plants in (B) and (C) were grown under LD conditions (16 h light/8 h dark) at 22°C. Seven of 10 transgenic Arabidopsis lines with 35SΩ : MdFT1 flowered significantly earlier than wild-type plants under LD conditions at a P-value of <0.01 for the number of rosette leaves (Fig. 4B, Table 1). For example, 35SΩ : MdFT1/wt (#7) flowered with 3.9 ± 0.2 rosette and 2.1 ± 0.1 cauline leaves, while wild-type plants flowered with 6.7 ± 0.3 rosette and 3.6 ± 0.2 cauline leaves (Table 1). Nine out of 10 transgenic lines with 35SΩ : MdFT2 also flowered significantly earlier than wild-type plants under LD conditions at a P-value of <0.01 for the number of rosette leaves (Fig. 4B, Table 1). For example, 35SΩ : MdFT2/wt (#3) exhibiting a strong phenotype flowered with 4.0 ± 0.2 rosette and 1.9 ± 0.2 cauline leaves, whereas wild-type plants flowered with 6.7 ± 0.3 rosette and 3.6 ± 0.2 cauline leaves (Table 1). No difference in the appearance of flowers and inflorescences was observed among 35SΩ : MdFT1/wt, 35SΩ : MdFT2/wt and wild-type plants (Fig. 4C). An early flowering phenotype was also observed in 35SΩ : MdFT1/wt and 35SΩ : MdFT2/wt under short-day conditions (data not shown). Table 1 Flowering time of transgenic lines ectopically expressing MdFT1 or MdFT2 Line  LD conditions (16 h light/8 h dark)   No. of plants  Rosette leaves  Cauline leaves  Total leaves  Wt (Col)  10  6.7 ± 0.3  3.6 ± 0.2  10.3 ± 0.4  Vector/Wt (Col)  8  7.3 ± 0.3  3.3 ± 0.2  10.5 ± 0.2  35SΩ:MdFT1/wt          #1  9  4.0 ± 0.3**  2.6 ± 0.2**  6.6 ± 0.4**  #2  10  3.9 ± 0.2**  2.6 ± 0.2**  6.5 ± 0.2**  #3  10  6.2 ± 0.2  3.1 ± 0.1*  9.3 ± 0.2*  #4  10  4.0 ± 0.2**  2.2 ± 0.1**  6.2 ± 0.2**  #5  10  4.3 ± 0.2**  2.1 ± 0.1**  6.4 ± 0.2**  #6  10  6.7 ± 0.3  2.6 ± 0.2**  9.3 ± 0.4  #7  10  3.9 ± 0.2**  2.1 ± 0.1**  6.0 ± 0.2**  #8  9  7.9 ± 0.5  2.3 ± 0.2**  10.2 ± 0.4  #9  10  4.0 ± 0.2**  2.8 ± 0.2**  6.8 ± 0.3**  #10  10  4.3 ± 0.3**  2.3 ± 0.1**  6.6 ± 0.4**  35SΩ:MdFT2/wt          #1  9  4.9 ± 0.3**  2.1 ± 0.1**  7.0 ± 0.3**  #2  10  5.0 ± 0.3**  3.0 ± 0.2*  8.0 ± 0.2**  #3  10  4.0 ± 0.2**  1.9 ± 0.2**  5.9 ± 0.3**  #4  10  4.5 ± 0.3**  2.3 ± 0.1**  6.8 ± 0.3**  #5  10  4.3 ± 0.2**  2.2 ± 0.1**  6.5 ± 0.2**  #6  10  4.1 ± 0.2**  3.3 ± 0.2  7.4 ± 0.4**  #7  9  5.3 ± 0.3*  2.3 ± 0.2**  7.7 ± 0.4**  #8  8  4.4 ± 0.4**  2.8 ± 0.3*  7.1 ± 0.3**  #9  10  4.9 ± 0.1**  2.1 ± 0.1**  7.0 ± 0.0**  #10  10  4.3 ± 0.1**  2.0 ± 0.0**  6.3 ± 0.1**  Line  LD conditions (16 h light/8 h dark)   No. of plants  Rosette leaves  Cauline leaves  Total leaves  Wt (Col)  10  6.7 ± 0.3  3.6 ± 0.2  10.3 ± 0.4  Vector/Wt (Col)  8  7.3 ± 0.3  3.3 ± 0.2  10.5 ± 0.2  35SΩ:MdFT1/wt          #1  9  4.0 ± 0.3**  2.6 ± 0.2**  6.6 ± 0.4**  #2  10  3.9 ± 0.2**  2.6 ± 0.2**  6.5 ± 0.2**  #3  10  6.2 ± 0.2  3.1 ± 0.1*  9.3 ± 0.2*  #4  10  4.0 ± 0.2**  2.2 ± 0.1**  6.2 ± 0.2**  #5  10  4.3 ± 0.2**  2.1 ± 0.1**  6.4 ± 0.2**  #6  10  6.7 ± 0.3  2.6 ± 0.2**  9.3 ± 0.4  #7  10  3.9 ± 0.2**  2.1 ± 0.1**  6.0 ± 0.2**  #8  9  7.9 ± 0.5  2.3 ± 0.2**  10.2 ± 0.4  #9  10  4.0 ± 0.2**  2.8 ± 0.2**  6.8 ± 0.3**  #10  10  4.3 ± 0.3**  2.3 ± 0.1**  6.6 ± 0.4**  35SΩ:MdFT2/wt          #1  9  4.9 ± 0.3**  2.1 ± 0.1**  7.0 ± 0.3**  #2  10  5.0 ± 0.3**  3.0 ± 0.2*  8.0 ± 0.2**  #3  10  4.0 ± 0.2**  1.9 ± 0.2**  5.9 ± 0.3**  #4  10  4.5 ± 0.3**  2.3 ± 0.1**  6.8 ± 0.3**  #5  10  4.3 ± 0.2**  2.2 ± 0.1**  6.5 ± 0.2**  #6  10  4.1 ± 0.2**  3.3 ± 0.2  7.4 ± 0.4**  #7  9  5.3 ± 0.3*  2.3 ± 0.2**  7.7 ± 0.4**  #8  8  4.4 ± 0.4**  2.8 ± 0.3*  7.1 ± 0.3**  #9  10  4.9 ± 0.1**  2.1 ± 0.1**  7.0 ± 0.0**  #10  10  4.3 ± 0.1**  2.0 ± 0.0**  6.3 ± 0.1**  Plants in the second generation (T2) were grown under long-day (LD) conditions. Numbers of rosette and cauline leaves were counted on the day floral organs became visible. Values are means ± SE. Student’s t-test was performed to compare the effects of treatments. One or two asterisks indicate a statistically significant difference from the wild-type plants in the same column (P <0.05, 0.01, respectively). View Large Overexpression of MdFT1 in apple Transgenic experiments using Arabidopsis revealed that both MdFT1 and MdFT2 had the potential to function as floral promoters in apple. Expression analysis by qRT–PCR suggested that MdFT1 played a role during floral transition, at least in tissues such as apical buds. Therefore, to clarify whether MdFT1 confers an early-flowering phenotype in apple, we produced transgenic apples carrying the 35SΩ : MdFT1 already used for Arabidopsis transformation, using micropropagated tissues of an apple rootstock cv. ‘JM2’. As a result, we obtained six independent transgenic lines designated MdFT1#1-1, MdFT1#1-2, MdFT1#3, MdFT1#5, MdFT1#6, and MdFT1#8. Five of six transgenic lines (MdFT1#1-1, MdFT1#1-2, MdFT1#3, MdFT1#6 and MdFT1#8) produced flower buds or opened flowers in vitro 8–12 months after Agrobacterium infection (2–6 months after regeneration). For example, MdFT1#3 produced terminal flower buds at the top of each shoot, and some of them came into flower (Fig. 5A, B), whereas the control plant never produced flower buds under the same culture conditions (Fig. 5C). In addition, the shoots of MdFT1#1-1 were grafted onto the apple rootstock ‘JM7’ and transferred to the greenhouse (Fig. 5D). The potted MdFT1#1-1 also produced a flower bud and opened a solitary flower; however, the flower had >8 petals (Fig. 5E), in contrast to the normal flower, which had five sepals, petals, carpels and approximately 20 stamens (Fig. 5F). Importantly, the in vitro flowering lines, such as MdFT1#3 and MdFT1#8, which produced in vitro flower buds more frequently than the others, showed a much less vigorous growth of shoots. Several lines with such severe phenotypes senesced, probably due to their inability to form new vegetative shoot meristems. The result of DNA blot analysis for transgenic lines MdFT1#1-1, MdFT1#5 and MdFT1#8 indicated that MdFT#1-1 had two copies and the other lines had one copy of the transgene (Fig. 5G). Fig. 5 View largeDownload slide Precocious flowering of transgenic apples overexpressing MdFT1. (A) Floral buds (indicated as an open triangle) formed at the top of the shoots of transgenic apple ‘JM2’ (line MdFT1#3) in the culture box. (B) In vitro flowering of a solitary flower (indicated as a filled triangle) at the top of the transgenic shoots in the culture box. (C) Cultured shoots of an apple ‘JM2’ control plant. A white bar indicates 1 cm. (D) Opened flower of transgenic apple ‘JM2’ (line MdFT1#1-1) in the greenhouse. A white bar indicates 3 cm. (E) Close-up view of (D). The number of petals increased compared with a normal flower with five petals. (F) Flower of an apple cv. ‘JM2’ control plant grown in the field. (G) DNA blot analysis of transgenic apples ‘JM2’ with 35SΩ:MdFT1 (MdFT1#1-1, MdFT1#5 and MdFT1#8). Genomic DNA (15 μg) was isolated from the cultured tissues of each transgenic line and a control plant, digested with HindIII, and then separated on a 0.8% (w/v) agarose gel. A blotted membrane was hybridized with a DIG-labeled nptII gene. Fig. 5 View largeDownload slide Precocious flowering of transgenic apples overexpressing MdFT1. (A) Floral buds (indicated as an open triangle) formed at the top of the shoots of transgenic apple ‘JM2’ (line MdFT1#3) in the culture box. (B) In vitro flowering of a solitary flower (indicated as a filled triangle) at the top of the transgenic shoots in the culture box. (C) Cultured shoots of an apple ‘JM2’ control plant. A white bar indicates 1 cm. (D) Opened flower of transgenic apple ‘JM2’ (line MdFT1#1-1) in the greenhouse. A white bar indicates 3 cm. (E) Close-up view of (D). The number of petals increased compared with a normal flower with five petals. (F) Flower of an apple cv. ‘JM2’ control plant grown in the field. (G) DNA blot analysis of transgenic apples ‘JM2’ with 35SΩ:MdFT1 (MdFT1#1-1, MdFT1#5 and MdFT1#8). Genomic DNA (15 μg) was isolated from the cultured tissues of each transgenic line and a control plant, digested with HindIII, and then separated on a 0.8% (w/v) agarose gel. A blotted membrane was hybridized with a DIG-labeled nptII gene. Influence of overexpression of MdFT1 on the other apple genes We examined the influence of the overexpression of MdFT1 on the expression of other endogenous genes, which had been shown or expected to play a role in flowering in apple. To avoid the fluctuation of the gene expression by artificial and other environmental factors, we used the cultured shoots of each line grown in a growth chamber under the same conditions. The result of qRT–PCR analysis showed that MdMADS12 [a FRUITFULL (FUL)-like gene of apple; van der Linden et al. 2002] was significantly upregulated in the in vitro flowering lines MdFT1#1-1 and MdFT1#8, where the MdFT1 transcripts accumulated in large numbers, whereas the expression level of MdMADS12 in a line not flowering in vitro, MdFT1#5, was almost the same as that in the control (Fig. 6A, B). In addition, MdSOC1a and MdSOC1b (putative apple orthologs of SOC1; accession Nos. AB501124 and AB501125, respectively), and MdAP1 (equivalent to MdMADS5, an apple ortholog of AP1) were more up-regulated in the lines MdFT1#1-1 and MdFT1#8, than in line MdFT1#5 and the non-transgenic control (Fig. 6C–E). The expression of MdLHP1a/MdLHP1b (MdLHP1a and MdLHP1b, apple homologs of LHP1/TFL2; Mimida et al. 2007) was not affected by the overexpression of MdFT1 (Fig. 6F). Fig. 6 View large Download slide Expression analysis for apple genes in the shoots of transgenic lines. (A–F) Expression of MdFT1 (A), MdMADS12 (B), MdSOC1a (C), MdSOC1b (D), MdAP1 (E) and MdLHP1a/MdLHP1b (F) in the transgenic shoots of in vitro flowering lines (MdFT1#1-1 and MdFT1#8) and a line not flowering in vitro (MdFT1#5), and in non-transgenic shoots as a control. Shoots were collected from each line cultured in a proliferating medium under LD conditions for 4 weeks to synchronize the growth state of each line. MdMADS12, MdSOC1a, MdSOC1b and MdAP1 are genes encoding a MADS-box transcription factor from apple; and MdLHP1a/MdLHP1b (MdLHP1a and MdLHP1b) are apple twin homologs of LHP1. Primers for MdLHP1a/MdLHP1b were designed to detect both MdLHP1a and MdLHP1b. HistoneH3 was used as an internal control of gene expression. In A–F, non-transgenic apple cv. ‘JM2’ was used as a control plant (control). Transcript levels were normalized against apple HistoneH3. Values are means ± SD of results from three replicates, and each box without a bar indicates that the standard deviation fell within the box. The primer sets used in quantitative real-time RT–PCR analysis and the PCR conditions are described in Supplementary Table S1. Fig. 6 View large Download slide Expression analysis for apple genes in the shoots of transgenic lines. (A–F) Expression of MdFT1 (A), MdMADS12 (B), MdSOC1a (C), MdSOC1b (D), MdAP1 (E) and MdLHP1a/MdLHP1b (F) in the transgenic shoots of in vitro flowering lines (MdFT1#1-1 and MdFT1#8) and a line not flowering in vitro (MdFT1#5), and in non-transgenic shoots as a control. Shoots were collected from each line cultured in a proliferating medium under LD conditions for 4 weeks to synchronize the growth state of each line. MdMADS12, MdSOC1a, MdSOC1b and MdAP1 are genes encoding a MADS-box transcription factor from apple; and MdLHP1a/MdLHP1b (MdLHP1a and MdLHP1b) are apple twin homologs of LHP1. Primers for MdLHP1a/MdLHP1b were designed to detect both MdLHP1a and MdLHP1b. HistoneH3 was used as an internal control of gene expression. In A–F, non-transgenic apple cv. ‘JM2’ was used as a control plant (control). Transcript levels were normalized against apple HistoneH3. Values are means ± SD of results from three replicates, and each box without a bar indicates that the standard deviation fell within the box. The primer sets used in quantitative real-time RT–PCR analysis and the PCR conditions are described in Supplementary Table S1. Discussion MdFT1 and MdFT2 are located in distinct linkage groups with duplicated segments MdFT1 and MdFT2 were mapped, respectively, on loci in distinct linkage groups, LG 12 and LG 4 (Fig. 3). Because there is a homoeology between the distal parts of LG 12 and LG 4 (Gardiner et al. 2007), MdFT1 and MdFT2 might have been derived from an ancient gene that ancestors of the original hybrid would have had. Coincidentally, MdFT1 was located on the distal part of LG 12, the same linkage group as MdTFL1. This finding suggests that MdFT1 and MdTFL1 have evolved in close relation to each other and that at least one chromosome, corresponding to LG 12, plays an important role in the regulation of flowering in apple. To date, it has been revealed that apple has at least six members of the TFL1/FT family in its genome (Kotoda et al. 2005, Mimida et al. 2009, this study), although members that belong to the BFT or MFT clade have not been reported. Other eudicots, such as Arabidopsis and tomato, have about six TFL1/FT family genes (Mimida et al. 2001, Carmel-Goren et al. 2003). However, the Lombardy poplar genome has nine (Igasaki et al. 2008), probably due to a recent gene duplication event 8–13 million years ago (Tuskan et al. 2006). Because apple has duplicated chromosomal regions as well, further study might reveal additional members falling into the clade of MFT or BFT in the TFL1/FT family in apple. MdFT1 and MdFT2 show different expression profiles in apple The expression pattern of MdFT1 and MdFT2 differed in that MdFT1 was expressed mainly in apical buds of FBS in the adult phase, whereas MdFT2 was expressed mainly in reproductive organs, including flower buds and young fruit (Fig. 3A, B). In addition, it was found that the transcripts of MdFT1 could hardly be detected in the tissues of 1-month-old seedlings in the juvenile phase, in contrast to the transcripts of MdFT2. In Satsuma mandarin, two citrus homologs of FT, CiFT1 and CiFT2, which are highly and specifically expressed in young fruit, were not involved in the transition to flowering, and only the third citrus homolog of FT, CiFT3, expressed at a relatively low level in leaves and stems but with no expression in reproductive organs, turned out to play a key role in regulating flower induction (Nishikawa et al. 2007). Considering that MdFT2 was highly expressed in the reproductive organs and that it was expressed similarly in the apical buds of both FBS and SS during the period of flower induction, MdFT1 might play a key role in the transition to flowering in apple (Fig. 3), although Arabidopsis FT and its paralog TSF are both highly expressed in flowers and developing siliques as well as in leaves (Kobayashi et al. 1999, Takada and Goto, 2003, Yamaguchi et al. 2005). The expression pattern of MdFT1 in various tissues was somewhat similar to that of tomato SFT (formerly SPD3, a tomato ortholog of FT; Carmel-Goren et al. 2003, Lifshitz et al. 2006) rather than that of Arabidopsis. In apple, flower induction occurs in late June [about 40–50 days after flowering (DAF)] and flower initiation with morphological changes at the apical meristem starts to form floral primordia in mid July (about 60 DAF) in Morioka, Japan (Kotoda et al. 2000). MdFT1 was expressed to relatively higher levels in the apical buds of FBS than in those of SS from the period of flower induction to the early stage of flower development. This result was consistent with the observation that the expression of MdTFL1 decreased rapidly from early June, although it is unclear where in apical buds MdFT1 was expressed (Fig. 3C, E). Recently, Hättasch et al. (2008) reported that the increase in transcription of MdFT (MdFT1 in this study), AFL1 and AFL2 (apple orthologs of LFY; Wada et al. 2002) in the apical meristems of current-year shoots of apple began 1–2 weeks earlier than flower initiation, suggesting that flower induction is initiated by these genes. The higher expression level of MdFT2 in mid September and early April might be related to the development of floral organs in apple (Fig. 3D), because the expression pattern of MdAP1 reflected the development of floral organs, such as sepals and receptacles (Fig. 3F, Kotoda et al. 2000). MdFT1 and MdFT2 have the potential to act as floral promoters and overexpression of MdFT1 confers precocious flowering in apple Transgenic Arabidopsis overexpressing MdFT1 and MdFT2 showed early flowering as compared with the control plant (Fig. 4, Table 1). Similar early-flowering phenotypes of transgenic Arabidopsis had been reported upon the ectopic expression of FT-like genes from heterologous plant species, such as citrus (Kobayashi et al. 1999), tomato (Teper-Bamnolker and Samach 2005, Lifschitz et al. 2006), grapevine (Sreekantan and Thomas 2006, Carmona et al. 2007), poplar (Hsu et al. 2006, Igasaki et al. 2008) and cucurbits (Lin et al. 2007), as well as the overexpression of FT and TSF. No differences in the extent of flower-promoting activity between MdFT1 and MdFT2 in transgenic Arabidopsis suggested that MdFT1 and MdFT2 both have the potential to act as floral promoters in apple (Table 1). However, expression analysis by qRT–PCR implied that, during floral transition, MdFT1 plays a more important role, at least in tissues within the apical buds of FBS, where transition to flowering occurs. Consequently, five out of six transgenic apples overexpressing MdFT1 flowered in vitro 8–12 months after Agrobacterium infection (Fig. 5). As a whole, the precocity of flowering in MdFT1-overexpressing lines was much higher than that in previously reported MdTFL1 down-regulated apples (Kotoda et al. 2003, Kotoda et al. 2006). The transgenic lines with 35SΩ : MdFT1 that produced flower buds in vitro had a tendency to have small, rounded leaves and show an extremely weak growth habit. In the transgenic lines MdFT1#1-1 and MdFT1#8, which showed in vitro flowering, MdFT1 was abundantly expressed, while the expression of MdFT1 was extremely low in line MdFT1#5, possibly due to co-suppression (Fig. 6A). As expected, line MdFT#5 has not flowered in vitro or in pots for at least 3 years after regeneration. These results suggest that there is a positive correlation between MdFT1 expression and flower induction in transgenic plants. Besides MdFT1, the heterologous gene PnFT3 (a poplar ortholog of FT; Igasaki et al. 2008) also promoted the transition to flowering when overexpressed in the apple cultivar ‘Greensleeves’ (Supplementary Fig. S2). Apple trees normally produce five flowers per cluster, and each flower has five sepals, petals and carpels and approximately 20 stamens. The changes in the number of floral organs in the transgenic plant might result from disturbance of the expression of floral organ identity genes, such as MADS-box genes, due to the overexpression of MdFT1 (Fig. 5E). Contrary to our results, Hättasch et al. (2009) reported that transgenic apples with MdFT did not flower for at least a year in vitro and in the greenhouse. This observation might be due to the fact that the regeneration competence of the transformed calli from some scion cultivars was severely impaired when FT-like genes were overexpressed, resulting in the selection of transgenic lines with little expression of those genes. Up-regulation of MADS-box genes could result in in vitro flowering in the transgenic lines overexpressing MdFT1 QRT–PCR analysis revealed that MdMADS12 and MdAP1 were significantly more up-regulated in the cultured shoots of transgenic lines that highly expressed MdFT1 than in those of line MdFT1#5, which had little expression of MdFT1, and a control (Fig. 6). These results seem to be consistent with the finding by Teper-Bamnolker and Samach (2005) that FUL and AP1 were highly misexpressed in the young seedlings and the older rosette leaves of Arabidopsis overexpressing FT. Interestingly, Flachowsky et al. (2007) reported that the overexpression of BpMADS4, a birch (Betula pendula) FUL-like MADS-box gene (Elo et al. 2001), induced early flowering in the apple cv. ‘Pinova’, with frequent in vitro flowering. In addition, the phenotype of the transgenic apple with 35S : BpMADS4 resembled that with 35SΩ : MdFT1 in that it produced solitary flower buds and rounded leaves in tissue culture. Considering that MdMADS12 is a putative ortholog of FUL, MdMADS12 might play an important role in flower induction of apple downstream of MdFT1. In Arabidopsis, FUL is required in several developmental processes, including silique and leaf development, and in the transition to flowering (Gu et al. 1998, Ferrandiz et al. 2000, Teper-Bamnolker and Samach, 2005). On the other hand, the up-regulation of MdAP1 in the transgenic lines indicated that the shoots had been florally induced because the expression of MdAP1 starts to increase in the apical buds of FBS after flower initiation (Figs. 3F and 6E; Kotoda et al. 2000). In addition, MdSOC1a and MdSOC1b were also up-regulated in the in vitro flowering lines with a similar expression pattern to MdAP1 (Fig. 6C–E), implicating them as the common targets of MdFT1, as SOC1 and AP1 are activated by the FT–FD complex in Arabidopsis (Abe et al. 2005, Wigge et al. 2005, Michaels 2009). Our results suggest that MdFT1 could function upstream of those genes to regulate flowering in apple, although whether or not the regulation is direct remains to be demonstrated. In future studies, the interaction of partner genes, such as transcription factors, with MdFT1 and/or MdFT2 will be investigated. Materials and Methods Plant materials The tissue samples of an apple (Malus × domestica Borkh.) cvs. ‘Fuji’ and ‘Jonathan’ in adult phase (age: 19–20 years) were collected from the experimental field at the National Institute of Fruit Tree Science in Morioka, Japan. One-month-old juvenile seedlings from a cross between apple cvs. ‘Fuji’ and ‘Orin’ were used for the expression analysis of apple FT-like genes in the juvenile phase. Micropropagated tissues of an apple rootstock cv. ‘JM2’ [Marubakaido ‘Seishi’ (Malus prunifolia Borkh. var. ringo Asami) × ‘Malling 9’ (‘M. 9’)] (Soejima et al. 1998) and an apple cv. ‘Greensleeves’ were kept at 24°C under LD conditions (16 h photoperiod; cool white fluorescent light, 50 μmol m−2 s−1) in the proliferating medium (MS medium containing B5 vitamin, 1 mg l−1 6-benzyl-aminopurine, 0.1 mg l−1 indole-3-butyric acid) in the culture box and subcultured every 4 weeks. The leaf explants of those apple cultivars were used for Agrobacterium-mediated transformation. For Arabidopsis transformation, wild-type plants of ecotype Columbia (Col) were used. Nucleic acid extraction and hybridization analysis The genomic DNA was isolated by a cetyltrimethylammonium bromide (CTAB)-based method modified by Yamamoto and Mukai as described in Kotoda et al. (2002). The genomic DNA (15 μg) was digested with EcoRI, HindIII or XbaI, electrophoresed on a 0.8% agarose gel, and then blotted onto Hybond-N+ nylon membranes (GE Healthcare Bio-Sciences Corp., Piscataway, NJ, USA). The blotted membrane was hybridized with digoxigenin- (DIG; Roche Diagnostics, Mannheim, Germany) labeled MdFT1 cDNA, which was amplified by PCR with a pair of primers, MdFT1(1→24) and MdFT1(508←528), and a DIG mixture (Roche Diagnostics). For DNA blot analysis of transformants, each genomic DNA of transgenic lines and a control plant was digested with HindIII and the blotted membrane was hybridized with a DIG-labeled nptII probe. The hybridization and washing was performed as described in Kotoda et al. (2002). Chemiluminescent signals were visualized using the LAS1000 image analyzer (Fuji Photo Film, Tokyo, Japan). The primer set is listed in Supplementary Table S1. Isolation of FT-like genes from apple The cDNA library derived from flower buds of ‘Fuji’ apple was constructed as described by Mimida et al. (2007). MdFT1 (accession No. AB161112; Kotoda and Wada 2005) labeled with [α-32P]dCTP (GE Healthcare Bio-Sciences Corp.) was used as a probe to screen 2.0 × 105 plaque-forming units (pfu) of the cDNA library. Positive plaques were excised to the pBluescript SK(−) phagemid (Stratagene, La Jolla, CA, USA), and approximately 20 clones were then sequenced using a DTCS Quick Start Kit for Dye Terminator Cycle Sequencing (Beckman Coulter, Fullerton, CA, USA) and an automated DNA sequencer CEQ 8000 (Beckman Coulter). To obtain genomic DNA of MdFT1, the cDNA of MdFT1 labeled with DIG was used as a probe to screen 2.0 × 105 pfu of the genomic library of ‘Fuji’ apple. The lambda clone DNAs from positive plaques were digested with restriction enzyme NotI and ligated into the corresponding site of pBluescriptII SK (+) (Stratagene). To obtain a MdFT2 genomic sequence from apple, on the other hand, PCR amplification of DNA fragments was performed in a mixture of a pair of primers MdFT2-EcoRI(−171→−148) and MdFT2(842←859)-XhoI, a high-fidelity DNA polymerase (KOD plus; Toyobo, Osaka, Japan) and 250–300 ng of genomic DNA of ‘Fuji’ apple. PCR was programmed for pre-heating at 94°C for 2 min followed by 30 cycles of 94°C for 30 s, 50°C for 30 s and 68°C for 2 min. The PCR-amplified fragments digested with EcoRI and XhoI were cloned into the corresponding site of pBluescriptII SK (+) (Stratagene), and then four clones were sequenced. The primer sets used in gene cloning are listed in Supplementary Table S1. Mapping of FT-like genes on the linkage map of apple Cleaved amplified polymorphic sequence (CAPS) markers for the MdFT1 and MdFT2 genes were developed, and genotyping was performed using an F1 mapping population with 72 seedlings from the cross between ‘Delicious’ and Mitsubakaido (Malus sieboldii Rehder) and another population with 83 seedlings from the cross between ‘Ralls Janet’ and Mitsubakaido (Igarashi et al. 2008). A population from ‘Delicious’ and Mitsubakaido was used for the mapping of MdFT1. PCR was performed with a pair of primers MdFT1-5f and MdFT1-4r. A 632 bp product was digested with HinfI, and the resultant 632 bp fragment specific to ‘Delicious’ was used for segregation analysis. On the other hand, a population from ‘Ralls Janet’ and Mitsubakaido was used for the mapping of MdFT2. PCR was performed with a pair of primers MdFT2-7f and MdFT2-8rm. A 1,064 bp product was digested with DraI, giving a 435 bp fragment specific to ‘Ralls Janet’, which was used for segregation analysis. The linkage map was constructed using JoinMap ver. 3.0 (Van Ooijen and Voorrips 2001). Primer sets, restriction enzymes and PCR conditions for the CAPS markers are described in Supplementary Table S2. Sequence analysis Amino acid sequences were analyzed using the Clustal X multiple sequence alignment program ver. 1.83 (Jeanmougin et al. 1998) and BioEdit ver. 7.7.0 (H. Hall, http://www.mbio .ncsu.edu/BioEdit/bioedit.html). The phylogenetic tree was displayed using the N–J plot unrooted (Perrière and Gouy, 1996) with bootstrap values for 1,000 resamplings in each branch. Expression analysis by q RT–PCR For analysis, total RNA was extracted from flower buds (at the balloon stage), sepals, petals, stamens, carpels, receptacles, peduncles, young fruit, mature fruit, peel, mature leaves (collected on June 24) and apical buds of FBS (collected on June 24 and August 7) from adult ‘Fuji’ apple trees; leaves, apical buds of vegetative shoots, stems and roots from 1-month-old juvenile seedlings and seeds from a cross between ‘Fuji’ and ‘Orin’ apples; apical buds of FBS (from June to the following April in 2004) and SS (from June to October in 2003) in ‘Jonathan’ apple trees; in vitro cultured shoots of ‘Fuji’ apple; and transgenic and non-transgenic lines of ‘JM2’ apple by using a PolyA Tract mRNA purification kit (Promega, Madison, WI, USA). For seasonal expression analysis, apical buds of all samples from the orchard were collected between 10 and 14 h. The first-strand cDNAs were synthesized from 1 μg of total RNAs in 20 μl of a reaction mixture using a QuantiTect Reverse Transcription kit (Qiagen GmbH, Hilden, Germany). The subsequent PCRs were performed with 1 μl of the first-strand cDNA as templates in a total volume of 12.5 μl by using ABI PRISM 7000 (Applied Biosystems, Foster City, CA, USA). Transcripts of MdFT1, MdFT2, MdLHP1a/MdLHP1b (MdLHP1a and MdLHP1b), MdTFL1, MdMADS12, MdSOC1a, MdSOC1b, MdAP1 and HistoneH3 were identified with the specific primers (the primer sets used in qRT–PCR analysis and the PCR conditions are described in Supplementary Table S1). An apple HistoneH3 gene (Kotoda et al. 2006) was used as an internal control. QRT–PCR was performed three times and transcript levels were normalized against apple HistoneH3. Construction of the transformation vector To construct a vector for the constitutive expression of MdFT1 and MdFT2, the coding region of each gene was amplified by PCR with thr pair of primers MdFT1-XbaI and MdFT1-SalI for MdFT1, and with thr pair of primers MdFT2-XbaI and MdFTa-SalI for MdFT2 (Supplementary Table S1). An amplified PCR product was subsequently digested with XbaI and SalI and then cloned into the XbaI/SalI sites of the modified pBI221 (pBI9526Ω) to be placed between the CaMV 35S promoter fused with the Ω sequence (Gallie and Walbot 1992) and the terminator of the nopaline synthase (nos) (Supplementary Fig. S1). The resultant plasmid was cut with ApaI, and then the fragment containing MdFT1 or MdFT2 was ligated into the same restriction enzyme site of the pSMAK312Blue binary vector (pSMAK312Blue; H. Ichikawa, in preparation). For vector construction of 35SΩ::PnFT3S, poplar PnFT3S amplified by PCR was inserted into the modified pSMAK193E (pSMAK193E; H. Ichikawa, in preparation) to be placed between 35SΩ and the 3′ region of the Arabidopsis rbcS-2B gene (TrbcS). Arabidopsis transformation Agrobacterium tumefaciens strain EHA101 was used to transform A. thaliana (Col) plants by the floral-dip method (Clough and Bent 1998). Kanamycin-resistant transformants were transplanted from the plate to moistened potting soil composed of vermiculite and perlite [1 : 1 (v/v)] after 2–5 adult leaves had developed and grown in the growth chamber (Biotron, Nippon Medical and Chemical Instruments Co., Ltd., Tokyo, Japan) set at 22°C under LD conditions (16 h photoperiod; cool white fluorescent light, 50 μmol m−2 s−1). Morphological analyses were performed on the second generation (T2). Apple transformation The apple cvs. ‘JM2’ and ‘Greensleeves’ were transformed with A. tumefaciens strain EHA101 bearing 35SΩ:MdFT1 and 35SΩ:PnFT3 (PnFT3; a poplar ortholog of FT), respectively. Agrobacterium tumefaciens was cultured overnight on a shaker in 20 ml of liquid LB medium with 100 mg l−1 trobicin (Pfizer Japan, Tokyo, Japan) at 28°C. After centrifugation, the pellet was resuspended with MS medium (Murashige and Skoog 1962) and further diluted to an optical density (OD) of 0.5–0.8 at 600 nm. Leaf explants were infected with the inoculum for 30 min. The selection of transgenic shoots for ‘Greensleeves’ was performed according to the procedure described by James et al. (1989) and Yao et al. (1995) and the procedure for ‘JM2’ will be published elsewhere (S. Takahashi, in preparation). Transformed shoots were multiplied in the proliferating medium supplemented with 50 mg l−1 kanamycin. Two to three shoots per line were grafted to apple rootstock ‘JM7’ [Marubakaido ‘Seishi’ (Malus prunifolia Borkh. var. ringo Asami) × ‘Malling 9’ (‘M. 9’)] (Soejima et al. 1998). Grafted apples were grown in an isolated greenhouse under natural day-length with the temperature set at 20–25°C during the growing season as described in Kotoda et al. (2006). Funding The Bio-oriented Technology Research Advancement Institution [Program for Promotion of Basic Research Activities for Innovative Biosciences (to N.K.)]; the Ministry of Agriculture, Forestry and Fisheries of Japan [‘Development of innovative plants and animals using transformation and cloning’ (to N.K.)]; the Ministry of Education, Sports, Science, and Technology of Japan [grant-in-aid for scientific research (to N.K.)]. Acknowledgments We thank Dr. H. Ichikawa for providing the binary vectors pSMAK312B and pSMAK193E, Dr. E. E. Hood for providing Agrobacterium tumefaciens EHA101, Ms. T. Sekita and A. Igarashi for their technical assistance, and Dr. E. Varkonyi-Gasic for critical reading of the manuscript. The nucleotide sequences reported in this paper have been submitted to the DDBJ under accession numbers AB458504 (MdFT2 mRNA), AB458505 (MdFT2 genomic DNA), AB458506 (MdFT1 genomic DNA), AB501124 (MdSOC1a mRNA), and AB501125 (MdSOC1b mRNA). References Abe M,  Kobayashi Y,  Yamamoto S,  Daimon Y,  Yamaguchi A,  Ikeda Y, et al.  FD, a bZIP protein mediating signals from the floral pathway integrator FT at the shoot apex,  Science ,  2005, vol.  309 (pg.  1052- 1056) Google Scholar CrossRef Search ADS PubMed  Ahn JH,  Miller D,  Winter VJ,  Banfield MJ,  Lee JH,  Yoo SY, et al.  A divergent external loop confers antagonistic activity on floral regulators FT and TFL1,  EMBO J. ,  2006, vol.  25 (pg.  605- 614) Google Scholar CrossRef Search ADS PubMed  Araki T.  Transition from vegetative to reproductive phase,  Curr. Opin. Plant Biol. ,  2001, vol.  4 (pg.  63- 68) Google Scholar CrossRef Search ADS PubMed  Benlloch R,  Berbel A,  Serrano-Mislata A,  Madueño F.  Floral initiation and inflorescence architechture; a comparative view,  Ann. Bot. ,  2007, vol.  100 (pg.  659- 676) Google Scholar CrossRef Search ADS PubMed  Bradley D,  Ratcliffe O,  Vincent C,  Carpenter R,  Coen E.  Inflorescence commitment and architecture in Arabidopsis,  Science ,  1997, vol.  275 (pg.  80- 83) Google Scholar CrossRef Search ADS PubMed  Carmel-Goren L,  Liu YS,  Lifschitz E,  Zamir D.  The SELF-PRUNING gene family in tomato,  Plant Mol. Biol. ,  2003, vol.  52 (pg.  1215- 1222) Google Scholar CrossRef Search ADS PubMed  Carmona MJ,  Calonje M,  Martínez-Zapater JM.  The FT/TFL1 gene family in grapevine,  Plant Mol. Biol ,  2007, vol.  63 (pg.  637- 650) Google Scholar CrossRef Search ADS PubMed  Chailahyan MK.  Internal factors of plant flowering,  Annu. Rev. Plant Physiol. ,  1968, vol.  19 (pg.  1- 37) Google Scholar CrossRef Search ADS   Chevreau E,  Lespinasse Y,  Gallet M.  Inheritance of pollen enzymes and polyploidy origin of apple (Malus × domestica Borkh.),  Theor. Appl. Genet. ,  1985, vol.  71 (pg.  268- 277) Google Scholar PubMed  Clough SJ,  Bent AF.  Floral dip: a simplified method for Agrobacterium-mediated transformation of Arabidopsis thaliana,  Plant J. ,  1998, vol.  16 (pg.  735- 743) Google Scholar CrossRef Search ADS PubMed  Corbesier L,  Vincent C,  Jang S,  Fornara F,  Fan Q,  Searle I., et al.  FT protein movement contributes to long-distance signaling in floral induction of Arabidopsis,  Science ,  2007, vol.  316 (pg.  1030- 1033) Google Scholar CrossRef Search ADS PubMed  Elo A,  Lemmetyinen J,  Turunen ML,  Tikka L,  Sopanen T.  Three MADS-box genes similar to APETALA1 and FRUITFULL from silver birch (Betula pendula),  Physiol. Plant. ,  2001, vol.  112 (pg.  95- 103) Google Scholar CrossRef Search ADS PubMed  Esumi T,  Tao R,  Yonemori K.  Isolation of LEAFY and TERMINAL FLOWER 1 homologues from six fruit tree species in the subfamily Maloideae of the family Rosaceae,  Sex. Plant Reprod. ,  2005, vol.  17 (pg.  277- 287) Google Scholar CrossRef Search ADS   Ferrándiz C,  Gu Q,  Martienssen R,  Yanofsky MF.  Redundant regulation of meristem identity and plant architecture by FRUITFULL, APETALA1 and CAULIFLOWER,  Development ,  2000, vol.  127 (pg.  725- 734) Google Scholar PubMed  Flachowsky H,  Peil A,  Sopanen T,  Elo A,  Hanke V.  Overexpression of BpMADS4 from silver birch (Betula pendula) in apple (Malus × domstica) induces early flowering,  Plant Breed. ,  2007, vol.  126 (pg.  137- 145) Google Scholar CrossRef Search ADS   Gallie DR,  Walbot V.  Identification of the motifs within the tobacco mosaic virus 5′-leader responsible for enhancing translation,  Nucleic Acids Res. ,  1992, vol.  20 (pg.  4631- 4638) Google Scholar CrossRef Search ADS PubMed  Gardiner SE,  Bus V.GM,  Rusholme RL,  Chagné D,  Rikkerink E.HA.  Kole C.  Apple,  Genome Mapping and Molecular Breeding in Plants, Vol. 4 Fruits and Nuts ,  2007 Berlin, Germany Springer-Verlag(pg.  1- 62) Giakountis A,  Coupland G.  Phloem transport of flowering signals,  Curr. Opin. Plant Biol. ,  2008, vol.  11 (pg.  687- 694) Google Scholar CrossRef Search ADS PubMed  Gu Q,  Ferrandiz C,  Yanofsky MF,  Martienssen R.  The FRUITFULL MADS-box gene mediates cell differentiation during Arabidopsis fruit development,  Development ,  1998, vol.  125 (pg.  1509- 1517) Google Scholar PubMed  Hackett WP.  Juvenility, maturation, and rejuvenility in woody plants,  Hortic. Rev. ,  1985, vol.  7 (pg.  109- 155) Hanke M.-V,  Flachowsky H,  Peil A,  Hättasch C.  No flower no fruit—Genetic potentials to trigger flowering in fruit trees,  Genes Genomes Genomics ,  2007, vol.  1 (pg.  1- 20) Hanzawa Y,  Money T,  Bradley D.  A single amino acid converts a repressor to an activator of flowering,  Proc. Natl Acad. Sci. USA ,  2005, vol.  102 (pg.  7748- 7753) Google Scholar CrossRef Search ADS   Hättasch C,  Flachowsky H,  Hanke M.-V,  Lehmann S,  Gau A,  Kapturska,  D..  The switch to flowering: genes involved in floral induction of the apple ‘Pinova’ and the role of the flowering gene MdFT,  Acta Hortic. ,  2009, vol.  839 (pg.  701- 705) Google Scholar CrossRef Search ADS   Hättasch C,  Flachowsky H,  Kapturska D,  Hanke M.-V.  Isolation of flowering genes and seasonal changes in their transcript levels related to flower induction and initiation in apple (Malus domestica),  Tree Physiol. ,  2008, vol.  28 (pg.  1459- 1466) Google Scholar CrossRef Search ADS PubMed  Hsu CY,  Liu Y,  Luthe DS,  Yuceer C.  Poplar FT2 shortens the juvenile phase and promotes seasonal flowering,  Plant Cell ,  2006, vol.  18 (pg.  1846- 1861) Google Scholar CrossRef Search ADS PubMed  Igarashi M,  Abe Y,  Hatsuyama Y,  Ueda T,  Fukasawa-Akada T,  Kon T, et al.  Linkage maps of the apple (Malus × domestia Borkh.) cvs. ‘Ralls Janet’ and ‘Delicious’ include newly developed EST markers,  Mol. Breed. ,  2008, vol.  22 (pg.  95- 118) Google Scholar CrossRef Search ADS   Igasaki T,  Watanabe Y,  Nishiguchi M,  Kotoda N.  The FLOWERING LOCUS T/TERMINAL FLOWER 1 family in Lombardy poplar,  Plant Cell Physiol. ,  2008, vol.  49 (pg.  291- 300) Google Scholar CrossRef Search ADS PubMed  Jack T.  Molecular and genetic mechanisms of floral control,  Plant Cell ,  2004, vol.  16 (pg.  S1- S17) Google Scholar CrossRef Search ADS PubMed  James DJ,  Passey AJ,  Barbara DJ,  Bevan M.  Genetic transformation of apple (Malus pumila Mill.) using a disarmed Ti-binary vector,  Plant Cell Rep. ,  1989, vol.  7 (pg.  658- 661) Google Scholar PubMed  Jeanmougin F,  Thompson JD,  Gouy M,  Higgins DG,  Gibson TJ.  Multiple sequence alignment with Clustal X,  Trends Biochem. Sci. ,  1998, vol.  23 (pg.  403- 405) Google Scholar CrossRef Search ADS PubMed  Kardailsky I,  Shukla VK,  Ahn JH,  Dagenais N,  Christensen SK,  Nguyen JT, et al.  Activation tagging of the floral inducer FT,  Science ,  1999, vol.  286 (pg.  1962- 1965) Google Scholar CrossRef Search ADS PubMed  Kobayashi Y,  Kaya H,  Goto K,  Iwabuchi M,  Araki T.  A pair of related genes with antagonistic roles in mediating flowering signals,  Science ,  1999, vol.  286 (pg.  1960- 1962) Google Scholar CrossRef Search ADS PubMed  Kojima S,  Takahashi Y,  Kobayashi Y,  Monna L,  Sasaki T,  Araki T, et al.  Hd3a, a rice ortholog of the Arabidopsis FT gene, promotes transition to flowering downstream of Hd1 under short-day conditions,  Plant Cell Physiol. ,  2002, vol.  43 (pg.  1096- 1105) Google Scholar CrossRef Search ADS PubMed  Koornneef M,  Alonso-Blanco C,  Peeters A.JM,  Soppe W.  Genetic control of flowering time in Arabidopsis,  Annu. Rev. Plant Physiol. Plant Mol. Biol. ,  1998, vol.  49 (pg.  345- 370) Google Scholar CrossRef Search ADS PubMed  Kotoda N,  Iwanami H,  Takahashi S,  Abe K.  Antisense expression of MdTFL1, a TFL1-like gene, reduces the juvenile phase in apple,  J. Amer. Soc. Hortic. Sci. ,  2006, vol.  131 (pg.  74- 81) Kotoda N,  Wada M.  MdTFL1, a TFL1-like gene of apple, retards the transition from the vegetative to reproductive phase in transgenic Arabidopsis,  Plant Sci. ,  2005, vol.  168 (pg.  95- 104) Google Scholar CrossRef Search ADS   Kotoda N,  Wada M,  Komori S,  Kidou S,  Abe K,  Masuda T,  Soejima T.  Expression pattern of homologues of floral meristem identity genes LFY and AP1 during flower development in apple,  J. Amer. Soc. Hortic. Sci. ,  2000, vol.  125 (pg.  398- 403) Kotoda N,  Wada M,  Kusaba S,  Kano-Murakami Y,  Masuda T,  Soejima J.  Overexpression of MdMADS5, an APETALA1-like gene of apple, causes early flowering in transgenic Arabidopsis,  Plant Sci. ,  2002, vol.  162 (pg.  679- 687) Google Scholar CrossRef Search ADS   Kotoda N,  Wada M,  Masuda T,  Soejima J.  The break-through in the reduction of juvenile phase in apple using transgenic approaches,  Acta Hortic. ,  2003, vol.  625 (pg.  337- 343) Google Scholar CrossRef Search ADS   Levy YY,  Dean C.  The transition to flowering,  Plant Cell ,  1998, vol.  10 (pg.  1973- 1990) Google Scholar CrossRef Search ADS PubMed  Lifschitz E,  Eviatar T,  Rozman A,  Shalit A,  Goldshmidt A,  Amsellem Z, et al.  The tomato FT ortholog triggers systemic signals that regulate growth and flowering and substitute for diverse environmental stimuli,  Proc. Natl Acad. Sci. USA ,  2006, vol.  103 (pg.  6398- 6403) Google Scholar CrossRef Search ADS   Lin M.-K,  Belanger H,  Lee Y.-J,  Varkonyi-Gasic E,  Taoka K.-I,  Miura E, et al.  FLOWERING LOCUS T protein may act as the long-distance florigenic signal in the cucurbits,  Plant Cell ,  2007, vol.  19 (pg.  1488- 1506) Google Scholar CrossRef Search ADS PubMed  Michaels SD.  Flowering time regulation produces much fruit,  Curr. Opin. Plant Biol. ,  2009, vol.  12 (pg.  75- 80) Google Scholar CrossRef Search ADS PubMed  Mimida N,  Goto K,  Kobayashi Y,  Araki T,  Ahn JH,  Weigel D, et al.  Functional divergence of the TFL1-like gene family in Arabidopsis revealed by characterization of a novel homologue,  Genes Cells ,  2001, vol.  6 (pg.  327- 336) Google Scholar CrossRef Search ADS PubMed  Mimida N,  Kidou S.-I,  Kotoda N.  Constitutive expression of two apple (Malus × domestica Borkh.) homolog genes of LIKE HETEROCHROMATIN PROTEIN1 affects flowering time and whole-plant growth in transgenic Arabidopsis,  Mol. Genet. Genom. ,  2007, vol.  278 (pg.  295- 305) Google Scholar CrossRef Search ADS   Mimida N,  Kotoda N,  Ueda T,  Igarashi M,  Hatsuyama Y,  Iwanami H, et al.  Four TFL1/CEN-like genes on distinct linkage groups show different expression patterns to regulate vegetative and reproductive development in apple (Malus × domestica Borkh.),  Plant Cell Physiol. ,  2009, vol.  50 (pg.  394- 412) Google Scholar CrossRef Search ADS PubMed  Murashige T,  Skoog F.  A revised medium for rapid growth and bioassays with tobacco tissue culture,  Physiol. Plant. ,  1962, vol.  15 (pg.  473- 497) Google Scholar CrossRef Search ADS   Nishikawa F,  Endo T,  Shimada T,  Fujii H,  Shimizu T,  Omura M.  Increased CiFT abundance in the stem correlates with floral induction by low temperature in Satsuma mandarin (Citrus unshiu Marc.),  J. Exp. Bot. ,  2007, vol.  54 (pg.  2439- 2448) Google Scholar CrossRef Search ADS   Notaguchi M,  Abe M,  Kimura T,  Daimon Y,  Kobayashi T,  Yamaguchi A, et al.  Long-distance, graft-transmissible action of Arabidopsis FLOWERING LOCUS T protein to promote flowering,  Plant Cell Physiol. ,  2008, vol.  49 (pg.  1645- 1658) Google Scholar CrossRef Search ADS PubMed  Ohshima S,  Murata M,  Sakamoto W,  Ogura Y,  Motoyoshi F.  Cloning and molecular analysis of the Arabidopsis gene Terminal Flower 1,  Mol. Gen. Genet. ,  1997, vol.  254 (pg.  186- 194) Google Scholar CrossRef Search ADS PubMed  O’Rourke D,  Janick J,  Sansavini S.  World apple cultivar dynamics,  Chron. Hortic. ,  2003, vol.  43 (pg.  10- 13) Pearl RT.  Apple rootstocks I to XVI,  J. South-Eastern Agric. Coll., ,  1932, vol.  30 (pg.  194- 213) Perrière G,  Gouy M.  WWW-Query: an on-line retrieval system for biological sequence banks,  Biochimie ,  1996, vol.  78 (pg.  364- 369) Google Scholar CrossRef Search ADS PubMed  Purwestri YA,  Ogaki Y,  Tamaki S,  Tsuji H,  Shimamoto K.  The 14-3-3- protein GF14c acts as a negative regulator of flowering in rice by interacting with the florigen Hd3a,  Plant Cell Physiol. ,  2009, vol.  50 (pg.  429- 438) Google Scholar CrossRef Search ADS PubMed  Ratcliffe OJ,  Amaya I,  Vincent CA,  Rothstein S,  Carpenter R,  Coen ES, et al.  A common mechanism controls the life cycle and architecture of plants,  Development ,  1998, vol.  125 (pg.  1609- 1615) Google Scholar PubMed  Ratcliffe OJ,  Bradley DJ,  Coen ES.  Separation of shoot and floral identity in Arabidopsis,  Development ,  1999, vol.  126 (pg.  1109- 1120) Google Scholar PubMed  Sadamori S,  Yoshida Y,  Murakami H,  Ishizuka S.  New apple variety ‘Fuji’,  Bull. Hortic. Res. Station, Jpn, Ser. (C) ,  1963, vol.  1 (pg.  1- 6) Sax K.  The origin of the Pomoideae,  Proc. Amer. Soc. Hortic. Sci. ,  1933, vol.  30 (pg.  147- 150) Soejima J,  Bessho H,  Tsuchiya S,  Komori S,  Abe K,  Kotoda N.  Breeding of Fuji apples and performance on JM rootstocks,  Compact Fruit Tree ,  1998, vol.  31 (pg.  22- 24) Sreekantan L,  Thomas MR.  VvFT and VvMADS8, the grapevine homologues of the floral integrators FT and SOC1, have unique expression patterns in grapevine and hasten flowering in Arabidopsis,  Funct. Plant Biol. ,  2006, vol.  33 (pg.  1129- 1139) Google Scholar CrossRef Search ADS   Sung S.-K,  Yu G.-H,  An G.  Characterization of MdMADS2, a member of the SQUAMOSA subfamily of genes, in apple,  Plant Physiol. ,  1999, vol.  120 (pg.  969- 978) Google Scholar CrossRef Search ADS PubMed  Takada S,  Goto K.  TERMINAL FLOWER2, a HETEROCHROMATIN PROTEIN1-like protein of Arabidopsis, counteracts the activation of FLOWERING LOCUS T by CONSTANS in the vascular tissues of leaves to regulate flowering time,  Plant Cell ,  2003, vol.  15 (pg.  2856- 2865) Google Scholar CrossRef Search ADS PubMed  Tamaki S,  Matsuo S,  Wong HL,  Yokoi S,  Shimamoto K.  Hd3a protein is a mobile flowering signal in rice,  Science ,  2007, vol.  316 (pg.  1033- 1036) Google Scholar CrossRef Search ADS PubMed  Teper-Bamnolker R,  Samach A.  The flowering integrator FT regulates SEPALLATA3 and FRUITFULL accumulation in Arabidopsis leaves,  Plant Cell ,  2005, vol.  17 (pg.  2661- 2675) Google Scholar CrossRef Search ADS PubMed  Tuskan GA,  DiFazio S,  Jansson S,  Bohlmann J,  Grigoriev I,  Hellsten U, et al.  The genome of Black Cottonwood, Populus trichocarpa (Torr. & Gray),  Science ,  2006, vol.  313 (pg.  1596- 1604) Google Scholar CrossRef Search ADS PubMed  van der Linden CG,  Vosman B,  Smulders M.JM.  Cloning and characterization of four apple MADS box genes isolated from vegetative tissues,  J. Exp. Bot. ,  2002, vol.  53 (pg.  1025- 1036) Google Scholar CrossRef Search ADS PubMed  Van Ooijen JW,  Voorrips RE. ,  JoinMap 3.0, Software for the calculation of linkage maps ,  2001 Wageningen Plant Research International Wada M,  Cao Q,  Kotoda N,  Soejima J,  Masuda T.  Apple has two orthologues of FLORICAULA/LEAFY involved in flowering,  Plant Mol. Biol. ,  2002, vol.  49 (pg.  567- 577) Google Scholar CrossRef Search ADS PubMed  Wigge PA,  Kim MC,  Jaeger KE,  Busch W,  Schmid M,  Lohmann JU, et al.  Integration of spatial and temporal information during floral induction in Arabidopsis,  Science ,  2005, vol.  309 (pg.  1056- 1059) Google Scholar CrossRef Search ADS PubMed  Wilkie JD,  Sedgley M,  Olesen T.  Regulation of floral initiation in horticultural trees,  J. Exp. Bot. ,  2008, vol.  59 (pg.  3215- 3228) Google Scholar CrossRef Search ADS PubMed  Yamaguchi A,  Kobayashi Y,  Goto K,  Abe M,  Araki T.  TWIN SISTER OF FT (TSF) acts as a floral pathway integrator redundantly with FT,  Plant Cell Physiol. ,  2005, vol.  46 (pg.  1175- 1189) Google Scholar CrossRef Search ADS PubMed  Yao J.-L,  Cohen D,  Atkinson R,  Richardson K,  Morris B.  Regeneration of transgenic plants from the commercial apple cultivars Royal Gala,  Plant Cell Rep. ,  1995, vol.  14 (pg.  407- 412) Google Scholar CrossRef Search ADS PubMed  Yao J.-L,  Dong Y.-H,  Kvarnheden A,  Morris B.  Seven MADS-box genes in apple are expressed in different parts of the fruit,  J. Amer. Soc. Hortic. Sci. ,  1999, vol.  124 (pg.  8- 13) Yeung K,  Seitz T,  Li SF,  Janosch P,  McFerran B,  Kaiser C, et al.  Suppression of Raf-1 kinase activity and MAP kinase signaling by RKIP,  Nature ,  1999, vol.  401 (pg.  173- 177) Google Scholar CrossRef Search ADS PubMed  Yoo SY,  Kardailsky I,  Lee JS,  Weigel D,  Ahn JH.  Acceleration of flowering by overexpression of MFT (MOTHER OF FT AND TFL1),  Mol. Cells ,  2004, vol.  17 (pg.  95- 101) Google Scholar PubMed  Yoshida Y,  Haniuda T,  Tsuchiya S,  Sanada T,  Masuda T,  Bessho H, et al.  New apple cultivar ‘Sansa’,  Bull. Fruit Tree Res Station, Ser. C (Morioka, Japan) ,  1988, vol.  15 (pg.  1- 12) Zeevaart JA.  Leaf-produced floral signals,  Curr. Opin. Plant Biol. ,  2008, vol.  11 (pg.  541- 547) Google Scholar CrossRef Search ADS PubMed  Zimmerman RH.  Juvenility and flowering in woody plants: a review,  HortScience ,  1972, vol.  7 (pg.  447- 455) Abbreviations Abbreviations AP1 APETALA1 ATC ARABIDOPSIS THALIANA CENTRORADIALIS BFT BROTHER OF FT AND TFL1 CaMV cauliflower mosaic virus CAPS cleaved amplified polymorphic sequence CEN CENTRORADIALIS DAF days after flowering DIG digoxigenin FBS fruit-bearing shoots FLC FLOWERING LOCUS C FT FLOWERING LOCUS T FUL FRUITFULL LD long day LFY LEAFY LG linkage group MFT MOTHER OF FT AND TFL1 N–J Neighbor–Joining RT–PCR reverse transcription–PCR SOC1 SUPPRESSOR OF OVEREXPRESSION OF CONSTANS1 SS succulent shoots SSR simple sequence repeat TFL1 TERMINAL FLOWER1 TSF TWIN SISTER OF FT. © The Author 2010. Published by Oxford University Press on behalf of Japanese Society of Plant Physiologists. All rights reserved. For permissions, please email: journals.permissions@oxfordjournals.org http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Plant and Cell Physiology Oxford University Press

Loading next page...
 
/lp/oxford-university-press/molecular-characterization-of-flowering-locus-t-like-genes-of-apple-H0NO15uHIC

References (88)

Publisher
Oxford University Press
Copyright
© The Author 2010. Published by Oxford University Press on behalf of Japanese Society of Plant Physiologists. All rights reserved. For permissions, please email: journals.permissions@oxfordjournals.org
ISSN
0032-0781
eISSN
1471-9053
DOI
10.1093/pcp/pcq021
pmid
20189942
Publisher site
See Article on Publisher Site

Abstract

Abstract The two FLOWERING LOCUS T (FT)-like genes of apple (Malus × domestica Borkh.), MdFT1 and MdFT2, have been isolated and characterized. MdFT1 and MdFT2 were mapped, respectively, on distinct linkage groups (LGs) with partial homoeology, LG 12 and LG 4. The expression pattern of MdFT1 and MdFT2 differed in that MdFT1 was expressed mainly in apical buds of fruit-bearing shoots in the adult phase, with little expression in the juvenile tissues, whereas MdFT2 was expressed mainly in reproductive organs, including flower buds and young fruit. On the other hand, both genes had the potential to induce early flowering since transgenic Arabidopsis, which ectopically expressed MdFT1 or MdFT2, flowered earlier than wild-type plants. Furthermore, overexpression of MdFT1 conferred precocious flowering in apple, with altered expression of other endogenous genes, such as MdMADS12. These results suggest that MdFT1 could function to promote flowering by altering the expression of those genes and that, at least, other genes may play an important role as well in the regulation of flowering in apple. The long juvenile period of fruit trees prevents early cropping and efficient breeding. Our findings will be useful information to unveil the molecular mechanism of flowering and to develop methods to shorten the juvenile period in various fruit trees, including apple. Introduction Apple (Malus spp.), which belongs to the Rosaceae subfamily Maloideae, is one of the most important fruit tree crops in the world. However, the long juvenile phase characteristic of woody plants, including apple, makes their breeding cycle slower (Zimmerman 1972, Hackett 1985). In apple, the juvenile phase generally lasts from 4 to 8 years or more. For example, the ‘Fuji’ apple (Malus × domestica Borkh.) from ‘Ralls Janet’ × ‘Delicious’, one of the most popular apple cultivars in the world (O’Rourke et al. 2003), set fruit for the first time 12 years after sowing (Sadamori et al. 1963), whereas the ‘Sansa’ apple from ‘Gala’ × ‘Akane’, top-grafted on fruit-bearing trees for early selection, set fruit after 7 years (Yoshida et al. 1988). In general, selection efficiency is limited until the seedlings set fruit. In the case of fruit breeding, the most important traits for selection, such as texture, flavor, ripening time and shelf life, are related to the fruit itself. Therefore, various practical techniques to accelerate flowering and fruiting of seedlings in the juvenile phase have been considered for years. In apple, grafting the seedling onto fruit-bearing trees (top grafting) or onto dwarfing rootstocks, such as ‘Malling 9 (M. 9)’ (Pearl 1932) and ‘JM1’ (Soejima et al. 1998), results in earlier flowering by 1 or 2 years. After the end of the juvenile phase, flower induction occurs in late June, and flower initiation with morphological changes at the apical meristem starts to form floral primordia in mid July in Morioka, Japan (Kotoda et al. 2000). However, little is known about the physiological and genetic factors involving the transition to flowering in apple. On the other hand, studies on Arabidopsis (Arabidopsis thaliana) and snapdragon (Antirrhinum majus) have led to the identification of many genes involved in flowering. In Arabidopsis, the transition from the vegetative to the reproductive phase is initiated by four independent pathways of signal transduction, i.e. the autonomous, and gibberellin-, vernalization- and light-dependent pathways (for reviews, see Koornneef 1998, Levy and Dean 1998). These signals are transmitted via integrator genes, such as FLOWERING LOCUS T (FT), SUPPRESSOR OF OVEREXPRESSION OF CONSTANS1 (SOC1)/AGAMOUS LIKE20 and FLOWERING LOCUS C (FLC), to the floral meristem identity genes APETALA1 (AP1) and LEAFY (LFY) at the apical meristems (for reviews, see Araki 2001, Jack 2004, Michaels 2009). In addition to the integrator genes, TERMINAL FLOWER1 (TFL1; Bradley et al. 1997, Oshima et al. 1997) is also a key gene that represses flowering and maintains the inflorescence meristem by preventing the expression of AP1 and LFY (Ratcliffe et al. 1998, Ratcliffe et al. 1999). TFL1 was the first member of the TFL1/FT family to be identified as a homolog of an animal phosphatidylethanolamine-binding protein (PEBP), considered to be a Raf-1 kinase inhibitor (Yeung et al. 1999). Thereafter, the Arabidopsis genome was found to contain six members of the TFL1/FT family, including ARABIDOPSIS THALIANA CENTRORADIALIS (ATC), BROTHER OF FT AND TFL1 (BFT), FT, MOTHER OF FT AND TFL1 (MFT) and TWIN SISTER OF FT (TSF) (Kardailsky et al. 1999, Kobayashi et al. 1999, Mimida et al. 2001, Yoo et al. 2004, Yamaguchi et al. 2005). FT was identified as a causative gene for a typical late- flowering mutant ft, the phenotype of which was opposite to that of tfl1 (Kardailsky et al. 1999, Kobayashi et al. 1999). Hanzawa et al. (2005) reported that the antagonistic functions of a floral activator and a repressor encoded by FT and TFL1, respectively, could be converted by a single amino acid substitution. Recently, it was found that the FT protein moves from the leaf to the shoot apex, where FT interacts with the bZIP transcription factor, FD, to activate AP1 in Arabidopsis (Abe et al. 2005, Wigge et al. 2005, Corbesier et al. 2007). In rice (Oryza sativa), it was demonstrated that the protein encoded by Hd3a (Kojima et al. 2002), a rice equivalent of the FT gene, similarly travels from the leaf to the shoot apex (Tamaki et al. 2007), probably in the absence of GF14c (a G-box factor 14-3-3c protein), a direct inhibitor of Hd3a (Purwestri et al. 2009). Furthermore, grafting experiments in cucurbit species (Cucurbita maxima and C. moschata) demonstrated that the FT protein, but not FT mRNA crossed the graft union in the phloem translocation stream (Lin et al. 2007). Experiments in Arabidopsis also revealed that the FT protein is graft transmissible (Notaguchi et al. 2008). The results of these studies suggest that the FT protein acts as a mobile flower-inducing signal, such as a florigen, as noted by Chailahyan (for reviews, see Chailahyan 1968, Giakountis and Coupland 2008, Zeevaart 2008). Based on studies of Arabidopsis and other plant species, much effort has been devoted to unveiling the molecular mechanism of flowering and to manipulate the flowering time in horticultural trees since the last decade (for reviews, see Benlloch et al. 2007, Hanke et al. 2007, Wilkie et al. 2008). Apple genes related to flowering have been investigated by analyzing its expression patterns (Yao et al. 1999, Kotoda et al. 2000, van der Linden et al. 2002, Esumi et al. 2005, Hättasch et al. 2008) and heterologous transgenic plants (Sung et al. 1999, Kotoda et al. 2002, Wada et al. 2002). In this context, we have searched for the genes that maintain the juvenile/vegetative growth or induce the transition to flowering in apple. As a result, transgenic apple expressing the antisense RNA of MdTFL1 (an apple ortholog of TFL1) showed precocious flowering (Kotoda et al. 2003). Therefore, we further investigated the genes of the TFL1/FT family as candidates that can shorten the juvenile/vegetative growth phase and manipulate the flowering time in apple (Kotoda and Wada 2005, Kotoda et al. 2006, Mimida et al. 2009). In this study, we have isolated and characterized the two FT-like genes of apple and examined their function using transgenic Arabidopsis. Furthermore, we investigated transgenic apple overexpressing the FT-like gene to clarify its function in apple. Results Identification of FT-like genes from apple We previously isolated MdFT, an apple homolog of FT, and classified it as an FT-like gene by phylogenetic analysis (Kotoda and Wada 2005). However, the result of DNA blot analysis suggested the existence of another MdFT-like gene in apple (Fig. 1A). Therefore, we screened the cDNA library derived from flower buds of ‘Fuji’ apple to obtain FT-like genes homologous to MdFT. As a result, positive clones converged to two kinds of cDNA. One corresponded to MdFT, which we renamed MdFT1, but the other seemed to be a novel gene homologous to MdFT, which we designated MdFT2. The MdFT1 and MdFT2 genes consisted of four exons of 198, 62, 41 and 224 bp encoding a putative protein of 174 amino acids, which resembled the genomic structure of the FT gene (Fig. 1B). The coding sequence of MdFT1 cDNA exhibited 94.2% identity to that of MdFT2 cDNA at the nucleotide level, although the sequence of each exon showed slightly different identities for MdFT1 and MdFT2 (93.4, 100.0, 95.1 and 95.1% for the first, second, third and fourth exons, respectively). The genomic sequence including the coding region of MdFT2 (1,963 bp) was longer than that of MdFT1 (1,620 bp) due to the extended introns of MdFT2 (Fig. 1B). Fig. 1 View largeDownload slide (A) DNA blot analysis for FT-like genes of apple. The genomic DNA (15 μg) of the ‘Fuji’ apple was digested individually with EcoRI, HindIII or XbaI and then separated on a 0.8% (w/v) agarose gel. DNA bands were transferred to Hybond N+ and hybridized with a digoxigenin (DIG)-labeled MdFT1 cDNA. Hybridization was performed in DIG Easy Hyb (Roche Diagnostics) at 42°C for 16 h followed by two washes in 0.5 × SSC containing 1% (w/v) SDS at 68°C for 20 min. The molecular size markers are shown in kb on the left. (B) Schematic representation of the genomic organization of FT-like genes of the apple and Arabidopsis FT gene. Boxes represent exons and lines represent introns. Numbers represent the lengths (bp) of exons (above the boxes) and introns (below the lines). The scale bar on the map represents approximately 200 bp. H represents a restriction enzyme site of HindIII in MdFT1 and MdFT2 genes. (C) Comparison of the deduced protein sequence of MdFT1 and MdFT2 (accession Nos. AB161112 and AB458504, respectively) with those of the TFL1/FT family from apple, Arabidopsis, tomato, grapevine, Lombardy poplar and pea. The AGI (Arabidopsis Genome Initiative; http://www.arabidopsis.org) code or accession number of each gene, as pertinent, is as follows: FT (AGI code At1g65480), TSF (AGI code At4g20370), SFT (tomato, accession No. AY186735), VvFT (grapevine, accession No. DQ871590), PnFT1 (Lombardy poplar, accession No. AB106111), TFL1 (AGI code At5g03840), ATC (AGI code At2g27550), SP (tomato, accession No. U84140), PsTFL1a (pea, accession No. AY340579), VvTFL1A (VvTFL1) (grapevine, accession No. DQ871591), MdTFL1 (apple, accession No. AB052994) and MdCENa (apple, accession No. AB366641). Amino acids in black and in gray are identical and similar, respectively, in at least seven of the 14 members of the TFL1/FT family. Amino acids common to genes in the FT clade are shown in pink and those in the TFL1 clade are shown in green. The gaps indicated by dashes are attributed to the lack of amino acids. The asterisks indicate the amino acid positions related to antagonistic functions between TFL1 and FT (Hanzawa et al. 2005, Ahn et al. 2006). The first square in red, called segment B, represents the region of a potential ligand-binding pocket in TFL1/FT family proteins, and the second square in red represents a region more conserved in FT than in TFL1 (Ahn et al. 2006). The triangles show the intron positions, which are conserved in the TFL1/FT family proteins. (D) Phylogenetic analysis of TFL1/FT family proteins. The tree was constructed by the Neighbor–Joining (N–J) method for the deduced amino acid sequence of the members of the TFL1/FT family from apple (MdFT1, MdFT2, MdTFL1, MdTFL1a, accession No. AB366643; MdCENa and MdCENb, accession No. AB366642), Arabidopsis (ATC, BFT, FT, MFT, TFL1 and TSF), grapevine [VvFT and VvMFT, accession No. DQ871594; VvTFL1A (VvTFL1) and VvTFL1B, accession No. DQ871592; VvTFL1C, accession No. DQ871593], Lombardy poplar (PnFT1 and PnFT3, accession No. AB110612; PnFTL1, accession No. AB369067; PnFTL4, accession No. AB181241), morning glory (InFT1, accession No. ABW73562), orange (CsTFL, accession No. AY344244), pea (PsTFL1a and PsTFL1c, accession No. AY343326), Satsuma mandarin (CiFT1, accession No. AB027456), snapdragon [CENTRORADIALIS (CEN), accession No. S81193) and tomato (SP, SFT and SP9D, accession No. AY186738). The protein sequence data were obtained from the DNA database (DDBJ/EMBL/GenBank). The N–J unrooted dendrograms were generated from the alignment of deduced amino acids with the Clustal X program, and the phylogenetic tree was displayed using the N–J plot unrooted program (Perrière and Gouy, 1996). Bootstrap values for 1,000 resamplings are shown in each branch. The unit for the scale bar displays branch lengths (0.02 substitutions/site). Fig. 1 View largeDownload slide (A) DNA blot analysis for FT-like genes of apple. The genomic DNA (15 μg) of the ‘Fuji’ apple was digested individually with EcoRI, HindIII or XbaI and then separated on a 0.8% (w/v) agarose gel. DNA bands were transferred to Hybond N+ and hybridized with a digoxigenin (DIG)-labeled MdFT1 cDNA. Hybridization was performed in DIG Easy Hyb (Roche Diagnostics) at 42°C for 16 h followed by two washes in 0.5 × SSC containing 1% (w/v) SDS at 68°C for 20 min. The molecular size markers are shown in kb on the left. (B) Schematic representation of the genomic organization of FT-like genes of the apple and Arabidopsis FT gene. Boxes represent exons and lines represent introns. Numbers represent the lengths (bp) of exons (above the boxes) and introns (below the lines). The scale bar on the map represents approximately 200 bp. H represents a restriction enzyme site of HindIII in MdFT1 and MdFT2 genes. (C) Comparison of the deduced protein sequence of MdFT1 and MdFT2 (accession Nos. AB161112 and AB458504, respectively) with those of the TFL1/FT family from apple, Arabidopsis, tomato, grapevine, Lombardy poplar and pea. The AGI (Arabidopsis Genome Initiative; http://www.arabidopsis.org) code or accession number of each gene, as pertinent, is as follows: FT (AGI code At1g65480), TSF (AGI code At4g20370), SFT (tomato, accession No. AY186735), VvFT (grapevine, accession No. DQ871590), PnFT1 (Lombardy poplar, accession No. AB106111), TFL1 (AGI code At5g03840), ATC (AGI code At2g27550), SP (tomato, accession No. U84140), PsTFL1a (pea, accession No. AY340579), VvTFL1A (VvTFL1) (grapevine, accession No. DQ871591), MdTFL1 (apple, accession No. AB052994) and MdCENa (apple, accession No. AB366641). Amino acids in black and in gray are identical and similar, respectively, in at least seven of the 14 members of the TFL1/FT family. Amino acids common to genes in the FT clade are shown in pink and those in the TFL1 clade are shown in green. The gaps indicated by dashes are attributed to the lack of amino acids. The asterisks indicate the amino acid positions related to antagonistic functions between TFL1 and FT (Hanzawa et al. 2005, Ahn et al. 2006). The first square in red, called segment B, represents the region of a potential ligand-binding pocket in TFL1/FT family proteins, and the second square in red represents a region more conserved in FT than in TFL1 (Ahn et al. 2006). The triangles show the intron positions, which are conserved in the TFL1/FT family proteins. (D) Phylogenetic analysis of TFL1/FT family proteins. The tree was constructed by the Neighbor–Joining (N–J) method for the deduced amino acid sequence of the members of the TFL1/FT family from apple (MdFT1, MdFT2, MdTFL1, MdTFL1a, accession No. AB366643; MdCENa and MdCENb, accession No. AB366642), Arabidopsis (ATC, BFT, FT, MFT, TFL1 and TSF), grapevine [VvFT and VvMFT, accession No. DQ871594; VvTFL1A (VvTFL1) and VvTFL1B, accession No. DQ871592; VvTFL1C, accession No. DQ871593], Lombardy poplar (PnFT1 and PnFT3, accession No. AB110612; PnFTL1, accession No. AB369067; PnFTL4, accession No. AB181241), morning glory (InFT1, accession No. ABW73562), orange (CsTFL, accession No. AY344244), pea (PsTFL1a and PsTFL1c, accession No. AY343326), Satsuma mandarin (CiFT1, accession No. AB027456), snapdragon [CENTRORADIALIS (CEN), accession No. S81193) and tomato (SP, SFT and SP9D, accession No. AY186738). The protein sequence data were obtained from the DNA database (DDBJ/EMBL/GenBank). The N–J unrooted dendrograms were generated from the alignment of deduced amino acids with the Clustal X program, and the phylogenetic tree was displayed using the N–J plot unrooted program (Perrière and Gouy, 1996). Bootstrap values for 1,000 resamplings are shown in each branch. The unit for the scale bar displays branch lengths (0.02 substitutions/site). Sequence analysis using the deduced amino acid sequences of TFL1/FT homologs revealed that MdFT1 and MdFT2 were the most similar to each other (94.3% identity), and they also had a high identity with FT-like members, such as VvFT, PnFT1, PnFT3, SFT, CiFT1, TSF and FT (86.7, 85.5, 85.1, 84.5, 83.6, 76.3 and 74.6%, respectively) at the amino acid level (for accession numbers, see the legend of Fig. 1C). The alignment of the deduced amino acid sequences of the TFL1/FT family showed that there were 19 amino acid residues specific to FT-like members and 15 residues specific to TFL1/CEN-like members (Fig. 1C). In particular, both MdFT1 and MdFT2 had Tyr84 and Gln139 in the positions corresponding to Tyr85 and Gln140 of Arabidopsis FT (Fig. 1C, see asterisks). Tyr85/His88 and Gln140/Asp144 are likely to be the most critical residues for distinguishing FT and TFL1 in Arabidopsis (Hanzawa et al. 2005, Ahn et al. 2006). In order to clarify the relationships among the TFL1/FT family members, we constructed a phylogenetic tree by Neighbor–Joining (N–J) distance analysis of apple, Arabidopsis, grapevine (Vitis vinifera), morning glory (Ipomoea nil), orange (Citrus sinensis), pea (Pisum sativum), Lombardy poplar (Populus nigra), Satsuma mandarin (Citrus unshiu), snapdragon and tomato (Solanum esculentum) as shown in Fig. 1D. The phylogenetic tree was divided into four major clades, represented by BFT, FT, MFT and TFL1. As expected, MdFT2, together with MdFT1, was classified into the FT clade. FT-like members of apple were more closely related to those of deciduous woody plants, such as grapevine and poplar. Localization of MdFT1 and MdFT2 on a linkage map The Rosaceae subfamily Maloideae, which includes the genus Malus, has a higher haploid base chromosome number, x = 17, than other members of the family Rosaceae, probably due to the hypothesized polyploid origin (Sax 1933, Chevreau et al. 1985). Since the DNA blot and sequence analysis suggested that MdFT1 and MdFT2 were not allelic (Fig. 1A), we attempted to show clearly on which linkage group MdFT1 and MdFT2 were located and to consider the evolutionary origin of those genes. Therefore, genetic mapping was performed for the genes using two kinds of F1 mapping populations from crosses between ‘Delicious’ (De) and Mitsubakaido (Malus sieboldii Rehder) and between ‘Ralls Janet’ (Ra) and Mitsubakaido (Igarashi et al. 2008). As a result, MdFT1 was assigned to the south end of linkage group 12 (LG 12) of ‘Delicious’ (De12), with 70.7 cM from the markers rD05A20–0.9 and E35M48–350 at the north end, whereas MdFT2 was assigned to the locus close to the marker E38M61–395 at the south end of LG 4 of ‘Ralls Janet’ (Ra04) (Fig. 2). Fig. 2 View large Download slide Location of the FT-like genes on the linkage map of the apple cultivars ‘Ralls Janet’ (Ra) and ‘Delicious’ (De). MdFT1 and MdFT2 were located on LG 12 (De12) and LG 4 (Ra04), respectively. Linkage groups (LGs) of ‘Ralls Janet’ (Ra04) and ‘Delicious’ (De12) are derived from segregation data of ‘Ralls Janet’ and ‘Delicious’ populations, respectively. LG numbers are according to Maliepaard et al. (1998). Loci are listed on the right side of the linkage maps. The left side shows genetic distances (cM). The linkage phase information is provided as + or −, indicating on which of the homologous chromosomes the marker/allele is located. In the case of SSR markers, the sizes of SSR bands are shown by three-figure numbers or the relative sizes of the SSR bands are shown by the letter ‘L (long)’ or ‘S (short)’. CAPS markers are shown with the gene name, the restriction enzyme (Dr, DraI; Hf, HinfI) and the length of the scored fragment (kbp). For a detailed explanation of molecular markers, see Supplementary Table S2 of this paper and the materials and methods of Igarashi et al. (2008). Fig. 2 View large Download slide Location of the FT-like genes on the linkage map of the apple cultivars ‘Ralls Janet’ (Ra) and ‘Delicious’ (De). MdFT1 and MdFT2 were located on LG 12 (De12) and LG 4 (Ra04), respectively. Linkage groups (LGs) of ‘Ralls Janet’ (Ra04) and ‘Delicious’ (De12) are derived from segregation data of ‘Ralls Janet’ and ‘Delicious’ populations, respectively. LG numbers are according to Maliepaard et al. (1998). Loci are listed on the right side of the linkage maps. The left side shows genetic distances (cM). The linkage phase information is provided as + or −, indicating on which of the homologous chromosomes the marker/allele is located. In the case of SSR markers, the sizes of SSR bands are shown by three-figure numbers or the relative sizes of the SSR bands are shown by the letter ‘L (long)’ or ‘S (short)’. CAPS markers are shown with the gene name, the restriction enzyme (Dr, DraI; Hf, HinfI) and the length of the scored fragment (kbp). For a detailed explanation of molecular markers, see Supplementary Table S2 of this paper and the materials and methods of Igarashi et al. (2008). Expression patterns of MdFT1 and MdFT2 in apple To identify when and where MdFT1 and MdFT2 were expressed, quantitative real-time reverse transcription–PCR (qRT–PCR) was performed on various tissues of apple in the adult phase, juvenile phase or tissue culture using primers specific to each gene (Supplementary Table S1). The transcript of MdFT1 accumulated mainly in apical buds of fruit-bearing shoots (FBS) (AB1, on June 24; AB2, on August 7), flower buds (at the balloon stage), floral organs, such as stamens, and whole young fruits. The level of MdFT1 transcript was very low in the tissues of 1-month-old seedlings in the juvenile phase, such as roots, stems, mature leaves and apical buds of vegetative shoots, with little detection in seeds and cultured shoots, which included apical buds, stems and leaves. The level of MdFT1 transcript in the mature leaves was relatively low in both the juvenile and the adult phase as compared with that detected in the apical buds and reproductive organs in the adult phase, although the expression of MdFT1 in mature leaves was higher in the adult phase than in the juvenile phase. The MdFT1 transcript level in mature fruit and peel was below the detection limit (Fig. 3A). On the other hand, the transcript of MdFT2 was detected mainly in reproductive organs, such as flower buds, sepals, petals, stamens, carpels, receptacles, peduncles and whole young fruits, with some expression also detected in mature fruit (Fig. 3B). In apical buds of FBS, we also detected the expression of MdFT2, but the expression level of MdFT2 was relatively lower than that of MdFT1. In the tissues of 1-month-old seedlings, MdFT2 was expressed at relatively higher levels than MdFT1; however, the transcript of MdFT2 in mature leaves was low in both the juvenile and the adult phases, similarly to MdFT1. We could not detect the expression of MdFT2 in seeds and peel (Fig. 3B). The transcripts of MdFT2 accumulated in shoots of cultured tissues more than those of MdFT1, although the transcription level of both was much lower than that in the apical buds of FBS. Fig. 3 View large Download slide Expression patterns of MdFT1 (A) and MdFT2 (B) in various tissues, and seasonal expression patterns of MdFT1 (C), MdFT2 (D), MdTFL1 (E) and MdAP1 (F) in apple by quantitative real-time RT–PCR. The samples for (A) and (B) from left to right are as follows: seeds (Sd), roots (Rt), stems (Sm), mature leaves (MLJ) and apical buds (ABJ) of 1-month-old seedlings (from a cross of cv. ‘Fuji’ × cv. ‘Orin’) in the juvenile phase; mature leaves (ML, collected on June 24) in the adult phase; apical buds of FBS on June 24 (AB1) and on August 7 (AB2) in the adult vegetative/reproductive phase; flower buds at the balloon stage (FB), sepals (Se), petals (Pe), stamens (St), carpels (Ca), receptacles (Re), peduncles (Pd), young fruit on June 15 (YF), mature fruit on November 18 (MF) and peel (Pl) from ‘Fuji’ apple in the adult reproductive phase; and shoots of tissue culture (CS). The samples for (C) to (F) from left to right are as follows: apical buds of fruit-bearing shoots (FBS) on June 6, June 27, July 24, August 22, September 19, October 24, November 7, December 19, January 23, Feburuary 20, March 19 and April 7; or succulent shoots (SS) on June 6, June 27, July 24, August 22, September 19 and October 24. Levels of detected amplicons were normalized by reference to amplified products that corresponded to apple HistoneH3. Values are means ± SD of results from three replicates. Each symbol without a bar indicates that the standard deviation fell within the symbol. The primer sets used in quantitative real-time RT–PCR analysis and the PCR conditions are described in Supplementary Table S1. Fig. 3 View large Download slide Expression patterns of MdFT1 (A) and MdFT2 (B) in various tissues, and seasonal expression patterns of MdFT1 (C), MdFT2 (D), MdTFL1 (E) and MdAP1 (F) in apple by quantitative real-time RT–PCR. The samples for (A) and (B) from left to right are as follows: seeds (Sd), roots (Rt), stems (Sm), mature leaves (MLJ) and apical buds (ABJ) of 1-month-old seedlings (from a cross of cv. ‘Fuji’ × cv. ‘Orin’) in the juvenile phase; mature leaves (ML, collected on June 24) in the adult phase; apical buds of FBS on June 24 (AB1) and on August 7 (AB2) in the adult vegetative/reproductive phase; flower buds at the balloon stage (FB), sepals (Se), petals (Pe), stamens (St), carpels (Ca), receptacles (Re), peduncles (Pd), young fruit on June 15 (YF), mature fruit on November 18 (MF) and peel (Pl) from ‘Fuji’ apple in the adult reproductive phase; and shoots of tissue culture (CS). The samples for (C) to (F) from left to right are as follows: apical buds of fruit-bearing shoots (FBS) on June 6, June 27, July 24, August 22, September 19, October 24, November 7, December 19, January 23, Feburuary 20, March 19 and April 7; or succulent shoots (SS) on June 6, June 27, July 24, August 22, September 19 and October 24. Levels of detected amplicons were normalized by reference to amplified products that corresponded to apple HistoneH3. Values are means ± SD of results from three replicates. Each symbol without a bar indicates that the standard deviation fell within the symbol. The primer sets used in quantitative real-time RT–PCR analysis and the PCR conditions are described in Supplementary Table S1. Seasonal expression pattern in apical buds of apple in the adult phase showed that MdFT1 was expressed to higher levels than MdFT2 in FBS from late June to August. In addition, MdFT1 was expressed relatively more highly in FBS than in succulent shoots (SS), which are vegetative shoots that do not produce flower buds, while the expression level of MdFT2 was similar between FBS and SS, from early June to mid August (Fig. 3C, D). The expression of MdFT2 peaked in mid September and thereafter decreased during the dormant period with a rapid increase from late March (Fig. 3D). On the other hand, the transcript of MdTFL1 decreased rapidly toward mid July before the peak of MdFT1 expression in apical buds of FBS, with the signal in those of SS showing a delayed decrease at late September (Fig. 3E). For MdAP1, it started to increase from late August after the period of flower induction, consistent with the onset of the development of floral organ primordia (Fig. 3F). Ectopic expressions of the apple FT-like genes in Arabidopsis To determine the effects of apple FT-like genes on flowering time and inflorescence morphology, we generated two kinds of transgenic Arabidopsis plants with MdFT1 or MdFT2 under the control of the cauliflower mosaic virus (CaMV) 35S promoter fused with the Ω sequence (35SΩ) (Fig. 4A). Consequently, we obtained >20 independent transgenic lines for 35SΩ : MdFT1 and 35SΩ:MdFT2 (designated 35SΩ : MdFT1/wt and 35SΩ : MdFT2/wt, respectively). Among them, 10 independent lines per construct were used for further analysis. Eight to 10 plants in the T2 generation per line were grown under long-day (LD, 16 h light/8 h dark) conditions, and their phenotypes were examined. Fig. 4 View largeDownload slide Phenotypes of transgenic Arabidopsis with apple FT-like genes under long-day conditions. (A) Schematic representation of the transformation vectors, 35SΩ:MdFT1 and 35SΩ:MdFT2. LB, left border; nptII, neomycin phosphotransferase II gene; P35S, promoter region of cauliflower mosaic virus 35S; Pnos, nos promoter; RB, right border; Tnos, 3′ region of nos. (B) Appearance of wild-type Arabidopsis (ecotype Columbia, left), transgenic Arabidopsis with 35SΩ:MdFT1 (middle) and transgenic Arabidopsis with 35SΩ:MdFT2 (right) 26 d after transfer to the growth chamber. (C) Inflorescence and flower of wild-type Arabidopsis (left), transgenic Arabidopsis with 35SΩ:MdFT1 (middle) and transgenic Arabidopsis with 35SΩ:MdFT2 (right). The photographs were taken after flowering for each plant. For details on the flowering time of transgenic lines, see Table 1. The wild type (Col) was used for Arabidopsis transformation. The plants in (B) and (C) were grown under LD conditions (16 h light/8 h dark) at 22°C. Fig. 4 View largeDownload slide Phenotypes of transgenic Arabidopsis with apple FT-like genes under long-day conditions. (A) Schematic representation of the transformation vectors, 35SΩ:MdFT1 and 35SΩ:MdFT2. LB, left border; nptII, neomycin phosphotransferase II gene; P35S, promoter region of cauliflower mosaic virus 35S; Pnos, nos promoter; RB, right border; Tnos, 3′ region of nos. (B) Appearance of wild-type Arabidopsis (ecotype Columbia, left), transgenic Arabidopsis with 35SΩ:MdFT1 (middle) and transgenic Arabidopsis with 35SΩ:MdFT2 (right) 26 d after transfer to the growth chamber. (C) Inflorescence and flower of wild-type Arabidopsis (left), transgenic Arabidopsis with 35SΩ:MdFT1 (middle) and transgenic Arabidopsis with 35SΩ:MdFT2 (right). The photographs were taken after flowering for each plant. For details on the flowering time of transgenic lines, see Table 1. The wild type (Col) was used for Arabidopsis transformation. The plants in (B) and (C) were grown under LD conditions (16 h light/8 h dark) at 22°C. Seven of 10 transgenic Arabidopsis lines with 35SΩ : MdFT1 flowered significantly earlier than wild-type plants under LD conditions at a P-value of <0.01 for the number of rosette leaves (Fig. 4B, Table 1). For example, 35SΩ : MdFT1/wt (#7) flowered with 3.9 ± 0.2 rosette and 2.1 ± 0.1 cauline leaves, while wild-type plants flowered with 6.7 ± 0.3 rosette and 3.6 ± 0.2 cauline leaves (Table 1). Nine out of 10 transgenic lines with 35SΩ : MdFT2 also flowered significantly earlier than wild-type plants under LD conditions at a P-value of <0.01 for the number of rosette leaves (Fig. 4B, Table 1). For example, 35SΩ : MdFT2/wt (#3) exhibiting a strong phenotype flowered with 4.0 ± 0.2 rosette and 1.9 ± 0.2 cauline leaves, whereas wild-type plants flowered with 6.7 ± 0.3 rosette and 3.6 ± 0.2 cauline leaves (Table 1). No difference in the appearance of flowers and inflorescences was observed among 35SΩ : MdFT1/wt, 35SΩ : MdFT2/wt and wild-type plants (Fig. 4C). An early flowering phenotype was also observed in 35SΩ : MdFT1/wt and 35SΩ : MdFT2/wt under short-day conditions (data not shown). Table 1 Flowering time of transgenic lines ectopically expressing MdFT1 or MdFT2 Line  LD conditions (16 h light/8 h dark)   No. of plants  Rosette leaves  Cauline leaves  Total leaves  Wt (Col)  10  6.7 ± 0.3  3.6 ± 0.2  10.3 ± 0.4  Vector/Wt (Col)  8  7.3 ± 0.3  3.3 ± 0.2  10.5 ± 0.2  35SΩ:MdFT1/wt          #1  9  4.0 ± 0.3**  2.6 ± 0.2**  6.6 ± 0.4**  #2  10  3.9 ± 0.2**  2.6 ± 0.2**  6.5 ± 0.2**  #3  10  6.2 ± 0.2  3.1 ± 0.1*  9.3 ± 0.2*  #4  10  4.0 ± 0.2**  2.2 ± 0.1**  6.2 ± 0.2**  #5  10  4.3 ± 0.2**  2.1 ± 0.1**  6.4 ± 0.2**  #6  10  6.7 ± 0.3  2.6 ± 0.2**  9.3 ± 0.4  #7  10  3.9 ± 0.2**  2.1 ± 0.1**  6.0 ± 0.2**  #8  9  7.9 ± 0.5  2.3 ± 0.2**  10.2 ± 0.4  #9  10  4.0 ± 0.2**  2.8 ± 0.2**  6.8 ± 0.3**  #10  10  4.3 ± 0.3**  2.3 ± 0.1**  6.6 ± 0.4**  35SΩ:MdFT2/wt          #1  9  4.9 ± 0.3**  2.1 ± 0.1**  7.0 ± 0.3**  #2  10  5.0 ± 0.3**  3.0 ± 0.2*  8.0 ± 0.2**  #3  10  4.0 ± 0.2**  1.9 ± 0.2**  5.9 ± 0.3**  #4  10  4.5 ± 0.3**  2.3 ± 0.1**  6.8 ± 0.3**  #5  10  4.3 ± 0.2**  2.2 ± 0.1**  6.5 ± 0.2**  #6  10  4.1 ± 0.2**  3.3 ± 0.2  7.4 ± 0.4**  #7  9  5.3 ± 0.3*  2.3 ± 0.2**  7.7 ± 0.4**  #8  8  4.4 ± 0.4**  2.8 ± 0.3*  7.1 ± 0.3**  #9  10  4.9 ± 0.1**  2.1 ± 0.1**  7.0 ± 0.0**  #10  10  4.3 ± 0.1**  2.0 ± 0.0**  6.3 ± 0.1**  Line  LD conditions (16 h light/8 h dark)   No. of plants  Rosette leaves  Cauline leaves  Total leaves  Wt (Col)  10  6.7 ± 0.3  3.6 ± 0.2  10.3 ± 0.4  Vector/Wt (Col)  8  7.3 ± 0.3  3.3 ± 0.2  10.5 ± 0.2  35SΩ:MdFT1/wt          #1  9  4.0 ± 0.3**  2.6 ± 0.2**  6.6 ± 0.4**  #2  10  3.9 ± 0.2**  2.6 ± 0.2**  6.5 ± 0.2**  #3  10  6.2 ± 0.2  3.1 ± 0.1*  9.3 ± 0.2*  #4  10  4.0 ± 0.2**  2.2 ± 0.1**  6.2 ± 0.2**  #5  10  4.3 ± 0.2**  2.1 ± 0.1**  6.4 ± 0.2**  #6  10  6.7 ± 0.3  2.6 ± 0.2**  9.3 ± 0.4  #7  10  3.9 ± 0.2**  2.1 ± 0.1**  6.0 ± 0.2**  #8  9  7.9 ± 0.5  2.3 ± 0.2**  10.2 ± 0.4  #9  10  4.0 ± 0.2**  2.8 ± 0.2**  6.8 ± 0.3**  #10  10  4.3 ± 0.3**  2.3 ± 0.1**  6.6 ± 0.4**  35SΩ:MdFT2/wt          #1  9  4.9 ± 0.3**  2.1 ± 0.1**  7.0 ± 0.3**  #2  10  5.0 ± 0.3**  3.0 ± 0.2*  8.0 ± 0.2**  #3  10  4.0 ± 0.2**  1.9 ± 0.2**  5.9 ± 0.3**  #4  10  4.5 ± 0.3**  2.3 ± 0.1**  6.8 ± 0.3**  #5  10  4.3 ± 0.2**  2.2 ± 0.1**  6.5 ± 0.2**  #6  10  4.1 ± 0.2**  3.3 ± 0.2  7.4 ± 0.4**  #7  9  5.3 ± 0.3*  2.3 ± 0.2**  7.7 ± 0.4**  #8  8  4.4 ± 0.4**  2.8 ± 0.3*  7.1 ± 0.3**  #9  10  4.9 ± 0.1**  2.1 ± 0.1**  7.0 ± 0.0**  #10  10  4.3 ± 0.1**  2.0 ± 0.0**  6.3 ± 0.1**  Plants in the second generation (T2) were grown under long-day (LD) conditions. Numbers of rosette and cauline leaves were counted on the day floral organs became visible. Values are means ± SE. Student’s t-test was performed to compare the effects of treatments. One or two asterisks indicate a statistically significant difference from the wild-type plants in the same column (P <0.05, 0.01, respectively). View Large Overexpression of MdFT1 in apple Transgenic experiments using Arabidopsis revealed that both MdFT1 and MdFT2 had the potential to function as floral promoters in apple. Expression analysis by qRT–PCR suggested that MdFT1 played a role during floral transition, at least in tissues such as apical buds. Therefore, to clarify whether MdFT1 confers an early-flowering phenotype in apple, we produced transgenic apples carrying the 35SΩ : MdFT1 already used for Arabidopsis transformation, using micropropagated tissues of an apple rootstock cv. ‘JM2’. As a result, we obtained six independent transgenic lines designated MdFT1#1-1, MdFT1#1-2, MdFT1#3, MdFT1#5, MdFT1#6, and MdFT1#8. Five of six transgenic lines (MdFT1#1-1, MdFT1#1-2, MdFT1#3, MdFT1#6 and MdFT1#8) produced flower buds or opened flowers in vitro 8–12 months after Agrobacterium infection (2–6 months after regeneration). For example, MdFT1#3 produced terminal flower buds at the top of each shoot, and some of them came into flower (Fig. 5A, B), whereas the control plant never produced flower buds under the same culture conditions (Fig. 5C). In addition, the shoots of MdFT1#1-1 were grafted onto the apple rootstock ‘JM7’ and transferred to the greenhouse (Fig. 5D). The potted MdFT1#1-1 also produced a flower bud and opened a solitary flower; however, the flower had >8 petals (Fig. 5E), in contrast to the normal flower, which had five sepals, petals, carpels and approximately 20 stamens (Fig. 5F). Importantly, the in vitro flowering lines, such as MdFT1#3 and MdFT1#8, which produced in vitro flower buds more frequently than the others, showed a much less vigorous growth of shoots. Several lines with such severe phenotypes senesced, probably due to their inability to form new vegetative shoot meristems. The result of DNA blot analysis for transgenic lines MdFT1#1-1, MdFT1#5 and MdFT1#8 indicated that MdFT#1-1 had two copies and the other lines had one copy of the transgene (Fig. 5G). Fig. 5 View largeDownload slide Precocious flowering of transgenic apples overexpressing MdFT1. (A) Floral buds (indicated as an open triangle) formed at the top of the shoots of transgenic apple ‘JM2’ (line MdFT1#3) in the culture box. (B) In vitro flowering of a solitary flower (indicated as a filled triangle) at the top of the transgenic shoots in the culture box. (C) Cultured shoots of an apple ‘JM2’ control plant. A white bar indicates 1 cm. (D) Opened flower of transgenic apple ‘JM2’ (line MdFT1#1-1) in the greenhouse. A white bar indicates 3 cm. (E) Close-up view of (D). The number of petals increased compared with a normal flower with five petals. (F) Flower of an apple cv. ‘JM2’ control plant grown in the field. (G) DNA blot analysis of transgenic apples ‘JM2’ with 35SΩ:MdFT1 (MdFT1#1-1, MdFT1#5 and MdFT1#8). Genomic DNA (15 μg) was isolated from the cultured tissues of each transgenic line and a control plant, digested with HindIII, and then separated on a 0.8% (w/v) agarose gel. A blotted membrane was hybridized with a DIG-labeled nptII gene. Fig. 5 View largeDownload slide Precocious flowering of transgenic apples overexpressing MdFT1. (A) Floral buds (indicated as an open triangle) formed at the top of the shoots of transgenic apple ‘JM2’ (line MdFT1#3) in the culture box. (B) In vitro flowering of a solitary flower (indicated as a filled triangle) at the top of the transgenic shoots in the culture box. (C) Cultured shoots of an apple ‘JM2’ control plant. A white bar indicates 1 cm. (D) Opened flower of transgenic apple ‘JM2’ (line MdFT1#1-1) in the greenhouse. A white bar indicates 3 cm. (E) Close-up view of (D). The number of petals increased compared with a normal flower with five petals. (F) Flower of an apple cv. ‘JM2’ control plant grown in the field. (G) DNA blot analysis of transgenic apples ‘JM2’ with 35SΩ:MdFT1 (MdFT1#1-1, MdFT1#5 and MdFT1#8). Genomic DNA (15 μg) was isolated from the cultured tissues of each transgenic line and a control plant, digested with HindIII, and then separated on a 0.8% (w/v) agarose gel. A blotted membrane was hybridized with a DIG-labeled nptII gene. Influence of overexpression of MdFT1 on the other apple genes We examined the influence of the overexpression of MdFT1 on the expression of other endogenous genes, which had been shown or expected to play a role in flowering in apple. To avoid the fluctuation of the gene expression by artificial and other environmental factors, we used the cultured shoots of each line grown in a growth chamber under the same conditions. The result of qRT–PCR analysis showed that MdMADS12 [a FRUITFULL (FUL)-like gene of apple; van der Linden et al. 2002] was significantly upregulated in the in vitro flowering lines MdFT1#1-1 and MdFT1#8, where the MdFT1 transcripts accumulated in large numbers, whereas the expression level of MdMADS12 in a line not flowering in vitro, MdFT1#5, was almost the same as that in the control (Fig. 6A, B). In addition, MdSOC1a and MdSOC1b (putative apple orthologs of SOC1; accession Nos. AB501124 and AB501125, respectively), and MdAP1 (equivalent to MdMADS5, an apple ortholog of AP1) were more up-regulated in the lines MdFT1#1-1 and MdFT1#8, than in line MdFT1#5 and the non-transgenic control (Fig. 6C–E). The expression of MdLHP1a/MdLHP1b (MdLHP1a and MdLHP1b, apple homologs of LHP1/TFL2; Mimida et al. 2007) was not affected by the overexpression of MdFT1 (Fig. 6F). Fig. 6 View large Download slide Expression analysis for apple genes in the shoots of transgenic lines. (A–F) Expression of MdFT1 (A), MdMADS12 (B), MdSOC1a (C), MdSOC1b (D), MdAP1 (E) and MdLHP1a/MdLHP1b (F) in the transgenic shoots of in vitro flowering lines (MdFT1#1-1 and MdFT1#8) and a line not flowering in vitro (MdFT1#5), and in non-transgenic shoots as a control. Shoots were collected from each line cultured in a proliferating medium under LD conditions for 4 weeks to synchronize the growth state of each line. MdMADS12, MdSOC1a, MdSOC1b and MdAP1 are genes encoding a MADS-box transcription factor from apple; and MdLHP1a/MdLHP1b (MdLHP1a and MdLHP1b) are apple twin homologs of LHP1. Primers for MdLHP1a/MdLHP1b were designed to detect both MdLHP1a and MdLHP1b. HistoneH3 was used as an internal control of gene expression. In A–F, non-transgenic apple cv. ‘JM2’ was used as a control plant (control). Transcript levels were normalized against apple HistoneH3. Values are means ± SD of results from three replicates, and each box without a bar indicates that the standard deviation fell within the box. The primer sets used in quantitative real-time RT–PCR analysis and the PCR conditions are described in Supplementary Table S1. Fig. 6 View large Download slide Expression analysis for apple genes in the shoots of transgenic lines. (A–F) Expression of MdFT1 (A), MdMADS12 (B), MdSOC1a (C), MdSOC1b (D), MdAP1 (E) and MdLHP1a/MdLHP1b (F) in the transgenic shoots of in vitro flowering lines (MdFT1#1-1 and MdFT1#8) and a line not flowering in vitro (MdFT1#5), and in non-transgenic shoots as a control. Shoots were collected from each line cultured in a proliferating medium under LD conditions for 4 weeks to synchronize the growth state of each line. MdMADS12, MdSOC1a, MdSOC1b and MdAP1 are genes encoding a MADS-box transcription factor from apple; and MdLHP1a/MdLHP1b (MdLHP1a and MdLHP1b) are apple twin homologs of LHP1. Primers for MdLHP1a/MdLHP1b were designed to detect both MdLHP1a and MdLHP1b. HistoneH3 was used as an internal control of gene expression. In A–F, non-transgenic apple cv. ‘JM2’ was used as a control plant (control). Transcript levels were normalized against apple HistoneH3. Values are means ± SD of results from three replicates, and each box without a bar indicates that the standard deviation fell within the box. The primer sets used in quantitative real-time RT–PCR analysis and the PCR conditions are described in Supplementary Table S1. Discussion MdFT1 and MdFT2 are located in distinct linkage groups with duplicated segments MdFT1 and MdFT2 were mapped, respectively, on loci in distinct linkage groups, LG 12 and LG 4 (Fig. 3). Because there is a homoeology between the distal parts of LG 12 and LG 4 (Gardiner et al. 2007), MdFT1 and MdFT2 might have been derived from an ancient gene that ancestors of the original hybrid would have had. Coincidentally, MdFT1 was located on the distal part of LG 12, the same linkage group as MdTFL1. This finding suggests that MdFT1 and MdTFL1 have evolved in close relation to each other and that at least one chromosome, corresponding to LG 12, plays an important role in the regulation of flowering in apple. To date, it has been revealed that apple has at least six members of the TFL1/FT family in its genome (Kotoda et al. 2005, Mimida et al. 2009, this study), although members that belong to the BFT or MFT clade have not been reported. Other eudicots, such as Arabidopsis and tomato, have about six TFL1/FT family genes (Mimida et al. 2001, Carmel-Goren et al. 2003). However, the Lombardy poplar genome has nine (Igasaki et al. 2008), probably due to a recent gene duplication event 8–13 million years ago (Tuskan et al. 2006). Because apple has duplicated chromosomal regions as well, further study might reveal additional members falling into the clade of MFT or BFT in the TFL1/FT family in apple. MdFT1 and MdFT2 show different expression profiles in apple The expression pattern of MdFT1 and MdFT2 differed in that MdFT1 was expressed mainly in apical buds of FBS in the adult phase, whereas MdFT2 was expressed mainly in reproductive organs, including flower buds and young fruit (Fig. 3A, B). In addition, it was found that the transcripts of MdFT1 could hardly be detected in the tissues of 1-month-old seedlings in the juvenile phase, in contrast to the transcripts of MdFT2. In Satsuma mandarin, two citrus homologs of FT, CiFT1 and CiFT2, which are highly and specifically expressed in young fruit, were not involved in the transition to flowering, and only the third citrus homolog of FT, CiFT3, expressed at a relatively low level in leaves and stems but with no expression in reproductive organs, turned out to play a key role in regulating flower induction (Nishikawa et al. 2007). Considering that MdFT2 was highly expressed in the reproductive organs and that it was expressed similarly in the apical buds of both FBS and SS during the period of flower induction, MdFT1 might play a key role in the transition to flowering in apple (Fig. 3), although Arabidopsis FT and its paralog TSF are both highly expressed in flowers and developing siliques as well as in leaves (Kobayashi et al. 1999, Takada and Goto, 2003, Yamaguchi et al. 2005). The expression pattern of MdFT1 in various tissues was somewhat similar to that of tomato SFT (formerly SPD3, a tomato ortholog of FT; Carmel-Goren et al. 2003, Lifshitz et al. 2006) rather than that of Arabidopsis. In apple, flower induction occurs in late June [about 40–50 days after flowering (DAF)] and flower initiation with morphological changes at the apical meristem starts to form floral primordia in mid July (about 60 DAF) in Morioka, Japan (Kotoda et al. 2000). MdFT1 was expressed to relatively higher levels in the apical buds of FBS than in those of SS from the period of flower induction to the early stage of flower development. This result was consistent with the observation that the expression of MdTFL1 decreased rapidly from early June, although it is unclear where in apical buds MdFT1 was expressed (Fig. 3C, E). Recently, Hättasch et al. (2008) reported that the increase in transcription of MdFT (MdFT1 in this study), AFL1 and AFL2 (apple orthologs of LFY; Wada et al. 2002) in the apical meristems of current-year shoots of apple began 1–2 weeks earlier than flower initiation, suggesting that flower induction is initiated by these genes. The higher expression level of MdFT2 in mid September and early April might be related to the development of floral organs in apple (Fig. 3D), because the expression pattern of MdAP1 reflected the development of floral organs, such as sepals and receptacles (Fig. 3F, Kotoda et al. 2000). MdFT1 and MdFT2 have the potential to act as floral promoters and overexpression of MdFT1 confers precocious flowering in apple Transgenic Arabidopsis overexpressing MdFT1 and MdFT2 showed early flowering as compared with the control plant (Fig. 4, Table 1). Similar early-flowering phenotypes of transgenic Arabidopsis had been reported upon the ectopic expression of FT-like genes from heterologous plant species, such as citrus (Kobayashi et al. 1999), tomato (Teper-Bamnolker and Samach 2005, Lifschitz et al. 2006), grapevine (Sreekantan and Thomas 2006, Carmona et al. 2007), poplar (Hsu et al. 2006, Igasaki et al. 2008) and cucurbits (Lin et al. 2007), as well as the overexpression of FT and TSF. No differences in the extent of flower-promoting activity between MdFT1 and MdFT2 in transgenic Arabidopsis suggested that MdFT1 and MdFT2 both have the potential to act as floral promoters in apple (Table 1). However, expression analysis by qRT–PCR implied that, during floral transition, MdFT1 plays a more important role, at least in tissues within the apical buds of FBS, where transition to flowering occurs. Consequently, five out of six transgenic apples overexpressing MdFT1 flowered in vitro 8–12 months after Agrobacterium infection (Fig. 5). As a whole, the precocity of flowering in MdFT1-overexpressing lines was much higher than that in previously reported MdTFL1 down-regulated apples (Kotoda et al. 2003, Kotoda et al. 2006). The transgenic lines with 35SΩ : MdFT1 that produced flower buds in vitro had a tendency to have small, rounded leaves and show an extremely weak growth habit. In the transgenic lines MdFT1#1-1 and MdFT1#8, which showed in vitro flowering, MdFT1 was abundantly expressed, while the expression of MdFT1 was extremely low in line MdFT1#5, possibly due to co-suppression (Fig. 6A). As expected, line MdFT#5 has not flowered in vitro or in pots for at least 3 years after regeneration. These results suggest that there is a positive correlation between MdFT1 expression and flower induction in transgenic plants. Besides MdFT1, the heterologous gene PnFT3 (a poplar ortholog of FT; Igasaki et al. 2008) also promoted the transition to flowering when overexpressed in the apple cultivar ‘Greensleeves’ (Supplementary Fig. S2). Apple trees normally produce five flowers per cluster, and each flower has five sepals, petals and carpels and approximately 20 stamens. The changes in the number of floral organs in the transgenic plant might result from disturbance of the expression of floral organ identity genes, such as MADS-box genes, due to the overexpression of MdFT1 (Fig. 5E). Contrary to our results, Hättasch et al. (2009) reported that transgenic apples with MdFT did not flower for at least a year in vitro and in the greenhouse. This observation might be due to the fact that the regeneration competence of the transformed calli from some scion cultivars was severely impaired when FT-like genes were overexpressed, resulting in the selection of transgenic lines with little expression of those genes. Up-regulation of MADS-box genes could result in in vitro flowering in the transgenic lines overexpressing MdFT1 QRT–PCR analysis revealed that MdMADS12 and MdAP1 were significantly more up-regulated in the cultured shoots of transgenic lines that highly expressed MdFT1 than in those of line MdFT1#5, which had little expression of MdFT1, and a control (Fig. 6). These results seem to be consistent with the finding by Teper-Bamnolker and Samach (2005) that FUL and AP1 were highly misexpressed in the young seedlings and the older rosette leaves of Arabidopsis overexpressing FT. Interestingly, Flachowsky et al. (2007) reported that the overexpression of BpMADS4, a birch (Betula pendula) FUL-like MADS-box gene (Elo et al. 2001), induced early flowering in the apple cv. ‘Pinova’, with frequent in vitro flowering. In addition, the phenotype of the transgenic apple with 35S : BpMADS4 resembled that with 35SΩ : MdFT1 in that it produced solitary flower buds and rounded leaves in tissue culture. Considering that MdMADS12 is a putative ortholog of FUL, MdMADS12 might play an important role in flower induction of apple downstream of MdFT1. In Arabidopsis, FUL is required in several developmental processes, including silique and leaf development, and in the transition to flowering (Gu et al. 1998, Ferrandiz et al. 2000, Teper-Bamnolker and Samach, 2005). On the other hand, the up-regulation of MdAP1 in the transgenic lines indicated that the shoots had been florally induced because the expression of MdAP1 starts to increase in the apical buds of FBS after flower initiation (Figs. 3F and 6E; Kotoda et al. 2000). In addition, MdSOC1a and MdSOC1b were also up-regulated in the in vitro flowering lines with a similar expression pattern to MdAP1 (Fig. 6C–E), implicating them as the common targets of MdFT1, as SOC1 and AP1 are activated by the FT–FD complex in Arabidopsis (Abe et al. 2005, Wigge et al. 2005, Michaels 2009). Our results suggest that MdFT1 could function upstream of those genes to regulate flowering in apple, although whether or not the regulation is direct remains to be demonstrated. In future studies, the interaction of partner genes, such as transcription factors, with MdFT1 and/or MdFT2 will be investigated. Materials and Methods Plant materials The tissue samples of an apple (Malus × domestica Borkh.) cvs. ‘Fuji’ and ‘Jonathan’ in adult phase (age: 19–20 years) were collected from the experimental field at the National Institute of Fruit Tree Science in Morioka, Japan. One-month-old juvenile seedlings from a cross between apple cvs. ‘Fuji’ and ‘Orin’ were used for the expression analysis of apple FT-like genes in the juvenile phase. Micropropagated tissues of an apple rootstock cv. ‘JM2’ [Marubakaido ‘Seishi’ (Malus prunifolia Borkh. var. ringo Asami) × ‘Malling 9’ (‘M. 9’)] (Soejima et al. 1998) and an apple cv. ‘Greensleeves’ were kept at 24°C under LD conditions (16 h photoperiod; cool white fluorescent light, 50 μmol m−2 s−1) in the proliferating medium (MS medium containing B5 vitamin, 1 mg l−1 6-benzyl-aminopurine, 0.1 mg l−1 indole-3-butyric acid) in the culture box and subcultured every 4 weeks. The leaf explants of those apple cultivars were used for Agrobacterium-mediated transformation. For Arabidopsis transformation, wild-type plants of ecotype Columbia (Col) were used. Nucleic acid extraction and hybridization analysis The genomic DNA was isolated by a cetyltrimethylammonium bromide (CTAB)-based method modified by Yamamoto and Mukai as described in Kotoda et al. (2002). The genomic DNA (15 μg) was digested with EcoRI, HindIII or XbaI, electrophoresed on a 0.8% agarose gel, and then blotted onto Hybond-N+ nylon membranes (GE Healthcare Bio-Sciences Corp., Piscataway, NJ, USA). The blotted membrane was hybridized with digoxigenin- (DIG; Roche Diagnostics, Mannheim, Germany) labeled MdFT1 cDNA, which was amplified by PCR with a pair of primers, MdFT1(1→24) and MdFT1(508←528), and a DIG mixture (Roche Diagnostics). For DNA blot analysis of transformants, each genomic DNA of transgenic lines and a control plant was digested with HindIII and the blotted membrane was hybridized with a DIG-labeled nptII probe. The hybridization and washing was performed as described in Kotoda et al. (2002). Chemiluminescent signals were visualized using the LAS1000 image analyzer (Fuji Photo Film, Tokyo, Japan). The primer set is listed in Supplementary Table S1. Isolation of FT-like genes from apple The cDNA library derived from flower buds of ‘Fuji’ apple was constructed as described by Mimida et al. (2007). MdFT1 (accession No. AB161112; Kotoda and Wada 2005) labeled with [α-32P]dCTP (GE Healthcare Bio-Sciences Corp.) was used as a probe to screen 2.0 × 105 plaque-forming units (pfu) of the cDNA library. Positive plaques were excised to the pBluescript SK(−) phagemid (Stratagene, La Jolla, CA, USA), and approximately 20 clones were then sequenced using a DTCS Quick Start Kit for Dye Terminator Cycle Sequencing (Beckman Coulter, Fullerton, CA, USA) and an automated DNA sequencer CEQ 8000 (Beckman Coulter). To obtain genomic DNA of MdFT1, the cDNA of MdFT1 labeled with DIG was used as a probe to screen 2.0 × 105 pfu of the genomic library of ‘Fuji’ apple. The lambda clone DNAs from positive plaques were digested with restriction enzyme NotI and ligated into the corresponding site of pBluescriptII SK (+) (Stratagene). To obtain a MdFT2 genomic sequence from apple, on the other hand, PCR amplification of DNA fragments was performed in a mixture of a pair of primers MdFT2-EcoRI(−171→−148) and MdFT2(842←859)-XhoI, a high-fidelity DNA polymerase (KOD plus; Toyobo, Osaka, Japan) and 250–300 ng of genomic DNA of ‘Fuji’ apple. PCR was programmed for pre-heating at 94°C for 2 min followed by 30 cycles of 94°C for 30 s, 50°C for 30 s and 68°C for 2 min. The PCR-amplified fragments digested with EcoRI and XhoI were cloned into the corresponding site of pBluescriptII SK (+) (Stratagene), and then four clones were sequenced. The primer sets used in gene cloning are listed in Supplementary Table S1. Mapping of FT-like genes on the linkage map of apple Cleaved amplified polymorphic sequence (CAPS) markers for the MdFT1 and MdFT2 genes were developed, and genotyping was performed using an F1 mapping population with 72 seedlings from the cross between ‘Delicious’ and Mitsubakaido (Malus sieboldii Rehder) and another population with 83 seedlings from the cross between ‘Ralls Janet’ and Mitsubakaido (Igarashi et al. 2008). A population from ‘Delicious’ and Mitsubakaido was used for the mapping of MdFT1. PCR was performed with a pair of primers MdFT1-5f and MdFT1-4r. A 632 bp product was digested with HinfI, and the resultant 632 bp fragment specific to ‘Delicious’ was used for segregation analysis. On the other hand, a population from ‘Ralls Janet’ and Mitsubakaido was used for the mapping of MdFT2. PCR was performed with a pair of primers MdFT2-7f and MdFT2-8rm. A 1,064 bp product was digested with DraI, giving a 435 bp fragment specific to ‘Ralls Janet’, which was used for segregation analysis. The linkage map was constructed using JoinMap ver. 3.0 (Van Ooijen and Voorrips 2001). Primer sets, restriction enzymes and PCR conditions for the CAPS markers are described in Supplementary Table S2. Sequence analysis Amino acid sequences were analyzed using the Clustal X multiple sequence alignment program ver. 1.83 (Jeanmougin et al. 1998) and BioEdit ver. 7.7.0 (H. Hall, http://www.mbio .ncsu.edu/BioEdit/bioedit.html). The phylogenetic tree was displayed using the N–J plot unrooted (Perrière and Gouy, 1996) with bootstrap values for 1,000 resamplings in each branch. Expression analysis by q RT–PCR For analysis, total RNA was extracted from flower buds (at the balloon stage), sepals, petals, stamens, carpels, receptacles, peduncles, young fruit, mature fruit, peel, mature leaves (collected on June 24) and apical buds of FBS (collected on June 24 and August 7) from adult ‘Fuji’ apple trees; leaves, apical buds of vegetative shoots, stems and roots from 1-month-old juvenile seedlings and seeds from a cross between ‘Fuji’ and ‘Orin’ apples; apical buds of FBS (from June to the following April in 2004) and SS (from June to October in 2003) in ‘Jonathan’ apple trees; in vitro cultured shoots of ‘Fuji’ apple; and transgenic and non-transgenic lines of ‘JM2’ apple by using a PolyA Tract mRNA purification kit (Promega, Madison, WI, USA). For seasonal expression analysis, apical buds of all samples from the orchard were collected between 10 and 14 h. The first-strand cDNAs were synthesized from 1 μg of total RNAs in 20 μl of a reaction mixture using a QuantiTect Reverse Transcription kit (Qiagen GmbH, Hilden, Germany). The subsequent PCRs were performed with 1 μl of the first-strand cDNA as templates in a total volume of 12.5 μl by using ABI PRISM 7000 (Applied Biosystems, Foster City, CA, USA). Transcripts of MdFT1, MdFT2, MdLHP1a/MdLHP1b (MdLHP1a and MdLHP1b), MdTFL1, MdMADS12, MdSOC1a, MdSOC1b, MdAP1 and HistoneH3 were identified with the specific primers (the primer sets used in qRT–PCR analysis and the PCR conditions are described in Supplementary Table S1). An apple HistoneH3 gene (Kotoda et al. 2006) was used as an internal control. QRT–PCR was performed three times and transcript levels were normalized against apple HistoneH3. Construction of the transformation vector To construct a vector for the constitutive expression of MdFT1 and MdFT2, the coding region of each gene was amplified by PCR with thr pair of primers MdFT1-XbaI and MdFT1-SalI for MdFT1, and with thr pair of primers MdFT2-XbaI and MdFTa-SalI for MdFT2 (Supplementary Table S1). An amplified PCR product was subsequently digested with XbaI and SalI and then cloned into the XbaI/SalI sites of the modified pBI221 (pBI9526Ω) to be placed between the CaMV 35S promoter fused with the Ω sequence (Gallie and Walbot 1992) and the terminator of the nopaline synthase (nos) (Supplementary Fig. S1). The resultant plasmid was cut with ApaI, and then the fragment containing MdFT1 or MdFT2 was ligated into the same restriction enzyme site of the pSMAK312Blue binary vector (pSMAK312Blue; H. Ichikawa, in preparation). For vector construction of 35SΩ::PnFT3S, poplar PnFT3S amplified by PCR was inserted into the modified pSMAK193E (pSMAK193E; H. Ichikawa, in preparation) to be placed between 35SΩ and the 3′ region of the Arabidopsis rbcS-2B gene (TrbcS). Arabidopsis transformation Agrobacterium tumefaciens strain EHA101 was used to transform A. thaliana (Col) plants by the floral-dip method (Clough and Bent 1998). Kanamycin-resistant transformants were transplanted from the plate to moistened potting soil composed of vermiculite and perlite [1 : 1 (v/v)] after 2–5 adult leaves had developed and grown in the growth chamber (Biotron, Nippon Medical and Chemical Instruments Co., Ltd., Tokyo, Japan) set at 22°C under LD conditions (16 h photoperiod; cool white fluorescent light, 50 μmol m−2 s−1). Morphological analyses were performed on the second generation (T2). Apple transformation The apple cvs. ‘JM2’ and ‘Greensleeves’ were transformed with A. tumefaciens strain EHA101 bearing 35SΩ:MdFT1 and 35SΩ:PnFT3 (PnFT3; a poplar ortholog of FT), respectively. Agrobacterium tumefaciens was cultured overnight on a shaker in 20 ml of liquid LB medium with 100 mg l−1 trobicin (Pfizer Japan, Tokyo, Japan) at 28°C. After centrifugation, the pellet was resuspended with MS medium (Murashige and Skoog 1962) and further diluted to an optical density (OD) of 0.5–0.8 at 600 nm. Leaf explants were infected with the inoculum for 30 min. The selection of transgenic shoots for ‘Greensleeves’ was performed according to the procedure described by James et al. (1989) and Yao et al. (1995) and the procedure for ‘JM2’ will be published elsewhere (S. Takahashi, in preparation). Transformed shoots were multiplied in the proliferating medium supplemented with 50 mg l−1 kanamycin. Two to three shoots per line were grafted to apple rootstock ‘JM7’ [Marubakaido ‘Seishi’ (Malus prunifolia Borkh. var. ringo Asami) × ‘Malling 9’ (‘M. 9’)] (Soejima et al. 1998). Grafted apples were grown in an isolated greenhouse under natural day-length with the temperature set at 20–25°C during the growing season as described in Kotoda et al. (2006). Funding The Bio-oriented Technology Research Advancement Institution [Program for Promotion of Basic Research Activities for Innovative Biosciences (to N.K.)]; the Ministry of Agriculture, Forestry and Fisheries of Japan [‘Development of innovative plants and animals using transformation and cloning’ (to N.K.)]; the Ministry of Education, Sports, Science, and Technology of Japan [grant-in-aid for scientific research (to N.K.)]. Acknowledgments We thank Dr. H. Ichikawa for providing the binary vectors pSMAK312B and pSMAK193E, Dr. E. E. Hood for providing Agrobacterium tumefaciens EHA101, Ms. T. Sekita and A. Igarashi for their technical assistance, and Dr. E. Varkonyi-Gasic for critical reading of the manuscript. The nucleotide sequences reported in this paper have been submitted to the DDBJ under accession numbers AB458504 (MdFT2 mRNA), AB458505 (MdFT2 genomic DNA), AB458506 (MdFT1 genomic DNA), AB501124 (MdSOC1a mRNA), and AB501125 (MdSOC1b mRNA). References Abe M,  Kobayashi Y,  Yamamoto S,  Daimon Y,  Yamaguchi A,  Ikeda Y, et al.  FD, a bZIP protein mediating signals from the floral pathway integrator FT at the shoot apex,  Science ,  2005, vol.  309 (pg.  1052- 1056) Google Scholar CrossRef Search ADS PubMed  Ahn JH,  Miller D,  Winter VJ,  Banfield MJ,  Lee JH,  Yoo SY, et al.  A divergent external loop confers antagonistic activity on floral regulators FT and TFL1,  EMBO J. ,  2006, vol.  25 (pg.  605- 614) Google Scholar CrossRef Search ADS PubMed  Araki T.  Transition from vegetative to reproductive phase,  Curr. Opin. Plant Biol. ,  2001, vol.  4 (pg.  63- 68) Google Scholar CrossRef Search ADS PubMed  Benlloch R,  Berbel A,  Serrano-Mislata A,  Madueño F.  Floral initiation and inflorescence architechture; a comparative view,  Ann. Bot. ,  2007, vol.  100 (pg.  659- 676) Google Scholar CrossRef Search ADS PubMed  Bradley D,  Ratcliffe O,  Vincent C,  Carpenter R,  Coen E.  Inflorescence commitment and architecture in Arabidopsis,  Science ,  1997, vol.  275 (pg.  80- 83) Google Scholar CrossRef Search ADS PubMed  Carmel-Goren L,  Liu YS,  Lifschitz E,  Zamir D.  The SELF-PRUNING gene family in tomato,  Plant Mol. Biol. ,  2003, vol.  52 (pg.  1215- 1222) Google Scholar CrossRef Search ADS PubMed  Carmona MJ,  Calonje M,  Martínez-Zapater JM.  The FT/TFL1 gene family in grapevine,  Plant Mol. Biol ,  2007, vol.  63 (pg.  637- 650) Google Scholar CrossRef Search ADS PubMed  Chailahyan MK.  Internal factors of plant flowering,  Annu. Rev. Plant Physiol. ,  1968, vol.  19 (pg.  1- 37) Google Scholar CrossRef Search ADS   Chevreau E,  Lespinasse Y,  Gallet M.  Inheritance of pollen enzymes and polyploidy origin of apple (Malus × domestica Borkh.),  Theor. Appl. Genet. ,  1985, vol.  71 (pg.  268- 277) Google Scholar PubMed  Clough SJ,  Bent AF.  Floral dip: a simplified method for Agrobacterium-mediated transformation of Arabidopsis thaliana,  Plant J. ,  1998, vol.  16 (pg.  735- 743) Google Scholar CrossRef Search ADS PubMed  Corbesier L,  Vincent C,  Jang S,  Fornara F,  Fan Q,  Searle I., et al.  FT protein movement contributes to long-distance signaling in floral induction of Arabidopsis,  Science ,  2007, vol.  316 (pg.  1030- 1033) Google Scholar CrossRef Search ADS PubMed  Elo A,  Lemmetyinen J,  Turunen ML,  Tikka L,  Sopanen T.  Three MADS-box genes similar to APETALA1 and FRUITFULL from silver birch (Betula pendula),  Physiol. Plant. ,  2001, vol.  112 (pg.  95- 103) Google Scholar CrossRef Search ADS PubMed  Esumi T,  Tao R,  Yonemori K.  Isolation of LEAFY and TERMINAL FLOWER 1 homologues from six fruit tree species in the subfamily Maloideae of the family Rosaceae,  Sex. Plant Reprod. ,  2005, vol.  17 (pg.  277- 287) Google Scholar CrossRef Search ADS   Ferrándiz C,  Gu Q,  Martienssen R,  Yanofsky MF.  Redundant regulation of meristem identity and plant architecture by FRUITFULL, APETALA1 and CAULIFLOWER,  Development ,  2000, vol.  127 (pg.  725- 734) Google Scholar PubMed  Flachowsky H,  Peil A,  Sopanen T,  Elo A,  Hanke V.  Overexpression of BpMADS4 from silver birch (Betula pendula) in apple (Malus × domstica) induces early flowering,  Plant Breed. ,  2007, vol.  126 (pg.  137- 145) Google Scholar CrossRef Search ADS   Gallie DR,  Walbot V.  Identification of the motifs within the tobacco mosaic virus 5′-leader responsible for enhancing translation,  Nucleic Acids Res. ,  1992, vol.  20 (pg.  4631- 4638) Google Scholar CrossRef Search ADS PubMed  Gardiner SE,  Bus V.GM,  Rusholme RL,  Chagné D,  Rikkerink E.HA.  Kole C.  Apple,  Genome Mapping and Molecular Breeding in Plants, Vol. 4 Fruits and Nuts ,  2007 Berlin, Germany Springer-Verlag(pg.  1- 62) Giakountis A,  Coupland G.  Phloem transport of flowering signals,  Curr. Opin. Plant Biol. ,  2008, vol.  11 (pg.  687- 694) Google Scholar CrossRef Search ADS PubMed  Gu Q,  Ferrandiz C,  Yanofsky MF,  Martienssen R.  The FRUITFULL MADS-box gene mediates cell differentiation during Arabidopsis fruit development,  Development ,  1998, vol.  125 (pg.  1509- 1517) Google Scholar PubMed  Hackett WP.  Juvenility, maturation, and rejuvenility in woody plants,  Hortic. Rev. ,  1985, vol.  7 (pg.  109- 155) Hanke M.-V,  Flachowsky H,  Peil A,  Hättasch C.  No flower no fruit—Genetic potentials to trigger flowering in fruit trees,  Genes Genomes Genomics ,  2007, vol.  1 (pg.  1- 20) Hanzawa Y,  Money T,  Bradley D.  A single amino acid converts a repressor to an activator of flowering,  Proc. Natl Acad. Sci. USA ,  2005, vol.  102 (pg.  7748- 7753) Google Scholar CrossRef Search ADS   Hättasch C,  Flachowsky H,  Hanke M.-V,  Lehmann S,  Gau A,  Kapturska,  D..  The switch to flowering: genes involved in floral induction of the apple ‘Pinova’ and the role of the flowering gene MdFT,  Acta Hortic. ,  2009, vol.  839 (pg.  701- 705) Google Scholar CrossRef Search ADS   Hättasch C,  Flachowsky H,  Kapturska D,  Hanke M.-V.  Isolation of flowering genes and seasonal changes in their transcript levels related to flower induction and initiation in apple (Malus domestica),  Tree Physiol. ,  2008, vol.  28 (pg.  1459- 1466) Google Scholar CrossRef Search ADS PubMed  Hsu CY,  Liu Y,  Luthe DS,  Yuceer C.  Poplar FT2 shortens the juvenile phase and promotes seasonal flowering,  Plant Cell ,  2006, vol.  18 (pg.  1846- 1861) Google Scholar CrossRef Search ADS PubMed  Igarashi M,  Abe Y,  Hatsuyama Y,  Ueda T,  Fukasawa-Akada T,  Kon T, et al.  Linkage maps of the apple (Malus × domestia Borkh.) cvs. ‘Ralls Janet’ and ‘Delicious’ include newly developed EST markers,  Mol. Breed. ,  2008, vol.  22 (pg.  95- 118) Google Scholar CrossRef Search ADS   Igasaki T,  Watanabe Y,  Nishiguchi M,  Kotoda N.  The FLOWERING LOCUS T/TERMINAL FLOWER 1 family in Lombardy poplar,  Plant Cell Physiol. ,  2008, vol.  49 (pg.  291- 300) Google Scholar CrossRef Search ADS PubMed  Jack T.  Molecular and genetic mechanisms of floral control,  Plant Cell ,  2004, vol.  16 (pg.  S1- S17) Google Scholar CrossRef Search ADS PubMed  James DJ,  Passey AJ,  Barbara DJ,  Bevan M.  Genetic transformation of apple (Malus pumila Mill.) using a disarmed Ti-binary vector,  Plant Cell Rep. ,  1989, vol.  7 (pg.  658- 661) Google Scholar PubMed  Jeanmougin F,  Thompson JD,  Gouy M,  Higgins DG,  Gibson TJ.  Multiple sequence alignment with Clustal X,  Trends Biochem. Sci. ,  1998, vol.  23 (pg.  403- 405) Google Scholar CrossRef Search ADS PubMed  Kardailsky I,  Shukla VK,  Ahn JH,  Dagenais N,  Christensen SK,  Nguyen JT, et al.  Activation tagging of the floral inducer FT,  Science ,  1999, vol.  286 (pg.  1962- 1965) Google Scholar CrossRef Search ADS PubMed  Kobayashi Y,  Kaya H,  Goto K,  Iwabuchi M,  Araki T.  A pair of related genes with antagonistic roles in mediating flowering signals,  Science ,  1999, vol.  286 (pg.  1960- 1962) Google Scholar CrossRef Search ADS PubMed  Kojima S,  Takahashi Y,  Kobayashi Y,  Monna L,  Sasaki T,  Araki T, et al.  Hd3a, a rice ortholog of the Arabidopsis FT gene, promotes transition to flowering downstream of Hd1 under short-day conditions,  Plant Cell Physiol. ,  2002, vol.  43 (pg.  1096- 1105) Google Scholar CrossRef Search ADS PubMed  Koornneef M,  Alonso-Blanco C,  Peeters A.JM,  Soppe W.  Genetic control of flowering time in Arabidopsis,  Annu. Rev. Plant Physiol. Plant Mol. Biol. ,  1998, vol.  49 (pg.  345- 370) Google Scholar CrossRef Search ADS PubMed  Kotoda N,  Iwanami H,  Takahashi S,  Abe K.  Antisense expression of MdTFL1, a TFL1-like gene, reduces the juvenile phase in apple,  J. Amer. Soc. Hortic. Sci. ,  2006, vol.  131 (pg.  74- 81) Kotoda N,  Wada M.  MdTFL1, a TFL1-like gene of apple, retards the transition from the vegetative to reproductive phase in transgenic Arabidopsis,  Plant Sci. ,  2005, vol.  168 (pg.  95- 104) Google Scholar CrossRef Search ADS   Kotoda N,  Wada M,  Komori S,  Kidou S,  Abe K,  Masuda T,  Soejima T.  Expression pattern of homologues of floral meristem identity genes LFY and AP1 during flower development in apple,  J. Amer. Soc. Hortic. Sci. ,  2000, vol.  125 (pg.  398- 403) Kotoda N,  Wada M,  Kusaba S,  Kano-Murakami Y,  Masuda T,  Soejima J.  Overexpression of MdMADS5, an APETALA1-like gene of apple, causes early flowering in transgenic Arabidopsis,  Plant Sci. ,  2002, vol.  162 (pg.  679- 687) Google Scholar CrossRef Search ADS   Kotoda N,  Wada M,  Masuda T,  Soejima J.  The break-through in the reduction of juvenile phase in apple using transgenic approaches,  Acta Hortic. ,  2003, vol.  625 (pg.  337- 343) Google Scholar CrossRef Search ADS   Levy YY,  Dean C.  The transition to flowering,  Plant Cell ,  1998, vol.  10 (pg.  1973- 1990) Google Scholar CrossRef Search ADS PubMed  Lifschitz E,  Eviatar T,  Rozman A,  Shalit A,  Goldshmidt A,  Amsellem Z, et al.  The tomato FT ortholog triggers systemic signals that regulate growth and flowering and substitute for diverse environmental stimuli,  Proc. Natl Acad. Sci. USA ,  2006, vol.  103 (pg.  6398- 6403) Google Scholar CrossRef Search ADS   Lin M.-K,  Belanger H,  Lee Y.-J,  Varkonyi-Gasic E,  Taoka K.-I,  Miura E, et al.  FLOWERING LOCUS T protein may act as the long-distance florigenic signal in the cucurbits,  Plant Cell ,  2007, vol.  19 (pg.  1488- 1506) Google Scholar CrossRef Search ADS PubMed  Michaels SD.  Flowering time regulation produces much fruit,  Curr. Opin. Plant Biol. ,  2009, vol.  12 (pg.  75- 80) Google Scholar CrossRef Search ADS PubMed  Mimida N,  Goto K,  Kobayashi Y,  Araki T,  Ahn JH,  Weigel D, et al.  Functional divergence of the TFL1-like gene family in Arabidopsis revealed by characterization of a novel homologue,  Genes Cells ,  2001, vol.  6 (pg.  327- 336) Google Scholar CrossRef Search ADS PubMed  Mimida N,  Kidou S.-I,  Kotoda N.  Constitutive expression of two apple (Malus × domestica Borkh.) homolog genes of LIKE HETEROCHROMATIN PROTEIN1 affects flowering time and whole-plant growth in transgenic Arabidopsis,  Mol. Genet. Genom. ,  2007, vol.  278 (pg.  295- 305) Google Scholar CrossRef Search ADS   Mimida N,  Kotoda N,  Ueda T,  Igarashi M,  Hatsuyama Y,  Iwanami H, et al.  Four TFL1/CEN-like genes on distinct linkage groups show different expression patterns to regulate vegetative and reproductive development in apple (Malus × domestica Borkh.),  Plant Cell Physiol. ,  2009, vol.  50 (pg.  394- 412) Google Scholar CrossRef Search ADS PubMed  Murashige T,  Skoog F.  A revised medium for rapid growth and bioassays with tobacco tissue culture,  Physiol. Plant. ,  1962, vol.  15 (pg.  473- 497) Google Scholar CrossRef Search ADS   Nishikawa F,  Endo T,  Shimada T,  Fujii H,  Shimizu T,  Omura M.  Increased CiFT abundance in the stem correlates with floral induction by low temperature in Satsuma mandarin (Citrus unshiu Marc.),  J. Exp. Bot. ,  2007, vol.  54 (pg.  2439- 2448) Google Scholar CrossRef Search ADS   Notaguchi M,  Abe M,  Kimura T,  Daimon Y,  Kobayashi T,  Yamaguchi A, et al.  Long-distance, graft-transmissible action of Arabidopsis FLOWERING LOCUS T protein to promote flowering,  Plant Cell Physiol. ,  2008, vol.  49 (pg.  1645- 1658) Google Scholar CrossRef Search ADS PubMed  Ohshima S,  Murata M,  Sakamoto W,  Ogura Y,  Motoyoshi F.  Cloning and molecular analysis of the Arabidopsis gene Terminal Flower 1,  Mol. Gen. Genet. ,  1997, vol.  254 (pg.  186- 194) Google Scholar CrossRef Search ADS PubMed  O’Rourke D,  Janick J,  Sansavini S.  World apple cultivar dynamics,  Chron. Hortic. ,  2003, vol.  43 (pg.  10- 13) Pearl RT.  Apple rootstocks I to XVI,  J. South-Eastern Agric. Coll., ,  1932, vol.  30 (pg.  194- 213) Perrière G,  Gouy M.  WWW-Query: an on-line retrieval system for biological sequence banks,  Biochimie ,  1996, vol.  78 (pg.  364- 369) Google Scholar CrossRef Search ADS PubMed  Purwestri YA,  Ogaki Y,  Tamaki S,  Tsuji H,  Shimamoto K.  The 14-3-3- protein GF14c acts as a negative regulator of flowering in rice by interacting with the florigen Hd3a,  Plant Cell Physiol. ,  2009, vol.  50 (pg.  429- 438) Google Scholar CrossRef Search ADS PubMed  Ratcliffe OJ,  Amaya I,  Vincent CA,  Rothstein S,  Carpenter R,  Coen ES, et al.  A common mechanism controls the life cycle and architecture of plants,  Development ,  1998, vol.  125 (pg.  1609- 1615) Google Scholar PubMed  Ratcliffe OJ,  Bradley DJ,  Coen ES.  Separation of shoot and floral identity in Arabidopsis,  Development ,  1999, vol.  126 (pg.  1109- 1120) Google Scholar PubMed  Sadamori S,  Yoshida Y,  Murakami H,  Ishizuka S.  New apple variety ‘Fuji’,  Bull. Hortic. Res. Station, Jpn, Ser. (C) ,  1963, vol.  1 (pg.  1- 6) Sax K.  The origin of the Pomoideae,  Proc. Amer. Soc. Hortic. Sci. ,  1933, vol.  30 (pg.  147- 150) Soejima J,  Bessho H,  Tsuchiya S,  Komori S,  Abe K,  Kotoda N.  Breeding of Fuji apples and performance on JM rootstocks,  Compact Fruit Tree ,  1998, vol.  31 (pg.  22- 24) Sreekantan L,  Thomas MR.  VvFT and VvMADS8, the grapevine homologues of the floral integrators FT and SOC1, have unique expression patterns in grapevine and hasten flowering in Arabidopsis,  Funct. Plant Biol. ,  2006, vol.  33 (pg.  1129- 1139) Google Scholar CrossRef Search ADS   Sung S.-K,  Yu G.-H,  An G.  Characterization of MdMADS2, a member of the SQUAMOSA subfamily of genes, in apple,  Plant Physiol. ,  1999, vol.  120 (pg.  969- 978) Google Scholar CrossRef Search ADS PubMed  Takada S,  Goto K.  TERMINAL FLOWER2, a HETEROCHROMATIN PROTEIN1-like protein of Arabidopsis, counteracts the activation of FLOWERING LOCUS T by CONSTANS in the vascular tissues of leaves to regulate flowering time,  Plant Cell ,  2003, vol.  15 (pg.  2856- 2865) Google Scholar CrossRef Search ADS PubMed  Tamaki S,  Matsuo S,  Wong HL,  Yokoi S,  Shimamoto K.  Hd3a protein is a mobile flowering signal in rice,  Science ,  2007, vol.  316 (pg.  1033- 1036) Google Scholar CrossRef Search ADS PubMed  Teper-Bamnolker R,  Samach A.  The flowering integrator FT regulates SEPALLATA3 and FRUITFULL accumulation in Arabidopsis leaves,  Plant Cell ,  2005, vol.  17 (pg.  2661- 2675) Google Scholar CrossRef Search ADS PubMed  Tuskan GA,  DiFazio S,  Jansson S,  Bohlmann J,  Grigoriev I,  Hellsten U, et al.  The genome of Black Cottonwood, Populus trichocarpa (Torr. & Gray),  Science ,  2006, vol.  313 (pg.  1596- 1604) Google Scholar CrossRef Search ADS PubMed  van der Linden CG,  Vosman B,  Smulders M.JM.  Cloning and characterization of four apple MADS box genes isolated from vegetative tissues,  J. Exp. Bot. ,  2002, vol.  53 (pg.  1025- 1036) Google Scholar CrossRef Search ADS PubMed  Van Ooijen JW,  Voorrips RE. ,  JoinMap 3.0, Software for the calculation of linkage maps ,  2001 Wageningen Plant Research International Wada M,  Cao Q,  Kotoda N,  Soejima J,  Masuda T.  Apple has two orthologues of FLORICAULA/LEAFY involved in flowering,  Plant Mol. Biol. ,  2002, vol.  49 (pg.  567- 577) Google Scholar CrossRef Search ADS PubMed  Wigge PA,  Kim MC,  Jaeger KE,  Busch W,  Schmid M,  Lohmann JU, et al.  Integration of spatial and temporal information during floral induction in Arabidopsis,  Science ,  2005, vol.  309 (pg.  1056- 1059) Google Scholar CrossRef Search ADS PubMed  Wilkie JD,  Sedgley M,  Olesen T.  Regulation of floral initiation in horticultural trees,  J. Exp. Bot. ,  2008, vol.  59 (pg.  3215- 3228) Google Scholar CrossRef Search ADS PubMed  Yamaguchi A,  Kobayashi Y,  Goto K,  Abe M,  Araki T.  TWIN SISTER OF FT (TSF) acts as a floral pathway integrator redundantly with FT,  Plant Cell Physiol. ,  2005, vol.  46 (pg.  1175- 1189) Google Scholar CrossRef Search ADS PubMed  Yao J.-L,  Cohen D,  Atkinson R,  Richardson K,  Morris B.  Regeneration of transgenic plants from the commercial apple cultivars Royal Gala,  Plant Cell Rep. ,  1995, vol.  14 (pg.  407- 412) Google Scholar CrossRef Search ADS PubMed  Yao J.-L,  Dong Y.-H,  Kvarnheden A,  Morris B.  Seven MADS-box genes in apple are expressed in different parts of the fruit,  J. Amer. Soc. Hortic. Sci. ,  1999, vol.  124 (pg.  8- 13) Yeung K,  Seitz T,  Li SF,  Janosch P,  McFerran B,  Kaiser C, et al.  Suppression of Raf-1 kinase activity and MAP kinase signaling by RKIP,  Nature ,  1999, vol.  401 (pg.  173- 177) Google Scholar CrossRef Search ADS PubMed  Yoo SY,  Kardailsky I,  Lee JS,  Weigel D,  Ahn JH.  Acceleration of flowering by overexpression of MFT (MOTHER OF FT AND TFL1),  Mol. Cells ,  2004, vol.  17 (pg.  95- 101) Google Scholar PubMed  Yoshida Y,  Haniuda T,  Tsuchiya S,  Sanada T,  Masuda T,  Bessho H, et al.  New apple cultivar ‘Sansa’,  Bull. Fruit Tree Res Station, Ser. C (Morioka, Japan) ,  1988, vol.  15 (pg.  1- 12) Zeevaart JA.  Leaf-produced floral signals,  Curr. Opin. Plant Biol. ,  2008, vol.  11 (pg.  541- 547) Google Scholar CrossRef Search ADS PubMed  Zimmerman RH.  Juvenility and flowering in woody plants: a review,  HortScience ,  1972, vol.  7 (pg.  447- 455) Abbreviations Abbreviations AP1 APETALA1 ATC ARABIDOPSIS THALIANA CENTRORADIALIS BFT BROTHER OF FT AND TFL1 CaMV cauliflower mosaic virus CAPS cleaved amplified polymorphic sequence CEN CENTRORADIALIS DAF days after flowering DIG digoxigenin FBS fruit-bearing shoots FLC FLOWERING LOCUS C FT FLOWERING LOCUS T FUL FRUITFULL LD long day LFY LEAFY LG linkage group MFT MOTHER OF FT AND TFL1 N–J Neighbor–Joining RT–PCR reverse transcription–PCR SOC1 SUPPRESSOR OF OVEREXPRESSION OF CONSTANS1 SS succulent shoots SSR simple sequence repeat TFL1 TERMINAL FLOWER1 TSF TWIN SISTER OF FT. © The Author 2010. Published by Oxford University Press on behalf of Japanese Society of Plant Physiologists. All rights reserved. For permissions, please email: journals.permissions@oxfordjournals.org

Journal

Plant and Cell PhysiologyOxford University Press

Published: Apr 8, 2010

Keywords: Apple FLOWERING LOCUS T ( FT ) Flowering time Juvenility Malus × domestica Borkh

There are no references for this article.