Mol Biol Rep (2012) 39:3737–3746 DOI 10.1007/s11033-011-1149-8

The expression level of Rosa Terminal Flower 1 (RTFL1) is related with recurrent flowering in roses Li-Na Wang • Yun-Feng Liu • Yu-Man Zhang Rong-Xiang Fang • Qing-Lin Liu

•

Received: 30 January 2011 / Accepted: 24 June 2011 / Published online: 8 July 2011 Ó Springer Science+Business Media B.V. 2011

Abstract We examined the relationship between the recurrent flowering character and the expression patterns of TERMINAL FLOWER 1 (TFL1) homologs in roses, using flower buds of Rosa multiflora, R. rugosa, R. chinensis, and six other rose species and nine rose cultivars. RTFL1 (Rosa TFL1) genes were amplified from rose genomic DNA using a combination of degenerate and gene-specific primers by thermal asymmetric interlaced-PCR and normal PCR, respectively. Their copy numbers in different species were determined by Southern blots. We used real-time PCR to analyze the expression patterns of RTFL1 genes at four developmental stages (pre-sprouting, young, mid-aged, and mature flower buds). Our results show that there are at least three RTFL1 homologs in roses; RTFL1a, RTFL1b, and RTFL1c. The sequences of the homologs were more similar among the same homolog in different species than among the different homologs in the same species. For RTFL1a, we detected two copies in R. multiflora, two copies in R. rugosa, and one copy in R. chinensis. For RTFL1c, we

Li-Na Wang and Y-F Liu contributed equally to this work.

Electronic supplementary material The online version of this article (doi:10.1007/s11033-011-1149-8) contains supplementary material, which is available to authorized users. L.-N. Wang  Y.-F. Liu  Q.-L. Liu (&) Department of Ornamental Horticulture and Landscape Architecture, China Agricultural University, Beijing 100193, China e-mail: [email protected] Y.-M. Zhang (&)  R.-X. Fang State Key Laboratory of Plant Genomics, Institute of Microbiology, Chinese Academy of Sciences, Beijing 100101, China e-mail: [email protected]

detected one copy in R. multiflora, two copies in R. rugosa, and three copies in R. chinensis. We detected only one copy of RTFL1b in R. chinensis. RTFL1c was expressed at high levels at all four flowering stages in R. multiflora and R. rugosa, which are non-recurrent flowering species, whereas it was barely detected in R. chinensis (a recurrent flowering species) at any stage. These results were further verified in six other non-recurrent flowering species and nine recurrent flowering cultivars. These results suggest that the recurrent flowering habit in roses results from lower expression of RTFL1c, which may be related to recurrent flowering character in roses. Keywords Flower development  Inflorescence meristem identity gene  Rosa chinensis  Recurrent flowering habit

Introduction As a crucial phase in plant development, flowering is a complex process including flower induction, flower evocation and flower bud formation. There are three gene classes such as inflorescence meristem genes, floral meristem genes and flower organ identity genes involved or interacted in flower development. The inflorescence meristem genes control the transition of shoot apical meristem from vegetative to reproductive (inflorescence), forcing or delaying the flowering time. The floral meristem genes control the derivation of floral meristem from inflorescence meristem. And the flower organ identity genes, i.e. homeotic genes or A, B, C function genes determine where and how many sepals, petals, stamens and carpels produced in flowers [1, 2]. However, few woody plants such as roses have the ability of flowering more than once or recurrent flowering in a year; which is an important horticultural and

123

3738

economical characteristics, but their molecular background is not clear [3–6]. The TERMINAL FLOWER 1 (TFL1) is an inflorescence meristem identity gene, and encodes phosphatidylethanolamine binding protein (PEBP) belongs to the TFL/CENTRORADIALIS (CEN) family. In Arabidopsis tfl1 mutant, the development of the inflorescence is determinate rather than indeterminate, and early flowering occurs due to the shortened vegetative growth. TFL1 contains four exons with conserved exon/intron boundaries, and the fourth exon plays a critical role in converting flowers into shoot-like structures [7]. There are four segments in the fourth exon (A, B, C, and D), and the B and C segments are sufficient for TFL1-like activity [8]. TFL1 maintains the inflorescence identity and inhibits floral development through suppressing the gene activities of LEAFY (LFY) and APETALA1 (AP1) in the central region of shoot meristems [9, 10]. The amino acid sequences (175 a-AA) encoded by TFL1, together with its homologous protein encoded by FLOWERING LOCUS T (FT) play antagonistic roles in mediating flowering signals. The flowering is inhibited by TFL1 and promoted by FT [11, 12]. The notable difference of FT and TFL1 is the presence of two critical amino acid residues [8, 13]. Some homologs of TFL1 have been isolated from many plants, include GMTfl1 in soybean, VvTFL1 in grapevine, Self-Pruning (SP) in tomato, CET in tobacco, CsTFL1 in citrus, MdTFL1 in apple, and PcTFL1 in pear [14–19]. There are two growth genotypes, indeterminacy (GMTfl1) and determinacy (GMtfl1) in soybean, just like wild-type and tfl1 mutant genotype in arabidopsis. GMtfl1 is derived from GMTfl1, due to a single-nucleotide substitution in GMTfl1. However, the introduction of GMTfl1 allele into tfl1 mutant can fully restore the wild type, but GMtfl1 allele did not result in apparent phenotypic change [8]. In grapevine, the overexpression of VvTFL1A, a homolog of TFL1, can lead to the reiterated reproductive meristem (RRM) of grapevine cultivar ‘Carignan’ and display alterations of the third and fourth whorls of RRM flowers, as a result of the lack of determination of flower meristems [20]. TFL1 homologs may have some functions other than determining the inflorescence meristem in different plants. Modern roses result from complex hybridizations of at least seven rose species, including Rosa chinensis, R. gigantea, R. damascena, R. gallica, etc. [21]. Various groups have been classified, such as hybrid tea (HT), floribunda, shrub, climbing (Cl), and miniature roses. The recurrent flowering character in HT roses was inherited from R. chinensis, which originated from China [22]. Other rose species, such as R. multiflora and R. rugosa, flower only once a year [23]. The recurrent flowering is an important character both for year-round production of cut flowers and for landscaping. In R. multiflora, the secondary shoots remain in the vegetative growth phase after the first blooming. In

123

Mol Biol Rep (2012) 39:3737–3746

R. chinensis and HT roses derived from this species, however, secondary flowering shoots and flower buds form after anthesis of the primary shoot as long as forcing temperatures are maintained at around 16°C. The main difference is the secondary shoots that are remained in the non-recurrent species in vegetative stage, while those in recurrent-flowering species have a ‘‘terminal flower’’ at the tip. The same process occurs in the third shoots, and so on, so that they flower continuously. This phenotype is similar to that of the tfl1 mutant in Arabidopsis. Therefore, if the non-recurrent flowering of R. multiflora is considered as the wild-type phenotype, the recurrent flowering of R. chinensis and HT roses could be hypothesized to be mutants of TFL1 homologs. There are some genes controlling qualitative and quantitative traits in the genetic maps of roses. Such as the major genes or QTLs (quantitative trait locus) of petal number, days to flower, inflorescence development, and soon have been located. A QTL for flowering time and many inflorescence traits were mapped to the same cQTL (common QTL) [24, 25]. The genes related to flowering process have different expression patterns during flower development in roses. Most of them express in all tissue, including root, leaf, shoot, floral bud, vegetative apices, vegetative pre-floral apices and floral apices; others express in more specific patterns. Among inflorescence meristem identity genes, RoELF8 expresses in root, apices and shoot, RoCOL2 in leaf, apices and floral buds, which is induced early, RoFT in floral apices and floral buds. Of floral meristem identity genes, RoLFY and RoAP1b express in induced and floral-induced apices as well as floral buds. As floral organ identity genes, MASAKO only induced in floral buds at the late stage of floral development, and RoAP1b only in floral apices [26, 27]. There was no information about TFL1 homologs from roses. To determine if the TFL1 is related with the recurrentflowering phenotype, we isolated the TFL1 homologs from roses, and comparatively analyzed their expression level in R. chinensis (recurrent flowering) and R. multiflora and R. rugosa (non-recurrent flowering). The expression differences of RTFL1c were further investigated in six nonrecurrent species and nine recurrent cultivars. The result showed that the low level expression of RTFL1 is related with the recurrent flowering habit.

Materials and methods Plant materials Flower buds from non-recurrent flowering species (R. multiflora ‘Albo-plena’, R. rugosa) and recurrent flowering species (R. chinensis ‘Slater’s Crimson China’)

Mol Biol Rep (2012) 39:3737–3746

3739

Table 1 Stages of flower bud development in rose species Stage

Flower buds

Collection date

Shoot length (cm) R. mutiflora

Time before flowering (weeks)

R. rugosa

R. chinensis

R. multiflora

R. rugosa

R. chinensis

I

Pre-sprout

12 Mar. 2008

0.67

0.48

0.25

9

9

10

II

Young

26 Mar. 2008

1.13

0.95

0.76

7

7

8

III

Middle

9 Apr. 2008

2.42

2.09

1.53

5

5

6

IV

Mature

22 Apr. 2008

9.77

5.23

5.01

3

3

4

were harvested from Beijing Botanical Gardens on 12 March, 26 March, 9 April, and 22 April 2008 corresponding to four stages of flower development (Table 1), pre-sprouting buds (Stage I), young flower buds (Stage II), mid-aged flower buds (Stage III) and mature flower buds (Stage IV). In order to further confirm the relationship between the expression level of RTFL1 and the recurrent flowering trait, the flower buds from more species were also collected, five non-recurrent rose species (R. banksiae, R. damascena, R. mairei var. plurijuga, R. primula, and R. roxburghii) from Beijing Botanical Gardens, one nonrecurrent variety R. chinenesis var. spontanea from Changping Nursery, Beijing Forestry University and nine recurrent-flowering rose cultivars (Altissimo (Cl.), Compassion (Cl.), Double Delight (HT), Parkdirektor Riffers (Cl.), Red Success (HT), Sonia (HT), Spectra (Cl.), Uncle Walter (HT) and Westernland (Shrub)) from The Institute of Vegetables and Flowers, Chinese Academy of Agricultural Sciences. The harvested rose buds were quick-frozen in liquid nitrogen and stored at -80°C until analyses. Isolation of RTFL1 Genomic DNA was extracted from flower buds of R. multiflora, R. rugosa, and R. chinensis by the CTAB method [28]. One pair of degenerate primers (TFL1-F-1/ TFL1-R-1, Table 2) was designed according to the conserved sequence of TFL1, and another pair, TFL1-F-2/ TFL1-R-2 was taken from the literature [19]. The TFL1 homologs from rose were amplified using the genomic DNAs as templates. The PCR conditions were as follows: 94°C for 5 min; 30 cycles of 94°C for 30 s, 52°C for 30 s, 72°C for 60 s; and 72°C for 10 min. The PCR products were purified, ligated into the pGEM-T vector (Promega, Madison, WI, USA) and transformed into competent cells of Escherichia coli strain JM109. The positive clones were sequenced (Invitrogen, Shanghai, China) and eight RTFL1 fragments were obtained; three RTFL1a, three RTFL1b, and two RTFL1c. Specific primers for TAIL-PCR were designed and synthesized based on the sequences of RTFL1a, RTFL1b, and RTFL1c fragments. Four AD (arbitrary degenerate)

primers were also synthesized [29, 30]. The TAIL-PCR reaction mixture contained 0.25 lM SP (specific) primer, 2.5 lM AD primer, and the genomic DNA of R. multiflora (as a template). The TAIL-PCR conditions are shown in Table S1. After the primary PCR, 1 ll aliquots of a 1:50 dilution of the primary TAIL-PCR amplification products were added to the secondary PCR. After the secondary PCR, 1 ll aliquots of 1:20 dilution of the secondary TAILPCR amplification products were added to the third PCR. The positive recombination clones were sequenced. Fragments of 1,008 and 387 bp, 442 and 340 bp, and 636 and 505 bp, corresponding to 50 and 30 sequences of RTFL1a, RTFL1b, and RTFL1c, respectively, were amplified by TAIL-PCR using the primers AD1/A-SP2-R and AD3/ASP3-F, AD2/B-SP3-R and AD1/B-SP2-F, and AD2/C-SP3R and AD4/C-SP2-F, respectively. Full-length sequences of RTFL1a, RTFL1b, and RTFL1c were deduced from the above fragments and amplified by PCR with degenerate primers and the sequences obtained by TAIL-PCR. Gene-specific primers were designed based on the deduced full-length sequences. The genomic DNAs of each of R. multiflora, R. rugosa, and R. chinensis were used as templates for PCR using the gene-specific primers (where the same primers were used for the same gene in all three species). The full-length sequences of RTFL1a, RTFL1b, and RTFL1c were amplified by PCR using the primers TFL1a-F-1/TFL1a-R-1, TFL1b-F-1/TFL1b-R-1, and TFL1c-F-1/TFL1c-R-1, respectively. The PCR conditions were as follows: 94°C for 5 min; 30 cycles of 94°C for 30 s, 54°C for 2 min, 72°C for 60 s; and 72°C for 10 min. RNA extraction and reverse transcription Total RNA was extracted using an EASYspin Plant RNA Quick Extraction kit (Galen Biotech, Beijing, China). For first-strand cDNA synthesis, the 20 ll reaction mixture contained 3 lg total RNA, 300 ng oligo (dT)18 (Takara, Tokyo, TM Japan), and 200 units SuperScript III reverse transcriptase (Invitrogen, Carlsbad, CA, USA). Primers used for RT-PCR were designed according to the deduced cDNA sequence based on the obtained genomic DNA sequence. The cDNA sequences of RTFL1a, RTFL1b, and RTFL1c were amplified

123

3740 Table 2 Primers used for PCR to isolate RTFL1 from roses

Mol Biol Rep (2012) 39:3737–3746

A Degenerate primers for random fragments TFL1-F-1

AATGGCCATGAGCTCTTTCCTTC

TFL1-R-1

AACGYCTKCKRGCGGCRGTTTC

TFL1-F-2 TFL1-R-2

TCCTGGCCCTAGTGA(T/C)CCTTAT GCGC(A/G)TTGAAGTA(A/G)A(C/G) GGCAG

B SP primers for TAIL-PCR A-SP3-F

AATGGTCTTGGCCTGCCTGTGG

A-SP2-R

TGGGGCTTGGGAAATCAGGATCT

B-SP2-F

GCCGAGGCCAAACATAGGGATCC

B-SP3-R

AGGAAGGAAAGAGCTCATGGCCATT

C-SP2-F

GCCAAGGCCAAACATAGGAATCCACAG

C-SP3-R

TCAAATAAGGGTCACTAGGGCCAGGAA

C AD primers for TAIL-PCR AD1

NTCGA(G/C)T(A/T)T(G/C)G(A/T)GTT

AD2

NGTCGA(G/C)(A/T)GANA(A/T)GAA

AD3

(A/T)CAGNTG(A/T)TNGTNCTG

AD4

(G/C)TTGNTA(G/C)TNCTNTGC

D Gene specific primers for DNA full length TFL1a-F-1 TFL1a-R-1

TAGGGTCTCAAGGAATAACAGCAAT CATTAGTACAGTACATGCCAGAAAGC

TFL1b-F-1

TCGAGAGAGATGAAGTACAAGACTATAATTAAT

TFL1b-R-1

TTATATATTGACTCTCACAACAGATGATCTATG

TFL1c-F-1

CCTCCTCTTATATACTCTTCTTCTTCTCAA

TFL1c-R-1

ACGGAGAGAGACCAACAACTAAATG

E Primers for cDNA cloning TFL1a-F-2

CCACTACTTCATTGATGTCGAGGA

TFL1a-R-2

CCCTGAGCTCCTTATGTTATTATCG

TFL1b-F-2

CTCAAGGAAAACAGAAAAGGATATATAATT

TFL1b-R-2

ATATTGACTCTCACAACAGATGATCTATGA

TFL1c-F-2

TCTTTGATATTTGTACGACCTTCTCTC

TFL1c-R-2

ACCGTTATAGCATGGAATCTTTTG

F Primers for real-time PCR TFL1a-F-3

CCAACGCAAAACACACATTCATCTTCTG

TFL1a-R-3

GAACATGTCCACAACCTCTCCTACCACTCT

TFL1b-F-3 TFL1b-R-3

TCTTCTCAACTATAAATATGGGGTCTGCTTCA AACACTGGGGGAGAAATAATCAACAACATC

TFL1c-F-3

TTTGATATTTGTACGACCTTCTCTCCTCTCTC

TFL1c-R-3

CATTGAAGACGAGTTTGGTGTTGTAAGTGAC

Act-F-3

CCCCAAGGCCAACAGAGAAAAGATG

Act-R-3

CGACCACTAGCATACAGGGAGAGAACAGC

by RT-PCR with the primers TFL1a-F-2/TFL1a-R-2, TFL1bF-2/TFL1b-R-2, and TFL1c-F-2/TFL1c-R-2, respectively. The PCR conditions were as follows: 94°C for 5 min; 30 cycles of 94°C for 30 s, 52°C for 30 s, 72°C for 60 s; and 72°C for 10 min. Sequence and phylogenetic analyses The identities of cDNA sequences of RTFL1a, RTFL1b, and RTFL1c from different rose species were compared

123

using GeneDoc version 2.7.0 (Free software foundation, Inc. Boston, MA, USA). The sequences of amino acids were predicted from the cDNA sequences and the deduced amino acids were analyzed by Vector NTI advanced version 10 (Invitrogen, Carlsbad, CA, USA) and GeneDoc version 2.7.0. To clarify the phylogenetic relationships between RTFL1 and other TFL1 homologs, a phylogenetic tree was constructed using MEGA 3.1 (molecular evolutionary genetics analysis software package version 3.1, the

Mol Biol Rep (2012) 39:3737–3746

Biodesign Institute, Tempe, AZ, USA). Before generating the phylogeny, predicted proteins were aligned using ClustalX version 1.83 [31] using the default parameters. The neighbor-joining (NJ) method of MEGA 3.1, based on p-distance, was used to construct the phylogenetic tree. The pairwise deletion option was used to handle gaps and missing data. The reliability of the obtained tree was tested using the bootstrap method with 1,000 replications. The tertiary structures of RTFL1 proteins were predicted using software at http://expasy.org and http://www.ncbi.nlm.nih. gov. Southern blotting Genomic DNA was extracted from pre-sprouting buds of R. multiflora, R. rugosa, and R. chinensis using a Nucleon Phytopure Genomic DNA Extraction kit (Code: RPN8510, GE Healthcare UK Limited., Buckinghamshire, UK). According to the known genomic sequences of these genes, restriction endonucleases with a single cut site located close to the 50 or 30 end of the probe sequence were chosen for Southern blot analyses. The digested DNA fragments were concentrated using ethanol and dissolved in 20 ll dH2O, separated by electrophoresis on a 0.7% agarose gel, and then transferred onto BiodyneÒB Nylon Transfer membranes (P/N: 60207, Pall Corporation, New York, USA). Prehybridization and hybridization were carried out using standard protocols [32]. The probes for RTFL1a (794 bp), RTFL1b (701 bp), and RTFL1c (875 bp) were each labeled with [a-32P]-dCTP (PerkinElmer, Boston, MA, USA) using the Prime-a-GeneÒ Labeling System (Code: U1100, Promega corporation, Madison, WI, USA) according to the manufacturer’s instructions. Real-time RT-PCR To analyze the expression of RTFL1a, RTFL1b, and RTFL1c, real-time PCR was conducted twice with different rose materials. First, we analyzed expression patterns in the flower buds of R. multiflora, R. rugusa, and R. chinensis at four developmental stages. Then, we examined the expression of RTFLc in six additional rose species and nine rose cultivars. All RNA extractions and PCR reactions were repeated at least three times for each experimental material. For real-time PCRs, the primers (TFL1a-F-3/ TFL1a-R-3, TFL1b-F-3/TFL1b-R-3, and TFL1c-F-3/ TFL1c-R-3) were designed from their respective cDNA sequences. Actin primers (Act-F-3/Act-R-3) were designed according to the Actin mRNA sequence from a hybrid cultivar [33]. Real-time PCRs were performed using the SYBRÒ Green PCR Master Mix kit (Code: QPK-201, Takara, Osaka, Japan) on a real-time PCR instrument (DNA Engine OpticonÒ2 Continuous Fluorescence

3741

Detector, MJ Research Inc, Waltham, MA, USA). The 25 ll reaction mixture contained a 1 ul aliquot of a 1:25 dilution of the reverse transcription product, 12.5 ll 29 SYBRÒ Green PCR Master mix, and 0.5 lM forward and reverse primers. The PCR conditions were as follows: 95°C for 3 min; 44 cycles of 94°C for 30 s, 61°C for 30 s, 72°C for 30 s; and 72°C for 10 min. The relative gene expression levels were calculated by threshold cycle (CT) values of the RTFL1 gene and the Actin gene, using the following formula [34]: relative mRNA expression concentration = 2-DCT, where DCT = CT RTFL1 gene - CT Actin gene. Data from three replicates were used to calculate standard errors and for t-tests.

Results Cloning and sequencing of RTFL1 from roses Eight RTFL1 homologs were cloned from rose genomic DNA by TAIL-PCR and PCR. RmTFL1a, b, and c were isolated from R. multiflora; RrTFL1a, b, and c from R. rugosa; RcTFL1a and b from R. chinensis. RmTFL1a, RrTFL1a, and RcTFL1a were 1,912, 1,758, and 1,908 bp in length, respectively. RmTFL1b, RrTFL1b, and RcTFL1b were 1,272, 1,267, and 1,272 bp in length, respectively, and RmTFL1c and RrTFL1c were 1,544, and 1,511 bp in length, respectively. The coding sequences (CDS) were compared among different genes within the same species, and among the same genes across different species (Table 3). There was only 77% identity among RmTFL1a, b, and c in R. multiflora; 78% identity among RrTFL1a, b, and c in R. rugosa; and 78% identity between RcTFL1a and b in R. chinensis. The identity among RmTFL1a, RrTFL1a, and RcTFL1a from the three different species was more than 99%; that among RmTFL1b, RrTFL1b, and RcTFL1b was more than 99%; and that between RmTFL1c and RrTFL1c was 98%. The alignments of deduced RTFL1 amino acid sequences (Fig. S1) indicated that the pattern of amino acid sequences was more variable within a species than among species for the same gene (Table 3). The identity of deduced amino acids of RmTFL1a, b, and c in R. multiflora was 81%; that of RrTFL1a, b, and c in R. rugosa was 81%; and that of RcTFL1a and RcTFL1b in R. chinensis was 66%. The identity of RmTFL1a, RrTFL1a, and RcTFL1a was more than 97%; that of RmTFL1b, RrTFL1b, and RcTFL1b was more than 98%; and that of RmTFL1c and RrTFL1c was 98%. These results indicated that there are higher identities of RTFL1 genes among different species than among the homologs within the same species. For this reason, we collectively referred to RmTFL1a, RrTFL1a,

123

3742

Mol Biol Rep (2012) 39:3737–3746

Table 3 Identity (%) of CDS of RTFL1 mRNAs and predicted amino acids of RTFL1 in rose species RTFL1 homologs

RmTFL1a

Identity (%) of CDS of mRNAs/Identity (%) of amino acids RrTFL1a

RcTFL1a

99/97

100/99

67/66

67/65

67/66

68/70

68/70

99/97

68/68

68/67

68/68

68/72

69/72

67/66

67/66

67/66

68/71

68/71

99/98

100/100

77/81

78/82

99/98

77/80

78/81

RrTFL1a RcTFL1a

RmTFL1b

RmTFL1b

RrTFL1b

RrTFL1b

RcTFL1b

RcTFL1b

RmTFL1c

77/81

RmTFL1c

RrTFL1c

78/82 98/98

CDS Coding sequence, RTFL1 Rosa Terminal Flower 1 gene, RRTFL1 is the protein encoded by RTFL1 Rm Rosa multiflora, Rr, Rosa rogusa, Rc Rosa chinensis, a, b, c indicate three homologs

and RcTFL1a as RTFL1a. Similarly, RmTFL1b, RrTFL1, and RcTFL1b were designated RTFL1b, and RmTFL1c and RrTFL1c were designated RTFL1c. Thus, there are at least three homologs of the RTFL1 gene in roses. Phylogenetic and structural analysis of RTFL1 proteins There were three major clusters in the TFL1 phylogenic tree (Fig. S2). RTFL1 proteins of R. multiflora, R. rugosa, and R. chinensis belonged to the TFL1 cluster and were not grouped with the FT homologs. Moreover, the amino acids His (H) and Asp (D) of RTFL1 were characteristic of TFL1, so they were classified as RTFL1 rather than FT. We concluded that the genes we have isolated from roses are TFL1 homologs. RTFL1a showed high identity with the tobacco CET1 protein, which was grouped with grapevine VvTFL1C in the phylogenic tree. RTFL1b showed high identity with apple MdCENa and MdCENb, which were classified with grapevine VvTFL1A and Antirrhinum majus CEN. RTFL1c showed high identity with apple MdTFL1-1 and MdTFL1-2, which were grouped with grapevine VvTFL1B and Arabidopsis TFL1 (Fig. S2). The tertiary structures of TFL1 (Fig. 1a) and RTFL1c proteins (Fig. 1b) were very similar. There are some differences, especially in segments B and C, among RTFL1a (Fig. 1c), RTFL1b (Fig. 1d), and TFL1. Copy numbers of RTFL1 in roses The sequences of probes for RTFL1a, RTFL1b, and RTFL1c had low identities with each other (44.2% between RTFL1a and RTFL1b, 44.7% between RTFL1a and RTFL1c, and 39.6% between RTFL1b and RTFL1c). Thus, there was no interference among these genes in the Southern blots. Southern blotting analyses revealed two copies of RTFL1a in R. multiflora, two in R. rugosa, and one in

123

R. chinensis (Fig. 2a), suggesting that the inflorescence identity genes in roses are well conserved. There was only one copy of RTFL1b in R. chinensis (Fig. 2b), and no copies were detected in the other two species. As shown in Fig. 2c, there were different copy numbers of RTFL1c in the three rose species; one in R. multiflora, two in R. rugosa, and three in R. chinensis, suggesting there are at least two pseudo or unfunctional TFL1c genes in R. chinensis. Different expression level of RTFL1 in roses Real-time PCR analyses showed that the expression levels of RTFL1a in R. multiflora, R. rugosa and R. chinensis were not significantly different (P B 0.05) at developmental Stages I, II, or III. At Stage IV, however, RTFL1a was expressed at high levels in R. multiflora, moderate levels in R. rugosa, and low levels in R. chinensis (Fig. 3a). The RTFL1b was expressed at the same level at all four stages in R. multiflora and R. rugosa. In R. chinensis, however, RTFL1b was expressed at higher levels at the pre-sprout (Stage I) and young (Stage II) stages (Fig. 3b). A parabolic expression pattern was observed for RTFL1c, with higher levels of expression at the mid-aged (Stage III) and mature flower bud (Stage IV) stages in R. multiflora, although there were no differences (P B 0.05) in expression of TFL1c between R. multiflora and R. rugosa at Stages I and II. No expression of TFL1c was detected at any stage in R. chinensis (Fig. 3c). These patterns of expression indicated that RTFL1c was more closely related to the flowering habit than RTFL1a and RTFL1b in roses. Furthermore, we analyzed the expression levels of RTFL1c in six rose species (non-recurrent flowering) and nine modern rose cultivars (recurrent flowering) (Fig. 4), and found that the expression levels of RTFL1c varied among species and cultivars. From cultivars such as Sonia and Westernland to species

Mol Biol Rep (2012) 39:3737–3746

3743

Fig. 1 Tertiary structures of RTFL1 protein and other TFL1 proteins. a Tertiary structure of TFL1 [8]. b, c, d Tertiary structures of RmTFL1c, RmTFL1a, RmTFL1b

Fig. 2 Southern blots of RTFL1 genes. a RTFL1a digested by EcoRI. b RTFL1b digested by EcoRV and BamHI. c RTFL1c digested by PstI and HindIII. Lanes 1, 5, and 12 contain positive controls. Lanes 2, 6,

9, 13, and 14 R. multiflora. Lanes 3, 7, 10, 15, and 16 R. rugosa. Lanes 4, 8, 11, 17, and 18 R. chinensis

such as R. mairei var. plurijuga and R. roxburghii, the expression levels increased gradually. There were significant differences (P B 0.05) in the expression levels of RTFL1c between recurrent and non-recurrent species/cultivars (except for R. damascena). The real-time PCR analyses indicated that the expression level of RTFL1c was very low in R. chinensis and other recurrent cultivars, but remarkably higher in six nonrecurrent flowering species, including R. chinensis var. spontanea. The exception of R. damascene need to be further investigated.

Discussion In this paper, three RTFL1 homologs, RTFL1a, RTFL1b, and RTFL1c, from three rose species were isolated and their copy number in each species was determined by Southern blot. We also analyzed their expression patterns at different stages of the three species, six other species and nine cultivars using real-time PCR. The three homologs showed different expression patterns among the various rose species. RTFL1c was not expressed or was expressed only at low levels in recurrent flowering species and

123

3744

Mol Biol Rep (2012) 39:3737–3746

Fig. 4 Expression of RTFL1c genes in non-recurrent and recurrentflowering roses. Recurrent cultivars: 1 Sonia (HT), 2 Westerland (Shrub), 3 Uncle Walter (HT), 4 Parkdirektor Riggers (Cl.), 5 Spectra (Cl.), 6 Compassion (Cl.), 7 Red Success (HT), 8 Double Delight (HT) and 9 Altissimo (Cl.). Non-recurrent species: 10 Rosa damascena, 11 Rosa banksiae, 12 Rosa primula, 13 Rosa mairei var. plurijuga,14 Rosa chinensis var. spontanea and 15 Rosa roxburghii. Relative RTFL1c mRNA concentrations with the same letters (a, b, or c) above columns were not significantly different (P B 0.05; t-test) among the different rose species/cultivars

Fig. 3 Expression of RTFL1 genes as detected by real-time PCR. a RTFL1a, b RTFL1b, c RTFL1c; Stages I, II, III and IV are presprouting buds, young flower buds, mid-flowering buds, and mature flower buds, respectively, which were collected at 9, 7, 5, 3 weeks before flowering date of R. mutiflora and R. rugosa, and 10, 8, 6 and 4 weeks before flowering date of R. chinensis. Relative RTFL1 mRNA concentrations with the same letters (a, b, or c) were not significantly different (P B 0.05; t-test) among the different developmental stages or among different rose species

cultivars (R. chinensis and nine modern rose cultivars), whereas it was expressed at higher levels in eight nonrecurrent flowering species and varieties. Thus, there are different expression patterns of RTFL1c in recurrent and non-recurrent flowering roses. This means functional divergence may have occurred among RTFL1 homologs in roses. The similar divergences exhibited in apple, pea and pear. Each of them has more than one TFL1 homolog may functioned redundantly, such as a flowering repressor and a regulator of vegetative and inflorescence meristem identity [35–37].We detected three copies of RTFL1c in

123

R. chinensis. Its cDNA was not cloned by RT-PCR with the degenerated primer for TFL1, nor was the DNA fragment amplified by PCR with the specific primer for RTFL1c. Moreover, its mRNA was not detected by real-time RTPCR with the specific primers for RTFL1c. This result indicates that the RTFL1c gene exists in the genome, but it is not transcribed into mRNA in R. chinensis, which is consistent to HvTFL1, a TFL1 homology in barley, there is no expression could be tested [39]. In R. chinensis var. spontanea (non-recurrent), the DNA fragments were not amplified with the specific primers as in R. chinensis, but it expressed normally as in other non-recurrent species. This phenomenon should be further investigated. In this study, we detected differential expression patterns of RTFL1 homologs between recurrent and nonrecurrent flowering roses. These results suggested that the non-recurrent flowering of R. multiflora and other rose species resembles a wild-type, while the recurrent flowering pattern of R. chinensis and some other cultivars resembles that of rtfl1 mutants. In non-recurrent flowering species, RTFL1c expresses normally, which is thought to inhibit the expression of LFY homologs [9]. The homeotic A, B, and C functional genes cannot be activated to develop flower buds and therefore, recurrent flowering is prevented. In the recurrent flowering species and cultivars, however, the RTFL1c gene is not or lower expressed. This releases the LFY gene, which is downstream of RTFL1, to function normally. Thus, there is continued development of flower buds, resulting in recurrent flowering. The flowering habits of modern roses are complex; however, most cultivars are recurrent flowering types because of continuous breeding and selection. In fact, even the non-recurrent flowering climbing roses and hybrid rugosa have recurrent-flowering cultivars. The recurrent

Mol Biol Rep (2012) 39:3737–3746

flowering habit in modern roses is controlled by a recessive monogene [38]. Our studies suggested that RTFL1c is strongly related to the recurrent flowering habit in roses. Even if it is not the recessive monogene, it must be a major gene for controlling the flowering habit. The variations in the expression levels of RTFL1c in recurrent and nonrecurrent flowering roses indicated there must be some factors that modify its expression. This implies the existence of some minor genes related to the recurrent flowering habit. Some desirable characteristics of wild rose species, including resistance to black spot (Diplocarpon rosae) and powdery mildew (Sphaerotheca pannosa var. rosae), are difficult to pass on to modern roses by hybridization, because recurrent flowering is a recessive character in the F1 generation [38]. As a preliminary study, this paper provides clues that the RTFL1c gene is related to recurrent flowering. If this is verified by further research, it will be possible to transform wild-type roses into recurrent flowering roses using antisense or RNAi technology to knock down RTFL1c. This has already been achieved in the apple ever-flowering line 705, which expresses an antisense MdTFL1 [40].

3745

10. 11.

12.

13.

14.

15. 16.

17.

18.

Acknowledgments This research was supported by the High-Tech Research and Development Plan (863), No. 2006AA10Z187, the National Natural Sciences Foundation Committee of China, No. 30871732, and a Guest Investigator Grant of the State Key Laboratory of Plant Genomics. We thank Miss Sadia Humaira (Institute of Microbiology, CAS), Dr. Yi Ma (UC, Davis), and Dr. Kate Donald (RHS Wisely Garden) for suggestions on the manuscript.

20.

References

21.

1. Coen ES, Meyerowitz EM (1991) The war of the whorls: genetic interaction controlling flower development. Nature 353:31–37 2. Weigel D, Meyerowitz EM (1994) The ABCs of floral homeotic genes. Cell 78:203–209 3. Semeniuk P (1971) Inheritance of recurrent blooming in Rosa wichuraiana. J Hered 62(3):203–204 4. Svejda F (1977) Breeding for improvement of flowering attributes of winterhardy Rosa kordesii Wuff hybrids. Euphytica 26(3):703–708 5. Debener T (1999) Genetic analysis of horticulturally important morphological and physiological characters in diploid roses. Gartenbauwissenschaft 64(1):14–20 6. Debener T, Malek BV, Mattiesch L, Kaufmann H (2001) Genetic and molecular analysis of important characters in roses. Acta Hortic 547:45–49 7. Ratcliffe O, Amaya I, Vincent C, Rothstein S, Carpenter R, Coen E, Bradley D (1998) A common mechanism controls the life cycle and architecture of plants. Development 125:1609–1615 8. Ji HA, David M, Victoria JW, Mark JB, Jeong HL, So YY, Stefan RH, Robert LB, Detlef W (2006) A divergent external loop confers antagonistic activity on floral regulators FT and TFL1. EMBO J 25:605–614 9. Lijegren SJ, Gustafson-Brown C, Pinyopich A, Ditta GS, Yanofsky MF (1999) Interactions among APETALA1, LEAFY,

19.

22. 23. 24.

25.

26.

27.

28.

and TERMINAL FLOWER1 specify meristem fate. Plant Cell 11:1007–1018 Ratcliffe OJ, Bradley DJ, Cone ES (1999) Separation of shoot and floral identity in Arabidopsis. Development 126:1109–1120 Bradley D, Carpenter R, Copsey L, Vincent C, Rothstein S, Coen E (1996) Control of inflorescence architecture in Antirrhinum. Nature 376:791–797 Kobayashi Y, Kaya H, Goto K, Iwabuchi M, Araki T (1999) A pair of related genes with antagonistic roles in mediating flowering signals. Science 286:1960–1962 Hanzawa Y, Money T, Bradley D (2005) A single amino acid converts a repressor to an activator of flowering. Proc Natl Acad Sci USA 102:7748–7753 Tian Z, Wang X, Lee R, Li Y, Specht JE, Nelson RL, McClean PE, Qiu L, Ma J (2010) Artificial selection for determinate growth habit in soybean. Proc Natl Acad Sci USA 107: 8563–8568 Carmona MJ, Calonje M, Martı´nez-Zapater JM (2007) The FT/ TFL1 gene family in grapevine. Plant Mol Biol 63:637–650 Pnueli L, Carmel-Goren L, Hareven D, Gutfinger T, Alvarez J, Ganal M, Zamir D, Lifschitz E (1998) The SELF-PRUNING gene of tomato regulates vegetative to reproductive switching of sympodial meristems and is the ortholog of CEN and TFL1. Development 125:1979–1989 Amaya I, Ratcliffe OJ, Bradley DJ (1999) Expression of CENTRORADIALIS (CEN) and CEN-like genes in tobacco reveals a conserved mechanism controlling phase change in diverse species. Plant Cell 11:1405–1417 Pillitteri LJ, Lovatt CJ, Walling LL (2004) Isolation and characterization of a TERMINAL FLOWER homolog and its correlation with juvenility in citrus. Plant Physiol 135:1540–1551 Esumi T, Tao R, Yonemori K (2005) Isolation of LEAFY and TERMINAL FLOWER 1 homologues from six fruit tree species in the subfamily Maloideae of the Rosaceae. Sex Plant Reprod 17: 277–287 Fernandez L, Torregrosa L, Segura V, Bouquet A, MartinezZapater JM (2010) Transposon-induced gene activation as a mechanism generating cluster shape somatic variation in grapevine. Plant J 61:545–557 Matsumoto S, Kouchi M, Yabuki J, Kusunoki M, Ueda Y, Fukui H (1998) Phylogenetic analyses of the genus Rosa using the matK sequence: molecular evidence for the narrow genetic background of modern roses. Scientia Hort 77:73–82 Chen JY (2001) Taxonomy of flower cultivars in China. China Forestry Publisher, Beijing Fei YL, Liu QL, Ge H (2008) Crops and their wild relatives in China—flowers. China Agricultural Press, Beijing Hibrand-Saint Oyant L, Crespel L, Rajapakse S, Zhang L, Foucher F (2008) Genetic linkage maps of rose constructed with new microsatellite markers and locating QTL controlling flowering traits. Tree Genet Genomes 4:11–23 Kawamura K, Hibrand-Saint Oyant L, Crespel L, Thouroude T, Lalanne D, Foucher F (2011) Quantitative trait loci for flowering time and inflorescence architecture in rose. Theor Appl Genet 122:661–675 Remay A, Lalanne D, Thouroude T, Le Couviour F, HibrandSaint Oyant L, Foucher F (2009) A survey of flowering genes reveals the role of gibberellins in floral control in rose. Theor Appl Genet 119:767–781 Foucher F, Chevalier M, Corre C, Soufflet-Freslon V, Legeai F, Hibrand-Saint Oyant L (2008) New resources for studying the rose flowering process. Genome 51:827–837 Xu Q, Wen XP, Deng XX (2004) A simple protocol for isolating genomic DNA from chestnut rose (Rosa roxburghii Tratt.) for RFLP and PCR analyses. Plant Mol Biol Report 22:301a–301g

123

3746 29. Liu YG, Mitsukawa N, Oosumi T, Whittier RF (1995) Efficient isolation and mapping of Arabidopsis thaliana T-DNA insert junctions by thermal asymmetric interlaced PCR. Plant J 8:457–463 30. Tsugeki R, Kochieva EZ, Fedoroff NV (1996) A transposon insertion in the Arabidopsis SSR16 gene causes an embryodefective lethal mutation. Plant J 10:479–489 31. Thompson JD, Gibson TJ, Plewniak F, Jeanmougin F, Higgins DG (1997) The ClustalX windows interface: flexible strategies for multiple sequence alignment aided by quality analysis tools. Nucleic Acids Res 24:4876–4882 32. Sambrook J, Russell DW (2001) Molecular Cloning: a laboratory manual, 3rd edn. Cold Spring Harbor, New York 33. Hennayake CK, Takagi S, Nishimura K, Kanechi M, Uno Y, Inagaki N (2006) Differential expression of anthocyanin biosynthesis genes in suspension culture cells of Rosa hybrida cv. Charleston. Plant Biotechnol 23:379–385 34. Hayeshi R, Hilgendorf C, Artursson P, Augustijns P, Brodin B (2008) Comparison of drug transporter gene expression and functionality in Caco-2 cells from 10 different laboratories. Eur J Pharm Sci 35:383–396 35. Mimida N, Kotoda N, Ueda T, Igarashi M, Hatsuyama Y, Iwanami H, Moriya S, Abe K (2009) Four TFL1/CEN-like genes on

123

Mol Biol Rep (2012) 39:3737–3746

36.

37.

38.

39.

40.

distinct linkage groups show different expression patterns to regulate vegetative and reproductive development in apple (Malus x domestica Borkh.). Plant Cell Physiol 50:394–412 Foucher F, Morin J, Courtiade J, Cadioux S, Ellis N, Banfield MJ, Rameau C (2003) DETERMINATE and LATE FLOWERING are two TERMINAL FLOWER1/CENTRORADIALIS homologs that control two distinct phases of flowering initiation and development in pea. Plant Cell 15:2742–2754 Esumi T, Tao R, Yonemori K (2008) Expression analysis of the LFY and TFL1 homologs in floral buds of Japanese pear (Pyrus pyrifolia Nakai) and Quince (Cydonia oblonga Mill.). J Jpn Soc Hortic Sci 77:128–136 Debener T (2003) Inheritance of characteristics. In: Roberts AV (ed) Encyclopedia of rose science. Elsevier Academic Press, Amsterdam, pp 286–292 Kikuchi R, Kawahigashi H, Ando T, Tonooka T, Handa H (2009) Molecular and functional characterization of PEBP genes in barley reveal the diversification of their roles in flowering. Plant Physiol 149:1341–1353 Kotoda N, Iwanami H, Takahashi S, Abe K (2006) Antisense expression of MDTFL1, a TFL1-like gene, reduces the juvenile phase in apple. J Am Soc Hort Sci 131:74–81