Introduction

During the last decade, a huge amount of genetic and molecular information about Arabidopsis, Antirrhinum, and rice has accumulated. These collective research efforts have led to increased understanding of the complexity of development in higher plants (Aswath and Kim 2005). The initiation of organized development is a complex morphogenetic phenomenon in which extrinsic and intrinsic factors are important (Aswath and Kim 2005; Prakash and Kumar 2002). The MADS-box genes, which encode transcription factors, have been studied mainly in flower development by identifying and characterizing flower homeotic mutants (Jack 2001; Kaufmann et al. 2005; Theissen et al. 2000). Although some of these genes have also been studied in vegetative tissues with regard to plant growth and development, a few genes expressed in vegetative tissues, specifically, have been identified in Arabidopsis (AGL24 and SVP) and in potato (STMADS11 and STMADS16) (Kim et al. 2002). In rice plants, a large number of MADS box genes were expressed in vegetative tissues and seed along with floral tissue (Arora et al. 2007), but it was still necessary to understand their function in plant growth and development.

Phylogenetic analyses performed for the MADS-box multigene family have distributed these genes into distinct groups or subfamilies (Munster et al. 1997; Purugganan et al. 1995; Theissen et al. 1996). Phylogenetic analyses have shown that the StMADS11 subfamily of the MADS-box gene family (Becker and Theissen 2003; Garcia-Maroto et al. 2000) includes JOINTLESS in tomato (which controls the formation of pedicel abscission zones; Mao et al. 2000), BM1 in barley (transcripts of which accumulate at high levels in both internodes and nodes; Schmitz et al. 2000), and IbMADS4 from sweet potato (which causes a highly branched phenotype in transgenic chrysanthemum; Aswath et al. 2004). Ectopic expression of the STMADS11 subfamily members in the development of the reproductive structure is a feature shared by genes that are not normally expressed, and which may have a vegetative role.

AGL24 and SVP are two Arabidopsis MADS-box genes in the STMADS subfamily. These genes also control floral transition. Despite their extensive sequence similarity, AGL24 acts as a promoter of this process (Michaels et al. 2003; Yu et al. 2002). However, SVP has a role as a repressor of flowering (Hartmann et al. 2000; Lee et al. 2007). Expression of AGL24 with the CaMV 35S promoter accelerates flowering, but accentuates floral abnormalities including alterations in floral organ number and the appearance of chlorophyll-containing petals (Michaels et al. 2003; Yu et al. 2004). Ectopic expression of the SVP gene is associated with a late-flowering phenotype (Masiero et al. 2004) and conversion of sepals and petals into leaf-like structures in a manner similar to that of 35S::AGL24 plants (Liu et al. 2007: Masiero et al. 2004).

Over-expression of OsMADS22 and OsMADS47, the orthologues of SVP and AGL24, respectively, in Arabidopsis produces the severe floral abnormalities similar to those observed in plants that over-express SVP or AGL24, and delays senescence. When StMADS16 from Solanum tuberosum is ectopically expressed in tobacco, the plants have a deferred flowering time, greater node numbers under short-day conditions only, and flowers with vegetative features (Garcia-Maroto et al. 2000). These results indicate that the STMADS gene subfamily might be involved in the regulatory network that directs vegetative organogenesis during plant growth (Garcia-Maroto et al. 2000) and may indirectly control the flowering process.

Three criteria can help evaluate whether orthologous genes of different species share a common function:

  • conservation of amino acid sequence;

  • similarity of mRNA transcription patterns during development; and

  • phylogenetic relationships.

The IbMADS3-1 protein shares 98% amino acid identity with STMADS16 (Garcia-Maroto et al. 2000), and 81 and 91% amino acid identity with STMADS11 (Carmona et al. 1998) and SVP (Hartmann et al. 2000), respectively, in the MADS-box sequence. STMADS11 and STMADS16 are found in vegetative and floral tissues (Carmona et al. 1998; Garcia-Maroto et al. 2000). Taken together, these observations suggest that the IbMADS3-1 genes from sweet potato and STMADS16 from potato might share functional similarity.

In this paper, we report integration of identification of the IbMADS3-1 gene isolated from the sweet potato and its functional characterization in tobacco. The results indicate that IbMADS3-1 may interfere with floral development and superimpose a vegetative program in tobacco.

Materials and methods

Plant materials

Sweet potato plants (Ipomoea batatas L. “Yulmi”, “White star”, “Elegantsummer”, and “Koganesengan”) and tobacco plants (Nicotiana tabacum “Xanthi”) were used in this study. Plants were grown in a climate chamber at 25 ± 3°C under a 16 h light/8 h dark photoperiod. In particular, tobacco transgenic plants T2 lines were grown under 460 nm blue and 630 nm red LED light (DYLED 45, Dynebio, Korea).

Cloning of the full length IbMADS3-1 cDNA

Total RNA was isolated from white fibrous roots and pigmented roots of sweet potato (Ipomoea batatas L. “Yulmi”) using the modified CTAB method (Kim and Hamada 2005). cDNA synthesis was performed with the 3′-RACE (Rapid Amplification of cDNA Ends) system (TaKaRa, Japan) in accordance with the manufacturer’s instructions. 3′-RACE adaptor primer and degenerate primer (VLCDADV (forward primer: GTI (C/T) TI TG(C/T) GA(C/T) GCI GA (A/G) GT)) (Kim et al. 2002) were used for isolation of the IbMADS3-1 gene. Moreover, detailed sequence comparison and alignment were performed with the CLC Free workbench program (CLC Bio, Denmark), the protein BLAST program in the NCBI database, and the ClustalW program (http://www.ebi.ac.uk/clustalw).

Structure analysis of the IbMADS3-1 gene

Specific primers for the open reading frame (ORF) were used in polymerase chain reaction (PCR) analysis. PCR amplification was done as follows: a total volume of 50 μl PCR reaction mixture assembled with 30 ng DNA as template, 5 μl 10× Buffer, 4 μl 2.5 mM dNTPs, 1 μl 10 pmol primers, and 2.5 U ExTaq polymerase (Takara). PCR reaction mixtures were heated at 94°C for 5 min, and then 35 cycles of amplification were performed as follows: denaturation for 30 s at 94°C, annealing for 30 s at 60°C, extension for 1 min at 72°C, with a 5 min final extension at 72°C. The amplified PCR products were assayed using 1% agarose gel electrophoresis. The purified PCR fragments were cloned into a T&A cloning vector (Real Biotech, Taiwan). Sequence analyses including sequence alignment and comparison were executed with the CLC Free workbench program (CLC Bio).

Plant transformation and molecular analysis

For A. tumefaciens-mediated tobacco transformation, tobacco leaves from in-vitro-grown plants were used. The coding region of IbMADS3-1 was fused to the cauliflower mosaic virus (CaMV) 35S promoter at the BamHI and XbaI sites of the pCAMLA vector (pCAMBIA 1300 + P35S-Tnos cassette) (Lee et al. 2005). The pCAMLA vector was transformed into A. tumefaciens strain EHA 105 by a freeze–thaw method (An 1987). Tobacco leaf discs inoculated with bacterial suspension at OD600 = 0.6–1.0 for 5 min were washed with distilled water (dH2O) and dried on sterile filter paper. The inoculated leaf discs were incubated in the dark for 2 days and then transferred on MS medium supplemented with 0.1 mg l−1 1-naphthaleneacetic acid (NAA), 1.0 mg l−1 benzylaminopurine (BAP), 50 mg l−1 hygromycin, and 500 mg l−1 cefotaxime for shoot induction. Regenerated shoots were rooted on half-strength MS medium containing 50 mg l−1 hygromycin and 500 mg l−1 cefotaxime. Transgenic lines selected on the basis of hygromycin resistance were further identified by IbMADS3-1 specific PCR and Southern hybridization analysis. Verified transgenic T2 generation seedlings were transplanted into pots and grown in the growth room for further experiments.

Southern hybridization analysis and RT-PCR analysis

Genomic DNA was isolated from tobacco leaves by a modified CTAB method (Kim and Hamada 2005). Twenty micrograms each of DNA samples was digested overnight with XbaI. Also, genomic DNAs were extracted from sweet potato root of the “White star”, “Elegantsummer”, “Koganesengan”, and “Yulmi” cultivars. These genomic DNAs were digested with EcoRI and BamHI. Digested products were purified and separated by 0.8% agarose gel electrophoresis. After electrophoresis, the gel was agitated in a tray containing 0.25 M HCl for 15 min and rinsed briefly with dH2O, and again agitated in a tray containing 0.4 M NaOH for 30 min. The DNAs were then transferred on to a positively charged Nylon membrane (Roche Molecular Biochemicals, Germany) by the capillary transfer method (Sambrook and Russell 2001). The membrane was rinsed in 2× SSC for 5 min, and then allowed to dry in a dry-oven set to 80°C. A 248 bp IbMADS3-1 fragment was labeled as a probe by two primers (5′-CAG GAC TGA GCC GTG TTA TAG AG-3′, 5′-GTC CTC CAC AAG ACT ACG ATG AC-3′) using a PCR DIG probe synthesis kit (Roche Molecular Biochemicals, Germany). Hybridization, washing, and detection were conducted according to the guidelines established by the manufacturer.

Total RNA was isolated from tobacco leaves and flower buds of IbMADS3-1 T2 lines using the modified cetyl trimethylammonium bromide (CTAB) method (Kim and Hamada 2005), and total RNAs were also isolated from six organs of sweet potato (Ipomoea batatas L. “Yulmi”): leaf, petiole, stem, tuberous root, thick pigmented root, and white fibrous root.

Each total RNA was treated with DNase to remove genomic DNA using the PrimeScript first strand cDNA Synthesis kit (Takara). One microgram of total RNA was synthesized to first-strand cDNA using Prime Script reverse transcriptase (Takara) and oligo dT primer. Then, reverse transcription (RT) was performed for 90 min at 42°C. The 1 μl RT reaction mixture was used as template in a 30 μl PCR-amplification reaction comprising 10× buffer, 200 mM dNTPs, 10 pmol primers, and 2.5 U exTaq polymerase (Takara, Japan).

PCR was run for 35 cycles of 30 s at 94°C, 30 s at 60°C and for 1 min at 72°C, with a 5 min final extension at 72°C. As an internal positive control of tobacco, a primer set (5′-GGA CTC TGG TGA TGG TGT C-3′, 5′-CCT CCA ATC CAA ACA CTG TA-3′) designed on the basis of the tobacco actin gene sequence was used in the experiment with tobacco. The primer set of sweet potato tubulin gene (5′-CAA CTA CCA GCC ACC AAC TGT-3′, 5′-CAA GAT CCT CAC GAG CTT CAC-3′) was used as the positive control in the sweet potato experiments. The amplified PCR products were visualized by electrophoresis on a 1% agarose gel.

To confirm the expression level of the floral organ identity genes, the following seven specific primer sets were used in the PCR: NAG1 (5′-CTG AAA CTG AGA GAG CTC AGC A-3′, 5′-GTA CAC TGC CAG ACA AGG AAC A-3′), NFL1 (5′-AAG GAG CGA GGA GAG AAT GTT G-3′, 5′-AGG AGG TAG ATA TAG GGG CAA GC-3′), NtGLO (5′-CAC AGA GAG CTC ATG ATG TTG G-3′, 5′-CTG AAC TCG GAA GGC AAA AG-3′), NtAP2 (5′-CAC CCT ACT ACC CCA TGG AAA T-3′, 5′-CTC ATC AGA AAA CGC ACA CC-3′), NtPLE36 (5′-CTT GAG AAG GCC ATT GGT AGA G-3′, 5′-GCG AGA GTA ATG GGG ATT AGG T-3′), NtDEF (5′-GGG GAC TAT AAC TCT GTG CTT GG-3′, 5′-GCT AAA GTC CGA CCG ATT CAC T-3′), and NtMADS5 (5′-GCA TCT TTC CTC CTA CAA CCA C-3′, 5′-GGG CAT AAC ATT GTA TTG AAC-3′).

Chlorophyll contents

Tobacco petals of IbMADS3-1 transgenic plants were collected after complete flowering. Petals (0.2 g) were immersed in and extracted with 80% acetone solution for 7 days at 4°C before measurement of absorbance. This experiment was repeated three times for each IbMADS3 transgenic line. The contents (Ca, chlorophyll a; Cb, chlorophyll b; Ca + b, total chlorophyll; and Cx + c, total carotenoids) were determined by use of Lichtenthaler’s equation (1982):

$$ Ca(\upmu \hbox{g}\,{\text{ml}}^{ - 1} ) = 12.25\,A\,663.2 - 2.79A\,646.8 $$
$$ Cb(\upmu \hbox{g}\,{\text{ml}}^{ - 1} ) = 21.50\,A\,646.8 - 5.10A\,663.2 $$
$$ Ca + b(\upmu \hbox{g}\,{\text{ml}}^{ - 1} ) = 7.15\,A\,663.2 + 18.71A\,646.8 $$
$$ Cx + c(\upmu \hbox{g}\,{\text{ml}}^{ - 1} ) = (1,000\,A\,470 - 1.82\,Ca - 85.02\,Cb)/198. $$

Scanning electron microscopy

Observation of tobacco T2 plants petal surface using scanning electron microscopy (SEM) was performed as previously described (Kim 2008). Pieces of petals were immersed in modified Karnovsky’s fixative consisting of 2% (v/v) glutaraldehyde and 2% (v/v) paraformaldehyde in 0.05 M sodium cacodylate buffer (pH 7.2) at 4°C overnight and washed with the same buffer three times each for 10 min. The specimens were fixed in second fixative containing 1% (w/v) osmium tetroxide in the same buffer at 4°C for 2 h. They were washed briefly two times with dH2O and dehydrated in a graded ethanol series at room temperature. Dehydration was performed using 10 min exposure to 30, 50, 70, 80, 95, and three times in 100% ethanol. The specimens were treated with isoamyl acetate two times each for 10 min, and dried in a CPD 030 critical point drier (BAL-TEC, Liechtenstein). Tobacco petal specimens were individually mounted on metal stubs, sputter-coated with gold, and observed by SEM.

Results and discussion

Sequence analysis of IbMADS3-1

To isolate IbMADS3-1 mRNA, a PCR amplification primer was designed for a highly conserved sequence in the MADS-domain of the STMADS family (IbMADS4, STMADS16, STMADS11, SVP, and AGL 24) (Kim et al. 2002). Sequence analysis showed that the full length IbMADS3-1 cDNA was 1,194 bp long and consisted of a 279 bp 5′-untranslated region (UTR), 231 bp 3′-UTR, and a 684 bp ORF encoding 227 amino acids (Fig. 1a). The whole IbMADS3-1 gene was further amplified from sweet potato genomic DNA, using specific primers for the ORF. Comparison of the genomic DNA sequence with the IbMADS3-1 cDNA revealed that it consisted of seven exons, which perfectly matched the cDNA sequence and first intron of 1,399-bp length and six introns (Fig. 1b). Sequence analysis of IbMADS3-1 grouped the gene as a member of the MADS-box transcription factor family. Its deduced amino acid sequence displayed high identity with those of SVP (AAG24508), STMADS16 (AAB94005), AGL24 (AAC63140), and STMADS11 (AAB94006) (Fig. 1c). The MADS-domain sequence of IbMADS3-1 was 90% identical with those of SVP, STMADS11, and STMADS16, and was 83% identical with that of AGL24. Phylogenetic analysis of the five MADS-box proteins–SVP and AGL24 from Arabidopsis, STMADS11 and STMADS16 from potato, and IbMADS3 from sweet potato–classified this group into the STMADS family (Kim et al. 2002) (Fig. 1d).

Fig. 1
figure 1

Structure and sequence analysis of the IbMADS3-1 gene. a Numbers on the right represent nucleotide and amino acid positions. The deduced amino acid sequence is shown in single-letter code above the nucleotide sequence. The start codon is indicated by a box and the termination codon is marked with an asterisk. The underlined nucleotide sequences are 5′-UTR and 3′-UTR. bIbMADS3-1 consists of seven exons and six introns. Especially, the first intron of IbMADS3-1 is much longer (about 12–19-fold) than the other introns. Filled boxes indicate the ORF, and lines between boxes indicate introns. The numbers represent nucleotides. c Comparison of the amino acid sequences of IbMADS3-1 and other homologous proteins. d Alignment of deduced amino acid sequences of IbMADS3-1 with high homologues including SVP (AAG24508), STMADS16 (AAB94005), AGL24 (AAC63140), and STMADS11 (AAB94006). The sequences were aligned using the ClustalW program. Identical amino acid residues in all five sequences are indicated by asterisks. Colons and dots indicate that amino acid residues are conserved in most sequences. Analysis was performed using the CLC workbench program

Full size image

STMADS11 and STMADS16 from potato are expressed during vegetative growth (Carmona et al. 1998). Although SVP and AGL24 are expressed in inflorescence meristems, stem, leaves, and roots, SVP transcripts disappear before emergence of the sepal primordia during flower development, and may positively regulate a repressor of flowering (Alvarez-Buylla et al. 2000). Subfamily members tend to share highly similar sequences and related functions (Doyle 1994; Purugganan et al. 1995; Theissen et al. 1996), so that IbMADS3-1 seems to be exerting its function in the vegetative phase.

These results were expected, considering that IbMADS3-1 is a sister gene to the STMADS-related genes that are specific for vegetative tissues, and that most previously characterized MADS-box genes cluster in phylogenetic clades of genes with similar functions and expression patterns (Alvarez-Buylla et al. 2000; Rounsley et al. 1995).

Southern hybridization and expression pattern analysis of IbMADS3-1

Southern hybridization suggested the presence of at least two copies of IbMADS3-1 in each cultivar of “Koganesengan”. The band patterns in EcoRI digestion within the “White star”, “Elegant summer”, and “Yulmi” cultivars were different. In the case of digested BamHI restrict enzyme, only “White star” exhibited a different band pattern among the cultivars (Fig. 2a). “Elegant summer”, “Koganesengan”, and “Yulmi” displayed similar band patterns. It is unclear at present whether this result was because of the hexaploid and heterozygous nature of sweet potato.

Fig. 2
figure 2

Southern blot and RT-PCR analysis of the IbMADS3-1 gene activity in sweet potato plants. a Genomic DNA of “White star” (W), “Elegant summer” (E), “Koganesengan” (K), and “Yulmi” (Y) cultivars was digested with EcoRI and BamHI. The genomic DNAs were separated on a 0.8% agarose gel, blotted, and probed with a DIG-labeled IbMADS3-1 fragment. The size marker (M) fragments of DNA are indicated on the left. b Total RNA was isolated from leaf (L), petiole (P), and stem (S), developing tuberous root (D), thick pigmented root (R), and white fibrous root (W). Tubulin was used as a control

Full size image

Total RNA was isolated from leaf, petiole, stem, and root tissues, and analyzed with the IbMADS3-1 specific primer. To determine the local expression pattern of IbMADS3-1, expression of the RNA from each organ was examined. RT-PCR revealed that each gene was expressed in all organs. In particular, the IbMADS3-1 gene was strongly expressed in white fibrous root (Fig. 2b). This result demonstrated that IbMADS3-1 transcripts are constitutively and differently expressed in various organs, suggesting a constitutive and/or specific function of IbMADS3-1 in vegetative tissue development. Additionally, the preferential expression of IbMADS3-1 more actively in root tissues (especially white fibrous root) implies that IbMADS3-1 may play an important role in root formation or development.

Nonetheless, previous studies of IbMADS3-1 had already suggested that this gene could function in vegetative organ development because of its high and apparently specific expression in roots (Kim et al. 2002). In this study, we confirmed that IbMADS3-1 is indeed expressed in vegetative tissues. In accordance with this pattern of expression, IbMADS3-1 is important for vegetative organ development.

Phenotype of IbMADS3 transgenic plants

Because of the minimum amount of plant tissue needed for PCR analysis, all the putative transformed leaves that exhibited hygromycin resistance were analyzed by PCR. A predicted 248-bp internal fragment for the IbMADS3-1 was amplified in all the leaves, consistent with integration of the transgene into the genomes. The genomic DNA from putative transgenic plants was subjected to gel blot analysis under high-stringency conditions. Hybridization with an IbMADS3-1 specific probe confirmed foreign DNA integration in the transgenic lines analyzed (Fig. 3a). Southern blot results indicated that the transformants selected had a single-copy insertion of T-DNA at different sites in the tobacco plant chromosome (Fig. 3b). Genomic DNA from the wild-type did not give any signal.

Fig. 3
figure 3

Confirmation of the IbMADS3-1 gene integration in transgenic plants. a PCR analysis of genomic DNA isolated from transgenic T2 lines. Abbreviations: M, marker; PC, Positive control (lines 2, 9, and 16 represent individual lines of transgenic tobacco plants). b Southern blot analysis of genomic DNA from transgenic T2 tobacco plants. Genomic DNA was digested with XbaI. Abbreviations: M, DIG-labeled DNA molecular weight marker. C, control (wild-type plant). Lines 2, 9, and 16 are individual lines of transgenic tobacco plants

Full size image

The IbMADS3-1 gene is expressed in vegetative tissue only, and IbMADS3-1 protein is a member of the STMADS subfamily (Kim et al. 2002). In general, gene expression correlates with the place where the gene exerts its function. In order to test this possibility, we created transgenic tobacco plants that constitutively transcribed IbMADS3-1 cDNA from the CaMV 35S promoter. The transgenic plants showed no morphological or growth differences in vegetative tissues, as compared with Wt plants. On the other hand, IbMADS3-1 over-expression lines exhibited remarkable morphological alterations in floral organs including sepal, petal, carpel, ovary, and peduncle, and the appearance of chlorophyll-containing petals (Fig. 4e). Only the stamen was unchanged to the naked eye (Fig. 4c). Sepals of IbMADS3-1 transgenic lines decreased compared with those of wild type. Furthermore, the wild-type sepal was connected with a pedicel of approximately 1-cm and transgenic plants pedicels were nearly degenerate (Fig. 4b).

Fig. 4
figure 4

Phenotypic alterations of the transgenic tobacco plants over-expressing IbMADS3-1 and arrangement of flower buds in the wild-type and over-expressing IbMADS3-1 tobacco plant. a Inflorescences of the tobacco wild type and IbMADS3-1 over-expressing line 9 and line 16. b Flower of the tobacco wild type and IbMADS3-1 over-expression lines. c Longitudinal sections of the tobacco flowers. The order is the same as in b. d The tip of petal of the tobacco wild type and IbMADS3-1 over-expression lines. e Chlorophyll content of wild type and IbMADS3-1 over-expression lines. Control is wild-type tobacco petal and lines 2, 9, and 16 are transgenic tobacco petals. Contents were similar in all lines. The density seemed to be higher for line 9 than for the other lines. f Flower bud arrangement of normal plant and transgenic plant. g Flower bud arrangement of wild-type plants was regular. h In contrast, the morphology of the flower bud arrangement of transgenic plants was altered and irregular

Full size image

The petal was altered in both shape and color. Petals of IbMADS3-1 transgenic lines grew longer than normal petal and color was yellowish green generally (Fig. 4b). Most of the transgenic plants retained pink color at the petal tip (Figs. 4f, 5d). By contrast, at the top of some petal tips of transgenic plants the white color disappeared, and just below the top of the petal tip was a yellowish color (Fig. 4f). The carpel length of wild-type flowers was the same as the petal length. The carpel length of transgenic plants was always reduced, but the extent of the decrease varied. Moreover, the stigma became smaller and had a reduced viscosity compared with the wild type (Fig. 4c). The ovary had an extended structure with a central furrow and a hardened internal structure, compared with the normal ovary. The nectaries, which normally appeared as an orange ring at the base, were lacking in the transgenic flowers (Fig. 4c, d). Such ovary shape alterations may reflect a lack of coordination in placental development at anthesis (stage 12 according to Garcia-Maroto et al. 2000; Koltunow et al. 1990). Past and previous observations support the notion that deformations of the ovary are produced by the absence of control in placental development. The arrangement of flower buds in the wild-type plant was very regular and displayed a circular pattern (Fig. 4g, h), while the arrangement in transgenic plants was slightly altered; both pedicel and sepal connecting with pedicel were shorter than those of normal plants (Fig. 4g). These observations bolster the view that the IbMADS3-1 gene product is concerned with suppression of pedicel and sepal development.

Fig. 5
figure 5

SEM analysis of characteristic cell surface morphology in wild-type tobacco petal and transgenic tobacco petals. Regions of petals of tobacco flowers sampled for SEM analysis from top of the tobacco petal tip (left) and middle of tobacco corolla tube (right). Surface cells of regions of each part of wild-type and transgenic petals were taken. a A wild-type petal tip. b A petal tip of IbMADS3-1 line 9. c A petal tip of IbMADS3 line 16. d A wild-type corolla tube. e A corolla tube of IbMADS3 line 9. f A corolla tube of IbMADS3 line 16

Full size image

Authentic surface cell types can be distinguished by SEM. Specified parts of tobacco flowers were used for SEM analysis. Cells from the petal tip were characterized by the presence of globular-tipped trichomes, stomata, and irregularly shaped cells. The trichomes of IbMADS3-1 transgenic line were altered in terms of an irregular size and increased density. In the petal tips of transgenic flowers, the epidermal cells were a dark color, rendering the stomata more clearly visible (Fig. 5a–c). The stomatal density was increased, as was the density of trichomes. Meanwhile, the surface of the corolla tube was similar to the petal tip surface in the density of trichomes and stomata, and the epidermal cells were elongated (Fig. 5d–f). In particular, IbMADS3-1 transgenic line 16 had remarkable visible changes in floral organ as contrasted with the other line and the wild-type plant. (Fig. 5c, f). This result showed that the petal tip surface morphology of transgenic flowers more closely mirrors that of the normal corolla tube (Fig. 5b–d).

Ectopic expression of IbMADS3-1 produced characteristic changes in organ identity carpel structure alterations and sepaloidy of the petals. One possible explanation of this observation is that the IbMADS3-1 product alone is capable of partially inducing the developmental pathways leading to petal and carpel formation. The IbMADS3-1 gene is exclusively expressed in the vegetative organs of normal plants, and phenotypic alterations are exclusively observed in the reproductive structures. It may be that over-expression of IbMADS3-1 in tobacco is able to trigger developmental events that are appropriate for vegetative tissues. In agreement with this suggestion, IbMADS3-1 expression is more abundant in vegetative tissues such as the white fibrous roots but not in the flower (Kim et al. 2002). It has previously been suggested that ectopic expression of STMADS16 (Garcia-Maroto et al. 2000), and OsMADS22 and OsMADS47 (Fornara et al. 2008) show vegetative characteristics, and similar alterations in tobacco and Arabidopsis flower development, respectively. These phenotypes could indicate that failure to establish floral identity results in enhancement of the vegetative traits of the flower.

Expression levels of the floral organ identity genes in flower buds of plants over-expressing IbMADS3-1

To confirm a possible genetic interaction between IbMADS3-1 and previously characterized genetic components of floral organ formation, mRNA expression of a number of genes known to be key regulators of floral organ development was assessed. It was confirmed that IbMADS3-1 had drastically increased levels of expression in transgenic lines (Fig. 6). Interestingly, the flowering promoter, NFL1, was clearly reduced at the RNA level compared with wild type in the background of transgenic lines 2, 9, and 16 (Fig. 6). In contrast, NAG1, NtGLO, NtPLE36, and NtDEF, which are upstream regulators in floral organ development (Davies et al. 1996; Hansen et al. 1993; Kang et al. 1995; Kempin et al. 1993; Theissen 2001), did not show significant alterations in mRNA expression in the transgenic lines. On the other hand, NtMADS5, which is principally required to direct a primordium to develop into a flower in tobacco (Smykal et al. 2007), showed a slight down-regulation compared with wild-type plants in transgenic lines. The results correlate with the abnormal shoot-like flower phenotypes. The transition from vegetative to reproductive development involves the coordination of several endogenous and environmental factors. Central to the transition to flowering is the increased expression of the LFY gene at the site of floral meristem formation in Arabidopsis, followed by the transcription of AP1 in the developing floral meristem (Mandel et al. 1992). The expression of AP1 is delayed and reduced in lfy mutant plants (Liljegren et al. 1999). In agreement with this, NtMADS5 is down-regulated by NFL1 in transgenic line backgrounds (Calonje et al. 2004). Therefore, our data suggest that IbMADS3-1 could be a repressor of NFL1 upstream of NtMADS5. However, complementation of ntmads5 and nfl1 mutants with IbMADS3-1 overexpression constructs should be pursued to confirm this proposed role of IbMADS3-1.

Fig. 6
figure 6

Expression pattern of the floral organ identity genes in flower buds of over-expressed IbMADS3-1 and schematic representation of the regulation of flowering transition. RNA was isolated from flower buds of a wild-type and three transgenic plants. The specific primer sets of six genes, NAG1 (L23925), NFL (U15798), NtGLO (X67959), AP2 (AJ299252), NtPLE36 (U63163), NtDEF (X96428), and NtMADS5 (AF068724) were used in the analysis

Full size image

There is one possibility of reconciling abnormal shoot-like flower phenotypes in the IbMADS3-1 over-expression line. Ectopic expression of IbMADS3-1 may interfere with floral development, especially repressed NFL1, and superimpose a vegetative program in tobacco.