Journal of Experimental Botany

JXB Advance Access originally published online on January 31, 2008
Journal of Experimental Botany 2008 59(2):247-259; doi:10.1093/jxb/erm305

This Article Abstract FREE Full Text (PDF) Supplementary Material OA All Versions of this Article: 59/2/247    most recent erm305v1 E-letters: Submit a response Alert me when this article is cited Alert me when E-letters are posted Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in ISI Web of Science Similar articles in PubMed Alert me to new issues of the journal Add to My Personal Archive Download to citation manager Disclaimer Google Scholar Articles by Song, J. Articles by Jameson, P. E. Search for Related Content PubMed PubMed Citation Articles by Song, J. Articles by Jameson, P. E. Agricola Articles by Song, J. Articles by Jameson, P. E. Social Bookmarking          What's this?

© 2008 The Author(s).
This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/by-nc/2.0/uk/) which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited. This paper is available online free of all access charges (see http://jxb.oxfordjournals.org/open_access.html for further details)

RESEARCH PAPER

Quantitative expression analysis of the ABC genes in Sophora tetraptera, a woody legume with an unusual sequence of floral organ development

Jiancheng Song^1,2, John Clemens^1,2 and Paula Elizabeth Jameson^1,*

¹School of Biological Sciences, University of Canterbury, Christchurch, New Zealand
²Institute of Molecular BioSciences, Massey University, Palmerston North, New Zealand

^* To whom correspondence should be addressed. E-mail: paula.jameson{at}canterbury.ac.nz

Received 17 September 2007; Revised 5 November 2007 Accepted 7 November 2007

	Abstract

Top Abstract Introduction Materials and methods Results Discussion Supplementary data References

 
Sophora is a woody genus of the Leguminosae in which an unusual order and process of floral organ development is often observed. The SEM results for Sophora tetraptera revealed precocious initiation of the carpel, delayed development of petals, and floral organ development interrupted by an unusual prolonged summer–autumn dormant period which occurred between organ initiation and organ differentiation. These observations provided an opportunity to track key floral identity genes over an extended developmental period. Homologues of LEAFY, APETALA1, PISTILLATA, and AGAMOUS were isolated from S. tetraptera. Real-time PCR enabled a simultaneous and quantitative analysis of both the temporal and spatial expression patterns of these four genes. Expression differences in the range of three to five orders of magnitude were detected between different genes and between different stages of flower development for the same gene. Although not functionally tested, the spatial expression patterns of the genes were consistent with expectations based on the ABC model of floral development. Their temporal expression patterns were consistent with the timing of flower initiation and the unusual order of organ development. Quantitatively, while the expression levels of the LFY homologue and the A-class gene were high during the periods of organ initiation and organ differentiation and low during the summer–autumn dormant period, high expression levels of the B- and C-class genes were detected only during the rapid, albeit delayed, phase of organ differentiation. Additionally, the sustained expression of the floral organ identity genes after differentiation reflects on-going roles for these genes during subsequent organ development.

Key words: Floral identity genes, floral ontogeny, legume, quantitative expression, qRT-PCR, real-time PCR, Sophora

	Introduction

Top Abstract Introduction Materials and methods Results Discussion Supplementary data References

 
In higher plants, the flowering process consists of several steps, including the formation of the floral meristem, the establishment of unique identities of floral organs in different whorls, and the development of these floral organs (Jack, 2004). The floral meristem identity genes, LEAFY (LFY) and APETALA1 (AP1) in Arabidopsis thaliana, and FLORICAULA (FLO) and SQUAMOSA (SQUA) in Antirrhinum majus, are necessary for the determination of floral meristem identity and the formation of floral primordia after floral induction (Coen et al., 1990; Weigel et al., 1992; Parcy et al., 1998). After having been activated by floral meristem identity genes, several groups of floral organ identity genes interact, either directly or indirectly, to control the formation of flowers (Cseke and Podila, 2004). Intensive studies of homeotic mutants of floral development in Arabidopsis and Antirrhinum have led to the ABC model of floral organ identity specification (Bowman et al., 1991; Coen and Meyerowitz, 1991).

The ABC model has been shown to be applicable to a wide range of angiosperm and gymnosperm species (Ng and Yanofsky, 2001; Becker and Theissen, 2003), but with variation in expression patterns and/or functions in many cases (Cseke and Podila, 2004). Although homologues of the floral meristem identity genes LFY and AP1 have been reported in a number of woody genera, including Eucalyptus (Kyozuka et al., 1997), Populus (Rottmann et al., 2000; Cseke and Podila, 2004), Actinidia (Walton et al., 2001), Malus (Yao et al., 1999; Wada et al., 2002), Vitis (Carmona et al., 2002; Calonje et al., 2004), and the New Zealand native species Metrosideros excelsa (Sreekantan et al., 2004), B- and C-class genes in woody angiosperms have been much less studied (Brunner et al., 2000; Yao et al., 2001; van der Linden et al., 2002; Lännenpää et al., 2005; Chaidamsari et al., 2006).

Sophora is a woody leguminous genus containing 45–50 species of small trees and shrubs of worldwide distribution, from south-east Europe across southern Asia, the Pacific, South Atlantic and Indian Oceans, and western South America (Bernardello et al., 2004). Floral development of Sophora is of particular interest because the order of floral organ initiation/development in some species of this genus, as well as in many other papilionoids, differs from the standard sequential pattern of sepal–petal–stamen–carpel (Tucker, 1994, 2003a, b, 2006).

The contrast in the pattern of organ initiation provides interesting material in which to track the expression of the ABC gene homologues, and to compare the genetic control of floral organ identity and specificity between legumes and the model dicot species (Tucker, 2003a). Given the fact that the sequences and functions of the floral identity genes are highly conserved across angiosperm species (Becker and Theissen, 2003), and that the expression pattern of a particular MADS-box gene is usually very well correlated with the functional domain of the gene (Ferrario et al., 2004), the expression patterns of these genes can be good predictors of their function (Zahn et al., 2006). Furthermore, the majority of Arabidopsis flowering genes are represented in Medicago, Glycine, and Lotus sequence databases (Hecht et al., 2005). Therefore, the hypothesis tested in this paper was that homologues of the herbaceous floral identity genes would be expressed in the corresponding reproductive organs of Sophora species.

Homologues of representative A-, B-, and C-class genes, AP1, PISTILLATA (PI), and AG, together with their upstream meristem identity gene, LFY, were isolated from Sophora tetraptera by direct sequencing of reverse transcriptase (RT)-PCR products. Other floral organ identity genes, such as APETALA2 and Sep1-4, were not suitable for this purpose because they are expressed in several floral organs and are less informative in identifying specific floral organs. Consequently, they were not included in this study. A real-time PCR strategy made it possible to study the temporal and spatial expression patterns of all four genes simultaneously, something not yet reported for any species.

	Materials and methods

Top Abstract Introduction Materials and methods Results Discussion Supplementary data References

 
Plant materials
Inflorescences or individual flower buds covering a range of developmental stages were harvested from 3-year-old S. tetraptera plants grown in a nursery plot under prevailing climatic conditions, with a monthly average temperature of 15–20 °C in summer and 5–10 °C in winter.

For histological studies, harvested samples were treated in FAA (formalin–acetic acid–ethyl alcohol) under partial vacuum for 2 h, then under room conditions for 20–24 h. Fixed samples were stored in 70% (v/v) ethanol at 4–7 °C until further preparation. For light microscopy, 10-µm-thick sections were prepared on a Leica RM2145 microtome and counterstained using safranin and fast green. For scanning electron microscopy (SEM), samples were dissected in 95% (v/v) alcohol, further dehydrated through an iso-amyl acetate series, critical-point dried, and coated with gold–palladium. Floral ontogeny and the developmental process were studied and micrographs taken on a Leica 440 scanning electron microscope.

For RNA isolation, harvested samples were immediately frozen in liquid nitrogen and stored at –80 °C until used.

Floral identity gene isolation and sequence analysis
Putative gene homologues were isolated through direct sequencing of RT-PCR products amplified from cDNA templates using degenerate primers because there was no previous molecular information on floral genes and housekeeping genes available for Sophora species. In addition, cloning of genes from New Zealand native species and transformation of these genes into other species are strictly regulated. Degenerate primers for each target gene were designed based on the amino acid and DNA sequences conserved among homologues/orthologues from different species, but not across the different target genes (paralogues) within the MADS-box gene family and were used for PCR amplification (see Table S1 in Supplementary data available at JXB online).

Total RNA was extracted from a mixture of inflorescences and flower buds at early, mid-, and late developmental stages, using a modified TRI Reagent procedure (Molecular Research Center, Cincinnati, OH, USA). In brief, 50–100 mg of frozen sample was ground in a 1.5 ml centrifuge tube with a plastic mini-pestle in the presence of liquid N₂. TRI reagent (800 µl) was added to the sample tube. All other reagents were proportionally decreased according to the TRI reagent protocol. The quantity and quality of RNA were determined using a NanoDrop spectrophotometer (Nyxor, ND-100). RNA integrity was checked by visualization on a 2% (w/v) agarose gel containing 2% (v/v) formaldehyde.

One microgram of total RNA was used for cDNA synthesis using 50 pmol of p(DT)15 primer and 50 U of Expand Reverse Transcriptase (Roche Diagnostics, Mannheim, Germany). Two microlitres of 5-fold-diluted cDNA was used as template for PCR with the following programme: 94 °C for 5 min followed by 30 cycles of 94 °C for 45 s; 56 °C for 45 s, and 72 °C for 1 min, with final extension at 72 °C for 5 min. PCR products were checked on a 1.5% (w/v) agarose gel with ethidium bromide staining. DNA bands with the expected size were excised and purified using the Concert^TM rapid gel extraction system (Life Technologies, Gaithersburg, MD, USA).

The purified fragments of all PCR products were directly sequenced at least once in an ABS 3730 sequencer (Applied Bioscience). At least four independent sequences of both DNA strands were obtained for each pre-selected putative fragment of the target gene and manually corrected to obtain the final sequence data. Both cDNA sequences and deduced amino acid sequences were BLAST searched against the GenBank database (http://www.ncbi.nlm.nih.gov/BLAST).

The deduced amino acid sequences of each putative target gene from S. tetraptera were multiple aligned with their homologues from the GenBank database using ClustalX programs (Thompson et al., 1997). For phylogenetic analysis, Neighbor–Joining trees were constructed with around 10 representative homologues for each of the four target genes, including dicots, monocots, and gymnosperms (see Table S2 in Supplementary data available at JXB online). All analyses were performed with 1000 bootstrap replicates. The graphic tree representation was generated using TreeView software.

The nucleotide sequences reported in this paper have been submitted to the GenBank database under the following accession numbers: DQ418760 [GenBank] (StLFY), DQ418761 [GenBank] (StAP1), DQ418762 [GenBank] (StPI), and DQ418763 [GenBank] (StAG).

Determination of gene copy numbers
A genomic Southern was carried out using a DIG-High Prime DNA Labeling and Detection Starter Kit (Roche Diagnostics, Mannheim, Germany) based on the manufacturer's protocol. PCR products generated using specific primers to StLFY, StAP1, StPI, and StAG were used for probe synthesis for each gene. Genomic DNA (5 µg) was digested at 37 °C for 3 h with 50 U of EcoRI and BamHI, separated through a 1% (w/v) agarose gel, and hybridized under low stringency conditions (2x SSC wash).

For expression Southern analysis, degenerate PCR primers were specifically designed based on the conserved regions of LFY/FLO homologues in a range of species, including dicots and monocots. The same cDNA used for gene isolation was used for PCR. PCR amplification was carried out on a real-time PCR machine (Stratagene, Mx3000P) and SYBR Green DNA dye, in a total volume of 30 µl, containing 1.5 µl of 10-fold-diluted cDNA template, 3.5–4.5 mM MgCl₂, 1.0 µl of each primer (15 µM), and 15 µl of 2x FastStart DNA Master SYBR^® Green I (Roche Diagnostics). The thermal cycling conditions were 95 °C for 10 min followed by 45 cycles of 95 °C for 30 s, 52 °C for 30 s, 72 °C for 30 s. A melting-curve analysis was conducted using one cycle of 95 °C for 0 s, and 50 °C for 40 s, followed by 95 °C for 0 s. Ten microlitres of PCR products were electrophoresed on a 1.5% (w/v) agarose gel. The same hybridization protocol was used as was used for the genomic Southern.

Floral identity gene expression studies
Spatial and temporal expression of putative floral identity genes and selected housekeeping genes was quantified using a LightCycler^TM real-time PCR instrument and SYBR^® Green DNA dye.

Gene-specific primers were designed based on the cDNA sequence of each target and housekeeping gene, spanning at least one intron to eliminate or distinguish amplified contaminating fragments from genomic DNA. All the primers were spaced such that the sizes of amplified products were as similar as possible and within the range of 100–350 nucleotides. The specificity of each primer pair was confirmed by BLAST searching against the GenBank database, by agarose gel electrophoresis, and by melting-curve analysis of PCR products. Sequences of these primers are listed in Table S3 in Supplementary data available at JXB online.

Total RNA samples were extracted from at least two independent tissue samples of each developmental stage or tissue type, which served as biological replicates. RNA samples were quantified using a NanoDrop spectrophotometer (Nyxor, ND-100) before reverse transcription so that similar amounts of total RNA were used in all reverse transcription reactions from different tissue samples. One microgram of total RNA was used with 50 pmol p(DT)15 and 100 pmol p(DN)6 random primers. Two separate reverse transcription assays were carried out for each tissue sample, and the resulting cDNAs were bulked to minimize error caused by variation in reverse transcription efficiency. A 20-fold dilution of the reverse transcription reaction was used for real-time analysis.

Real-time PCR amplifications were carried out in glass capillaries in a total volume of 15 µl, containing 1.5 µl of 20-fold diluted cDNA template, 3.5–4.5 mM MgCl₂, 1.0 µl of each primer (15 µM), and 1.5 µl of FastStart DNA Master SYBR^® Green I (Roche Diagnostics). The thermal cycling conditions were 95 °C for 10 min followed by 45 cycles of 95 °C for 10 s, 52 °C for 20 s, 72 °C for 20 s, and 80–84 °C for 1 s, in order to detect and quantify the fluorescence at a temperature above the denaturation of primer dimers, using a slope of 20 °C s^–1. A melting-curve analysis was conducted using one cycle of 95 °C for 0 s, and 50 °C for 40 s, followed by 95 °C for 0 s. Relative template abundance was quantified using the second derivative maximum method described in the LightCycler manual. At least two PCR runs were carried out for each cDNA to serve as technical replicates.

A serial dilution of 10-, 100-, 1000-, 10 000-, and 100 000-fold of the same cDNA was used to determine the amplification efficiency of each target and housekeeping gene. A standard curve was obtained by plotting the threshold cycle (C_T) value versus the logarithm of the concentration. The PCR efficiency (E) was calculated according to the formula: E=10^(–1/slope)–1 (Pfaffl, 2001).

Quantification of mRNA was based on the C_T value of each sample using the relative quantification method. To minimize experimental error, PCRs containing cDNAs from a series of experimental samples were generally included in a single PCR run, with genes of interest and housekeeping genes sharing the same cDNA template in each of the samples investigated. A negative control with the cDNA template replaced by water was included in the same PCR run for each primer pair.

The most frequently used housekeeping genes for gene expression studies, GAPDH, β-actin, and 18S rRNA, were used as multiple internal controls. For the genes of interest, the C_T value of each sample was adjusted according to the normalization factor or the geometric mean of the housekeeping genes, which were calculated using geNorm v3.3 software (Vandesompele et al., 2002). The adjusted C_T values were then transformed to relative quantities. The highest relative quantity for each gene of interest served as the calibrator and was set to 100%. Each of the normalized values for the same gene of interest was divided by the calibrator to generate the relative expression value, ranging from 0% to 100%.

To compare the expression levels of samples that could not be included in the same PCR run, separate runs were carried out using the same PCR master mix prepared for both runs. In this case, at least two overlapping samples were included in each PCR run for both the gene of interest and the housekeeping genes so that the C_T values of one run were adjusted based on the relative C_T values of these overlapping samples. To compare the expression levels between different genes in the same tissue sample, the C_T value of each gene was first corrected by its amplification efficiency.

Data analysis
Experimental data were analysed as a randomized block experiment by analysis of variance, where PCR run was the block factor, the stage of development was the treatment factor, and relative transcript abundance was the response variable.

	Results

Top Abstract Introduction Materials and methods Results Discussion Supplementary data References

 
Flower ontogeny and development
Inflorescence initiation in S. tetraptera started in late spring (mid-October to November) from leaf axils of the growing shoot, shortly after the previous flowering season. A single flower was initiated in the axil of each bract of the inflorescence 1 week after inflorescence initiation (Fig. 1A). Each floral apex produced paired bracteoles (Fig. 1B). The floral organs were initiated in an atypical acropetal order: sepals, petals, outer stamens plus carpel, and inner stamens. Within each whorl, the order of initiation was unidirectional starting from the abaxial side.

View larger version (112K): [in this window] [in a new window] [Download PowerPoint slide]   Fig. 1. Floral ontogeny and development in Sophora tetraptera using SEM (B–O) and light microscopy (A, P, Q, safranin and fast green counter-stained): (A) inflorescence with floral primordia; (B) single floral meristem before organ initiation; (C) sepal initiation; (D) petal initiation; (E) outer stamen and carpel initiation; (F) mass increase of petals, outer stamens, and carpel, with similar height; (G) inner stamen initiation; (H) completed organ initiation; (I) flower bud before and/or during summer–autumn dormancy; (J) flower bud after summer–autumn dormancy; (K–O) organ differentiation and rapid enlargement; (P) ovule formation; (Q) pollen formation; (R) flower maturation; (S) several inflorescences showing flowers in bloom. The abaxial side is at the base of the figure in (C) to (Q). Some or all sepals were removed in (D) to (O). Some petals were removed in (L), (M), and (N). Some stamens were removed in (L) and (N). Scale bars: B–H=50 µm; A, I, J=100 µm; K–O=200 µm; P, Q=1 mm; R, S=10 mm. Abbreviations: Br=bracteole, C=carpel, Fl=floral primordium, K=keel petal, O=ovule, P=petal, S=outer stamen, s=inner stamen, Sp=sepal, V=vexillum petal, W=wing petal.

 
The first sepal primordium initiated at the median position on the abaxial side, followed by two lateral sepal primordia, and lastly the two adaxial sepal primordia. Long, filamentous trichomes formed on the outside surface of each sepal soon after its initiation (Fig. 1C). The five petals were initiated only after all sepals were present and had enlarged significantly (Fig. 1D). The five outer stamen primordia formed very shortly after initiation of the petals and at the same time as a carpel formed as a distinct central dome (Fig. 1E). Within the petal and outer stamen whorls, the initiation time of these organs was so close that the unidirectional nature of initiation was barely discernible (Fig. 1D, E). After initiation, petal, outer stamen, and carpel primordia developed at more or less the same rate over the next couple of weeks, resulting in floral organs of similar height in these three whorls (Fig. 1F). The five inner stamens initiated much later at the inner side of the petal primordia, at a time when petal, outer stamen, and carpel primordia had started to enlarge markedly (Fig. 1G). The size of the inner stamens remained much smaller than the other floral organs during the remainder of flower development (Fig. 1H, L, M). The organogenesis of a flower bud occurred quite rapidly, normally within 2–3 weeks.

After all the organs within a floral bud had initiated, the floral organs enlarged markedly but remained undifferentiated (Fig. 1H). Synchronization of flower buds initiated at different times occurred at this stage, which was mid- to late summer (January to February) (Fig. 1I). During the next 4–5 months, the development of the floral organs all but ceased and the floral buds remained at a similar developmental stage until early winter (June) (Fig. 1J). However, by July, floral organs resumed their development with a rapid increase in size and differentiation of the petals, stamens, and carpels (Fig. 1K, L); stamens further differentiated into distinct anthers and filaments, carpel into stigma and ovaries (Fig. 1M, N), and petals into vexillum, wing, and keel petals, although with little distinction in shape (Fig. 1O). By August, ovules and pollen grains were well formed and petals elongated rapidly (Fig. 1P, Q). At these late stages, the inner and outer stamens were similar in shape, although the former were about half the height of the latter (Fig. 1N, P, Q). By September, the length of the flower bud reached 10–12 mm, and underwent further enlargement through a series of stages towards their full size at flowering in early October (Fig. 1R).

Sophora tetraptera inflorescences are determinate racemes with 5–10 flowers per inflorescence. Flowers have an unlobed calyx, free petals, and stamens. The corolla of S. tetraptera is non-papilionaceous and its petals are unusually long (40–50 mm) and lack wing sculpturing (Fig. 1S). A colour change with age from greenish to yellow to bright yellow was also observed (Fig. 1R, S).

Isolation of floral identity genes
RT-PCR with degenerate primers was used to isolate partial cDNAs of LFY, AP1, PI, and AG homologues from S. tetraptera using flower bud cDNA as template. One or several PCR products were produced from different primer pairs of each target gene. The BLAST search result of all sequenced PCR products showed that only one putative fragment for each target gene was obtained. The 403 bp putative LFY fragment was named StLFY. Results of BLASTn and BLASTp searches against the GenBank database revealed that StLFY had sequence characteristics typical of a LFY/FLO homologue and its deduced amino acid sequence was most similar to the FLO/LFY-like protein UNIFOLIATA (UNI) in Pisum sativum, with 97.0% identity. Comparison of intron/exon split sites with LFY homologues from a wide range of species showed that the StLFY fragment contained 43 amino acids of the second exon and 90 amino acids of the third exon (see Fig. S1 in Supplementary data available at JXB online). Multiple sequence alignment and the unrooted Neighbor–Joining phylogenetic tree showed that StLFY was well grouped into the LFY/FLO gene class and was closely related to LFY homologues from eudicot species, especially leguminous species (Fig. 2; see Fig. S1 in Supplementary data available at JXB online).

View larger version (8K): [in this window] [in a new window] [Download PowerPoint slide]   Fig. 2. Unrooted Neighbor–Joining phylogenetic tree of representative LEAFY homologues generated with 1000 bootstrap replicates. AFL1, Malus domestica; ALF, Actinidia deliciosa; FLO, Antirrhinum majus; GinLFY, Ginkgo biloba; LFY, Arabidopsis thaliana; LjLFY, Lotus corniculatus var. japonicus; LtLFY, Lolium temulentum; MEL, Metrosideros excelsa, NEEDLY, Pinus radiata; PtLF, Populus trichocarpa; UNI, Pisum sativum; StLFY, Sophora tetraptera; VvLFY, Vitis vinifera; ZFL1, Zea mays. The bar represents 10% amino acid differences.

 
Similarly, a fragment of 441 bp, StAP1, was also isolated. The BLAST search results showed that the deduced amino acid sequence of StAP1 was most similar to the AP1-like protein LjAP1a in Lotus corniculatus var. japonicus, with 95.2% identity. Comparison of intron/exon split sites with AP1 homologues from other species showed that StAP1 fragments spanned exons 1–6 of the AP1 homologue, containing the major part of the MADS-box, the entire sequences of both the I region and the K-box, and partial sequences of the C-terminus (see Fig S2 in Supplementary data available at JXB online). Multiple sequence alignment and the unrooted Neighbor–Joining tree with AP1 homologues from other species showed that StAP1 was closely related to AP1 homologues from leguminous species and other eudicot species (Fig. 3; see Fig. S2 in Supplementary data available at JXB online).

View larger version (14K): [in this window] [in a new window] [Download PowerPoint slide]   Fig. 3. Unrooted Neighbor–Joining phylogenetic tree of representative floral organ identity gene homologues generated with 1000 bootstrap replicates. A-class genes: AAP1, Actinidia deliciosa; AP1, Arabidopsis thaliana; BOAP1, Brassica oleracea; LjAP1a, Lotus corniculatus var. japonicus; MEAP1, Metrosideros excelsa; PEAM4, Pisum sativum; PrMADS2, Pinus radiata; PTAP1-1, Populus trichocarpa; SQUA, Antirrhinum majus; StAP1, Sophora tetraptera; TmAP1, Triticum monococcum. B-class genes: AP3, Arabidopsis thaliana; DEF, Antirrhinum majus; FBPI, Petunia hybrida; GLO, Antirrhinum majus; LjPIa, Lotus corniculatus var. japonicus; MdPI, Malus domestica; NTGLO, Nicotiana tabacum; OrcPI, Orchis italica; PEAM1, Pisum sativum; PI, Arabidopsis thaliana; StPI, Sophora tetraptera; WPI, Triticum aestivum. C-class genes: AG, Arabidopsis thaliana; HvAG1, Hordeum vulgare; LjAGa, Lotus corniculatus var. japonicus; LLAG1, Lilium longiflorum; PEAM7, Pisum sativum; PLE, Antirrhinum majus; PTAG1, Populus trichocarpa; SAG1a, Picea mariana; StAG, Sophora tetraptera; TAG1, Lycopersicon esculentum. The bar represents 10% amino acid differences.

 
A 323 bp fragment of the putative PI gene was also obtained and named StPI. The BLAST search results showed that the deduced amino acid sequence of StPI shared 90.7% identity with PEAM1 in P. sativum. Comparison of intron/exon split sites with PI homologues from other species showed that these fragments spanned exons 1–4 of the PI homologues and contained the complete sequences of the MADS-box and I region, and partial sequences of the K-box (see Fig. S3 in Supplementary data available at JXB online). Multiple sequence alignment and the unrooted Neighbor–Joining tree with PI homologues from other species showed that StPI belongs to the PI/GLO subfamily, but not the AP3/DEF subfamily, of B-class genes, and that it was closely related to PI/GLO homologues from eudicot species, especially leguminous species (Fig. 3; see Fig. S3 in Supplementary data available at JXB online).

An isolated fragment of 343 bp was named StAG. The deduced amino acid sequence was most similar to the AG protein LjAGa from L. corniculatus var. japonicus, with 93.8% identity. However, the sequence similarity of StAG to AG homologues from other species was relatively low, with similarities to other eudicot species ranging between 70% and 85% and between 61% and 67% to those from monocot species available in the GenBank database. Comparison of intron/exon split sites with AG homologues from other species showed that the StAG fragment spanned exons 2–6 of the AG gene, containing short sequences of the MADS-box and the major part of the K-box (see Fig. S4 in Supplementary data available at JXB online). Multiple sequence alignment and the unrooted Neighbor–Joining tree with AG homologues from other species showed that StAG was closely related to AG homologues from other leguminous and eudicot species (Fig. 3; see Fig. S4 in Supplementary data).

The combined information obtained from the BLAST search against the GenBank database, multiple sequence comparison, and phylogenetic analysis strongly suggests that the four isolated fragments from S. tetraptera are homologues of LFY/FLO, AP1/SQUA, PI/GLO, and AG/PLE in Arabidopsis and Antirrhinum, respectively.

To determine the copy number of these genes, Southern hybridization was performed using S. tetraptera genomic DNA digested with two restriction enzymes, EcoRI and BamHI. The hybridization patterns obtained using a specific probe to each gene under low-stringency conditions resulted in a single band for StAP1, StPI, and StAG. This suggests that each of the three genes occurs as a single copy in this species. However, two bands presented when blotted with the StLFY probe using both enzymes, indicating that two copies of a StLFY-like gene may exist in the S. tetraptera genome (see Fig. S5 in Supplementary data available at JXB online).

An expression Southern combined with a melting-curve analysis from the real-time PCR assay was used to determine if both copies of StLFY were expressed during floral development. Using degenerate primers specifically designed within conserved regions of LFY/FLO homologues, all of the four primer pairs generated PCR products with only one discrete peak in the melting-curve plot. Hybridization with the StLFY-specific probe also resulted in a single band (see Fig. S6 in Supplementary data available at JXB online). This may suggest that only one of the two StLFY copies is expressed during floral development.

Protocol optimization and real-time PCR efficiency determination
Two to four primer pairs of each target and housekeeping gene were tested for their efficacy in amplifying the specific target cDNA. Only those amplifying single PCR products of the expected size with an efficiency value (E-value) over 0.85 were chosen for further quantitative expression studies. The determined E-values for the selected primer pairs, between 0.87 and 0.98, are summarized in Table 1, and were taken into account in subsequent quantifications. Other PCR conditions including MgCl₂ concentration and cDNA dilution were also optimized (Song, 2005).

View this table: [in this window] [in a new window]   Table 1. PCR efficiencies of floral identity genes and selected housekeeping genes

 
The C_T values varied between 17 and 35 cycles between different genes, organs/tissues, and developmental stages, indicating a detection power of up to five orders of magnitude for gene transcript levels. PCR products, if any, detected between 35 and 40 cycles were non-specific to the target genes as verified by melting-curve and gel analysis. In some cases, non-specific PCR products amplified between 30 and 35 cycles but were excluded from further analysis.

Spatial expression patterns of floral identity genes
To test the tissue-specific expression of the gene homologues from S. tetraptera, their relative mRNA accumulation was quantitatively examined in various tissues including shoot tips, adult leaves, inflorescences, flower buds, and seed pods, using real-time PCR analysis.

The lowest C_T values for different floral identity genes ranged between 22 and 27, indicating an mRNA concentration difference of up to 30-fold between these genes at their highest expression levels. Transcripts of StLFY were detected in all tissues tested but the expression levels varied substantially between tissue types (Fig. 4). The highest expression level (100%) was observed in early-stage flower buds, followed by inflorescences, being 16.2% of its highest levels. In vegetative shoot tips, mRNA accumulation of 5.1% of its highest level was also detected. Expression of StLFY in adult leaves, mid-stage flower buds, and young seedpods was very low, being 500- to 1500-fold lower than that in early-stage flower buds (Fig. 4).

View larger version (10K): [in this window] [in a new window] [Download PowerPoint slide]   Fig. 4. Expression of floral identity genes in selected vegetative and reproductive tissues of Sophora tetraptera. Relative mRNA levels were determined using real-time PCR and normalized using housekeeping genes 18S, GAPDH, and β-actin as internal controls. Values are mean and standard error (error bars ±1 SE) of four replicates. Inflo, Inflorescences 2–3 mm in length; Flo buds 1, early-stage flower buds 2–3 mm in length; Flo buds 2, mid-stage flower buds 18–20 mm in length. Seed pods are 2–3 weeks after flowering.

 
The highest StAP1 mRNA accumulation was observed in early-stage flower buds, being 6-fold and 16-fold higher than that in inflorescences and mid-stage flower buds, respectively. No mRNA was detected in adult leaves, vegetative shoot tips, or seedpods, even after 40 PCR cycles (Fig. 4).

Similarly to StAP1, StPI expression was only detected in flower buds and inflorescences, with the highest mRNA levels accumulated in mid-stage flower buds. However, by contrast to StLFY and StAP1, expression of StPI in early stage flower buds was relatively low, representing 38% of that in mid-stage flower buds. Only 0.6% of StPI expression was detected in the inflorescence (Fig. 4).

The expression profile of StAG was very similar to that of StPI, with 74% mRNA accumulation in early-stage flower buds and 1.2% in inflorescences compared with its highest expression level (100%) in mid-stage flower buds. In addition, a low level of StAG expression (2.1%) was also detected in seedpods (Fig. 4).

Organ-specific expression of floral organ identity genes
To test the specificity of expression of the A, B, and C gene homologues in floral organs in Sophora, expression of AP1, PI, and AG homologues, and an internal control, St18S, were quantified in individually dissected floral organs of late-stage flower buds (18–20 mm in length, showing the petal tip (Fig. 1R2). Relative mRNA levels were calculated across all the genes tested (Fig. 5).

View larger version (15K): [in this window] [in a new window] [Download PowerPoint slide]   Fig. 5. Expression of floral organ identity genes and the 18S homologue in different floral organs of Sophora tetraptera. Ten flower buds at late developmental stage (18–20 mm in length showing petal tip) were individually dissected to obtain a mixed sample of each floral organ. Relative mRNA levels were determined using real-time PCR, calculated, and presented on a logarithmic scale across all the genes tested. Values are mean and standard error (error bars ±1 SE) of four replicates.

 
The expression of the internal control gene, St18S, was relatively stable in the four floral organ tissues examined, with a detected variation within 40% between organ types. The A-class gene StAP1 was highly expressed in sepals and petals. Although an expression of over 1000-fold lower than that in sepals was detected in carpels, there was no detectable StAP1 expression in petals. The B-class gene StPI was highly expressed in petals and stamens while no expression was detected in sepals and carpels. A high expression level of the C-class gene, StAG, was detected in stamens and carpels. While a trace expression (10 000-fold lower than that in stamens) of the AG homologue was detected in petals, no StAG expression was observed in sepals.

Temporal expression patterns of floral identity genes
Detailed expression profiles of the four floral identity genes during a complete reproductive growth cycle, from the initiation of inflorescence primordia in spring (October) to flowering time the following spring, were analysed using real-time PCR. Here, the aim was to elucidate the role of the floral identity genes of Sophora in floral development, especially their potential influence on the long dormant period during floral organogenesis, and the unusual order of floral organ initiation and development.

Expression patterns of the four floral identity genes in S. tetraptera grouped into two categories (Fig. 6). Expression of StLFY and StAP1 was characterized by a bimodal pattern: high mRNA levels were first detected during inflorescence and early floral developmental stages in summer (December to January) and, subsequently, again during mid-stage flower bud development in winter (June to August). By contrast, both StPI and StAG showed one main phase of expression during mid- and late-stage floral development in winter. Their expression remained very low or undetectable during inflorescence and early floral developmental stages in summer and autumn. All genes were expressed at very low levels in fully open flowers, being over 500-fold lower than the peak level for StLFY, over 15-fold lower for StAP1 and StAG, and nearly 4-fold lower for StPI (Fig. 6).

View larger version (16K): [in this window] [in a new window] [Download PowerPoint slide]   Fig. 6. Expression profiles of floral identity genes in Sophora tetraptera during a complete growth cycle. Relative mRNA levels were determined using real-time PCR and normalized using housekeeping genes 18S, GAPDH, and β-actin as internal controls. Values are mean and standard error (error bars ±1 SE) of four replicates. Arrows show the starting points of: I, floral meristem initiation; II, floral organ initiation; III, floral organ differentiation and development. Sample descriptions: 6 Oct, Shoot tips with inflorescence primordia; 3 Nov, inflorescences (2–3 mm in length); 19 Dec, inflorescences (10–12 mm) with floral primordia; 8 Jan, inflorescences (20–30 mm) with flower bud (1 mm) and all floral organs initiated; 28 Jan to 1 May, flower buds (1–1.5 mm) with undifferentiated floral organs; 26 Jun, flower buds (1.5–2 mm), with differentiated floral organs; 24 Jul, flower buds (2–3 mm) with elongated anthers and pistils; 15 Aug, flower buds (5–6 mm) starting ovule formation; 18 Sep, flower buds (18–20 mm) showing petal tip, with well-formed pollen and ovules; 7 Oct, mature flower buds before opening.

 

	Discussion

Top Abstract Introduction Materials and methods Results Discussion Supplementary data References

 
The SEM and light microscopy results showed that floral ontogeny in S. tetraptera shared some distinctive features with related species. These included the precocious carpel initiation and advanced sepal development previously noted within the genus Sophora (Tucker, 1994) and, within the subfamily Papilionoideae (Tucker, 2006), the acropetal initiation order among whorls, overlapping initiation between whorls, and unidirectional order starting from the abaxial side in sepal, petal, and stamen whorls (Tucker, 1994, 2003a, 2006; Ferrándiz et al., 1999; Benloch et al., 2003).

In S. tetraptera the unidirectional order among floral organs of the same whorl was barely discernible for petals and stamens, although very evident for sepals (Fig. 1D, E). This almost simultaneous initiation of all organs within petal and stamen whorls has not been reported in other Sophora species (Tucker, 1994). Although S. tetraptera has unusually long petals of 40–50 mm, petal development was significantly delayed in comparison with the outer stamens and carpel. Full petal size was attained during a short period of rapid enlargement during the maturation of the flower buds. Such delayed development of petals has been noted in a few of the other species in the genus Sophora (Tucker, 1994), and in other species in the subfamily Papilionoideae (Tucker, 2006).

More interestingly, after all organ initiation has been completed, floral organ development is then interrupted by a prolonged summer–autumn dormant period, during which little further development of any floral organ occurs.

In summary, precocious carpel initiation, delayed petal development, and the interrupted floral organ development observed in the present study in S. tetraptera are all significantly different from the pattern of floral organ initiation and development in Arabidopsis and Antirrhinum on which the ABC model has been based. These unusual patterns of initiation and development in S. tetraptera provided an interesting system in which to track the spatial and temporal expression of the ABC genes in relation to initiation, differentiation, and development of the different organ types.

Analysis of the partial sequences of the putative floral identity genes strongly suggests that these genes are homologues of LFY/FLO, AP1/SQUA, PI/GLO, and AG/PLE for Arabidopsis/Antirrhinum, respectively. The results were also consistent with the taxonomic relationship of the species analysed, with the newly isolated fragments being most closely related to those of other leguminous species, and more related to dicots than to monocots and gymnosperms.

Southern blotting revealed a single copy of StAP1, StPI, and StAG, while two copies of StLFY may exist in the S. tetraptera genome. Multiple copies of LFY homologues have been reported in a number of species, both dicots and monocots. While the two copies of LFY homologues in maize (Bomblies and Doebley, 2006) and apple (Wada et al., 2002) are functional and both play a similar role to LFY, only one of the two copies of the LFY homologues in Eucalyptus globulus (Southerton et al., 1998) and Nicotiana tabacum (Ahearn et al., 2001) is functional; the other copy is considered to be a pseudogene (Eucalyptus) or with undetermined function (Nicotiana).

It was expected that expressed copies of StLFY would be amplified using degenerate primers designed according to the conserved sequence segments among homologues from a wide range of species, and that the amplified fragments would be distinguishable based either on size difference on an agarose gel or on melting-point difference by melting-curve analysis from the real-time PCR program. The combined data from these techniques revealed a single expressed copy of StLFY during flower development. The other copy may be either a pseudogene or a paralogue with a role other than in floral identity.

A two-step quantitative real-time RT-PCR protocol was established. Expression differences of up to five orders of magnitude between different genes, and different stages of flower development and tissue types of the same gene were effectively detected. This detection ability is comparable with that obtained by Czechowski et al. (2004) for over 1400 Arabidopsis transcription factors using a similar real-time PCR protocol, suggesting that real-time PCR is a very powerful tool for precise gene expression studies.

For precise quantification of target gene expression, the non-specific DNA-binding nature of SYBR^® Green dye was efficiently overcome by analysing the melting-curve of PCR products (Madani et al., 2005) in the current study. The problem of genomic DNA contamination was resolved by using primers that spanned one or several introns. Since the MADS box genes are highly conserved across angiosperm (Montag et al., 1995), gymnosperm (Sundstroem et al., 1999), and non-flowering plant species (Henschel et al., 2002), the estimated structure of the isolated fragments provided sufficient information for designing intron-spanning primers without further verifying their precise exon–intron boundaries by additional sequencing of intron-containing genomic DNA.

Accurate normalization of experimental data is an absolute prerequisite for correct measurement of gene expression using real-time RT-PCR. The most commonly used normalization strategy involves standardization to a constitutively expressed housekeeping gene as the internal control (Bustin and Nolan, 2004). However, Song (2005) showed that despite their relatively stable expression among different floral organs of the same developmental stage, the three most frequently used housekeeping genes, 18S rRNA, GAPDH, and β-actin, had substantial expression variations across different developmental stages of the same tissue type in S. tetraptera, consistent with a number of recent reports (see Czechowski et al., 2005). Therefore, it is not recommended that these housekeeping genes be used independently as internal controls for normalizing the expression of target genes during different developmental stages. By using the geometric mean of these three housekeeping genes, calculated using the method described by Vandesompele et al. (2002), variations between developmental stages were reduced to an acceptable level compared with the substantial differences in the expression levels of the target genes. Therefore, the geometric mean was used to normalize target gene expression in the current study.

Real-time PCR enabled a unique simultaneous and quantitative analysis of both the temporal and spatial expression patterns of four key floral identity genes. The spatial expression patterns of the A-, B-, and C-class floral organ identity gene homologues in S. tetraptera were very similar to the expression of their counterparts in Arabidopsis and Antirrhinum, with the expression territory of the A-class gene homologue, StAP1, strictly limited to floral primordia during floral commitment and to sepals and petals after floral organ differentiation. The expression of the B-class gene homologue, StPI, was limited to petals and stamens, and the C-class gene homologue, StAG, to stamens, carpels, and seedpods. Although a detection power of up to five orders of magnitude was demonstrated in the current study, activity of these genes was detected only in restricted tissues or floral organs. Apart from the extended low-level expression of StAG in seedpods, these expression patterns are completely consistent with the defined roles of these genes in Arabidopsis and Antirrhinum (Bowman et al., 1991; Coen and Meyerowitz, 1991).

The observed expression of StAG in seedpods is not in agreement with the model species Arabidopsis. Although AG is a co-determinant of ovule identity and is functionally redundant to SHATTERPROOF (SHP) and SEEDSTICK (STK) in Arabidopsis (Pinyopich et al., 2003), the role of AG in Arabidopsis fruit development has not yet been described. More recently, Ordidge et al. (2005) observed that the AG homologue in Impatiens balsamina, IbAG, was expressed in the floral meristem until the production of the ovules, but was not expressed after ovule production. The extended expression of AG homologues in developing fruits has been observed in several other species, including Solanum lycopersicum (Giovannoni, 2004), Eschscholzia californica (Zahn et al., 2006), and Theobroma cacao (Chaidamsari et al., 2006). Consequently, the extended function of AG homologues in fruit development warrants further investigation.

Floral development in S. tetraptera was characterized by a significant temporal separation between organ initiation and organ differentiation of several months’ duration during the summer–autumn period. Both StLFY and StAP1 showed a bimodal seasonal expression pattern. The expression peaks were detected initially during floral organ initiation and subsequently during mid-stage flower bud development, and were separated by a period of low expression during the late summer and autumn period. This low expression period was coincident with the summer–autumn dormant period of floral development. Bimodal expression patterns have also been described in a number of woody species including M. excelsa (Sreekantan et al., 2004), A. deliciosa (Walton et al., 2001), V. vinifera (Carmona et al., 2002), and P. tremuloides (Cseke and Podila, 2004). However, by contrast to S. tetraptera, the two expression peaks in all other species were separated by a winter dormancy period before floral organ initiation (Walton et al., 2001; Carmona et al., 2002; Cseke and Podila, 2004; Sreekantan et al., 2004).

Previous studies of LFY homologues in other woody perennials showed a peak of expression at the time of inflorescence initiation (Walton et al., 2001; Sreekantan et al., 2004). By contrast, StLFY expression was relatively low at the equivalent developmental stage, being only 15% of the peak level. This low expression in inflorescences compared with the high expression during floral initiation and development suggests that StLFY might play a more important role in the inflorescence to floral transition than in the vegetative to inflorescence transition, and that the quantitative regulation of LFY homologues is important in fulfilling their role of establishing floral meristem identity and floral organ development. Indeed, the present data showed that StLFY was expressed in both reproductive and vegetative tissues, albeit at a lower level in the latter. The precise quantification of the expression of LFY homologues in vegetative tissues supports the suggestion that expression of the S. tetraptera LFY homologue is not sufficient to confer reproductive fate.

The data strongly suggest that all of the A-, B-, and C-class genes are also required during the later stages of flower organ development. Consequently, the roles of floral organ identity genes in organ identity and development may also be separate, and the dose effect of these genes may be more crucial for organ development than for organ specification. This was clearly demonstrated in S. tetraptera in which StAP1, StPI, and StAG were expressed at very low levels, within 10% of their peak levels, from December to May, during the period when organ initiation was completed but floral organs remained undifferentiated (Fig. 1H, I). Such low expression levels of the floral identity genes might be the main barrier for organ differentiation, although this remains to be verified. High expression levels of these genes were detected only after the onset of floral organ differentiation and volume expansion. Further, these levels remained high until the late stages of flower development, indicating that only low expression levels of floral identity genes are required for floral meristem/organ initiation, but high levels are required not only for floral organ differentiation but also for development.

The temporal expression patterns of PI and AG homologues in S. tetraptera reflected accurately the precocious carpel initiation and development and the delayed petal development. Had S. tetraptera followed the classic floral organ initiation order of sepal–petal–stamen–carpel, a markedly earlier and elevated PI expression relative to that of AG would have been expected. However, in S. tetraptera, very similar onset times and expression levels of StPI and StAG during floral organ initiation and early development were observed (Fig. 6). Further, it was considered unlikely that the expression profile of the other B-class gene, the AP3 homologue, would affect the explanation and conclusion in this study, because the AP3 homologue(s) would not be functional without forming heterodimer(s) with StPI. These data also suggest that the expression territory of the floral organ identity genes may be strictly limited to their defined floral organs regardless of the timing and spatial variations during initiation and development of these organs.

In conclusion, this study has provided a comprehensive quantitative expression profile of representatives of the A-, B-, and C-class genes, together with their upstream regulating gene, LFY, in a woody legume, a profile which fits both the timing and sequence of organ development in S. tetraptera and matches expression profiles of the A-, B-, and C-class genes of Arabidopsis. Because of the length of time separating floral initiation and developmental events, the relative importance of floral identity genes in meristem/organ specification and floral organ development for S. tetraptera homologues is clearly demonstrated.

	Supplementary data

Top Abstract Introduction Materials and methods Results Discussion Supplementary data References

 
The following supplementary data can be found at JXB online.

Table S1. Degenerate primers used for floral identity gene isolation from Sophora tetraptera.

Table S2. Floral identity genes used for multiple alignment and phylogenetic analysis.

Table S3. Real-time PCR primers for expression analysis of floral identity genes and housekeeping genes in Sophora tetraptera.

Fig. S1. Comparison of amino acid sequences of StLFY with representative FLO/LFY-like proteins.

Fig. S2. Comparison of amino acid sequences of StAP1 with representative AP1/SQUA-like proteins.

Fig. S3. Comparison of amino acid sequences of StPI with representative PI-like proteins.

Fig. S4. Comparison of amino acid sequences of StAG with representative AG/PLE-like proteins.

Fig. S5. Southern blot analysis of StLFY, StAP1, StPI and StAG genes in S. tetraptera.

Fig. S6. Real-time melting curve analysis and expression Southern blot of StLFY in S. tetraptera.

	Acknowledgements

 
This work was supported by The New Zealand Foundation for Research, Science & Technology Public Good Science Fund under the Native Ornamental Plants Programme through subcontract C02X0015 to Crop & Food Research (JC), and a Massey University PhD Scholarship (JS). We acknowledge Mr Neil Andrew for his technical assistance in SEM analysis and Dr Rod King for constructive criticism of our manuscript.

	References

Top Abstract Introduction Materials and methods Results Discussion Supplementary data References

 
Ahearn KP, Johnson HA, Weigel D, Wagner DR. NFL1, a Nicotiana tabacum LEAFY-like gene, controls meristen initiation and floral structure. Plant and Cell Physiology (2001) 42:1130–1139.[Abstract/Free Full Text]

Becker A, Theissen G. The major clades of MADS-box genes and their role in the development and evolution of flowering plants. Molecular Phylogenetics and Evolution (2003) 29:464–489.[CrossRef][ISI][Medline]

Benloch R, Navarro C, Beltran JP, Canas LA. Floral development of the model legume Medicago truncatula: ontogeny studies as a tool to better characterize homeotic mutations. Sexual Plant Reproduction (2003) 15:231–241.[ISI]

Bernardello G, Aguilar R, Anderson GJ. The reproductive biology of Sophora fernandeziana (Leguminosae), a vulnerable endemic species from Isla Robinson Crusoe. American Journal of Botany (2004) 91:198–206.[Abstract/Free Full Text]

Bomblies K, Doebley JF. Pleiotropic effects of the duplicate maize FLORICAULA/LEAFY genes zfl1 and zfl2 on traits under selection during maize domestication. Genetics (2006) 172:519–531.[Abstract/Free Full Text]

Bowman JL, Smyth DR, Meyerowitz EM. Genetic interactions among floral homeotic genes of Arabidopsis. Development (1991) 112:1–20.[Abstract]

Brunner AM, Rottmann WH, Sheppard LA, Krutovskii K, DiFazio SP, Leonardi S, Strauss SH. Structure and expression of duplicate AGAMOUS orthologues in poplar. Plant Molecular Biology (2000) 44:619–634.[CrossRef][ISI][Medline]

Bustin A, Nolan T. Pitfalls of quantitative real-time reverse-transcription polymerase chain reaction. Journal of Biomolecular Techniques (2004) 15:155–166.[Medline]

Calonje M, Cubas P, Martínez-Zapater JM, Carmona MJ. Floral meristem identity genes are expressed during tendril development in grapevine. Plant Physiology (2004) 135:1491–1501.[Abstract/Free Full Text]

Carmona MJ, Cubas P, Martinez-Zapater JM. VFL, the grapevine FLORICAULA/LEAFY ortholog, is expressed in meristematic regions independently of their fate. Plant Physiology (2002) 130:68–77.[Abstract/Free Full Text]

Chaidamsari T, Samanhudi, Sugiarti H, Santoso D, Angenent GC, de Maagd RA. Isolation and characterization of an AGAMOUS homolog from cocoa. Plant Science (2006) 170:968–975.

Coen ES, Meyerowitz EM. The war of the whorls: genetic interactions controlling flower development. Nature (1991) 353:31–37.[CrossRef][Medline]

Coen ES, Romero JM, Doyle S, Elliot R, Murphy G, Carpenter R. FLORICAULA: a homeotic gene required for flower development in Antirrhinum majus. Cell (1990) 63:1311–1322.[CrossRef][ISI][Medline]

Cseke LJ, Podila GK. MADS-box genes in dioecious aspen. II. A review of MADS-box genes from trees and their potential in forest biotechnology. Physiology and Molecular Biology of Plants (2004) 10:7–28.

Czechowski T, Bari RP, Stitt M, Scheible WR, Udvardi MK. Real-time RT-PCR profiling of over 1400 Arabidopsis transcription factors: unprecedented sensitivity reveals novel root- and shoot-specific genes. The Plant Journal (2004) 38:366–379.[CrossRef][ISI][Medline]

Czechowski T, Stitt M, Altmann T, Udvardi MK, Scheible WR. Genome-wide identification and testing of superior reference genes for transcript normalization in Arabidopsis. Plant Physiology (2005) 139:5–17.[Abstract/Free Full Text]

Ferrándiz C, Navarro C, Gomez MD, Canas LA, Beltran JP. Flower development in Pisum sativum: from the war of the whorls to the battle of the common primordia. Developmental Genetics (1999) 25:280–290.[CrossRef][ISI][Medline]

Ferrario S, Immink RG, Angenent GC. Conservation and diversity in flower land. Current Opinion in Plant Biology (2004) 7:84–91.[CrossRef][ISI][Medline]

Giovannoni JJ. Genetic regulation of fruit development and ripening. The Plant Cell (2004) 16:S170–S180.[Free Full Text]

Hecht V, Foucher F, Ferrándiz C, et al. Conservation of Arabidopsis flowering genes in model legumes. Plant Physiology (2005) 137:1420–1434.[Abstract/Free Full Text]

Henschel K, Kofuji R, Hasebe M, Saedler H, Münster T, Theißen G. Two ancient classes of MIKC-type MADS-box genes are present in the moss Physcomitrella patens. Molecular Biology and Evolution (2002) 19:801–814.[Abstract/Free Full Text]

Jack T. Molecular and genetic mechanisms of floral control. The Plant Cell (2004) 16:S1–17.[Free Full Text]

Kyozuka J, Harcourt R, Peacock WJ, Dennis ES. Eucalyptus has functional equivalents of the Arabidopsis AP1 gene. Plant Molecular Biology (1997) 35:573–584.[CrossRef][ISI][Medline]

Lännenpää M, Parkkinen S, Järvinen P, Lemmetyinen J, Vepsäläinen S, Savola T, Keinonen K, Keinänen M, Sopanena T. The expression and promoter specificity of the birch homologs for PISTILLATA/GLOBOSA and APETALA3/DEFICIENS. Physiologia Plantarum (2005) 125:268–280.[CrossRef]

Madani M, Subbotin SA, Moens M. Quantitative detection of the potato cyst nematode, Globodera pallida, and the beet cyst nematode, Heterodera schachtii, using real-time PCR with SYBR green I dye. Molecular and Cellular Probes (2005) 19:81–86.[CrossRef][ISI][Medline]

Montag K, Salamini F, Thompson RD. ZEMa, a member of a novel group of MADS box genes, is alternatively spliced in maize endosperm. Nucleic Acids Research (1995) 23:2168–2177.[Abstract/Free Full Text]

Ng M, Yanofsky MF. Function and evolution of the plant MADS-box gene family. Nature Reviews Genetics (2001) 2:186–195.[CrossRef][ISI][Medline]

Ordidge M, Chiurugwi T, Tooke F, Battey NH. LEAFY, TERMINAL FLOWER1 and AGAMOUS are functionally conserved but do not regulate terminal flowering and floral determinacy in Impatiens balsamina. The Plant Journal (2005) 44:985–1000.[CrossRef][ISI][Medline]

Parcy F, Nilsson O, Busch MA, Lee I, Weigel D. A genetic framework for floral patterning. Nature (1998) 395:561–566.[CrossRef][Medline]

Pfaffl MW. A new mathematical model for relative quantification in real-time RT-PCR. Nucleic Acids Research (2001) 29:e45.[Abstract/Free Full Text]

Pinyopich A, Ditta GS, Savidge B, Liljegren SJ, Baumann E, Wisman E, Yanofsky MF. Assessing the redundancy of MADS-box genes during carpel and ovule development. Nature (2003) 424:85–88.[CrossRef][Medline]

Rottmann WH, Meilan R, Sheppard LA, Brunner AM, Skinner JS, Ma C, Cheng S, Jouanin L, Pilate G, Strauss SH. Diverse effects of overexpression of LEAFY and PTLF, a poplar (Populus) homolog of LEAFY/FLORICAULA, in transgenic poplar and Arabidopsis. The Plant Journal (2000) 22:235–245.[CrossRef][ISI][Medline]

Song J. Genetic diversity and flowering in Clianthus and New Zealand Sophora (Fabaceae). PhD thesis, Massey University, Palmerston North, New Zealand. (2005) 152–176.

Southerton SG, Strauss SH, Olive MR, Harcourt RL, Decroocq V, Zhu XM, Llewellyn DJ, Peacock WJ, Dennis ES. Eucalyptus has a functional equivalent of the Arabidopsis floral meristem identity gene LEAFY. Plant Molecular Biology (1998) 37:897–910.[CrossRef][ISI][Medline]

Sreekantan L, Clemens J, McKenzie MJ, Lenton JR, Croker SJ, Jameson PE. Flowering genes in Metrosideros fit a broad herbaceous model encompassing Arabidopsis and Antirrhinum. Physiologia Plantarum (2004) 121:163–173.[CrossRef][Medline]

Sundstroem J, Carlsbecker A, Svensson ME, Svenson M, Johanson U, Theissen G, Engström P. MADS-box genes active in developing pollen cones of Norway spruce